Underwater Technology 34.3 by SUT

Vol. 34 32 No. No. 332 2017 2014 Vol.

UNDERWATER TECHNOLOGY

101

Editorial Underwater Technology accepted into the Emerging Sources Citation Index (ESCI)

Dr Martin DJ Sayer

103

A Personal View... AUV development trends and their implications for risk management strategies

M P Brito

107

Trust model for cluster head validation in underwater wireless sensor networks

Nitin Goyal, Mayank Dave and Anil K. Verma

115

Handling free gas in deep and ultra-deep water drilling risers: a technical review and safety case explanation

Paul A Potter

129

Technical Briefing Application of the threat matrix to improving the efficiency of risk assessments for the integrity management of subsea pipeline systems

JHA Baker

135

Technical Briefing Underwater monitoring system for body temperature and ECG recordings

Andreas Schuster, Olivier Castagna, Bruno Schmid, Tobias Cibis and Arne Sieber

ISSN 1756 0543

141

Book Review Modern Observational Physical Oceonography: Understanding the Global Ocean

143

Book Review Robot Fish: Bio-inspired Fishlike Underwater Robots

www.sut.org

0A-SUT-34(3)-OFC.indd 1

29/07/17 10:31 am

UNDERWATER TECHNOLOGY Editor Dr MDJ Sayer Scottish Association for Marine Science Associate Editor G Griffiths MBE Autonomous Analytics Associate Editor Subsea Engineering LJ Ayling Maris International Ltd Assistant Editor E Azzopardi SUT Editorial Advisory Board Chairman Dr MDJ Sayer Scottish Association for Marine Science Gavin Anthony GAVINS Ltd LJ Ayling Maris International Ltd Prof DS Cronan Imperial College London G Griffiths MBE Autonomous Analytics Prof C Kuo FRSE Strathclyde University Dr WD Loth WD Loth & Co Ltd Dr S Merry Renewable Energy Association & Focus Offshore Ltd Prof J Penrose Curtin University of Technology Prof WG Price FRS FEng Southampton University Prof MF Randolph University of Western Australia Dr R Rayner Sonardyne International Ltd Prof R Sutton Plymouth University Prof P Wadhams University of Cambridge Cover Image (top): zoonar.com/syrist

Society for Underwater Technology Underwater Technology is the peer-reviewed international journal of the Society for Underwater Technology (SUT). SUT is a multidisciplinary learned society that brings together individuals and organisations with a common interest in underwater technology, ocean science and offshore engineering. It was founded in 1966 and has members in more than 40 countries worldwide, incIuding engineers, scientists, other professionals and students working in these areas. The Society has branches in Aberdeen, London and South of England, and Newcastle in the UK, Perth and Melbourne in Australia, Rio de Janeiro in Brazil, Beijing in China, Kuala Lumpur in Malaysia, Bergen in Norway and Houston in the USA. SUT provides its members with a forum for communication through technical publications, events, branches and specialist interest groups. It also provides registration of specialist subsea engineers, student sponsorship through an Educational Support Fund and careers information. For further information please visit www.sut.org or contact: Society for Underwater Technology 1 Fetter Lane EC4A 1BR London UK e info@sut.org t +44 (0)20 3440 5535 f +44 (0)20 3440 5980

Scope and submissions The objectives of Underwater Technology are to inform and acquaint members of the Society for Underwater Technology with current views and new developments in the broad areas of underwater technology, ocean science and offshore engineering. SUT’s interests and the scope of Underwater Technology are interdisciplinary, covering technological aspects and applications of topics including: diving technology and physiology, environmental forces, geology/geotechnics, marine pollution, marine renewable energies, marine resources, oceanography, salvage and decommissioning, subsea systems, underwater robotics, underwater science and underwater vehicle technologies. Underwater Technology carries personal views, technical papers, technical briefings and book reviews. We invite papers and articles covering all aspects of underwater technology. Original papers on new technology, its development and applications, or covering new applications for existing technology, are particularly welcome. All papers submitted for publication are peer reviewed through the Editorial Advisory Board. Submissions should adhere to the journal’s style and layout – please see the Guidelines for Authors available at www.sut.org.uk/journal/default.htm or email elaine.azzopardi@sut.org for further information. While the journal is not ISI rated, SUT will not be charging authors for submissions.

Cover Image (bottom): Steve Crowther

Publication and circulation

Cover design: Quarto Design/ kate@quartodesign.com

Underwater Technology is published in March, July and November, in four issues per volume. The journal has a circulation of 2,400 copies to SUT members and subscribers

0B-SUT-34(3)-IFC.indd 1

in more than 40 countries worldwide, including over 190 Corporate Members of the Society.

Disclaimer and copyright The Society does not accept responsibility for the technical accuracy of any items published in Underwater Technology or for the opinions expressed in such items. The copyright of any paper published in the journal is retained by the author(s) unless otherwise stated. All authors are supplied with a PDF version of their papers once published. Authors are encouraged to make the PDF version of their papers free to download from their own websites.

Open Access Underwater Technology is available as Open Access. PDF versions of all published papers from Underwater Technology may be accessed via ingentaconnect at www. ingentaconnect.com/content/sut/unwt. All issues from Volume 20 (1995) onwards are available as Open Access. The Society for Underwater Technology also encourages Underwater Technology authors to make their papers available online on their personal and/or institutional websites for Open Access. Through this arrangement, the Society supports the Open Access policy not only in the UK (the Research Councils UK (RCUK) policy) but also the drive towards Open Access in other countries.

Abstracting and indexing Abstracting and indexing services covering Underwater Technology include: American Academy of Underwater Sciences (AAUS) E-Slate; Aquatic Sciences and Fisheries Abstracts (Biological Sciences and Living Resources; Ocean Technology, Policy and Non-Living Resources; and Aquatic Pollution and Environmental Policy); Compendex; EBSCO Discovery Service; Fluidex; Geobase; Marine Technology Abstracts; Oceanic Abstracts; Scopus; and WorldCat Discovery Services.

Subscription Subscription to the print version of Underwater Technology is available to non-members of the Society at the following rates per volume (single issue rates in brackets). Prices are given in GBP. Accepted methods of payment are cheque or credit card (MasterCard and Visa). Foreign cheques must be in GBP and drawn on a British bank otherwise a currency conversion surcharge is incurred. UK subscription Overseas subscription

£102.00 (£25.50 per issue) £108.00 (£27.00 per issue)

Underwater Technology is also available in electronic format via ingentaconnect as Open Access. To subscribe to the print version of the journal or for more information please email Elaine Azzopardi at elaine.azzopardi@sut.org

Advertising To book an advert or for more information please contact Elaine Azzopardi at elaine.azzopardi@sut.org

28/07/17 7:17 pm

www.sut.org

Editorial

doi:10.3723/ut.34.101 Underwater Technology, Vol. 34, No. 3, pp. 101–102, 2017

Underwater Technology accepted into the Emerging Sources Citation Index (ESCI) Dr Martin DJ Sayer, Editor

In 2013 I wrote an editorial which explained what the ISI Impact Factor was, why the Society’s journal Underwater Technology wasn’t ISI-rated and the measures we were taking to correct that (Sayer, 2013). Since then, as well as maintaining a regular publication rate, we have made a number of improvements to how the journal is indexed but have also changed its publication format with it now having full digital open-access. Over the same period, the company that maintains the ISI database, now called Thomson Reuters/Clarivate Analytics, have created a secondary tier to their citation databases with the introduction of the Emerging Sources Citation Index (ESCI). This new index is described as a method for providing earlier visibility for sources under evaluation as part of the wider journal selection process for the ISI databases. As a result of demonstrating consistent publication, and the steady increase in the number of citations our published articles are receiving, the journal was accepted into the ESCI on 7th April 2017. This means that the journal is now under formal consideration for regaining its Impact Factor and, after two years of coverage in ESCI, will be evaluated for re-entry into the flagship indexes, including the Science Citation Index Expanded (SCIE). Journals indexed in the ESCI do not receive Impact Factors. However, the citations from the ESCI are now included in the citation counts for the Journal Citation Reports and, therefore, contribute to the Impact Factors of other journals. The journal will now be discoverable via the Web of Science with an identical indexing process to any other indexed journal. This includes full citation counts and author information. Therefore, because Underwater Technology articles are now indexed in ESCI, they will contribute to an author’s H-Index calculation and be included in any analyses conducted on Web of Science data or

related products such as InCites. This means that, from this year onwards, publishing in Underwater Technology carries all the benefits to authors that would be expected in an official Impact Factor journal. At this time, the Society for Underwater Technology is continuing to make the journal completely and instantly open access with no article publication charges. The journal is hosted in electronic format by Ingenta Connect (www.ingentaconnect.com/ content/sut/unwt), it is available in e-reader format (issuu.com/sut7), and all articles are provided with DOI (digital object identifier) numbers. The journal is now licensed under a Creative Commons Attribution 4.0 International License (CC-BY). As such the Society for Underwater Technology encourages authors to make their full articles freely available as soon as they are published, both online on their personal and/or institutional websites and on research social media sites such as, for example, ResearchGate. Since becoming open access in 2014, and with no author charges, the rate of citing articles in Underwater Technology has increased markedly. The journal’s CiteScore for 2016 was 1.11 compared with scores of 0.58 and 0.22 for the years 2015 and 2014, respectively. This ranks the journal in the 64th percentile in the Scopus “Ocean Engineering” category (ranked number 29 out of 82 journals). Likewise, the 2016 ScimagoJR two-year citations per document rating (similar to the ThomsonReuters Impact Factor) has increased to 1.529, compared to 0.964 and 0.400 for years 2015 and 2014, respectively. The journal was last listed on the Web of Science in 2006. It has been a significant task to get it readmitted to this new index and many thanks must go to all the authors who have submitted to the journal in that time and especially to the support that has come from the SUT’s own special interest groups. With this new listing on the ESCI combined

101

01-SUT_Editorial_34(3).indd 101

28/07/17 7:17 pm

Dr Martin DJ Sayer. Underwater Technology accepted into the Emerging Sources Citation Index (ESCI)

with free gold status open access publishing, the future for the journal looks encouraging. There is now genuine academic value in publishing in Underwater Technology. For those interested in publishing with us then I should encourage prospective authors to visit the “scope and submissions” webpage (www.sut.org/publications/underwatertechnology/scope-submissions/) while also taking

note of our guidelines for authors plus the mandatory submission form.

