FUTURE STEEL FORUM - JUNE 2022 Since 1866
w w w. s t e e l t i m e s i n t . c o m
TURN INFORMATION INTO VALUE Building the Learning [Steel] Plant SMS digital develops innovative solutions to boost your business. Benefiting from cutting-edge development methods, our solutions for plant and process condition, product quality,production planning, and energy management contribute in streamlining your maintenance efforts,decrease quality deviations and optimize plant utilization, even down toa short-term rescheduling. The digital future has already begun
Also visit us online: www.sms-digital.com/solutions/the-learning-steel-plant
Learning Steel Plant Ad onepage.indd 3
21.10.21 18:04
FUTURE STEEL FORUM - JUNE 2022
w w w. s t e e l t i m e s i n t . c o m
Since 1866
Cover courtesy of Badische Stahl Engineering
Contents 3 Welcome by Matthew Moggridge, programme director 4
Editorial
Editor / Programme Director Matthew Moggridge +44 1737 855151 matthewmoggridge@quartzltd.com Editorial Assistant Catherine Hill +44 1737 855021 catherinehill@quartzltd.com
Production Editor Annie Baker
Advertisement Production
Conference programme 8 Floor plan and exhibitor list 10 Speakers biographies 28 Exhibitor profiles 38
Martin Lawrence
Succeeding with AI in steel manufacturing
Sales
43
International Sales Manager Paul Rossage +44 1737 855116 paulrossage@quartzltd.com
Beginners guide to AI in the metal industry 49 Four roadblocks to generating RPO through machine learning
Sales Director
Ken Clark +44 1737 855117 kenclark@quartzltd.com
Corporate
Managing Director Tony Crinion Published by: Quartz Business Media Ltd Quart House, 20 Clarendon Road Redhill, Surry RH1 1QX, UK +44 1737 855000 www.steeltimesint.com © Quartz Business Media, 2022
55 Eliminating CO2, energy and quality efficiencies with AI 62 Tracing sustainability information with ‘Internet of Metals’ technology 67 AI and ML applications for EAF optimization 72 2D, 3D or both? 78 Safe removal of high-strength steel straps 82 Classifiers for surface inspection
Steel Times International
contents page.indd 3
FUTURE STEEL FORUM 2022
1 13/05/2022 07:31:11
E N G A G E
A C C E L E R A T E
A C H I E V E
www.digi-met.com
Danieli Automation Spa Via Bonaldo Stringher, 4 33042 Buttrio (UD) Italy Phone +39 0432 518 111
Matthew Moggridge, programme director, Future Steel Forum
Welcome Welcome to the first live Future Steel Forum since 2019 in Budapest. You could say that a lot of water has passed under the bridge since then and now, three years later, we are back and we are live and, let’s hope, that normality returns for everybody pretty soon. We appear to be living in world that can only cope with one big news story at a time. First there was Brexit, then there was Trump, then there was COVID and now it’s Russia’s invasion of Ukraine. I dread to think what comes next. There are, however, many constants in life and one of them is the continual development of AI- and ML-based technologies designed to make the steelmaking process that little bit more efficient and, of course, greener. In fact, almost by default, there appears to be a theme to this year’s Future Steel Forum and that is Digitalisation and Decarbonisation and whether one can help the other. It’s a question that some of our 2022 participants, including H2GreenSteel’s chief commercial officer Mark Bula, will be discussing across the duration of the conference. This year’s presentations have a distinct sustainability bias too, with leading steelmakers and production technologists examining how ‘hitech’ can change things for the better on a global scale. I wish all delegates well and I hope you all enjoy the conference.
Matthew Moggridge, programme director.
Steel Times International
welcome.indd – read MM..indd 3
FUTURE STEEL FORUM 2022
3 11/05/2022 09:44:52
CONFERENCE PROGRAMME
Day One: Tuesday 8 June 09:00 Welcome to the Future Steel Forum by Matthew Moggridge, Programme Director and Editor, Steel Times International. 09:10 Opening Keynote Address by Mark Bula, Chief Commercial Officer, H2GreenSteel. 09:40 ArcelorMittal’s Global Approach to Digitalisation by Carlos Alba, Chief Digital Officer, ArcelorMittal Global Research & Development. 10:10 The Future of Steelmaking: Decarbonized and Digitalized by Professor Katja Windt, CDO, SMS group GmbH. 10:40 Tea & Coffee break 11:10 Aiding Human Decision Making and Standardizing Practices Using Artificial Intelligence. By Robert Vandlik, Head of Digital Studio, US Steel Kosice, and Juraj Sabol, General Manager for Strategy, US Steel Kosice. 11:40 Automatic Slab Yard for a Greenfield 4.4Mt/yr Hot Strip Mill by Monti Fernando, Automation New Technologies Director, Ternium, and Lorenzo Bacchetti, Senior Manager, Cranes and Automatic Yard, Danieli. 12:10 Promotions and Challenges in JFE’s DX on Steelmaking Process by Junichi Yotsuji, JFE Steel. 12:40 Lunch Break
4
FUTURE STEEL FORUM 2022
a a Programme.indd – read MM needs Forum logos.indd 2
Steel Times International
11/05/2022 09:50:06
CONFERENCE PROGRAMME
13:45 Digitalization and Automation as Crucial Success Factors in Modern Steelmaking – New Special Steel Plant Sets a Global Benchmark, by Dr. Roman Stiftner, Managing Director, Austrian Non-Ferrous Metals Federation and the Austrian Mining & Steel Association. 14:15 Building a Sustainable Future for Steelmaking Using Modern Technology by Sambit Beborta, Director Technology, Liberty Steel Europe. 14:45 Tea break 15:15 Digital Assistants – Companions for a New Way of Operating Metals Plants Towards a Dark Plant. By Dr. Alexander Thekale, Technology Concepts and Digital Solutions, Primetals Technologies. 15:45 How Digitalisation can Aid Decarbonisation Part One – Panel Discussion Panellists include: • Mark Bula, Chief Commercial Officer, H2GreenSteel • Sambit Beborta, Director Technology, Liberty Steel Group • Carlos Alba, ArcelorMittal (awaiting confirmation) • Dr. Thomas Pfatschbacher, Head of Digital Transformation and Smart Production, Primetals Technologies. Panel Chair: Dr. Nils Naujok, Partner, Oliver Wyman’s Energy & Natural Resources Project. 16:30 Decarbonizing Steel: How Demand for Greener Steel will Upend the Supply Chain and how Digitalisation can Help by Dr. Nils Naujok, Partner, Oliver Wyman’s Energy & Natural Resources Project and Holger Stamm, Principal, Oliver Wyman, Dusseldorf, Germany. 17:00 How to transform the steel industry for efficiency, sustainability and future competitiveness by Bertrand Orsal, steel experience senior director, Dassault Systemes. 17:30 Closing remarks 17:45 Conference closes Steel Times International
a a Programme.indd – read MM needs Forum logos.indd 3
FUTURE STEEL FORUM 2022 5
11/05/2022 09:50:23
CONFERENCE PROGRAMME
Day Two: Wednesday 9 June 09:00 Welcome to Day Two of the Future Steel Forum by Matthew Moggridge, Programme Director and Editor, Steel Times International. 09:10 Risk-Based Inventory Management, by Diego Diaz Fidalgo, Global R&D expert (Artificial Intelligence) at ArcelorMittal. 09:40 Digital Twin Modelling and Optimisation Technology Helps Reduce Cost and Carbon Emissions in Stainless Steelmaking Operation, by Yale Zhang, Global Director, Analytics and Decision Solutions, Hatch Digital. 10:10 WQR: The Most Innovative Wire Rod mill in the World by Enrico Plazzogna, Executive Vice President, Danieli Automation SpA Italy. 10:40 TEA & NETWORKING BREAK 11:10 Aligning Production to Available Energy with an Energy Consumption Forecast, by Dr. Stefan Albers, Business Consultant, PSI Metals. 11:40 Eliminating CO2, Energy, and Quality Inefficiencies through Application of Artificial Intelligence, by Dr. Falk-Florian Henrich, Smart Steel Technologies. 12:10 Towards a Fully Connected and Autonomous Steel Melt Shop that’s Safer, Smarter and More Sustainable by Tarun Mathur, Global Product Manager, Metals Digital, ABB. 12:40 LUNCH & NETWORKING BREAK
6
FUTURE STEEL FORUM 2022
a a Programme.indd – read MM needs Forum logos.indd 4
Steel Times International
11/05/2022 09:50:27
CONFERENCE PROGRAMME
13:45 Using Eco-Responsible Combustion Systems to Transform Steel Manufacturing Approaches by Jean-Paul Nauzin, Technology & Innovation Director, Fives Group. 14:15 Integrating Plant Logistics with Production Processes – Fully Automatic Logistic Operations at Acciaierie Bertoli Safau SpA (ABS) by Tony Leikas, CEO, Pesmel, and Lorenzo Bacchetti, Senior Manager, Danieli Centro Cranes 14:45 TEA & NETWORKING BREAK 15:15 Artificial Intelligence Applications for EAF Optimization by Mariana Viale, Applications and Development Engineer, AMI Automation. 15:45 How Digitalisation Aids Decarbonisation – Panel Discussion Pt2 Panellists include: • Diego Diaz Fidalgo, Global R&D Expert (Artificial Intelligence) at ArcelorMittal. • Jean-Paul Nauzin, Technology & Innovation Director, Fives Group. • Dr. Falk-Florian Henrich, Smart Steel Technologies. • Mariana Viale, Applications and Development Engineer, AMI Automation. Panel Chair: Holger Stamm, Principal, Oliver Wyman, Dusseldorf, Germany. 16:30 Optimizing Heat Chemistry in Real Time using Explainable Machine Learning. By Berk Birand, Founder, FeroLabs. 17:00 What Will We still Sell in the Next Weeks? Using AI and Demand Sensing for a Smarter Conversion of Forecast in the Order Book by Dr. Luc Bongaerts, Senior Advisory Manager, OM Partners. 17:30 CLOSING REMARKS 17:45 CONFERENCE CLOSES
Steel Times International
a a Programme.indd – read MM needs Forum logos.indd 5
FUTURE STEEL FORUM 2022
7 11/05/2022 09:50:35
EXHIBITOR LIST
ABB Metals AMI Automation
(Sponsor) Stand 7
LAP Midrex
Stand 8
Badische Stahl-Engineering GmbH
Stand 12
Nextsense GmbH
Beda Oxygentechnik Armaturen GmbH
Stand 14
OMP BV
Danieli Automation Spa
Stand 10
Optris GMBH
Dassault Systèmes
(Sponsor)
Pesmel
Dr Schenk GmbH
Stand 19
Primetals Technologies Austria GmbH
Stand 17
Endress And Hauser
Stand 20
PSI METALS
(Sponsor)
Fero Labs Inc.
Stand 23
QuantoLux Innovation GmbH
Fives Group
Stand 6
FrigorTec GmbH IMS Messsysteme GmbH
Stand 24 Stand 4
Stand 10a Stand 3 Stand 18 Stand 9
Stand 5
Quinlogic GmbH
Stand 22
SAP SE GmbH
Stand 13
Smart Steel Technologies GmbH
Stand 1
ISRA Parsytec GmbH
Stand 11
SMS Group GmbH
Stand 21
KELLER ITS
Stand 25
Tebulo Industrial Robotics
Stand 16
Koeppern
Stand 26
TeamViewer
Stand 15
TMEIC Inc
8
Stand 27
FUTURE STEEL FORUM 2022
a c EXHIBITOR list and Floor plan.indd 2
Stand 2
Steel Times International
11/05/2022 09:38:31
FLOOR PLAN
Steel Times International
a c EXHIBITOR list and Floor plan.indd 3
FUTURE STEEL FORUM 2022
9 11/05/2022 09:38:33
SPEAKER PROFILES
MARK BULA
CHIEF COMMERCIAL OFFICER, H2GREENSTEEL The equity founders of H2 Green Steel recruited Mark to serve as its chief commercial officer in January 2021. They saw in Mark critical experience that was missing in Europe: minimill start-up knowledge. Mark has participated in multiple start-up companies in the American steel industry, including as a co-founder of Big River Steel (BRS), Osceola Arkansas. Mark and the BRS team raised over $1.3 billion USD to build the world’s first Flex Mill®, blending the advancements of US minimill technology and the broader product mix superior grade capabilities of an integrated mill. Serving as the company’s spokesperson and chief commercial officer, Mark and the BRS marketing team developed a radical corporate image, brand, and go-to-market strategy. BRS became the world’s fastest profitable steel start-up. Mark has over 25 years of experience in the US steel industry, including roles at Nucor Steel and as a director and board member of a steel pipe and tube trade association. His experience, however, is not limited to steel. Mark served as managing director for a strategic consulting firm where he advised global Fortune 500 companies such as Toyota, Frito-Lay, Boeing, Interbrew and BHP on strategic growth initiatives. He also helped launch an artificial intelligence company focused on industrial safety solutions and remains a member of that company’s advisory board. In connection with his position at H2 Green Steel, Mark, his wife Megan, and his son Tim relocated to Stockholm, Sweden, in 2021. Their daughter, Kathleen, remained in the US, a military cadet at The Citadel, in Charleston, SC.
CARLOS ALBA
CHIEF DIGITAL OFFICER, ARCELORMITTAL GLOBAL RESEARCH & DEVELOPMENT
Carlos Alba is chief digital officer of ArcelorMittal Global Research & Development and leads its digital strategy. Digital core technologies such as artificial intelligence and mathematical optimization are systematically merged with ArcelorMittal’s value chain. This includes manufacturing (and mining), procurement, commercial, supply chain, logistics, finance, strategy and product development. Carlos joined ArcelorMittal Global Research & Development’s corporate team in 2007 focusing his activity on artificial intelligence models applied to business optimization across the company (Europe, Americas and ACIS). He holds a PhD Master Science Degree in Computer Science from the University of Oviedo, Spain, where in parallel he taught for 14 years until 2018 in its engineering schools (computer science, mining, industrial and telecommunications). He is a frequent speaker at international events.
10
FUTURE STEEL FORUM 2022
a b Speakers biographies.indd 2
Steel Times International
11/05/2022 10:41:33
SPEAKER PROFILES
PROFESSOR DR.-ING KATJA WINDT
MEMBER OF THE MANAGEMENT BOARD OF SMS GROUP Born in Bonn, Germany, Katja Windt studied mechanical engineering at Leibniz University of Hannover. She was a visiting scholar at MIT and in 1995 graduated as a mechanical engineer specialising in production technology. She received her doctorate (Dr.-Ing) from the Institute of Production Systems and Logistics at Leibniz University of Hannover and has held a number of senior academic positions, including Full Professor of Global Production Logistics at the Jacobs University, Bremen. Professor Windt has held a number of supervisory board memberships at Deutsche Post AG, Fraport AG and was appointed to the Managing Board of SMS group in 2018. She was awarded Professor of the Year in 2008 from the German Association of University Professors and Lecturers and the Alfred Krupp Prize for Young University Teachers in 2008.
ROBERT VANDLIK
HEAD OF DIGITAL STUDIO, US STEEL KOSICE, SLOVAKIA
JURAJ SABOL GENERAL MANAGER FOR STRATEGY, US STEEL KOSICE, SLOVAKIA Steel Times International
a b Speakers biographies.indd 3
FUTURE STEEL FORUM 2022
11 11/05/2022 10:41:51
SPEAKER PROFILES
MONTI FERNANDO
AUTOMATION NEW TECHNOLOGIES DIRECTOR, TERNIUM Fernando J. Monti is global director of Ternium New Automation Technologies and leads the use of emerging technologies for process control across the facilities of the organization. Mechatronics, applied physics and data science are the main areas of interest for these emerging technologies. Fernando joined Ternium Automation’s corporate team in 2018 to assemble a team capable of investigating, selecting, developing and testing the best suitable technologies for problems that can’t be solved using a conventional approach. From 1996 to 2006 he worked at Tenaris Siderca as automation project manager in Argentina, after that he moved to Italy where he took on the role of automation project manager for Tenaris Dalmine until 2008. Later on he became the automation manager for Tenaris Silcotub until 2010. During that same year, 2010, he moved to the US where he remained until 2016 as automation manager and project director for the Houston area. Later in 2016, he joined Danieli as manager of the electrical and automation division for Danieli Breda, Danieli Centro Tube, where he remained until 2018 before moving to Mexico and starting his current position in Ternium. Fernando holds a Master Science Degree in numerical simulation and control from Buenos Aires University and a degree in electronic engineering from Rosario National University, both institutions located in Argentina.
LORENZO BACCHETTI
SENIOR MANAGER, CRANES AND AUTOMATIC YARD, DANIELI Lorenzo Bacchetti is senior manager of the Danieli Cranes Automation and Automatic Yards business unit. Hleads the sales and technical development of autonomous cranes and automatic yards in Danieli’s portfolio. Automation and Mechatronics are the main areas of interest for this technology. Lorenzo joined Danieli in 2012, covering different roles inside the Danieli Cranes organization, as software engineer, commissioning engineer and manager, cranes automation leader and technical manager for cranes automation and automatic yards.
12
FUTURE STEEL FORUM 2022
a b Speakers biographies.indd 4
Steel Times International
11/05/2022 10:42:05
SPEAKER PROFILES
JUNICHI YOTSUJI
DATA SCIENCE PROJECT DEPARTMENT, JFE STEEL Junichi Yotsuji worked in the Steel Laboratory of JFE Steel from 1992 to 2019 and was primarily working on control and instrument sensors and non-destructive inspection. Currently, he is working in the data science project department of the company on planning, installing and education in data science technologies. Mr Yotsuji holds a PhD in engineering from the University of Tokyo in Japan.
