Contemporary ENERGY Vol2 No1 (2016) by REMOO

International Journal of Contemporary ENERGY Peer-reviewed open-access E-journal

ISSN 2363-6440

Vol. 2, No. 1 (2016) February 2016 www.Contemporary-ENERGY.net

Publisher

Get It Published Verlag e.k. Allee am Röthelheimpark 14 91052 Erlangen GERMANY

T 00 49 (0)9131 917 96 14 E info@get-it-published.de W www.get-it-published.de

Copyright This journal and all published articles, including all illustrations contained in authors’ papers block, are protected by copyright. Upon an article being accepted for publication, all rights of publication, for translation, further reproduction, distribution, transmission, display, broadcast, of storage in any electronic form and producing photocopies are transferred to the publisher. Without the written permission of the publisher, any usage outside the limits of the copyright act is forbidden.

© Copyright by Get It Published Verlag e.K. Notice The publisher does not assume any responsibility for any harm and/or injury to property and persons resulting from any ideas, instructions, methods or products contained in the material published in this journal, as well as a matter of inattention or creation liability, or from any use or operations.

Cover Illustration Gewerbegebiet Illustration; Source: http://de.123rf.com; Copyright: Gui Yongnian

Founding Editor & Editor-in-Chief Zoran V. Stosic

–

editor@contemporary-energy.net

Director RENECON International, GERMANY; Former Vice President ICO South East Europe at AREVA, GERMANY

Editorial Board Prof. Jan Blomgren

–

Uppsala University; CEO of INBEx, SWEDEN

Ass. Prof. Leon Cizelj

–

University of Ljubljana; Head of Reactor Engineering Division at IJS, SLOVENIA

Ass. Prof. Davor Grgić

–

Faculty of Electrical Engineering and Computing, University of Zagreb, CROATIA

Prof. Nikola Čavlina

–

Faculty of Electrical Engineering and Computing, University of Zagreb, CROATIA

Dr. Ludger Mohrbach

–

Head of Competence Center Nuclear Power Plants at VGB PowerTech e.V., Essen, GERMANY

Dr. Maximilian Emanuel Elspas

–

Head of Energy Law and Lawyer Partner at Beiten Burkhardt Law Munich, GERMANY

Dr. Dietmar O. Reich

–

Co-Head of Competition Practice Group and Lawyer Partner at Beiten Burkhardt Law Brussels, BELGIUM

Dr. Miodrag Mesarović

–

Secretary General of the SerbianWEC Member Committee; Senior Advisor to Energoprojekt-ENTEL, Belgrade, SERBIA

Prof. Ana M. Lazarevska

–

Faculty of Mechanical Engineering, University of Skopje, MACEDONIA

LL.M. Ana Stanič

–

Lawyer Principal at E&A Law London, UNITED KINGDOM

Prof. Li Ran

–

School of Engineering, University of Warwick, UNITED KINGDOM; Deputy Director of China State Key Lab in Power Transmission Apparatus Security, Chongqin University, CHINA

Dr. Changxin Liu

–

Deputy Director General of China National Nuclear Corporation – CNNC, Beijing CHINA

Prof. Xu Cheng

–

Institute of Fusion and Nuclear Technology, Karlsruhe Institute of Technology – KIT, GERMANY; School of Nuclear Sciences and Engineering, Shanghai, Jiao Tong University, CHINA

Prof. Josua P. Meyer

–

Department of Mechanical and Aeronautical Engineering, University of Pretoria, SOUTH AFRICA

Prof. Zhao Yang Dong

–

Chair Professor and Head of School of Electrical and Information Engineering, University of Sidney, AUSTRALIA

M.Sci.Engng. Jukka Tapani Laaksonen

–

Vice President ROSATOM Overseas, Moscow, RUSSIA; Former Director General of the STUK, FINLAND

M.Sci.Engng. Jože Špiler

–

Head of TechnicalServices and Investments at GEN-energija, Krško, SLOVENIA

Prof. Michael Narodoslawsky

–

Institute for Process and Particle Engineering, Technical University of Graz, AUSTRIA

Dr. Raffaella Gerboni

–

Post-Doc Fellow Researcher, Energy Department, Politecnico di Torino, ITALY

Prof. Henryk Anglart

–

Deputy Head of Physics Department, KTH Royal Institute of Technology, Stockholm, SWEDEN

Dr. Suna Bolat

–

Assistant Professor, Eastern Mediterranean University – EMU, Famagusta, North Cyprus, TURKEY

Prof. Nikola Popov

–

Faculty of Engineering Physics, McMaster University, Hamilton; President DENIPO Consulting Ltd., Toronto, Ontarion, CANADA

Prof. Milovan Perić

–

Managing Director of CoMeT Continuum Mechanics Technologies GmbH, GERMANY; Senior Corporate Consultant CD-adapco, UNITED KINGDOM

Prof. Umberto Desideri

–

Department of Energy Engineering, University of Pisa, ITALY

Prof. Chul-Hwa Song

–

University of Science and Technology – UST, Seoul; Director of Thermal-Hydraulics Safety Research Div., KAERI, Daejeon, SOUTH KOREA

Prof. Shpetim Lajqi

–

Faculty of Mechanical Engineering, University of Prishtina, KOSOVO

Dr. Camila Braga Viera

–

Post-Doc Researcher, SCK-CEN Belgian Nuclear Research Center, Boeretang, BELGIUM

Ass. Prof. Manuel Ruiz de Adana Santiago

–

Department of Applied Thermodynamics, University of Cordoba, SPAIN

Ass. Prof. Roonak Daghigh

–

Department of Mechanical Engineering, University of Kurdistan, Sanandaj, IRAN

Dr. Cristina Cornaro

–

Assistant Professor, Department of Enterprise Engineering, University of Rome “Tor Vergata”, ITALY

Dr. Naseem Udin

–

Principal Lecturer, Institute Teknology Brunei, BRUNEI

Prof. Gordana Laštovička -Medin

–

Faculty of Science and Mathematics, University of Montenegro, Podgorica, MONTENEGRO

Prof. Serkan Dag

–

Department of Mechanical Engineering, Middle East Technical University – METU, Ankara, TURKEY

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

A WORD FROM THE EDITOR-IN-CHIEF It is nice to say that with this issue the Journal is somehow celebrating its first birthday. We are trying, and based on feedbacks we are getting, we are succeeding to be better than a year before. In addition, we have more than 30 high-qualified Associate Editors from all around the world helping in reviewing and selecting the articles to be published. Most of the articles published in the Journal are still selected from the best papers presented at the International Conference & Workshop REMOO. At the beginning of 2016 the REMOO conference was certified by ACE (Academic Conference Excellence) thus proving also the scientific quality of the international event, lasting more than 6 years and feeding the Journal. For this issue two Associate Editors, Jan Blomgren and Ana Lazarevska, have written very good editorials about “Energy – A Truly Multi–Faceted Field” and an overview of the 5th REMOO conference “Technological, Modelling and Experimental Achievements in Energy Generation Systems”, held on 23rd and 24th of September 2015 in Budva, Montenegro. Founding Editor & Editor–In–Chief Zoran V. Stosic

A selection of authors’ contributions part begins this time with a short summary of a PhD Thesis dealing with important aspects of energy market in Europe. Finally, REMOO–2016 “Science and Engineering for Reliable Energy” will be held on 18–20 May 2016 again in Budva, Montenegro. Seven worldwide known experts will give remarkable keynote speeches on important issues covering most updated topics in energy field. Corresponding information may be found at the end of this issue. So, enjoy in flipping …

___________________________________________________________________________________________________________ A Word from the Editor–in–Chief

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________ A Word from the Editor–in–Chief

Source: http://de.123rf.com; Copyright: Bruce Rolff

___________________________________________________________________________________________________________

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Energy, Safety and Profitability – The Thinking Behind by Jan Blomgren

REMOO–2015: Overview of the 5th International Conference & Workshop by Ana M. Lazarevska

Simon Pezzutto Analysis of the Space Heating and Cooling Market in Europe

Editorial

CONTENT

Shpetim Lajqi, Naser Lajqi, Beqir Hamidi Design and Construction of Mini Hydropower Plant with Propeller Turbine

Filip V. Stojkovski, Marija Chekerovska, Risto V. Filkoski, Valentino Stojkovski Numerical Modelling of a Solar Chimney Power Plant

Martin Felix Pichler, Alexander Arnitz, Markus Brychta, Andreas Heinz, René Rieberer Small Scale PV-Power – On Site Use Maximization through Smart Heat Pump Control

Eugen Bancheş, Ionut Purica Long Term Assessment of Nuclear Technology Penetration using MESSAGE – The Case of Romania

Jarkko Ahokas, Kristiina Söderholm The Role of Nuclear Power in the Future Energy System

Bashir Busahmin, Brij Maini, Hossein Hejazi, Amin Sharifi, Mohammad Tavallali Application of Foamy Mineral Oil Flow under Solution Gas Drive to a Field Crude Oil

Ana M. Lazarevska, Daniela Mladenovska Multi-Criteria Assessment of Natural Gas Supply Options – The Macedonian Case

Drenusha Krasniqi, Risto Filkoski, Fejzullah Krasniqi An Approach towards Thermal Power Plants Efficiency Analysis by Use of Exergy Method Kristina Šarović, Arben Abrashi, Damir Božičević Modelling and Analysis of Thermal Energy Storage Implementation in the District Heating Systems of the City of Zagreb

Giuseppe Conte, Luca Paciello, David Scaradozzi, Anna Maria Perdon A Software Tool to support Design and Upgrade of Energy Production and Storage Systems

About the Journal

e10

Instructions for Authors

e12

Advertisements

The Journal

Authors‘ Papers

(Short Summary of the PhD Thesis at the University of Natural Resources and Life Sciences, Vienna, Austria, 2014)

___________________________________________________________________________________________________________ Content

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Editorial

.e1

Energy — A Truly Multi-Faceted Field by Jan Blomgren

.e3

REMOO–2015 Overview of the 5th International Conference & Workshop “TECHNOLOGICAL, MODELLING AND EXPERIMENTAL ACHIEVEMENTS IN ENERGY GENERATION SYSTEMS” by Ana M. Lazarevska

___________________________________________________________________________________________________________

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

ENERGY — A TRULY MULTI–FACETED FIELD If you want simple and quick answers, look in another direction. If you are attracted to multi-faceted, notoriously challenging issues, energy provides plenty of mouth-watering opportunities. Contemporary energy involves technology, economy, ethics, environment, national security and a whole range of other aspects. More than so, we also need to take human values into account, resulting in that also when we agree on the facts, we can still draw different conclusions about the way forward, simply because we give different weight to these various factors. Does this mean all problems are impossible to solve, and we should just give up? The right-out answer is NO. We need, however, to address the challenges more intelligently than looking for binary answers of the type right or wrong, or yes or no. The first consequence of the complexity of energy issues is that reasonable solutions more or less always are systems, not single solutions. Essentially all countries on Earth has a production mix. Very few, if any, rely on a single production technology. Finding a mix that optimizes all the factors mentioned above and many more is then either a political problem, a market challenge, or both (or none, could a cynic add…). The systemic nature of solutions in the energy sector naturally calls for system modelling to support decision-making on all levels; on a global scale, nationally, regionally as well as for optimization of energy supply and use in a single house. This was a common underlying theme in the recent REMOO conference held September 23-24 in Budva, Montenegro, with the title ”Technological, Modelling and Experimental Achievements in Energy Generation Systems”. In fact, it turned out that the conference attracted an even wider scope of contributions, not limiting to production but where consumption also was part of the agenda. Some of the contributions are presented in the present issue of the International Journal of Contemporary Energy. A notable common theme in all these papers is improved efficiency, although they might seem to be on disparate topics at first glance. On the renewable energy side, Lajqi, Lajqi and Hamidi presents design of a mini hydro power plant. Stojkovski, Chekerovska, Filkoski and Stojkovski have performed modelling of a solar chimney for energy production, whereas Pichler, Arnitz, Brychta, Heinz and Rieberer have studied how to improve the efficiency in solar photovoltaics joining forces with a heat pump. Bancheş and Purica presents modelling useful for strategy considerations concerning future nuclear power production in Romania, and Ahokas and Söderholm contributes a similar study concerning the Nordic region. In the conventional energy realm, Busahmin, Maini, Hejazi, Sharifi and Tavallali have investigated the potential for improved extraction from oil wells, and Lazarevska and Mladenovska have studied optimization of natural gas supply. Krasniqi, Filkoski and Krasniqi have performed an exergy analysis aiming at improved performance of a thermal power plant, whereas the paper by Šarović, Abrashi and Božičević deals with improved efficiency by storing thermal energy in a district heating system. Finally, Conte, Paciello, Scaradozzi and Perdon presents optimized energy use in a single house. This sums up the contributions in this issue of the International Journal of Contemporary Energy. You can look forward to most interesting reading.

Jan Blomgren Associate Editor ___________________________________________________________________________________________________________ “Energy – A Multi–Faceted Field”

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Jan Blomgren is CEO and founder of INBEx (Institute of Nuclear Business Excellence), providing independent nuclear executive advice and business leadership training globally. The INBEx team comprises over 20 former CEOs, Director Generals and similar. He was the youngest professor ever in Sweden in nuclear physics, holding the chair in applied nuclear physics at Uppsala University. His research was focused on neutron-induced nuclear reactions, an area in which he has published over 200 papers in refereed international journals and conference proceedings. When plans to build new nuclear power in Sweden were initiated, he was recruited to Vattenfall, one of the largest nuclear power operators in Europe. At Vattenfall, he was responsible for planning the competence development needed for nuclear new-build, as well as coordinating training for nuclear power plant personnel. In addition, he was Director of the Swedish Nuclear Technology Centre, which is the coordination organization for nuclear research and education involving universities, industry and the regulator. He was involved in the creation of ENEN, the European Nuclear Education Network, in which essentially all European universities in nuclear engineering collaborate. Moreover, he has recently established a large collaboration with France on research and education. Finally, he is the father of several industry-sponsored university programs, as well as having started a number of nuclear business training programs in industry. Jan Blomgren is alone in Europe to have upheld high-ranked positions both at university and nuclear industry. He is frequently invited speaker at conferences, and was recently invited as expert advisor to the French Senate.

___________________________________________________________________________________________________________ “Energy – A Multi–Faceted Field”

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

REMOO–2015

OVERVIEW OF THE 5TH INTERNATIONAL CONFERENCE & WORKSHOP

“TECHNOLOGICAL, MODELLING AND EXPERIMENTAL ACHIEVEMENTS IN ENERGY GENERATION SYSTEMS” In continuation of a successful series of International Regional Energy Mix and Outlook Options (REMOO) Conferences and Workshops held for the first time in 2010 in Belgrade, Serbia covering the topic of the “Energy Mix and Outlook Options for Serbia and the Region”; then in 2012 in Zagreb, Croatia focusing on “Regional Co-Operation with Focus on CO2-Free Energy Options” 1 and in Ljubljana, Slovenia, 2012 addressing the “Nuclear Energy Development and New Build Prospects”2; followed by the ones in 2013 again in Belgrade, Serbia covering the topic of “Modern Management and Safety Culture for the Sustainable Energy Mix”3 and in Ljubljana, Slovenia, 2014 focussing on “Energy Infrastructure Development”4; in 2015, the founder of REMOO – conferences, RENECON International from Germany marked a small but significant and rememberable jubilee by organizing REMOO–20155: The 5th International Conference & Workshop, held on 23rd and 24th of September 2015, in Budva, Montenegro. Since, the greatest source of energy for the future is to commend its utilization in a more prudent, responsible and efficient manner, not forgetting we owe the future generations its current sustainable use by meeting “…the needs of the present without compromising the ability of future generations to meet their own needs…”6, the main focus of REMOO–2015 was set on “The Technological, Modelling and Experimental Achievements in Energy Generation Systems”. In that sense and in order to commend this 5th jubilee, the Editorial Board of the International Journal of Contemporary Energy (IJCE) decided to commit this issue of IJCE to a series of selected key-note speeches and papers presented during REMOO–2015. The objective of the REMOO Conferences and Workshops is to tackle all aspects of energy production, transportation and usage at a regional level, whereas “the region” originally was localized to countries that emerged after the disintegration of Yugoslavia. However, as the time went by, and in concordance to the integrative idea promoted throughout all European countries, this initial concept of “the region”, spread not only to the immediate neighbouring countries, but as well countries well beyond this initial “region”. In line with a five-year history of an established positive practice, as well this time, a set of carefully selected distinguished lecturers – all leading experts in their fields of research – were invited to prepare keynote lectures providing up-to-date insights relating to modelling and experimental achievements in the energy generation systems and serving as a corresponding introduction to each conference section.

LEITMOTIF FOR THE CONFERENCE THEME The objective of the 5th International Conference and Workshop REMOO-2015 on “The Technological, Modelling and Experimental Achievements in Energy Generation Systems” was to provide an international platform and forum for discussing important issues affecting further development of energy and electricity generation systems, with a focus on 1

http://www.remoo.eu/html/2011.html http://www.remoo.eu/html/2012.html 3 http://www.remoo.eu/html/2013.html 4 http://www.remoo.eu/html/2014.html 5 http://www.remoo.eu/html/2015.html 6 https://www.iisd.org/sd/ and http://www.worldbank.org/depweb/english/sd.html 2

___________________________________________________________________________________________________________ “REMOO-2015: Overview of the 5th International Conference & Workshop”

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

present technological, modelling and experimental achievements including their future expectations and outlook. Through invited keynote lecturers, the leading experts provided up-to-date information and the newest insights that helped delegates and their institutions to successfully navigate through the complex and changing workplace of different energy technologies as well as of modelling and experimental techniques being important and indispensable supportive tools for their further development. As highlighted by the first conference key-lecturer Jan Blomgren, “…Energy is a notoriously difficult area to reach consensus about … ” since it involves a whole range of aspects starting from technology, economy, covering environmental protection and security, interweaving social issues such as ethics, corruption, national security etc. Such a variety of attributes, merits and points of view, makes the process of energy management in real environments – starting from generation, ending with end-users consumption –a highly unpredictable and challenging one. Moreover, different cultures view challenges and corresponding priorities in connection to energy management differently. What the energy sector is in urgent need is a holistic and integrated approach that shall enable relevant stakeholders in the energy sector to tackle the rising and more complex issues deriving from the continuous rise of the energy demand. Thus, with the carefully selected key-note lectures and dedicated participants, the 5th International Conference and Workshop REMOO–2015, strongly contributed to this serious and demanding task to provide insight in the technological, modelling and experimental achievements in energy generation systems, on one side, and to propose selected contemporary solutions towards more sustainable utilization of the energy in any form it is required to be used, on the other. Therefore, we invite you to continue this odyssey towards engaging science and engineering for reliable energy by taking part in the 6th REMOO Conference and Workshop, to be held 18-20 May, 2016 in Budva, Montenegro.

TOPICS THAT HAVE BEEN DISCUSSED ON REMOO–2015 The Topics and Themes encompassed and discussed in the frames of REMOO–2015, describing the structure of the submitted papers and presentations are listed below. As well this time, although some submissions did not strictly relate ___________________________________________________________________________________________________________ “REMOO-2015: Overview of the 5th International Conference & Workshop”

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

to those topics, by addressing the original REMOO objective, i.e. discussions and analyses regarding all aspects of energy production, transportation and usage, they earned their place and were included in the frames of REMOO–2015. Topics and Themes that have been discussed within REMOO–2015 are as follows: T1 POWER PLANTS ENGINEERING with the corresponding sub-topics: Operations & Safety, Maintenance; Load Change Flexibility; Efficiency & Reducing Energy Losses; Balance of Plant (BoP) & Instrumentation and Controls (I&C): Modifications and Upgrades: Power Up-Rate & Life-Time Extension: Technological Innovations & Innovative Designs; New Technologies and their Applications. T2 COMMODITIES & ENERGY MIX with the corresponding sub-topics: Fuel Cycles and Commodity Mix; Energy Density and Environmental Load; Carbon Emissions T3 ELECTRICAL GRIDS with the corresponding sub-topics: Upgrades; Transmission and Distribution; Smart Grids/Cities T4 MATERIALS & STRUCTURAL ANALYSES with the corresponding sub-topics: Durability, Ageing & Life Prediction Methodologies; Structural Integrity; Advanced & Smart Materials; Design and Monitoring for Seismic, Dynamic and Extreme Events. T5 MODELLING & SIMULATION with the corresponding sub-topics: Heat Transfer & Thermal Hydraulics; MultiPhase &Multi-Fluid Flows; Computational Fluid Dynamics (CFD); Mechanical Interactions of Fluid and Structures; Codes and Numerics &their Developments and/or Improvements; Coupled Codes &used Methods; Component(s) &Plant System Codes. T6 EXPERIMENTAL SUPPORT with the corresponding sub-topics: Experiments and Databases for Assessment and Codes Validation; Experimental Methods in Complex Flows; Innovative Experimental Techniques; Imaging, Processing and Analysis. T7 DATA ANALYSIS with the corresponding sub-topics: Accuracy and Uncertainty Analysis; Data Post-Processing; Visualisation and Animation; Virtual Reality and Graphical Simulations. T8 REGULATIONS / LEGISLATIONS / COLLABORATION with the corresponding sub-topics: Regulatory Frameworks & Common Issues; Standards & Licensing; International Collaboration & Know-How / Know-Why Transfer.

PARTICIPATION AND SUBMISSIONS To mark this momentous 5th jubilee, the organizers of REMOO–2015 invited two distinguished speakers at the opening of the conference; eight key-note lecturers, and facilitated sharing contributions of more than 100 authors of 43 papers and posters from all around the World and active involvement of 71 registered participants, originating from 26 countries. Thus, the REMO0–2015 statistics described in numbers is as follows: • 2 speakers at the opening; • 8 invited key-note lecturers from 8 countries from around Europe as an overture to the 8 corresponding sessions covering the eight selected conference topics T1 through T8; • 51 works have been submitted, out of which 8 were rejected and the remaining 43 were accepted; • A total of 43 contributed works, i.e. 9 under topic T1, 7 under topic T2, 5 under topic T3; 1 under topic T4, 13 under topic T5, 2 under topic T6, 3 under topic T7 and 2 under topic T8; • Out of them 15 were authors’ posters from 12 countries around Europe and beyond, while the remaining 28 were presented orally by their authors originating from 19 countries around the World; • A total of 71 registered participants. REMOO–2015: The 5th International Conference and Workshop on the “Technological, Modelling and Experimental Achievements in Energy Generation Systems” was solemnly opened by Zoran Stosic from Germany, the REMOO Conference Chairman and Founder, Managing Director of RENECON International followed by the speech of Sreten Skuletic, professor at the Faculty of Electrical Engineering, University of Montenegro. The latter, again expressed the ___________________________________________________________________________________________________________ “REMOO-2015: Overview of the 5th International Conference & Workshop”

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

firm and continuous support the academia and the research institutes provide to facilitate activities around and deriving from REMOO and RENECON International. The number and the quality of invited key-note lectures again prove the added value REMOO conferences hold and why REMOO also qualifies for the category of workshop. Namely, key-note lecturers are selected to provide important information to the conference participants and to expose their specific knowledge and experience in the subject. Nevertheless, it is never a “one-way street!” Conference participants are invited and expected to actively take part in the discussions and to share their specific knowledge, important facts, views and experience relating to the presented lectures leading to a win-win synergy between the lecturer and the conference auditorium. Some highlights from the invited lectures follow.

___________________________________________________________________________________________________________ “REMOO-2015: Overview of the 5th International Conference & Workshop”

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

With his exceptional key-note lecture, on the issue of “Energy, Safety and Profitability – The Thinking Behind”, Jan Blomgren (Sweden), from the Uppsala University opened the first conference session. By confronting the traditional approach in assessing investments which utilizes cost-benefit analysis combined with risk-informed analysis, versus the newly proposed approach which integrates technology challenges and management from a philosophical point of view, Blomgren provided an insight on how to quantify consequences a single accident can cause towards threatening the existence of entire companies should incorrect risk assessment is performed. A superior contribution towards the concept of sustainability and how to measure it in terms of assessing optimal energy-mix for a country, region or globally was provided in the systematic and thorough review of Zoran Stosic (Germany), the Managing Director of RENECON International, in his key-note lecture “A Selection of Criteria for Sustainability of Energy Mix”. Nikola Cavlina (Croatia), professor at FER of the University of Zagreb focused on the comparison among energy produced in newly built coal-fired, gas-fired, solar, wind and in nuclear power plants. In his excellent speech “Energy Options and Electricity Generation Costs” he presented results from a comparative competitiveness assessment based on the Levelized Unit of Electricity Cost (LUEC) performed by means of the Monte Carlo method. Zalan Bacs (Hungary) Regional Director of ROSATOM gave the presentation instead of Jukka Laaksonen (Russian Federation), who is the Advisor to Director General of ROSATOM Overseas. He provided a realistic and accurate portrait of the activities encompassed within the Nuclear Power Plants construction projects. His notable lecture titled “An Overview and Update on Worldwide Nuclear Power Plant Construction Today” was based on information contained in direct contracts with representatives of nuclear facilities, vendors, or regulators considered as relevant stakeholders eligible to provide reliable information on the status of a respective project. My personal favourite was the lecture of Milovan Peric (Germany / UK), the Managing Director of CoMeT and Senior Corporate Consultant of CD–adapco, titled “Continuum Mechanics Simulation in the Development of Power Technology”. In his noteworthy lecture, Peric presented the rising approach of Computational Continuum Mechanics (CCM), which attempts to incorporate the fluid-interaction in simulations of solid mechanics. This integrated approach of optimal use of theoretical, numerical and experimental analysis provides new developments in modelling and numerical simulations relating to energy generation and consumption with a higher goal to bring better products to market in a shorter period of time.

___________________________________________________________________________________________________________ “REMOO-2015: Overview of the 5th International Conference & Workshop”

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

In his outstanding lecture on “Experimental Support for Safer and More Reliable Power Generation”, Leon Cizelj (Slovenia), who is the Head of Reactor Engineering at the “Josef Stefan” Institute and professor at the University of Ljubljana, explained the challenges the traditional experimental techniques face while studying heat transfer and fluid flow phenomena occurring at frequencies invisible for the conventional techniques. On the other side, adopting a combination of the Particle Image Velocimetry (PIV) and advanced imaging and processing for assessment of the temperature profiles provide significantly more effective synergy. Addressing the new trends of urbanization which automatically imply a rapid increase in the energy supply and consumption in the mega-cities, Igor Kuzle (Croatia), professor at FER of the University of Zagreb and the IEEE Region-8 Vice Chair and Member of CIGRE, provided a sound elaboration on the challenges and issues with reference to the “Smart Grid and Smart City – State-of-the-Art and Future Trends in Development”. This significant and comprehensive lecture focused on the recent and expected advancements in the smart grid with the final goal to quantify the ability of the smart-grid components to provide the necessary flexibility. Last but not least, was the astonishing lecture of Hakan Johansson (Sweden), the Senior Executive, Smart Grid and Consulting at ABB, who again proved that exactly we as humans are the ones who are to take responsibility both in enjoyable and in challenging times. He has practically shown that only a “smart mind” can produce and deliver a “smart system”. Although challenged by the power break-down during his lecture titled “Implementation of Smart Grid Systems & Services to Optimise System Performance” he managed to vividly convey all points of his speech. Poster part of the conference was another important opportunity to exchange views, experience, knowledge and information. The significant number of posters, 15 in total, were more focused on specific problems described by detailed technical information, whereby maintained was a high graphics quality. Significant number of the papers in this issue of International Journal of Contemporary Energy is based on expanded versions of posters presented during the REMOO–2015 conference and workshop.

(All images reprinted with permission from RENECON International) Ana M. Lazarevska Associate Editor

Ana M. Lazarevska (BSc, 1994: MSc, 2001, in mechanical engineering; DSc, technical sciences, 2008, at the University of Ss Cyril & Methodius (UKIM), Skopje, Macedonia) is an associate professor at UKIM, Skopje (Faculty of Mechanical Engineering, Institute for Hydraulic Engineering & Automation), Macedonia. Her areas of main interest are environmental protection, climate change, sustainability and sustainable development, decision making in particular Multicriteria Decision Making (MCDM), energy management, resource efficient and cleaner production (RECP), risk assessment. Further interests are in the fields of air pollution modelling, theory of system’s control, and persons with limited abilities (or disabilities). DSc, technical sciences – environmental protection, Faculty of Mechanical Engineering, UKIM, Skopje, Macedonia, 2008. MSc, mechanical engineering, Faculty of Mechanical Engineering, UKIM, Skopje, Macedonia, 2001. BSc, mechanical engineering, Faculty of Mechanical Engineering, UKIM, Skopje, Macedonia, 1994. Current applications and research: Multicriteria Decision Making Theory and its implementation on various topics, like contributing to mitigating climate change, primarily through implementing the concepts of SD, Energy Management, Resource Efficient & Cleaner Production and their applications in the sectors energy (generation/consumption side) and integrated (solid) waste management; Risk Identification, Modelling & Assessment; etc.