References Sayer, M.D.J. (2013). Editorial: Underwater Technology and the ISI Impact Factor. Underwater Technology 31, 161–163. [doi:10.3723/ut.31.161]

102

01-SUT_Editorial_34(3).indd 102

28/07/17 7:17 pm

www.sut.org

A Personal View...

doi:10.3723/ut.34.103 Underwater Technology, Vol. 34, No. 3, pp. 103–105, 2017

AUV development trends and their implications for risk management strategies Autonomous underwater vehicles (AUV) have been under substantial development since the 1980s. The first AUV, the self-propelled underwater research vehicle (SPURV), was built in 1957, at the University of Washington’s Applied Physics Laboratory (Widditsch, 1973). Other early AUVs were built in the 1980s, such as the L’Epaulard and the ARCS built by the Institut Français de Recherche pour l’Exploitation de la Mer (IFREMER, 2017) and International Submarine Engineering (ISE, 2017) respectively. Here I argue that the risk management strategy adopted in the early days is still in use for most AUV operations but is unsuitable for informing decision making for modern AUV operations. A risk management strategy, or strategic framework, is a multifaceted set of design considerations that underpin the implementation of the risk management process (Ward, 2005). It is partly concerned with the philosophical and cultural context for risk management practice, and seeks to influence and improve how people engage with problems or situations. For example, one concept commonly identified as a vital enabler for early and effective responses to possible risk is ‘mindfulness’ (Weick and Sutcliffe, 2001) which is perhaps best known as a state of mind advocated by the teachings of Buddhism where it promotes meditation in order to reflect on experiences. Mindfulness, when considered as a risk management strategy, comprises psychological techniques aimed at ensuring constant vigilance

against the unexpected. It consists of a combination of on-going scrutiny of existing expectations, and continuous refinement and differentiation of expectations based on new experiences. Arguably, mindfulness was the risk management strategy adopted by the early AUV owners. One of the dangers of following a mindfulness risk management strategy is that it consumes a great deal of resources in attending to what often turn out to be false positive errors. Many AUV pioneers had only one vehicle to operate and this understandably influenced a conservative operational mindset. There was relatively little scope for experimental learning through flexibility (Hamblin, 2002). This is a risk management strategy that advocates the definition of alternative states of success and ongoing experimentation to learn and re-evaluate what success can mean. The exception to conventional AUV deployments are the long endurance missions carried out underneath ice covered areas, such as the missions of Autosub 3 under the Pine Island Glacier in 2009 and 2013 and the missions of ISE Arctic Explorer as part of the Cornerstone Project (Brito et al., 2010; 2012). Here a resilience risk management strategy was adopted, which favoured mitigation rather than a constant review of objectives. For these missions, mitigation was applied in terms of improving the robustness of design vulnerabilities and introducing a monitoring distance. The resilience philosophy seeks to manage the entire cycle of unexpected events from first detection

Dr Mario P Brito Dr Mario P Brito is a Lecturer in Risk Analysis and Risk Management at the University of Southampton. He has a degree in Aeronautical Engineering from the University of Beira Interior, Portugal and a PhD in Software Reliability Assurance and Management, awarded by the University of Bristol. Dr Brito has conducted several risk and reliability analyses of autonomous underwater vehicles and delivers courses to the industry on how to use these methods. He has also chaired several accident investigations of underwater technology. Dr Brito is the Deputy chair of the SUT/ECOR panel on Underwater Robotics, he is Co-Chair of the European Safety and Reliability Association Committee on Marine and Offshore Technology and is a member of the Institute of Engineering and Technology (IET).

through crisis management and eventual return to normalcy. Such mitigations were planned on a combined ex ante and ex post basis that is, through the planning of both precautionary and remedial risk controls. In the last five years, technological developments and substantial investment from both government

103

02-SUT289_34(3).indd 103

28/07/17 7:17 pm

M Brito. AUV Development trends and their implications for risk management strategies

and industry have led to an increased confidence that AUV technology can be used successfully for more complex applications. These trends, some of them highlighted in the early days of AUV development, force us to question the suitability of risk management strategies adopted thus far. Some trends by their own nature imply the use of a certain risk management strategy, whereas for other trends there is currently a lack of a risk management strategy. There are three key new trends in AUV development that could benefit from a new risk management strategy: 1. Higher autonomy. Several large research projects have sought to increase the autonomy of AUVs. The aim is to augment AUV capability by enabling platforms to conduct intervention tasks such as valve turning (PANDORA, 2017), to search for explosive mines and to identify and track physical, chemical and biological features (Monterey Bay Aquarium Research Institute, 2017; Natural Environment Research Council, 2017). 2. Multiple vehicle operations. Self-configuration AUVs have been proposed as solutions for optimal sampling of ocean fronts, large eddies and other mesoscale ocean processes (L’Hévéder et al., 2013). Virtual moorings also provide a solution for expensive mooring deployments (Alvarez and Mourre, 2011; Hodges and Fratantoni, 2009; Smeed et al., 2013). 3. Deeper and long endurance AUVs. Both commerce and science have a stake in increasing the depth and range of platforms. Ocean explorations and sea mining are expected to reach deeper regions of the ocean than ever and there is now a strong emphasis on developing AUVs capable of

operating at depths of 6000 m (McPhail, 2008). Mindfulness is of course important for any new technology development. However, for a wider group of AUV users, developers must adopt a fuller range of risk management strategies that must be hybridised and sometimes traded off against one another. A criticism raised about the first trend listed above is the uncertainty associated with the platform behaviour at a given point of the mission. To address this problem a combination of other general strategies such as anticipation and resilience is required. Both imply that the mission objective must be met. In contrast to resilience, anticipation does not favour mitigation. The mind set for anticipation is one that favours precise identification of possible problems so that specific remedies can be designed or recalled. The use of this risk management strategy alone can be dangerous, because it presumes a level of understanding that is impossible to achieve when dealing with unknowable and unpredictable environments. A resilience strategy can also be applied to higher level autonomous systems by the implementation of mitigation functions in the design or operation phases. Multiple vehicle operations favour the use of flexibility or slack. Here we have both the ability to reach other acceptable states and excess capability (Nãslund, 1964). In technical terms going deeper and for longer missions only implies more robust pressure housings and connections, more powerful energy supplies and more energy efficient systems and design. Of course, developers must be mindful of the technological limitations. However, resilience is also crucial

particularly in mission planning, where the mitigation actions and recovery procedures must be defined for different stages of the deployment. These arguments do not consider any overlap between trends. If there is overlap, then the risk strategies suggested for the trends involved must be applied. For example, if for one application, the developer is considering a long endurance mission with high autonomy then anticipation, mindfulness and resilience strategies should be applied. Once there is agreement on a risk management strategy, the next stage is to consider how to implement the strategy in a given organisation or project. Here several risk management frameworks can be adopted. The IRM framework provides guidelines for implementing risk management processes (IRM, 2002). This and other risk management frameworks have been developed by risk managers, engineers, accountants and financial analysts. These can be tailored to any technology or organisation. In my view, a fundamental feature in persuading others to adopt AUVs is the level of transparency regarding risks as well as opportunities. Naturally, being very enthusiastic about technology makes developers better at articulating the opportunities rather than the risks. Identifying the risk management strategy before building the risk management process enables developers to objectively communicate the advantage of autonomous underwater vehicles.

References Alvarez AA and Mourre BB. (2012). Optimum Sampling Designs for a Glider–Mooring Observing Network. Journal of Atmospheric and Oceanic Technology 29: 601–612. Brito MP, Griffiths G and Challenor P. (2010). Risk Analysis for Autonomous

104

02-SUT289_34(3).indd 104

28/07/17 7:17 pm

Underwater Technology Vol. 34, No. 3, 2017

Underwater Vehicle Operations in Extreme Environments. Risk Analysis 30: 1771–1788. Brito M, Griffiths G, Ferguson J, Hopkin, D, Mills R, Pederson R and MacNeil E. (2012). A Behavioral Probabilistic Risk Assessment Framework for Managing Autonomous Underwater Vehicle Deployments. Journal of Atmospheric and Oceanic Technology 29: 1689–1703. Hamblin DG. (2002). Rethinking the management of flexibility—a study in the aerospace defence industry. Journal of the Operational Research Society 53: 272–282. Hodges BA and Fratantoni DM (2009). A thin layer of phytoplankton observed in the Philippine Sea with a synthetic moored array of autonomous gliders. Journal of Geophysical Research 114: C10020. Institut Français de Recherche pour l’Exploitation de la Mer (IFREMER). (2017). Institute of Risk Management (IRM). (2002). A risk management standard. London: Institute of Risk Management. Available at: www.theirm.org/ media/886059/ARMS_2002_IRM. pdf <last accessed on 25/05/2017>.

L’Epaulard. Available at: http://wwz. ifremer.fr/grands_fonds/Les-moyens/Les-engins/Les-robots/RobotsIfremer/L-Epaulard <last accessed 3 May 2017>. International Submarine Engineering (ISE). (2017). ARCS. Available at: www.ise.bc.ca/arcs.html <last accessed on 03/05/2017>. L’Hévéder B, Mortier L and Testor P. (2013). A glider network design study for a synoptic view of the oceanic mesoscale variability. Journal of Atmospheric and Oceanic Technology 30: 1472–1493. McPhail S. (2008). Autosub6000: a deep diving long range AUV. In: Proceedings of the 2nd International Conference on Underwater System Technology Theory and Applications (USYS), 4–5 November 2008, Bali, Indonesia. 6pp. Monterey Bay Aquarium Research Institute (MBARI). (2017). Tracking drifting algal blooms and the nutrients that keep them going. Available at: www. mbari.org/tracking-drifting-algalblooms-and-the-nutrients-that-keepthem-going <last accessed 3 May 2017>. Nãslund B. (1964). Organizational Slack. Scandinavian Journal of Economics 66: 26–31.

National Environment Research Council (NERC). (2017). NERC Autonomy Joint Strategic Research. Available at: www.nerc.ac.uk/research/ funded/programmes/autonomous/ news/miaos-ao/dstl/ <last accessed 25/5/2017>. PANDORA (2017). Persistent Autonomous Robots (PANDORA). Available at: http://persistentautonomy. com/ <last accessed 3 May 2017>. Smeed D, McCarthy G and White D. (2013). Underwater gliders as virtual moorings; lessons from the RAPID program. In Proceedings of the European Geosciences Union General Assembly 2013, 7–12 April 2013, Vienna, Austria. Ward S. (2005). Risk Management Organization and Context. London: Witherlby & Co. 222pp. Weick KE and Sutcliffe KM. (2001). Managing the Unexpected: resilient performance in an age of uncertainty. San Francisco: John Wiley & Sons, 194pp. Widditsch HR. (1973). SPURV- The First Decade, APL-UW 7215. University of Washington. Available at: http://dtic. mil/dtic/tr/fulltext/u2/a050816.pdf <last accessed 25/5/2017>.