DR. ROMAN STIFTNER MANAGING DIRECTOR, AUSTRIAN NON-FERROUS METALS FEDERATION AND THE AUSTRIAN MINING & STEEL ASSOCIATION Dr Roman Stiftner, CSE, is managing director of the Austrian Non-Ferrous Metals Federation and the Austrian Mining and Steel Association, as well as director-general of EUMICON. He is president of the Austrian Logistic Association (BVL), president of the European Shippers’ Council (ESC), vice president of Euromines and the European Logistics Association (ELA), and a member of the executive committee of European Aluminium and Eurometaux. Roman was CEO of the logistics automation company Dematic GmbH and senior vice president of Siemens AG Austria. He represented as a spokesman transport, infrastructure, environmental and energy affairs as a member of the Vienna State Parliament from 2005 to 2015. Roman studied at the Technical University Vienna and at the Vienna University of Economics and Business and graduated in 1994. He holds an academic doctorate degree from the Technical University Vienna. Since 2018 he has been a Certified Supervisory Expert (CSE) and is a lecturer at the University of Applied Science in Vienna.
Steel Times International
a b Speakers biographies.indd 5
FUTURE STEEL FORUM 2022
13 11/05/2022 10:42:34
SPEAKER PROFILES
SAMBIT BEBORTA
DIRECTOR TECHNOLOGY, LIBERTY STEEL EUROPE Sambit Beborta is director technology for LIBERTY Steel’s primary operations in Europe, operating out of LIBERTY Ostrava in the Czech Republic. Sambit has over 29 years of integrated steel plant experience and before joining LIBERTY in 2020, he worked as the chief technology officer for ArcelorMittal Temirtau in Kazakhstan. Prior to joining ArcelorMittal in 2011, he spent seven years with Safal Steel in Africa as head of manufacturing and sales, following 17 years serving in a variety of integration and improvement roles for Tata’s upstream and downstream facilities in India and across Europe. Sambit holds a Bachelor of Engineering degree and a Diploma in Steel Plant.
DR. ALEXANDER THEKALE TECHNOLOGY CONCEPTS AND DIGITAL SOLUTIONS, PRIMETALS TECHNOLOGIES April 2019 to present day: Head of digital solutions, EA Downstream, Primetals Technologies Germany GmbH. February 2011- March 2019: Special engineeer, models and concepts, EA Downstream, Siemens AG/Primetals Technologies, Germany GmbH May 2006 to January 2011: Doctoral studies in Applied Mathematics, University Erlangen-Nuremberg, Germany. October 2001 to April 2006: Diploma Studies in Applied Mathematics, University Erlangen-Nuremberg, Germany.
14 a b Speakers biographies.indd 6
FUTURE STEEL FORUM 2022
Steel Times International
11/05/2022 10:42:50
SPEAKER PROFILES
DR. THOMAS PFATSCHBACHER, SENIOR VICE PRESIDENT SMART PRODUCTION & DIGITAL TRANSFORMATION, PRIMETALS TECHNOLOGIES Dr. Thomas Pfatschbacher completed his master’s degree in technical physics at the technical university of Linz and his PhD in material science at the University of Leoben and Technical University of Munich. From 1998 until 2007 he worked at voestalpine as senior expert and later head of electrical steel, after that he was business unit manager at Vatron GmbH until 2010. In 2010 he joined Primetals Technologies in Linz as global vice president and head of mechatronics. In 2013 he became the vice president of technology and innovation for casting and rolling, ESP and mechatronics, as well as through process optimization. From 2019 Mr. Pfatschbacher has been acting as senior vice president of smart production and CDO of Primetals Technologies.
BERTRAND ORSAL
DASSAULT SYSTÈMES’ MAIN STEEL EXPERT For 13 years Bertrand Orsal developed breakthrough processes for ArcelorMittal, some of which were patented and published. Then, after two years managing new works on a Seveso site, he came back to steel within the Fives group in 2019. As a digital innovation director, he led the development of an award-wining automatic operations management solution (MES) for flat carbon steel continuous galvanizing lines. Orsal joined Dassault Systèmes in 2021 as steel experience senior director, to lead the offer strategy of the group towards the steel segment.
Steel Times International
a b Speakers biographies.indd 7
FUTURE STEEL FORUM 2022
15 11/05/2022 10:42:59
SPEAKER PROFILES
DR. NILS NAUJOK
PARTNER, OLIVER WYMAN’S ENERGY AND NATURAL RESOURCES PRACTICE Nils is a partner in Oliver Wyman’s Energy and Natural Resources Practice. His work focuses on chemical, engineering and steel industry innovations and technologies. At its core, Nils’ work is always about increasing sustainability and improving operational performance as companies redefine their goals and realign their organizations. Nils supports Oliver Wyman’s clients with in-depth, expert advice coming from over 20 years of industry experience and his solid academic background (he holds a PhD in strategic management from the University of Bamberg and a Master of Science degree in Industrial Engineering from TU Berlin). He emphasizes the importance of creating the kind of entrepreneurial momentum that produces bottom-line results and durable new capabilities.
DIEGO DIAZ FIDALGO
GLOBAL R&D EXPERT (ARTIFICIAL INTELLIGENCE) AT ARCELORMITTAL Diego is a global referent within ArcelorMittal for artificial intelligence, supporting strategic research lines and high-impact projects across Global R&D. Additionally, he has been a researcher at the Business and TechnoEconomic Department (KiN) of ArcelorMittal Global R&D since its inception in 2004 developing artificial intelligence and advanced analytics solutions across the value chain of the steel industry: line scheduling, internal and external logistics, yard management, strategy, purchasing and sales. His main focus is on mathematical optimization, metaheuristics, and machine learning; and to a lesser degree on simulation, algorithmic game theory, and other fields of artificial intelligence. Prior to that, Diego was a postgraduate researcher in the systems engineering and automation department at Oviedo University for two years, working on data-driven predictive models of inclusions in steel stemming from the secondary metallurgy and continuous casting processes. Diego has a PhD in applied metaheuristics. He obtained a master’s degree in industrial engineering from Oviedo University in 2002, and completed the artificial intelligence and advanced control postgraduate programme, also at Oviedo University, in 2004. He was visiting scholar in the electrical engineering department of Stanford University for eight months in 2015, joining the Convex Optimization group led by Professor Stephen Boyd.
16 a b Speakers biographies.indd 8
FUTURE STEEL FORUM 2022
Steel Times International
11/05/2022 10:43:03
SPEAKER PROFILES
YALE ZHANG
PRINCIPAL CONSULTANT, DIGITAL IN METALS, HATCH DIGITAL Yale Zhang is principal consultant at Hatch Digital with responsibility for Industry 4.0 and digital technology consulting and development for the steel and metals industry. His technical expertise focuses on intelligent decision optimization of integrated value chain, big data analytics, discrete event simulation and model-based advanced process control. Prior to joining Hatch, Yale worked in both manufacturing and management consulting roles, including ArcelorMittal and Vale. His global industrial experience includes the iron and steel, nickel and copper, energy and pharmaceutical sectors. He holds a degree in control engineering from Tsinghua University, China.
ENRICO PLAZZOGNA
EXECUTIVE VICE PRESIDENT, DANIELI AUTOMATION SPA Born in Udine and living there today, Enrico Plazzogna graduated in electronic engineering, specialising in industrial controls at Padova University, winning a scholarship from Consorzio Padova Ricerche for a graduation thesis focused on robotic applications. He joined DANIELI as a proposal engineer in 1994 and was then area manager for Europe, then for Middle East and for Eastern Europe and Russia. He was then appointed executive manager (sales) for minimills and turnkey plants. Today he is executive vice president (sales) and a member of the board of Danieli Automation.
Steel Times International
a b Speakers biographies.indd 9
FUTURE STEEL FORUM 2022
17 11/05/2022 10:43:11
SPEAKER PROFILES
DR. STEFAN ALBERS BUSINESS CONSULTANT, PSI METALS After working four years as head of production planning and IT at a PSI Metals customer, Stefan Albers joined PSI in 2007, working in various positions as software engineer, solution architect, project manager and business consultant for planning, scheduling and production execution. He holds a PhD in electrical engineering from Dortmund University.
DR. FALK-FLORIAN HENRICH
FOUNDER & CEO, SMART STEEL TECHNOLOGIES Dr. Falk-Florian Henrich is founder and managing director of Smart Steel Technologies. The company offers AI-based software products to optimize steel production. Leading steel manufacturers apply SST software 24/7 in production. After a very short set-up project, production quality is substantially increased and energy consumption is reduced. Prior to founding Smart Steel Technologies, Dr. Henrich built the high-tech company CeleraOne (2011-2018), established it as the market leader in the paid content sector and sold it to Axel Springer SE. Dr. Henrich holds a PhD in mathematics and contributed substantial research to the theory of loop spaces of Riemannian manifolds and artificial intelligence.
18
FUTURE STEEL FORUM 2022
a b Speakers biographies.indd 10
Steel Times International
11/05/2022 10:43:19
®
With over 50 years in the Steel Industry, we have a wide variety of solutions to keep your mill rolling.
Traversing Restbars allow for quick and accurate pass changes
· Honours existing mill attachment points and guide base, no machining on your stands required · Made from stainless steel and specifically tailored to your mill ensuring perfect fit, operation and longevity
Our roller entry guides keep your product on the pass · Single point, centralized adjustment during operation · 2, 3 and 4 roller configurations · Rigid, stainless steel construction · Broad size range
® spiral cooling system
Maximize your mill speed while meeting your quenching requirements on all bar sizes. Not just controlled cooling, but correct cooling
box configuration
· Fewer surface defects and better scale control · Available in Box, Trough and Restbar-mounted configurations
Contact Us www.fabris.com
info@fabris.com
905.643.4111
trough configuration
SPEAKER PROFILES
TARUN MATHUR
GLOBAL PRODUCT MANAGER, METALS DIGITAL, ABB Tarun Mathur joined ABB in 2006 and has held several positions in research and development, specializing in the development of advanced model-based solutions for process industries. In his current role, Tarun focuses on projects applying new digital technologies to optimize steel plant performance, process and quality. Tarun graduated from the Indian Institute of Technology, Mumbai, and holds a master’s degree in mathematical modeling and process control.
JEAN-PAUL NAUZIN TECHNOLOGY & INNOVATION DIRECTOR, STEEL GLOBAL SALES, FIVES GROUP April 1, 2020 – present: Technology & innovation director, Steel Global Sales, Fives Group Jan 2019 – April 1, 2020: Vice president – marketing & technology, Fives’ Steel Business Unit 2013 – 2018: CEO, Fives KEODS (France) and automotive expert 2002-2013: Metallic material expert, PSA Peugeot Citroën (France)
EDUCATION 1995: Master’s degree in materials science and forming (Sofia Antipolis, France) 1993: Engineer degree in materials science and processing at Mines de Nancy (France)
1996-2001: Metallic material expert, ArcelorMittal (France)
20
FUTURE STEEL FORUM 2022
a b Speakers biographies.indd 12
Steel Times International
11/05/2022 10:43:29
Committed to Client Success As a global leader in system solutions, TMEIC serves industry and social infrastructure worldwide. With almost 100 years of experience in the metals industry, we offer: • Complete mill automation • A full range of low voltage and medium voltage motors and drives for mill stands and all auxiliary applications • Controls hardware and software for Level 1, Level 2, and mathematical models
WWW.TMEIC.COM +1-540-283-2000 2060 Cook Drive Salem, Virginia 24153 U.S.A.
JAPAN | NORTH AMERICA | SOUTH AMERICA | EUROPE | SOUTHEAST ASIA | INDIA | CHINA | MIDDLE EAST | AUSTRALIA
SPEAKER PROFILES
TONY LEIKAS
CEO, PESMEL With an engineering degree in machine automation, Tony Leikas has held various positions in the technology industry areas of engineering, sales and management. Over the past 20 years he has worked for the Finnish company Pesmel and since 2011 has held the position of chief executive officer. His experience in sales combined with a firm technical background has given him a strong basis for leadership of the company. Leikas has been a member of the management group of Pesmel for 15 years. He is also chairman of the board in Pesmel Taiwan and the other subsidiary companies. Leikas is active in the Finnish business and technology fields as a member of the board in the regional chamber of commerce and in the Federation of Finnish Technology Industries.
MARIANA VIALE
APPLICATIONS AND DEVELOPMENT ENGINEER, AMI AUTOMATION Mariana Viale is a computer science engineer from the Technological Institute of Buenos Aires. She holds a master’s degree in robotics from the University of Tsukuba, Japan and has more than 20 years of experience in research, development, and implementation of vision and artificial intelligence systems applied to steelmaking processes. Ternium has implemented some of these developments, such as caster sticker detection, blast furnace torpedo weighting using vision, EAF slag carry over measurement. Development of T-Expert, Real Time Expert for production processes, and optimization of fly tundish using a genetic algorithm. Ten years ago, Mariana joined AMI Automation leading the company’s advanced solutions department. Among the new product developments we can find the Abnormal Water Vapour Detection (AWVD), which uses water vapour prediction models in the EAF comparing their output against the off gas sensor systems; the IoTrode measures graphite electrodes consumption per phase and per heat using vision; and one of the latest developments, the IoConveyor, which measures a ConSteel scrap feed rate in real time, providing EAF energy models with the exact mix of scrap each minute.
22
FUTURE STEEL FORUM 2022
a b Speakers biographies.indd 14
Steel Times International
11/05/2022 10:43:48
IT‘S MORE THAN JUST A MACHINE.
KOCKS rolled ®
focused
R ELIABILITY
unique
S PECIALISTS
ingenious
B ELIEF
Do you want more?
WE MAKE YOUR PRODUCT GOLD
SPEAKER PROFILES
HOLGER STAMM PRINCIPAL, OLIVER WYMAN, DUSSELDORF, GERMANY Holger Stamm works with senior management in the steel and chemicals industries. He specializes in transformation programmes in manufacturing and supply chain, especially as they relate to ESG considerations. Holger supports Oliver Wyman’s clients with his in-depth expertise coming from over 25 years of industry experience with focus on the digital and green transformation of the steel industry. He holds a diploma in physics (laser technology) from RWTH Aachen University and he has worked in both the process industry and strategy consulting.
DR. BERK BIRAND
CEO, FERO LABS
Berk is the CEO of Fero Labs, an industrial process optimization software company based in New York. He is passionate about helping large industrial companies advance their digital transformation goals using explainable machine learning. Birand holds a Ph.D. in electrical engineering and computer science from Columbia University. His academic research focused on optimizing wireless and optical networks with efficient cross-layer algorithms. He developed scheduling algorithms for optimizing cellular base stations in 5G networks and has several patents in IoT systems for resilient fibre-optic networks.
24
FUTURE STEEL FORUM 2022
a b Speakers biographies.indd 16
Steel Times International
11/05/2022 10:43:59
Predictive Modelling to Enhance Quality
• SmartLine, automatic strip processing line control • Eyeron™, real-time quality control system • Virtuo™, thermal performance optimizer
STEEL DIGITAL FACTORY www.fivesgroup.com
SPEAKER PROFILES
DR. LUC BONGAERTS
BUSINESS DEVELOPMENT MANAGER, OM PARTNERS
Luc Bongaerts has been business development manager at OM Partners since 2009. He has a PhD in mechanical engineering, which concentrated on the integration of scheduling and shop floor control of holonic manufacturing systems. It is a subject that is particularly relevant for Industry 4.0 as holonic manufacturing was part of the Intelligent Manufacturing Initiative that focused on autonomous and co-operating agents organising themselves to form agile, adaptive and high-performance production systems for the 21st century. Luc has been active in supply chain management for 25 years. His experience includes several SCM projects at metals companies, focusing on delivering true value through integrated supply chain planning.
26 a b Speakers biographies.indd 18
FUTURE STEEL FORUM 2022
Steel Times International
11/05/2022 10:44:00
Is Your SAFETY Application Is Your SAFETY Solution SMART, Connected and Robust? Pneumatic
•
Hydraulic
•
System Safety Status Predictive Maintenance Signals • Redundant Safety Logic
Hazardous Locations
Electric
Systems
www.rosscontrols.com
The Global Leader in Fluid Power Safety
EXHIBITOR PROFILES
ABB SPONSOR
Website: www.new.abb.com/metals Email: se-metals_mailbox@abb.com ABB is a trusted partner and leading supplier to the metals industry, offering a complete range of products, services and end-to-end solutions that improve productivity, quality, safety and cost-efficiency in iron, steel, aluminium and other metals production processes. With over a century of experience, ABB’s offerings are expertly tailored and highly processspecific, serving the entire industry from material yard and primary manufacturing to casting, rolling and processing lines. Across the whole metals value chain, ABB demonstrates a commitment to optimizing operations with high performance products and digital solutions.
AMI AUTOMATION STAND 7
Phone: +528110014050 Website: www.amiautomation.com Email: info@amiautomation.com AMI Automation is an international automation and control solutions company with a focus on designing, manufacturing, and implementing innovative technology solutions to provide process improvements that help make companies more efficient. Industries served by AMI Automation include steel, mining, cement, pulp and paper, and oil and gas. AMI Automation combines its 30+ years of experience with automation and controls including artificial intelligence to assist customers with the challenges of implementing and upgrading to new technology solutions. In addition to new equipment, AMI can add updates to older existing equipment for cost-effective process improvements, provide repairs and spares, and produce preventative maintenance programmes. Its team of experienced, specialized engineers is available for consulting, custom design, training, and technical support. AMI Automation consists of two groups: the industrial systems group’s focus is on drive solutions while the meltshops solutions group is focused on electric arc furnace optimization.