___________________________________________________________________________________________________________ “REMOO-2015: Overview of the 5th International Conference & Workshop”

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Summaries / Reviews

Simon Pezzutto

Analysis of the Space Heating and Cooling Market in Europe (Short Summary of the PhD Thesis at the University of Natural Resources and Life Sciences, Vienna, Austria, 2014)

___________________________________________________________________________________________________________

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.20160100

Analysis of the Space Heating and Cooling Market in Europe1 Simon Pezzutto Institute for Renewable Energy, EURAC research Viale Druso 1, 39100 Bolzano, Italy; simon.pezzutto@eurac.edu

There are major knowledge gaps concerning the European space heating and cooling market. While thorough data is available for the space heating and domestic hot water markets over the past two decades, the data regarding space cooling is still inadequate. Similarly, while the data for the residential sector is complete, the services (trade, hotels, offices, education, hospitals, and bars) sectors are virtually unexplored. Significantly more data is available for the EU-15 group than for the EU-13 countries (member states which entered the EU more recently than 1996). Furthermore, not all of the available data is considered reliable, even though the data has been collected solely from trustworthy sources. The data from the space cooling market in the EU-13 member states is particularly questionable. As European citizens continue to become more and more sensitive to energy utilization the demand for energy usage to be accurately estimated becomes more urgent. Data describing the space heating, cooling and domestic hot water markets has been collected, elaborated statistically and combined using bottom-up techniques. Information regarding the space cooling market was also retrieved using a top-down approach, and the results from the two methods were compared. According to both methods, the total space cooling energy utilization is approximately 70 TWh/y. The European Commission reports a significantly lower value, of around 30 TWh/y. Results from the bottom-up approach show a 9:1 ratio between the existing energy consumption and total potential for air-conditioning. That same ratio is approximately 1:1 for space heating. Almost 90% (about 63 out of 70 TWh/y) of the space cooling consumption occurs in the EU-15 states. Correspondingly, the potential of EU-13 member states is enormous; almost four times larger than their current energy

consumption. There is a high potential for growth in the European space cooling market, particularly in residences. Contrarily, there is almost no room for growth in the space heating market. Space cooling is responsible for a significant part of the electricity consumption in European Union households, especially in the service sector. There has been a steady increase in the European specific and total energy consumption for space cooling during recent decades. Both the cooled floor area and the sales volume of space cooling equipment have dramatically increased since 1990. There is a huge difference between the EU-15 and EU13 member states space cooling consumption in residences. The EU-15 countries use approximately 30 kWh/m² y for space cooling, and the EU-13 countries utilize around 10 kWh/m² y. The EU-15 countries use approximately 17 TWh/y while the EU-13 countries consume around 1 TWh/y in households. This is primarily because most residential buildings in the EU-13 countries do not use space cooling equipment. Surprisingly, the EU-13 country with the most space cooling consumption in the residential sector is Cyprus, using around 0.3 TWh/y. It is also highly ranked in the EU-28, tying Portugal for fourth place despite Portugal having about ten times Cyprus’ population. This occurs primarily because more than 74%, a very high percentage for European homes, of floor area in Cyprus’ households is cooled. The elevated specific value for airconditioning purposes in Cyprus contributes to that point as well. The specific space heating consumption values for Europe have been steadily decreasing for the past two decades. The specific space heating consumption values for the EU-13 and EU-15 states are very similar, with

Short summary of the PhD Thesis: Pezzutto, S., “Analysis of the Space Heating and Cooling Market in Europe”, University of Natural Resources and Life Sciences, Vienna, Austria, 2014; http://permalink.obvsg.at/bok/AC10776806?&con_lng=eng

___________________________________________________________________________________________________________ S. Pezzutto: “Analysis of the Space Heating and Cooling Market in Europe”, pp. r1–r3

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

respective weighted averages of 156 kWh/m² y and 160 kWh/m² y. Regarding European space heating, the potential energy consumption and actual space heating consumption are very similar, with a ratio of 1.2:1. Most of the energy consumption in the European Union is used for space heating, which represents 3,169 TWh/y. Within the residential sector, the EU-15 countries consume 1,544 TWh/y more space heating energy than the EU-13 countries. The EU-15 countries use approximately 1,982 TWh/y while the EU-13 countries utilize about 438 TWh/y. There is a significant difference between the specific space heating consumption values per occupant in the EU-15 and EU-13 countries. Each occupant in the EU-15 countries uses approximately 5,457 kWh/y for space heating while the average occupant in the EU-13 countries consumes 3,787 kWh/y. This occurs primarily because of the lower economic power of people in the EU-13 countries. Across the entire European Union, the average space heating consumption per occupant is 4,681 KWh/y. The specific domestic hot water consumption is 26 kWh/m² y for the EU-15 countries and 33 kWh/m² y in the EU-13 countries. This occurs primarily because the climates in the EU-13 countries tend to be colder than the EU-15 countries, and because the boilers in EU-15 countries are more energy efficient. In total, the EU-15 countries consume 409 TWh/y for domestic hot water while the EU-13 countries use approximately 92 TWh/y. This represents a huge difference of 317 TWh/y. A comparison between the potential and actual domestic hot water consumption values shows that there is limited room for growth in the domestic hot water market. The ratio is approximately 1:1, with each value around 501 TWh/y. Similarly to the space heating occupant behaviour data, the domestic hot water occupant behaviour data shows that people in the EU-15 countries use more energy than the people in the EU-13 countries. EU-15 occupants and EU-13 occupants use 1,079 kWh/inhabitant y and 942 kWh/inhabitant y respectively. The energy use patterns of Poland are interestingly similar to those of Germany. The two countries return the highest energy consumption values for the majority of cases, usually representing around 1/3 of each type of energy consumption (space heating, cooling and domestic hot water). The economic strength of a country has a significant impact on their energy consumption, and the Polish and German economies are both the strongest within their group of European Union countries (EU-13 and EU-15 respectively).

Comparisons between the total (combining the service and residential sectors) energy consumption of space heating, cooling and domestic hot water within the European Union show that space heating consumes the most energy, with domestic hot water consumption the second most and space cooling the least. Space heating consumes around 3,169 TWh/y, domestic hot water consumes approximately 501 TWh/y and space cooling consumes about 68 TWh/y. Thus space heating uses around six times as much energy as domestic hot water and about 46 times as much energy as space cooling. A Porter´s five forces analysis indicates that the space cooling market has been fairly steadily rising for about two decades, and that it is currently in a healthy state. In contrast, a number of reports concerning the same topic show the treated sector to be a booming market. Taking into consideration that in the majority of cases air-conditioning is not a commodity of primary necessity and that the space cooling market is suffering significantly due to the economic crisis, this result appears to be trustful. The main driver behind the investigated market is research and development: however, the sector secures only a small part (7%) of the total European Union funding for heating and cooling research. Funding for heating and cooling research, when expressed as a percentage, decreased between the Sixth and Seventh Framework Programmes. In the Sixth Framework Programme it received 8% of the funding, while it got only 7% in the seventh. A fundamental and technical analysis indicates that the European Union financial support for the heating and cooling sector, as a percent of total research funding, will decrease until 2020: to 6%. The latter indicated time frame equals to the end of the recently begun follow up of the Seventh Framework Programme: Horizon 2020. However, the latter finance has been indicated to rise in terms of absolute amount of money. A Multiple-criteria decision analysis shows once more a slightly positive evolvement of the space cooling market and that it will be especially sensitive to: i) Research and development; ii) The economic crisis and iii) Higher comfort standards requested by the European population. The space cooling market forecasts, provided by the aforementioned Porter five forces and Multiple-criteria decision analysis, show both a high dependence on research and development activities and a moderate increase of the European air-conditioning market until 2020. The two carried out economic models – a quantitative and a qualitative one – indicate practically the same result.

___________________________________________________________________________________________________________ S. Pezzutto: “Analysis of the Space Heating and Cooling Market in Europe”, pp. r1–r3

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

The outcomes from the performed investigations show that the space heating market is almost fully saturated, as it is a commodity of primary necessity. In contrast, the space cooling market in Europe is expected to gradually increase in future. The main factors influencing this tendency lie in the fact that:

i) Space cooling is only an asset of primary necessity for specific building types and climates; ii) The customers’ purchasing power reduction and iii) The growth of electricity prices.

___________________________________________________________________________________________________________ S. Pezzutto: “Analysis of the Space Heating and Cooling Market in Europe”, pp. r1–r3

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Authors’ Papers 1

Shpetim Lajqi, Naser Lajqi, Beqir Hamidi

Design and Construction of Mini Hydropower Plant with Propeller Turbine 14

Filip V. Stojkovski, Marija Chekerovska, Risto V. Filkoski, Valentino Stojkovski

Numerical Modelling of a Solar Chimney Power Plant 22

Martin Felix Pichler, Alexander Arnitz, Markus Brychta, Andreas Heinz, René Rieberer

Small Scale PV-Power – On Site Use Maximization through Smart Heat Pump Control 31

Eugen Bancheş, Ionut Purica

Long Term Assessment of Nuclear Technology Penetration using MESSAGE – The Case of Romania 39

Jarkko Ahokas, Kristiina Söderholm

The Role of Nuclear Power in the Future Energy System 47

Bashir Busahmin, Brij Maini, Hossein Hejazi, Amin Sharifi, Mohammad Tavallali

Application of Foamy Mineral Oil Flow under Solution Gas Drive to a Field Crude Oil 54

Ana M. Lazarevska, Daniela Mladenovska

Multi-Criteria Assessment of Natural Gas Supply Options – The Macedonian Case 63

Drenusha Krasniqi, Risto Filkoski, Fejzullah Krasniqi

An Approach towards Thermal Power Plants Efficiency Analysis by Use of Exergy Method 69

Kristina Šarović, Arben Abrashi, Damir Božičević

Modelling and Analysis of Thermal Energy Storage Implementation in the District Heating Systems of the City of Zagreb 77

Giuseppe Conte, Luca Paciello, David Scaradozzi, Anna Maria Perdon

A Software Tool to support Design and Upgrade of Energy Production and Storage Systems ___________________________________________________________________________________________________________

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.20160101

Design and Construction of Mini Hydropower Plant with Propeller Turbine Shpetim Lajqi, Naser Lajqi*, Beqir Hamidi Faculty of Mechanical Engineering, University of Prishtina “Hasan Prishtina” Bregu i Diellit p.n., 10000 Prishtina, Kosovo; naser.lajqi@uni-pr.edu

Abstract

1. Introduction

Nowadays, the hydropower plant is considered as one of the more desirable sources of producing electrical energy due to its environmentally-friendly nature and extensive potential available throughout the world. On the other hand the hydropower plant allows the autonomous production of quantities of electrical energy capable of meeting the requirements of individual users starting from water resources which would otherwise be wasted.

Energy is one of the more fundamental elements in our universe. It is based for survival and inevitably for the development of activities for promoting education, health, transportation and infrastructure for attaining a reasonable standard of living and is also a critical factor for economic development and employment [1, 2, and 3].

Based on the continuous requirement for renewable energy, a mini hydropower plant by using a propeller turbine is discussed due to its simple structure and easy production. The key parameters are studied for designing a hydropower plant, like water head, water flow-rate and turbine speed. The propeller turbine is considered for working under the best working conditions. During the detailed design of a hydropower plant some parameters are known and give us some indication about the geometry of the turbine and this is the starting point. The indication parameters are: turbine power, runner diameter, turbine speed, turbine housing design, draft tube, etc. A detailed design and construction of a mini hydropower plant was done for a recreational center in the village Sllakovc, Vushtria, Republic of Kosovo, financed by European Union. The hydropower plant consists of water intake, penstock, hydro turbine, control system, and a hydropower house. A synchronous generator is connected directly to the turbine which converts the hydraulic energy into electrical energy.

Keywords:

Design; Hydropower plant; Propeller turbine; Design parameters; Efficiency

Article history:

Received: 24 December 2015 Revised: 29 January 2016 Accepted: 29 January 2016

Over the past decade, as well as now, problems with energy supply are present everywhere in the world. Problems coming from different sources like the oil crisis, climate change, technical capacity limits, continuously growing demands and restrictions on the whole sale. These difficulties are continuously growing, which represent an urgent need for using alternative sources which would enable assurances of their solutions. One of these alternative sources is to generate electricity as close as possible to consumption demands by using renewable sources that do not cause environmental pollution. The renewable sources are considered to be energies produced by wind, solar, geothermal, hydropower, etc., [2]. Hydropower energy is part of a renewable energy resource which comes from the motion of water through hydropower device in order to generate electricity. When the water is flowing by the force of gravity, its potential energy converts into kinetic energy. This kinetic energy of the flowing water turns blades or vanes in hydro turbines, and then energy is changed to mechanical energy. The turbine turns the generator rotor which converts this mechanical energy into electrical energy [4]. Many other components may be used in a hydropower system to produce energy. It is perhaps the oldest renewable energy technique. Depending on the capacity of water sources and flow of water by the force of gravity, hydropower plants may be large, small, mini, and micro. The larger hydropower plants would supply many consumers with electricity,

___________________________________________________________________________________________________________ Sh. Lajqi, N. Lajqi, B. Hamidi: “Design and Construction of Mini Hydropower Plant with Propeller Turbine”, pp. 1–13

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

while small mini and micro hydropower plants operate individually for their own energy needs or to sell power to utilities. These hydropower plants can provide effective cost of energy to remote rural communities, recreational centers and restaurants which have suitable water sources. Actually, several schemes for small, mini and micro hydropower plant have been proposed, designed and successfully implemented, including Pelton, Turgo, Francis, Kaplan, propeller and cross flow hydro turbines. The Pelton, Turgo and Francis turbines work with high and medium water heads with less water flow, while Kaplan, propeller and cross flow turbines works with lower heads and larger flow rates. Nowadays, the propeller turbine is gaining in popularity because it works with very low heads and larger water flow-rates. Many places have good potential with low water heads from 2 to 10 [m] but only a few have been developed because there has been a lack of appropriate turbine design. Typically components of a propeller hydro turbine are shown in Figure 1. In Kosovo, electricity is generated mainly by coal power plants and only around 5% of consumption comes from hydropower plants. There are some summer houses, recreation centers, restaurants, etc., in the mountains and hills where public electrification is not yet available but water sources exist. Therefore the availabilities of water sources are a main factor in electricity generation from hydro sources.

This paper describes a design and process of construction of a mini hydropower plant. The design of the mini hydropower plant was done by taking into account a lot of requirements during study, like the designing of the turbine and hydropower plant. This hydropower plant was installed at a recreational center in the village of Sllakovc, Vushtria, Republic of Kosovo and financed by the European Union. The hydropower plant consists of a water intake, penstock, hydro turbine, control system, and hydropower house.

2. Design process of hydropower plant The design process of the hydropower plant, especially for a mini power system with propeller turbine, involves the following steps.

2.1. Measuring of the site data For determining power generation, there are two major parameters: water flow- rate and water fall head. If these parameters are available then it involves calculations and measuring the net head of the hydropower plant. If the actual flow-rate is lower than the turbine design flow, the turbine will generate very little power. In view of this, it is very important to determine as precisely as possible the water flow-rate and water fall head.

4 2

Legend: 7 6

1. Generator 5. Guide and runner vane 2. Connection piece 6. Rubber joint 3. Turbine housing 7. Butterfly valve 4. Support steel plate 8. Draft Tube

Figure 1: Design of a propeller hydro turbine with generator

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Dynamic pressure curve Static pressure curve

Penstock

Gross head, Hg

Net head, Hn

Working pressure curve

Surge pressure, Hdyn

Hlos

∆H

Water intake

Power house

Figure 2: Principal scheme of hydro-electric power system a. Estimation of the water flow- rate

b. Estimation of the net head

The water flow rate (Q) can be estimated in different ways but a more suitable method could be measuring the river water flow velocity and river cross-section areas at the same measuring place by employing the following expression:

Water fall head is also called water pressure which is created by the difference in elevation between the intake of the level of water and the hydro turbine power house, Figure 2.

Q = Ar ⋅ vr [m3/s]

Water head can be measured as vertical distance or as pressure. Regardless of the size of your stream, a higher head will produce greater pressure, and therefore higher power output at the hydro turbine.

(1)

where: Ar [m2] – river cross-section, vr [m/s] – river water-flow velocity. As is known, the river levels change throughout the seasons, so it is important to measure water flow-rates at various intervals of the year. The best estimation is considered when available data for many years observation are at your disposal. If this is not possible, attempts can be made to determine various annual river flow-rates by discussing with a neighbor, or finding hydrological flow data for your river or a nearby larger river. In most cases these data should be provided from the National Hydrological Institute which is responsible for calculating water flow-rate, quality of water, rainfall, etc. Using of the all river’s water for a hydropower plant is not allowed because certain amounts of water need to flow across the river bed as fish, birds, plants, and other living things rely on your river for survival. There are more known methods for measuring water flow-rate like: measuring by flow—container, float, and weir [5].

An altimeter can be useful in estimating the head for preliminary site evaluation but should not be used for the final measurement. GPS altimeters are often used even though they provide less accurate measuring. Topographic maps can also be used for providing a very rough idea of the vertical drop along a section of a river’s sources. The best way when measuring water head can be done by employing modern electronic digital levels. Net head (Hn) is calculated by employing the following expression:

H n = H g − H los [m]

(2)

where: Hg [m] – The gross head; the vertical distance between water surface level at the intake of water and at the turbine site, Hlos [m] – head losses due to the open channel, trash rack, intake, penstock and gate or value. These losses were approximately in some cases around 6% of gross head [2],

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Dhub

2.2. Calculation of output power generation from hydro turbine The output power generation from the hydro turbine (Pgen), can be estimated by the following expression:

Pgen = ρ ⋅ g ⋅ H n ⋅ Q ⋅η t ⋅η gen [W]

(3)

where:

ρ = 1000 [kg/m ] – water density, – gravity acceleration,

ηt [%]

– turbine efficiency,

ηgen [%]

– generator efficiency.

a turbine with low efficiency. Where the head is low and the flow high, it might be a good idea to design for parallel turbines, each operating on part of the total flow. Otherwise it is necessary to choose a lower speed for the turbine, which will result in a larger physical size. Two smaller turbines running at higher speed are LESS costly than one large turbine.

2.3. Calculation of the hydro turbine specific speed The specific speed gives an indication of the geometry of the turbine and it is the starting point for detailed design. There are many different ways for determining the specific speed (Ns) of a hydro turbine. For our case, the following expression was used:

Ns =

N⋅ Q Hn

3/ 4

Drunner Figure 3: Design of propeller runner performed in Autodesk Inventor 3D

g = 9.81 [m/s2]

Hhub

∆H, Hdyn [m] – surge pressure, dynamic pressure appeared during emergency stop of hydro turbine.

(4)

where: N [rpm] – turbine’s rotation speed. The choices of the turbine rotation speed depend on the speed of the generator and the type of drive used. Often it is possible to use a direct drive, with the turbine runner attached to the end of an extended generator shaft. On the other hand, using a single stage belt drive allows for the possibility of changing the turbine operating speed. This gives more flexibility in the turbine design and when matching to site conditions. According to Simpson and Williams the expected range of specific speed for propeller turbine values is 70 < Ns < 300 [6, 7]. If the specific speed is Ns < 70, then you should look at other alternative type of turbine – e.g. cross-flow (Mitchell-Banki), pump as turbine or turgo turbine. Other criteria during designing of turbine should avoid specific speed Ns > 300 because of this will tend to have

2.4. Calculation of the diameter of the turbine runner and hub Determining of diameter of the runner (Drunner) can be done in different ways. One of the more popular expressions for calculating of the diameter of a propeller turbine runner is:

Drunner = 84.5 ⋅ (0.79 + 1.602 ⋅ N s ) ⋅

Hn [m] (5) 60 ⋅ N

The hub diameter of runner (Dhub) can be estimated by the following expression:

 0.0951   ⋅ Drunner [m] Dhub =  0.25 + N s  

(6)

Figure 3 shows the design of a propeller runner by introducing the diameter of runner & hub and hub height.

3. Design of a mini hydropower plant in Kosovo The of a mini hydropower plant was made by several visits during 2013 to the recreational center “Trofta e

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Estimated water flow rate: Q = Ar ⋅vr [m3/s]

Float Float travelling, L = 10 [m]

Water flow velocity, vr [m/s]

Cross section area, Ar = b ⋅ h [m2] b

Figure 4: Measuring of water flow-rate by the floating approach

Lumit”, Village Sllakoc, Vushtria, Republic of Kosovo to gather important information regarding water flowrate, water fall height, owner of property, form of investment, etc. The following presented the approach for determining water flow-rate and water head.

3.1. Determining of water flow-rate and gross water head The water flow-rate is estimated by measuring the river’s water flow velocity and river cross-section areas in the place where it is planned building of the water intake, Figure 4. The river flow changes throughout the seasons, so it is important to measure water flow-rate at various intervals of the year. In most of the cases this data are provided from The National Hydrological Institute. Unfortunately, this kind of data was not available; therefore it was necessary to develop the measuring approach by floating. The best estimation of water flow-rate is considered when measuring was done covering many years observations of river but this requires a long time to do. In our case, measuring was done once per month during 2013. The critical flow was during three months, like July, August and September 2015 where the levels of water flow were very low and water was used only for ecology and irrigation purposes. Figure 5 presents the water balance of the Sllakoc River as well as the availability of water for the generation of electricity from water. From the graphs can be seen the available water for generation of energy is Qavailable = 0.165 [m3/s] for nine months of the other months the hydropower plant should be stopped. From the graph

can be seen, the river water for generation of energy is available only for nine months (Qvailable = 0.165 [m3/s]), while for other months (July, August and September) is used for irrigation of agriculture’s land, therefore the hydropower plant should be stopped. For determining the sustainability curve of the annual production of electricity by water, by employing data present in Figure 5 and considering ecological flow and the irrigation period, it can be concluded that the mini hydropower plant will be stopped for around 90 days therefore water will flow to its bed and could be used by the community for agricultural irrigation. Despite the use of water for power generation and other needs for communities, determined ecological flow must be respected. There are different approaches for determining the minimum ecological approach which can be used. Figure 6 shows how much water is available for generating energy and the sustainability of the River Sllakoc vs days of 2013 year.

3.2. Determining of penstock diameter for the hydropower plant The penstock (pipeline) diameter depends on these factors: -

Losses in the penstock as a result of friction between the water flow in the penstock and the inside walls of the penstock,

The thickness of the walls to cover static and dynamic pressure, and

The cost and installation of the penstock.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

0.60

Available flow

Sllakoc river water flow rate, Q [m3/s]

0.55

Ecologic flow

0.50

Water for irrigation

0.45 0.40 0.35

Qavailable = 0.165 [m3/s] – 9 month

0.30 0.25 0.20 0.15 0.10

December

November

October

September

August

July

June

May

April

March

December

January

0.00

February

0.05

Water flow rate vs months of 2013 year Figure 5: Water balance of Sllakoc River during 2013

0.60

Available flow

0.55

Ecologic flow

0.50

Water for irrigation

Water flow rate, Q [m3/s]

0.45 0.40

Available water for generation energy

0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00

122

152

183

213

243

274

304

335

365

Sustainability curve for water source for observed day of 2013 year Figure 6: Sustainability of Sllakoc River observed during 2013 year ___________________________________________________________________________________________________________ Sh. Lajqi, N. Lajqi, B. Hamidi: “Design and Construction of Mini Hydropower Plant with Propeller Turbine”, pp. 1–13

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Table 1: Water flow velocities for different penstock diameters and water flow-rate Q = 0.165 [m3/s]

Internal penstock diameter, D [mm] Water flow velocity, v [m/s]

200 5.25

250 3.36

300 2.33

350 1.71

400 1.31

450 1.04

Table 2: Coefficient of friction depending on the Reynolds Number

If Re ≤ 2000 flow is Laminar

If 2000 < Re < 4000 flow is transition zone

If Re ≥ 4000 flow is turbulent

  R  λ = 1.8 ⋅ Log e    7  

64 λ= Re As a result of reducing of penstock diameter for a certain amount of water flow, velocity of water in the penstock increases, and consequently the energy losses increase. The greater the energy losses mean the smaller the energy generation. On the other hand, by increasing the penstock diameter, the energy losses will decrease then the energy generated is greater. The cost of the penstock drastically increases when increasing the penstock diameter. So we are pushed to have balance between energy loss, pipe diameter, and material and wall thickness of the pipe. For the initial design, the water flow speed can be obtained from v = 1 ... 2 [m/s]. The internal diameter of the penstock (D) can be determined by the following expression:

−2

Longitudinal losses Several authors have developed different expressions for determining the longitudinal losses in penstock but as a universal expression which is valid for different penstock diameter is the Darcy-Weisbach expression with Colebrook coefficient for friction. The longitudinal losses are determined by the following expression:

Hlong =

λ L v2

⋅ ⋅ [m] 2 D g

(9)

where:

– friction coefficient of water with penstock (Colebrook coefficient),

L [m] – length of penstock.

4⋅Q [m] π ⋅v

(7)

Table 1 shows the estimated water speeds through the penstock for different tube diameters and average flow rate Q = 0.165 [m3/s]. According to the calculations it attempted to select the optimal diameter. Based on the recommendations of the experts, the water flow velocity can be obtained from v = 1 ... 2 [m/s], the diameters of the penstock 400 or 450 will be eligible for further calculations.

3.3. Determining of energy losses in penstock The total energy losses that arise in the pipeline can be: longitudinal (Hlong) and local losses (Hloc) and is written as follow:

H los = H long + H loc [m]

(8)

The coefficient of friction (λ) depends on the value of the Reynolds Number (Re), and the type of flow. The coefficient of friction is determined by the expression shown in Table 2. Local losses Expression for determining local losses is as follow:

k v2 H loc = ⋅ [m] 2 g

(10)

where: k – coefficient of local losses through bends, fittings, etc.

3.4. Determining of net head and output bends, fittings, etc. Finally, the net head of water is determination by the following expression:

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Hn = H g − (Hlong + Hloc ) = 2  λ L k  v [m] Hg −  ⋅ +   2 D 2 g

(11)

In order for the energy losses to be as small as possible, by decision the penstock diameter should be Ø 450 [mm] and then provide 3.88% energy losses as well as create net head by Hn = 9.61 [m].

For water flow rate Q = 0.165 [m3/s] and gross head Hg = 10 [m], total energy losses (Hlos) in penstock and net head for different diameter of pipe is presented in Figure 7.

By assuming the efficiency of the propeller, the hydro turbine should be ηt = 80% and efficiency of generator ηgen = 85%, then the expectation output generation power as function of penstock diameter is shown in Figure 8.

From curves presented in Figure 7 is shown that, with increasing diameter of penstock, the energy losses in penstock decrease while net head increase too.

From diagram shown in Figure 7 it is observed that, output power from generator will increase by increasing penstock diameter.

Net head and energy losses [m]

Net head

Energy losses in penstock

10 5 0,20

0,25

0,30

0,35

0,40

0,45

(5) (10)

Internal penstock diameter, D [m] Figure 7: Net head and energy losses vs internal penstock diameter

Output power generation, Pgen [kW]

12 9 6 3 0,20

0,25

0,30

0,35

0,40

0,45

(3) (6) (9)

Internal penstock diameter, D [m] Figure 8: Output power generation vs internal penstock diameter ___________________________________________________________________________________________________________ Sh. Lajqi, N. Lajqi, B. Hamidi: “Design and Construction of Mini Hydropower Plant with Propeller Turbine”, pp. 1–13

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Table 3: Indication parameters for propeller turbine Description of indication parameter Water flow rate Gross head Net head Power generation Generator speed Specific speed Turbine runner diameter Hub Diameter

Symbol Q Hg Hnet Pgen N Ns Drunner Dhub

values 0.165 10 9.61 10.58 1500 116.63 232.00 124.00

Unit [m3/s] [m] [m] [kW] [rpm] [-] [mm] [mm]

Figure 9: Design of propeller hydro turbine with 12.5 [kVA] synchronous generator performed in Autodesk Inventor 3D

4. Design of mini hydro turbine By taking in consideration the availability of water flowrate: Q = 0.165 [m3/s] and net head Hn = 9.61 [m], the suitable turbine for working under such conditions is foreseen as the propeller hydro turbine. Some indication parameter for determining geometry of the turbine should be determined.

4.1. Design of propeller turbine The indication parameters for a propeller turbine are: specific speed, turbine runner diameter, hub diameter, turbine housing design, draft tube, etc. In order to determine such indicative parameters some input data is required. The generator rotation speed is selected to be with four poles and rotation speed is: N = 1500 [rpm]. By employing know values elaborated on previously and substations in expressions (4), (5) and (6), then the output results will be as shown in Table 3. Figure 9 presents the propeller hydro-generator performed in Autodesk Inventor 3D. An synchronous generator is connected to the turbine shaft which converts the hydraulic energy into electric energy.

4.1. Design of hydro turbine power house In order for the hydro turbine power house to be useful and suitable as much as possible, there is selected such a design where the mini hydro turbine, control box and discharge of water are in the same place, Figure 10. Figure 11 present front view of the hydro turbine power house for the recreational centre, in the village Sllakovc, Vushtrri.

5. Construction works Performing of the civil works was not difficult due suitable terrain with a minor slope. Work started by removing the old water wheel turbine which had never fulfilled the demand for energy and had a lot of problems regarding maintenance, Figure 12. Reason for removing the old water turbine wheel was done in order to build a power house in the same place. Figure 13 presents the excavation work and discharge place by removing the old water wheel turbine in order to build a hydro turbine power house in the same place.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 10: Design of the hydro turbine power house in 3D view for recreational center, in the village Sllakovc, Vushtrri

Figure 11: Front view of the hydro turbine power house in the village Sllakovc, Vushtrri ___________________________________________________________________________________________________________ Sh. Lajqi, N. Lajqi, B. Hamidi: “Design and Construction of Mini Hydropower Plant with Propeller Turbine”, pp. 1–13

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 12: Old water wheel turbine for generating energy from water

Figure 13: Excavation works and construction works for hydro turbine power house in the village of Sllakovc, Vushtrri

Figure 14: Open channels with excavators and connecting penstock with hydropower house ___________________________________________________________________________________________________________ Sh. Lajqi, N. Lajqi, B. Hamidi: “Design and Construction of Mini Hydropower Plant with Propeller Turbine”, pp. 1–13

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 14 present activity for open channels for installing penstock and connection penstock with hydro turbine power house.

Mini hydropower plants usually are run-of-river systems, which do not require a dam, and are installed on the water flow available on a year round basis,

6. Discussion of results

Construction work and production of mini hydro turbines can be done without higher qualified personnel,

Investment in a mini hydropower plant is not that high and period of turnover is shorter when compared with other alternative sources,

The investment costs increase drastically by increasing penstock diameter but energy losses decrease too.

Due to limitation of budget, there has not been possible installing water flow meter and output power meter for measuring generated energy from generator. Therefore, to determine the hydro turbine efficiency was difficult to do it. But, according to some measuring instruments that are installed in mini hydropower plant such as ampere meter, voltage meter and consumption energy meter is determined that the maximum generated energy from this plant when river flow is maximal has reached the value of 9.0 kW. The investment cost for building of this mini hydropower plant has value from 25 thousand Euros. According to existing feeding tariffs in Kosovo, investment payback period for this plant is about 6 years. The investing cost for producing of the one kW energy is ~ 2,800 €. Due to the limitation of water resources, the mini hydropower plant works only 9 months and planning for remount is foreseen to perform only during the period when the turbine is stopped, therefore installing of other turbine in order to increase readiness of plant is not reasonable.

Therefore, studying the potential of water for installing mini hydropower plants creates opportunity for employing new workers and enables the developing of new business and will be a good source for financial support in order to improve the lives of the population. Developed project performed in this paper has achieved the predicted result in power generation and has been working from May 2014.

References [1]

Delson Josel, Lini Varghese, Renjini G., Design of Small Hydro Electric Project Using Tailrace Extension Scheme, International Journal of Advanced Research in Electrical and Electronics Engineering, Volume 3, (2014), Issue 1, pp. 79-87.

[2]

Bilal Abdullah Nasir, Design of Micro - Hydro Electric Power Station, International Journal of Engineering and Advanced Technology, Volume 2, (2013), Issue 5, pp. 39-47.

[3]

Mohibullah, M. A. R. and MohdIqbal Abdul Hakim, Basic design aspects of micro-hydropower plant and its potential development in Malaysia, National Power and Energy Conference (PECon) Proceedings, Kuala Lumpur, Malaysia, 2004.

[4]

Celso Penche, Layman's guidebook on how to develop a small hydro site", Published by the European Small Hydropower Association (ESHA), Second edition, Belgium, June, 1998.

[5]

http://www.zonhan.com/eproducts/167.html (25.07.2015).