105

02-SUT289_34(3).indd 105

28/07/17 7:17 pm

CALL FOR PAPERS Underwater Technology: InternaƟonal Journal of the Society for Underwater Technology The Society for Underwater Technology is calling for papers for its internaƟonal journal, Underwater Technology. The journal publishes peer-reviewed technical papers on all aspects and applicaƟons of underwater technology, including: • • • • • • • • • • • • •

diving technology and physiology environmental forces geology/geotechnics marine polluƟon marine renewable energies marine resources oceanography subsea systems underwater acousƟcs underwater roboƟcs underwater science underwater vehicle technologies salvage and decommissioning

Original papers on new technology, its development and applicaƟons, and papers covering new applicaƟons for exisƟng technology, are parƟcularly welcome. Submissions should adhere to the journal’s guidelines available at www.sut.org/publicaƟons/underwater-technology/guidelines-for-authors/ For more informaƟon or to make a submission, please contact the Assistant Editor, Elaine Azzopardi, at Elaine.Azzopardi@sut.org

02-SUT289_34(3).indd 106

28/07/17 7:17 pm

www.sut.org

Technical Paper

doi:10.3723/ut.34.107 Underwater Technology, Vol. 34, No. 3, pp. 107–114, 2017

Trust model for cluster head validation in underwater wireless sensor networks Nitin Goyal*, Mayank Dave1 and Anil K. Verma2 1 Department of Computer Engineering, National Institute of Technology, Kurukshetra, Haryana, India 2 Department of Computer Science and Engineering, Thapar University, Patiala, Punjab, India Received April 2017; Accepted May 2017

Abstract In underwater wireless sensor networks (UWSNs), there are a number of sensor nodes that perform collaborative monitoring tasks over a deﬁned area. The sensor nodes move into a compromised stage during network operations for various adverse reasons. These compromised nodes are unable to trust each other’s capabilities to run a network, thus creating a hindrance in network operations. In the past, various trust models based on different parameters were designed to enhance trust among nodes, but these models ignored the existence of redundant data. In this paper, a trust-based security model for UWSNs is proposed to enhance the trust among the nodes in the network. The total trust of each cluster head (CH) is taken as a sum of its direct trust and its recommendation trust value, which are a combination of communication trust, data collection trust and energy trust. If the trust value of a CH is below a threshold value, the CH will be replaced by a CH with a higher trust value. This paper provides a comparative analysis of the proposed trust model with an existing technique, and discusses the efﬁciency of the proposed technique on the basis of increased trust and lower risk of attack among the nodes. Keywords: communication, security, trust, underwater wireless sensor network, UWSN

1. Introduction Underwater wireless sensor networks (UWSNs) have proven their strength in various applications of monitoring, exploration, surveillance, attack protection and tracking in harsh underwater environment (Das and Thampi, 2015). The underwater channel is characterised by prolonged propagation time and frequency-dependent attenuation that are both highly affected by the distance between * Contact author. Email address: er.nitin29@gmail.com

deployed nodes as well as by the link orientation (Dhurandher et al., 2009). An UWSN consists of a number of nodes that process the data and transmit them to other nodes or a sink (Goyal et al., 2016). Radio signals do not work underwater due to high attenuation, and so acoustic signals are used in underwater networks (Goyal et al., 2014). Cluster-based UWSNs have been investigated by researchers to achieve the network scalability and efficiency that maximise network lifetime and reduce bandwidth consumption by using local collaboration among sensor nodes (Kumar and Goyal, 2014). Because of the broadcast nature of acoustic communication, cluster-based UWSNs are vulnerable to many critical security attacks, including Sybil attacks, replay attacks and message manipulation attacks (Senel et al., 2015). Ensuring node security is a basic and essential requirement to improving node location accuracy and reliability (Zenia et al., 2016). Cluster heads (CHs), which are elected to manage local clusters, become adversaries’ prime targets. If one CH is captured or compromised, the entire local cluster will be affected by attacks. These compromised sensor nodes cannot trust one another to run a given network protocol in order. This lack of trust between sensor nodes is detrimental to the operations of the whole network (Albagory and Said, 2015; Jia and Meng, 2016; Bahl et al., 2015; Liu et al., 2015; Sabet and Naji, 2015). A node has to follow specific rules in a network and this helps in building trust among the nodes. The trust is qualitative and asymmetric in nature (Xu et al., 2015). If the rules are not followed by a particular node, then it is identified as an errant node – i.e. to be eliminated from the network to restrict further communication. Thus for smooth

107

03-SUT286_34(3).indd 107

28/07/17 7:17 pm

Goyal et al. Trust model for cluster head validation in underwater wireless sensor networks

working of a network and better communication, a trust management framework is required. In a trust management system, each node observes the neighbour node and evaluates the trust factors of forwarded data packets and broadcasted data packets. The sum of these trust factors indicates the trustworthiness of a neighbouring node. A trust management framework must consider different trust factor parameters that build trust in the network for faster data transmission. Various security challenges occur owing to weak communication in a network, and the performance of a network becomes a major issue in such a difficult environment (Yavuz and Ning, 2012). In a cluster-based UWSN environment, if a CH performs attacks by non-cooperating selective dropping and injecting false data repeatedly, the entire data aggregation process is affected. Hence the trustworthiness of CHs should be ensured before aggregation initiates. This can be achieved by assigning suitable trust or reputation value to the CH. The existing trust models are based on packet loss and packet error parameters caused by selective forwarding and selfish behaviour of a node. However, it is noted that due to lack of trust there is a transmission of false or redundant data which should also be taken care of. This paper proposes a trust-based security model for UWSN.

2. Related work To date, researchers have designed various trustbased models for fast and accurate data transmission. Han et al. (2016) developed the collaborative secure localisation trust model (CSLT) to ensure location security based on trust model. The model divides localisation into five processes based on selection, localisation and trust evaluation of a node. An attack-resistant trust model based on multidimensional trust metrics (ARTMM) is a trust model for underwater acoustic sensor networks (UASNs) that was introduced by Vennila and Madhura (2016). It considered multidimensional trust metrics along with energy levels and communication. The system adopts the autoregressive integrated moving average (ARIMA) model to predict packet loss. The system achieves trustworthiness by correction and similarity analysis. Geetha and Chandrasekaran (2014) proposed a trust management system for wireless sensor networks (WSNs) to identify different parameters and trust-based factors to influence trust variation. Kaur et al. (2016) introduced a secure trust-based key management system (STKF) that relies on the past and present interaction between the nodes. In this framework, the faulty node is isolated from the

network route and a link is created between the rest of the nodes. Ren et al. (2013) have presented an approach for data distribution that considers forward and backward secrecy along with data reliability. An improvement in iterative filtering (IF) techniques is proposed by Rezvani et al. (2015). This is achieved by inserting an approximation of error parameters such as variance and biasing in WSNs, thus making network collision robust along with accurate and faster converging. Jadidoleslamy et al. (2016) designed a scheme for trust estimation and forecasting using fuzzy logic to combine direct and indirect trust. However, the overhead is high for large networks because each node monitors the behaviours of its neighbours. Yavuz and Ning (2009) aimed to mitigate the ongoing adverse effects that put sensors into a compromising stage and to extract the data stored. Their approach allows a signer to create a publicly verifiable signature in sequence, along with nominal cost of computation. Jahanshahi et al. (2013) describes the algorithm based on fuzzy c-means in order to group the jammed signals of wireless cellular networks. The jamming detection performance is effectively improved at the base station by considering Doppler shift. Assessing the credibility of clusters by using fuzzy logic is evaluated by Thuc and Insoo (2011) who use the event detection algorithm to maximise detection accuracy. Existing works on trust-based data transmission techniques concentrate on packet loss and packet error parameters, thereby improving the reliability of the network, ignoring the trust of the network related to false data or redundant data injection. This motivated the authors to develop a technique that enhances the node reliability by decreasing the packet drop and increasing accuracy in faulty packet detection and packet delivery ratio, along with improved energy efficiency.

3. Trust model for cluster head validation In a cluster, some of the nodes that are termed ‘malicious nodes’ or ‘attackers’ behave improperly by attacking other nodes or sending false data. These attackers create disturbance in the network operations and their behaviour directly or indirectly hampers the performance of the whole network. In this way, the attackers or the malfunctioning nodes put the network at greater risk, thus it becomes essential to build trust among the nodes so that they can accept the data to the extent of a minimum threshold value. Various schemes have been proposed that calculate the trust value of the node based on different parameters. This paper proposes

108

03-SUT286_34(3).indd 108

28/07/17 7:17 pm

Underwater Technology Vol. 34, No. 3, 2017

Cluster formation and CH selection

Trust value estimation

Communication trust

Data collection trust

Energy trust

Trusted cluster based secure communication

Fig 1: Block diagram of proposed security model TMCHV

a trust based security model for cluster head validation (TMCHV) for UWSN.

3.1. System model Fig 1 shows the block diagram of the proposed TMCHV. In this model, initially cluster formation and cluster head (CH) selection are performed as illustrated in our previous work (Goyal et al., 2014; 2016). The total trust value of each node is estimated as the sum of direct trust (DT, which is the trust value of a particular node for another node) and the recommendation trust (RT, which is the trust value of a node recommended by the remaining nodes). The main feature of the proposed model is that DT and RT are individually based on three parameters: communication trust, data collection trust and energy trust. A trusted cluster based secure communication algorithm is then proposed. In this algorithm, a node or a CH selectively accepts the data transmitted by another node and forwards them to the CH or the sink. This results in selectively accepting the incoming data and forwarding them to the destination. Furthermore, if the trust value of any CH falls below a minimum threshold, the sink will replace the relevant CH with another CH having high trust value.

UW-SINK CH

O BCH

BCH

G J K C2

UW-SINK CH

O BCH

BCH G

J K

3.2. Cluster formation along with cluster head and backup cluster head selection In this technique, backup cluster heads (BCHs) are also elected along with primary CHs using fuzzy logic. The nodes deployed in the network initially broadcast â&#x20AC;&#x2DC;HELLOâ&#x20AC;&#x2122; messages to its neighbours. They include parameters such as residual energy (Eres), distance(D), node density (ND), load(L) and link quality (LQ). Utilising the parameter values, each node analyses the parameters condition using fuzzy logic technique. The chance of a node being a CH or BCH is estimated by using fuzzy rules. The selected CHs must behave like aggregator nodes. For a node to become a CH, it must possess at least four of the ideal conditions, and to become a BCH it must possess at least three

Fig 2: Cluster formation along with CH and BCH selection

of the ideal conditions. However, the priority to become a CH or BCH will be given to the node with maximum ideal conditions. In all circumstances, the formation of a CH will be given priority over the formation of BCH. Fig 2 demonstrates the cluster formation along with CH and BCH selection phase. The nodes N and I are selected as CHs, and nodes O and J are selected as the BCH. Each CH gathers the data from its members (indicated by red or dark grey

109

03-SUT286_34(3).indd 109

28/07/17 7:17 pm

Goyal et al. Trust model for cluster head validation in underwater wireless sensor networks

arrows) as well as from its lower level CHs (indicated by blue or light grey arrows) and forwards them towards the sink.

3.3. Trust value estimation The trust value of each CH is categorised into communication trust, data collection trust and energy trust. 3.3.1. Communication trust Communication trust reflects the number of fair attempts made towards delivering the packets from one node to another node. The factors that generally affect the correct delivery of packets include packet loss (which are completely lost) and packet error (corrupted packets), and communication trust is explicitly based on these two events. Both data packets and control packets are considered. The communication trust (Tc) is derived by Han et al. (2016) as: P1 + 1 Tc = P1 + P2 + 2

3.3.2. Data collection trust Data collection trust reflects the number of honest attempts made towards collecting and forwarding the data packets from one node to another node. The factors that generally affect the data collection process are aggregation and packet forwarding ratio. Aggregation ratio is the ratio of packets aggregated correctly to the total number of packets aggregated. Forwarding ratio is the ratio of the number of packets correctly forwarded to the total number of packets forwarded. Thus, it is clearly stated that data collection trust is based on aggregation and packet forwarding ratio. Then data collection trust (Tdc) is derived by: Tdc =

qr =

Pfc Pf

(7)

ar =

Pac Pa

(8)

(1)

Here,

where:

P1 = Ps + Pc

(2)

P2 = Pm + Pr

(3)

ar Aggregation ratio;

Pm = P − K

(4)

Pfc Packets forwarded correctly;

Ps = P − Pm

(5)

Pf Total number of packets forwarded; Pa Total number of packets aggregated.

Number of successfully received packets; Number of correct packets; Number of packets lost due to malicious nodes; Number of wrong packets; Packet loss caused by unreliable acoustic channel.