28 a d Exhibitors profiles.indd 2
FUTURE STEEL FORUM 2022
BSE STAND 12 Phone: +49 7851 877 - 0 Website: www.bse-kehl.de Email: info@bse-kehl.de Badische Stahl Engineering (BSE) – the sister company of steelmaker Badische Stahlwerk GmbH (BSW) – provides technology that is proven and tested in its own melt shop in Germany, where the company produces over 2.2 Mt of steel every single year. BSE is not only operating one of the most productive mills in the world, but also focuses on safety. The holistic safety concept includes hardware solutions geared towards the safety of operators around the melt shop. The latest addition to this concept is the SandMan; installed in January 2021 at BSW. This manipulator for ladle sand filling not only allows for a safe and ergonomical operation, but also ensures process safety and a high opening rate. Historically, the company’s focus has been on equipment around the EAF, adding the TapHoleManipulator (THM) to its portfolio in 2020. In the meantime, BSE sold five units within a years’ time to steelmakers all over the world. With efficiency being its other main focus, BSE also offers modernisation solutions of EAF and secondary metallurgy, environmental technology and engineering services. BSE has proven its capabilities in many EAF steel mills around the world.
BEDA OXYGENTECHNIK ARMATUREN GMBH STAND 14
Phone: +49 (0) 2102 - 910 90 Website: www.beda.com Email: info@beda.com BEDA was founded 50 years ago in a small village near Duisburg in the centre of the German steel industry – the Ruhr area. BEDA became the first company in the world to produce lance holders, safety devices and special valves for oxygen. These new BEDA products offered both better handling and improved safety. Against this background, the company grew steadily, developed patents, and offered Steel Times International
11/05/2022 12:18:07
EXHIBITOR PROFILES
its customers excellent customer service. Thus, BEDA Oxygentechnik became Europe’s leading manufacturer of oxygen lance equipment and is today the world leader in oxygen lance equipment. BEDA currently offers its globally known, comprehensive range of advanced security products and solutions that include: oxygen lancing equipment, ladle stirring solutions, and EAF safety.
DASSAULT SYSTEMES
SPONSOR
Website: www.3ds.com Contact form: www.3ds.com/how-to-buy
DANIELI AUTOMATION STAND 10
Phone: +39 0432 518 111 Website: www.dca.it Email: info@dca.it Danieli Automation is the company, within the Danieli Group, responsible for the transfer of technological know-how from other Danieli technological divisions to end users, supplying the interface between plant process and operator. It’s mission is to provide process automation and control systems for the metals industry covering the wide spectrum of Danieli technology, ranging from iron ore to long and flat products. Software algorithm models, computerized quality and production control systems are developed in house and are the means to transfer Danieli technologies to end users, thanks to the synergistic relationship with Danieli. This co-operation with the mechanical designers has led to optimised and standardised solutions, resulting in better performances and quicker plant start-up timings. The motors and drives department designs and selects the best solutions for motors and their respective drives, dependent upon the technical requirements. Daniel Automation designs and supplies complete electrical distribution systems, with installation engineering based on extensive experience. In addition, the company produces special instruments and sensors, designed and engineered to meet the demand from steel producers for sophisticated controls, quality certification, cost optimisation and quick adaptation to the latest systems. Finally, DIGI&MET is the cross-functional business unit Danieli Automation has created to develop and implement new plant design concepts, based on digital innovation, servitization, and outcome economy principles, to ensure consistency in quality, plant utilisation, OpEx and faster deliveries.
Steel Times International
a d Exhibitors profiles.indd 3
Dassault Systèmes provides businesses and people with virtual universes to envision sustainable innovations. These leading-edge solutions, powered by the 3DEXPERIENCE® platform, are driven to understand the impact of decisions on the environment and economy, and facilitate decision support, scenario management, and innovation, in order to reimagine new business models and ways of working—without compromising real-world operations. Dassualt Systèmes’ solutions are capable of modelling the entire metals value network in order to optimize companies’ business goals, supporting them to tackle the current market volatility with consideration to the environment. Customer service, capacity utilization, inventories, production alternatives, and cost and sustainability are all part of the DELMIA QUINTIQ solution suite for the metals industry in the form of key performance indicators, which can then be leveraged to make trade-off decisions during planning and reporting.
DR. SCHENK GMBH STAND 19
Phone: +49-89-85695-0 Website: www.drschenk.com Contact form: www.drschenk.com/contact.html Dr. Schenk GmbH offers inspection and measurement solutions for automated quality assurance and production process control – a key success factor in the making and converting of plastics, non-wovens, textile materials, paper, metal, and glass, for a multitude of markets such as display glass, automotive, packaging, medical, renewable energy, and many more. From modular standard units to highly customised systems – Dr. Schenk’s solutions have your material in focus!
FUTURE STEEL FORUM 2022
29 11/05/2022 12:18:16
EXHIBITOR PROFILES
ENDRESS+HAUSER STAND 20
Phone: +41 61 715 7700 Website: www.endress.com Email: info@endress.com Founded in 1953 by Swiss engineer Georg H Endress and German banker Ludwig Hauser, Endress+Hauser has been a reliable partner for the process industry for 69 years, providing comprehensive support to customers, from the laboratory to the process. The company‘s core expertise lies in the fields of process instrumentation and laboratory analysis. With products, solutions and services, Endress+Hauser helps its customers design safe, reliable, efficient and economically friendly processes across the entire life cycle. The company maintains a close presence to customers worldwide, with a large network of sales centres, as well as select representatives – ensuring competent support around the globe. Production centres on four continents ensure fast and flexible delivery to its customers, wherever they are located. The Endress family, the sole shareholder since 1975, plays an influential role in the company’s development to this day, with a key objective: Endress+Hauser shall remain a successful family company.
FERO LABS STAND 23
Website: www.ferolabs.com Email: info@ferolabs.com Fero offers factory optimization software that enables steel manufacturers to stay ahead of ever-changing production conditions. With intuitive software guiding them, engineers can confidently take informed action that gets results without wasting time or resources. Powered by machine learning that makes decisions clearer, Fero gets leading global steelmakers better production results in less time. With Fero, you see every situation – and every solution – more clearly.
30 a d Exhibitors profiles.indd 4
FUTURE STEEL FORUM 2022
FIVES STAND 6
Website: www.fivesgroup.com Fives, a global industrial engineering group, provides process expertise, advanced technologies, and digital solutions, supporting steel manufacturers’ ability to achieve the ultimate in performance excellence. Within the steel industry, Fives is renowned for: � Digital solutions. � Process expertise. � Ultra-low NOx reheating furnaces. � Cold rolling mills for carbon, stainless and silicon steels. � Complete strip processing lines, including cutting-edge galvanizing lines for automotive applications. Fives has been recognised as a ‘Technology Provider of the Year’ by Fastmarkets for several breakthrough technologies: � FlashCooling® for ultra-rapid cooling of AHSS and UHSS � CELES EcoTransFlux™, a high-power transverse flux induction technology for rapid heating cycles. � Eyeron™, an intelligent qualification system. � SmartLine, a fully automatic strip processing line control based on predictive modeling.
FRIGORTEC STAND 24
Phone: +49 7520 914 82 0 Website: www.frigortec.com Email address: info@frigortec.de For over 50 years, FrigorTec has been providing solutions around the world for grain cooling as well as CRANEFRIGORTM crane cooling. FrigorTec produces the CRANEFRIGORTM crane air-conditioning devices for crane cooling and crane cabin cooling in hot operating environments like smelting works, foundries, and mills. The devices, which operate world-wide, are available in many installation variations and with various output stages. The series of CRANEFRIGORTM crane cooling
Steel Times International
11/05/2022 12:18:25
EXHIBITOR PROFILES
devices and crane cabin cooling devices cover a wide range of cooling capabilities, with cooling possible at ambient temperatures of -40°C to 90°C. The solutions are designed for extreme conditions and corrosive environments. With CRANEFRIGORTM cooling devices, FrigorTec offers a wide range of operating voltages, with a UL/CSAcompliant version also available. Several thousands of CRANEFRIGORTM cooling devices are currently operating successfully around the world.
IMS
systems for quality control and software tools since 1984. As a member of the global operating ISRA VISION group, the world’s largest and most qualified supplier of surface inspection solutions, ISRA PARSYTEC supports the steel industry for full quality monitoring throughout complete production and processing – from 3D inspection of slabs and plates to inline inspection for hot rolling, blanks and plates. The company’s systems are easy to integrate at every production step, from the slab to the coated sheet. The collected data not only provides a solid decision-making basis for enhancing quality and processes but also leads to waste reduction. ISRA’s yield management solutions support the entire quality cycle in metal production, detecting defects, automatically classifying final product quality, and simplifying the control of process quality and yield. This comprehensive reporting resultantly aims to ensure maximum transparency.
STAND 4
Phone: +49 (0) 2056 / 975 - 0 Email: info@ims-gmbh.de Websites:www.ims-gmbh.de www.ims-experts.com Precision out of passion, quality out of conviction and innovation out of tradition is the IMS mantra. The company has been manufacturing x-ray measuring systems, isotope measuring systems and optical measuring systems since 1980. Its non-contact detection systems are used in the steel, metal and aluminium industries wherever material testing is required to guarantee the highest standards of quality under the toughest operating conditions. In hot production, such as continuous casting plants, hot rolling mills and tube mills, where shimmering surfaces, heat, dirt and moisture are common, as well as in cold rolling mills and service centres, measuring systems from IMS measure and detect with precision. Exactly reproducible measurements and evaluations optimize production lines in real time, while simultaneously reducing production costs and reject rates.
KELLER ITS STAND 25
Phone: +49 (0) 5451 850 Website: www.keller.de/en/its/ Email: its@keller.de Since 1967, KELLER ITS (infrared temperature solutions) – a business unit of KELLER HCW – has been developing and producing precision measuring instruments and application solutions for optical temperature measurement in industrial applications ranging from -50 to 3500 °C. Today, KELLER ITS is one of the market leaders for infrared thermometers and pyrometers worldwide, setting the standard with its durable measuring systems. KELLER ITS’ broad product portfolio of more than 350 models includes stationary and portable pyrometers, infrared temperature switches, complete measuring solutions for customer-specific tasks, applications and application-oriented software and accessories.
ISRA PARSYTEC STAND 11
+49 (6151) 948-0 Email: info@isravision.com Website: www.isravision.com ISRA PARSYTEC has provided automatic surface inspection
Steel Times International
a d Exhibitors profiles.indd 5
FUTURE STEEL FORUM 2022
31 11/05/2022 12:18:31
EXHIBITOR PROFILES
KÖPPERN
Among the customers of the company, which today employs more than 300 people, are steel makers such as Baosteel, Posco, ThyssenKrupp and Vallourec. LAP staff support the company’s worldwide customer base from headquarters in Lüneburg/Germany and through an international network of branches and technical agencies.
STAND 26
Phone: +49 (2324) 207-0 Website: www.koeppern-international.com Email: info@koeppern.de Founded in 1898 in Hattingen, Germany, Köppern remains a family-run enterprise reflecting its traditional values of technology leadership and highly dependable manufacturing quality, coupled with a unique regard for the individual needs of its customers. Köppern’s global network of subsidiaries provides customised service on all continents. Köppern has also successfully expanded into the cement industry, building a strong position through innovative press designs as well as the unique HEXADUR® wear protection system. Köppern has sold several hundred roller presses in over 60 countries for briquetting, compaction and comminution. In collaboration with its customers, optimized processes have been developed by Köppern’s dedicated staff of process engineers and tested in its own pilot plant, backed up by trials involving more than 2,000 different materials.
LAP STAND 27
Phone: +49 4131 9511 12 Website: www.lap-laser.com Email: d.meuser@lap-laser.com Since 1984, LAP has supplied laser-based systems for highprecision measurements of geometric dimensions, such as width, thickness, length, diameter and flatness. LAP systems excel at providing ultra-precise measurement results under the harshest operating conditions. Hundreds of LAP systems are tried and tested every day in steel and rolling mills worldwide. The company’s aim is to save resources in the production process of long products with intelligent laser-based measuring gauges for automatic inline detection of rolling defects. The LAP solution has been designed to support the aims of achieving higher production efficiency, of resource savings through scrap minimisation, and of reduced energy consumption in the overall production processes.
32 a d Exhibitors profiles.indd 6
FUTURE STEEL FORUM 2022
MIDREX STAND 8
Phone: (704) 373-1600 Email address: info@midrex.com
Midrex Technologies is an international process engineering and technology company providing steelmakers with commercially proven solutions for greater profitability and has been the leading innovator in the direct reduction of iron ore for more than 40 years. The company offers eco‐friendly technologies for ironmaking that provide high productivity, outstanding product quality, and cost competitiveness. Midrex has built its foundation upon the MIDREX® Direct Reduction Process that converts iron ore into high‐purity direct reduced iron (DRI) and hot briquetted iron (HBI) for use in steelmaking, ironmaking, and foundry applications. Each year since 1987, MIDREX® plants produce about 60% or more of the world’s DRI and 80% of the world’s DRI produced by means other than rotary kilns. Midrex continues to develop innovative DRI solutions that improve their new plant designs as well as support their existing plants. It is this continual support and improvement that has made the MIDREX® process unsurpassed in the direct reduction industry. The company’s headquarters and technical centre are located in North Carolina, US, with offices in the UK, China, and India.
NEXTSENSE STAND 10a
Phone: +43 (0) 316 232400-0 Website: www.nextsense-worldwide.com Email: office@nextsense-worldwide.com With its unique CALIPRI® principle, NEXTSENSE is a world
Steel Times International
11/05/2022 12:18:41
EXHIBITOR PROFILES
leader in mobile profile measurement and surface inspection, especially in the fields of wear measurement in the railway industry, gap and flush measurement in the automotive industry, and the measurement of hot metal profiles. Its customers include all major railway companies, numerous well-known international automobile manufacturers and major steel producers, including Deutsche Bahn, SNCF, China Railways, Daimler, JLR, Audi, BMW, voestalpine and ArcelorMittal. NEXTSENSE was founded in 2007 as a spin-off of the Austrian research company Joanneum Research. With 90 employees, more than 40 sales partners, and sales and service centres in Atlanta (US) and Shanghai (China), the high-tech company is represented all over the world. Since May 2018, Nextsense has been part of Hexagon AB, a leading global provider of information technologies.
OPTRIS STAND 18
Phone: (603) 766-6060 Website: www.optris.com Email: sales@optris-ir.com Established in 2003, Optris is a leading innovative company in the field of non-contact temperature measurement and infrared radiation. The product range covers portable thermometers, fixed industrial thermometers and infrared cameras.
PESMEL OMP STAND 3
Phone: +32 3 650 2211 Website: https://omp.com Email: LBongaerts@ompartners.com OMP helps companies facing complex planning challenges to excel, grow and thrive by offering the best digitized supply chain planning solution on the market. Its Unison PlanningTM concept has a unique approach, handling all supply chain planning challenges in a unified way, and synchronizing planning stages, horizons, functions and roles. The combination of services and technology boosts collaboration throughout any value chain, from forecasters to schedulers, and from business leaders to technology experts. Unison PlanningTM is a cloud-based, out-of-the-box solution for industry-specific challenges. Hundreds of customers in consumer goods, life sciences, chemicals, metals, paper and packaging utilise Unison PlanningTM in order to make the right decisions at the heart of their business. Valued as a thought leader by experts such as Gartner, OMP invests one out of every three dollars earned into innovation.
Steel Times International
a d Exhibitors profiles.indd 7
STAND 9
Phone: +358 20 7009 600 Website: https://pesmel.com Email: pesmel@pesmel.com Pesmel has more than 40 years of experience in delivering solutions that improve material flows and logistics at different types of manufacturing facilities. The company focuses on serving customers around the world in the pulp and paper, metals and tyre manufacturing industries. At the heart of Pesmel’s offering is the unique Material Flow How® concept which covers the customers’ entire spectrum of needs from smart, customer-centric design and engineering to active lifecycle service and maintenance. The right combination of intelligent packaging and handling solutions along with a one-of-a-kind automated high bay storage is always custom-built to meet customers’ specific needs and requirements. Pesmel is focused on bringing customers benefits that materialise through improved space efficiency and material flows between processes and shipping, work safety, enhanced loading and shorter turnaround times, as well as quality and accuracy of deliveries. Pesmel is committed to continuous development and engineering excellence, with operations being based on five cornerstones: � Innovation � Engineering know-how � System know-how � Tailoring and flexibility � Life-long support and services
FUTURE STEEL FORUM 2022
33 11/05/2022 12:18:45
EXHIBITOR PROFILES
PRIMETALS
QUANTOLUX
STAND 17
STAND 5
Primetals Technologies, Limited, headquartered in London, UK, is a world leader in the fields of engineering, plant building, and the provision of lifecycle services for the metals industry. The company offers a complete technology, product, and services portfolio that includes integrated electrics and automation, digitalization, and environmental solutions. This covers every step of the iron and steel production chain—from the raw materials to the finished product – and includes the latest rolling solutions for the non-ferrous metals sector. Primetals Technologies is a joint venture of Mitsubishi Heavy Industries and partners, with around 7,000 employees worldwide.
Quantolux is an expert in optical emission spectroscopy (OES), with its products being used to perform elemental analysis on all kinds of materials, and is determined to drive customer success with innovative analytical solutions. With decades of OES experience, QuantoLux designs, produces, and services high-end laser spectrometers for various applications worldwide. QuantoLux’s products enable customers to: � earn money on the scrap yard by improving steel scrap quality with stable, radiation free analysis � establish in-situ process control in the melt shop � take advantage of ultra-high speed, safe and locationindependent identification (PMI) Furthermore, the laser enables extraordinary long-term stability, low argon consumption, and cross-contaminationand maintenance-free operation.