[6]

Pradhumna Adhikari, Umesh Budhathoki, Shiva Raj Timilsina, Saurav Manandhar, Tri Ratna Bajracharya, A Study on Developing Pico

7. Conclusion Nowadays requirements for energy continue growing more and more as a result of population growth and the rapid development of technology. Many countries have serious problems with supply regarding energy, especially for green energy. There are still many countries in the world that don’t have electrification. In order to reduce as much as possible the problem with generating energy and to improve peoples’ lives as well as environmental protection, we can draw the following conclusions: -

Many places in World have good potential for developing mini hydropower plants,

In locations where public grids do not exist, mini hydropower plants are crucial,

The only requirements for mini hydropower plants are water sources, turbines, generators, penstock and power houses, which not only helps each individual person but also helps the world and environment as a whole,

The choice of turbine will depend mainly on the available water head and the water flow rate,

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Propeller Turbine for Low Head Micro Hydropower Plants in Nepal, Journal of the Institute of Engineering, Volume 9, (2013), Issue, pp. 36–53, http://dx.doi.org/10.3126/jie.v9i1.10669

[7]

Robert Simpson & Arthur Williams, Design of propeller turbines for pico hydro, http://www.eee.nottingham.ac.uk/picohydro/d ocs/Pico%20propeller%20guidelines%20(Apr%2 02011)%20v11c.pdf (25.07.2015).

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.20160102

Numerical Modelling of a Solar Chimney Power Plant Filip V. Stojkovski1, Marija Chekerovska2, Risto V. Filkoski1*, Valentino Stojkovski1 1

Faculty of Mechanical Engineering, University ”Sts Cyril and Methodius” 1000 Skopje, Macedonia; risto.filkoski@mf.edu.mk 2 Faculty of Mechanical Engineering, University “Goce Delchev”, Shtip, Macedonia

Abstract

1. Introduction

A solar chimney power plant is a relatively new concept for power generation, based on renewable energy, combining the greenhouse effect with the chimney suction. The present study involves mathematical modelling of the system, based on the computational fluid dynamics (CFD) approach. The technical features of solar chimney power plant are analysed by use of CFD technique, as a way for effective optimisation of the object’s geometry and thermo-fluid aspects. The created numerical domain represents the complete volume of the object under consideration, with total height of 100 m, chimney’s base radius of 6.25 m, chimney’s top radius of 10.5 m and roof radius of 100 m. The numerical grid consists of 276000 volume cells, 729242 faces and 190156 nodes. The governing equations for mass, momentum and energy are solved using a commercial CFD code. The computation is performed using the assumption of steady 3-D flow and the turbulence is taken into consideration with the krealizable model. The discrete ordinates (DO) model is selected as thermal radiation model, since it represents properly the physicality of the radiation energy transfer phenomenon and due to the opportunity of applying a solar load directly to the radiation model. The obtained initial results demonstrate the capability of the CFD, as a powerful research and engineering tool for analysis of complex aerodynamic and thermal systems.

The solar energy represents inexhaustible nonpolluting energy resource. The solar chimney power plant is one of the relatively new methods aimed for large-scale generation of electricity, based on quite simple that combines solar energy and the greenhouse effect with the chimney suction. Basically, as it is illustrated in Figure 1 (modified from [1]), air and soil, or other layer, water-filled pipes, bags etc., are heated underneath the collector transparent roof by solar radiation. As a consequence, in combination with the chimney effect, a strong upward air draft is created due to a density difference that drives a turbine connected to an electrical generator.

Keywords:

Solar energy; Solar chimney; Heat transfer; Convection; Thermal radiation; Power generation

Article history:

Received: 31 July 2015 Revised: 19 October 2015 Accepted: 29 January 2016

The idea of solar chimney power plant initially was proposed by the German engineers Jorg Schlaich and Rudolf Bergermann in 1976 [2]. The solar chimney consists of three essential components: a solar roof collector, a chimney and a turbine, which have been familiar from time immemorial, but in this case, combined in a new way [2]. Research efforts on solar chimney are characterised with a number of theoretical studies, but with insufficient experimental work. The tests conducted on the first prototype in Manzanares, Spain, (a chimney radius 5 m, a height 195 m and a collector radius of 120 m), with a designed peak output of 50 kW, have shown that the concept is technically viable [3, 4]. Some of the works, conducted in the meantime, are directed towards development of mathematical models, in combination with experimental research [5, 6]. Others, such as [7], are devoted to development of specific modelling approach, in order to study important effects, such as wall friction, internal drag and area change. Pretorius and Kröger in [8] solved a convective heat transfer equation, evaluated a more accurate turbine inlet loss coefficient and the effects of various types of soil on the performance of a large-scale solar chimney power plant. The paper [9] presents some

___________________________________________________________________________________________________________ F. Stojkovski, et al: “Numerical Modelling of a Solar Chimney Power Plant”, pp. 14–21

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 1: Schematic layout of a solar chimney power plant (modified from [1])

theory aspects, technical issues, practical experience and basic economy analysis of solar updraft towers, giving results from designing, construction and operating of a small scale prototype in Spain. The feasibility of solar chimney power plants as an energy source for settlements, islands and remote places of countries in regions with specific weather conditions, is analysed in the works [1, 10]. The accuracy of different theoretical models for solar chimney power plants, as well as the dependence of the plant’s power output and efficiency on the geometry: chimney height and radius, collector radius and roof height is analysed in [11]. The common general findings of all the works regarding the efficiency can be summarised as follows: -

The plant overall efficiency is very low, but it increases with the plant size [11];

The investment cost per MW installed power and the levelised electricity cost decrease with the plant size [9];

In particular, a significant reduction of electricity generation cost is associated with increasing the plant size, leading to values comparable to the ones for electricity generated in conventional power plants [9];

The optimisation of the plant geometry is essential in order to design an economically and technically practical system [3, 4, 9, 11].

The present work is a continuation of the study presented in [12], with the main objective to investigate a solar chimney power plant operation by use of a computational fluid dynamics (CFD) approach. The CFD technique, as a proven powerful engineering tool, has been extensively used for modelling and investigation of operational behaviour of aerodynamics and thermal energy systems, helping researchers and engineers in performing their work more efficiently. In combination with experimental research and on-site measurements, CFD offers multiple benefits, such as time and cost reduction, possibility to reproduce the operating conditions, analysis of geometry variations etc. The Fluent CFD software is used in the present study for performing the numerical modelling and simulation procedure [13].

2. Methodology of the research 2.1. Theoretical background The operation principle of a solar chimney plant is shown in Fig. 1 (modified from [1]). Air enters the space under a low circular transparent or translucent roof that is open at the periphery and receives heat by solar radiation. The covering roof and the ground below it form a solar air collector. The height of the collector roof varies from the inlet to the junction with the chimney in accordance to the law of meridian flow theory. As a consequence, the air under the roof does not accelerate,

___________________________________________________________________________________________________________ F. Stojkovski, et al: “Numerical Modelling of a Solar Chimney Power Plant”, pp. 14–21

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

state of air at the collector outlet; 1r-2r – real expansion in the turbine(s); 3r – real state of air at the top of the chimney; 3r-4 – real heat rejection.

2.2. Model set-up and simulation This sub-section presents briefly the mathematical model, describing the solar chimney system geometry, the modelling and the numerical approach. The governing equations for mass, momentum and energy are solved using the commercial Fluent CFD code [13]. The object geometry is presented in Figure 3. The dimensions of the solar chimney under consideration are as follows: the collector roof inlet height is 2 m, the collector radius is 100 m, the chimney height is 100 m, the chimney’s base radius is 6.25 m and the chimney’s top radius is 8.75 m. The numerical mesh is given in Figure 4, consisting of 276003 volume cells, 729242 faces and 190156 nodes. Figure 2: The ideal and real air process in solar chimney power plant

so that the hydraulic energy losses are minor. A vertical chimney, also called tower, with large air inlets at its base, is placed in the middle of the roof. The chimney directs and intensifies the air flow to the top, causing air acceleration in the zone behind the wind turbine, which results with a pressure drop and making it suitable for placing the turbine at that point. Hot air flows up the tower due to its smaller density, compared to the cold air. Then, more hot air from the collector is drawn in by the tower suction, and cold air comes in from the outer perimeter. Continuous dayand-night operation can be achieved by placing a heat accumulator, water-filled tubes or tight bags, on the ground. In that case, the water heats up during day-time and releases heat at night, causing a relatively constant updraft in the tower. The energy contained in the air updraft is converted into mechanical energy by pressure-staged turbines located at the base of the chimney, which is then transformed into electrical energy by conventional generator. The theoretical and real processes of air in a solar chimney power plant are illustrated in the enthalpyentropy chart, Fig. 2 (slightly modified from [1]). The process 0-1-2t-3t-4-0 represents the ideal Brayton cycle. Due to aerodynamic losses, there is a small pressure drop in the real process, represented with the cycle 0-1r2r-3r-4-0. The annotation used in the chart has the following meaning: 0 – the state of air at the inlet; 1 – theoretical state of air at the collector outlet; 1-2t – theoretical expansion in the turbine(s); 2t-3t – air flow to the top of the chimney; 3t-4 – heat rejection; 4-0 – isentropic downdraft air movement to the inlet; 1r – real

The numerical simulations were carried out using segregated implicit pressure based solver for steady state conditions. The governing differential equations for mass and momentum are solved for steady incompressible flow. Turbulence is taken into account by the realizable k-ε model, with inclusion of standard wall function for near wall treatment. The velocitypressure coupling has been effectuated through the SIMPLEC algorithm. Second order upwind scheme was chosen for the solution scheme. The interpolation formulae for the air thermo-physical properties, used in the calculations, depending on absolute temperature, are given in Table 1. Two modelling approaches were considered within the present research, one taking into account natural convection without radiant heat transfer within the envisaged system (model 1) and the second one taking into consideration natural convection with radiant heat transfer (model 2). In the framework of the model 1, it was assumed that heat is coming from the collector roof and the heat accumulation system under the collector roof, situated at the ground surface. The adopted pressure values at the inlet and the chimney’s outlet are p1=p2=0 [Pa], at atmospheric pressure of patm=101325 [Pa]. In this case, the solver is density based, using the Boussinesq approximation theory for natural convection [13]. The density is set to change by the ideal gas law. The heat flux given at the collector roof surface and the heat accumulation is set as Φ = 1000 W/m2. In the model 2, in the present study, the discrete ordinates (DO) model is selected as a thermal radiation model, due to the opportunity of applying a solar load directly to the DO model [13, 14, 15]. The DO radiation model considers the radiate transfer equation (RTE) in the direction s as a field equation:

___________________________________________________________________________________________________________ F. Stojkovski, et al: “Numerical Modelling of a Solar Chimney Power Plant”, pp. 14–21

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 3: The object geometry

Figure 4: The numerical mesh of the object

Table 1: Thermo-physical properties of the air (T in K) Property Density, kg/m3 Dynamic viscosity, kg/ms Specific heat capacity, J/kgK Thermal conductivity, W/mK

Interpolation function ρ = 2.829 – 0.0084Т + 0.00001Т2 – 5·10–9Т 3 μ = 5·10–8 + 4·10–10Т cp = 1014.7 – 0.0005Т + 0.0001Т 2 λ = 0.0228 + 8·10–5Т

___________________________________________________________________________________________________________ F. Stojkovski, et al: “Numerical Modelling of a Solar Chimney Power Plant”, pp. 14–21

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

dI (r, s) +(a+ss)I(r,s)= ds σ T 4 σ 4π an 2 0 + s  I (r, s' )Φ (s ⋅ s' ) dω ' π 4π 0

(1)

where: Il(r,s) [W/m2srad] is the spectral intensity, λ is the wavelength, a [m-1] is the spectral absorption coefficient, r [-] is the location vector, s [-] is the direction vector (defined as s= μ i + η j + ξ k ), s’ [-] is the vector of scattering direction, σs [m−1] is the scattering coefficient at wavelength λ, σ0 [W/m2K4] is the Stefan-Boltzmann constant, σ0 =5,672⋅10-8 W/m2K4, ω’ [] is the solid angle, Φ is the phase function, which represents the probability that a ray with frequency ν’ from the direction s in a finite discrete solid angle dω’ will veer in the direction s’ inside the angle dω, with frequency ν. The irradiation flux is applied directly to the semitransparent walls (the collector roof) as a boundary condition, and the radiative heat transfer is derived from the solution of the DO radiative transfer equation. The DO radiation method considers the radiative transfer equation (RTE) as a field function. The RTE is solved for a finite number of discrete solid angles, each associated with a vector direction s, fixed in the Cartesian system. The fineness of the angular discretization can be changed accordingly and the DO model solves as many transport equations as there are directions s. In this case, the so-called S6 approximation was applied, corresponding to 48 flux approximations [13, 14, 15]. This approach gives sufficiently reasonable results for the amount of the numerical work. The higher-order approximations, such as the S8, with 80 flux approximations, require considerably more numerical effort. Imposing appropriate boundary conditions on the numerical domain is very important step in the creation of the CFD model. A ‘pressure inlet’ boundary condition is specified for the air inlet at the periphery bellow the collector roof and for the chimney outlet, an ‘outflow’ condition is specified. The collector roof is defined as semi-transparent cover, to which the irradiation flux is applied directly, as a boundary condition. The solar irradiation to the collector roof was changed in the range 400-1000 W/m2, based on actual weather conditions for selected season and period of the day. The grid independence test was performed in order to check the impact of the mesh density on the numerical solution, for meshes in range between 160000 and 330000 cells. Since the refinement from 276000 to 330000 cells did not change the results by more than 0.5 %, it was concluded that the influence of eventual further grid refinement would be negligible and,

Figure 5: Velocity vectors obtained with the model 1

therefore, the present mesh quality was taken as appropriate for computation.

3. Results and discussion The established mathematical model was used to conduct numerical simulations for different seasons of the year, periods of day and various weather conditions. Regarding the meteorological conditions, it was assumed that the object considered in this study is located in area with latitude 41o 32’ and longitude 22o 07’ (Krivolak region, Republic of Macedonia). Some results of the provided investigation, derived presuming average weather conditions during middle of April, are presented with the illustrations given in the next subsections.

3.1. Results of the model 1 The results of the model 1 show that the heat is evenly distributed under the collector roof and the velocity profile of the flow is uniform. The maximum velocity of the air in the plant, obtained by the simulations, is 8.06 m/s. The velocity vectors in two vertical cross-sections in this case are shown Figure 5. However, by applying different heat fluxes at the surfaces, there was no significant change of the intensity of the air flow, as the heat energy appeared to be insufficient to intensify the flow and to generate higher rates of natural convection, which leads to a conclusion that this modelling approach is not sufficiently accurate.

3.2. Results of the model 2 The model 2 is considered as successful, because by changing the heat intensity in the model, there was a

___________________________________________________________________________________________________________ F. Stojkovski, et al: “Numerical Modelling of a Solar Chimney Power Plant”, pp. 14–21

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 6: Contours of static pressure (in Pa)

Figure 7: Velocity vectors coloured by velocity magnitude (in m/s) in central vertical intersection

Figure 8: Contours of static temperature (in K), with implemented ground heat accumulator ___________________________________________________________________________________________________________ F. Stojkovski, et al: “Numerical Modelling of a Solar Chimney Power Plant”, pp. 14–21

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

significant change of the velocity increasing rate, which is regarded as a proof that the natural convection effect was established numerically. Figure 6 shows contours of static pressure in a central axis vertical intersection. Velocity vectors coloured by the velocity magnitude in a vertical intersection are given in Figure 7. Contours of static absolute temperature, in a case with implemented ground heat accumulator, are presented in Figure 8. The comparison between the results regarding the air velocity, obtained by CFD simulation and calculations based on the Boussinesq approximation theory for natural convection [13], are given in Table 2 and Chart 1.

It can be concluded from Chart 1, that the trend of the velocity calculated with CFD is following the curve patter calculated from the Boussinesq approximation theory, so the results can be assumed as correct. The total enthalpy, as an energy "input" in the system, is transformed into enthalpy of air, pressure energy, kinetic energy, potential energy and energy loss. The maximum value of the total enthalpy in this case was Emax = 16253 J/kg. By far the largest share of the energy is transformed into enthalpy of air (over 93 %) and just a small part is utilised in the wind turbine, which results with output electric power between 15 and 25 kW, depending on the air flow rate.

Table 2: Comparison between the air velocities obtained by CFD simulation and calculations based on the Boussinesq approximation T-coll. (K)

Mass flow (kg/s)

v0 (m/s)

v2 (m/s)

v_max - CFD (m/s)

v_max - Boussinesq (m/s)

350

993.92

0,787

4,281

8.378

9.881

345

981.34

0,777

4,223

8.259

9.748

340

969.01

0,766

4,166

8.141

9.613

335

956.54

0,756

4,108

7.979

9.475

330

944.11

0,746

4,05

7.905

9.335

325

931.55

0,736

3,991

7.785

9.192

320

918.63

0,725

3,932

7.660

9.047

315

905.74

0,714

3,871

7.539

8.898

310

892.62

0,703

3,809

7.417

8.744

305

879.05

0,692

3,743

7.294

8.585

300

864.87

0,681

3,674

7.165

8.421

Maximum air velocity at natural convection (comparison between the CFD calculated velocity and Boussinesq velocity)

v max (m/s)

12 10

y = 2,5659x0,5 R² = 1

y = 2,2268x0,4905 R² = 0,9979

6 v_max_NS 4

v_max_Boussinesq

2 0 10

∆T (K)

Chart 1: Comparison of the CFD calculated velocity and Boussinesq velocity at natural convection ___________________________________________________________________________________________________________ F. Stojkovski, et al: “Numerical Modelling of a Solar Chimney Power Plant”, pp. 14–21

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

4. Closing remarks The present study demonstrates the capabilities of the CFD technique, as a powerful research and engineering tool for analysis of complex aerodynamic and thermal systems, such as the solar chimney power plants. The CFD approach should enable an experimental study of a solar chimney power plant to be simpler and more economical, so this approach can be further developed and upgraded for more detailed analysis. Comparison between the two models applied in the present study, clearly shows that for further CFD investigation of such objects, a proper radiant heat transfer model should be included in the overall model, changing the air thermo – physical properties with other recommended values, or even changing the aerodynamic geometry of the object can lead to obtaining more accurate results. Objects in this study are assumed to be achieved, so that a successful numerical model for a solar chimney power plant is developed, i.e. the natural convection phenomena is obtained and proved with the CFD simulations (figure 9) which follows the trend of the Boussinesq theoretical calculated velocity. The accuracy of the results is relevant for this subject, but not so important, because the effort is aimed for developing a successful CFD simulation which can be used for further investigation of the flow and convection phenomena and energy capacities of such objects. Although, it can be concluded from this CFD approach, from the efficiency viewpoint, that it is much better to include the heat accumulation (water pipes, water bags or other layer on the ground) that will release heat from the ground during the night time, enabling significant increase of the overall plant efficiency, which is otherwise quite low and the natural convection can not be developed.

[4]

Haaf W., Solar chimneys. Part II: Preliminary test results from the Manzanares plant, International Journal of Solar Energy, 2, (1984), pp. 141–161.

[5]

Pasumarthi N., Sherif S. A., Experimental and theoretical performance of a demonstration solar chimney model – Part I: Mathematical model development, International Journal of Energy Research, 22, (1998), 3, pp. 277–288.

[6]

Pasumarthi N., Sherif S. A., Experimental and theoretical performance of a demonstration solar chimney model – Part II: Experimental and theoretical results and economic analysis, International Journal of Energy Research, 22, (1998), 5, pp. 443–461.

[7]

von Backström T.W., Gannon A. J., Compressible flow through solar power plant chimneys, ASME Journal of Solar Energy Engineering, 122, (2000), 3, pp. 138–145.

[8]

Pretorius, J.P., Kröger, D.G., Solar chimney power plant performance, Journal of Solar Energy Engineering, 128, (2006), pp. 302–311.

[9]

Schlaich J., Bergermann R., Schiel W., Weinrebe G., Design of Commercial Solar Updraft Tower Systems – Utilization of Solar Induced Convective Flows for Power Generation, Trans. of the ASME, Journal of Solar Energy Engineering, 127, (2005), pp. 117-124.

[10]

Hamdan M. O., Analysis of a solar chimney power plant in the Arabian Gulf region, Renewable Energy, Vol. 36, (2011), pp. 2593-2598.

[11]

Koonsrisuk A., Chitsomboon T., Accuracy of theoretical models in the prediction of solar chimney performance, Solar Energy, Vol. 83, (2009), pp. 1764-1771.

[12]

Filkoski V. Risto, Stojkovski V. Filip, Stojkovski Valentino, A CFD study of a solar chimney power plant operation, Proceedings, (Editor A. Olabi), 6th Int. Conf. on Sustainable Energy and Environmental Protection SEEP 2013, Maribor, Slovenia, 2013, pp. 631-636.

[13]

Fluent Inc., Fluent 6.2 User’s Guide, Lebanon NH 03766, USA, 2005.

[14]

Fiveland W. A., Three-dimensional radiative heattransfer solutions by the discrete-ordinates method, Journal of Thermophysics, Vol. 2, No. 4, pp. 309-316, 1988.

[15]

Chekerovska Marija, Filkoski V. Risto, Efficiency of solar-tracking liquid flat-plate solar energy collector, Accepted for publication in (Int. Journal of) Thermal Science, (2015).

References [1]

Nizetic S., Ninic N., Klarin B., Analysis and feasibility of implementing solar chimney power plants in the Mediterranean region, Energy, Vol. 33, (2008), pp. 1680-1690.

[2]

Schlaich J., Schiel W., Solar Chimney, Encyclopaedia of Physical Science and Technology, Third Edition, 2000.

[3]

Haaf W., Friedrich K., Mayr G., Schlaich J., Solar chimneys. Part I: Principle and construction of the pilot plant in Manzanares, International Journal of Solar Energy, 2, (1983), pp. 3–20.

___________________________________________________________________________________________________________ F. Stojkovski, et al: “Numerical Modelling of a Solar Chimney Power Plant”, pp. 14–21

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.20160103

Small Scale PV-Power – On Site Use Maximization through Smart Heat Pump Control Martin Felix Pichler1*, Alexander Arnitz1, Markus Brychta2, Andreas Heinz1, René Rieberer1 1

Graz University of Technology Inffeldgasse 25/B, 8010 Graz, Austria; martin.pichler@tugraz.at, martin.pichler@gmx.eu 2 University of Innsbruck

Abstract

1. Introduction

Small scale grid-connected photovoltaic (PV) power plants may lead to unwanted disturbance to the electricity grid. In addition, low feed-in tariffs motivate the homeowner and operator of the PV plant to maximize the self-consumption. A PV system in connection with a compression heat pump (HP) for heating (and cooling) purposes of a single family house equipped with some kind of thermal storage poses an interesting optimization problem in this context. Load management is possible through an intermediate thermal storage such as a simple water storage tank. However, in addition the building may comprise a thermally activated building system (TABS) such as a floor heating or an activated ceiling. Thermal storage capacity enables to maximize the utilization of PVpower to pre- or ‘overheat’ the water tank or the whole building, and at the same time the grid is prevented from overcharge and thus regulating (power) requirements are reduced. This research investigates a system consisting of a small grid connected PV plant in connection with a HP charging a thermal storage. The load of the storage comprises domestic hot water draw offs and the space heating demand of a single family house. The main challenge is to maximize the PV-electricity self-consumption. The work presented herein is a preliminary study for a more complex system described in the introduction.

Small scale grid-connected photovoltaic (PV) systems may represent an unwanted disturbance to the grid. In addition, low feed-in tariffs motivate the operators of the PV plant to maximize the self-consumption and minimize the consumption from the grid. The volatile character of solar irradiance, typical for Central Europe, poses a challenge in terms of stability and control at grid level, if numerous small scale PVsystems potentially feed into the grid. Assume a compression heat pump (HP) installed in a single family house for heating (and cooling) purposes. A predictive weather-forecast-utilizing control framework may be used for load management to maximize the PV-electricity self-consumption and prevent from overcharge at grid level and thus reduce regulating (power) requirements. This becomes possible, due to the relatively high thermal time constants of current standard buildings.

1.1. Short review on related research

Keywords:

PV; Load change flexibility; Smart control, MPC; Thermal storage

Article history:

Received: 17 August 2015 Revised: 10 September 2015 Accepted: 29 January 2016

Wimmer [1] has investigated the model predictive control (MPC) of a heat pump for a single family house with different approaches, but for constant compressor speed. Simulations have shown cost savings up to 13% – compared to a standard on/off controller – and an electricity consumption reduction up to 3.5%, on an annual base. Bianchi [2] continued this research and investigated an adaptive framework for the same application. Young [3] deals with the topic demand side management (DSM) with heat pumps for single family houses with floor heating – he found a prediction horizon of 24 h to outperform smaller horizons. The electrical load shifting potential of a smart heat pump has also been analyzed in Danny [4] with the finding of a

___________________________________________________________________________________________________________ M. F. Pichler, et al: “Small Scale PV-Power – On Site Use Maximization through Smart Heat Pump Control …”, pp. 22–30

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

load shifting potential between 19% and 33%, however, the annual electricity consumption increased by approximately 9%.

1.2. The project TheBat A research project dealing with a smart control of a HPsystem in connection with a PV-system, for the purpose of space heating (SH) and domestic hot water (DHW) preparation is currently ongoing at Graz University of Technology and University of Innsbruck. The project “TheBat” comprises simulation, prototype test, and real hardware in the loop (HIL) tests. Based on the building used in IEA SHC Task 44 (compare Figure 1) a reference building was defined. The scope is to analyse different concepts of thermal storage, also including the building mass itself. The gable roof is asymmetric, with a steeper south orientated part (26 m²) which can be easily used for solar applications. For further details see [5]. The original building was slightly adapted in order to meet the specific requirements of the project. The main differences are the implementation of thermally activated building systems on the ground level and the first floor. Three different building types were considered (see Table 1). The buildings RES 45 and OFF 45 will be investigated in connection with the HP-, PVsystem operated by a model predictive control (MPC). The planned actuating variables are the HP-power

(a)

(frequency) and valve positions to divide the heat supply to the storage and/or the building. Controlled variables are a storage tank- and the air-temperature in the building. The aim is to apply a linear MPC concept which requires a linear model for the thermal storage and the building dynamics. In addition, the expected PVgenerated electricity must be estimated by given weather forecast data. Finally the HP-characteristic has to be modelled using a suitable approach. Research presented herein focusses on three stages on the way to the final solution for the complex system. First, it deals with modelling related to PV-generated electricity, second, it covers the thermal storage tank modelling part, and finally this research reports on definitions made to evaluate the controller and the system performance.

2. Reference system and boundary conditions The parametrization of a predictive controller, which means setting a reference trajectory, minimum and maximum bounds and weights for the actuated and controlled variable etc. requires certain simulation experiments to assure a reasonable control behavior. The reference system described in the following and the research reported in this manuscript is tailored to gain insight into the heat pump related MPC parametrization. Building related modelling for MPC purpose is not covered in this report.

(c)

(b)

Figure 1: 3D view of the reference building (a), illustration of thermally activated floors (b), and related cross-section of the floor (c) [5]

Table 1 Characterization of the buildings investigated in the research project TheBat Building Type RES 45 OFF 45 RES 15

Usage Residential Office Residential

Heat demand approx. 45 kWh/(m² a) 45 kWh/(m² a) 15 kWh/(m² a)

U-Value Walls 0.29 W/(m² K) 0.29 W/(m² K) 0.18 W/(m² K)

U-Value ceiling (1st floor) 0.69 W/(m² K) 0.60 W/(m² K) 1.27 W/(m² K)

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 2: Scheme of the system controlled by the MPC. Abbreviations not explained in the further text: Domestic hot water (DHW), cold water (CW)

2.1. Hydraulic and electric scheme This section introduces the reference simulation system illustrated in Figure 2. It consists of a heat pump (7 kW @B0W35), a thermal storage and a PV array. By contrast to the original system of the project TheBat, the SH demand is incorporated by means of a load profile, which acts on an increased thermal storage tank to mimic the thermal inertia of the building. The ground is assumed to serve as a heat reservoir for the HP. The evaporator (e) supply temperature is assumed as constant 10 °C. The HP cycle consists of a scroll-compressor with variable speed (f) and an economizer which injects medium pressure vapor into the compressor. Heat delivery from the HP to the storage occurs via the condenser (c) which connects to the storage at half height and via the desuperheater (d) at the top. More details on the HP may be found in [6]. The thermal storage with a volume of 1.2 m³ represents the heat sink. It incorporates five equally spaced sensors to measure the temperature stratification (ϑSi). The counter flow heat exchanger for DHW preparation is connected at the top of the storage as is the desuperheater outlet of the HP. The SH connection is situated below these connections. The DHW and SH return flows are reinjected in the lower storage region. During ordinary heat pump operation, the circulation pump (P1) ensures a mass flow rate in order to reach a 5 K higher temperature at the outlet of the condenser

(ϑco = ϑci + 5 K). A share of this mass flow is diverted with the valve (V1) to pass through the desuperheater, however, the bigger part is charged into the storage. The mass flow further heated in the desuperheater is adjusted such to reach the set outlet temperature (ϑdo,set = ϑco +25 K) – characteristic diagrams are utilized to determine the required mass flows.

2.2. PV power estimation and climate data The suggested control scheme requires a model to estimate the PV power given a forecast time series for the solar irradiance. A few models with different degree of complexity may be found in literature – see [7-9]. These approaches are based on a method to derive PV electrical characteristics from generally provided test data, see [10]. The four-parameter model from Eckstein [7] is used in TRNSYS for annual simulations. The approach, facilitated to estimate the expected PV power from irradiance for control purpose relies on a simple efficiency model, which is briefly described in the following – a detailed description may be found in [7] or [11]. The basic assumption of this approach is the existence of a maximum power point tracker (MPPT). This device assures an operation of the PV array, such that in the IV graph, the area below the operating point becomes a maximum. Finally, the efficiency of the PV array is given as a function of the actual cell- and outside ambient-

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

temperature ( ), and of irradiance on the aperture ) may be area ( ). The cell temperature ( approximated by an energy balance considering the incoming solar energy, the generated electricity and the dissipated heat energy. The formula is given by Eq. 1 with = 0.9 as mean absorbance transmission and being the heat loss constant, and , coefficient and the efficiency at reference conditions, both derived from manufacturers data. As reference conditions Nominal Operating Cell Temperature (NOCT) conditions are used. =

(

)

1−

(

)

(1)

Given the actual efficiency at MP is calculated using and , characterizing the open Eq. 2 with , circuit voltage and short circuit current and the ’s the according temperature coefficients at reference conditions. Figure 3 shows the relatively good ( ) and the maximum power for approximation of different irradiance values ( ), when comparing the efficiency- against the four parameter model results.

3. Methods and performance indicators Simulations conducted with TRNSYS [12] use a time step of 2 min. The maximization of the direct utilization of in situ generated PV electricity can be interpreted as a motivation for the advanced control strategy.

3.1. Base case scenario Table 2 provides an overview for annual simulation results obtained with a standard hysteresis control. The values fit to a typical single family home. , , ° is the total incident solar radiation on the inclined PV cell aperture area and WPV the total PV generated electricity. Wel is the total compressor electricity demand, and Qcond and Qdesup represent the heat delivered by the condenser and the desuperheater. QDHW and QSH give the hot water and the heating energy demand and QS,loss indicates the thermal storage losses. The total demand QDHW + QSH must be covered by the totally generated heat, Qcond + Qdesup.