Note 1: The value of Tc lies in the range of 0 to 1 (i.e. 0 ≤ Tc ≤ 1). Note 2: Based on the signal to noise ratio (SNR) and the modulation mode (m) of acoustic communication, the average bit error rate (BER) can be calculated as: (5a)

where, l is the communication distance between the nodes and f is the frequency of the acoustic signal. Then the packet loss caused by unreliable acoustic channel K is given by: K = ψ F (b, BER(I, f )) where, b is the length of packet.

Forwarding ratio;

Pac Packets aggregated correctly;

where:

BER (l, f ) = φM (SNR(I, f ))

(6)

Here:

where P is the number of broadcasted packets.

Ps Pc Pm Pr K

qr + ar 2

(5b)

Note 3: The value of Pdc lies in the range of 0 to 1 (i.e. 0 ≤ Tdc ≤ 1). 3.3.3. Energy trust Energy trust reflects the amount of total energy spent by a node in genuine operations. The factor that generally affects the energy trust is the remaining battery capacity at a specific time interval. The remaining battery energy is calculated by subtracting the total energy consumption from the initial battery energy. Therefore, energy trust is the ratio of remaining battery energy to the total battery energy. If we are given the electronics energy (Ee) and amplifier energy (Ea), then the energy consumed by the sender and receiver for transmitting x bits message through distance d is derived as follows: The energy consumed by the sender (Etx) is given by: Etx = Ee∙x + Ea∙x∙d 2

(9)

110

03-SUT286_34(3).indd 110

28/07/17 7:17 pm

Underwater Technology Vol. 34, No. 3, 2017

The energy consumed by the receiver (Erx) is given by: Erx = Ee∙x

(10)

Then the residual energy of each node (Eres) following one data communication is derived as: Eres = [Ei − Etx +(Erx)]

(11)

where, Ei is initial energy of the node. The energy trust (Te ) is then given by: Eres (12) Ei Note 4: The value of Te lies in the range of 0 to 1 (i.e.) 0 ≤ Te ≤ 1 Te =

3.3.4. Total trust value The total trust value (T ) is given by the sum of communication, data collection and energy trusts: T = Tc + Tdc + Te

(13)

Based on notes 1, 3 and 4, the total trust value lies in the range of 0 to 3 (i.e. 0 ≤ T ≤ 3).

3.4. Trusted cluster-based secure communication algorithm In a cluster-based underwater network, the moment when a CH intends to transfer the data to sink, it establishes a trusted secure communication in the network. The same is illustrated in the form of algorithm by assuming S as a sink. Let DTij be the direct trust value of CHi observed by another CHj and assuming that DTth and RTth are the threshold values of DT and RT. Another CH may be a primary or secondary CH. Let RTik be the recommendation trust value of CHi recommended by the CHk, where k = 1, 2, 3, . . . Both DT and RT are determined based on Equation 13. Based on this logic, Table 1 provides a proposed algorithm for trusted cluster-based secure communication.

4. Simulation results The AquaSim tool of NS2 version 2.30 is used to simulate the proposed TMCHV algorithm. To simulate the proposed scheme in UWSN, UnderwaterChannel and UnderwaterPropagation are used, while UnderwaterMAC is used as a media access control (MAC) layer protocol. In this simulation, 50 underwater sensor nodes are deployed randomly and remain static. The packet sending rate is varied from 50 Kb to 250 Kb. The simulation parameters and settings are summarised in Table 2.

4.1. Performance metrics The CSLT model (Han et al., 2016) is based on packet loss and packet error, but it does not consider data collection and energy trust. Since the communication trust of proposed TMCHV is also based on these parameters, the performance of TMCHV is compared with that of CSLT. The performance metrics considered are: • Average packet delivery ratio: the ratio of data packets received successfully to the total number of packets transmitted. It reflects the efficiency and reliability of the network. Table 2: Simulation parameters used Parameter

Value

Number of nodes in the network Test Area Simulation time Antenna type Channel capacity Trafﬁc source Range Packet size Initial energy Transmission power Receiving power Trafﬁc rate Number of attackers

50 500 m2 x 500 m2 50 s Omni antenna 2 Mbps CBR 100 m 50 to 250 bytes 1000 J 2.0 W 0.75 W 50 Kbps 1 to 5

Table 1: Proposed algorithm for trusted cluster-based secure communication Step 1: Step 2: Step 3: Step 4: Step 5: Step 6: Step 7: Step 8: Step 9: Step 10: Step 11: Step 12: Step 13: Step 14: Step 15: Step 16: Step 17:

For each CHj, j = 1, 2, 3, . . . If (CHj receives data from CHi at time tl) then CHj estimates DTij based on T = Tc + Tdc + Te If (DTij < DTth) then CHj collects RTik from CHk, k ≠ j If (RTik < RTth) then CHi s considered as malicious CHj invokes BCHi of CHi BCHi informs its cluster members about cluster change Cluster members retransmit the data to BCHi End If End If End If Similarly If (S receives data from CHi at time t2) then Repeat steps 3 to 13 in respect of S End If End For

111

03-SUT286_34(3).indd 111

28/07/17 7:17 pm

Goyal et al. Trust model for cluster head validation in underwater wireless sensor networks

4.2. Results and analysis For the simulation, the malicious nodes or numbers of attackers are taken and randomly varied from 1 to 5, along with the performance metrics individually. The results of the proposed TMCHV and existing CSLT technique are represented graphically as shown in Fig 3. It shows the detection accuracy for both the techniques. During simulation, CSLT and TMCHV techniques calculate the trust value of each node based on two and three parameters, respectively. As TMCHV calculates the trust value on more parameters, the probability of selecting the true nodes increases, thus achieving the level of 9 % higher detection accuracy in comparison to CSLT. Further, the graphical representation shows that with increasing number of attacking nodes in a medium, the detection accuracy of the proposed scheme increases. Fig 4 shows the packet delivery ratio for both the techniques. The delivery ratio of TMCHV is 49 % higher in comparison to CSLT because, in the proposed scheme, each data packet is accepted based on the sum of three parameters. Thus the probability

of a data packet having higher trust value than the prescribed threshold value increases. The total trust value is the sum of trust value based on two parameters, and so there may be the chance of rejecting the data packets that may have been accepted if the trust value was calculated on more than two parameters. Fig 4 shows that the delivery ratio of the proposed scheme is higher than the existing technique although with an increase in the number of attacking nodes it goes slightly downwards. As described earlier, TMCHV accepts the data based on the trust value of three parameters. Thus there are chances that packet drop decreases during the simulation, which is also shown through graphical representation. Fig 5 shows the packet drop for both the techniques and illustrates that, when the number of attackers increases the packet drop ratio will also increase. However, TMCHV has 9.8 % fewer packet drops than CSLT. 0.08 0.07

Pocket delivery ratio

• Average energy consumption: the amount of energy consumed by the nodes for the data transmission. It is expressed as the average energy consumption of all the nodes in the networks during the simulation. • Packet drop: the number of packets dropped due to attacks. • Detection accuracy: the ratio of number of attacks detected correctly to the total number of attack attempts. • Network lifetime: the time until the first sensor node or any member of the cluster completely runs out of energy.

0.06 0.05 0.04 0.03 0.02

CSLT

0.01

TMCHV

0 1

Number of attackers

Fig 4: Packet delivery ratio with varying number of attacker nodes

0.94 26

0.9 0.88

CSLT

0.86

TMCHV

0.84 0.82 0.8

Number of packets dropped

Detection accuracy

0.92

25 24 23 22 21

CSLT 20

TMCHV

0.78 1

Number of attackers

19 1

Number of attackers

Fig 3: Detection accuracy with increasing number of attackers

Fig 5: Packet drop versus increasing number of attackers

112

03-SUT286_34(3).indd 112

28/07/17 7:17 pm

Energy consumed (J)

Underwater Technology Vol. 34, No. 3, 2017

400

5. Conclusion

350

It is a well-known fact that data transmission plays an important role in the efficient operations of the network. However, some of the nodes may send false data due to some adverse reasons, thereby weakening the trust level in the network. In this paper, a trust-based security model is designed to build up or enhance the trust among the nodes in an UWSN. This paper defines the total trust of a node or CH as the sum total of DT and RT values. Further, each out of DT and RT are the total of the communication trust, data collection trust and energy trust. The proposed scheme is so designed that if a trust value of a node falls below a threshold value, the CH will be replaced by another CH having higher trust value. In this way, the trust of the whole network increases and lessens attacks in the network, thereby increasing the efficiency of the network in terms of packet detection accuracy, delivery ratio, packet drop, average energy consumption and network lifetime. Through a simulation technique, the performance of the proposed scheme TMCHV is analysed and compared with the existing CSLT technique. The graphical representation of analysis shows the outstanding performance of TMCHV over CSLT on the basis of parameters outlined in this paper.

300 250 200 150

CSLT

100

TMCHV

50 0 1

Number of attackers

Fig 6: Energy consumption along with increasing number of attacker nodes

Network lifetime(s)

CSLT 10

TMCHV

0 1

Number of attackers

Fig 7: Network lifetime with increasing number of attackers

Fig 6 shows that less energy is consumed by the proposed scheme in comparison to existing techniques. The energy consumption decreases when the number of attackers increases. When comparing the performance of the two algorithms, TMCHV has 49.38 % less energy consumption than CSLT. Fig 7 shows the average network lifetime of TMCHV and CSLT. When comparing the performance of the two algorithms, TMCHV has 20 % higher lifetime than CSLT. However, when there is an increase in number of attackers, the network lifetime decreases linearly but still remains higher than when using the existing technique. The proposed model is also considered for a larger area where the number of nodes is 200, the test area is 2000 m2 × 2000 m2 and the transmission range is increased. After the simulation, it is concluded from the results that the proposed solution is proven effective. There is no cumbersome effect on the performance.

References Albagory Y and Said O. (2015). Performance enhancement of high-altitude platforms wireless sensor networks using concentric circular arrays. AEU-International Journal of Electronics and Communications 69: 382–388. DOI: http:// dx.doi.org/10.1016/j.aeue.2014.10.013. Bahl N, Sharma AK and Verma HK. (2015). Asymptotic multicast throughput analysis and energy efficiency in WSN under double Nakagami fading channel using extreme value theory. AEU- International Journal of Electronics and Communications 69: 1346–1353. DOI: http:// dx.doi.org/10.1016/j.aeue.2015.05.021. Das AP and Thampi SM. (2015). Secure communication in mobile underwater wireless sensor networks. In: Proceedings of the IEEE 4th International Conference on Advances in Computing Communications and Informatics (ICACCI), Kerala, India, 2164–2173. DOI: http://dx.doi.org/10.1109/ ICACCI.2015.7275937. Dhurandher SK, Khairwal S, Obaidat MS and Misra S. (2009). Efficient data acquisition in underwater wireless sensor Ad Hoc networks. IEEE Wireless Communications 16: 70–78. DOI: http://dx.doi.org/10.1109/MWC.2009. 5361181. Geetha V and Chandrasekaran KA. (2014). Distributed trust based secure communication framework for wireless sensor network. Wireless Sensor Network 6: 173–183. DOI:http://dx.doi.org/10.4236/wsn.2014.69017. Goyal N, Dave M and Verma AK. (2014). Fuzzy based clustering and aggregation technique for under water wireless sensor networks. In: Proceedings of the IEEE International

113

03-SUT286_34(3).indd 113

28/07/17 7:17 pm

Goyal et al. Trust model for cluster head validation in underwater wireless sensor networks