Phone: 020 8996 4300 Website: www.primetals.com Email address: contact@primetals.com
Phone: +49 (0) 2821 899 399-0 Website: www.quantolux.de Email: info@quantolux.de
PSI SPONSOR
Phone: +49 211 60219-271 Website: www.psimetals.de/en/met-home/ Email: info@psimetals.com PSI is the leading partner for digital production in the metals industry. Its software solutions enable producers of aluminium and steel products to ensure their competitive edge by delivering products as agreed in quantity, quality and time while considering inventory, productivity and performance targets. The PSI metals software line is an end-to-end approach for the overall supply chain caring for all the needs of the primary metals industry. From supplier to customer, PSI metals offers powerful and highly configurable standard products to support all processes from planning to execution while respecting the complexity of metal production. Combining 45 years of experience in implementing production management software with innovation, PSI supports numerous metals producers around the globe in achieving their competitive edge.
34 a d Exhibitors profiles.indd 8
FUTURE STEEL FORUM 2022
QUINLOGIC STAND 22
Phone: +49 (2405) 47 999 40 Website: www.quinlogic.com Email: info@quinlogic.de QuinLogic was established in Aachen in 2008 and is operating within SMS digital as a wholly owned subsidiary of SMS group since 2019. To enable industrial mass production and at the same time highest flexibility, SMS-QuinLogic developed software products by applying the latest generation of computing power and storage capabilities. Based on this technical evolution, SMS-QuinLogic provides an ‘easy to use’ quality assurance solution for steel and aluminium production facilities. The SMS-QuinLogic software secures reliable quality for challenging markets such as automotive, aviation, wind energy, by producing high-strength and wear-resistant material grades. It also provides proactive support in order Steel Times International
11/05/2022 12:18:51
EXHIBITOR PROFILES
to immediately address production problems and, therefore, minimises deficiencies in customer products. Its seamless communication interfaces to existing gauging systems allows fully integrated systems in all existing but also new production lines. SMS-QuinLogic focuses on key value drivers such as: improving revenues, reducing cost, and increasing yield. The company combines its experience with digital assistance, and provides solutions that make more out of its customers’ quality and process data.
SAP
the transformation towards intelligent AI-supported optimised production. SST delivers AI software products that boost quality, improve energy demand and ensure accurate management of CO2 efficiency. Leading steel manufacturers in Germany and abroad successfully use SST software throughout the production process. The range covers the production of high-quality automotive exposed grades to maximizing efficiency in the construction steel sector. SST excels in a portfolio of professional AI-powered optimization packages, which lead to a permanent performance increase of 5-10% per process stage in 24/7 use. At the same time, SST accompanies the steel manufacturer from the integration of the software to the complete achievement of any optimisation goals with a high service level. To make this happen, SST relies on its well-co-ordinated team of outstanding metallurgists, process experts and AI specialists.
STAND 13
Phone: 0800 0852 631 Website: www.sap.com SAP has 40 years’ experience working closely with hundreds of metals manufacturers across the globe, using digital innovation to anticipate real-time demand, and enhancing processes for operational efficiency. In addition, SAP can help metals companies manage stakeholder expectations in producing environmentally responsible products and services, reducing GHG emissions, minimizing waste, and using socially responsible business practices across product lifecycles and value chains. As the market leader in enterprise application software, SAP helps companies of all sizes and in all industries run at their best. SAP customers generate 87% of total global commerce. The company’s end-to-end suite of applications and services enables its customers to operate profitably, and adapt continuously. SAP aims to simplify technology for companies so they can consume software the way they want – without disruption. With a global network of customers, partners, employees and thought leaders, SAP both helps the world to run better, and improves people’s lives.
SMS STAND 21
Phone: +49 2161 350 1603 Website: www.sms-group.com Email: simone.grimm@sms-grop.com The SMS group unites global players in plant and machinery construction for processing steel and NF metals, operating under the roof of SMS Holding GmbH. The family-owned company – now run by the fourth generation – stands out with its strong market position and corporate culture of responsibility, together with highperformance products and services tailored to individual customer requirements. The SMS group combines the flexibility of medium-sized company units with the vast resources of a global group. The company builds on the continuous training and comprehensive expertise of its employees to develop groundbreaking technologies, as well as electrical systems, and automation services. The SMS group is your partner for new plant and machinery as well as modernisations and upgrades.
SMART STEEL TECHNOLOGIES STAND 1
Phone: +49 30 403673720 Website: www.smart-steel-technologies.com Smart Steel Technologies (SST) supports the steel industry in Steel Times International
a d Exhibitors profiles.indd 9
FUTURE STEEL FORUM 2022
35 11/05/2022 12:18:56
EXHIBITOR PROFILES
TEAMVIEWER
delivery reliability being a key part of their approach. Tebulo is an ambitious and driven employer and offers its team a challenging international working environment with room for personal input and development. The company’s team consists of qualified and customer-oriented professionals who strive for the best result, as taking responsibility and constantly improving is in Tebulo’s DNA.
STAND 15
Website: https://www.teamviewer.com/ TeamViewer is a leading global technology company that provides a connectivity platform to remotely access, control, manage, monitor, and repair devices of any kind – from laptops and mobile phones to industrial machines and robots. Although TeamViewer is free of charge for private use, it has more than 625,000 subscribers and enables companies of all sizes and from all industries to digitalize their business-critical processes through seamless connectivity. Against the backdrop of global megatrends like device proliferation, automation and new work, TeamViewer proactively shapes digital transformation and continuously innovates in the fields of augmented reality, Internet of Things and artificial intelligence. Since the company’s foundation in 2005, TeamViewer’s software has been installed on more than 2.5 billion devices around the world. The company is headquartered in Goppingen, Germany, and employs around 1,500 people globally. TeamViewer AG (TMV) is listed on the Frankfurt Stock Exchange and belongs to the MDAX.
TMEIC STAND 2
Phone: +44 (0)3300 58 44 60 Website: www.tmeic.com/industry/metals TMEIC is an industry leader in steel mill automation and drive systems, with almost 100 years of experience in process control and automation technology for hot mills, cold mills, long products, and processing lines. Technical capabilities include complete engineering and design, mill modernisations and revamps, simulation and testing, and comprehensive project management. Site services include mill audits, start-up, and commissioning. TMEIC’s complete range of steel mill automation products includes large AC and DC main drives and motors, auxiliary drive equipment, level 1 controllers, level 2 process models, and HMI systems.
TEBULO INDUSTRIAL ROBOTICS STAND 16 Phone: +31 72 20 05 500 Website: www.tebulorobotics.com Email address: info@tebulorobotics.com Tebulo Industrial Robotics is the leading and innovative specialist in high-tech robot integrations, working closely with top players in the steel, aluminium, and construction industry, and the sustainable energy sector. Tebulo’s expert and multidisciplinary teams prioritise an efficient approach, with clear advice, realization and aftersales, always seeking the most effective solution to make robots work for their customer’s businesses. This way, customers are assured of a significant improvement in safety and productivity. In addition, Tebulo takes care of optimization and ‘troubleshooting’ throughout the entire lifespan of their products, with sustainability and high
36 a d Exhibitors profiles.indd 10
FUTURE STEEL FORUM 2022
Steel Times International
11/05/2022 12:19:01
ARTIFICIAL INTELLIGENCE
Succeeding with AI in steel manufacturing
By deploying AI in the right way and selecting the right projects, steel manufacturers can gain critical insight into continuous real-time processes to unlock hidden value at speed and scale. By Shreebhooshan Badrinarayan* AI applied to solve complex operational problems can not only improve availability, quality and performance, but can also support plant optimization and continuous improvements in production processes. And yet, despite being widely accepted as a game changer for the steel industry, many organizations are still struggling to make notable progress with the technology onground. This is primarily because AI projects are perceived to be highly complex, time consuming and expensive, preventing most projects to scale beyond the proof-of-concept stage. The slower adoption of AI is somewhat understandable – given the fact that many mills are running for several decades, leading to deeply entrenched processes. However, there is no doubt that leaders who are embracing AI and ML (machine learning) are the ones gaining a competitive
advantage; and it is, therefore, important that organizations start walking now before sprinting for benefits later. Over the years, we have developed a deep understanding of the challenges from the steel manufacturer’s point of view and have helped the industry address them. Below are our responses, following research and analysis. How to manage the data deluge? Today huge amounts of data are being generated by the automation of steel plant operations. Modern integrated steel plants produce over 15 trillion data points every year, yet only a small fraction of this data is used to both understand the operational bottlenecks and solve them. Forensic methods help understand some problems post facto, but can be time consuming and unscalable. Another approach organizations often
take is to concentrate on building data lakes first, and relegating plans for the utilization of that data to some nebulous point in the future. By the time the organization aligns on how the data is to be used, the data might be rendered unusable because it is no longer relevant, or it is realized that it lacks context. Hence, there needs to be a shift from collecting and storing the data to drawing facts and actionable information out of the available data using advanced analytics. In order to harness time-series data that contains granular information on the physical performance of the mill, we need to have a system that can analyze vast amounts of data in real time and provide instant insights that are actionable for the end users. For instance, in a run-out or hot run table, hundreds of motor-driven rollers work in tandem to move the steel plate or strip while it cools down. The
*Senior manager, Falkonry
38
FUTURE STEEL FORUM 2022
Falkonry.indd – read MM..indd 2
Steel Times International
09/05/2022 07:35:48
ARTIFICIAL INTELLIGENCE
slightest deterioration of even one motorroller could introduce surface defects into the passing slab of steel. It is impossible for a human to analyze and process the sensor and motor data for all the rollers in real time. An automated AI engine, on the other hand, can ingest the data of the entire line and look for emerging patterns that are associated with failure events. By putting the available data to use, the frontline personnel get a real time view of the operations and become aware of anomalies before a major failure can occur. Alternatively, we also hear from our customers that their mill is producing too little data for AI. Contrary to popular belief, AI does not need large amounts of data or specialized big data infrastructure. What is needed instead is good quality data that corresponds to the asset or line that is to be monitored. Where to start? A common conundrum when starting an AI project is finding the right application area or the ‘use-case’ for AI to start monitoring. Even digitally mature organizations find this difficult as it is not straightforward to figure out which use-case will translate to the highest business value in the future. Often the use-case focuses on addressing only the frequently occurring failure modes, instead of also discovering the rare and novel failures. The pressing question when starting with any use-case is – what if the complexity of the selected use-case does not allow us an opportunity to prove success?
A solution to this challenge is to shift from a use-case-based approach to a ‘monitor-first’ approach. By connecting the available data of the entire line to an advanced analytics system like Falkonry, various plant stakeholders get an aggregated view of entire plant operations at one place. This approach helps by casting a plant-wide net in catching the recurring as well as novel anomalies in the now. For instance, a continuous caster consists of various components and sub-components, each with a different nature of operation. A mold oscillator operates continuously, whereas a shear for slab cutting operates periodically. Each of these components capture data at different sampling rates, ranging from milliseconds to seconds
depending on the operations. By connecting the data from the entire continuous caster to the time series AI, Falkonry is able to develop over 50 different machine learning use-cases and monitor the health of every component in real time. As opposed to the conventional approach, this approach minimizes the AI implementation risk, which could stall the value realization. (Fig 1) What to expect? Operational visibility: The foremost outcome from the intelligence-first approach is to get operational visibility of the entire line or plant by avoiding the disruptive blindspots. By working in the now, it becomes possible to visualize plant scale trend analysis by scoring
Fig 1. Falkonry enables smarter decisions with AI on time series data
Steel Times International
Falkonry.indd – read MM..indd 3
FUTURE STEEL FORUM 2022
39 09/05/2022 07:36:19
Our innovations aiming at decarbonizing the steel industry: CO2-free steel production process Siderwin, Furnace Electrification, Carbon Capture, Heat Recovery, Jet Vapor Deposition (JVD), E-Qual® Electrical Steel Processing Technologies
TrainLab™ Virtual reality & immersive e-training platformfor considerably improved learning retention.
JOHN COCKERILL INDUSTRY METALS Reheating Furnaces Pickling Acid Regeneration Cold Rolling Strip Processing Automation
Eagle Eye Coating® A unique and intelligent zinc coating technology.
Technologies for cleaner and smarter steel making Innovation is part of our Group’s DNA. For over 200 years John Cockerill Industry has demonstrated its ability to think differently and its capacity of seeing things from another angle. Still today, we are combining talent and technology, history and modernity, engineering and services. Virtues that help us to accompany our steelmaking clients in their digital transformation journey and their transition to responsible, zero carbon steel making.
Follow us on
johncockerill.com/industry
ARTIFICIAL INTELLIGENCE
each and every signal against anomalous and reference trends. For instance, with one of our customers, our AI was able to focus the attention of reliability engineers to problems that were not known issues. For a particular sub-system in question, the known issues were electrical or mechanical. The AI surfaced an alert with an explanation that indicated an issue with hydraulics. Upon inspection, a leaking valve was discovered, which was not influencing production at that time, but if the attributes of the finished product (steel width in this case) were to change, that leaky valve would have caused problems. Maintenance engineers acted upon this and prevented downtime. Ease of use: The automated nature of AI hides the complexities of the data management and analysis from the end user, while presenting the insights to augment human decisions. These insights should be accompanied by explanations that enable Steel Times International
Falkonry.indd – read MM..indd 5
faster root cause analysis and prioritization of actions. The engagement of the subject matter experts is crucial in understanding and solving any production challenges. By putting the next generation of AI capabilities directly in the hands of these end-users, the human decisions and actions are enriched by the fact-based insights provided by AI. Proactive data-driven mindset: A capable AI engine should be able to unearth facts from the data that can be utilized in bringing positive changes in the plant operations. An AI-identified anomaly does not limit its attribution to failure modes for the maintenance team. One must ask the right questions: Is there something else that an AI-identified anomaly can mean? Is the sensor difference in value related to loss of calibration? Is there a product quality implication we are not seeing because we are focusing on predictive maintenance? AI facilitates the analysis of all these unusual
conditions before they impact the operation, and aids in the continuous improvement of the plant production operations. �
References: C. Waters, “Insights into Steel Industrial Transformation” Falkonry Inc., 16 Feb 2022. [Online] Available at: https://falkonry. com/blog/insights-into-steel-industrialtransformation-with-crick-waters/ C. Lee, “Database-First vs Intelligence-First: The Cart Before the Horse” Falkonry Inc., 10 Jun 2020. [Online] Available at: https:// falkonry.com/blog/database-first-vsintelligence-first-the-cart-before-the-horse/ C. Waters, R. Talla, P. Jain, N. Mehta, “Transforming Metal Production by Maximizing Revenue Generation With Operational AI” Falkonry Inc., 29 June 2021. [Online]. Available at: https://info.falkonry. com/aistech-technical-paper FUTURE STEEL FORUM 2022
41 09/05/2022 07:36:52
ARTIFICIAL INTELLIGENCE
Beginners guide to AI in the metals industry Technological advances have developed rapidly, and artificial intelligence, once a little understood technology, is now allowing businesses to streamline data, create seamless interactions between machines and humans, and drive effective results. By Humera Malik* The world is swiftly transforming to an era where digital technologies will be the catalyst to safeguard long-term business resilience. The pandemic has pushed businesses to accelerate their digitalization goals, and if anyone had any doubts about the importance of building technology capabilities, the unprecedented events of the last two years have put them to rest. A recent survey, titled The new digital edge: Rethinking strategy for the postpandemic era conducted by McKinsey, affirmed that the COVID-19 pandemic has fundamentally changed the pace of business, and companies with superior technology capabilities will leapfrog others. Many of the survey respondents identified that their companies’ business models are becoming outdated. According to the survey, only 11% trusted their current business models to be economically viable through 2023, while another 64% said their companies needed to develop new digital businesses to help them get there. According to McKinsey, metal players across the globe are being forced to take a fresh look at their business strategies to align with growing demand for carbon-friendly steel products, tightening carbon emission regulations and increasing public and investor interest in sustainability. Moreover, with fluctuating market demand, shrinking margins and ageing technology, metal producers are finding it challenging to maintain efficiencies.
Addressing challenges in the metals industry Data is the new lifeblood of organizations. With rapid growth in digital investments, industrials are looking to tap into Industry 4.0 by keeping data at the centre of their operations. According to IDC, metal companies generate an estimated 1.3TB of data per day, which is expected to nearly double by 2025. However, while collecting data is the first step, its true potential can only be realised if companies are able to derive valuable insights that can drive decision making, thereby improving performance and productivity. Advanced AI solutions can equip companies to streamline their data, allowing
them to forecast any changes in their operating environment in real-time to predict machine failure, and control processes to keep production within specification. While more and more companies are investing in digital solutions, they are struggling to execute at speed. Ambiguous vision, poor data quality, outdated technology infrastructure, and scarcity of talent are some of the reasons for this. For organizations to realise the true potential of AI and maximize the return on investment of their digital investments, it is essential to create an operational framework that brings together people, processes, and technology on a common platform.