3.2. Predictive controller =

(2)

1 + −

The climate data set for simulation represents the conditions for Strasbourg, compare [5]. Although the PV array, with a total cell aperture area of 20.4 m² facing south at an angle of 45° is connected to the grid, its main purpose is to drive the HP.

Ambient temperature in °C

The goal for the investigated predictive controller is to with . The control cover as much as possible of is mostly concerned with temperatures being the controlled variables and thermal power or a related variable (compressor speed) being the actuating variable. The refrigerant cycle itself is not in focus of the research. The controller is implemented in MATLAB [13] and relies on ideal forecast data for the weather, meaning that the used prediction values match the simulation data. A (disturbance) prediction for the DHW and SH demand is currently not incorporated. The MPC is called every 10 minutes to update the power of the HP

Solar irradiance in W/m²

Figure 3: PV module efficiency as a function of the ambient temperature (left) and PV power as a function of the irradiance (right) – PV module Solarwatt 60P ___________________________________________________________________________________________________________ M. F. Pichler, et al: “Small Scale PV-Power – On Site Use Maximization through Smart Heat Pump Control …”, pp. 22–30

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

indicator for the HP control. It deviates from Eq. 3a only in the denominator. The integral represents the pure grid consumption due to HP operation. The sub index at the closing bracket indicates that only positive values are allowed, a negative value would mean feeding into the grid, which is not taken into consideration. A high SPFCtr indicates a good control concerning a high degree of electricity self-consumption: = Figure 4: Overview of the predictive control scheme. The measured storage temperature provides feedback to the MPC

and the division into condenser and desuperheater heat flux. The general reference value for ϑ is 50 °C.

3.3. Predictive controller

For the classic reference case simulation the compressor frequency is set to 90 Hz constantly during HP operation. < 45 ° and turns off at > The HP turns on for 50 ° . During MPC operation the compressor frequency is adjusted according to the required power.

3.4. Performance indicators The HP coefficient of performance (COP) is given as the sum of heat flux delivered by the condenser and the + ) divided by the desuperheater ( required power to drive the compressor (Pel) and the pump P1 ( , ): =

(3)

The main performance indicator is the seasonal performance factor (SPF) of the HP (Eq. 3a). It is defined + ) as the totally delivered thermal energy ( divided by the auxiliary (P1) and compressor electr. + consumption ( , ) of the HP, during the same interval: =

(3a)

Concerning the control, a controller performance related SPF is defined (Eq. 4), which presents an interesting

(4)

Extending towards the whole system one may define a system specific SPFCtr (Eq. 5), which indicates the generated useful energy, for DHW and SH purpose + ), per kWh of grid consumed electricity. ( This is the most interesting indicator at system level:

The principal compressor frequency control range is from approximately 30 Hz to 117 Hz. The heat pump coefficient of performance (COP, Eq. 3) depends mainly on the frequency, the condenser inlet temperature and the source temperature (assumed constant 10 °C).

∙

= ,

∙

(5)

The term incorporates the required power , , over all auxiliary devices of the system.

4. Simulation results for MPC- and Base-Case The characteristic behavior of the MPC in comparison to the classic base case control is demonstrated by means of temperature and heat flux trajectories for seven sequential days. Summer simulation results are depicted in Figure 5 and Figure 6. The smart MPC and the relatively low heat demand lead to HP operation only at times with PV power being available (Figure 5). For the MPC operated HP the COP at operation times is higher than for the classic control (Figure 6), compare also Table 3. However, this depends on the heat exchanger design and the compressor efficiency characteristic, which varies among HPs. Winter simulation results look slightly different due to increased total heat demand compared to the summer; this is demonstrated by Figure 7 and Figure 8. The MPC in winter is also characterized by HP operation at times where PV power is available. This PV-led operation causes overheating in the storage as visible in the top graph of Figure 7. However, by contrast to the summer, the HP is also operated at times where PPV=0. Suddenly decreasing storage temperatures indicate SHor DHW- heat demand. Sometimes (Pel+Paux,el) is smaller than the available PV power, this indicates potential for improvement. The classic control depicted in Figure 8 leads to shorter HP operation intervals with increased power and the temperature trajectories show less variation.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 5: MPC simulation results for the heat fluxes and PV power for seven summer days

Figure 6: Classic control (base case) for the heat fluxes and PV power for seven summer days

Figure 7: MPC results; the top graph shows the top, middle and bottom storage temperature, and the lower graph shows the relevant heat fluxes and PV power, for seven winter days ___________________________________________________________________________________________________________ M. F. Pichler, et al: “Small Scale PV-Power – On Site Use Maximization through Smart Heat Pump Control …”, pp. 22–30

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 8: Classic control, base case results for top, middle and bottom storage temperature (top graph) and relevant heat fluxes and PV power (lower graph) for seven winter days

Table 2: Base case and MPC scenario simulation results. Note: The specific PV power is with respect to the cell aperture area, and the DC/AC conversion efficiency is assumed 1 , ,

ℎ

Base Case ℎ/

ℎ

MPC ℎ/

25264

4272

2689

6027

3236

2127

6475

640

1236

209

132

295

158

104

317

25264

4272

2352

6982

2363

2132

6464

744

1236

209

115

342

116

104

316

Table 3: Performance indicators for the base case and the MPC scenario ,

COP mean ± sigma

mean ± sigma

Base Case

3.38

3.66

5.24

3.30

3.38 ± 0.49

0.17 ± 0.01

MPC

3.77

7.80

10.75

6.73

3.92 ± 1.04

0.17 ± 0.01

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

4.1. Annual Results Table 2 provides annual simulation results for relevant amounts of heat and generated electricity for both, the base case and the MPC case. The situation in terms of ), incident solar radiation ( , , ° ), PV output ( generated heat ( + ) and heat consumption ( + ) is approximately the same for both cases. The required electricity demand for the compressor ) is higher for the base case since the compressor in ( that case operates constantly at 90 Hz for which the COP is mostly lower than the COP at MPC operation, where the dominant operation frequency is close to the lowest frequency approximately 35 Hz (compare COP values in Table 3, notice also the higher empirical standard deviation (σ) for the MPC case).

incorporated into the MPC framework. In addition, several sensitivity analyses with respect to relevant sizing parameters will be conducted.

Acknowledgement The project “The thermal battery in the smart grid in combination with heat pumps – an interaction optimization (TheBat)“ is sponsored by the Austrian Climate and Energy Fund and conducted within the framework of the program „ENERGY MISSION AUSTRIA“ (Project-Number 838657).

References [1]

Wimmer, R. W., Regelung einer Waermepumpenanlage mit Model Predictive Control (Model predictive control of a heat pump system), PhD thesis, ETH Zuerich, 2004.

[2]

Bianchi, M. A., Adaptive Modellbasierte Praediktive Regelung einer Kleinwaermepumpenanlage (Adaptive model based predicitive control of a small scale heat pump system), PhD thesis, ETH Zuerich, 2006.

[3]

Young, J. Y., Demand-side-management with heat pumps for single family houses, Proceedings of BS2013, 13th Conference of International Building Performance Simulation Association, Chambery, France, 2013.

[4]

Danny, G., Wapler, J., Miara, M., Simulation and analysis of demand-side-management effects on operating behavior and efficiency of heat pumps, 11th IEA Heat Pump Conference Proceedings, Montreal, Quebec, 2014.

[5]

Dott, R., Haller, M. Y., Ruschenburg, J., Ochs, F. & Bony, J., The Reference Framework for System Simulations of the IEA SHC Task 44 / HPP Annex 38 Part A: General Simulation Boundary Conditions; Part B: Buildings and Space Heat Load, Technical report, IEA, 2013.

[6]

Hengel, F., Heinz, A., Rieberer, R., Analysis of an air source heat pump system with speed controlled compressor and vapor injection, Applied Thermal Engineering, in Review, 2014.

[7]

Eckstein, J. H., Detailed modelling of photovoltaic system components, Master’s thesis, University of Wisconsin – Madison, 1990.

[8]

Fry, B., Simulation of grid tied building integrated photovoltaic systems, Master’s thesis, University of Wisconsin – Madison, 1998.

[9]

Soto, W. D., Klein, S. & Beckman, W., Improvement and validation of a model for

The significant “overheating” for the MPC case, as visible in the top graph of Fig. 7, leads to approximately 15% higher storage losses ( , ) for the MPC case. The performance of the suggested MPC framework may ) defined be expressed by the control related SPF ( by Eq. 4. This SPF indicates to which degree the HP uses the available PV electricity. The upper limit is given by , which is an ideal theoretical value. This , value is based on the assumption, that Pel+Paux,el – at times where available PV power overlaps with the HP operation time – is entirely provided by the PV plant. As Table 3 shows the control related SPF is much higher for the MPC case than for the base case. The number of 7.8 means that a consumption of 1 kWhel from the grid, leads to 7.8 kWh of generated heat. The system and ) shows that the effect control related SPF ( , concerning the whole system is smaller, which is mainly due to the increased storage losses.

5. Conclusion and future work The self-consumption tailored MPC framework of the heat pump shows a relatively high degree of operation of time during available PV power, leading to a of 6.73. These SPF’s 7.8 and a slightly smaller , neglect the consumption of PV generated electricity. Young [3] found values ranging from 8.6 to 17.0 for a is, but similar performance indicator as considering solar and wind generated electricity from the grid instead of purely on-site PV generated electricity. Future work will focus on the extension and improvement of the MPC framework. This means incorporation of a heating demand prognosis through a building model, which entails also a further control variable (room air temperature). Further, the COP as a function of the compressor frequency of the HP will be

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

photovoltaic array performance, Solar Energy 80(1), 78 – 88, 2006. [10]

[11]

Townsend, T., A Method for Estimating the LongTerm Performance of Direct-Coupled Photovoltaic Systems, Master’s thesis, University of Wisconsin – Madison, 1989. Duffie, J. A., Beckman, W., Solar Engineering of Thermal Processes, Wiley, 2006.

[12]

Klein, S., Beckman, W., Mitchell, J., Duffie, et al., A TRaNsient SYstems Simulation Program – TRNSYS 17.00.0019, Manual, Solar Energy Laboratory, University of Wisconsin-Madison, 2010.

[13]

Bemporad, A., Morari, M., Ricker, N. Lawrance, Model Predictice Control Toolbox Users’s Guide V3.2, Release 2010a, Mathworks, Natick, MA, 2010.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.20160104

Long Term Assessment of Nuclear Technology Penetration using MESSAGE – The Case of Romania Eugen Bancheş1*, Ionut Purica2 1

Nuclear Agency & Radioactive Waste 21-25 D.I.Mendeleev St., 010362 Bucharest, Romania; eugen.banches@gmail.com 2 Romanian Academy and AOSR Casa Acadenmiei, Calea 13 Septembrie nr.13, 050711 Bucharest, Romania; puricai@yahoo.com

Abstract

1. Introduction

The paper is doing a long-term simulation of the nuclear technology penetration for the Romanian power system using the IAEA MESSAGE optimisation model. The horizon taken into consideration is 2050 and 2070. The production and the demand are considered with various scenarios and the emissions of CO2 are also evaluated. Given that the program is using a simplex optimisation algorithm only the optimal scenarios are retained in the given restrictions and specific objective function. This provides a useful support for the decision maker in selecting only optimal scenarios. The results are destined to assess the impact of the nuclear technology on the implementation of the EU energy and climate change policy on a long term basis such that to eliminate short term effects in the power system. Given the specifics of the Romanian power system both electrical energy and thermal energy (CHPs) are considered in the main scenarios.

The energy EU policy recently launched by the Commission is bringing along with the existing renewables penetration, emissions reduction and efficiency increase two more pillars i.e. interconnection of 15% of the energy consumption and the research in energy systems. This approach gives a new area of research perspective, especially on the long-term energy systems development that needs to assess the impact of low or no emission energy technologies penetration beyond the short term spikes such that the one of renewables based on subsidies. These subsidized technologies are saturating after a period of fast penetration and need to be correlated with the rest of the system is clearly showing up. Moreover, an integrated view is necessary on a longterm horizon for the system in order to assess the need to prepare investment and to secure the emissions reduction. Also, out of the various technological combination scenarios one should have the capability to select the optimal ones, given specific restrictions in the general frame of the economical optimization function. The restrictions are reflected in the requirements of the EU policy summarized below based on the Road Map 2050 and other documents e.g. [6], [7], [8], [9].

Keywords:

Article history:

Nuclear energy system; Modelling energy system, CO2 emissions mitigation; Electricity and heat cogeneration; Energy system security Received: 31 July 2015 Revised: 19 October 2015 Accepted: 29 January 2016

2. Transforming the energy system 2.1. Energy saving and managing demand: A responsibility for all Improving energy efficiency is a priority in all of the new scenarios related on the energy system decarburization. Current initiatives need to be implemented swiftly to

___________________________________________________________________________________________________________ E. Banches, I. Purica: “Long Term Assessment of Nuclear Technology Penetration using MESSAGE – The Case of Romania”, pp. 31–38

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

achieve change. Higher energy efficiency in new and existing buildings is the key. Buildings – including homes - could produce more energy than they use. Investments by households and companies will have to play a major role in the energy system transformation. Energy efficiency has to follow its economic potential. This includes using the waste heat of electricity generation in combined heat and power (CHP) plants.

2.2. Switching to renewable energy sources The second major pre-requisite for a more sustainable and secure energy system is a higher share of renewable energy beyond 2020. Renewables will move to the centre of the energy mix in Europe, from technology development to mass production and deployment, from small-scale to largerscale, integrating local and more remote sources, from subsidized to competitive. Many renewable technologies need further development to bring down costs. Storage is currently often more expensive than additional transmission capacity, gas backup generation capacity, while conventional storage based on hydro is limited. With sufficient interconnection capacity and a smarter grid, managing the variations of wind and solar power in some local areas can be provided also from renewables elsewhere in Europe. This could diminish the need for storage, backup capacity and base load supply. Renewable heating and cooling are vital to decarburization. A shift in energy consumption towards low carbon and locally produced energy sources (including heat pumps and storage heaters) and renewable energy (e.g. solar heating, geothermal, biogas, biomass), including through district heating systems, is needed.

2.3. Gas plays a key role in the transition Gas will be critical for the transformation of the energy system. Substitution of coal (and oil) with gas in the short to medium term could help to reduce emissions with existing technologies until at least 2030 or 2035. On the other hand, gas heating may be more energy efficient than electric heating or other forms of fossil fuel heating, implying that gas may have growth potential in the heating sector in some.

2.4. Transforming other fossil fuels Coal in the EU adds to a diversified energy portfolio and contributes to security of supply.

Oil is likely to remain in the energy mix even in 2050 and will mainly fuel parts of long distance passenger and freight transport. The challenge for the oil sector is to adapt to changes in oil demand resulting from the switch to renewable and alternative fuels and uncertainties surrounding future supplies and prices.

2.5. Nuclear energy as an important contributor Nuclear energy is a decarburization option providing today most of the low-carbon electricity consumed in the EU. Some Member States consider the risks related to nuclear energy as unacceptable. Since the accident in Fukushima, public policy on nuclear energy has changed in some Member States while others continue to see nuclear energy as a secure, reliable and affordable source of low-carbon electricity generation. New nuclear technologies could help to address waste and safety concerns. As a large scale low-carbon option, nuclear energy will remain in the EU power generation mix.

2.6. Smart technology, storage and alternative fuels Whichever pathway is considered, the scenarios show that fuel mixes could change significantly over time. Much depends on the acceleration of technological development. It is uncertain which technological options might develop, at what pace, with what consequences and trade-offs.

3. Brief description of the MESSAGE model The MESSAGE (Model for Energy Supply Strategy Alternatives and their General Environmental Impacts) was first developed as model in IIASA (International Institute for Applied system Analysis) and later on taken over and extended by the IAEA (International Atomic Energy Agency) [4]. The model is based on a simplex (linear programing) algorithm [1] that finds the optimal scenario of energy technology penetration based on an optimality function and on several restrictions specified for the system under consideration. The fact that the model allows discerning if a given scenario is optimal or not is an advantage at least for the capability to eliminate those scenarios that do not meet the optimality criterion, thus concentrate on the optimal ones. [3], [4], [15].

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

More evolved models may be used e.g. as developed in [2], but for the purpose of an initial assessment we decided to use the simplex that is at the basis of the MESSAGE initial program.

Romanian energy system was developed by the ANDR and ICN-Pitesti team leaders using a preliminary model agreed by AIEA-PESS experts. The model improvement was implemented step by step based on the IAEA experts’ recommendations.

4. Scenarios, parameters and variables

The main improvements refers to use an updated study [5] on the classical and renewable long term contribution in national energy mix (as competitors up to 2070 to old and new nuclear technologies) and to use an Intermediary Level of energy between the Secondary Level and Final Level (in order to allow the independent control of the distribution and storage the electricity produced by the nuclear and renewable technologies, and the electricity and heat produced by the classical technologies). In the intermediary level works the hydro-pumping storage technology and the district heat storage and distribution technology were considered. These technologies are critical to ensure the security of total energy supply in the National Energy Mix (NEMix).

In the determination of scenarios several parameters and variables were considered as presented in the Appendix 2. Our paper is only considering selected scenarios that reflect some of the requirements of the EU policy such as considering the distributed generation of heat from electrical energy sources or analysing the penetration of pumped storage hydro power plant, in the general framework of nuclear technology penetration in the power system of Romania. These considerations have reduced the number of selected scenarios that were run out of all possible ones. Several scenarios were run and only one has been chosen to be presented in here as a random example of the model possibilities. It should be underlined that the run scenarios are optimal scenarios that the model produces using a simplex algorithm given the specific scenario restrictions. The linear programming optimisation algorithm, in some cases, resulted in no optimal scenario with the given parameters and imposed restrictions. Having these optimal scenarios in place allows the decision maker to choose, based on economic criteria, a scenario; but whatever scenario is chosen it is an optimal one in its specific restrictions and objective function. Present status of the energy system – in order to better understand the presented scenarios the present status of the energy system is given briefly (a list of recent data may be found in [5], [11], [12], [13], [14]). The generation of electrical energy in Romania is done from a balanced portfolio of technologies that includes 30% hydro, 20% coal, 18% nuclear, 15% hydrocarbures and the rest of 17% renewables (combining PV and wind and 0.7% biofuels). There is a sizeable CHP generation (actually the CHP size has reduced continuously in the last 25 years still remaining important). Heat is also produced from heat only boilers either distributed residential or integrated into the heat networks serving to take over the heat demand peaks. Related to heating with renewable generation a massive introduction of heat pumps was considered as a potential extension of the heating options once the classic units were retired also by the penetration of nuclear units. This situation is reflected in the technology penetration figures in Appendix 1.

All classical, renewable and nuclear technologies are selected as possible competitors in the national long term energy mix, only if the associated technical and economic data are validated and defined in the general frame of the accessible references until February 2015. In respect with this principle, in this stage of the study, some innovative nuclear technologies (as eg. ALFRED – LFR) and the new proposal for carbon capture technologies (not expected to be validate until 2070), are not considered. The national energy mix considered in the model for the preliminary study, includes over 50 technologies. The Energy Chain is developed on 6 Energy Levels: Resources, Primary, Secondary, Intermediary, Final and Demand. There are 31 technologies “competitors” in energy production. There are 5 technologies in NEMix that transport, store and distribute to the demand level the energy in order to ensure the stability of the system on the all 24 load regions considered in NEMix. The preliminary version of the reference case study was tested and agreed by ANDR and ISPE team in February 2015. This version was improved by ANDR team by taking into account the available internet information about the contribution of Hot Water Boilers and District Heat Mix and Distribution System used by the main national companies implied in provided the electricity an heat in the NEMix.

5. General MESSAGE case study development

The code of the case studies, given in Appendix 2 (NESαdβehϑsγδεvζyθ), includes the possibility to identify over the 2000 versions of the model in order to provide results for a specific selected of input data setting:

The general MESSAGE case study for the long term assessment of nuclear technology penetration in

• 3 versions of the Nuclear Energy System – option/ scenario (α=Reference; Low; High)

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

• 3 versions of the Discount rate of investments (β=8%- reference; 5%- low; 10%- high) • 3 version of the Scenario for the evolution of the demand of electricity and heat (ϑ=1 – reference; 2 – low; 3 – high) • 3 versions of the “Specific options” to use support technologies & systems for energy transport storage & distribution (γ) like: „district station for Hot Water Mix and Distribution”, “Hot Water Boilers, support for peak of gas CHP over-load heat requirements in the system” or „Electric Heat for cooking and for domestic boilers”) • 3 versions of the “Specific options” to use tax of CO2 emissions (δ = 0 for 10 [USD/tone CO2]; 1 for 5 [USD/tone CO2]; 3 for 30 [USD/tone CO2]) • 2 versions of the “Specific options” to bound renewable contribution in energy mix • 3x2 version (ζ=23 –reference; 25 – low; 29 high) of new nuclear investment cost vs the basic national concept (N) or new innovative concept (Q) • the option to record the last year of the period of modelling (y) and to record if the program run in integer mode all the modelling period (eh – option) or was set to run in integer mode only until 2050 (ieh – option). In a more comprehensive report conducted by Romania in an IAEA-INPRO project the modelling team have selected only 50 versions from over 150 optimal versions performed. The primary information about the case studies selected to be analysed in the INPRO national report are objective function, and elapsed time used by MESSAGE to run the specific case study. Elapsed time is important, when selecting the representative versions used to perform sensitivity NEMix assessment (some versions run over 180 hours – other only 20 minutes). In what follows the results of a selected scenario (code [NES3d5eh2s200v23Ny70]) are presented in detail. It must be underlined that each of the considered scenarios has a similar set of results and the purpose of this presentation is to show one in detail. This scenario was chosen for demonstration purposes just to give the reader a flavour of the capability of the model. As coded this is a rather complex scenario including: 4 CANDU units (operational 2 ones and those in construction – being approx. 40% ready/after 2019 + life extension)+ NEW NPP (Gen III+: LWR/ PWR/after 2035 – the LWR 1000MW units were chosen as a balance between the economy of scale and the size of the power system to ensure operational security)) + Gen

IV(ALFRED)/after 2080 (not yet in the model until 2070); 5% discount rate; low demand; NO bounds of Hot Water Mix and Distribution (HWMixD) and Hot Water Boilers (HWB); 10 [USD/tone CO2] penalties of CO2 emissions; Reference investment cost for Nuclear; 2070 is the last year of modelling the CASE STUDY in MESSAGE. It may be seen that the case is considering the penetration of nuclear technologies both the existing CANDU and advanced LWR technologies. The demand was chosen low given the trend of the energy efficiency policies and the price of the t of CO2 emissions taken at a low value to be conservative. This is the reason for choosing the power of the LWR units at 1000 MW due to operational reasons of the power system. Also, the investment cost for nuclear is at the reference value and the time horizon is 2070. Some things to notice are the sharp increase of nuclear after the lifetime of some wind power parks reaches its limit and the associated reduction in CO2 emissions. Also, a sharp increase of the Uranium for nuclear fuel is seen accompanied by the need for more spent fuel deposit space. This aspect can affect in the future the public acceptance for nuclear. The figures in Appendix 1 are giving the basic results in terms of the dynamics of the system.

6. Conclusions In the framework of the EU energy policy the MESSAGE model was used to assess various scenarios of energy system development with time horizons in 2070. The main objective of the analysis was to assess first the optimality of the scenarios considered based on a simplex algorithm, second the impact of nuclear technology penetration such as emissions of CO2, need of Uranium for nuclear fuel, production of spent fuel for final repositories. The capability of the model to run several versions of a scenario and to find by a simplex algorithm the optimal ones is a definite advantage that allows not only scanning various potential dynamics but also to cluster the optimal ones and provide the decision maker with a cluster of optimal versions to select from based on economic, financial, etc. considerations. Thus the importance of the nuclear energy technologies penetration can be understood in the context of the whole energy system – the combined heat & power technologies and their district-heat facilities, support the increasing of nuclear energy technologies penetration in NEMix.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

References [1]

Dantzig, G. B., The Simplex method, Chapter 25, System Engineering Handbook, edited by R.Machol, McGraw Hill, New York, 1965.

[2]

Purica, I., Nonlinear Models for Economic Decision Processes, Imperial College Press, London, 2010.

[3]

An Introduction to MESSAGE, IAEA, Vienna.

[4]

MESSAGE, Model for Energy Supply Strategy Alternatives and their General Environmental Impacts, User Manual, IAEA, June 2007.

[5]

ISPE Study code 8118/2014-1-S0071921-B2, Bucharest, December 2014.

[6]

Directive 2003/87/EC of the European Parliament and of the Council of 13 October 2003 establishing a scheme for greenhouse gas emission allowance trading within the Community and amending Council Directive 96/61/EC.

[7]

Directive 2009/29/EC of the European Parliament and of the Council of 23 April 2009 amending Directive 2003/87/EC so as to improve and extend the greenhouse gas emission allowance trading scheme of the Community.

[8]

Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources.

[9]

Decision no.406/2009/EC of the European Parliament and of the Council of 23 April 2009 on the effort of Member States to reduce their greenhouse gas emissions to meet the Community’s greenhouse gas emission reduction commitments up to 2020.

[10]

IAEA Tools and Methodologies for Energy System Planning and Nuclear Energy System Assessments, IAEA, August 2009, https://www.iaea.org/OurWork/ST/NE/Pess/ass ets

[11]

http://www.transelectrica.ro/sistemulenergetic-national

[12]

http://www.anre.ro/ro/energieelectrica/rapoarte]

[13]

http://media.cnssnc.ca/nuclear_info/candu_performance.html

[14]

http://www.elcen.ro/

[15]

https://www.iaea.org/OurWork/ST/NE/Pess/

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Appendix 1 Basic MESSAGE result for case code [NES3d5eh2s200v23Ny70] (Objective function = 689955 [US$’00] (MINimum)/ Elapsed time: 187h27m)

Figure 1: Total energy demand

Figure 2: Energy production

Fig. 3: Total energy production by competitors in NEMix and nuclear & renewable contribution

Fig. 4: Nuclear new installed capacities [MW] ___________________________________________________________________________________________________________ E. Banches, I. Purica: “Long Term Assessment of Nuclear Technology Penetration using MESSAGE – The Case of Romania”, pp. 31–38

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Fig. 5: Storage and re-distribution system (HydroPS & district heat distribution), [MWyr]

Fig.6: Re-Distribution system share in NEMix vs. competitor energy share production, [share]

Fig.7: Evolution of CO2 emissions, [kton]

Fig.8: Nuclear Electricity production, [MWyr]

Fig.9: Cumulative Uranium consumption, [ton]

Fig.10: Spent nuclear fuel sent annually to intermediate wet storage, [kton]

Fig.11: Volume of spent nuclear fuel in intermediate dry storage, [kton]

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Appendix 2 CASE STUDY CODES code

Meaning/ comments

[NESαdβehϑsγδεvζyθ] NES 1 2 α 3

General code Nuclear Energy System – option/ scenario Only 2 CANDU units (existing ones + life extension) 4 CANDU units – national energy strategy in force 2013-2020 (existing ones and those in construction/after 2019 + life extension) 4 CANDU units (existing ones and those in construction/after 2019 + life extension)+ NEW NPP (Gen III+: LWR/ PWR/after 2035) + Gen IV(ALFRED)/after 2080 (not yet in the model until 2070) Discount rate of investments 5% 8% 10% Scenario for the evolution of the demand of electricity and heat Reference Demand Low Demand High Demand Specific options Option to use support technologies & systems for energy transport storage & distribution NO bounds of: HWMixD*, HWB**, El.HeatNucl*** NO bounds of HWMixD and HWB, All additional innovative technologies are bounded Option to use tax of CO2emmisions 10 [USD/tone CO2] 5 [USD/tone CO2] 30 [USD/tone CO2] Option to use a minimum contribution of renewable energy in mix NO bounds % of Renewable in the Total Electricity Production Mix, considered “fix/year” : 31,33,34,34,35,35,35,35,35,38…; last year:2035 Version of new nuclear investment cost vs. reference scenario and innovative scenario Reference investment cost for Nuclear +18% Reference investment cost for Nuclear -10% Reference investment cost CANDU34 New Innovative System proposal to use Nuclear and Renewable electricity pick production in power the HWB for District Station (not in included in the model for the preliminary report) Last year of modelling the CASE STUDY in MESSAGE 2070 is the last year of modelling the CASE STUDY in MESSAGE 2050 is the last year of modelling the CASE STUDY in MESSAGE

d 5 8 10

β eh****

1 2 3

ϑ s γ

1 2 3

y δ

0 1 2

δ ε

0 5

ε v

23N 25N 29N 23Q;25Q;29Q

y y

* ** *** ****

70 50

HWMixD – district station for Hot Water Mix and Distribution (HeatInFlow) HWB – Hot Water Boilers, support for peak of gas CHP over-load heat requirements in the system El.HeatNucl – Electric Heat for cooking and for domestic boilers (solution not yet developed in Romania) ieh – refers to the option to limit the optimisation process as integer only until 2050 – in order to extract faster partial results if the optimization process take over 100 hours

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.20160105

The Role of Nuclear Power in the Future Energy System Jarkko Ahokas*, Kristiina Söderholm Fortum Power and Heat Oy Keilaniementie 1, 00048 Espoo, Finland; jarkko.ahokas@fortum.com, kristiina.soderholm@fortum.com

Abstract

1. Introduction

Currently widely accepted consensus is that greenhouse gas emissions produced by the mankind have to be reduced in order to avoid further global warming. The European Union has set a variety of CO2 reduction and renewable generation targets for its member states. Nuclear power is a low carbon energy technology and it seems to have many attractive properties in regards to carbon free energy system aspirations. The study presented in this paper is condensed version of a Master's thesis [1]. In this study, the role of nuclear power in the future Nordic energy system is the main research question. The basic characteristics and properties of nuclear power are presented in the study. Also other selected carbon free energy production methods are presented, as well as energy storages. Carbon free energy system in Nordic region is modelled with different scenarios. The model is used to analyse whether the constructed energy system consisting only renewable and nuclear generation can function. This paper summarizes the aforementioned study and presents the main results.

Climate change mitigation has emerged as one of the most popular and important targets for mankind. Widely accepted consensus is that greenhouse gas emissions have to be reduced to avoid, or even halt, further global warming. The energy sector has an important role in reducing these emissions, especially carbon dioxide emissions. Globally, fossil fuels dominate the energy sector and they are likely to do so in the foreseeable future. Emerging economies, such as China and India, are unlikely to abandon fossil fuelled energy generation anytime soon. In more advanced economies, a transition to different low carbon technologies, such as renewables, is being encouraged and even demanded. The energy system in the Nordic countries is already one of the most carbon free in the world [2]. The majority of the CO2 emissions in the Nordic power sector come from coal, peat and natural gas power plants in Finland and Denmark. Finland generated around 46% and Denmark 33% of the 67 million tonnes of CO2 that the Nordic power sector generated in 2010 [2]. The overall share of renewables, mainly hydropower, in Nordic power generation was around 60% in 2010. According to various IEA scenarios, the share of renewables in Nordic power generation will increase to 80% by 2050. In this study, the Nordic energy sector is assumed to be carbon free by 2050 and this is achieved with using nuclear power together with a large share of renewables [1].