Conference on Electronics and Communication System (ICECS), Coimbatore, India, 1–5. DOI: http://dx.doi. org/10.1109/ECS.2014.6892804. Goyal N, Dave M and Verma AK. (2016). Energy efficient architecture for intra and inter cluster communication for underwater wireless sensor networks. Wireless Personal Communications 89: 687–707. DOI: http://dx.doi.org/ 10.1007/s11277-016-3302-0. Han G, Liu L, Jiang J, Shu L and Rodrigues JJ. (2016). A collaborative secure localization algorithm based on trust model in underwater wireless sensor networks. Sensors 16: 229. DOI: http://dx.doi.org/10.3390/s16020229. Jadidoleslamy H, Aref MR and Bahramgiria H. (2016). A fuzzy fully distributed trust management system in wireless sensor networks. AEU International Journal of Electronics and Communications 70: 40–49. DOI: http://dx.doi.org/ 10.1016/j.aeue.2015.09.017. Jahanshahi JA, Ghorashi SA and Eslami M. (2013). Fuzzy c-means clustering-based jamming detection algorithm at base station. Arabian Journal of Science and Engineering 38: 2125–2133. DOI: http://dx.doi.org/10.1007/s13369013-0578-1. Jia J and Meng J. (2016). Impulsive noise rejection for ZigBee communication systems using error- balanced wavelet filtering. AEU International Journal of Electronics and Communications 70: 558–567. DOI: http://dx.doi. org/10.1016/j.aeue.2016.01.004. Kaur J, Gill S and Dhaliwal BS. (2016). Secure trust based key management routing framework for wireless sensor networks. Journal of Engineering 2016: 1–9. DOI: http:// dx.doi.org/ 10.1155/2016/2089714. Kumar M and Goyal N. (2014). Reviewing underwater acoustic wireless sensing networks. International Journal of Computer Science and Technology 5: 95–98. Liu Y, Liu A and He S. (2015). A novel joint logging and migrating traceback scheme for achieving low storage requirement and long lifetime in WSNs. AEU International Journal of Electronics and Communications 69: 1464–1482. DOI: http://dx.doi.org/10.1016/j.aeue.2015.06.016. Ren Y, Oleshchuk VA and Li FY. (2013). Optimized secure and reliable distributed data storage scheme and performance evaluation in unattended WSNs. Computer Communications 36: 1067–1077. DOI: http://dx.doi.org/10.1016/ j.comcom.2012.08.001.

Rezvani M, Ignjatovic A, Bertino E and Jha S. (2015). Secure data aggregation technique for wireless sensor networks in the presence of collusion attacks. IEEE Transactions on Dependable and Secure Computing 12: 98–110. DOI: http:// dx.doi.org/10.1109/TDSC.2014.2316816. Sabet M and Naji HR. (2015). A decentralized energy efficient hierarchical cluster-based routing algorithm for wireless sensor networks. AEU International Journal of Electronics and Communications 69: 790–799. DOI: http:// dx.doi.org/10.1016/j.aeue.2015.01.002. Senel F, Akkaya K, Erol-Kantarci M and Yilmaz T. (2015). Self-deployment of mobile underwater acoustic sensor networks for maximized coverage and guaranteed connectivity. Ad Hoc Networks 34: 170–183. DOI: http://dx. doi.org/10.1016/j.adhoc.2014.09.013. Thuc KX and Insoo K. (2011). A collaborative event detection scheme using fuzzy logic in clustered wireless sensor networks. AEU International Journal of Electronics and Communications 65: 485–488. DOI: http://dx.doi.org/10.1016/ j.aeue.2010.05.002. Vennila C and Madhura M. (2016). An energy-efficient attack resistant trust model for underwater wireless sensor networks. Middle-East Journal of Scientific Research 24: 33–39. DOI: http://dx.doi.org/10.5829/idosi.mejsr.2016.24. S2.109. Xu M, Liu G and Guan J. (2015). Towards a secure medium access control protocol for cluster-based underwater wireless sensor networks. International Journal of Distributed Sensor Networks 11: 1–11. DOI: http://dx.doi. org/10.1155/2015/325474. Yavuz AA and Ning P. (2009). Hash-based sequential aggregate and forward secure signature for unattended wireless sensor networks. In: Proceedings of the IEEE 6th Annual International Conference on Mobile and Ubiquitous Systems: Networking and Services, Toronto, Canada. 1–10. DOI: http://ieeexplore.ieee.org/document/5326402/. Yavuz AA and Ning P. (2012). Self-sustaining efficient and forward-secure cryptographic constructions for unattended wireless sensor networks. Ad Hoc Networks 10: 1204–1220. DOI: http://dx.doi.org/10.1016/j.adhoc.2012.03.006. Zenia NZ, Aseeri M, Ahmed MR, Chowdhury ZI and Kaiser MS. (2016). Energy-efficiency and reliability in mac and routing protocols for underwater wireless sensor network: a survey. Journal of Network and Computer Applications 71: 72–85.

114

03-SUT286_34(3).indd 114

28/07/17 7:17 pm

www.sut.org

Technical Paper

doi:10.3723/ut.34.115 Underwater Technology, Vol. 34, No. 3, pp. 115–127, 2017

Handling free gas in deep and ultra-deep water drilling risers: a technical review and safety case explanation Paul A Potter* SUBC Engineering Ltd., Aberdeen AB12 3AX Received June 2016; Accepted May 2017

Abstract The Macondo well blowout, also referred to as the Deepwater Horizon Disaster, changed the safety case towards the capable and safe handling of the free gas present in ultra-deep water riser strings. This paper is a technical review that discusses the fundamental physical phenomena of free gas contained in drilling ﬂuids, its behaviour under the conditions present in ultra-deep water drilling risers, and its migration to the surface. The technical advances of post-Macondo riser gas handling (RGH) systems are presented and the safety criticality of such systems as an integral component of the safety case for deep water drilling installations is also explained. Finally, a parallel architecture aspect of a RGH system when installed, and operated in conjunction with a closed managed pressure drilling system is described. Keywords: Macondo blowout, deep water drilling, drilling riser, safe gas handling, degassing, process ﬂowpaths, evolving systems

1. Introduction In the previous decade and a half, the offshore drilling industry has ventured into increasingly deep and ultra-deep waters in search of recoverable hydrocarbons. Following classical physical gas laws, contractors and operators alike have identified a growing risk from gas cut drilling fluids in the well and marine riser annulus above the drilling blowout preventer (BOP) stack. The precise mechanisms of different gas migration characteristics are still not fully understood but will be dependent upon a number of factors such as, drilling fluid composition and gas type. However, the inverse pressure to volume relationship fundamentally results * Contact author. Email address: Paul@subceng.com

in the exponential increase in the gas volume as pressure decreases towards the surface. The Macondo well blowout (Sutherland et al., 2016), also referred to as the Deepwater Horizon Disaster, was preceded two years previously by another event which resulted in an uncontrolled unloading of the marine drilling riser driven by well gases. The free gas compositions in the two cases differed in that the Macondo gases were predominantly methane-based while gases in the prior event were largely inert, being mainly carbon dioxide. During the violent unloading of the marine drilling riser throughout the prior event, the identification of the gas composition was not considered as ‘life-threatening’. The best recorded first inferences that a RGH system would be advantageous as an installed supplementary safety system came in 2002 with the publication of the Deepwater Well Control Guidelines by the International Association of Drilling Contractors (IADC, 2002). Although the guidelines made a number of pivotal statements, the traditional subsea system remained short of the minimum requirements for a safe gas handling system.

2. Gas management without any form of gas handler Initially, a subsea well is ‘top hole’ drilled without any mechanical link, in the form of a marine drilling riser between the subsea wellhead and the drilling installation. Shallow geo-pressured formations may be encountered whilst drilling top hole and the resulting release of shallow gas may constitute a hazardous threat to the drilling installation (IADC, 2002). More than 20 years ago, for floating anchored

115

04-SUT252_34(3).indd 115

28/07/17 7:17 pm

Paul A Potter. Handling free gas in deep and ultra-deep water drilling risers: a technical review and safety case explanation

drilling installations and drilling in depths usually shallower than 150 m, a so-called pin connector was installed in order to bring drilling fluid returns back to the rig ahead of installing a drilling BOP stack (American Petroleum Institute (API), 2001). For anchored rigs drilling top hole without a riser installed in the water column, shallow gas posed two distinct threats: ignition should the composition of escaping surface gas be flammable; and reduced floating drilling installation stability caused by the decreased buoyancy factor of the seawater directly above the well. Today, deepwater drilling operations with platforms stationed on location by dynamic-positioning azimuthing thrusters, traditionally use a total loss drilling fluid system known as the ‘pump and dump’ technique. One member of the rig crew is positioned in the moonpool area to alert the driller of any visual sightings of gas bubbles appearing. One dominant drilling contractor has a standing order procedure to intentionally ‘drop’ the drill string, enabling the rig to move off location using its dynamic-positioning capabilities without the drill pipe string in the event of a shallow gas release event. Once the drilling BOP stack is installed on the subsea wellhead, the mechanical link with the well is established with a marine drilling riser string extending from the top of the drilling BOP stack to the rig floor rotary kelly bushings (RKB) on the drilling installation. This bridging mechanical link remains in place for the remainder of the subsea well drilling programme. For a conventional single gradient open drilling system, the only possibility of handling gas that has migrated above the subsea drilling BOP stack after a hard well shut-in has been performed is to use the rig’s diverter and an annular preventer on the drilling BOP stack (Fig 1). In the scenario illustrated in Fig 1, the riser has been shut in by the upper annular on the subsea drilling BOP stack and the standard diverter at the rig floor elevation. The shut-in of the diverter packer is auto-sequenced with the outlet isolation valves (API, 2001). In this situation, the diverter packer is closed/energised with the flowline to the shale shakers shut and the starboard overboard line open to ‘divert overboard’. This is not a pressure-containing boundary and it should be noted that the 356 mm (14 in) nominal diameter through bore isolation valves on the diverter outlets are full bore open/ closed with no choking intermediate positions. There are many shortfalls to this system arrangement when considering riser gas handling. Most importantly, handling, in respect of any form of control, is not possible and there are also limitations in terms of releasing trapped gas from the drilling riser. The diverter system is typically rated

at 500 psi (ca. 34.5 bar) with only the latest build diverter systems rated at 2000 psi (ca. 137.9 bar). Original equipment manufacturers stress that diverter systems are not designed to contain pressure but divert it to a safe area. With free gas in the riser annulus following a well shut-in with the subsea drilling BOP stack, the final pressures within the top portion of the riser string annulus can be significant. The final pressures will depend on: the volume of gas at the bottom of the riser; the gas composition; the degree of gas expansion; the gas bubble point depth below sea level; the water depth and height of the riser annulus column; and the liquid composition of the drilling fluid within the riser annulus (oil or water based drilling fluids). Irrespective of gas pressures, weak links in the riser system include the lower and upper flex joints, the telescopic slip joint packers and the diverter packing element (Fig 1). The weakest link overall is the telescopic slip joint packers which, at best, can tolerate no more than 200 psi (13.8 bar) internal pressure before blowing out to the atmospheric conditions within the moonpool. Of note in Fig 1 is the mechanical connection, with isolation valve, to the rig’s mud gas separator (MGS). This was the arrangement on the Deep Water Horizon: once the rig floor had activated the diverter packer during the loss of well control, the wellbore fluids driven by flammable gas were routed to the rig’s MGS, which was almost immediately overwhelmed and ignition occurred shortly afterward. This connection is now forbidden. The rig’s diverter is located at an elevation above sea level (known as the air gap) of approximately 15 m (Fig 1). This means that the migrating free gas in the riser will expand to its full extent prior to being diverted overboard. Therefore, the advantage of bridging the flowpath below the sea level is not possible with only the standard diverter assembly. Taken overall, potential pressure peaks experienced in the marine drilling riser annulus with free gas migrating upwards, may be in excess of the maximum design load limits of the riser couplings used in the string. Mud boost lines are fitted to all deep water marine risers to assist in the circulation of returning fluids to the drilling unit. In conventional riser drilling, they are also used to introduce fresh base drilling fluid into the bottom of a shut-in riser annulus in order to aid in circulating out free gas in the riser (Fig 1). However, the controlled circulation can only be achieved if some form of choke valve is present at the top of the system. In the scenario shown in Fig 1 where only the standard rig diverter is available, controlled circulation of riser contents is not possible since the design basis for diverter systems does not include a choking valve/device.