* CEO, CANVASS Steel Times International
Canvass.indd – read MM.indd 3
FUTURE STEEL FORUM 2022
43 09/05/2022 07:40:40
ARTIFICIAL INTELLIGENCE
How can manufacturers accelerate their time to AI impact? Optimizing assets: With AI, metal producers can ensure that their assets are operating at the most effective operating range and that uptime is being optimized. AI can also help monitor the health-status of different assets by identifying key parameters that can signal a maintenance issue, so organizations are ready in advance to reduce unplanned downtime. Optimizing processes: Manufacturers use a diverse set of processes to make the final product. All these processes send a wide range of data to the cloud at every stage of manufacturing. An integrated AI solution can help analyze these large-scale datasets and get visibility into production processes in real time, which can result in smarter decision making. Optimizing costs: How metal manufacturers leverage their utilities cost effectively can dramatically reduce their production costs. AI can help pinpoint the optimal resources required in order to reduce operating costs, improve margins and maximize production yield. How to create a winning AI strategy The successful deployment of AI requires companies to take a broader, businessdriven view and create an ecosystem where machines and humans are able to seamlessly interact within working environments. It demands long-term effort in reskilling employees and enhancing the learning curve to drive impactful results and maximise a return on investment. According to Gartner, organizations should make efforts to gain people’s trust to succeed in AI. “Technological advances are often historically associated with a reduction in staff headcount. While reducing labour costs is attractive to business executives, it is likely to create resistance from those whose jobs appear to be at risk. Organizations can miss out on real opportunities to use the technology effectively in pursuing this way of thinking.” Why AI projects fail: Lack of collaboration: The majority of AI projects fail because of a lack of coordination between data scientists, SMEs, and operations staff.
44
FUTURE STEEL FORUM 2022
Canvass.indd – read MM.indd 4
Poor data strategy: Before setting objectives, enterprises should prepare a clear data strategy unique to their business case. Even if an organization is generating a lot of data, it may not be relevant for the processes or use case where the transformation is needed. No end goal in sight: The data collected needs to be relevant for a particular use case or for a problem an organization is trying to solve. It is important to identify the end objective of the AI transformation for the outcome to be measurable. So, how can organizations successfully deploy AI? How can businesses construct agile, integrated, simplified, and smarter AI programmes to achieve their end objectives? Five steps for a successful AI transformation Set realistic expectations: One of the primary reasons for the failure of AI projects is the unreasonable expectations set by leadership. Businesses need to look at AI
realistically as a strategic enabler in achieving overall business objectives. It is important to examine what you are doing, what the purpose is, and what the various limitations are. Identify the right use cases: Not all functions are conducive to AI. For organizations to advance in their AI journey and work smartly, data-driven business decision-making models are essential. The processes should enable algorithms to compare current performance with past performance and identify the opportunities to improve. Most importantly, AI implementation should not compromise an existing process or develop unnecessary unforeseen risks. Selecting the right use case sets the ball rolling for your AI roll-out. Analyze the quality of data: AI processes run on data and having the right data set that is representative of a particular use case is essential. According to an IDC survey, the lack of relevant data remains a significant challenge. Data security, governance, performance, and latency (transfer rate) are the top data integration challenges. Just because a process delivers plenty of data does not mean the data variables are helpful for AI implementation. In many AI deployments, often the complexity of the data is misunderstood. And this becomes the primary reason why AI projects fail before they even get started. Empower your subject matter experts: Not every organization has the capability to build AI expertise. Implementing AI requires looking extensively at a massive amount of
Steel Times International
09/05/2022 07:40:43
ARTIFICIAL INTELLIGENCE
data to find and define problems and then design a model to address these challenges. However, shortage of data science talent and a lack of skilled AI professionals is a key concern. Choose a platform that is designed to address this skill gap by empowering engineers and SMEs to leverage AI capabilities without any coding or data science experience. Develop an effective change management strategy: A successful AI initiative involves changes across the organization. It demands commitment from all teams and functions and for this reason, it is crucial to get buy-in from senior management. Without gaining the trust of all users and stakeholders, metal producers cannot realize the productivity and efficiency gains from AI. An effective change management plan needs to be put in place to communicate what is changing, why it is changing, and how it will impact people. To build an AI organization, people using the technology must be comfortable and have baseline knowledge about how AI will benefit
them, as well as the business. How metal producers are using AI to improve their operations While AI adoption in the metals industry is still in its infancy, some metals companies are already achieving impactful results from their investment. For example, metal companies are applying AI across their end-to-end production processes to improve production quality, reduce energy consumption, and reduce waste across core stages of the steel production process (such as raw material processing, steel making and casting, and hot rolling.) With predictive AI insights, metal producers are digitizing their plant operations floor by giving process engineers the real-time insights they need to control their production processes, by making the necessary adjustments before production is impacted. In addition to increasing throughput, metal companies are experiencing greater stability in their downstream processes, faster sales cycles, and less deterioration of their refractories. Outside of the direct production processes, other metals companies are
using AI to control the co-generation boilers to optimise fuel source, consumption, and energy supply to plant demand, as well as reduce energy wastage and fuel costs. Not only is this contributing to reducing the plant energy costs, but it also provides important contributions to lowering CO2 emissions. Key enablers Amidst these transformational times, AI and ML technologies have proven themselves as key enablers of redefining and realigning business processes, and helping industries address the diverse set of challenges they face today. An ageing workforce, highly complex processes, and responding to the challenge of climate change are some of the key challenges plaguing many industries. Metal manufacturers can overcome these challenges by transforming their operations using data-driven solutions. By leveraging AI to extract value from operational data, many forward-thinking companies have strengthened their decision-making capabilities, improved profitability, and created a strong innovation ecosystem. �
SIGN UP TODAY TO RECEIVE YOUR FREE COPY Furnaces International brings readers a selection of technical features focusing on all aspects of the international furnaces market, as well as industry news, investments, and the latest products and projects Published quarterly in a digital format, Furnaces International and the new monthly newsletter, are sent to the inbox of over 25,000 industry professionals. As publishers of Aluminium International Today, Steel Times International and Glass International, we are able to compile this knowledge and bring you the latest developments on: • Energy Efficiency • Hot Repairs
• Maintenance • Heat Treatment
• Thermal Processes • Testing and Measurement
Look out for the December issue which contains The Furnaces International Buyers’ Guide. It is the essential guide to furnace manufacturers and suppliers of furnace equipment and services to the industrial heating/process industry.
CONTACT US NOW: Esme Horn
Sales Manager +44 (0) 1737 855136 esmehorn@quartzltd.com
Nadine Bloxsome
Editor, Furnaces International +44 (0) 1737 855115 nadinebloxsome@quartzltd.com
Zahra Awan
Editorial Assistant +44 (0) 1737 855038 zahraawan@quartzltd.com
WWW.FURNACES-INTERNATIONAL.COM
46
FUTURE STEEL FORUM 2022
Furnaces_HalfPage_Ad_Sept21.indd 1
Canvass.indd – read MM.indd 6
11:46 Steel Times 27/10/2021 International
09/05/2022 07:40:44
AI-Based Operating Model for Steel
Applying artificial intelligence (AI) and machine learning (ML) algorithms to optimize energy consumption, productivity and quality in steel manufacturing has the potential to lower the carbon footprint without the need to invest into new equipment. Save CO2, lift quality and increase efficiency along all process steps of steel production Smart Steel Technologies is the category leader for AI in steel. The company offers readyto-use software solutions for optimizing energy consumption, productivity and quality in steel manufacturing. The metallurgical expertise and process know-how of Smart Steel Technologies’s team is programmed into the core of its software. Metallurgically or technically irrelevant conclusions are avoided from the start.
Get in Touch:
Smart Steel Technologies GmbH Willi-Schwabe-Straße 1, 12489 Berlin, Germany Phone: +49 30 403 673 720 Email: request@smart-steel-technologies.com USA Office: Smart Steel Technologies Inc. 4555 Lake Forest Dr, Suite 650 Cincinnati, OH 45242, USA Email: request@smart-steel- technologies.com
We understand how it is important to efficiently extract and process precious metals and minerals.
PROCESS + PROTECT You save valuable resources while keeping your employees and the environment safe.
Improve your processes with our comprehensive portfolio of measuring instruments:
Liquiline CM44 and CPS11D: Versatile transmitter and Memosens pH sensors ensure reliable measurement and safe calibration in the lab.
Do you want to learn more? www.de.endress.com/primaries-metal
FieldPort SWA50: Intelligent Bluetooth or WirelessHART adapter for the easy connection of all HART field devices to the Netilion Cloud via Edge Devices.
Cerabar PMC51B: Absolute and gauge pressure transmitter combining measurement accuracy with IIoT functionalities.
ARTIFICIAL INTELLIGENCE
Four roadblocks to generating ROI through machine learning Machine learning has the capacity to transform the inner workings of factories and businesses, with potential to lower overall costs, and optimize production. However, roadblocks can remain, despite investments in success. With this in mind, FeroLabs’ software has been designed to maintain the best features of machine learning, and safeguard against bad values and inaccuracy. By Berk Birand* Plants across the steel sector have embraced machine learning technology, with managers increasingly seeing it as a method to help them boost volume and profitability in this highly competitive industry. Indeed, machine learning has vast potential to lower production costs and scrap rates, as well as optimize the overall production process to ensure that the resulting steel is of the highest potential volume and quality. However, fully benefitting from the power of machine learning requires a complex system far beyond creating a handful of models. Machine learning models themselves are only as good as the data that is put into them. If you put in bad data, you will get a bad result; furthermore, as the factory changes, the models must adapt, otherwise they can break or provide inaccurate predictions. If you build false assumptions into the code, you will similarly get an inaccurate output. Unless you are using dedicated industrialgrade factory optimization software, you will need to build a whole complex set of systems around machine learning models to make them production-ready, in addition to what may already be expensive machine learning investments. Without building this system, you
will encounter significant roadblocks. In this article, we’ll address four major roadblocks that will prevent you from generating a return on investment (ROI): 1. Preparing industrial data for machine learning analyses When many people think of machine learning, they think about building and developing models—in other words, algorithms that take in raw data, and use it to generate analysis, prediction, or recommendation. Of course, building models is a
fundamental element of any machine learning process. But the process of creating the models is only one step in the complex ‘MLOps’ system that is required to achieve ROI. Alongside code, one also needs to think about a multitude of areas including configuration, feature extraction, and machine resource management. And let’s not forget data cleaning. Real-world industrial data is rarely perfect, making the machine learning model’s job a challenge. To address this challenge, data scientists typically spend hundreds of hours
* CEO, FeroLabs Steel Times International
Ferolabs.indd – read MM..indd 3
FUTURE STEEL FORUM 2022
49 06/05/2022 14:50:41
ARTIFICIAL INTELLIGENCE
writing data cleaning code – and this is only for one-time cleaning. When a machine learning model is deployed, the cleaning has to happen in real-time, by streaming data, which is a much more complex and difficult process. If the process is not done correctly, your model will be trained and evaluated on garbage, and the output of your model will also be garbage, yielding zero ROI. If you want to build your own MLOps system, you will need to pay a lot of data scientists to spend a lot of time dealing with the tediousness of data cleaning. This is particularly important in steel – given that different grades of steel go into different products, there is a lot of work that needs to be done on the data cleaning end to define alloy bounds and aims for each product in the grade book. If you are not looking to build your own system, you may want to take advantage of a software that has such functionality built in. At Fero, we automatically remove measurements that are outliers to remove false sensor readings or test results. Fero also automatically processes data from different historians and merges them. For example, the raw material spectrometer readings get merged with rolling mill process readings, product specifications, and the final test results to build a comprehensive, clean dataset. In addition, Fero software’s data cleaning code is optimized for both one-time use (when creating and evaluating the model for the first time) and streaming use. This means that less garbage enters the model, making the ROI of the models that much higher. 2. Preventing bad predictions for deployed models As we said before, real industrial data is rarely perfect. Machine learning models are trained on certain values, and when real-world values diverge dramatically from the values provided during training, the system will not be able to adapt accordingly and provide accurate predictions. From a managerial standpoint, you would either need to hire a huge engineering team to deal with this or accept a fickle system with frequently inaccurate predictions. When you first train a machine learning model, it uses all the data available in the factory at that moment. Since factories
50
FUTURE STEEL FORUM 2022
Ferolabs.indd – read MM..indd 4
change dynamically, the future values that are sent to the model may be different than the past ones. As a result, machine learning models must often evaluate data they’ve never seen before. Let’s say your nitrogen reading comes in unusually high, and this leads to manganese increasing your tensile strength more than it is supposed to. Machine learning models can pick up on this unique change and recommend a lower manganese addition for a particular heat so that you avoid exceeding your maximum tensile limits. This is the genius of machine learning – that you can train it on data and then let it ‘guess’ the result. But when the training values are too different from the actual ones, the prediction won’t work. Fero adds safeguards. It catches when real-world values are too different from training values and prevents bad values from being evaluated by the model, so bad values don’t cause bad predictions. Essentially, it’s like having that huge engineering team, in one piece of software. 3. Factories change over time If you keep the same model without retraining it, predictive accuracy will decline as the plant produces new types of products, or when the plant is run in a different way. Retraining is key to making sure that the predictive accuracy is as high as it can be. However, retraining wrongly can also cause the models to perform badly.
And furthermore, even if you consistently retrain your models, the underlying assumptions may change if your data sources begin to look different. A factory produces different product mixes over the year. In addition, raw material sourcing practices change, and machines deviate from their original calibration. Retraining allows the models to learn the most recent results and relationships, thereby dramatically increasing predictive accuracy within a campaign. Whenever a process is updated with Fero software, the software automatically searches for schema updates and notifies you of potential changes you might not be aware of. Additionally, Fero’s means of displaying and quantifying uncertainty allows you to evaluate if your models are still performing as inspected and debugs your data feeds. 4. Process engineers and data scientists aren’t collaborating For machine learning investments to be successful, data scientists need to speak to process experts. Any issues in this communication stream will cause the models to perform badly. In the steel industry, any data science teams are typically located within a central unit. These folks must work with the plants to understand the exact characteristics of the product, which requires communication with operations teams and quality engineers at the plant. Any issue in this communication results
Steel Times International
06/05/2022 14:50:43
Ceran
Beyond premium greases
Highest performance calcium sulfonate complex technology used to reduce costs (TCO) and CO2 footprint (TCO2) in the Steel Industry. TotalEnergies Lubrifiants, your preferred lubricating grease partner.
lubricants.totalenergies.com
TotalEnergies Industry Solutions
ms.industry@totalenergies.com
TotalEnergies 552 006 454 RCS Nanterre - France.
ARTIFICIAL INTELLIGENCE
in bad assumptions, thereby leading to lower predictive accuracy or outright failures. Any of these roadblocks will prevent you from seeing ROI. This is obviously bad. But what is even more dangerous is the fact that a loss of ROI, in turn, will cause teams at the factories and organization to lose trust in machine learning. Many machine learning solutions marketed to manufacturers are black boxes. Operators and engineers have no way to know what’s inside, any more than one can know why Google’s image recognition software might erroneously pin a cat as a potato. This has many drawbacks. A black-box model isn’t built for partial data, nor does it change with time. You can train an image recognition model with a database of cats from the past 50 years, but factory behaviour changes dramatically within months. So, if you take data from a decade and put it in a model, it’s like you’ve trained it for two different factories. Even more problematically, conventional
ML is not built to prescribed inputs. In the industrial world, you might want to change melt shop setpoints, so the mechanical properties at the end of the line can be sure to meet specs. Most ML methods are not built to answer this question. Perhaps you’ve seen the example of a Tesla ‘thinking’ that it’s being attacked by traffic lights, when in fact it’s driving behind a truck carrying traffic lights. With all the car’s intelligence, it can’t tell the difference between
a stationary traffic light and one that’s moving on the highway. One bad analysis simply affects trust in that analysis. But if ML keeps producing bad analyses, teams will lose trust in the technology and miss out on a potential opportunity to boost their competitive value and production quality, which will be fundamental as the industry shifts towards Industry 4.0 and the interconnected factory. �
The event format is a niche trade exhibition where people can arrange meetings with a number of suppliers and industry experts in one place. Visitors and exhibitors can attend the free conference sessions to hear from industry experts. These exhibitions bring together international experts, hollow and container manufacturers and businesses that use glass containers, to discover the latest innovations which include energy efficiency, quality control, packaging, logistics and decorative possibilities.