Keywords:

Nordic energy system; Energy system model, Nuclear power

Article history:

Received: 17 July 2015 Revised: 15 October 2015 Accepted: 29 January 2016

Nuclear power is the world's second-largest source of low carbon electricity after hydropower and in OECD countries it is the largest source of low carbon electricity [3]. It is important to consider the role of nuclear power in the future carbon free energy system. The model in this study calculates generation mixes, produced and consumed total energies and hourly energy balances in the Nordic energy system. Electricity prices, market mechanisms or detailed production and

___________________________________________________________________________________________________________ J. Ahokas, K. Söderholm: “The Role of Nuclear Power in the Future Energy System”, pp. 39–46

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

construction costs of the generation fleet are not considered. Rudimentary cost comparisons between different generation technologies are performed but these are only indicative as the aforementioned financial aspects are not in the scope of this study. The different scenarios presented are not forecasts, rather they present alternative targets and avenues for a carbon free energy system. Generation mixes used in the thesis are either based on the literature or they are chosen somewhat arbitrarily to construct and study different carbon free energy systems. Future scenarios are set for the year 2050 following the various IEA scenarios and targets found in the Nordic Energy Technology Perspectives 2014 [2].

2. Basic characteristics of nuclear power In regards to carbon free energy system aspirations, nuclear power seems to have many attractive properties. The basic characteristics and properties of nuclear power are presented more widely in the study and include e.g. current and future reactor technologies, resources and resource efficiency, emissions, and flexibility of nuclear power plants [1]. The impact of nuclear power to energy security and security of supply is presented below.

2.1. Energy security and security of energy supply It is clear that security of energy supply and the continuous availability of energy at an affordable price is invaluable for society. It not only provides essential services for production, communication and trade but is also invaluable in maintaining basic human needs such as heating, ventilation and food and water supply. Historically the introduction of nuclear power to the

country's electricity generation mix has improved the energy security of that country. This can be seen, for example, in the evolution of the generalized Simplified Supply and Demand Index (SSDI) which indicates the security of supply for a defined region. Figure 1 shows the SSDI values in selected OECD countries from 1970 to 2006. For example, the United Kingdom's switch from coal to gas and the introduction of nuclear programmes in Finland, France, Sweden and the United States improve the value of the SSDI [4]. Generally, the country's security of energy supply and energy security improve with the introduction of nuclear power and decreases often relate to increases in imports. Security of energy supply and energy security can be divided into external and internal dimensions, seen in Figure 2. Nuclear power has several characteristics which improve the energy security of a certain nation, especially regarding the external dimension. Few examples as how the nuclear power can improve the energy security are its positive attributes in following categories: geopolitical risk, safety and adequacy of international infrastructures, price stability and operational reliability. Uranium resources are available from diverse sources, both geographically and politically, which in turn lessens the geopolitical risk of acquiring nuclear fuel or uranium [5]. Global nuclear fuel supply chain has yet to experience a serious disruption and nuclear power involves long lead times allowing the nuclear industry to have ample time to anticipate and respond to changes in uranium demand. Generating costs for nuclear are less sensitive to changes in fuel costs than those of fossil fuelled generation and nuclear power can increase the price stability in the region. Nuclear power plants also traditionally have high capacity factors which suggests good operational reliability [3].

Figure 1: The evolution of the SSDI in selected OECD countries [4] ___________________________________________________________________________________________________________ J. Ahokas, K. Söderholm: “The Role of Nuclear Power in the Future Energy System”, pp. 39–46

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 2. Dimensions of energy supply security [4]

3. Energy system model Globally, the current energy system is based on large, centralised generation using mainly fossil fuels. The future low carbon energy system will have greater diversity of technologies and fuels, and especially more renewables will be used. In addition to renewables, nuclear power could be a major contributor to the decarbonisation of the electricity supply. Renewable energy output has a tendency to fluctuate depending on the weather. Today's energy systems uses mainly hydropower and fossil fuelled power plants as regulating power to stabilise the power grid and to meet the supply and demand challenges. In this study, a future energy system is envisaged to be carbon free and thus, the share of fossil fuel dependent energy generation is significantly lower, or zero, in the system. Decarbonising the energy system requires alternative grid stabilising methods and technologies. Increased volumes of variable production from wind and solar will highlight issues related to regulating power. In Nordic countries, the large share of hydropower will help the transition and will become increasingly valuable in regulating electricity systems in these countries and Northern Europe in general [2]. The Nordic countries have set their ambitions and energy targets on a Carbon-Neutral Scenario (CNS), in which CO2 emissions in the region are reduced by 85% by 2050 compared to emission levels in 1990. Within this strategy, some Nordic countries would achieve a carbonneutral energy system by 2050 [2]. The future energy system in the Nordic countries in 2050 is studied and analysed by constructing a basic model

representing the energy system in the Nordic countries [1]. The model uses real 2013 Nordic load data from Nord Pool Spot which is then scaled up to estimated 2050 consumption levels [6]. The IEA estimates that Nordic countries will have a load between 430 and 450 TWh in 2050 and that Nordic countries will become net exporters of electricity [2]. Every hour of the year has a load value, i.e. how much electricity was consumed in any given hour in the Nordic countries in 2013. The load demand is then satisfied with different types of electricity generation in a particular order. In this study, the energy system has electricity generation from the following sources: hydropower, wind power, nuclear power and biomass fuelled generation. The Nordic grid connections between Denmark, Finland, Norway and Sweden are considered to be more than adequate in 2050, meaning that Nordic countries and their electricity grid is considered as a single unit. International grid connections, possible heat storage and the geographic distribution of wind generation are not directly analysed in the model. In all future scenarios, both the IEA's and the ones analysed in the model, growth in electricity generation outpaces electricity demand in the Nordic countries, implying a rise in the net exports from the Nordic region. The model logic is shown in Figure 3. The generation mix, capacities and shares for different generation sources vary depending on scenario. The generation mix used in the model is based on the IEA's scenarios found in their publication Nordic Energy Technology Perspectives 2013 (NETP) but with little modifications [2]. The Base scenario, discussed in chapter 3.1, in the model is based on IEA's Carbon-

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Neutral Scenario (CNS) found in the NETP. The CNS scenario still has some fossil fuelled generation, opposed to none in the model, and about 2 GW of solar capacity. According to the IEA, solar generation is not significant in Nordic countries in 2050. This may prove to be wrong, but one purpose of the model was to consider whether system consisting of large amount of intermittent renewable generation, traditional renewable generation and nuclear can function. In the model, wind power represents intermittent renewable generation and solar generation is not modelled. Hourly output data in 2013 for the Nordic wind power was extracted from various sources. This allows to use realistic wind profile in the model. Hourly installed capacities and peak load hours were calculated and modified for the year 2050 in order to take into account better windmills and turbines, larger wind farms and the possibly larger share of offshore wind power. Nordic wind power has 3 265 peak load hours in the model.

Nordic electricity generation is currently dominated by hydropower and hydropower will also be the largest generation source in 2050. In the model, a portion of the total hydroelectric power capacity is dispatchable and the remaining portion non-dispatchable. Some of the Nordic hydropower plants are run-of-river plants with limited water basins and these types of hydropower plants have limited manoeuvring and adjusting capabilities. According to the NETP, around 35 GW of the 60 GW of hydropower capacity in the Nordic countries in 2050 could be considered dispatchable [2]. Dispatchable hydropower is the flexible element in the model and it responds to the possible electricity deficits with no other limitations than the maximum capacity it has. The nondispatchable portion of the hydroelectric power produces electricity evenly over the year. At the time of writing, only Finland and Sweden of the Nordic countries have nuclear generation; 2.7 GW and 9.5 GW respectively. However, the two majority

Figure 3. Illustration of model logic [1] ___________________________________________________________________________________________________________ J. Ahokas, K. Söderholm: “The Role of Nuclear Power in the Future Energy System”, pp. 39–46

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

operators of nuclear power plants Ringhals and Oskarshamn in Sweden have announced intentions to close down some production capacity. The projected loss of capacity (around 3 GW) is not considered in this study. It is unlikely that Norway or Denmark will build nuclear power generation in the future. The IEA estimates in the Nordic Energy Technology Perspectives that the capacity for nuclear generation in Sweden will remain the same in 2050 as it is at the time of writing but they anticipate that Finland's nuclear generating capacity will rise from the current level to 6.4 GW [2]. These same assumptions are used in the Base scenario of this study, meaning that in 2050 Nordic nuclear generating capacity is estimated to be 15.9 GW. In the model, all the Combined Heat and Power (CHP) generation is estimated to be biomass fuelled in 2050. Fossil fuelled CHP is non-existent in 2050. Combined Heat and Power generation is assumed to follow the heat demand in the Nordic countries and is not flexible in the model. CHP plants feeding different industries are assumed to run evenly throughout the year while district heating is used mainly only in the winter months. Therefore, it is assumed that CHP produces double the amount of energy in winter compared to summer. The model basically shows hourly energy balances and system is considered to function properly when there are no deficit hours present. Simply put, the energy system in the model has to produce more energy than it consumes. All the balancing needs in the Nordic scenarios are handled with hydro power.

3.1. Base scenario In this Base scenario (year 2050), the Nordic energy system has about 76 % renewable generation and 24 %

nuclear and the whole system is carbon free. The only intermittent element in the system is the large share of wind power. A portion of hydropower is considered nondispatchable and acts as a base load power source in the system. The total energy of the nuclear power is divided evenly throughout the year reflecting its role as a base load power source in the energy system. There is, however, the possibility to use nuclear power as a flexible power source, but this is not analysed in the Base scenario. CHP is assumed to generate 100% more energy in winter months (November to April) than in summer months and the output is stable. The energy generated by CHP in the model is electricity; heat energy or heat flows in the energy system are not included in the model. Fixed hydropower, nuclear power and biomass based CHP all generate electricity evenly throughout the year. Finally, the remaining hydropower is dispatchable and is the only flexible form of power generation in the Base scenario [1]. The overall electricity generation increases by 30% compared to 2013 levels, from 384 TWh to 506 TWh and the generating capacity with about 30 GW. These values are on the same scale as presented in the NETP. In the Base scenario, the constructed energy system has a generation surplus of 76 TWh. The IEA estimates that, depending on scenario, the Nordic countries will have a combined electricity exports between 40 TWh and 100 TWh [2]. In Europe, intermittent wind and solar power are becoming more popular and it is intended that fossil fuelled power generation will be abandoned in many countries, thus the need for regulating power capacity will increase. The large hydropower capacity in the Nordic system is valuable and can help to regulate the electricity systems in Northern and Central Europe.

Figure 4: The generation and load month to month in the Base scenario [1] ___________________________________________________________________________________________________________ J. Ahokas, K. Söderholm: “The Role of Nuclear Power in the Future Energy System”, pp. 39–46

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

The monthly generation profile for the Base scenario is presented in Figure 4. The black line represents the load in Nordic countries. The different electricity generation sources are, from bottom to top, in the same order as in the model [1]. As can be seen from the figure, wind generation and dispatchable hydropower are the only intermittent electricity generation sources. Non-dispatchable hydro, nuclear and CHP generation have no hourly variation. Nuclear generation, non-dispatchable hydro and biomass CHP provide the base load generation to the energy system. The Nordic countries have more load in the winter months and therefore dispatchable hydropower is needed to compensate for deficits. Overall, the energy system constructed in the Base model has no deficit hours over the whole year, meaning that there is always at least the same amount of electricity generation as there is consumption. Some specific situations were also analysed by modifying the Base scenario. Modifying the levels of wind generation allows to see whether the energy system can handle low wind hours. Zero wind situation is unrealistic but shows that remaining generation sources in the Base scenario can generate enough electricity to satisfy the demand. This situation highlights the role of different electricity sources in the system. This also highlight the importance of having versatile generation mix. Dispatchable hydropower is used when needed while nuclear power provides reliable and stable electricity regardless of weather conditions. Nordic hydropower generation is also somewhat affected by the water levels. The Base scenario was also modified by replacing 15.9 GW of nuclear power with either more wind power, biomass fuelled generation or combination of these two generation sources. Aim was to generate as much energy in the system as in the non-modified Base scenario. In order to do so without nuclear, a total of 73 GW of wind power capacity was needed, 33 GW more than in the Base scenario. For biomass fuelled generation, 25 GW of

additional biomass fuelled capacity would be needed to replace nuclear generation. This would bring the biomass fuelled generating capacity to 37 GW and the strain on the biomass production and fuel transportation system would be greatly increased.

3.2. Low hydro scenario The energy system in this scenario is not strictly based on any real life country, region or any IEA scenario. However, as the model uses real data for the Nordic electricity load also this scenario uses the Nordic load pattern [1]. This means that the largest loads are in the winter months. This scenario is an illustration of a situation with limited hydro power resources and thus, is not relevant in Nordic energy system. The purpose of this scenario was to construct a carbon free energy system in 2050 without a large amount of hydropower to see if a system containing only wind, biomass and nuclear power together with energy storage can form a working energy system and to see whether hourly balancing can be achieved with other elements of the system. If the energy system has large amount of hydropower capacity, the challenges that arise from large shares of intermittent renewable generation, such as wind and solar, are more easily mitigated. The Nordic energy system has significant amounts of hydropower generation available and it has the largest impact on the Nordic energy system. In this scenario, the share of hydropower compared to other generation sources is insignificant. The Low Hydro scenario includes electricity generation from wind, nuclear and biomass. Hourly balancing and flexibility is ensured by incorporating energy storage into the energy system. The shares and capacities for different generation sources and storage are selected in such a way that there are no deficits in the system after energy storage. Wind, nuclear and biomass capacity values were chosen in such a way that the system generates around the same

Figure 5: Electricity generation mixes. Energy storage not included [1] ___________________________________________________________________________________________________________ J. Ahokas, K. Söderholm: “The Role of Nuclear Power in the Future Energy System”, pp. 39–46

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Table 1: Capital cost and O&M costs of different technologies [7] Technology

Nominal capacity [kW]

Capital cost [$/kW]

Fixed O&M [$/kW-yr]

Variable O&M [$/MWh]

Advanced nuclear

2 234 000

5 530

93.28

2.14

Wind, onshore

100 000

2 213

39.55

Wind, offshore

400 000

6 230

74.00

Biomass, BFB

50 000

4 114

105.63

5.26

Biomass, CC

20 000

8 180

356.07

17.49

Figure 6: Total capital cost and O&M costs in one year for different technology combinations. Here 1$ = 0.833€

amount of electricity as the IEA estimated in the Nordic Energy Technology Perspectives [2]. If the energy system had surplus energy in any given hour that energy was stored in long and short term storage facilities. Energy stored in these facilities was utilized in the hours when electricity generation from wind, nuclear and biomass was not enough to satisfy the demand. The constructed energy system has no deficit hours during the year and shows that a carbon free energy system can function without hydropower. Figure 5 presents electricity generation mixes in 2013 and in the Base and Low hydro scenarios. Nuclear energy has an important role in this system as it produces almost 49% of the total energy while its generating capacity is 33% of the total capacity connected to the grid. Replacing this much nuclear energy with other low carbon technologies would

require significant investments not just in generating capacity but also in technologies involved in balancing the grid. For example, solar and wind power do not introduce rotating inertia naturally to the grid while nuclear power does. Power grid inertia is one parameter the synchronized operation of the grid is based on and it determines the immediate frequency response. If the overall inertia in the power grid is low, the grid frequency reacts nervously to sudden changes in generation and load patterns. To put it simply, inertia is the grid's resistance to change [1]. Basically all thermal power plants, whether they are fossil, nuclear or biomass fuelled, introduce rotating inertia to the power grid as their synchronous machines, generators and turbines, are connected to the grid directly. This rotating inertia is essential as it helps balancing the grid.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

3.3. Cost comparison Financial considerations were not the main focus of this study but some comparisons were made. Table 1 gathers capital cost and operation & maintenance costs of different technologies with a certain nominal capacity. The information in Table 1 is used to calculate the total capital cost and O&M costs in one year for different technology combinations. These combinations produce the same amount of energy. The amount of energy selected for this comparison is the amount that seven nuclear power plants (total of 15.6 GW) from Table 1 produce over the year. This is about the same as in the Base scenario (15.9 GW of nuclear power). Figure 6 shows these costs. The second and third columns have 15 GW of wind and 14 GW of biomass power. Each column produces the same amount of energy and Figure 6 shows how much it would roughly cost to replace 15.6 GW of nuclear power with alternative low carbon technologies. This is just a rudimentary comparison as financing, interests and market mechanisms are not considered. Still, it shows that nuclear power is cost effective option when comparing purely total capital costs and operation & maintenance costs.

abandon the use of fossil fuelled generation, nuclear power can replace it as a low carbon technology while maintaining grid stability. The model and scenarios in this study show that it is possible to form a functioning, 100% carbon neutral energy system by combining greenhouse gas free renewables and nuclear energy. Wind power has large share of the total produced energy, but its generation profile fluctuates. Production from wind turbines is highly dependent on the prevailing weather conditions. Nuclear power produces electricity evenly and predictably throughout the year. Biomass fuelled CHP also produces electricity and heat evenly throughout the year. Their roles are equally as important as hydropower is; however wind and hydropower capacities cannot satisfy the consumption demand by themselves. Out of these generation sources, nuclear power produces the most energy per installed capacity.

References [1]

Ahokas Jarkko, 2015. The role of nuclear power in the future energy system. Master's thesis. Lappeenranta: Lappeenranta University of Technology, School of Technology, Energy Technology. Available at: http://urn.fi/URN:NBN:fi-fe201504222785

[2]

IEA, 2013. Nordic Energy Technology Perspectives. OECD/IEA. Paris: IEA Publishing. ISBN 978-82-92874-24-0. License: http://www.iea.org/t&c/termsandconditions/

[3]

IEA, 2014. World Energy Outlook 2014. OECD/IEA. Paris: IEA Publishing. ISBN 978-92-6420805-6. License: http://www.iea.org/t&c/termsandconditions/

[4]

NEA, 2010. The Security of Energy Supply and the Contribution of Nuclear Energy. OECD Nuclear Energy Agency. ISBN 978-92-64-09634-9.

[5]

IAEA, 2014. Nuclear Technology Review 2014. Vienna: International Atomic Energy Agency (IAEA). Available at: https://www.iaea.org/About/Policy/GC/GC58/D ocuments/ Nord Pool Spot, 2015. Electrical power market, Nord Pool Spot AS. [web page accessed 10.7.2015]. Available at: http://www.nordpoolspot.com/#/nordic/table

4. Conclusions All the scenarios constructed in the model have working energy systems where there are no deficits after all the different elements of the systems have made their contributions. A broad general conclusion from the analyzed scenarios is that nuclear power provides stable energy generation for the system without direct greenhouse gas emissions. Nuclear power generally has high capacity factors and peak load hours and its high stable electricity output helps to offset the intermittent profile of wind power. Nuclear power offers grid management services which traditionally are not offered by wind or solar power. These include primary and secondary frequency control, predictable and controllable availability and rotating inertia. All of these are needed in order to have a functioning energy system making an energy system with 100% share of renewables hard to realise. As of now, these grid management services are performed by fossil fuelled generation, nuclear generation and hydropower. The Nordic countries are in the fortunate position that they can utilize the vast hydropower capacity of Norway. Hydropower is an excellent form of power generation for regulating power and for grid balancing purposes, but there is natural limit for hydropower capacity. As future energy systems

[6]

[7]

U.S. Energy Information Administration (EIA), 2014. Updated Capital Cost Estimates for Utility Scale Electricity Generating Plants. Available at: http://www.eia.gov/forecasts/capitalcost/

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.20160106

Application of Foamy Mineral Oil Flow under Solution Gas Drive to a Field Crude Oil Bashir Busahmin1*, Brij Maini2, Hossein Hejazi2, Amin Sharifi2, Mohammad Tavallali3 1

Institut Teknologi Brunei-University Jalan Tungku Link Gadong Brunei Darussalam, BE 1410; bashir.abusahmin@itb.edu.bn 2 University of Calgary, Canada 3 IHS, Canada

Abstract

1. Introduction

Heavy oil flow in the form of foamy oil under solution gas drive is widely observed in many Canadian reservoirs. Despite the importance of such phenomenon, complexity involved in foamy oil flow in porous media is not well understood. Series of numerical simulations were performed to model experiments that were carried out in a two meter long Sand pack to investigate the conditions required to increase oil production under solution gas drive mechanism. Through these experiments the solution gas drive performance at different depletion rates were analyzed.

Foamy oil flow is a process which against many experimental and theoretical investigations still is not completely clear in many aspects [1, 2, 3, and 4]. An important question in foamy oil flow is whether or not the behaviour observed in the laboratory scale primary depletion tests can be simulated with available multiphase flow simulators. Among the various commercial reservoir simulators, CMG-STARSTM is considered the most versatile for modelling foamy oil flow. As discussed by Bayon et al. [5], it permits modelling of foamy flow by defining separate components to represent dissolved gas, dispersed gas, and the free gas. Artificial chemical reactions with associated reaction kinetics are defined to model the rate processes involved in formation of bubbles and separation of dispersed gas from the oil to form free gas [6, 7].

Creation of foamy heavy oil is thought to be responsible for higher recovery factors compared to what is expected from the conventional solution gas drive theory. However, the complex nature of foamy oil and different transport parameters are yet to be understood.The results of this study can be used to numerically model foamy-oil mechanism in heavy oil reservoirs. Furthermore, the results can be applied for reservoir production optimization as well as management. A new model has been developed using commercial numerical simulator, computer modeling group, (CMG-STARSTM). By using the experimental data, different experimental production histories have been matched. Effect of different parameters such as fluid and reservoir properties and depletion rate on foamy oil recovery have been evaluated. The results reveal that despite many difficulties, foamy oil flow through porous media can be numerically modeled. However these models will strongly depend on a good understanding of many different parameters including rock-fluid interaction, as well as the depletion rates. Given the complex nature of such systems, this numerical model can be used to simulate and predict the oil and gas production from heavy oil reservoirs under foamy oil conditions.

Keywords:

Article history:

Modelling and simulation; Multi-phase and multi fluid flows; Codes numeric and their developments and/or improvements Received: 08 July 2015 Revised: 23 July 2015 Accepted: 29 January 2016

This manuscript presents the simulation model implemented in CMG-STARSTM to history match the performed experimental results, and address the question of whether or not such commercial reservoir simulators can be used to history match foamy solution gas drive tests. An important idea in this context is to determine whether the simulation parameters tuned to history-match a specific experiment can truly represent the rock-fluid properties of the system, i.e. is it possible to simulate different depletion tests that have been ran under same rock-fluid system, with a unique set of parameters?

2. Reservoir simulation model A one dimensional (1D) model was used to simulate the primary depletion tests using two types of methane saturated oils; mineral oil and crude oil. Since the sandpack was placed horizontally for all the experimental runs, the flow direction is assumed to be horizontal. A

___________________________________________________________________________________________________________ B. Busahmin, et al: “Numerical Modelling of Foamy Oil Flow under Solution Gas Drive in Heavy Oil Reservoirs”, pp. 47–53

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Producer

100 grid blocks No flow boundary

…….…... …….…... …….…... Figure 1: Schematic of simulation model

Table 1: Sand pack properties

Table 2: Dead oil properties

Property

Value

Property

Crude Oil

Mineral Oil

Sand-pack length, cm Cross-sectional area, cm2 Pore volume, cm3 Porosity, % Absolute permeability, Darcy Overburden pressure, psi Sand grain size, mesh Pore space compressibility, psi-1

200 23.82 1619.38 34 24 1000 30-50 6.49 x 10-6

Density, kg/m3

936 2800 600 28 800 6.9 x 10-6

896 1876 441 245 700 6.8 x 10-6

Viscosity at 21◦C, cp Viscosity at 40◦C, cp Viscosity at 50◦C, cp Molecular weight, kg/mol Compressibility, psi-1

Table 3: Fluid properties used in reservoir simulation Property

Crude Oil-CH4

Mineral Oil-CH4

Live oil density, kg/m3

928

891

Live oil viscosity at 21◦C, 23◦C cp

1300

1080

Solution GOR, std. scm3/scm3

Density of gas at SC, g/cm3

0.00067832

Density of water at SC, g/cm3

1.070223

Oil compressibility, 1/psi

4.37 x 10-6

4.52 x 10-6

Water formation volume factor

0.999305

Water compressibility, 1/kPa

4.74x 10-7

Water viscosity, cp Initial pressure, psi Bubble point pressure, psi

1.26459 525 500

1.26459 520 500

“producer” well was defined at one end of the model while the opposite end was a no-flow boundary. In the model, 100 grid blocks were defined along the length of the sand-pack. Therefore, the number of grid blocks in “X” direction is 100, and in “Y” and “K” directions are one. The average grid block size was about 2 cm by 4.88 cm by 4.88 cm. Figure 1 shows the schematic of the simulation model.

3. The analysis of a heat accumulator’s operation The rock and fluid properties of the physical model that experiments were conducted are used in the numerical model as presented in Table 1 to 3. All the parameters

such as porosity (34%) and permeability (24 Darcy) were kept constant for all depletion rate tests.

4. Pressure and production data In the experiment, the pressure at the production well was set by the back-pressure regulator and was an independently controlled parameter. In the numerical model, the pressure at the production well of the sandpack (P1) was considered the bottom hole production well pressure (BHP), which was entered into the simulator as an operating constraint. The oil and gas production data at different depletion rates were used in the history match analysis of the experiments.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

5. Relative permeability curves In order to achieve a reasonable history match for the primary depletion tests conducted at different depletion rates, the relative permeability curves had to be adjusted individually for each test. The relative permeability curves were based on Corey’s exponent model for oil and gas as follows: = =

(1) 1−

(2)

(3)

Table 4 presents the summary of the relative permeability and other parameters used for history matching the depletion tests. Also reaction rates for aggregation of dispersed gas to bubbles and then bubbles into free gas phase are also tuned in simulation model (RF1 and RF2 respectively)

6. Simulation results Four depletion rates were designed for each oil type during experimental studies, which ranges from 0.021 to 0.434 psi/min. Table 4 shows different depletion rates

as applied on the sand-pack. In order to model each test with multi-phase flow simulator relative permeability end points and exponents in the power-law model are used as matching parameters, and other parameters presented in Tables 1-3 are kept fixed with high level of confidence for their values. Figures 2 and 3 present the experimental and simulated values of cumulative oil production at different depletion rates for mineral oil. It is readily apparent that the simulated results can capture the general trend of oil production history and the final recovery values reasonable well. The cumulative gas production for these four tests is compared in Figures 4 and 5. Here also the production trends and the final values are reasonably well matched. So, it is apparent that the simulation model can be tuned to provide decent matches of the cumulative production. However, the cumulative production often hides small differences in the behavior. The production rate is usually a more sensitive measure of the goodness of history matches. Figure 6 presents a comparison of the experimental and simulated oil rates for the fastest depletion. The start of oil production is fairly well matched but the experimental data shows a much higher peak and faster decline. It was not possible to obtain a very good match between the experimental and simulated oil rates for this test. The gas rate match is shown in Figure 7. The gas rate was matched somewhat better than the oil rate. However, the simulated rate shows none of the wild swings in gas rate observed in the experiment.

Table 4: Summary of the parameters used for history matching the eight tests Parameters

Mineral Oil

Crude Oil

Depletion rate (psi/min)

0.406

0.247

0.086

0.021

0.434

0.226

0.048

0.023

Krwro

Krocw

0.128

0.05

0.008

0.003

0.18

0.043

0.007

0.001

Krgro

0.009

0.008

0.4

0.08

0.004

0.008

0.4

0.08

Krocg

Swcon

0.08

Sgcon

Swc

0.1

Sgc

0.02

0.05

0.04

0.025

0.05

0.04

Sorg

0.3

Soirw

0.3

1.5

2.5

RF1

0.019986

0.009993

0.00045

0.000138

0.018986

0.0001978

0.00025

0.000147

RF2

0.00095

0.00048

0.00045

0.00037

0.00085

0.0007

0.00045

0.0004

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________ Sim.1: DR = 0.406 psi/min Sim.3: DR = 0.086 psi/min

Exp.2: DR = 0.247 psi/min Exp.4: DR = 0.021 psi/min

450

400

350

350 Oil Production (cm /min) (cm3)

Oil Production (cm /min) (cm3)

Exp.1: DR = 0.406 psi/min Exp.3: DR = 0.086 psi/min

300

Sim.2: DR = 0.247psi/min Exp.4: DR = 0.021 psi/min

250 200 150

100

50 0

0 0

4000

8000

12000

16000

20000

24000

4000

8000

12000

Time (min)

Figure 2: Experimental cumulative oil production versus time for depletion tests at different rates with methane saturated mineral oil at room temperature, and at 3447 kPa

Exp.1: DR = 0.406 psi/min Exp.3: DR = 0.086 psi/min

16000

20000

24000

Time (min)

Figure 3: Simulation of cumulative oil production against time for depletion tests at different rates with methane saturated mineral at room temperature, and at 3447 kPa

Sim.1: DR = 0.406 psi/min Sim.3: DR = 0.086 psi/min

Exp.2: DR = 0.0247 psi/min Exp.4: DR = 0.021 psi/min

14000

12000

Sim.2: DR = 0.0247 psi/min Sim.4: DR = 0.021 psi/min

Gas Production (std.cm3)

Gas Production (std.cm )

14000

10000

8000

6000

10000

8000

6000

4000

2000

0 0

4000

8000

12000

16000

20000

24000

4000

8000

Time (min)

Figure 4: Experimental cumulative gas production against time for methane saturated mineral oil at room temperature, and at 3447 kPa

Sim Oil rate

12000

16000

20000

24000

Time (min)

Figure 5: Simulation of cumulative gas production against time for methane saturated mineral oil at room temperature, and at 3447 kPa

Exp Oil rate

Exp gas rate

Sim gas rate

1.2

Gas Production Rate, (cm3/min) Std. cc/min

Oil Production Rate, (cm3/min mL/min

140

0.8

0.6

0.4

0.2

120 100 80 60 40 20 0

0 0

200

400

600 800 Time, min

1000

1200

1400

Figure 6: Typical oil production rate vs. time for very fast depletion test for methane saturated mineral oil system

200

400

600

800

1000

1200

1400

Time, min

Figure 7: Typical gas production rate vs. time for very fast depletion test for methane saturated mineral oil system

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Sim oil rate

Exp oil rate

Exp gas rate

Sim gas rate

Gas Production Rate, (cm3/min) Std. cc/min

Oil Production Rate, (cm3/min) mL/min

1.40 0.03

0.02

0.01

1.20 1.00 0.80 0.60 0.40 0.20 0.00

0.00 0

4000

8000

12000

16000

20000

24000

4000

8000

12000

Figure 8: Oil production rate against time for a slow depletion test – methane saturated mineral oil system

Exp.2: DR = 0.13 psi/min Exp.4: DR = 0.022 psi/min

Sim.1: DR = 0.35 psi/min Sim.3: DR = 0.048 psi/min 450

400

350

350 Oil Production (cm3) (cm/min)

450

24000

Sim.2: DR = 0.13 psi/min Sim.4: DR = 0.022 psi/min

300

300 250 200 150

250 200 150

100

5000

10000

15000

20000

25000

5000

Time (min)

14000

12000

10000

8000

6000

4000

2000

0 5000

10000

15000

Sim.1: DR = 0.35 psi/min Exp.3: DR = 0.048 psi/min

"@0.048 psi/min"

14000

20000

25000

20000

0 25000

Time, min

Figure 12: Experimental of cumulative gas production against time for methane saturated crude oil at room temperature, and at 3447 kPa

Exp.2: DR = 0.13 psi/min Exp.4: DR = 0.022 psi/min

12000

10000

8000

@0.226 psi/min

Oil Production (cm3) (cm/min)

"@0.023 psi/mi"

15000

Figure 11: Simulation of cumulative oil production against time for methane saturated crude oil at room temperature, and at 3447 kPa

Cumulative Gas, std. cc

@0.434 psi/min

10000 Time (min)

Figure 10: Experimental Cumulative of oil production against time for methane saturated crude oil at room temperature, and at 3447 kPa

Cumulative Gas, std. (cm3)cc

20000

Figure 9: Gas production rate vs. time for slow depletion test – methane saturated mineral oil

Oil Production (cm (cm3) /min)

Exp.1: DR = 0.35 psi/min Exp.3: DR = 0.048 psi/min

16000

Time, min

6000

4000

2000

0 0

5000

10000

15000

20000

25000

Time (min)

Figure 13: Simulation of cumulative gas production against time for methane saturated crude oil at room temperature, and at 3447 kPa

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 8 shows the oil rate history match of a slow depletion in the mineral oil-methane system. The oil rate match is somewhat better in the slow depletion. The gas production rates are compared in Figure 9. Here the match is not as good. So, in general, it is extremely difficult to get very good history match for oil and gas production rates, even though the cumulative production history can be reasonable well matched. It should also be mentioned here that different simulation parameters had to be used to get the matches shown here. These included different relative permeability endpoints and different rate constants.

except that the experimental profiles show lot more fluctuations. Therefore it can be stated that it is possible to obtain reasonably good history matches by tuning the rockfluid parameters. However, with crude oil also, tests carried out at different depletion rates required different rock-fluid parameters. This limits the usefulness of such history matching in using the laboratory experiments to predict the field scale behavior. This is due to the complexity of the foamy oil flow in porous media which involves bubbly oil flow and sand production and cannot be easily modeled such as conventional oil reservoirs. It suggests more investigation is needed to understand the foamy oil flow to model the physics of the process.