116

04-SUT252_34(3).indd 116

28/07/17 7:17 pm

Underwater Technology Vol. 34, No. 3, 2017

Riser fill up line (from HP standpipe manifold)

Pressure rated at 500 psi

Rig floor Port overboard

Diverter assembly To shale shakers

LP flowline knife valve

To mud gas separator

Upper flex joint

Air gap mean sea level to rotary Kelly bushings

Starboard overboard Telescopic slip joint inner barrel Master overboard discharge non-adjustable isolation valve 14in. ISO 150 Line

Tension/gooseneck ring

Surface

TJ dual packing box (200 – 500 psi)

Mean sea level

Telescopic slip joint outer barrel

Mud boost line

Water column

8-10K feet (height of interest) Marine drilling riser

Mud boost valve

Riser adapter SSBOP

Lower flex joint

Mudline Upper annular SSBOP

Fig 1: Marine drilling riser system designed for conventional riser drilling; the well is shut-in with suspected free gas in riser; the diverter system has been activated to release increasing riser annulus pressure.

If a late decision is made to shut in the riser and divert the pressure, the increasing internal riser pressure may be so great through the diverter system overboard lines that the aggressive mass flow stream (cuttings laden) may ‘wash through’ outer radii on long angle sweeps within a relatively short time period. For floating drilling units without any form of gas handler, reverse circulation has been proposed as a remedial contingency. This involves pumping fresh base mud into the top of the riser annulus via the diverter system through a riser fill-up line, which displaces the gas down the annulus and up the choke or the choke and kill lines back to the rig’s choke and kill manifold. This proposed

contingency has largely been dismissed as too high risk for the following reasons: • Reverse circulation mass flow rate, via the riser fill-up line, does not guarantee that some gas fraction could ‘channel’ past the downward annulus flow and continue its ascent (and expansion) towards the top of the riser system. • The diverter system control system autosequence must be overridden for full closure of the diverter system which is beyond its fundamental design basis. • The telescopic slip joint packer could be exposed to excessive internal pressure beyond its design limit.

117

04-SUT252_34(3).indd 117

28/07/17 7:17 pm

Paul A Potter. Handling free gas in deep and ultra-deep water drilling risers: a technical review and safety case explanation

• The shut-in drilling BOP stack must have variable or fixed bore ram type preventers closed in around the pipe (assuming it is in the hole), enabling BOP side outlets to be opened to choke and/or choke/kill lines. As such, the previously closed annular, activated initially at the well shut-in, must now be re-opened to establish a reverse flowpath through the BOP wellbore to the choke/kill outlets. If there is no pipe in the hole, the well system will ‘see’ the dynamic reverse circulation since the drilling BOP stack wellbore will not be capable of well shut-in since the shear/blind rams, located uppermost in the stack configuration, will need to be in the open position. In summary, any drilling installation that is rated for deep and ultra-deep water (nominally deeper than 1524 m), but not fitted with some form of gas handler, is at high risk of a safety violation through its inability to cope with a gas in riser event. Management of any such gas volume in the riser annulus will be, at best, hit and miss with little or zero accurate control or real time monitoring of the progressive and dynamic parameters. For these reasons, drilling standards state that a gas handler should be installed and used on a rig drilling in deep water (e.g. Norsok, 2012).

3. Gas handling systems 3.1. System design rationale: minimum required components The central design briefs for a modern gas handler system are twofold. Firstly, the riser system top closure device should be subsea (or more precisely near mean sea level elevation). Secondly once the subsea diverter device is energised closed and diverting the annulus contents, the control of this flowpath should be both easily measured with realtime monitoring (RTM) devices and the residual pressure control with the riser annulus should be practically achieved using a choking device (pressure control valve). It is difficult to make comparisons between the elevations of a standard rig diverter system and a subsea diverter closure device located beneath the telescopic slip joint. The vertical difference in displacement is not more than 30 m. However this small difference is critical in the effective management of gas volumes that are expanding rapidly in an exponential relationship as the pressure drops rapidly towards the surface as related to Boyle’s Ideal Gas Law. 3.2. First generation riser gas handler system In all planned approaches to gas handling, the time needed to shut in a well influx is the dominant factor.

This is driven by the combination of time taken to confirm a genuine influx and the time needed to close the subsea BOP stack (SSBOP) and the gas handler. The first of these time dependent events has been dramatically shortened by the advent of early and deep water kick detection systems. These are often used in conjunction with a managed pressure drilling method although they are equally effective when used in conventional drilling, provided that sufficient high definition sensors are installed throughout the topsides drilling fluid system volume. The effectiveness and overall sensitivity of these relatively new systems is such that well influxes can be realistically detected as early as between 0.5 to 1 barrel equivalent; this compares with influxes of 5 to 8 barrels with a conventional detection system installed. This discrepancy translates directly to time and gas migration. If the well is shut-in as a result of a confirmed influx then there is a positive assumption that there is free gas in the riser until proven otherwise. During the initial well shut-in, the riser gas handler is usually activated simultaneously with the SSBOP upper annular preventer. The riser annulus is now isolated from the top portion of the riser string inventory culminating at the diverter assembly and the housing diverter system outlets (Fig 2). Alternatively, an externally mounted carbon steel gas vent line is opened (by the upward stroking action of the hydraulically operated gas handler packing element operating piston), providing a flowpath route directly to the high pressure section of the rig’s choke and kill manifold (Fig 2). At this point, the riser main annulus beneath the telescopic slip joint is closed off with annular packing element capable of sustaining 1000 psi (68.9 bar) of pressure beneath; the gas handler gas vent line is now open to the rig’s choke and kill manifold; the SSBOP upper annular preventer is closed isolating the riser annulus column throughout the water depth column height; and the mud boost line circulation capability is available (Fig 2). It is good practice to ‘adjust’ the regulated hydraulic closing pressure for all annular type preventers, dependent on the drilling phase of the well (in respect of drilling or running casing), including stack-mounted annular preventers, diverter packers and gas handling packers. Reduced regulated hydraulic closing pressure charts are available for different tubulars used in all drilling well processes. Shutting in the riser annulus below mean sea level reduces the potential and full expansion of gas if it is allowed to reach the elevation of the surface diverter packer. Possible redundancy is available for the gas handler packer in the form of a double unit: one packer box mounted on top of the other.

118

04-SUT252_34(3).indd 118

28/07/17 7:17 pm

Underwater Technology Vol. 34, No. 3, 2017

Gas vent

Rig mud gas separator

Shakers

Three way flow diverter

Discharge overboard

Aux. line termination ring

Gas handler reel

High-pressure standpipe manifold

Choke and kill manifold

Telescopic slip joint

Mud boost pump

Mean sea level

Gas handler

Marine drilling riser

18.75 m–10 m annular

18.75 m–15 m drilling lower BOP stack

Fig 2: A block process diagram for a ﬁrst-generation gas handling system.

The riser annulus has the capability of being shut-in with a subsea diverter and automatically re-routing the wellbore fluids/gases to the rig’s choke and kill manifold. The top of the closed-in system culminates with dual remotely-operated chokes that can hold back pressure in the shut-in riser annulus and together with a controlled circulation can displace free gas with base drilling fluid. This also controls the bubble point and hence the depth at which gas breaks out of solution although this can be modified given the significantly greater solubility of gas in oil-based compared to water-based drilling fluids. A number of unknowns continue to exist with this system. For instance, the monitoring and interpretation of pressures within the riser annulus, if wholly dependent on choke and kill manifold instrumentation, are not guaranteed to be wholly accurate throughout the entire period of circulating out free gas in the riser annulus. However, the

possibility of a violent unloading event is now impossible with a closed in annulus unless the subsea diverter packer fails while energised. The possibility of a partial pressure-driven evacuation of the top portion of the riser as a major gas influx expands nearing the surface is also highly unlikely with the gas handler subsea diverter energised and the system closed in. Any gas volume downstream of the on-line choke valve on the choke and kill manifold during the gas circulation process will further expand as it passes through the choking device and all downstream elements (free gas, drilling fluids and cuttings) will be directed to the rig’s MGS. There is a quantifiable likelihood that the downstream pressure of the gas passing to the MGS may exceed its design peak gas flow rate. The immediate subsequent risk is that the MGS mud seal leg could be blown through by the excessive pressure in the MGS and hence flammable gas may

119

04-SUT252_34(3).indd 119

28/07/17 7:17 pm

Paul A Potter. Handling free gas in deep and ultra-deep water drilling risers: a technical review and safety case explanation

pass to the rig’s shale shakers with the attendant high risk of topsides ignition. Both the incidents mentioned in section 1 involved the rig’s MGS being overwhelmed with a very low degree of rig control. In addition, there was no gas handler of any description installed or in use on either rig. When there is a gas handling system in place, any increasingly abnormal condition detected in the rig’s MGS can be mitigated simply by shutting in the flowpath at the manifold choke valve. The possible reverse consequence of this abrupt shut-in could result in an over-pressurisation situation in the marine drilling riser immediately below the gas handler joint.

4. Justiﬁcation for a high-capacity gas handling system Today, contractors and operators alike recognise the market advantage of high-capacity gas handling systems and ultra-deep water new-builds are expected to demonstrate such abilities. Rig owners now universally identify with the marketability advantage of specifying in a Schedule C technical description of their new build, a capable highcapacity gas handling system. This is often quoted in terms of equivalent barrels of free gas in the bottom of a deep or ultra-deep water marine riser, following a well influx. As discussed in section 3, some level of applied surface back pressure in marine drilling riser annulus (containing the gas influx) contributed to finite control of the free gas migrating through the riser annulus column. This phenomenon of gas migration control and breakout can only now be allied with managed pressure drilling (MPD) systems, particularly the pressurised mud cap drilling system variant. In developing these new systems capable of handling gas in much higher capacities emphasis has shifted from their original classification as secondary well control systems towards them being safety system enhancements.