LATIN AMERICA 2024
2023 | TURKEY
2023 | KOREA
2024 | MEXICO
glassmanevents.com/europe
glassmanevents.com/asia
glassmanevents.com/latin-america
Ken Clark Sales Director +44 (0)1737 855117 kenclark@quartzltd.com
52
Manuel Martin Quereda International Sales Manager +44 (0)1737 855023 manuelm@quartzltd.com
FUTURE STEEL FORUM 2022
GlassmanEvents_2022_Half_Page_Ad.indd 1
Ferolabs.indd – read MM..indd 6
Go online to find out more at WWW.GLASSMANEVENTS.COM
13:15 Steel Times 13/04/2022 International
06/05/2022 14:50:47
Digitalization
Explore the Nature of Your Supply Chain Smart Plans for Sustainable Metals Production
www.psimetals.com/planning
Software Excellence for Steel & Aluminium Producers
ARTIFICIAL INTELLIGENCE
Eliminating CO2, energy, and quality inefficiencies with AI Steel is the backbone of our modern civilization, but the production of one ton of steel emits an average of almost two tons of CO2. The steel industry, therefore, is responsible for about 8% of anthropogenic CO2 emissions. Due to rising average global temperatures, the urgency to reduce greenhouse gas emissions is increasing rapidly. The steel industry has understood the challenge of a fundamental and necessary transformation. Customer requirements have changed, and the demand for carbon-friendly steel products is growing. Consuming industries, such as automotive and aerospace, are pushing themselves and their supply chains to reduce their carbon footprint. Growing investor and public interest in sustainability further foster the transformation towards CO2 friendly steel production. By Michael F Peintinger*, Falk-Florian Henrich*, Otmar Jannasch* and Lucas Corts* The pressure to reduce CO2 emissions and energy consumption in steel production is rapidly and legitimately increasing. Most of the direct emissions (Scope 1) in integrated steel manufacturing originate from coking, the sinter process, and the blast furnace process. Additional direct and indirect emissions (Scope 1 and Scope 2) can be associated with the melt shop and with downstream processes like reheating and rolling. The big lever in CO2 reduction lies undoubtedly in alternative steel making equipment, for example direct reduction plants based on hydrogen technology followed by EAF routes, also based on green energy. However, it will take several years before the new plants and alternative process routes make a significant contribution to steel production. Artificial Intelligence-(AI) and Machine Learning (ML)assisted production has the potential to lower energy consumption, increase yield and lower the carbon footprint now for existing plants
and newly constructed steel mills. In 24/7 production use, the software helps to minimize inefficiencies across various production routes. Each reduction of quality deviations, energy inefficiencies and CO2 inefficiencies minimize the CO2 footprint of steel products. Software solutions based on AI and ML increase energy efficiency and thus reduce CO2 emissions along the entire process chain from iron ore reduction and liquid steel to the finished long or flat product. For example, they make it possible to reduce temperatures in the liquid phase through optimized processes, thus minimizing energy requirements. They also increase yield, so that more semi-finished products of prime quality can be sold with the same energy input, fewer coils are downgraded, and less scrap must be remelted. They improve the metallurgical properties of the steel and the quality of the surface. And finally, based on current process data, they precisely calculate the energy input
and CO2 emissions for each ton of liquid steel and for each product. However, isolated applications of AI to individual processing steps could even counteract valuable achievements in an upstream or downstream process, due to imposed requirements on input material, or a single objective cost optimization at the expense of output quality. To fully unleash the potential and increase AI acceptance in steel manufacturing, the AI approach must be extended from individual processing steps to a plant-wide strategy. Establishing such an end-to-end optimization requires a systematic approach taking into account data integration, quality monitoring and process optimization. Also, the transition needs to take place in parallel with the existing steel manufacturing systems and must not interfere with ongoing steel production. We have successfully implemented our AI-based software modules in different steel mills – integrated facilities and EAF
* Smart Steel Technologies GmbH and Smart Steel Technologies Inc Steel Times International
SST.indd – read MM.indd 3
FUTURE STEEL FORUM 2022
55 12/05/2022 10:54:42
ARTIFICIAL INTELLIGENCE
steelmakers – producing long and flat high-quality steel. They specifically target efficiency improvements of the production process regarding energy consumption and yield increase by means of a big data-driven approach. In the following, we describe how artificial intelligence-based emission assignment and tracking, temperature guidance, defect classification and casting optimization interplay to achieve this pressing goal. Greenhouse gas emission tracking Implementing transparent CO2 tracking in steel production is one of the major challenges to address, since transparent tracking of material in combination with machine learning allows for the assignment of accurate CO2 footprints to individual pieces of material and their processing steps. Therefore, a regression model was developed that tracks all parameters and, thus, correlates the total energy consumption to tapped hot metal. In this way the CO2 emissions can be assigned to every single ton of hot metal. The model predicts all influencing parameters and their impact. This is the basis for a model that helps to reduce energy consumption and CO2 emissions. (Fig. 1) Process control from BOF/EAF to continuous casting Taking the entire steelmaking process into account, the optimum BOF or EAF tapping temperature is calculated by a global recommendation model for each heat in the sequence to precisely hit the specified ladle furnace and tundish superheat temperature. In order to tap exactly at the recommended temperature, the SST temperature AI also calculates a prediction of the tapping temperature at the start of the refining phase (in the case of an EAF) or the main blow (in the case of a BOF), taking into account the numerous relevant input variables such as quantity, temperature and analysis of the pig iron, composition of the scrap, the refining phase schedule as well as the addition of slag formers and other additives during treatment. This enables operators to react to any unwanted deviations. The treatment at the ladle furnace is supported by comparable models; in particular, the
56 SST.indd – read MM.indd 4
FUTURE STEEL FORUM 2022
Fig 1. Regression analysis
software also calculates the ideal discharge temperature. The whole temperature guidance is not only optimized to save energy and CO2 emissions but to supply the liquid in the most suitable condition for prime-quality solidification to the casting machine. Therefore, SST’s temperature AI optimizes either possible route in even complicated secondary metallurgy processes, i.e., it considers treatment and purging stands, ladle furnaces, vacuum degassers. The whole course of the temperature from BOF/EAF tapping to the caster is designed to save energy and CO2, time, and finally supply the ladle at the required temperature to the caster. SST’s temperature AI modules recommend the exit temperature for every single process step and guides operators to achieve the recommended temperatures. In addition, the casting operation is supported by the tundish end-temperature model, which calculates the expected tundish end-temperature for the current heat during casting. The prediction is continuously updated in real time and considers, among many other parameters, the temperature of the preceding and the homogeneity of the current heat. The interplay of these global recommendation and local prediction models stabilizes the process across all stations, so, for example, Marienhütte is now able to minimize temperature buffers and consequently, lower the tapping temperature in the EAF. On average, the temperature is reduced by 8K (kelvins). The result is permanent energy and CO2 savings.
(Fig.2) These models are successfully implemented and in 24/7 real-time operation in the basic oxygen furnace (BOF) melt shop at ArcelorMittal Eisenhüttenstadt and Duisburg in Germany and in the electric arc furnace (EAF) melt shop of Marienhütte in Graz, Austria. This proves that this approach is suitable for large and complex melt shops, but also smaller plants, and can result in increased profitability and a reduced carbon footprint. Recommendations and predictions calculated by SST’s temperature AI are displayed within the original melt shop user interface as well as in the SST web-based human machine interface, together with the most important influencing variables. (Fig. 3) Reliable defect classification Additional optimization potential with AI algorithms lies in further downstream production processes where a clear optimization goal or ‘target signal’ must be identified. Reliable defect classification in terms of quality data is the basis for process optimization. The main reason for rejecting material produced for high-end quality products, such as automotive exposed or electrical sheet metal, are surface defects, which in the worst case scenario might lead to yield loss (if the material cannot be repaired or reapplied). This has a particularly high impact on the carbon footprint. Therefore, the reduction of these surface defects serves as an excellent target signal for process optimization, for example, in continuous casting.
Steel Times International
12/05/2022 10:54:44
Tenova_Corporate_Green_210x297_FM_10152021_eo.pdf 1 10/15/2021 1:57:35 PM
C
M
Y
CM
MY
CY
CMY
K
Temperature
ARTIFICIAL INTELLIGENCE
Reclassified defects are shown in the webbased centralized coil map, including defect tracking, clustering, and full coil genealogy. The optimized casting schedule is also shown in a web-based user interface.
SST temperature Ai Traditional Approach
Liquidus Temp.
BOF/EAF
LF Entry
LF Exit
RH Entry RH Exit
CC
Time
Fig 2. Minimizing temperature buffer
Steelmakers invest in automated surface inspection systems (ASIS) at the end of the hot strip mills, continuous pickling lines and continuous galvanizing lines. These systems take images of the top and bottom of the strip as it passes by. Due to outdated classification algorithms, most of the currently deployed ASIS classifiers are not robust or accurate enough to be used in automated process optimization. However, we have achieved accurate classification using our deep convolutional neural network (CNN) classifiers specifically designed for steel surface images taken at individual steel processing steps. The network topologies of our SST surface AI are finetuned with plant-specific training and test data and beat any other method in terms of labeling accuracy[6]. Integrating the full material genealogy (see section 2.5) enables potential defects to be cross-referenced to preceding manufacturing steps to achieve even higher labeling accuracy. Data integration and human machine interface implementation Correctly classified surface defects from the automated surface inspection system (ASIS) are the mandatory basis for optimizing the casting sequence. In order to enable the machine learning models to calculate precise recommendations and predictions for each slab, the correctly classified defects which are detected on the finished strip have to be
58 SST.indd – read MM.indd 6
FUTURE STEEL FORUM 2022
accurately mapped back to their position at the slab and thus, the parameters at the time of their origin. All relevant L1, L2 and ASIS data are integrated into the SST data platform in real time [5]. It aggregates and validates data from all sources in an integrated steel mill and stores data in specialized databases depending on the type and usage. This includes relational databases (scalars, genealogy), columnar OLAP databases (time series), a dedicated vector database (features) as well as low-cost media storage (images, video). Any transformation occurring during the production process is stored during data aggregation to allow full material tracking and data transformation depending on the reference point.
Casting optimization High quality steels used for automotive exposed products are specifically susceptible to surface quality defects such as slivers [3–8]. These defects often only become visible towards the end of the value chain. In previous studies we have shown how ArcelorMittal Eisenhüttenstadt reduced specific surface defects by more than 50% [2]. The input of the SST casting AI is a full set of casting parameters, such as casting speed, submerged-entry nozzle, submersion level and mold width as well as melt shop parameters, such as superheat temperature and chemistry. The models compute optimal values for continuous and discrete casting as well as melt shop parameters that lead to the best quality, for example, minimization of slivers, while considering highly complex constraints originating from both business requirements and physical limits of the casting equipment. The system is operating in a 24/7 real-time mode. The parameters are automatically fed back into the casting planning system. This results in an optimized sequence that minimizes quality deviations, and downgrading and reallocation. The yield increase translates directly into reduced greenhouse gas emissions.
Fig 3. HMI for tapping temperature recommendation at the BOF
Steel Times International
12/05/2022 10:54:47
You’ve got it • • • •
Tundish Skulls Pig iron Heavy scrap
We break it!
The Fractum technology ensures high precision breaking of steel scrap in a safe and efficient way Erritsø Møllebanke 11 7000 Fredericia Denmark
Email: info@fractum.com Phone: +45 7262 7300
ARTIFICIAL INTELLIGENCE
Results The overall temperature level from EAF tapping to LF exit was reduced by 8K. The result is permanent energy and CO2 savings. Surface defect classification was improved to 95% accuracy. Applying the SST casting, AI achieved a permanent reduction of the rate of sliver defects for automotive exposed grades by up to 50 % and more, and consequently less downgrading. Conclusions The presented solutions are actively used in steel production at various facilities, which all have individual product mixes, equipment set-ups, process peculiarities and data structures. By combining the different recommendation and prediction models, which support the process steps from the converter via the ladle furnace to the last station of the secondary metallurgy treatment, and considering the planned sequences at the casting machine, the SST temperature AI allows precise temperature control during the entire secondary metallurgical treatment. Temperature-related disturbances are thus immediately reduced. In addition, the more precise control of the temperature leads to
better plannability, also since, for example, treatment times at the ladle furnace are homogenized. We plan to further improve predictability by using off-gas data and chemistry data from taken samples for situations involving heats of high chemical energy (i.e., containing higher amounts of Cr, Mn, Si). Data-driven casting optimization Reliable surface defect classification and mapping enables producers to implement data-driven casting optimization with the SST casting AI, leading to less scrap, rejects and reallocation. Implementing such an integrated and systematic approach in a steel mill is a challenging project and requires close collaboration between metallurgy, steelmaking, data science, machine learning and IT experts. This cross-disciplinary expertise is the key to the success of such a project. The motivation to undertake the endeavour of implementing temperature guidance, defect classification and casting optimization is mainly economic. However, lower energy consumption, enhanced quality, and, therefore, increased yield, directly
translate into the reduction of carbon dioxide emissions. � References 1. M. Peintinger, “Big Data”, Iron & Steel Technology, Dec Issue, 2021, http://digital. library.aist.org/pages/PR-DA1221-1.htm. 2. R. Bösler, F. Henrich, O. Jannasch and J. Daldrop, “Optimized Production of Automotive Steel Sheet Through Application of AI,” Steel Times International 2021, Future Steel Forum 2020. 3. M. Lüttenberg, S. Hilterscheid, F. Henrich, O. Jannasch, J. Daldrop and T. Wessels, “Präzise Stahltemperaturführung mit künstlicher Intelligenz,” Stahl und Eisen, No. 04, March 2021. 4. D.H. Kindt et al., “Steelmaking Practices to Improve the Surface Quality of Cold Rolled Sheet at Bethlehem’s Burns Harbor Plant,” Steelmaking Conf. Proc., 1990. 5. M. Peintinger, “Application of Highly Specialized Database Technology Within a Unified Data Landscape”, AIST Digital Transformation Forum 2022. 6. F. Henrich et al., “Classifying Defects More Reliably,” STEEL + TECHNOLOGY, Vol. 4, 2019.
LIGHTENING THE IMPACT OF HEAVY INDUSTRY Brought to you by a coalition of industry experts, Sustainable Industrial Manufacturing (SIM) is a global series of regional exhibitions and conferences supporting and facilitating the transition towards cleaner manufacturing GLASS
ALUMINIUM
CEMENT
around the world.
STEEL
CHEMICALS
BOOK TO ATTEND AT www.SustainableIndustrialManufacturing.com
ORGANISED BY:
José Sebastião Commercial Director josesebastiao@quartzltd.com Tel: +44 (0) 1737 855 013
60
FUTURE STEEL FORUM 2022
SIM_Europe_Half_Page_Ad_185x128mm_v2.indd 1
SST.indd – read MM.indd 8
FROM THE PRODUCERS OF:
Nadine Bloxsome Event Director nadinebloxsome@quartzltd.com +44 (0)1737 855115
Steel Times International 19/04/2022 11:50
12/05/2022 10:54:49
YO U M A K E S T E E L , W E R E D U C E YO U R E N E R G Y CONSUMPTION
As simple as that.
Drive costs down with ready to use AI. Being sustainable is no longer a dream. Start your journey today!
info@wizata.com
INTERNET OF METALS
Tracing sustainability data coilDNA, an Austrian startup company based in Linz, has developed a technology that gives individual pieces of metal an identity – aiming to improve the properties of the final product, monitor quality, and take producer-processor integration to new levels. By Dr. Werner Aumayr*
Industrial metals such as steel are processed by means of melting, casting, rolling, and using different kinds of heat and surface treatment, such as galvanizing, pickling or passivation. In each step, countless sets of data are generated which are relevant for the quality, sustainability, and other properties of the semi-finished product (coils, strips, plates, profiles). During further processing steps, single pieces of metal are usually cut from these semi-finished products. In this process step, the data link between the single piece and the parent product is broken. The single piece usually does not carry any information about its material composition, the manufacturer, or the position within the parent product and thus any sustainability information is lost.
The main challenge is making it possible to trace the manufacturing conditions and sustainability information for each part of a steel coil. Sheet metal blanks supplied to the automotive industry are an excellent example. Let’s consider a 1mm thick coil with a length of four metres, which was produced from a rolling slab having itself a thickness of 570 mm and a length of seven metres. Tolerances are in place for every single process parameter in each step of production from melting, casting, and rolling, to heat and surface treatment. This means that during the production of a single coil, parameters vary with time given within the allowed tolerance, resulting in a variation of material properties with length. There are also transitional phases in casting especially during the start and
end of a cast sequence, or during ladle and tundish exchange. In these phases of a cast sequence, it takes a certain time for process parameters to stabilize at the predetermined level. This is also valid for any deviation from stable production conditions such as speed variations during casting, leading to different liquid steel flow patterns in the mould and cooling conditions in the strand guide. All these factors have an affect on the internal and surface quality of the slabs, and ultimately on the material properties and quality parameters of hot and cold rolled coils. Let us now assume that the aforementioned coils yield 4,000 automotive panels with a panel length of 1m. Each of these sheets was manufactured under slightly
* Managing director, coilDNA
62
FUTURE STEEL FORUM 2022
Coil DNA.indd – read MM.indd 2
Steel Times International
09/05/2022 07:55:41
INTERNET OF METALS
different conditions and, therefore, have slightly different properties within the product specification. These minor deviations might have an impact on further processing and consequently, on final part properties. In addition, processors use steel strips from different manufacturers for a single component (such as a fender). Coils are tracked by an immense amount of related data. Length-based data is rarely, if at all, available. Once cut into a single piece, all the information about the origin, product and quality data, as well as sustainability indicators are lost and no longer assigned to the piece. If quality issues arise during processing, it is almost impossible to identify the single piece to follow up the issues with the material supplier. There are numerous other examples of continuously manufactured semi-finished steel products that lose their identity in a further dimension-changing processing step, with a good example of this being hot rolled steel beams used for construction in bridges. Once cut to length and welded into the final supporting element of a bridge, all the identity of the single beam element usually gets lost. Life cycle tracking becomes a particular challenge. The same is valid for roll forming operations, where coils are slit and processed to a final shape. CoilDNA has developed a patented technology that gives individual pieces of metal an identity. By making these metal products smart, they may be connected to the Internet as well. So IoT (Internet of Things), the concept of connecting smart devices to the internet, now applies to coilDNA-enabled smart steel and thus IoM (Internet of Metals) is taking shape. To make this revolutionary development tangible, let’s look back in retail history. In 1974, a pack of Wrigley’s chewing gum heralded a revolution in retail. It was the first product in the world to be issued with a barcode for scanning at a supermarket checkout. Today, everyone uses barcodes and associated apps to look up consumer product information such as nutritional data, allergens, vitamin content and similar. For retailers, this not only makes data maintenance much simpler, it is also an invaluable tool for warehousing, reordering and sales forecasting. Like that pack of chewing gum, steel semi-finished Steel Times International
Coil DNA.indd – read MM.indd 3
products marked with a coilDNA code, open up opportunities for users that are as revolutionary as those offered by the barcode 45 years ago. The human DNA is an excellent role model for coilDNA. Every single cell of a human body can be used to identify the individual.