In general it is not possible to history match experiments done at different rates with the same set of rock-fluid properties and rate constants. Therefore, the simulation parameters obtained by history matching a specific experiment cannot be used for predicting the behavior of depletions involving substantially different operating conditions.

7. Conclusions This paper presents the conclusions drawn from the simulation of primary depletion tests conducted in a two-meter long sand-pack to study the effects of several variables that influence foamy oil flow under solution gas drive. Conclusions drawn from the limited simulation work carried out to history match several depletion tests are also included.

Figures 10-13 present the experimental and simulated oil and gas production histories of the four solution gas drive tests carried out with the crude oil system. Here also, it was not difficult to obtain reasonable histories matches for the cumulative oil and gas productions. However, the real test of the history match is in comparing the rates of oil and gas production.

1. The solution gas drive performance in all systems declined with decreasing rate of pressure reduction at the production end.

The experimental and simulated oil production rates for the fastest depletion are compared in Figure 14. As in the case of mineral oil, the high peak in oil production rate observed in the experiment is not well matched by the simulation. The oil rate declines more rapidly in the experiment than in the simulation. The gas production rates are compared in Figure 15. The gas rates are quite well matched. The matches for the slow experiments are shown in Figure 16-17. Here the matches are quite good

Exp oil rate

2. The solution gas drive recovery factor, in heavy oil systems, depends strongly on the pressure drawdown (as the driving force for the oil production) that develops in the system as a result of pressure reduction or fluid withdrawal at the production port.

Sim oil rate

0.8

Sim gas rate

1000

2000

0.7

16 Gas Rate (cm3/min)

0.6

Oil Rate (cm /min)

Exp gas rate 20

0.5 0.4 `

0.3 0.2

14 12 10 8 6 4

0.1

0 0

1000

2000

3000

4000

5000

Time (min)

Figure 14: Oil production rate versus time for a fast depletion test for methane saturated crude oil system

500

1500

2500

3000

Time (min)

Figure 15: Typical gas production rate vs. time for fast depletion test for methane saturated crude oil system

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Sim oil rate

Sim. gas rate

Exp oil rate

7.5 Gas Rate (cm3/min)

Oil Rate (cm3/min)

0.15

0.1

0.05

2.5

0 0

5000

10000

15000

20000

25000

30000

Figure 16: Oil production rate against time for a slow depletion test – methane saturated crude oil system

3. The foamy solution gas drive performance is negatively affected by increased solution gas-oilratio.

[2]

[3]

Arora, P., Kovscek, A. R., Mechanistic Modeling of Solution Gas Drive in Viscous Oils, Proceedings, SPE International Thermal Operations and Heavy Oil Symposium, 12-14 March, Porlamar, Margarita Island, Venezuela, 2001, SPE 69717. Alicia, N. Ostos, Effect of capillary Number on Performance of Solution Gas Drive in heavy oil Reservoirs, Master Thesis, Chemical and

15000

20000

25000

Petroleum Engineering, University of Calgary, Calgary, Canada, 2003. [4]

Andarcia, L., Kamp, A.M., Vaca, P., Heavy Oil Solution Gas Drive in the Venezuelan Orinoco Belt: Laboratory Experiments and Field Simulation, Proceedings, SPE International Thermal Operations and Heavy Oil Symposium, 12-14 March, Porlamar, Margarita Island, Venezuela, 2001, SPE 69715.

[5]

Bayon, Y.M., Cordelier, Ph.R., Coates, R.M., Lillico, D.A., Sawatzky, R.P., Application and Comparison of two Models of Foamy oil Behavior of Long core Depletion Experiments, Proceedings, SPE International Thermal Operation and Heavy Oil Symposium and International Horizontal Well Technology Conference, 4-7 November, Calgary, Alberta, Canada, 2002, SPE 78961-MS.

[6]

Bayon, Y. M., Cordelier, Ph.R., Nectoux, A., A New Methodology To Match Heavy-Oil Long – core Primary Depletion Experiments, Proceedings, SPE/DOE Improved Oil recovery Symposium, 1317 April, Tulsa, Oklahoma, USA, 2002, SPE 78961.

[7]

Bondino, I., MacDougall, S.R., Hamon, G., Pore Network Modelling of Heavy Oil Depressurisation: a Parametric Study Affecting critical Gas Saturation and 3-phase Relative Permeability, SPE Journal, 10, 2005, 2, doi, http://dx.doi.org/10.2118/78976-PA

References Albartamani, N.S., Farouq Ali, S.M., and Lepski, B., Investigation of Foamy Oil Phenomena in Heavy Oil Reservoirs, Proceedings, International Thermal Operation/Heavy Oil Symposium, 17-19 March, Bakersfield, California, USA, 1999, SPE 54084.

10000

Figure 17: Gas production rate against time for slow depletion test – methane saturated crude oil

4. Both mineral and crude oil systems displayed similar decline in the oil recovery performance with decreasing rate of pressure depletion. 5. Foamy solution gas drive simulation parameters tuned by history matching a specific experiment do not provide good history matches of other experiments carried out at different rates in the same rock-fluid system.

5000

Time (min)

[1]

Exp. gas rate

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.20160107

Multi-Criteria Assessment of Natural Gas Supply Options – The Macedonian Case Ana M. Lazarevska1*, Daniela Mladenovska2 1

Faculty of Mechanical Engineering, Ss Cyril and Methodius University Karposh II b.b. POBox 464, 1000 Skopje, Macedonia; ana.lazarevska@gmail.com 2 JSC “Macedonian Power Plants”, Skopje, Macedonia; dmladenovska@gmail.com

Abstract

1. Introduction

The number and reliability of the natural gas supply sources have a significant influence over the security of its supply. The specific geographic location that implies natural gas supply through a single gas-pipeline originating from a single source defines Macedonia’s geostrategic position related to natural gas as quite vulnerable. As a 100% importer of natural gas, the country faces serious challenges in relation to the security of the natural gas supply, as well as necessity of defining strategies for mitigating related risks. Efforts should be oriented towards tackling infrastructure and transit related risks, while the highest priority should be diversification of supply sources.

Energy infrastructure in many parts of the Western Balkans is still fragile and in need of serious investment. One of the main priorities among policy makers in the region is to identify and put in place institutions, infrastructure and policies that can support and facilitate the provision of reliable, affordable, diversified and sustainable energy [1].

Relevant indicators and corresponding weights for the natural gas supply to Macedonia have already been defined, assessed and calculated as part of a previous analysis. Herein, a set of potential optional concepts and routes for natural gas supply to Macedonia are investigated and relevant scenarios are examined, discussed and evaluated. The obtained results are ranked to attain optimal scenario for the natural gas supply for Macedonia.

In the case of Macedonia, the natural gas consumption and the corresponding infrastructure are facing severe risks. Namely, since 1997, when Macedonia started to use natural gas, investments and advancements in the gas infrastructure, thus increases in its consumption were negligible. Further, it is important to emphasize that in Macedonia there is no domestic natural gas production, whereby gas supply is fully sourced from Russia [2]. Hence, the position regarding security of supply is quite vulnerable. Another very important issue for decision makers in the country is developing a concept for diversification of the natural gas supply. Therefore, the main focus in this paper is set on investigating several potential natural gas supply options (as alternatives), as well as on defining corresponding criteria and indicators [3] necessary for their overall assessment and ranking. In this paper, the Multi-Criteria Decision Making (MCDM) theory, and the Analytic Hierarchy Process (AHP) as one of its techniques are utilized to address the problem of determining optimal options and routes for natural gas supply to Macedonia.

Keywords:

Article history:

Multi-criteria analysis; Indicators; Weights; Natural gas; Supply routes Received: 13 July 2015 Revised: 12 September 2015 Accepted: 29 January 2016

2. The Multi-Criteria Decision Making theory Multi-criteria Decision Making (MCDM) can be defined as a methodology that facilitates evaluating real world situations, based on various qualitative and/or

___________________________________________________________________________________________________________ A. M. Lazarevska, D. Mladenovska: “Multi-Criteria Assessment of Natural Gas Supply Options – The Macedonian Case”, pp. 54–62

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 1: Components of the MCDM and their interrelations (Source, Lazarevska et al. [5])

quantitative criteria in certain, uncertain, or risky environments to suggest a suitable course of action, choice, strategy, or policy among the available options. This problem becomes even more complex when conflicting and non-commensurable criteria are present and when significant number of decision-makers are involved [4]. The main components of the MCDM are -

the attributes – represented through the corresponding attributes’ set X ,

the criteria or objectives and their relevant indicators – represented through the corresponding criteria set S and

the alternatives – represented through the corresponding alternatives’ set A .

Figure 1 depicts the three main component sets of the MCDM and their interrelations and dependencies

2.1. Criteria and indicators There are many definitions for indicators, but, in fact, they are information tools and should be selected as a measure to quantify criteria, on which basis alternatives are to be compared and ranked. Indicators summarize data for complex and often conflicting problems, with the objective to determine actual conditions, and to create forecasts for future trends. In the context of the policy making processes, indicators represent key signals for necessary improvements in legislation. Indicators serve to four main functions: simplification, quantification, standardization and communication, and usually they contain assessments and analyses contextual to the defined goals [6, 7].

2.2. Alternatives Alternatives represent decision makers’ possible choices. When alternatives are to be identified, initial ideas usually derive from a brainstorming session performed among a carefully selected relevant pool of stakeholders, i.e. the pool of decision maker(s). These ideas must then be enriched to arrive at a respectable choice set. This enrichment is twofold: firstly, the list of alternatives has to be enlarged, and secondly each alternative must be clothed with information [8] Mintzberg (1975) [9] clarified that “actual decision– makers very quickly eliminate large number of possible alternatives to limit their choice to a very small number of alternatives that later are to be examined in details. This may be related to the cost and difficulty involved in analyzing all alternatives.” However, quick elimination of alternatives is a “double-edged sword” which could result in neglecting alternatives that at first glance might seem as a bad solution, but in fact hold potential to resolve the analyzed problem. Therefore, thorough investigation of all potentially relevant alternatives is advised.

2.3. Mathematical definition of the MCDM Here, a brief mathematical formulation to solving a MCDM problem is presented. Further details are thoroughly elaborated in Saaty (1986, 1990) (1986) [10]. The decision (or the goal achievement) matrix , X MxN , aggregates the complete problem related information and is a basis for the problem solution. In the so defined decision matrix we consider that the subjective mapping

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

of the attributes’ set ( X ) onto the criteria set ( S ) has already been performed, i.e. N is the number of the mapped criteria relevant for the calculation of weights and thus the decision making. That is

[

]

X = xij = f j ( Ai ) MxN , i = 1, M , j = 1, N

consideration of multiple criteria or alternatives. Its ability (1) to logically incorporate data and experts’ judgments in the decision making model; (2) to provide a scale for measuring intangibles and method of establishing priorities;

(1)

(3) to deal with interdependence of elements in a system;

i.e. X1 w1

X2 w2

... ...

Xj wj

... ...

XN wN

x11 = f1 ( A1 )

x12 = f 2 ( A1 )

...

x1 j = f j ( A1 )

...

x1N = f N ( A1 )

... Ai

... xi1 = f1 ( Ai )

... xi 2 = f 2 ( Ai )

... ...

... xij = f j ( Ai )

... ...

... xiN = f N ( Ai )

...

(4) to allow revision of judgments in a short time; (2)

(6) to accommodate group judgements if the groups cannot reach a natural consensus,

xM 1 = f1 ( AM ) xM 2 = f 2 ( AM ) ... xMj = f j ( AM ) ... xMN = f N ( AM )

where, M and N are the number of alternatives and criteria, respectively, while xij = f j ( Ai ) indicate the value of the x j criterion with respect to the alternative

Ai .

S = { f1 , f 2 ,..., f N } is the set of criteria, defined as

(∀x ∈ X)(∃f ( x) ∈ S ) : X  S = { f ( x) | x ∈ X} (3) where X = {x | g ( x ) ≤ 0} and g ( x ) ≤ 0 are the problem related set of attributes and the corresponding vector of constraints, respectively. A = { A1 , A2 ,..., AM } is the set of the identified feasible alternatives. A weighting factor, w j , can be associated to each criterion indicating its importance. Then, the “best” solution to a MCDM problem can be defined as:

i =1

makes this method a valuable contribution to the field of MCDM [4, 11, 12]. The technique is convenient for breaking down a complex, unstructured situation into its component parts, then arranging them in a hierarchic order (criteria, sub-criteria, indicators) and assigning numerical values from 1 to 9 [13] (see Eq. (6)) (or an otherwise predefined scale) to subjective judgments based on the relative importance of each criterion (thus indicator), using pair-wise comparison [4]. Hence, in accordance to the selected number of criteria ( N ) corresponding to the defined decision making problem, a square reciprocal matrix of order N is formed as follows:

x1 x1 x2 ... xN

1 a12 a21 1 ... ... aN 1 aN 2

...

... a1 N , ... a2 N ... ... ... 1

(5)

i.e.

max/ x min U ( f ) =  wi ⋅ u i ( f j ( x))

(5) to monitor consistency in the decision–maker’s judgements; and

(4)

where, U i ( f ) is the overall utility function calculated for the alternative Ai , while w j and u j are the weighting factor and the utility related to a particular criteria and the corresponding alternative Ai . [5, 10].

2.4. Analytic Hierarchy Process The Analytic Hierarchy Process (AHP) is a technique based on priority theory, utilized by the MCDM. It deals with complex problems which involve simultaneous

1  aik = a  ki  aii = 1

i ≠ k , where i, k = 1, N and aik ∈ {1,2,...,9} . (6) ∀i

To retrieve the resulting weights, w j , corresponding to the selected criteria, the normalized principle (i.e. the largest) eigenvector of the matrix (eq. (5)) is computed. It corresponds to the vector of the criteria-related weights. In addition, AHP provides a measure of the evaluator’s inconsistency. It is based on the fact that the stakeholders participating in the questionnaire-based survey perform pair-wise comparisons of the criteria in

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Table 1: Random Index, RI , of a randomly-generated pair-wise comparison matrix for n = 1,...,15 . (Source: based on Saaty (1980) [14] and [15])

0.00

0.58

0.90

1.12

1.24

1.32

1.41

1.46

1.49

1.51

1.48

1.56

1.57

1.59

a circle. They compare criterion A with B, B with C, but also C with A. Building on this information, Saaty (1980) , to define a [14] uses the largest eigenvalue, λmax consistency index, CI, as follows:

CI =

(3) Identifying optimal natural gas supply options for the specific case-study.

λmax − N N −1

(7)

where N is the number of criteria. The consistency index is divided by the random index, RI , to compute the final inconsistency measure referred to as consistency ratio, CR

CR =

CI , RI

(2) Assessing and ranking those alternatives versus the criteria-based preferences previously defined from a relevant pool of stakeholders by means of AHP; and

(8)

whereby, RI depends on the order of the matrix defined by Eq. (5). Saaty (1980) [14, 15] elaborated this dependency for matrices of the various order using a randomly-generated pair-wise comparison matrix and, accordingly, proposes the following values for RI given in Table 1. A consistency ratio, CR , of 0.1 or less is considered as acceptable.

3. Case study: Assessing natural gas supply options for Macedonia

The overall hierarchy used for the herein presented case-study is organized in four hierarchy levels, whereby the first level is classified in four categories: economic, environmental, social and technical in concordance with Wang et al. (2006) [16] recommendations for sustainable energy resource management. The subjective mapping (Figure 1) towards identifying the relevant set of criteria (i.e. corresponding indicators) in terms of policymaking, related to the natural gas supply chain in Macedonia, as well as determining corresponding weights, is elaborated in detail in [17]. The objective mapping from the attribute set, in particular the category “technical factors”, onto the alternatives set results in defining alternatives relevant for this case study. In order to avoid overburdening the analysis and final ranking of the alternatives, the overall problem hierarchy presented herein is reduced to the second hierarchy level preserving its sustainability frame. Criteria related weights are recalculated, whereby due to inconsistency of answers in 10 questionnaires, the number of relevant questionnaires was reduced to 24. Final weighting of the criteria is presented on Figure 2.

3.1. Selection and ranking of alternatives 3.1. Problem definition and hierarchy Security of supply, especially in terms of diversification of the supply sources and the corresponding investment costs are definitely among the most important issues in creating policies related to natural gas supply chain in a specific country. However, in order to understand relevant obstacles and drivers for natural gas consumption, and thus to provide consistent and secure supply, a set of additional relevant factors must be taken in consideration. Thus, the focus of our study, carried out between 04–08/2014 (w.r.t. the weighting factors) and 01 – 03/2015 (w.r.t. the alternatives) is set on (1) Determining a set of possible natural gas supply options for a certain country – in this case Macedonia;

Considering the geographic location, geostrategic position, capacities and availability of existing and potential natural gas projects in the region, herein the following supply options are identified as relevant alternatives [2, 18]: − A1: Existing pipeline (business as usual (BaU) scenario); − A2: South Stream (concerning Turkish stream as its alternative); − A3: Trans Adriatic pipeline (TAP); − A4: Energy Community (EC) Gas Ring; − A5: Liquid Natural Gas (LNG); and − A6: Gas storage.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

Figure 2: Problem hierarchy and corresponding weights of indicators

___________________________________________________________________________________________________________

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

A decision making software was utilized to calculate final ranking of the selected alternatives versus the 16 analyzed indicators (Figure 3) in accordance to data in relevant literature. Following ratings were used: 3 for low, 7 for medium and 10 for best rank (Table 2).

that during transport, compared to gas from pipelines [20], LNG is more energy consuming. With respect to GHG emissions (indicator 2.3), since during transport LNG produces more GHG compared to the pipeline gas, it is ranked low, while all remaining are ranked medium. With respect to resource efficiency (indicator 2.4), since for the BaU scenario no additional resources are required, it is ranked best, while low ranking goes to LNG (compared to pipeline gas it requires less space and materials but significantly higher consumption energy during transportation).

With regards to new investments costs (indicator 1.1), the BaU scenario is ranked high, while the TAP, South Stream and EC Gas Ring are ranked low. LNG and Gas storage scenarios are ranked medium due to the fact that although they fall among the most expensive types of projects [19], in Macedonia there is no potential to engage in such projects, thus capacities in the region shall have to be utilized. With regards to indicator 1.2, i.e. “economy’s energy intensity”, all alternatives are assessed equal with low ranking, since this indicator is not related with diversification of supply sources. With respect to natural gas consumption, energy mix and carbon intensity (i.e. indicators 1.3, 1.4 and 1.5, respectively), due to source diversification and price competition, EC Gas Ring, TAP and LNG are ranked medium, while because they neither affect the increase of the natural gas consumption nor the carbon intensity (tCO2/1000USD), the existing pipeline (BaU), Gas storage and South Stream are ranked as low.

In terms of quality of life (indicator 3.1), due to significant diversification of sources and price competition, EC gas Ring and LNG are ranked highest. South Stream prevails over the BaU, because it enriches the quality of life through creating new jobs. With regards to air quality (indicator 3.2), only LNG option is ranked low (due to the increased GHG emissions), while the remaining are ranked medium. In terms of price competition (indicator 3.3), South Stream and the existing pipeline both have low ranking because none of them contribute to diversification of supply.

With respect to compliance with EU Directives (indicator 2.1), all alternatives are ranked low. As per the energy efficiency (indicator 2.2), LNG is ranked low, while all remaining are ranked medium, due to the fact

With respect to import dependency (indicator 4.1) all options are ranked low, because Macedonia is 100% importer of gas and none of the alternatives contributes to changing this status. Gas intensity (indicator 4.2) is in

Technical factors

Social factors

Environmental factors

Economic factors

Table 2: Ranking of the alternatives versus relevant criteria

Indicator 1.1: New investment in gas and energy infrastructure 1.2: Economy’s energy intensity

List of Alternatives A2 A3 A4 A5 Ranking of the alternatives

1.3: Natural gas consumption

1.4: Energy mix

1.5: Carbon intensity 2.1: Legislation (% of the transposed EU Directives)

2.2: Energy efficiency

2.3: Greenhouse Gases (GHG) Emission

2.4: Resources efficiency

3.1: Quality of life

3.2: Ambient air quality

3.3: Prices of other competitive fuels

4.1: Net gas import dependency

4.2: Gas intensity

4.3: Geopolitical risk

4.4: Transit Risk Index (TRI)

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

close relation with economy’s energy intensity (indicator 1.2), hence, all alternatives are ranked low. In terms of geopolitical risk (indicator 4.3) LNG is ranked best because it offers numerous possibilities for diversification [21]. Finally, from the TRI point of view, EC Gas Ring and LNG are ranked high, while South Stream and BaU have low rating due to the number of transiting countries and their geopolitics [22].

4. Results and discussion The quality of the set of indicators deriving from the stakeholders’ preferences is closely related to the relevancy of the analysis. As presented in Figure 2, the survey presented herein confirmed that economic factors are the most important. This was to be expected because a potential investment requires considering all related financial aspects. Furthermore, technical factors that guarantee security of gas supply are closely behind. Such ranking of the criteria-related weights contributed to the final ranking of alternatives. As indicated in Figure 3, EC Gas Ring is the first ranked, while South Stream is ranked as the lowest option. Such ranking clearly shows that advantage is given to the more flexible options and to options providing

diversification of natural gas supply. Due to low ranking versus the environmental factors, LNG is second ranked although with respect to the remaining three categories it is better or similar ranked than the remaining alternatives. EC Gas Ring is a project that holds a potential to cause serious regional impact and, as well, could enable significant price-competition due to diversification of supply. Thus, its significant advantage with reference to the social indicators is quite expected. From the technical factors (security of supply) point of view, LNG scenario is best ranked because it offers diversification of supply. It is interesting that primarily due to identical reasons, TAP prevails over the South Stream project, because it facilitates decreasing the country’s 100% dependency on Russian gas.

5. Conclusion The herein presented case-study provides an interesting point of view with respect to identifying a set of scenarios for natural gas supply, in this case in Macedonia, and ways how, based on the stakeholders’ preferences and opinions, decision makers should act

Figure 3: Ranking of the alternatives vs. the corresponding criteria-based and weighted indicators ___________________________________________________________________________________________________________ A. M. Lazarevska, D. Mladenovska: “Multi-Criteria Assessment of Natural Gas Supply Options – The Macedonian Case”, pp. 54–62

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

with regards to selecting the optimal scenario (option) among them. With respect to overall ranking of the alternatives, as per the herein conducted analysis, EC Gas Ring project was ranked best, South Stream finished last, while TAP is ranked third. It is interesting to note that decision makers in Macedonia actually decided to engage in the South Stream project1 – at least at the time when it still represented a realistic option – while, no initiative existed to engage in the TAP project. Similarly, for the time being, top level policy makers in Macedonia do not consider LNG as a serious option. Further analysis shall be necessary to investigate the reasons for both discrepancies. Such results further indicate that in order to contribute towards higher objectivity and transparency when delivering decisions regarding energy advancements in a country, especially when gas supply and infrastructure are at stake, a thorough and multiattribute approach is a sound tool that facilitates preventing subjectivity among all concerned parties. This is the single and justified approach whilst optimal scenario towards stabile and sustainable energy future of a country is desired.

[2]

Македонска академија на науките и уметностите, Стратегија за развој на енергетиката во Република Македонија до 2030; Министерство за економија, Скопје 2010. (Macedonian Academy of Sciences and Arts, Strategy for Energy Development in the Republic of Macedonia until 2030, Ministry of economy, Skopje, 2010, original in Macedonian language).

[3]

Mladenovska Daniela, Kochov Atanas, Identification of technical indicators for creating natural gas supply policies – Macedonian case, Proceedings, Proceedings of the Conference for International Energy and Environmental Protection in South Eastern Europe, Zlatibor, Serbia, 2015.

[4]

Kumar D. Nagesh, Raju K. Srinivasa, Multicriterion analysis in Engineering and Management, PHI Learning New Delhi, 2010.

[5]

Lazarevska M. Ana, Fischer Nicole, Münnich Kai, Haarstrick Andreas, A multi-criteria decision making conceptual approach to optimal landfill monitoring; GeoSpatial Visual Analytics: Geographical Information Processing and Visual Analytics for Environmental Security”, (Eds. De Amicis, R., Stojanovic, R., Conti, G.), Springer + NATO Public Diplomacy Division, Science for Peace and Security Series C: Environmental Security, Science + Business Media B.V., Dordrecht, The Netherlands, 2009, pp. 85-96.

[6]

Birkmann Jörn, Indicators and criteria for measuring vulnerability: Theoretical bases and requirements, Measuring Vulnerability to Natural Hazards: Towards Disaster Resilient Societies, (Ed. Birkmann Jörn), 2006, pp. 55-77.

[7]

Mladenovska Daniela, Lazarevska M. Ana, Determining relevant attributes and corresponding indicators in a decision making concept for site-selection of a coal fired thermal power plant, Proceedings of the 5th Balkan mining congress, Ohrid 18th-21st September, 2013, pp. 464-469.

[8]

Pomerol, Jean-Charles and Romero S. Barba, Multicriterion Decision in Management, Principles and Practice; Kluwer Academic Publishers, Netherlands, 2000.

[9]

Mintzberg Henry, The manager’s job: Folklore and fact, Harvard Business Review, 53(4), pp. 49– 61, cited in [8].

List of abbreviations, acronyms and symbols AHP BaU EC GHG IEA LNG MCDM OECD SD TAP TRI UNDP UNEP

Analytic Hierarchy Process Business as Usual Energy Community Greenhouse Gases International Energy Agency Liquefied Natural Gas Multi-criteria Decision Making Organisation for Economic Co-operation and Development Sustainable Development Trans Adriatic Pipeline Transit Risk Index United Nations Development Programme United Nation Environment Programme

References [1]

OECD/IEA & UNDP, Energy in the Western Balkans: The Path to Reform and Reconstruction, IEA, Paris, France, 2008. http://www.iea.org/publications/ freepublications/publication/balkans2008.pdf

http://www.finance.gov.mk/en/node/3641

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

[10]

Saaty Thomas, L., Axiomatic foundation of the Analytic hierarchy process, Management Science, Vol.32, No.7, (1986), pp. 841-855.

[11]

Saaty Thomas, L., How to make a decision: The Analytic Hierarchy process, European Journal of Operational Research, 48, (1990), pp. 9-26.

[12]

Saaty Thomas, L. and Gholamnezhad Abdol H., High-Level Nuclear Waste Management: Analysis of Options, Environmental Planning, 9, (1982), pp. 181-196, cited in [4].

[13]

Saaty Thomas, L., Decision making with the analytic hierarchy process, International Journal of Services Sciences, Vol. 1, No. 1 (2008), pp.83– 98.

[14]

Saaty Thomas, L., Multicriteria Decision Making: The Analytic Hierarchy Process, McGraw-Hill, New York, 1980, Revised and published by the authors in 1988; cited in [11], and in Sutter, Christoph, Sustainability Check-up for CDM projects, How to assess the sustainability of international projects under the Kyoto Protocol, Swiss Federal institute of Technology Part II: Theory, 2003.

[15]

Donegan, H. A., Dodd, F. J., A Note on Saaty's Random Indexes, Mathematical and Computer Modelling, Vol. 15, Issue 10, (1991), pp. 135-137.

[16]

Wang Tien-Chin, Liang Ling- Jia and Ho Chun-Yen, Multi-criteria decision analysis by using fuzzy VIKOR, Proceedings of International Conference on Service Systems and Service Management, 2, pp. 901-906, 2006 cited in: Demirtas, O., Evaluating the Best Renewable Energy Technology for Sustainable energy Planning,

International Journal of Energy Economics and policy, Vol.3, Special Issue, (2013), pp. 23-33. [17]

Mladenovska Daniela, Lazarevska M. Ana, Decision making concept for creating policies for natural gas supply chain in Macedonia, (accepted for publishing in the Proceedings of the Conference for Sustainable Development of Energy, Water and Environment Systems (SDEWES)), 2015.

[18]

Енергосистем, Простор, Гастек, Петрол, Хрвоје Пожар, Физибилити студија за гасоводен систем во Република Македонија со идеен проект, Министерство за транспорт и врски, 2010 (Energosistem, Prostor, Gastek, Petrol, Hrvoje Požar, Feasibility study for the gas distribution system in the Republic of Macedonia with a Preliminary project, Ministry for transport and communications, 2010, original in English language).

[19]

US Department of Energy, The Global Liquefied Natural Gas Market: Status & Outlook; US Department of Energy, Energy Information administration, 2003.

[20]

Kavalov Boyan, Petrić Hrvoje, Georgakaki Aliki, Liquefied Natural Gas for Europe – some important issues for consideration, JRC Reference report, 2009.

[21]

Van Der Linde Coby et al., Study on Energy Supply Security and Geopolitics, Final report, Clingendael International Energy Programme, The Hague, 2004, p. 65.