4.1. Fundamental requirements of a highcapacity riser gas handling system The underlying design rationale behind a highcapacity RGH system is based on containing pressure, maintaining adequate mass flow rate capabilities, and having the ability to control the pressure. When specifying the design of a new build ultradeep water drilling unit, it is now typical for the client to issue a mandate for two MGSs. One is intended for use during secondary well control circulations while the other is meant to be an in-line degasser. The standard MGS fitted to seventh generation ultra-deep water drilling units is generally rated at 15.5 million

standard cubic feet per day gas flow rate (MMSCFD) with a liquid maximum rate flow rate of around 1550 gallons per minute (gpm). This is the inferred rating for the in-line degasser while the rig’s standard secondary well control MGS should be at least 10 MMSCFD peak gas flow rate capacity. Neither capacities are considered adequate today. Fig 3 is a block process diagram for a dedicated high-capacity gas handling system that illustrates the differences from the first generation gas handling system detailed in Fig 2.

4.2. Sizing considerations for high-capacity liquid/gas separators In terms of both empirical and theoretical computational flow dynamics, internal pipework diameters are critically important. Not only do generated friction losses in degasser gas vent lines have a direct influence on degasser design sizing but they also impact a wide variety of other pipework systems commonly encountered in both drilling and production. Generated friction losses in a number of safety critical pipework systems are pivotal to the responsiveness of today’s secondary well control systems during the circulation of well influxes. The diameter of gas vent lines fitted to degassing vessels found in RGH systems will affect steady-state pressure drops (Fig 4). These influence maximum and minimum peak flow rates which, in turn, dictate design dimensions of components such as the high-capacity separators (MacDougall, 1991). During the detailed design of the rig, the precise effective length of the gas vent line(s) for both vessels must be calculated taking into account any changes in direction resulting from practical routing constraints. The mud outlet should nominally be 30 cm diameter and the length no longer than 30 m; these dimensions should ensure that frictional losses do not need to be calculated for controlled circulation rates but only for uncontrolled maximum liquid surge rates. The positioning and installation of the in-line degasser should be at least the same elevation as the shaker system/gumbo box; this will ensure that the liquid leg hydrostatic pressure calculations are not unduly affected. The elevation of the mud seal leg top should coincide with the nominal working fluid level within the inline degasser vessel, whose lower section is fitted with low level alarm instrumentation and will be a 60 deg conical section. 4.3. Parametric factors in de-gasser sizing This design basis study for the appropriate sizing of the in-line degasser for use in RGH systems for rigs drilling in deep and ultra-deep water may be split

120

04-SUT252_34(3).indd 120

28/07/17 7:17 pm

Underwater Technology Vol. 34, No. 3, 2017

Gas vent line

Overboard

Gas cut multiphase fluids

Gas handler reel

HP standpipe manifold

Topsides valve distribution manifold

Telescopic slip joint

Pressure control valve manifold

Gas handler

Drillng riser

In-line degasser

Three way valve

Mud boost pump

18.75 m–10 m lower marine riser package annular preventer closed

Shakers

18.75 m–15 m drilling LBOP stack

Fig 3: A block process diagram for a high-capacity gas handling system. Several core components of this system differ from a ﬁrst generation RGH (Fig 2). The system components, commencing with the take-off diversions from the riser annulus, through to the disassociation of liquid/gas phases in the In-line degasser are bigger to accommodate the higher liquid and gas phase peak ﬂow rates and the greater working maximum pressures. In the illustrated scenario, the upper annular on the drilling BOP stack is closed as is the gas handler, with the gas vent line discharging through the moonpool hose to topside valving and in-line degasser.

Pressure losses in nominal pipework for identical arbitrary conditions 1100

1100 1000

Pressure losses (psi)

900 800 700

yi 600 500 400 300 200 100 0 4

5 4

12 12

Nominal pipe diameter (in)

Fig 4: Pressure losses in nominal pipework for identical arbitrary conditions based on line length of 45.72 m; mud density of 1.15 speciﬁc gravity; outlet pressure of 1 bar absolute; temperature of 26.7 °C; gas density of 0.7 speciﬁc gravity; and plastic viscosity of 8 centiPoise.

into two categories: static and dynamic (Kozizc, 2012). The main points to be considered relate to: concerns surrounding the peak gas flow rate at the surface; the external geometry of the degasser vessel being sufficient to support optimum separation of liquid and gas phases during the through-flow circulation of a RGH event; the design of the internal arrangement of the degasser vessel in relation to optimising separation rates required; and the diameter of the vessel liquid inlet, gas outlet and liquid outlet to minimise retention rates. In-line degassers built for high-capacity throughput and with significant RGH, need to be structurally supported to handle the worst case operational conditions. These are: the vessel is liquid-filled; there is significant drilling installation movement caused by limiting ocean conditions; and there are high force winds. Two current options exist for the installation of such equipment on floating drilling installations: the

121

04-SUT252_34(3).indd 121

28/07/17 7:17 pm

Paul A Potter. Handling free gas in deep and ultra-deep water drilling risers: a technical review and safety case explanation

equipment is installed topsides and integrated into existing infrastructure as a ‘retrofit’; or it is installed during a shipyard new build programme. The engineering challenges for a retrofit are influenced by the size of the vessel and its supporting framework and for it having to penetrate the various levels of the drilling rig structure. The impact of the added mass also affects the drilling unit metacentre height and centre of gravity. These factors can be better accounted for when designing a new build.

5. Implications of the riser gas handling/ managed pressure drilling interfaces The manufacturers of RGH systems frequently state that their system is equally effective to recover from a gas-in-riser event, whether the drilling installation is in a conventional riser drilling mode or some variant of an MPD mode. Both the first generation riser and the high-capacity gas handling systems (Figs 2 and 3, respectively) require some form of diverter outlet from beneath the closed subsea diverter once the RGH system has been activated. The requirement for a diverter system is unavoidable as all RGH systems feature a closure device that is installed near the surface and beneath the rig’s telescopic slip joint. The stated equal effectiveness is correct, if and only if, a subsea diverter has been installed beneath the telescopic slip joint and below that there is some form of diverter spool piece to route the riser contents from the riser annulus to the in-line degasser. If a drilling unit installs the subsea equipment, integrated into the marine riser system at the beginning of a subsea well (in coincident with deployment of the subsea drilling BOP stack), it does not necessarily follow that the drilling programme will dictate that all well sections will be drilled in an MPD mode. It is a simple re-configuration matter to switch between a mud cap MPD mode of drilling and the conventional open system of subsea drilling.

5.1. Minimum requirements for high-capacity ultra-deep water riser gas handling systems Minimum hardware requirements to establish a working high-capacity RGH system fall into two separate categories of identification: surface and subsea. The surface requirements consist of: a mud boost pipework flowpath (tie in); a flow distribution manifold; a pressure control manifold; a highcapacity in-line degasser; a cuttings agitation pump; a subsea control and data acquisition (SCADA) container; a hydraulic power supply; and interconnecting reel supplies to subsea installed equipment.

Subsea requirements are: a subsea diverter closure device; a flowspool joint assembly with diverter outlets; flexible high-capacity diverter hoses for riser to topsides connections; connection capability subsea for telemetry and power; and riser crossover joints. The availability of the riser gas handling system depends wholly on the physical deployment of the subsea hardware and it’s hook-up with the surface equipment. However, its installation is required irrespective of whether there is an intention to drill with an MPD method or remain in the open conventional mode of drilling. The minimum subsea equipment insertion into the existing marine drilling riser string demands a number of considerations. These are: • Scopes of supply from vendors for this system are required to ‘adapt’ the equipment to align it to the specific rig. Essentially, this involves the manufacture of riser crossover joints for installation at the top and bottom limits of the RGH equipment stack-up. Rarely are two marine drilling riser strings identical and such uniqueness in the radial configuration of the auxiliary lines (e.g. choke, kill, mud boost, hydraulic conduits) must be geometrically accounted for in the external passage of these lines longitudinally for the length of the RGH equipment stack. • The additional effects of the RGH hardware in all the mandated conditions/scenarios studied in riser analyses need to be considered, including: (i) internal pressure limits in accordance with riser gas handling procedures; (ii) end loading on riser couplings resulting from internal pressure loading; (iii) increased worst case emergency disconnect events with respect to the rig’s marine riser tensioner system (riser recoil); (iv) the moonpool hose population torque build-up during a planned rig rotation (monohull drillship) using it’s dynamic positioning system. • Subsea components for RGH must pass, in an unobstructed manner, through the rig’s diverter housing. This is because the diverter assembly is not installed in the housing until the SSBOP stack has been deployed subsea on its marine drilling riser, and has landed and latched on the subsea wellhead. • Safe working loads (SWL) are not exceeded either on the rig floor itself or equipment in use off the rig floor (e.g. rig cranes, handling yokes, conveyors); • There is a reliable hydraulic supply and control system.

122

04-SUT252_34(3).indd 122

28/07/17 7:17 pm

Underwater Technology Vol. 34, No. 3, 2017

5.2. Speciﬁcs on RGH subsea and surface equipment components Many of the equipment components are dual role in that they are an integral part of an impressed annulus pressure managed pressure drilling system as shown in Fig 5. Subsea diverter. A number of subsea diverter closure devices exist, primarily targeted at RGH systems. Their functionality and dynamic capabilities do not differ dramatically for rotating control devices (RCD) used in MPD systems. Dynamic capabilities refer to both stripping and rotating the drill pipe in the hole: a normal condition for a well shut-in and the following assumption that free gas is present in the bottom of an extended marine drilling riser annulus in deep and ultra-deep water. As with RCDs, these closure devices require surface control with a hydraulic power supply to drive the packing elements. Flowspool assembly. The flowspool joint enables the riser annulus contents to be ‘diverted’ to rig topsides pipework and manifolds with an RCD or RGH diverter packer energised. Flowspool assemblies are fitted with diverter goosenecks, similarly engineered to the style adopted for the flexible hose/hard gooseneck interfaces for choke, kill, mud boost and hydraulic conduits on conventional termination rings or hands-free gooseneck rings. The main differences between the flowspool

Mud boost flow for circulating out gas in riser

Diverter

diverter outlets and the standard goosenecks employed at the top of the marine drilling riser system are threefold: • the diverter outlets are nominally 152.4 mm (6 in) diameter in comparison to 63.5 mm – 127 mm (2.5 in -– 5 in) in conventional drilling; • the pressure rating of the RGH diverter outlets (including moonpool hoses) is around 3000 psi (307 bar) compared to the choke/kill which may be either 10 000 psi, 15 000 psi or 20 000 psi (690 bar, 1034 bar or 1379 bar, respectively); • following the two mechanical barrier provisions mandated post-Macondo, the diverter outlets are fitted with tandem remotely and hydraulicallyoperated gate valves. For choke and kill lines, the two-barrier provision exists in the form of SSBOPmounted choke and kill valves. Diverter flow hoses. At a nominal length of 60 m – 75 m (200ft – 250 ft), these 267 mm (10.5 in) outside diameter (OD), 140 mm (5.5 in) inside diameter (ID) diverter hoses accommodate any vertical displacements of the drilling unit caused by rig heave. The topside termination is via a 2/4 bolt clamp/hub arrangement to the rigid carbon steel topsides pipework. The flowspool end is fitted with a dynamic swivel to offset the adverse torque forces that build during weather vaning of monohull drill ships. Installation and disconnection of these hoses

Assigned highpressure mud pump as booster pump

Gas vent line

Boost line tie-in and distribution Flowmeter

Telescopic slip joint

RCD

Alternative flow PMCD mud boost flow

Distribution manifold

Coriolis upstream of pressure control valves In-line mud gas separator

Shakers

Flowspool

Riser Pressure control

Overboard

BOP

Fig 5: A block process diagram for a high-capacity RGH system. Equipment ‘blocks’ shaded light grey represent those already present on all ultra-deep water drilling units. Green or dark grey shaded equipment blocks are those that are required for retroﬁt if the rig is in service.