DNA sequencing allows the re-construction of all the DNA information obtained from only parts of a single DNA molecule. The coilDNA technology uses comparable mathematical algorithms. A unique coilDNA information code gets continuously printed on the surface of a parent product (Fig 1)
FUTURE STEEL FORUM 2022
63 09/05/2022 07:55:48
INTERNET OF METALS
e.g., a coil, a tube or a profile by laser or inkjet marking systems (Fig 2). This code uniquely identifies the position within the parent product and subsequently allows the assignment of the production data recorded at this position. Regardless of how this parent product is cut in subsequent production steps, the item-related and even the position-related information is always available. With only 14 eligible characters of the coilDNA code, all the information about the respective piece of metal can be retrieved. This coilDNA code is direction detecting – this is important because the direction of a coil is reversed in each production step – and can even detect errors. Conventional identification methods based on bar- or QR-Codes fail as soon as the material is divided, and the barcode is cut. In addition, after these markings have been attached at certain intervals, it cannot be guaranteed that each individual part shows a marking and can thus be assigned to the original mother product, or even to the manufacturer An example: A steel producer which is proud of its ‘green steel’, wants to make sure that every part of the further processing chain can unambiguously identify this green material with an ordinary smart phone, even if there are only a couple of inches left. The steel producer applies coilDNA’s code after galvanizing, then adds sustainability information, the product name, and more data. It is of no relevance how the steel is divided into further steps – if there are 14 remaining characters, identification and data retrieval is possible.
64
FUTURE STEEL FORUM 2022
Coil DNA.indd – read MM.indd 4
coilDNA thus offers a web-service platform for distributing the patented codes to producers and processors of metals to exchange product-related data in extra fine granularity. Features of coilDNA include: � • Efficient processing of metal parts by considering the local properties of the semifinished product used. �• Seamless tracking of products and their sustainability properties throughout the whole supply-chain from producer to processor. � • Data driven communication with the producers using a simple picture of a product labelled with a coilDNA code, taken, for example, by a cell phone. The coilDNA CHAT App allows users to identify the product and give direct feedback to the producer. • Sustainability information checks by simply using a smart phone (Fig 3). • Checking of the validity of productrelated paper documents using the coilDNA CHECK App. • Combining physical products and documents, like quality or sustainability reports, in a forgery-proof manner, for example, with block chain technologies. Summary coilDNA is the key to the Internet of Metals (IoM), where metal parts have an identity in the form of a unique and eligible divisionindependent code. This enables users to communicate with the manufacturers and processors simply by entering 14 consecutive characters of the visible code in various smart coilDNA apps, manually or by taking a picture. Any kind of data generated during
About coilDNA coilDNA is a start-up company, founded in 2019 as a spin-off from Austrian aluminium producer AMAG in Ranshofen. The company is located in Linz. The main goals of the company are the development and marketing of the patented coilDNA technology. coilDNA supplies the unique product code, data services and application support, with industrial partners in the fields of: � manufacturing execution systems (MES) for the integration of the coilDNA technology into MES systems. � industrial printing and high-speed inline character recognition. � block chain technologies.
production and processing are clearly assigned to each segment of metal semifinished products by means of the patented coilDNA code. Special web services make this data available to all parties involved in the supply and production chain, at any time, and when required. As a result, each single step of the value chain could be linked to the other in order to improve the properties of the final product, seamlessly monitoring quality and taking producer-processor integration to new levels. � Contact: www.coildna.com hello@coildna.com
Steel Times International
09/05/2022 07:56:01
PRI_ANZ_Green-Steel-Ad_A4-vek.indd 2
02.05.2022 14:31:05
ARTIFICIAL INTELLIGENCE
Artificial intelligence and machine learning for EAF 0ptimization The amount of data available in a modern electric arc furnace operation, even if only using standard instrumentation and automation, has enormous potential for process analysis and improvement using the latest artificial intelligence tools and algorithms. By Guillermo Fernandez1, Mariana Viale2 and Emmanuel Placier3 AMI’s SmartKnB platform is a graphic programming interface that makes it possible to develop solutions that merge data acquisition, complex process logic and machine learning models, all in the same user-friendly environment. Several applications have been developed and deployed in the field of electric steelmaking using SmartKnB, for instance, to follow the process from raw material intake, analyzing its characteristics in advance to optimize melting and final steel composition. The software is also capable of continuously evaluating correlations between the process and usage of consumables to find the most favourable operating point. An image-based scrap type classification has been implemented using a Convolutional Neural Network model in the software. The possibility of adding image processing and using advanced cameras capable of withstanding the harsh environment of a steel plant, will give immediate feedback to the AMI platform, opening a vast field of opportunities to put eyes on the process closer than ever. 1. SmartKnB platform SmartKnB, like its predecessor Visual KB, is designed for the development of advanced control applications. For interaction with the real world, the software is capable of
Pic A
reading and writing data from different PLCs using native PLC communications for some brands and OPC communications for others. It can also connect in the same way to databases to obtain the required information about the process. The code in SmartKnB is on-line programmable and supports multiple users working simultaneously in different parts of the system. The code, similar to a flow chart diagram, is designed to visually represent the rules and algorithms, allowing the tool to accurately implement the knowledge base behind the control system. It allows the creation of several levels of sub-modules to
encapsulate complex logic processes, which can be easily replicated or transferred to a different area of the programme, or even to an entirely different operation. An example of the actual programming code is shown in Pic A. The SmartKnB system includes several artificial intelligence tools such as clustering algorithms, multiple stage fuzzy logic with profile-based parameters, and advanced instructions to achieve non-linear approximations, to name just a few. If the user requires a very specific application development or prefers to use other programming languages like Python or C, it is
1. Technology director, AMI Automation. 2 Senior application and development engineer. 3. Business development manager. Steel Times International
AMI automation.indd 3
FUTURE STEEL FORUM 2022
67 10/05/2022 11:04:02
ARTIFICIAL INTELLIGENCE
IoTrode
Pic B
Conductivity and temperature anomalies
• Red tip measurements (length and width) • Butt loss detection • Gapping or cracking detection
Clear image of roof
Pic C
also possible to create new instructions within the SmartKnB system environment using these languages. Another important addition to the SmartKnB tools are the machine learning
modules. These modules allow SmartKnB to develop and import models using machine learning. The models can be created on any platform the user prefers like Python, Math Lab, etc. which can then be imported into
SmartKnB. Several applications have already been developed using these tools, as it is described in the following sections. SmartKnB is also able to create different sets of data which can be used to develop machine learning models, or just to have historical records of the process. Several data sets can be created at the same time. These data logs can be accessed later for analysis or review, but they can also be used for simulation in the SmartKnB software, thus providing the capability to test newly developed algorithms with historical data sets, getting a simulated response. Finally, in order to facilitate the interaction with the system and make it more visual, SmartKnB has a tool for custom HMI development. This HMI enables the user to access instructions directly with the control parameter, turn different switches to modify system behaviour, watch data trends in graphs, or use different types of indicators. 2. Machine vision developments The latest advances in machine vision and artificial intelligence are being gradually adopted by the steel industry. AMI is leading this process, integrating these new technologies to the existing EAF optimization models, helping to improve process safety, and providing additional tools to optimize overall performance. IoTrode For analyzing the consumption of the EAF electrodes, AMI developed the IoTrode. This system uses high-definition cameras to measure the electrodes’ red tip oxidation.
Pic E Pic D
68 AMI automation.indd 4
Sensor fusion unit
FUTURE STEEL FORUM 2022
Steel Times International
10/05/2022 11:04:12
Contact us! indalgo.com Perttu Laurinen perttu.laurinen@indalgo.com +358 40 560 5616
ARTIFICIAL INTELLIGENCE
Pic F
Plant PLC
Pic G
SFU
AMI merge
3D camera and RGB module
Interface and SmartFurnace server
Electric Arc Furnace Direction of movement>>> Scrap profile scanning
By registering the electrode tip deformation and oxidation, it is then possible to measure the real electrode consumption, per heat and per phase. This information is used by an adaptative control system to better use the electrode water cooling in order to minimize the red tip generation. Furthermore, all the data generated by the electrode high-definition images can be used to evaluate the quality of the different providers, as well as for inferring which process variable is affecting electrode consumption the most. (Pics B and C) Scrap classification for continuous and bucket charging For the automatic scrap classification, AMI has developed a system that uses sensor fusion technology and deep learning. The principal challenge was to classify and measure more than 30 different types of scrap, without human intervention, before the raw material was fed into the EAF. To achieve this, a 3D camera was used. This camera provides stereo vision using two monochrome cameras with an infrared spectrum projector,
70 AMI automation.indd 6
FUTURE STEEL FORUM 2022
Off Gas Extraction
Preheating
together with high resolution colour images. By using this technology and developing sensor fusion algorithms, a high-resolution colour image of the scrap together with the scrap volume was calculated in real time, either when preparing the buckets, or when feeding the scrap through a conveyor (Pics D and E). In order to automatically classify the scrap type based on 3D images, data-driven analytics and machine learning were used. With having thousands of images from more than 30 different scrap types, a data-driven ML was the most direct solution. 70% of the data set was used as inputs for a transfer learning technique based on a convolutional neural network (CNN) model and the 30% remaining was used for validation. Once a CNN trained model validation accuracy surpassed 95%, the model was considered fitted and it was exported as an open neural network exchange model (ONNX). ONNX is an open format to represent artificial intelligence models that is widely supported and can be found in many frameworks.
The trained scrap classifier ONXX model was imported into the SmartKnB platform and directly integrated into the rest of the EAF optimization process. The versatility of the software allowed the EAF to develop complex machine learning solutions and integrate them easily using its existing control systems (Pic F).
Energy optimization Knowing the exact scrap mix that is being charged in real time into the EAF allows workers to optimize the available energy inside the furnace and improve the end point carbon and temperature estimations. The scrap classifier, named IoConveyor, was implemented in a furnace continuously feeding scrap. The system also measures the scrap speed on the conveyor and estimates the real feed rate. The objective was to generate this data with greater detail than what was previously available, leading to better control of the feeding speed to avoid overheating or unmelted scrap. The Steel Times International
10/05/2022 11:04:18
ARTIFICIAL INTELLIGENCE
Operator HMI Taping station
Pic H
Infrared camera LWIR 7 to 14 micron Temp up to 2000C
Image processor VKB and cloud connection
Plant Network
Plant PLC
Pic I
system was installed in two major steel manufacturing plants in the USA (Pic G). AMI is also developing a solution for scrap classification and mass estimation for bucket charge facilities named the IoBucket, based on the gained experience in the continuous conveyor to address the bucket charge challenge. This system will be tested in mid-2022. Pic J IoTap For detecting the slag carry-over during tapping, AMI developed the IoTap. This system uses a high-resolution long wave infrared camera suitable to measure temperatures of up to 3000 deg C. The infrared images provide valuable information regarding the tapping process that cannot be seen by the human eye. In knowing the emissivity of the steel, the slag can be differentiated, and the system can integrate the amount of slag that is passing from the EAF to the ladle. The steel producer can then assure clean steel in the caster to avoid stickers, as well as decide when it is necessary to remove the extra slag Steel Times International
AMI automation.indd 7
in the ladle (Pics H and I). Freeboard detection An extra LWIR camera focuses on the ladle free board. The actual freeboard is measured using image processing of the IR image and the tapping operator has online feedback of the ladle level to avoid splashing and brick wash-out (Pic J). Taking a closer and more detailed look on existing electric arc furnaces using smart
vision systems, coupled with powerful data analysis tools, has shown that potential for improvements and optimization can always be found. The SmartKnB optimization platform is capable of processing data with much greater volume than ever before, opening new opportunities for understanding and improving process efficiency and safety, while simultaneously making it available to the user in a meaningful way. � FUTURE STEEL FORUM 2022
71 10/05/2022 11:04:32
SURFACE INSPECTION
IMS surcon 3D Surface Inspection Systems with integrated 3D calibration and blower solution
2D, 3D, or both? A trial surcon 2D/3D surface inspection system even for most challenging environments will help you to make the right decision, says Thomas Köpsel* The development of IMS surcon surface inspection started back in 2008, with the intent of creating innovative and flexible solutions for a vast variety of applications. As a result, full scale prototypes for 3D slab inspection and 2D systems for hot and cold strip inspection were being implemented quickly during the following years. Now, more than 10 years later, the system has matured and provides advanced and specialized solutions for the automated inspection of slabs, heavy plates, coils, tubes
and even complex profiles like I-beam or sheet piles. For decades, IMS gauges have been successfully operating in mills all over the world. Over the years, the company has built a reputation for operating measuring systems safely and reliably even in the most difficult environments. Nevertheless, it has turned out that sometimes the users need to be convinced that automated surface inspection is even possible for their application. Sturdily built field test units provide the best
solution to achieve this. These units, specially designed to cope with local challenges, can prove the feasibility of surface inspection under challenging conditions and provide the user with a first impression of the capabilities of a fully automated surface inspection. Harsh environments that use aggressive kerosene-based oil or extreme heat require special modifications in order to protect the sensitive equipment. Camera and illumination windows need to stay free of oil while gaskets need to withstand the corrosive
* Sales and product manager, IMS GmbH
72 IMS.indd – read MM..indd 2
FUTURE STEEL FORUM 2022
Steel Times International
09/05/2022 10:30:05
SURFACE INSPECTION
environment. Tight space restrictions often require a highly customized design for the system. Thus, engineering teams, local staff, and system designers need to work closely together, not only making the system fit for use, but also ensuring easy access for maintenance and long-life operation. They also need to keep in mind that fitting a full-scale system is usually the necessary long-term goal. A test system predominantly covers one measuring field of about half a metre and can ideally be placed freely along the width of the material for testing its performance at different locations. This small and flexible set-up allows users to easily switch between different configurations for camera and illumination angles. Multiple cameras allow fast and objective comparisons of different set-ups. The verification of test results and additional on-site evaluations will enable you to deduce the best resulting set-up for a permanent inspection system covering the whole width. So what are the main reasons for implementing a trial system of an IMS surcon 2D or 3D surface inspection system? The main questions which must be answered in this first step are: • Is a surcon 2D or 3D system, or maybe even both required for the task? • What is the optimal camera configuration and layout? • Are multiple configurations required? The environmental point-of-view must be clarified as well: • What are the required countermeasures? • What about heat protection? • Is a blow-off required? • Are additional sprays or air knives needed? The trial system is already equipped with a variety of countermeasures to withstand the harsh conditions of a typical production environment. Depending on the actual set-up, it is possible to retrofit existing trial units or build special units – reflecting IMS’ experience in building reliable measuring systems within the metal production processes. The system is intended to provide a hasslefree installation and already provides all the features of a full-scale surface inspection – including adjustable defect detection, Steel Times International
IMS.indd – read MM..indd 3
IMS surcon 2D surface inspection system
Configuration 1:
Configuration 2:
The modular design allows you to retrofit existing surface inspection systems with dark field configurations
While invisible in the 2D image, the critical surface defect can be clearly distinguished in the 3D image
FUTURE STEEL FORUM 2022
73 09/05/2022 10:30:16
SURFACE INSPECTION
The 3D image allows objective defect assessment based on the depth of the defect
But the imagination is not limited to this general guideline. Any combination is possible. And to determine the best setting sample evaluation in our laboratory as well as in field experience by reference systems or trial runs provide a solid base for decision making.
of this and offers further optimization of defect detection. Parallel image recording and processing ensures that no surface defect is overlooked. The camera and lighting geometry is adapted individually to the surfaces of the customer’s products for an optimal result. The modular design allows the user to retrofit existing surface inspection systems with dark field configurations, meaning systems remain completely flexible even after installation. The flexible camera and illumination setup of surcon 2D surface inspection systems show that different types of defects appear differently depending on the illumination set-up. In the following pictures the same sample is detected, but with different camera configurations. It is important to get knowledge about both defects detected by the same surface inspection system. The solution, in this particular case, is additional camera arrays as part of the set-up of the surcon 2D surface inspection system. The flexible camera layout can be extended up to three parallel channels.
Surcon 2D surface inspection system Surcon 2D surface inspection uses high power LED and fast line scan cameras to capture every detail of the surface. The sensitivity of the system can be tuned using various settings and parameters, and a powerful and quick self-learning algorithm allows immediate defect identification based on previously analysed training images. The appearance of certain defects can change drastically under different angles of illumination. An additional dark field configuration, for example, takes advantage
Unveiling the invisible with a surcon 3D surface inspection For certain defects which are invisible to the human eye, a surcon 3D surface inspection system is the solution. Surcon 3D surface inspection systems are based on laser triangulation using specially developed IMS line lasers for optimal line quality. Based on brightness information alone, surface defects cannot be reliably distinguished from the surface itself. Only the additional height information – the 3D – makes surface defects visible and allows
auto-classification, and a tool set to fine-tune the defect classification to the specific task. An intuitive user interface allows easy access to recorded inspection data and provides tools to tune the system performance. In general, IMS proposes that surcon 2D surface inspection, meaning high-speed line cameras and high-power LED lights, is used for: • Hot strip mills • Cold rolling mills • Processing lines Surcon 3D also provides the height information of the detected defects which is essential for: • Slabs, billets and blooms • Heavy plates, tubes • Long products
74 IMS.indd – read MM..indd 4
FUTURE STEEL FORUM 2022
for a reliable detection. In the past, it was only possible to guess. Now, the surcon 3D inspection system delivers solid results (Pic F). 3D surface inspection allows an objective way for identifying and grading defects that possess a detectable height deviance. Hard thresholds that can be adapted depending on grade, material or target customer allow full control of the quality evaluation. The surcon trial system allows a linespecific evaluation of the optimal solution for your application, be it 2D, 3D, or parallel settings. The possibility to check the performance of the IMS inspection technology provides the certainty that customers need to pick the perfect fit. Mobile Inspector App (MIA) All surcon inspection systems can be viewed remotely with the mobile inspection app MIA to get the best out of the data. With the help of the IMS Mobile Inspector App (MIA) for surcon 2D and 3D surface inspection systems (available for iOS and Android), inspection results can be viewed while on the move. Regardless of location – whether in a conference room or in the plant – all data is always just a finger swipe away. The surface map, which can be controlled via gestures, enables fast, intuitive navigation through the inspection results of all surcon 2D and 3D surface inspection systems. In this way, detected surface defects can be compared directly with the actual surface of the material. The MIA Inspector App supports all surcon 2D and 3D surface inspection systems. Even simultaneous access to several systems is possible, thereby bundling together all inspection results required by the customer.