[22]

Le Coq Chloe, Paltseva Elena, Assessing Gas Transit Risks: Russia vs. the EU, Stockholm Institute of Transition Economics, 2011.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.20160108

An Approach towards Thermal Power Plants Efficiency Analysis by Use of Exergy Method Drenusha Krasniqi1*, Risto Filkoski2, Fejzullah Krasniqi1 University ,,HasanPrishtina’’, Faculty of Mechanical Engineering 10000,Prishtina, Kosovo; drrenusha@gmail.com, fejzullahkrasniqi@ashak.org Faculty of Mechanical Engineering, University ,,Sts Cyril and Methodius’’ 1000 Skopje, Macedonia; risto.filkoski@mf.edu.mk

Abstract

1. Introduction

This paper presents the energetic and exergetic analysis of the working cycles in power plants in particular in power plant Kosovo B. The energetic analysis does not take into account the changed quality of heat and mechanical work and as a result it reflects only one part of the system losses. The losses cannot be calculated by this balance because of the outside irreversibility of the processes in particular of the working cycle and that’s why the thermodynamic analysis according to the energetic balance is necessary but not complete. The exergetic analysis of the working cycles identifies losses during the processes in the parts of the power plants, in particular, besides the inner losses (inner irreversibility), it also takes into account the outer irreversibility. This is the reason why the analysis of Rankin working cycle analyzes the losses in power plants. This includes the analysis of the exergy of gases in the furnace of the steam generator (heat exergy of the smoke of gases), exergy of the steam produced in the steam generator, the change of exergy in the regulatory parts of power plants, in the turbine, condenser etc. In the plants where are present big losses, there is a need to find ways to reduce them. The energetic and exergetic analysis of the power plant Kosovo B will be made by different working regimes respectively for different physic and thermal parameters such as temperature and pressure of fresh steam, quantity of the fresh steam, temperature and pressure in the condenser (vacuum change in different regimes), change of parameters of the supplying water and in different points of power plants, quantity of the burning fuel etc. As a result of this analysis, it is possible to identify where the biggest loss happens during the working cycle, the obtained power in each period of time and the electrical energy production.

In relation with the reflection of literature relevant in this field, this paper has given an appropriate contribution, so that, for example, exergy efficiency to EPP Kosovo B is set to change the temperature of the surroundings. This provides a practical approach with regard to optimizing the work of the EPP. This enables simultaneously deepening the problems concerned in professional and scientific terms. The optimization of reheat regenerative thermal-power plants has been analysed for the subcritical pressure range. The exergetic analysis and optimization has been done for the supercritical Rankine cycle. Generalized Thermodynamic Analysis of Steam Power Cycle with adequate number of Feed Water Heaters has been analysed. In addition to the energy analysis, a full-exergy analysis helps to identify components where high inefficiencies occur. Improvements should be done to these components to increase efficiency (exergy yield). The thermodynamic cycle is optimized by minimizing the irreversibilities, so that it is followed to change the temperature of the surroundings (T0). Firstly, the steam turbo-aggregate is analysed with two regulated steam makers and then it is given the appropriate analysis of EPP Kosovo B.

2. The exergy analysis for several examples

Keywords:

Energy; Exergy; Rankin cycle; Steam

Article history:

Received: 28 July 2015 Revised: 13 September 2015 Accepted: 29 January 2016

Amongst various related problems, here are elaborated some concrete cases. Figure 1 presents the regenerative scheme of the heat of the feeding water from the steam taken from turbine. Thermal efficiency of cycle, Figure 2, with regenerative heating of feeding water is:

___________________________________________________________________________________________________________ D. Krasniqi, R. Filkoski, F. Krasniqi: “An Approach towards Thermal Power Plants Efficiency Analysis by Use of Exergy Method”, pp. 63–68

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

1 kg steam SH 1

α1, Pm1, tm1

E EG

α2,Pm2,tm2 2’ 9 5 FWH1

4 P2

3 FWH2

Figure 1: The Scheme of the regenerative heating of the feeding water: SG-steam generator; SH-superheat; T-turbine; C- condenser; P-pump; FWH-open regenerative feedwater heating; EG-electrical generator

1 T

1 kg

9 10 8

6 (1-α1)kg

α1,Pm1 α2,Pm2

7 (1-α1-α2)kg

4 3 2’

αK,PK

s Figure 2: Regenerative heating process diagram (s, T) ___________________________________________________________________________________________________________ D. Krasniqi, R. Filkoski, F. Krasniqi: “An Approach towards Thermal Power Plants Efficiency Analysis by Use of Exergy Method”, pp. 63–68

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

i −i6 +(1−α1 )( i6 −i7 ) +(1−α1 −α2 )( i7 −i2 )

l qf

ηt = c = 1

i1 −i10

Where lT- is useful work which is obtained from turbine, lP - is work which is lost in the pump, ex1- is exergy of steam at the entrance of steam turbine, ex10 – exergy of the steam at the entrance of the steam generator.

i −i −α ( i −i ) −α ( i −i ) =1 2 1 6 2 2 7 2 i1 −i10

For the case is: (1)

Under the scheme, it may be noted that at issue is the regenerative heat in two grades, so that the fraction of steam taken off the turbine is the α1 and α2.

i −i2 −α1 ( i6 −i2 ) −α2 ( i7 −i2 ) +i8 −T0s10 + 2.5

ηex = 1

(8) where:

Considering that applies:

−i0 + T0 s0 = −84, 2 + 293.0, 296 = 2.5( kJ / kg )

i + i8 i' + i ; α2 = (1−α1 ) 2 4 i4 + i6 i2 + i7

α1 = 4

(2) In Figure 3 is presented graphically the exergy efficiency in function of temperature surroundings. In Figure 4 is a topological diagram where exergy efficiency is given in function of steam intakes off the turbine α1 and α2. By analysing the diagram in Figure 4, according to the topological diagram, can be concluded:

following can be derived:

(3)

As regards exergetic efficiency as a constant value, by increasing α1intake, the value α2 is decreased;

Parametric curves (in this case the parametric directions) separated with a linear trend in the reduction (with a trend in reduction);

If the value of exergy energy is greater then, parametric direction is limited, namely the number of solutions is less favourable.

For optimal working regimes:

P P1 = m1 Pm1 Pm 2

;

i1 −T0s1 + 2.5

Pm1 Pm 2 = Pm 2 PK

(4)

Related of (4) is:

(

Pm1 = P12 PK

)

1/3

(

; Pm 2 = P12 PK

)

1/6

(5)

Provided deals: P1 = 80(bar), t1 = 500(0C), PK = 0.05(bar), following is obtained:

α1 = 37.17(%), α2 =15.31(%), ηt = 47.57(%)

(6)

Considering that the energy characteristics of the cycle can be given as a ratio between thermal efficiency and thermal performance of Carnot cycle, for this case is obtained: Θ = 78.74(%). Exergy efficiency per cycle according to Figure 1 and 2 can be determined by:

l − lP + ex10 ex1

ηex = T

(7)

Figure 3: Exergy efficiency ηex in function of temperature T0 (K) of the surrounding

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

High pressure turbine gives this power:

N HPT = m1 (i1 − i2 )

(9)

From the medium pressure turbine steam enters the medium pressure turbine, and the obtained power:

(10) From low pressure turbine steam passes into low pressure where it is obtained:

N LPT = m1 (i4 − iG ) + (m4 − mG )(iG − i5 )

(11)

General power turbine is: Figure 4: Exergetic efficiency ηex as a function of α1 and α2

(12)

Let us now discuss the working regime of EPP Kosovo B. The scheme of the steam turbine is shown in Figure 5. In this figure is shown the reheating water vapour after leaving the high pressure turbine.

Mechanic efficiency, ηm = 0.97 to 0.99;

Electrical generator efficiency, ηgj = 0.98 to 0.99,

and approved values, ηm = 0.987 and ηgj = 0.987, then it is obtained:

Thermal parameters of water vapour of turbine in EPP Kosovo B (working parameters) are given in the Table 1, while the number of intakes of EPP Kosovo B is given in Table 2.

(13) This is the nominal power of one block of EPP Kosovo B.

SH1 MPT

LPT

HPT

G B

Figure 5: The scheme of the steam turbine for EPP Kosovo B (SG-steam generator, SH-super heater, SG- steam generator, HPT-high pressure turbine, MPT-medium pressure turbine, LPT- low pressure turbine) ___________________________________________________________________________________________________________ D. Krasniqi, R. Filkoski, F. Krasniqi: “An Approach towards Thermal Power Plants Efficiency Analysis by Use of Exergy Method”, pp. 63–68

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Table 1: Working parameters of EPP Kosovo B

Table 2: Number of intakes of EPP Kosovo B

Efficiency of Rankin cycle to EPP is:

The heat which enters the process is comprised of: -

Heat which is given in the steam generator for water vapour state (225.5 bar pressure and enthalpy of i= 1076.1 kJ / kg);

Heat which is given in the steam generator for steam superheat from the dry steam to the superheated steam state 1.

ηt =

N T 323.4 = = 0.458 706.8 Qf

(15)

Exergy efficiency for Electrical Power Plants (EPP) Kosovo B is determined by: ηex =

1 [D0 (i1 − i2 ) + (iA − iB )(D0 − DA ) + (iB − iC )(D0 − DA − DB ) + D0 (i1 − T0 s1 + 2.5)

+(iC − iD )(D0 − DA − DB − DC ) + (iD − iEF )(D0 − DA − DB − DC − DD ) + (iEF − −iG )(D0 − DA − DB − DC − DD − DEF ) + (iG − iK )(D0 − DA − DB − DC − DD − −DEF − DG ) − D0 (i1' − ia ' ) + D0 (i1' − T0 s1' + 2.5)]

(16)

(14)

Taking everything into account, the values tabulated by Tables 1 and 2 are presented in the Figure 6, where ___________________________________________________________________________________________________________ D. Krasniqi, R. Filkoski, F. Krasniqi: “An Approach towards Thermal Power Plants Efficiency Analysis by Use of Exergy Method”, pp. 63–68

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

analysis and thermal-energy devices are based on exergy and they are closely related to the first and the second law of thermodynamics. It is important to note that instead of classic diagram to Grassmann (diagram of exergy flows), this paper proposes a chart where respective exergy losses (in the elements of the aggregate), as the exergy efficiency can be given to the effect of temperature environment (T0 ). So is the complete analysis, dynamic and depth in professional and scientific terms. Additionally thus offered the opportunity to optimize the issues in question even using linear and nonlinear programming.

References [1]

Figure 6: Exergy efficiency ηex to EPP Kosovo B in function of temperature T0 (K) of the surrounding

Abdel-Rahim Y. M., Exergy analysis of radial inflow expansion turbines for power recovery. Heat Recovery System & CHP, 1995; 15(8): 775– 85.

[2]

exergy efficiency of Electrical Power Plants Kosovo B is given in function of temperature T0 of surroundings.

Aljundi, I. H., Energy and exergy analysis of a steam power plant in Jordan-Applied Thermal Engineering, Volume 29, Issues 2-3, February 2009, Pages 324-328.

[3]

Bošnjaković, F., Nauka o toplini, I i II dio, TK, Zagreb, 1978.

3. Conclusions

[4]

Brayton/Rankine combined power cycle with reheat- Applied Energy 78 (2004) 179–197.

[5]

Cengel, Y. A., Boles, M. A., Thermodynamics-An Engineering Approach, 5th Ed. McGraw-Hill, New York. 2006.

[6]

Dincer I, Rosen M., Exergy, Oshawa, Canada, 2007.

[7]

Krasniqi, Fejzullah., Kosova Power Plants, Kosova Academy of Sciences and Arts, Prishtina, 2014.

[8]

Kwak, H. Y, Kim, D. J. and Jeon, J. S., Exergetic and thermo-economic analyses of power plants, Energy 28 (2003) 343–360.

[9]

Sokolov J. Jakovlevic, District heating and thermal networks, Моskva, 1982.

[10]

Berichterssammlung “Energie und Exergie”, Dusseldorf, VDI-Verlag, 2005.

[11]

Baehr, H., Thermodynamik, Berlin, HeidelbergNew York, Springer-Verlag, 2006.

[12]

Cerbe, G., Hofimann, H., Einfuhrung in die Warmelehre, Munchen, Carl Hanser, Verlag, 1973.

The exergy analysis of the steam power plants presents a qualitative overview of exergy losses against any energy process and offers the possibility of defining the losses into question. Such analysis is essential not only, specifically, in a theoretical research, but also in relation to the design of thermal-energy devices, as well as for the provision of the possibility of deepening the analysis in terms of professional and scientific research. The rapid and quick industry, livestock, communication, and human growth based on contemporary standards makes a great impact on the development of energy. Nowadays, it cannot be imagined the planning of the development of any industrial branch, and that this mostly because of the implication in the problem of exploitation and energy production. Therefore, out of the large number of energy wells which are available to contemporary man, only a small number of wells into question can be used in a comprehensive manner and can be reasonably economical. For this reason, the contemporary man is bound to define its activities in the so-called classical field with respect to the benefits of energy and requires efficient technological principles regarding their utilization. Understandably, the energy

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.201601009

Modelling and Analysis of Thermal Energy Storage Implementation in the District Heating Systems of the City of Zagreb Kristina Šarović*1, Arben Abrashi1, Damir Božičević2 1

EKONERG – Energy and Environmental Protection Institute Koranska 5, p.p. 14, 10000 Zagreb, Croatia; kristina.sarovic@ekonerg.hr 2 HEP Proizvodnja, 1000 Zagreb, Croatia

Abstract

1. Introduction

Thermal energy storage (TES) is not a novel concept and is already used at many locations, especially in systems with an abundance of renewable energy sources (RES). TES is favoured since it is much easier to store thermal energy than electricity.

Thermal energy storage (TES) is not a novel concept and is already used at many locations, especially in systems with an abundance of renewable energy sources (RES) [1], [2]. TES is favoured since it is much easier to store thermal energy than electricity.

When it comes to district heating (DH) systems whose primary purpose is heat energy supply and which are not directly affected by RES, a question arises: is the implementation of TES financially viable and under which conditions? Complex DH systems usually consist of different types of production units (e.g. gas/fuel oil boilers, electric boilers, cogeneration units, combined cycle cogeneration units). Implementation of TES in this kind of DH systems represents a complex task since the characteristics of the TES system (i.e. power and heat capacity) have to be determined with consideration to the different production units and their ability to interact with the TES. It is almost intuitively clear that implementation of a TES as a part of a DH system improves the system stability and security of supply. It is a little less obvious that TES implementation may also increase the profitability of the DH system through utilisation of the diurnal variations of heat consumption and electricity prices. However, this is not necessarily true but depends on the characteristics of the heat production units, the heat consumption profile, electricity and heat energy tariffs etc. This paper is based on techno-economic analyses dealing with implementation of TES at the locations of the two heat production plants in the city of Zagreb. The analyses were conducted in order to quantify the financial and technical advantages of TES implementation in these specific DH systems (east and west DH system of the city of Zagreb). The goal was to determine the optimal capacity of the TES to be built and to prove their financial viability. In the paper, a short description of the TES and the basic assumptions are given. The advantages and disadvantages of the different kinds of production units available coupled with the TES are explored. Furthermore, an optimisation method for production unit priority determination and TES work mode is proposed.

Keywords:

Thermal energy storage; Zagreb DH systems; Security of supply

Article history:

Received: 29 July 2015 Revised: 06 October 2015 Accepted: 29 January 2016

The implementation of TES as a part of a DH system improves the system stability and security of supply while reducing the necessary maximum production capacity by reducing the maximum load. However, the profitability of TES implementation highly depends on multiple factors, which makes it hard to predict by simply using some rule of thumb. This paper proposes an optimisation method for production unit priority determination and TES work mode definition. Based on the results obtained with this method, it is possible to quantify the financial and technical advantages of a TES implementation. The algorithm is developed specifically for finding the optimal TES capacity at the location of the production plant EL-TO Zagreb, but with minor changes it can be applied for evaluating TES implementation profitability at other locations as well [3].

2. Basic principles TES units can simply be described as huge water reservoirs. The TES are filled with stratified water - a hot water layer above a cold water layer, with a narrow boundary layer in between. The water layer arrangement (also including the protective layers and the top vapour layer) is shown in Figure 1. The terms charging and discharging do not apply to the water content of the TES but to the heat energy content. When charging, hot water enters the storage unit while pushing out the same amount of cold water and vice versa; when discharging, cold water enters the storage

___________________________________________________________________________________________________________ K. Šarović, A. Abrashi, D. Božičević: “Modelling and Analysis of Thermal Energy Storage Implementation in the District …”, pp. 69–76

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 1: Basic scheme of a TES unit

unit while pushing out hot water. Both hot and cold water have to enter the TES via special diffusers which ensure that the vertical stratification stays intact, which is crucial for the TES functioning. The big TES implemented in DH systems are generally not pressurised, but atmospheric. This limits the charging water temperature to around 98 ⁰C.

3. Factors influencing the TES implementation profitability Generally, implementation of TES results in increased security of supply, system availability and stability. It reduces the necessary maximum production capacity by reducing the maximum load thus reducing the need for building extra production units. An increase in profitability of the DH system is also expected, but this is not necessarily true. The profitability is increased trough utilisation of the diurnal variations of the heat consumption and the prices of the fuel, electricity, heat energy, etc. Furthermore, TES usage reduces the number of annual hours of operation of the peak boilers and the number of unit start-ups. The characteristics of the available production units also have significant impact on the profitability. As an

example we can consider usual diurnal cycles of heat consumption at the EL-TO production plant in Zagreb and electricity prices. Both electricity prices and heat consumption reach minimum values during the night. Since the goal is to charge the TES during the night and discharge it during the peak hours, the heat production during the night would be increased while the peaks during the day will be reduced. By doing this, the heat production fluctuations would be partially or completely eliminated. If production units such as condensing steam turbines with steam extraction for DH steam exchangers are used, the heat production increase would result in electricity production reduction. In that case, the usage of TES would yield positive financial results [4]. However, if production units with positive correlation between the heat and electricity production are used, such as backpressure steam turbines with heat exchangers at the outlet or gas turbines coupled with HRSG, the TES usage would result in increased electricity production during the night when the electricity prices are low and reduced electricity production in the peak hours when electricity prices are high. In that case, the TES usage may even yield negative financial results [5]. Third type of units that deserve special consideration is electric boilers. Lately, we are witnesses of relatively

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

high natural gas prices and low electricity prices. Electricity prices are often extremely low during the night. On the other hand, electric boilers have low capital and maintenance costs, high efficiency (>99 %) and precise load control [6]. Therefore, electric boilers represent a great option for TES charging during the night. If building a TES at a location of a production plant consisted of units with positive correlation between the heat and electricity production is considered, building electrical boilers in addition to the TES should almost certainly be considered as well. Finally, the heat capacity and the charging/discharging power of the TES itself significantly influence the profitability. The optimal values of those variables should be carefully chosen within the physical design limitations. It should be noted that not only production unit characteristics are site specific. The electricity, fuel and heat prices/tariffs depend on the contracts the production plant has with its suppliers and customers. For example, some production plants sell the heat at a unique daily price while others have a double tariff system. Heat consumption profiles, maximum load and daily fluctuations also significantly differ from site to site. Furthermore, some production plants are obligated to deliver heat (or domestic hot water) during the night while others shut down. All these factors combined produce a huge number of combinations making it hard to predict the profitability of a TES implementation without a comprehensive analysis.

4. TES implementation calculation model As previously explained, predicting the profitability of a TES implementation by some rule of thumb would be imprecise and unreliable. For that reason, a model for calculating the TES implementation profitability was developed. In the following subchapters a description of the algorithm and the input data is given. This model was used for predicting the TES implementation profitability at the location of the production plant ELTO Zagreb. Some of the obtained results will be shown in chapter 5.

4.1. Input data Input data of the model consist of: 1) Heat (and process steam) consumption profile -

In the mentioned example hourly data from one year period was used

2) Characteristics of the production units available at the site

Max./min. production capacity, efficiency, fuel type, start-up expenses, etc.

3) Prices and tariffs -

Buying prices/tariffs of natural gas, fuel oil and electricity

Selling prices/tariffs of heat energy, process steam and electricity

4) TES characteristics (existing or planned TES) -

Heat capacity and charging/discharging power

5) Parameters for the financial analysis -

Investment costs, expenses for additional employees, maintenance costs, etc.

Cost of capital, inflation, financing methods, depreciation rate, etc.

4.2. Model for determining the production unit engagement priority The production unit engagement priority is closely dependant on the production profitability of the units. A problem arises if the energy and fuel prices are variable, which is usually the case. Then, the production profitability and consequently the engagement priority of the units are also variable. To determine the unit engagement priority on an hourly basis, a variable expressing the specific heat production expenses was defined. This variable is appropriate for determining the production unit engagement priority since heat supply is usually the primary obligation of the production plants while the electricity is a by-product of the cogeneration and combined cycle cogeneration units. The specific heat production expenses are calculated by the following expression =

∗

(

)∗ ∗(

∗ )

(1)

denotes the specific heat production where expenses [€/MWh], the fuel price [€/m3], the fuel the electricity market price consumption [m3/h], the electricity transportation costs [€/MWh], the electricity consumption [MWh], [€/MWh], the electricity production [MWh], the heat energy production [MWh], the process steam production [t], ℎ the process steam enthalpy ,[MWh/t] and ℎ the feedwater enthalpy [MWh/t]. Obviously, in the periods of low electricity prices, the electric boilers would probably have lowest specific heat production expenses and thus highest priority. On the other hand, in the periods of high electricity prices, the

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

cogeneration units would probably have lowest specific heat production expenses and thus highest priority. The model then engages the units according to their specific heat production expenses, starting with the unit with lowest value of that variable. If the first unit cannot satisfy the whole heat demand, the second unit is engaged and so on. If at one point, the remaining heat demand is lower than the minimum capacity of the next unit according to the priority order, than the production load of the previous unit is lowered so that the next one can be engaged. At the end, the algorithm makes additional corrections in order to avoid frequent startups and turndowns. The priority determination and the engagement algorithm are identical in cases with or without TES.

4.3. TES working mode and security of supply The next question that needs to be answered is how we want or plan to use the TES. The answer is again case dependant. In the case of the production plant EL-TO Zagreb, the goal was to flatten the production curve, in other words to eliminate the extreme production fluctuations. As previously explained, by doing this, the number of units’ start-ups and operating hours of the peak boilers would be reduced. Full load engagement of the electric boilers during the night when the electricity price is low would also be possible. For predicting the ideal target production, the statistical function moving average for periods of 48 hours was used on the available heat consumption data. The areas above the moving average curve and below the heat consumption curve are approximately equal to the areas below the moving average curve and above the heat consumption curve (Figure 2). These areas represent the discharging and charging energy respectively. However, since the TES capacity and charging/discharging power are limited, in reality the TES is not always able to respond to the demand defined by the difference between the heat consumption and the ideal targeted production. Therefore, the real production will not always coincide with the ideal targeted production, but will adjust to fulfil the heat demand. The heat demand (blue area), the ideal targeted production (black dashed line), and the real production (red line) in a sample 10 day period are shown in Figure 2. Also note the yellow transparent area showing the TES charging/discharging energy. Four different cases are shown. In the first diagram, which shows the case without a TES, the real production curve coincides with the heat consumption curve. The second diagram shows the case with a 500 MWh

capacity TES. The production curve somewhat follows the ideal targeted production curve but it can be seen that in some hours of peak demand (e.g. around the 138th or 160th hour) the TES has already given out all the accumulated heat and in order for the heat demand to be satisfied the real production has to be increased. Hours of extremely low demand can be almost equally problematic. The most extreme example in the represented sample period is around the 120th hour when the TES is fully charged and cannot take in any more heat so the heat production has to be reduced. This may require turning down some of the production units which we actually want to avoid. In the next two diagrams the cases with 1,000 MWh capacity and 1,500 MWh capacity TES are shown. It can be noticed that with the increase of the TES capacity the number of hours in which the TES is fully charged or fully discharged and thus cannot cover the difference between the heat demand and the ideal targeted production, is reduced. The frequent occurrence of situations in which the TES cannot cover the difference between the heat demand and the ideal targeted production, or in other words when the real targeted production curve does not coincide with the ideal targeted production curve, is a sign of insufficient capacity of the TES. In the case of the production plant EL-TO Zagreb, a period of one year was analysed. The number of hours in which the TES could not cover the difference between the heat demand and the ideal targeted production as a function of the TES capacity is shown in Figure 3 (left). It is shown that this number for 500 MWh capacity TES is around 1,100 hours, but it significantly drops to around 370 hours for a 750 MWh capacity TES. It further drops to around 160 hours for a 1,000 MWh capacity TES. No significant drop of the number of hours in which the TES could not cover the difference between the heat demand and the ideal targeted production is observed with further increase of the TES capacity above 1,000 MWh. Another interesting parameter is the maximum production load. In the sample period shown in Figure 2, the maximum production load in the case without TES equals around 350 MW while it drops for about 100 MW in the case with the TES with capacity of 1,500 MWh. The results of the one year period analysis are shown in Figure 3 (right). The reduction of the maximum production load with the increase of the TES capacity is not as dramatic as the reduction of the number of hours in which the TES could not cover the difference between the heat demand and the ideal targeted production, but it still reaches significant 15-30 % depending on the TES capacity. This clearly shows how TES implementation reduces the necessary installed production capacity.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 2: Heat demand and production in a sample 10 day period

Figure 3: Number of hours in which the TES could not cover the difference between the heat demand and the ideal targeted production in a one year period (left) and maximum production load (right)

5. Results for the case of the production plant EL–TO Zagreb The goal of the analysis conducted in the case of the production plant EL-TO Zagreb was to determine the optimal capacity of a TES for that location if the implementation of TES proves to be profitable. Cases with TES capacities from 500 MWh to 1,500 MWh and a case without TES installed were analysed. All the analyses were repeated using electricity market prices from two different years to determine the TES profitability in long term periods of low and high electricity prices. Cases with different combinations of available production units and with different heat demand, based on the expected future conditions, were also analysed. In total, there were 48 different scenarios. The annual amount of the produced heat energy, process steam and electricity, as well as the annual

amount of the consumed fuel and electricity were calculated and used to predict the annual income and expenses. The unit start-up costs were also taken into account. These data were used as an input for the financial analysis.

5.1. Production unit engagement model results Two sample 10 day periods of the production unit engagement in the case without TES and with 1,000 MWh capacity TES are shown in Figure 4. Both sample periods are from the winter season and are characterised by high heat demand and big fluctuations of the heat demand. In the first period the electricity prices were generally high while in the second one the electricity prices are lower. In both periods, continuous engagement of the CCGT block (red area) is observed. The reason for this is the incentive price of electricity

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

produced by this block which makes it most profitable by far, thus always having highest priority. In the first period, more frequent engagement of cogeneration units (green and blue areas) is observed because of higher electricity prices. These units are turned down during the night because of electricity price drop, as well as heat demand drop. In the night periods of low electricity prices, electric boilers (orange areas) are turned on. If the mentioned units cannot satisfy the heat demand, peak boilers (purple areas) are started up. In the second period, more frequent engagement of electrical boilers (orange areas) is observed. The peak

demand is again covered by the peak boilers (purple areas).

5.2. Financial analysis results Based on the production unit engagement during the one year period, the annual profit was calculated for each of the analysed cases. But, what is relevant for the financial analysis is the incremental profit – the difference between the annual profit in a case including TES of certain capacity and the profit in a corresponding case without TES installed. The results are shown in Figure 5 (left).

Figure 4: Production unit and TES engagement in a sample 10 day period, no TES installed (left) and 1,000 MWh capacity TES (right)

Figure 5: Annual incremental profit for the production plant EL-TO Zagreb (left) and prediction of the investment costs (right) ___________________________________________________________________________________________________________ K. Šarović, A. Abrashi, D. Božičević: “Modelling and Analysis of Thermal Energy Storage Implementation in the District …”, pp. 69–76

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 6: Financial analysis results, NPV (left) and IRR (right)

Detailed description of each specific case with respect to the available production units is beyond this article. However, from the presented results, it can be concluded that the electricity prices reduction has a positive effect on the incremental profit. This shows the dominant positive influence of the electric boilers usage in the periods of low electricity prices. An exception is the year 2018 in which the analysis with higher electricity prices returns better financial results. The explanation for this discrepancy is that in 2018 the CCGT unit will still not be commissioned. As a result, the engagement of the cogeneration units in that case is much more frequent and that type of unit is more profitable when the electricity prices are higher. The increase in the heat demand results in an increase in the incremental profit. This can be most clearly noticed by comparing the results for the years 2023 and 2030 since in those two years the same production units are available while the heat demand is increased. Since the analysis objective was to determine the optimal TES capacity, the parameter of greatest interest was actually the TES capacity. From the diagram in Figure 5 (left) it can be noticed that the incremental profit increases with the increase of the TES capacity to 750 MWh. After that, the increase of the incremental profit with the further increase of the TES capacity is almost negligible. To complete the financial analysis it is also necessary to predict the investment costs and to determine the rest of the economic parameters needed (expenses for additional employees, maintenance costs, cost of capital, inflation, financing methods, etc.). In the case of the TES implementation at the location of the production plant EL-TO Zagreb, the investment costs were predicted according to the investment costs for the reference accumulator with a capacity of 750 MWh (at the location of the production plant TE-TO Zagreb). The investment costs for the TES units of capacity different than 750 MWh were predicted based on the assumption that one part of the investment costs (e.g. auxiliary equipment, design, preparatory work) are

independent of the TES dimensions, the other part is proportional to the volume of the TES (e.g. foundation costs) and the rest are proportional to the TES surface (e.g. steel construction and skin fabrication and assembly). The predicted investment costs, together with the capacity specific investment costs are shown in Figure 5 (right). Finally, the results of the financial analysis performed over a period of 20 years are presented in Figure 6. They show that the TES implementation at the location of the production plant EL-TO Zagreb is justified and financially viable for all analysed cases except for the case with TES with capacity of 1,500 MWh in the analysis with higher electricity prices. The optimal TES capacity is the one for which the maximum NPV and IRR values are reached. According to Figure 6, the optimal TES capacity is around 750 MWh. It is also interesting to notice that in this case the electricity price has a significant effect on the profitability of the TES implementation, however, it does not affect the optimal capacity of the TES.

5.3. Other consideration The financial analysis results clearly show that the optimal TES capacity is around 750 MWh. Nevertheless, other factors such as the security of supply and system stability also have to be taken into account in the decision-making process. Eventual supply interruption or sudden unplanned unit turndown would directly or indirectly generate additional expenses. So, the security of supply and system stability, although not considered and quantified in the model itself, may have significant financial effects. That means that the number of hours in which the TES could not cover the difference between the heat demand and the ideal targeted production as a function of the TES capacity shown in Figure 3 (left) as well as the maximum production load as a function of the TES capacity shown in Figure 3 (right) should most definitely be considered. Finally, it should be noticed that in the case with 750 MWh capacity TES the model operates the TES in a way

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

that it is almost fully charged and fully discharged in each cycle (in the winter period). However, the general attitude of the experts in the field is that during normal operation the TES should always contain some reserve of stored energy. The amount of the stored energy to be kept as a reserve depends on the size of the DHS.