Steel Times International
09/05/2022 10:30:18
AR for Industry Processes that Simply Work Better U p to 40% increased process speed
U p to 50% saved training time
U p to 100% output accuracy
U p to 100% satisfied workers
Frontline is a fully integrated augmented reality solution platform. Our solutions improve manual work processes across industries on the basis of the latest wearable computing technologies. Frontline is designed to consciously empower the human worker in an increasingly digitized working environment.
www.teamviewer.com
2022-04_Frontline_Ad_Future_Steel_Forum_A4_K1.indd 1
05.05.22 13:49
SURFACE INSPECTION
The Mobile Inspection App (MIA) rounds off the individuality of IMS surcon 2D and 3D surface inspection systems
IMS Messsysteme GmbH has been a guarantee for highest product quality in the production and processing of steel, aluminium, and non-ferrous metals for more than 40 years. Exactly reproducible measurements and evaluations in real-time optimize
76 IMS.indd – read MM..indd 6
FUTURE STEEL FORUM 2022
customers’ production lines all over the world. The surcon 2D or 3D surface inspection trial system provides easy access to evaluate system performance and select the necessary features based on the individual characteristics of a processing line, whether
the settings are 2D, 3D or a combination of both. A trial will determine what the individual customer requires to best understand the surface quality, acquire the information needed, and then take the necessary steps for process optimization. �
Steel Times International
09/05/2022 10:30:32
MOVE FAST
Marking & Labelling Coils & Slabs
Dross Removing
www.tebulorobotics.com
MOVE FORWARD
Destrapping of High Strength Steel Straps
Eye Strapping
Product Handling & Specials
Sample Plate Handling
ROBOTICS
Fig 1
Safe removal of highstrength steel straps International robotics company Tebulo Industrial Robotics has developed a safe method of removing high strength steel straps from coils of steel. The De-Strapper is a robotic system that detects the straps over the full width of the coil, and removes them safely, says Henriëtte van Norel* The remarkable increase in the demand for high-strength steel in recent years can be explained by the fact that this material offers substantial cost reductions. Just consider the
material’s weight savings potential as well as its capacity to improve the performance qualities of a great variety of products. Compared to traditional steel, the benefits
of high strength steel are evident. However, from a safety perspective, materials handling procedures definitely need an upgrade. Special precautions are required to remove
* Director, Tebulo Industrial Robotics
78 Tebulo.indd 2
FUTURE STEEL FORUM 2022
Steel Times International
09/05/2022 10:34:43
ROBOTICS
the strapping from high-strength steel rolls. A company with global reach, Tebulo Industrial Robotics, has developed a new robotic De-Strapper, particularly designed for the safe de-strapping of rolls of steel. Aside from this, the company manufactures tailor-made robotic solutions for material handling, welding, hot and cold marking and/or tagging of rolls, sheets or slabs. Hans Spaans, director of technology with Tebulo Industrial Robotics, explains: “In a production process, strapping high strength coils of steel requires a substantially greater number of straps than where coils of conventional steel are concerned. Whenever the strapping is made of a much tougher steel quality, a much lower amount of straps are required. Although this may seem to be an obvious solution, the removal of high-tensile steel straps is an art in and of itself. Advantages “High-strength steel is a generic term for steel qualities that are tougher than standard steel qualities. In recent years, steel producers continuously focused on high-tensile steel materials. These are the so-called ‘highstrength’ and ‘ultra high-strength’ materials, used to create better products from a thinner metal sheet, with a substantially lower weight, but with the same strength as conventional steel. This sheet metal is utilised among other things within the automotive and transport industry. It improves safety and forms the basis for weight-saving solutions, resulting in more energy-efficient vehicles. Even within the shipping and offshore industry, these materials are successfully employed. Increasing demand Spaans continues his explanation: “High-strength steel will not only provide advantages, it will also bring some risks, since the material is similar to a tensioned coil spring. In other words: If such a steel sheet is rolled into a 30 to 40 ton coil, then much internal tension will be constrained within the material. For the safe transportation of a coil, steel manufacturers may need to wrap 10 to 12 straps on a single coil. The extensive strapping process highly increases cost and time. As an alternative, manufacturers have developed thicker, higher strength straps. A substantially lower number of straps is required of such high-strength straps.” Steel Times International
Tebulo.indd 3
Fig 2
Removable straps A general disadvantage is that this procedure entails a great risk when strapping material is cut in an unprofessional manner. Not only may the straps jump away in an uncontrolled manner, the steel coil itself can also suddenly break open with all of the related dangerous consequences. Patrick Stam, sales manager with Tebulo Industrial Robotics commented: “For the purpose of safety, our organisation developed the so-called De-Strapper many years ago. It is a robotic system that detects the straps over the full width of the coil, and removes them (Fig 1). The first De-Strapper was supplied for a pickling line in Canada. Although this type of system was very successfully utilized in recent years, it is impossible to deploy it for cutting and processing the new, thicker and stronger
straps. Among other things, this relates to the fact that the material has a double breaking force compared to traditional strapping materials. Moreover when cut, new strapping material shoots away with explosive strength, due to the extremely high internal tension of the strapped roll of steel. Therefore, the time was right to optimize and further develop the De-Strapper so that the new high-strength steel straps could be removed safely. Development of the head According to Spaans, “For the development of the new De-strapper head, our engineers searched for the optimal balance between a blade with the proper geometry on the one hand and the correct cutting angle on the other, with the optimal force and the minimum wear and tear. In order to determine the FUTURE STEEL FORUM 2022
79 09/05/2022 10:34:59
ROBOTICS
Fig 3
correct cutting and deformation behaviour, we, therefore, started by recording the process on a high-speed video. Subsequently, we have created a complex calculation method and conducted some force analyses, which eventually led to the new De-Strapper with a substantially improved cutting head. We have improved the shape of the entire construction and stiffened it with hardly any weight increase. Moreover, the cutter geometry (i.e. the cutting function) has changed. In order to cut the new straps, the knife has to somehow slide underneath the straps. This is quite a challenge since these flexible, new straps are extremely tightly wrapped around the steel coils. As soon as the cutter slides under the strapping material, the strap is lifted and cut with minimum damage to the coil. In the same motion, the other side of the strap is clamped down and thus prevented from shooting away. Stam explains: “In order to flexibly carry out these three functions of sliding, cutting and holding,
80 Tebulo.indd 4
FUTURE STEEL FORUM 2022
which require much strength in a single movement, we have made use of a hydraulic system.” Long tool life It is interesting to note that we have produced a higher strength material cutter by modifying the geometry with which the blade slides under the straps. If we had made the blade tip too sharp, the tip of the blade would
definitely break. If we make the blade tip too dull, then it is impossible to slide it under the strapping. Thanks to the cutting movement’s smarter geometric design, for which another type of material is combined with a unique hardening process, the wearable parts boast a longer lifespan, guaranteeing the cutter’s long tool life. This equals the lifespan of the cutter in the De-Strapper for standard steel rolls. From a safety perspective, we have modified the principle of the counter knife in order to prevent strapping material from shooting away. Tebulo developers, therefore, went from a one-sided clamping principle to a two-sided clamping principle. In the new De-Strapper with its innovative cutting and clamping unit, the straps are held down with a stamp. The new De-Strapper utilizes laser technology in conjunction with powerful software to detect the straps. As soon as the strap is cut, it is automatically pulled away by the De-Strapper and handed over to the ‘strapwinder’ (Figs 2, 3 and 4).
Steel Times International
09/05/2022 10:35:32
ROBOTICS
Fig 4
Disposal of waste materials In conjunction with the De-Strapper, the strapwinder has been upgraded, since it is difficult if not impossible to bend high-strength steel straps together into a compact wastematerial package. Spaans explains: “We succeeded in developing a ‘strapwinder’ that may be utilised for standard straps as well as for high-strength-steel strap material. For this purpose, we have exhaustively tested our concept designs, eventually settling for the best. An important difference with the former strapwinder model is that the distortion ratio has been completely overhauled for the new design. Moreover, our engineers have also managed to minimize the risk of strapping material getting caught within the strapwinder’s press area, so it is virtually impossible for straps to get stuck. Also noteworthy is the fact that the new strapwinder design is very low maintenance with a minimum of wearing parts. This is contrary to the popular strap choppers. Aside from the
Steel Times International
Tebulo.indd 5
new model, the former De-Strapper remains available in our product range for the removal of conventional straps. First models Spaans explains that the first two DeStrappers, including control units and robot tools, have been delivered to a high-strength steel producer in Finland, in conjunction with the strapwinder, robotic handled bin for the disposal of the waste packages and a hydraulics aggregate.” Stam elucidates: “Per project each new De-Strapper will be specifically tailored to customer requirements. The De-Strapper may be equipped with various options. Just think of its scanning functionality for roll material shape recognition, reel detection, width scanning detection, a functionality for bar code scanning and/or a camera for access control or product number identification. As standard, each De-Strapper consists of a clamping and cutting unit, as well as a laser system
for the detection of the number and exact position of straps and to perform a diameter measurement”. According to Spaans, “With this new De-Strapper on the market, we expect that the use of high-strength steel for straps will really take off, since strap removal may from now on be handled in a safe and affordable manner.” �
FUTURE STEEL FORUM 2022
81 09/05/2022 10:35:48
MACHINE VISION SYSTEMS
Classifiers for surface inspection Steelmakers wanting to increase production yield, the quality of their end products and customer satisfaction view machine vision systems as an essential piece of equipment for their processing lines.
Fig1. Combining several classification methods to increase the overall classification performance
Today, machine vision systems are an essential asset on many processing lines. Investments have been soaring and multiple new applications in the metal industry have been developed. The driving factors for these investments are process optimization, increasing quality, customer satisfaction, and production yield by sorting out defects in real time besides CO2 footprint reduction. But, given all these driving factors, what is the right inspection set-up? Inspection systems remain complex technologies, and some critical decision factors are not always well understood – especially the classification performance and its impact on return on investment (ROI) during operation. Let´s take a look at potential inspection set-ups to better understand the challenge
82
FUTURE STEEL FORUM 2022
ISRA Partytec.indd – read MM.indd 2
of quality inspection automation. Compared to an automatic inspection system, a manual inspection has a lower repeatability for key performance indicators. This is due to human factors: each operator can interpret defects differently, and manual inspection is less consistent in time. Therefore, an automatic inspection system is usually considered to be superior. But not all automatic inspection systems play in the same league. A traditional inspection system consists of machine vision technologies with a limited number of optical channels or low-quality sensors, but is especially limited in terms of defect detection –particularly on structured surfaces and defect classification. Detection deficits lead to overlooked defects, wrong defect sizing, and a low classification performance. On
the other hand, a high-end inspection system is optimized for a specified application, integrating high-quality data acquisition, selflearning detection algorithms, and best-inclass defect classification. A high amount of defect features allows superior classification performance. Practically, this classification performance is a key indicator and determines the final systems’ performance and ROI. High-end machine vision systems see a huge number of events during inspection – and not all of them are relevant defects. The classification has to divide these events into not relevant, relevant and critical. A bad discrimination leads to potential overkill and missed defects; best practice correctly determining the severity of the defects detected. This forms the basis for Steel Times International
09/05/2022 10:38:55
MACHINE VISION SYSTEMS
Fig 2. Condition monitoring with all surface inspection systems at a glance
proper quality decisions: complex grading rules will support the final decision if material is rejected, or dispatched to customers with different quality levels. Machine vision systems typically integrate software tools that can determine process flaws, analyze defect patterns and decide if the produced coil fulfills the given quality requirement. High-performance surface inspection includes a huge quantity of defect information as well as advanced classifier technologies. Classifying using simple image features like minimum and maximum grey value is no longer an option. Advanced classifiers will compare hundreds of image features of different defects to find similarities and define rules for differentiation. AI-based classifiers just require correctly classified training data, and the more different examples, the better the resulting classifier will be. In a machine learning approach (feature-based classification), the system is continuously trained according to its own needs. This is primarily via an evaluation branch or via the next similar defect – ie, the classification is generated via features such as geometry, statistics, frequency or neighborhood defects. The result is a classifier that explicitly contains the differentiating factors. Advanced surface inspection systems, dedicated for certain process line types,
Steel Times International
ISRA Partytec.indd – read MM.indd 3
already offer optimized and performant classifiers based on the experience of hundreds of installations. (Fig. 1) In contrast to the feature-based approach, deep learning (image-based classification) classifies defects without pre-calculated features – it is the task of the neural network to learn the relevant features. Deep learning works directly on representative images. Given a sufficiently large and qualitatively good image data set, the system learns the distinguishing features and classification with high precision. Relative features will not have to be defined in advance. As classification performance and simplicity is key, deep learning has become the most popular approach to developing artificial intelligence (AI) – machines that perceive and understand the world. The steelmaking industry needs the best classification performance to allow integrated quality rules that realize a fully automatic coil release as part of the smart factory. Simplicity, therefore, isn’t the optimal choice. Both classification methods’ decision trees and neural networks have their own advantages. To get the best overall precision, it is necessary to combine the results. IVAI – ISRA VISION artificial intelligence combines the results of a decision tree classifier (C5) – matured over years – with ISRA’s PEARL,
an up-to-date convolutional neural network (CNN) classifier that allows not only the most precise classification, but also root cause analysis, predictive maintenance and industry 4.0. The combination of best-in-class classification with deep learning algorithms already leads to an increase of the classification rate by up to 30% in performance tests against well-tuned inspection systems in practical use. But there is more to achieve. Even though classification is one major building block of the selflearning steel factory, there are additional complementing technologies and solutions to optimize production, delivery quality, and sustainability. Even though classification is one major building block of the self-learning steel factory, there are additional complementing technologies and solutions to optimize production, delivery quality, and sustainability. With condition monitoring, for example, steel producers are continuously informed on current machinery status via live surveillance of measures in one central database and gain the ability to react immediately on upcoming problems. • Live information gathering from worldwide production equipment via standard interfaces and interlink quality data
FUTURE STEEL FORUM 2022
83 09/05/2022 10:38:57
MACHINE VISION SYSTEMS
Fig 3. CO2 footprint seen through the SURFACE MASTER for hot strip mills’ lens, preventing defects during casting and scrap in the downstream processes
and process information. • Extraction of KPIs for quality and production. • Online visualization of inspection system and quality data with predefined rule sets via condition monitoring and KPI monitoring. • Highest system availability with reporting and alarming on KPIs for hardware, defect detection, classification and trends (Fig. 2). As a perfectly working surface inspection is a precondition for a reliable quality decision, condition monitoring forms the basis for SQMS (Surface Quality Management System). In addition, condition monitoring enables you to permanently monitor the current state of the inspection system’s health (hardware, image acquisition including illumination and cameras, detection classification), the produced quality – including defect class statistics and quality decisions – and the millwide system state with all production lines on one screen. As the working environment gets more dynamic, access to all of this information is granted on any web-capable device from mobile phone to tablets to PCs. Increasing productivity and optimizing processes is not only a cost saving measure, it also helps steel factories to significantly
84
FUTURE STEEL FORUM 2022
ISRA Partytec.indd – read MM.indd 4
reduce their carbon footprint. The European Green Deal commits the EU to carbon neutrality by 2050 and lawmakers want the industry to ‘pave the way’ with a transition towards net zero emissions for a cleaner, healthier future. The whole European steel industry is being driven to reduce its direct and indirect CO2 emissions. 2050 may sound like a long way off, but innovations from ISRA Parsytec today are already supporting customers in reducing their overall CO2 footprint by • Lowest possible energy consumption per system. • Reducing the overall scrap rate.
Increasing the entire yield. Most efficient re-routing of material with the link to the factory´s ERP system (Fig. 3). Early recognition of quality-relevant problems in steel production allows rapid intervention and ensures the highest saleable throughput, and lower costs. However, while there is no single solution, ISRA´s holistic view from different angles and on different areas of production, is the right path towards sustainable and cost-optimized production that goes hand-in-hand with the necessary emissions through energy and raw material savings. � • •
Steel Times International
09/05/2022 10:39:02
www.harald-wehrle.de
The Future is Manless
®
Technology Protecting People Spray Technology
TSM/LM2
SLD Smart Leakage Detector
SandMan
SAFE STEEL
EBT Balcony Flap
MultiROB
SAFE EBT EBT Monitoring
THM Tap Hole Manipulator
EBT Sand filling
Badische Stahl-Engineering GmbH Robert-Koch-Straße 13 D-77694 Kehl/Germany Phone (+49) 7851/877-0 Fax (+49) 7851/877-133 eMail info@bse-kehl.de www.bse-kehl.de
We are Steelmakers!
BSE0393_anz_4_2022_safety_A4_GB.indd 1
18.03.22 17:12
Integrated logistics.
Shortened lead times. Automated handling, packing and storing systems for the metal industries.
Pesmel – The Material Flow How® company pesmel.com