References [1]

Akkaya, B.M., Romanchenko, D., Modelling and analysis of a district heating system containing thermal storage – Case study of the district heating system of Borås, Master’s Thesis, Chalmers University of Technology, Göteborg, Sweden, 2013.

[2]

Harris, M., Thermal Energy Storage in Sweden and Denmark, Master’s Thesis, Lund University, Lund, Sweden, 2011.

[3]

Šarović, K., Abrashi, A., Mužek, A., Komerički, Novelacija opravdanosti izgradnje akumulatora topline na lokaciji EL-TO Zagreb (Revision of the feasibility study for building heat accumulator at the location of EL-TO Zagreb), I-07-0373/14, EKONERG, Zagreb, Croatia, 2015.

[4]

Abrashi, A., Kitarović, P., Vidak, D., Komerički, Z., Abramac, J., Studija izvodljivosti izgradnje akumulatora topline na lokaciji TE-TO Zagreb (Feasibility study for building heat accumulator at the location of TE-TO Zagreb), I-07-0193/12, EKONERG, Zagreb, Croatia, 2013.

[5]

Group of authors, Construction of New Combined Cycle Gas Turbine Cogeneration Unit in El-To Zagreb-Feasibility Study, Consortium Leader EKONERG, Zagreb, Croatia, 2013.

[6]

Jakšić, D., Studija izvodljivosti ugradnje visokonaponskih niskotlačnih kotlova u pogonu EL-TO Zagreb (Feasibility study for implementation of high-voltage low-pressure boilers in the production plant EL-TO Zagreb), EIHP-14106900073, Energy Institute Hrvoje Požar, Zagreb, Croatia, 2014.

Considering all of the above, a TES with a capacity of 1,000 MWh was recommended for the

6. Conclusion TES implementation in a DHS would in any circumstances improve the system stability and the security of supply. Furthermore, it reduces the maximum production capacity needed by reducing the maximum load, thus reducing the need for building extra production units. However, the profitability of the TES implementation highly depends on multiple factors, most important of which are the heat (and process steam) consumption profile, the characteristics of the available production units, the fuel and energy prices/tariffs and the characteristic of the TES itself. All these factors combined produce a huge number of combinations making it hard to predict the profitability of TES implementation without a comprehensive analysis. An algorithm to analyse available data from a reference long term period, to predict the production units and the TES engagement and to calculate the TES implementation profitability was created. The model was developed specifically for the production plant ELTO Zagreb, but with minor changes it can be applied for evaluating TES implementation profitability at other locations as well. However, at this point it is difficult to give any further general suggestions since the problem is highly case specific.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

DOI: 10.14621/ce.20160110

A Software Tool to support Design and Upgrade of Energy Production and Storage Systems Giuseppe Conte*, Luca Paciello, David Scaradozzi, Anna Maria Perdon Dipartimento di Ingegneria dell’Informazione – Università Politecnica delle Marche Via brecce Bianche, 60131, Ancona; Italy; gconte@univpm.it

Abstract

1. Introduction

Home energy systems that can produce, store and manage electricity are efficient solutions to satisfying the growing demand of power in home installations by the exploitation of renewable sources. The design of systems that assure autonomy without oversizing the production with respect to expected requirements can be greatly facilitated by the use of tools for modelling and evaluating performances by simulation in a simple way. Here, we propose a method to model home energy systems by means of Petri Nets and, on that basis, we describe conceptually a software simulation tool, called EPSS (Energy Production and Storage Simulator), that can be used for analyzing the system’s behavior and optimizing its structure with respect to autonomy and self-consumption.

In the last years, variations in cost of fuels and changes in the policy of European countries led to a growth of interest for home energy systems that can locally produce, store and manage electric energy as effectively and efficiently as possible. Indeed, in the recent past, the governments massively incentivized the production of energy from renewable sources, promoting the construction of many plants [1]. After about 15 years of massive incentives, electricity produced in this way has exceeded the thresholds that were indicated as objective in development plans, causing an excess of energy fed into the grid during the peak production hours of photovoltaic plants (more than 17GW were installed in Italy at the beginning of 2013 [2]). Then, many governments changed the mechanism of incentives, in order to promote the installation of new systems, smaller and appropriately sized to maximize self-consumption and/or autonomy. Italy, in particular, has chosen to implement a hybrid policy, with tax reductions for new photovoltaic installations and a new net-metering scheme that gives credits for the energy fed into the grid, instead of money as previously done. Due to incentives, but also to other reasons, like the market availability of electric cars, the idea of achieving autonomy by means of home plants is getting more and more consensus. To-day, the design and sizing of home production and storage system must be very accurate in order to profit of the current incentives, since they are economically efficient only if the electricity produced, or available from the storage elements, equals than that required by the home appliances, devices and systems on any time period [3].

Keywords:

Article history:

Home energy system; Modelling and simulation; Petri Net Received: 22 July 2015 Revised: 14 September 2015 Accepted: 29 January 2016

Actually, the best solution would be that of having a production and storage system that provides total autonomy, so to avoid the necessity of buying power from the grid, and whose production does not exceed the needs, so to avoid unproductive installation costs.

___________________________________________________________________________________________________________ G. Conte, et al: “A Software Tool to support Design and Upgrade of Energy Production and Storage Systems”, pp. 77–85

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Design and sizing with respect to expected requirements is not easy because of the number of involved variables and difficulty increases if, together with solar photovoltaic panels (PV) for production and Lead or Lithium batteries for storage, other production and storage systems are considered (e.g. wind generators and thermal accumulators). Studies, algorithms and software tools about design and analysis of home energy systems, usually focusing on a micro-grid point of view, have already been presented for example in [4], [5], [6],[7], [8] [9] . In this paper, we propose and describe conceptually a versatile and easy to use software simulation tool, called EPSS (Energy Production and Storage Simulator), and the system modelling strategy that works behind its definition, that can be used for several different purposes in designing, virtual prototyping and analyzing home energy systems. The EPSS can be used in the system design phase in order to determine, in a given range, the values of nominal power of the energy production devices and the values of capacity of the storage devices that better satisfy the requirements about autonomy and self-consumption with respect to an expected consumption. Practically, the EPSS evaluates the performance indices of the system with respect to different configurations (i.e. different nominal values of power production and storage) and with respect to different production and storage technologies (i.e. different dynamical models of the production and storage devices) by simulating its operations over a given time period. The results of the simulations allow the optimization of the system structure by a simple and intuitive procedure. Although we focus in this paper on the issues of autonomy and self-consumption, it is worth noting that the EPSS, by providing a complete description of the flows of energy within the modeled home system and between this and the electric network it is connected to, produces all data on the production and consumption of electricity that, together with information on the costs of components, their maintenance and mortgage costs, the cost of energy, the environmental impact due to the construction of the system components and to their management, makes possible to evaluate and to compare the performances of different configurations with respect to a number of economic or environmental criteria. The EPSS employs a Petri Net to model home energy systems as multi-agent systems (HAS-Sim, [10], [11], [12]) and to describe the internal energy flows. Petri Nets give the possibility to model systems of concurrent agents in a simple way and to represent intuitively the energy flow by the exchange of tokens (see [13]). The paper is organized as follows. In Section 2, we describe the modelling methodology for home energy systems that uses Petri Nets. In Section 3, we introduce

performance indices for self-consumption of the produced energy and autonomy. In Section 4, we illustrate the structure of the EPSS, we discuss its performances and we illustrate the way in which it can be used as a design tool to optimize the choice of components with respect to given criteria. Section 5 contains conclusions.

2. Modelling home energy systems by Petri Net The home energy systems we consider include four different kinds of elements: -

External energy suppliers (typically represented by the Electric Company);

Local energy producers (e.g. photovoltaic panels; wind generators);

Energy storage devices (e.g. batteries; thermal accumulators);

Energy consumers or loads (appliances; heating and air conditioning systems; systems and devices for home care, entertainment, communication, safety).

Electric energy is produced, stored or consumed according to the internal state of each involved element and, in turn, states evolve according to specific dynamics and in response to external inputs. The home energy system operates by allowing exchange of energy between the elements, according to current production and demand. It is useful to represent the behavior of the home energy system in discrete time and to quantize by multiples of e.g. 100 W the energy produced and made available, stored or consumed during each time interval between consecutive sampling instants. Exchanges of energy between producers and consumers require availability and, therefore, they may generate conflicts. In addition, they must follow rules and policies in order to assure that the system’s behavior satisfies requirements of efficiency and cost. From this assumptions, it is quite natural to use a Petri Net (PN) (see [13] for basic notions on PN) to construct a model of the home energy system. In the PN model, energy is represented by tokens (1 token = 100 W), which at each sampling instant are moved from places, where they represent the amount of produced/stored energy or the amount of required energy, to places that act as counters. The presence of tokens at different places is indicated by numbers, which form the marking of the net at each sampling instant. Tokens move along arcs that connect places to transitions and transitions to places. The transfer of each token between two places is regulated by the firing of involved transition. Firing can be inhibited if there are tokens on any place that is connected to the transition by an inhibitory arc. If it is

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

not inhibited, a transition can fire if it is enabled, which means that there are tokens at all places from which an arc goes to the transition. Simulating the home energy system’s behavior by a PN model, we assume that all transitions that are enabled and not inhibited fire at each sampling instant, causing the marking to change. In graphic rendering, places are represented by circles and denoted by specific names; transitions are represented by black rectangles and denoted by T#; arcs, from places to transitions and from transitions to places, are represented by arrows; inhibitory arcs, from places to transitions, are represented by dotted lines ending in small circle. Marking is represented, when needed, by numbers inside the circles that represent places. In order to understand how the energy flow can be modelled and simulated, let us consider a simple home energy system that consists of an energy consumer L, connected to a domestic energy producer P and to an external energy supplier S (see Figure 1). The marking EL(t) of the consumer at each sampling instant t represents the (variable) load it generates in the period of time [t, t+1). The markings ES(t) of the supplier and EP(t) of the producer represent the amount of energy they can make available in the same period of time. We impose by a suitable inhibitory arc that the load is satisfied primarily by the domestic producer. Assuming that the external energy supplier is able to supply 3300 W, the marking ES(t) = ES is 33, denoting the presence in S of 33 tokens, each one corresponding to 100 W, while in P and L there are respectively EP(t) and EL(t) tokens. If EP(t)≠ 0, the transition T2 is inhibited and the transition T1 fires n = min(EP(t), EL(t)) times. At each firing, one token moves from P and from L and one token is added to O1. If EP(t) is smaller than EL(t), the marking of P1 goes to 0, transition T2 is no longer inhibited and it can fire m = min(ES(t), EL(t) - EP(t)) times. At each time, one token moves from L and from S and one token is added to O2. T2 stops firing when either the marking of L or that of S is 0. If L ≠ 0 when firing stops, the model signals that in [t,t+1) the home energy system has not been able to accomplish its basic task, namely to satisfy the load, and it must stop. Alternatively, new markings ES=33, EP(t+1) and EL(t+1) are entered to restore values for the time period [t+1, t+2) and operation continues. The marking of O1 and O2 keep growing since these places are counters that, at the end of the simulation, measure respectively the energy produced by P and the energy supplied by S that have been used to satisfy the load. By the same tools, it is possible to model the behavior of the elements that have been mentioned above. External Energy Supplier The External Energy Supplier represents an external source that provides energy to the home energy system by connection to the electric grid of an Electric Company. Its characteristic is that of making available

the same amount of energy in each time interval. It is modelled by the simple net consisting of places S, L and O2 and transition T2 and by the connecting arcs already depicted in Figure 1. The marking of S represents the available energy (e.g. 3300 W). The place L connects the supplier to a load, while O2 records the amount of energy coming that has been used to satisfy the load. Local Energy Producer Local Energy Producers are components of the home energy system that produce energy to be used or stored within the system or, possibly, to be fed into the grid for sale. There may be different kind of producers, whose behaviour is governed by specific dynamics. A common characteristic is that the amount of energy they provide in each time interval varies according to external inputs and cannot exceed a given threshold. Photovoltaic (PV) panels behave in that way and they are modelled by the net shown in Figure 2. The marking PVMAX of PV1 represent the threshold that limit the available energy on each time interval. This is a physical characteristic of the device, which does not change over time, but can be modified to model PV’s with different characteristics. The marking of PV2 represent the energy EPV(t) actually produced in the time period [t, t+1) and it varies at each sampling instant. The places PV3 and PV4 are used to store temporarily the available energy and to separate the flows. The place PL is used to connect the PV panel to a load whose value is indicated by the marking L(t) and PVO1 is the counter that records the amount of produced energy that has been used to satisfy the load. The place PVS is used to connect the PV panel to a storage device and PVO2 is the counter that records the amount of produced energy that has not been used to satisfy the load and that has possibly been stored. The place PVO3 is used to connect the PV panel to the grid and it acts as a counter that records the amount of produced energy that has not been used or stored and that has possibly been fed to the grid. Inhibitory arcs force the use of the produced energy so to satisfy primarily the load, then to charge the storage device and possibly to feed the grid. Different decoupled loads, storage devices and grids can easily be connected by replicating the corresponding connection places and priority in servicing can be set by suitable additional inhibitory arcs. A different kind of local energy producer is the Airborne Wind Energy (AWE) system considered in [14]. Differently from PV panel, the AWE system alternatively produces and consumes energy, in such a way that the balance is positive. A Petri Net model of the AWE system can be constructed by considering it alternately as a producer (that can be modelled as the PV panel) at time t and as a load at time t+1. The length of the sampling interval, in such case, must agree with the length of the operating cycle of the AWE system.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 1: Simple home energy system

Figure 2: PV model

Energy Storage Device Energy produced locally that is not directly consumed can be fed to grid or stored. The most common storing devices are batteries of different kind. In general, their behaviour alternates charging cycles, in which they act as loads, and discharging cycles, in which they supply the stored energy. Such behaviour can be modelled by the net of Figure 3. The marking ES(t) of PB1 represents the total energy stored in the battery at the sampling instant

Figure 3: Battery model

t and, at t=0, it is initialized with the maximum value that characterize the device at issue. The marking of PB2 indicates the energy ED(t) that the device can supply in the time period [t, t+1) and it may vary, according to the battery status, at each sampling instant. The place PL is used to connect the battery to a load whose value is indicated by the marking L(t); PB3 is a temporary counter of the energy that has been used, while the place PBO is the general counter of the used energy. When ES(t) reaches 0, the battery cannot supply energy and a charging cycle starts. Transition T1 is no longer inhibited and all tokens temporarily stored in PB3 are moved to PB5, whose marking EN(t) (initialized as EN(0) = 0) denotes the energy required to recharge the battery. The marking of PB6 indicates the energy EC(t) that the battery can absorb in the time period [t, t+1) and it may vary, according to the battery status, at each sampling instant. The place PS is used to connect the battery to a recharging source, which, in the time period [t, t+1), can supply the amount indicated by the marking S(t). The places PB4 store temporarily the absorbed energy. When the marking EN(t) of PB5 reaches 0, the charging cycle stops and a discharging cycle may start. Transition T2 is no longer inhibited and all tokens temporarily stored in PB4 may move to PB1. The index R(t) associated to the arcs connecting PB4 to T2 indicates that tokens can be moved only in group of R(t) elements and it may vary, according to the battery status, at each sampling instant. Depending on the difference between ES(t) at the beginning of each discharging cycle and R(t) during the next charging cycle, some tokens may remain trapped in PB4 and this models in a simple way hysteresis in the battery’s behaviour. Electric energy can also be transformed in such a way to be more conveniently stored, e.g. by thermal accumulators. In general, they may consist of a

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Figure 4. An example of a home energy system model

thermally insulated tank, in which water or another fluid is stored. In some installations, electric energy produced e.g. by PV during the day (or during the whole Summer) is used to heat the fluid and this, by a heat exchanger, is used to regulate the home temperature during the night (or during the whole Winter). If, instead of heating the fluid, electric energy is used to cool it, it is possible to use the stored energy to reduce the home temperature. We can consider thermal accumulators as devices that can supply the stored energy only to a Heating, Ventilating and Air Conditioning (HVAC) system, which acts as a load either for them or for a producer/supplier of electric energy. We can therefore model also this kind of storage devices as batteries, providing the only load they can be connected to is that representing the HVAC system. Using the elements described above, it is possible to model home energy systems that contain several components. An example is given by the net of Figure 4,

which describes a home energy system which integrates PV, a battery and a thermal accumulator. Load is represented by a generic load, due to all devices that require electricity to work except the HVAC system. This is represented as a special load, which can be connected either to the thermal accumulator or to other sources of electricity. Note that inhibitory arcs force a specific behaviour of the net, imposing e.g. that the generic load must be satisfied primarily be the photovoltaic production or by the battery. At the same time, the HVAC system load is primarily satisfied by the thermal accumulator. Only in case all this is not sufficient, electricity is taken from the grid. Similarly, the energy produced by PV is primarily used for satisfying the load, then for charging the battery and the thermal accumulator and only if there is no load and the storage devices are fully charged, it is fed to the grid. Inhibitory arcs can be modified, varying the structure of the net, in order to implement different policies in managing electricity.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

3. Self–consumption and autonomy To evaluate the performances of a home production and storage system, one of the most commonly used indexes is the Self-Consumption index. As described in literature, (for example in [4]) the Self-Consumption index (SC) measures the rate of produced energy that is consumed or stored locally. Let us consider a system connected to a provider grid and composed e.g. by a PV generator and a battery, in which the energy produced is used to satisfy the domestic electric load and to charge the battery. Exceeding energy is feed to the grid and, on the other hand, exceeding load is satisfied by getting power from the grid. In this case, the SC index is defined as E

SC= PV,L

+EPV,S

EPV,T

(1)

where -

EPV,T is the total energy produced by the PV generator during a given time period;

EPV,L is the part of energy produced by the PV generator that is used to satisfy the domestic electric load during the same time period;

EPV,S is the part of energy produced by the PV generator that is stored in the Battery during the same time period.

Obviously, SC cannot exceed 1 and it is lower than 1 if, during the considered period, the load, including that generated by charging the battery, is lower than the production. As it’s easy to understand, maximizing SC without considering other constraints does not make sense, since this objective can be easily reached by keeping the production lower than the load. So, together with SC, it is important to consider another index, which is related to autonomy. The Autonomy index (AUT) actually measures the percentage of the consumed energy that has been provided by the production and storage system. In the case at issue, the AUT index is defined as: E

AUT= PV,L

+EPV,S

ET,C

(2)

where -

ET,C is the total energy consumed, given by

ET,C =ET,L +൫EPV,S -E஻,L ൯

(3)

ET,L is the total energy required by domestic loads during a given period;

EB,L is the total energy provided by the battery to satisfy the domestic load.

Clearly, a high value of AUT characterizes a system in which the domestic load is largely satisfied by the production and little or no energy is got from the grid. Since the load is not constant, increasing the AUT index will generally result in a decrement of the SC index and conversely. The design of an efficient home production and storage system needs therefore to address a practical optimization problem that consists in maximizing both indices, while containing the costs of the various components, with respect to a (variable) expected load.

4. Simulating home automation systems A software tool, called Energy Production and Storage Simulator (EPSS), has been developed in the NI LabView 2013 environment [15]. Basically, the EPSS implements the dynamics of the Petri Net that, as described in Section 1, models a home energy system activating the transfer of tokens. In the simplest operating mode, the EPSS can be used to simulate the operation of a home energy system over a given time interval. The input of the EPSS consists of look-up tables that describe the behavior of the local energy producers and of the energy storage devices, together with a number of files that describe the behavior of the load(s) and of some external variables. For the PV, the look-up table indicates PVMAX and the energy EPV(t) produced by the PV element in each time interval in correspondence of a given value of the solar radiation S(t) and possibly of the external temperature T(t). For the battery, the look-up table indicates the initial charge ES(0); the values of ED(t), EC(t) that correspond to the actual values of ES(t) and EN(t) and that, possibly, depend also on the number of executed charge/discharge cycles and on the external temperature T(t); the value of R(t) that possibly depends on the number of executed charge/discharge cycles and on the external temperature. A similar look-up table can be used for the thermal accumulator. The look-up tables are constructed off-line, either on the basis of experimental data or using suitable models of the considered devices. The load(s) file contains the values L(t) of the load(s) at each sampling time. The solar_ radiation and the ex_temp files contain the values of the solar radiation S(t) and of the external temperature T(t) at each sampling time. If other variables that may influence the behaviour of the above elements are considered in the look-up tables (e.g. internal temperatures), their values are entered by means of additional input files. Simulation is performed by executing a sequence of steps, in which each step corresponds to the time interval between consecutive sampling instants.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Marking is initialized at the beginning of each step by using the look-up tables and the input files. The EPSS gives the possibility to see the temporal evolution of the marking of the counter places, representing the energy that is produced, stored, consumed and fed to grid and it computes the self-consumption and autonomy indices. In this way, the EPSS works as a tool for analyzing the performances of the modelled home energy system, using measured or predicted data in the input files. Energy producers and storage devices with different physical characteristics, performances and costs will correspond to different look-up tables. Therefore, any choice of the look-up tables in a given set defines a specific configuration of the home energy system. In a more complex operating mode, the EPSS can be instructed to perform a series of simulation over the same time period and with the same input files (that is: in the same conditions), changing, in a programmed way, the look-up tables (that is: the system configuration) by choosing them in a given set. Comparison of the resulting values of self-consumption and autonomy, then, gives the possibility to select the configuration of the home energy system that provides the best performances for the considered conditions. In this second operating mode, the EPSS explores the parameter space using a discretization step that can be set by the user. Procedures that search for values of the parameters that try to optimize the performances can be implemented by exploiting this operating mode of the simulator. An example of a home system that can be modeled and simulated is given in Figure 4. The system includes a PV panel, a battery, a thermal accumulator, an external energy supplier, a generic load, a HVAC load. The set of places of the PN model which correspond to each one of these elements are identified by circles of different colors: red for the PV panel, green for the battery, blue for the thermal accumulator, solid yellow for the external energy supplier, solid red for the generic load, solid green for the HVAC load. Connections are such that the thermal accumulator can supply energy only to the HVAC system. Inhibitory arcs are such that the HVAC exploits primarily the energy stored in the thermal accumulator. If none is available, the HVAC load adds to the generic load. This is satisfied primarily by the PV, then by the battery and finally, if needed, by energy coming from the external energy supplier (Electric Company). The exceeding energy from the PV is used primarily to charge the battery or to increase that stored in the thermal accumulator (if the battery is charged) or it is fed to the grid for sale (if the battery and the thermal accumulator are charged). To evaluate the performances of the EPSS, we compared the real behavior of a home energy system with that obtained by simulation from a model of the same system

Table 1: Real behavior vs. simulated behavior

Leaf House EPSS

SC 49.11% 52.55%

AUT 24.61% 27.38

with regard to the SC and AUT indices. The real system is that of the Leaf House in Angeli di Rosora, Ancona, Italy. The Leaf House is a complex of apartments, divided in energetically independent blocks. Each block is connected to a PV system with a nominal power of 6kW and to a Lithium battery with a nominal capacity of 5.8kWh. Energy production and consumptions are continuously monitored and an average value for each of them is recorded every 15 minutes. The system is governed by an inverter in such a way to maximize selfconsumption. The model used in simulation is like that of Figure 4, without the thermal accumulator. Data about real loads and solar radiation have been collected for one year in 2013 and they have been used to construct the load and the solar radiation files The lookup tables for the installed PV and battery systems have been constructed experimentally. Results given in Table 1 show a substantial agreement between the real and the simulated behavior, with small discrepancies that are due to the non-ideal behavior of the real components. As a design tool, the EPSS has been tested to find the best configuration, in terms of nominal power of the PV panel and nominal battery capacity, for the same system as above, using the same data files, with respect to a desired level of autonomy and self-consumption. In a first series of simulations, look-up tables for PV panels with nominal power between 2000 W and 10000 W (with interval of 1000W) and for battery with nominal capacity between 2000 Wh and 10000 Wh (with interval of 1000 Wh) have been used. Considering only the system configurations, in terms of PV nominal power and battery nominal capacity, that assure, for instance, values of the SC index greater than or equal to 40% and values of the AUT index greater than or equal to 45%, simulations gave, in particular, the results contained in Table 2. Simulation can also be refined, for instance considering configurations with PV nominal power between 3500 W and 4500 W with interval of 100 W and battery capacity between 3500 Wh and 4500 Wh with interval of 100 Wh, obtaining the results shown in Table 3. Then, results can be used together with information about the market prices of the PV panels and of the batteries to find the configuration that, while achieving the desired performances in terms of SC and AUT indices, minimizes the costs of one or more of those components. In this case, choosing for instance to minimize the cost of the battery irrespectively from the cost of the PV panel, the configuration corresponding to 4000W and 4000Wh resulted to be the preferable one (see [14], [15]).

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Table 3: Simulation results

Table2: Simulation results PV nominal power

Battery nominal power

AUT

PV nominal power

Battery nominal power

AUT

4000

41.69%

45.03%

4000

41.69%

45.03%

3000

5000

58.17%

45.60%

4100

41.05%

45.27%

4000

5000

46.76%

48.27%

3900

4200

42.80%

45.01%

3000

6000

62.70%

47.81%

4100

4200

41.03%

45.24%

4000

6000

51.73%

51.23%

3800

4300

44.71%

45.50%

5000

6000

42.18%

52.22%

3900

4300

43.38%

45.39%

3000

7000

66.44%

49.50%

4000

4300

42.30%

45.46%

4000

7000

54.29%

52.71%

4100

4300

41.29%

45.45%

5000

7000

44.07%

53.58%

3500

4400

49.06%

45.68%

2000

8000

90.69%

45.84%

3600

4400

47.39%

45.48%

3000

8000

70.52%

51.30%

3700

4400

45.61%

45.22%

4000

8000

57.37%

54.32%

3800

4400

45.09%

45.80%

5000

8000

46.91%

55.46%

3900

4400

44.00%

45.84%

6000

8000

40.21%

56.57%

4000

4400

43.28%

46.12%

2000

9000

92.00%

46.26%

4100

4400

42.51%

46.27%

3000

9000

72.71%

52.11%

4200

4400

41.14%

46.12%

4000

9000

58.99%

55.31%

3500

4500

48.78%

45.52%

5000

9000

48.54%

56.46%

3600

4500

47.26%

45.39%

6000

9000

42.78%

58.26%

3700

4500

45.55%

45.16%

2000

10000

93.28%

46.77%

3800

4500

44.78%

45.60%

3000

10000

73.65%

52.45%

3900

4500

44.29%

45.97%

4500

43.24%

46.07%

4000

10000

60.48%

55.94%

4000

5000

10000

51.18%

57.98%

4100

4500

42.47%

46.22%

6000

10000

44.54%

59.35%

4200

4500

41.14%

46.09%

5. Conclusions A simple and intuitive way of modelling the flows of energy in home energy systems by using Petri Net models has been illustrated. By simulating the model behaviour, components of the system can be optimized with respect to self-consumption and autonomy, facilitating the design of a cost efficient system. Since the EPSS provides also the values of the quantities EPV,T, EPV,L, EPV,S, ET,C, ET,L, EB,L defined in Section 3, which completely characterize the flows of energy in the home system and between the home system and the grid, additional information about the price of electric energy, either bought from the electric company or sold to it through the grid, and about the mortgage costs of the components can be used to analyze the performances of

the home system from an enlarged economic point of view. This makes possible to find the configuration that achieves given economic objectives or that better responds to economic constraints. Analogously, information about the environmental aspects that are connected to the life cycle of the system components (PV panels, battery and thermal accumulator) and/or to the production of electric energy that the home system gets from the grid may be used to evaluate the environmental impact of different system configurations.

Acknowledgements The authors would like to thank the Loccioni Group for providing facilities and data and the unknown reviewers for useful comments.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

Renewable Power Generation IET/RPG, vol.6, 2012, n. 5, pp. 358 -371.

References [1]

Brown P., EU Wind and Solar Electricity Policies: Overview and Considerations, Congressional Research Service, 2013. https://fas.org/sgp/crs/row/R43176.pdf

[2]

Roney J. M., China Leads World to Solar Power Record in 2013, in Cumulative Installed Solar Photovoltaics Capacity in Leading Countries and the World 2000-2013, Earth Policy Institute, 2014.

[3]

D. M. 05/07/2012 Attuazione dell’art. 25 del decreto legislativo 3 marzo 2011, n. 28, http://www.acs.enea.it/doc/dlgs_28-2011.pdf

[4]

[5]

Layadi T.M., Mostefai M., Champenois G. and Abbes D., Dimensioning a hybrid electrification system (PV / WT / DG + battery) using a dynamic simulation, Proceedings International Conference on Electrical Engineering and Software Applications (ICEESA), 2013. Dufo-Lopez R., Bernal-Agustin J.L., Design and control strategies of PV-Diesel systems using genetic algorithms, Solar Energy, 79, 2005, no. 1, pp.33-46.

[6]

Nelson D.B., Nehrir M.H. and Wang C., Unit sizing and cost analysis of stand-alone hybrid wind/PV/fuel cell power generation systems, Renewable Energy, vol 31, 2006, n. 10, pp. 1641– 1656.

[7]

Belfkira R., Zhang L. and Barakat G., Optimal sizing study of hybrid wind/PV/diesel power generation unit, Solar Energy, vol. 85, 2011, n. 1, pp. 100-110.

[8]

Abbes D., Martinez A. and Champenois G., Ecodesign Optimisation of autonomous hybrid windphotovoltaic system with battery storage,

[9]

Kolhe M., Techno-Economic Optimum Sizing of a Stand-Alone Solar Photovoltaic System, IEEE Transactions on Energy Conversion, vol. 24, 2009, n. 2, pp.511–519.

[10]

Conte G., Scaradozzi D., Donnini R. and Pedale A., Building Simulation/ Emulation Environments for Home Automation Systems, Proceeding 19th Mediterranean Conference on Control and Automation, Corfu, Greece, 2011, pp. 31-38.

[11]

Conte G., Scaradozzi D., Viewing home automation systems as multiple agents systems, in Multi-agent system for industrial and service robotics applications, Proceedings RoboCUP2003, 2003.

[12]

Scaradozzi D., Methodologies and techniques for analysis and design of home automation systems, Ph.D. dissertation, Università Politecnica delle Marche.

[13]

Peterson J. L., Petri Net Theory and the Modeling of Systems, Prentice-Hall, 1981.

[14]

Paciello G. L., Scaradozzi D., Conte G., Coleman J., Toal D., A Simulation Tool for Home Energy Production and Storage System Using Photovoltaic and Airborne Wind Energy, Proceedings QUAEST2014 - The 2nd Virtual Multidisciplinary Conference, Zilina, Slovakia, 2014.

[15]

Paciello L., Pedale A., Scaradozzi D., Conte G., A Design Tool for Modelling and Sizing Energy Production/Storage Home System, Proceedings EESMS2014, Naples, Italy, 2014, pp. 1-6.

International Journal of Contemporary ENERGY, Vol. 2, No. 1 (2016)

ISSN 2363-6440

___________________________________________________________________________________________________________

The Journal

.e9

About the Journal

.e10

Instructions for Authors

.e12