123

04-SUT252_34(3).indd 123

28/07/17 7:17 pm

Paul A Potter. Handling free gas in deep and ultra-deep water drilling risers: a technical review and safety case explanation

constitutes a major procedural step in deployment/ retrieval of the subsea drilling system. Flow distribution manifold. The diverter outlets on the flow spool are connected to this manifold via their respective moonpool hoses and a portion of suitably rated hard rigid pipework on the topsides of the drilling installation. The buffer manifold serves as a distribution manifold between the diverter outlets and the pressure control manifold in the MPD mode of drilling. This switches its modus operandi in the RGH mode, upstream of the ultra high-capacity mud gas separator (UMGS). Incorporated in the manifold capabilities are two high over-pressurization routings utilising pressure safety valves (PSVs). One of three flowspool hoses for over-pressurisation relief is sourced from the dedicated over-pressure diverter line from the flowspool. The other is sourced from the high-pressure booster pumps and manifold for mud cap drilling (MCD). Consequently there are two PSVs; all pipework is nominally 152 mm (6 in) diameter. At minimum for required redundancy reasons, two inlets from a port/starboard flowspool diverter outlet must input to the flow distribution manifold via respective moonpool catenary hoses and associated rigid topsides pipework. Booster tie-in manifold. A gas handling system must feature a circulatory system that enables drilling fluid volume to be replenished in the riser annulus as the free gas is circulated out. This is achieved by using a separate mud booster valve and line tie-in and utilising the rig’s existing mud boost system. This admits pressurised drilling fluids at the mud boost valve at the elevation of the subsea BOP stack riser adapter. A dedicated high-pressure mud pump is generally assigned for continuous duty. It provides the required mass flow rate of fresh base drilling fluids to the bottom of the extended marine drilling riser string, both for stimulated circulation of well fluids in the marine drilling riser annulus in the MPD mode of drilling and to ‘circulate out’ the free gas in the isolated riser annulus following a well shut-in in deep and ultradeep water. The tie-in is configured to offer either an appropriate pumped mass flow rate down the riser mounted auxiliary mud boost line (entering the marine riser annulus at the top of the SSBOP) or else directed to the flow distribution manifold to set up marine annulus reverse circulation for pressurised MCD, the details of which are beyond the scope of this paper. The pressure control valve manifold. This is the final manifold in the RGH general arrangement upstream of the in-line degasser (Fig 5). This manifold may be considered the ‘core’ of the RGH system in that it is within this manifold that finite

control of the conditions within the riser annulus containing migrating gas in the RGH operation are achieved. The manifold contains a minimum of two identical choking devices, both of which can be isolated by positively pressurising valves on the upstream and downstream side of the chokes. Throughput nominal diameter, according to peak liquid and gas rates, is (as for MPD) 152 mm (6 in), although smaller bores may be considered a designredundant shortfall. Some of today’s manifolds feature a Coriolis flow meter downstream of the choking devices, while others install the flowmeter upstream. The underlying rationale for this upstream selected installation location for the flowmeter is that if gas flashes off through a choke valve, a Coriolis flowmeter may likely misread the mass flow rate and become ‘confused’ with the multiphase product passing through it. Hydraulic supply. The hydraulic power unit (HPU) provides power to activate valve actuators, packing elements in annulus closure devices and interlock mechanisms featured throughout the general assembly of the MPD/RGH system. Initial design philosophy proposed utilisation of the rig’s SSBOP control system to provide the hydraulic requirements of the system. However, that system was vetoed on the grounds that the SSBOP control system is a safety critical system and, as such, cannot be subject to extended end-use whatever the functionality may be. HPUs vary system to system; open and closed systems have been used with different hydraulic mediums, ranging from 10 weight (10W) mineral oil in a closed system to a water-based control fluid in an exhaust-to-sea system. Gas flow rate downstream: in-line degasser. The amount of gas being vented through a high-capacity in-line degasser is required to be measured, both for RGH operations and drilling ahead with the MPD system. For this purpose, an optical flow meter (OFM) is installed in the lower section of the in-line degasser gas vent line. Laser beams are used to measure the gas flow rate by sensing the velocity of microscopic particles that co-exist naturally in gas. Flow rate measurements need to be calibrated to the specific diameter of the gas vent line and a pressure/temperature probe is required to compensate for ambient conditions. Data corrections are applied from the gas probe to improve accuracy. Hence, a second probe, adjacent to the gas sensing probe and which does not impact linear gas flow, is installed in the gas vent line. SCADA and telemetry. The SCADA and the attendant telemetry devices are integrated into the MPD system. All aspects of data collection, system monitoring and automatic control, are collectively illustrated on human/machine interfaces, which

124

04-SUT252_34(3).indd 124

28/07/17 7:17 pm

Underwater Technology Vol. 34, No. 3, 2017

are commonly termed the ‘Cyberbase’ on a modern ultra-deep water drilling unit. Programmable logic controllers feature prominently in the architecture, with most of the inter-process communications being achieved by fibre-optic cable interconnections. A significant proportion of the MPD equipment controlling microprocessor boards feeds the RGH remote control capability of the system. All circuitry is designed for 100 % redundancy to boost reliability and reduce failures to as low as reasonably practicable (ALARP). Although RGH systems are safety systems, today they are considered safetycritical systems.

6. RGH system rig retroﬁt philosophies The IADC have categorised RGH (and MPD) systems into two discrete categories: reactive and proactive. Reactive systems are those that may be utilised intermittently during the course of a drilling programme. This underlying philosophy has been practiced by rig owners and operators alike for a number of years. However, today’s trend is for proactive systems that advocate the continuous or near-continuous use of an adaptive new drilling method. This philosophy can be carried on to the installation and ready-capability of both RGH systems and early/deep water kick detection systems. Today, drilling contractors are more likely to include a high-capacity RGH system in a new build design than to consider retrofitting. This modern trend has been influenced by recent incidents, specifically the Macondo well incident (Sutherland et al., 2016). These events have highlighted the extreme potential dangers associated with extended lengths of marine drilling riser and the relatively high likelihood of free gas ingressive to the riser annulus while drilling in deep and ultra-deep water. Regarding the current trend of RGH installation on ultra-deep water drilling units, unlike prior trends the prospect of removing such equipment (either between wells, drilling campaigns, contracts or rig transits) is now not considered a viable option. In other words, this relatively new and innovative system is now considered to be standard fit for all ultra-deep water drilling installations.

7. Evolving gas handling methods for high-capacity riser gas handling In all cases of confirmed wellbore influxes, the assumption is that there is a presence of free gas in the base of the extended marine drilling riser string. The driller, as part of their standing instruction upon confirmation of a well influx, will shut in the SSBOP using an annular preventer, normally the

uppermost unit in the SSBOP stack. Immediately following (or simultaneously with) this action, the RGH system will be activated thereby shutting in the riser annulus. In order to eliminate the risk of hydrate formation, the riser will be de-gassed ahead of circulating out the well influx, using the standard secondary well control circulation methods and equipment. Both the well and the riser will be continuously monitored for changing conditions (pressure build-up) and a change of procedural priority may be instigated at any time. It is not considered feasible that the well influx can be circulated out of the well concurrent to circulating out the free gas in the riser annulus. The suspected size of the barrel equivalent ingress of free gas in the bottom of the extended riser string in ultra-deep water will influence which of two, approaches are used: the constant surface back pressure method (CSBP), or the constant riser booster pump pressure method (CRBPP).

7.1. Constant surface back pressure method A high level procedure should adopt the following general steps: (i) RGH system activated: well shut-in at upper annular on the SSBOP, subsea diverter closure device closed and one diverter outlet of the flowspool assembly open to the topsides RGH system. (ii) Pumping through the mud boost line, into the riser annulus, bring the mud boost pump on-line to the kill rate or slow circulating rate (SCR). (iii) Simultaneously to ramping up the mud booster pump to the desired pump rate adjust the online pressure control valve to around 300 psi applied surface back pressure (ASBP). This is achieved automatically through the RGH software algorithm, providing the RGH mode was selected on the Cyberbase control. (iv) The circulation duration is the entire volume of the enclosed riser annulus with the ASBP held constant. The initial riser booster circulating pressure (IRBCP) observed on the driller’s Cyberbase should be the SCR pressure plus the ASBP (as read at the RGH pressure control manifold). (v) On completion of the riser kill (in accordance with volumetric calculations), the circulation is stopped and a static system check is performed to ensure that the riser annulus is stable with no residue of expanding gas present. A list of variables in the gas-in-riser event can influence this procedure. Most notably, these are: the drilling

125

04-SUT252_34(3).indd 125

28/07/17 7:17 pm

Paul A Potter. Handling free gas in deep and ultra-deep water drilling risers: a technical review and safety case explanation

fluid type (water- or oil-based mud); the current mud density; the ASBP throughout the circulation duration; possible pressure threat to lower flex joints; the maximum allowable surface back pressure; the maximum mass flow rate of the booster pump; the optimal ASBP values for effective circulation without exceeding any design limits; monitoring of peak liquid surge rates; and the break-out points and bubble point for dissolved gas variances caused by the compositions of the gas and drilling fluid.

7.2. Constant riser booster pump pressure method A high level procedure should adopt the following general steps: (i) RGH system activated: well shut-in at upper annular on the SSBOP and RGH activated. (ii) Apply 300 psi ASBP using on-line pressure choke valve. (iii) Start riser boost pump and bring to IRBCP, which is the SCR and the 300 psi ABSP. This is then the initial circulating pressure (ICP). (iv) Switch over control to the automatic system to maintain the IRBCP. The control system will now ‘track’ the cumulative pump strokes versus 1.5x the riser annulus volume and signal the operator once the required volume has been displaced. (v) When the operator is satisfied that 1.5x volume has been displaced, they can check for dead riser by ramping down the riser boost pump rate whilst the on-line choke will compensate for reducing circulation rate and maintain a 300 psi ASBP. (vi) On completion of the riser kill, the circulation is finally stopped and a static system check is performed to ensure that the riser annulus is stable with no residue of expanding gas present. If the riser annulus condition shows an increasing pressure over time duration of the flow check, then the RGH will be re-activated and further circulation will commence through the system and in-line degasser. As with the CSBP method, conditions in the gasin-riser event can influence the CRPBB procedure. With larger influxes in a water based mud (WBM) system this method will be limited in its application. The increase in return flow rate caused by the immiscibility (i.e. insolubility) of gas and gas expansion in a WBM in the riser, increases the flowing back pressure on the system. Drilling fluid will surge through the pressure control valve/surface choke at the riser gas handling manifold, causing the surface pressure to potentially exceed the pressure that it is trying to maintain. As a result of the increased

back pressure for surge rates, the pressure control valve will attempt to open to compensate and maintain the set constant pressure allowing gas expansion to increase thereby further reducing the total pressure applied to the influx. There will be a point where, as the influx volume becomes larger, the pressure control valve will not be able to maintain a constant surface pressure because of the increased pressure from the liquid surge rate. Eventually, it will move to a fully open position to try to compensate for the increases in pressure from liquid surge while maintaining the set applied surface back pressure. In a circulation, it is noted that the mud flow rate reaches its maximum liquid surge rate first and shortly after the gas peak rate is reached. During this period, the pressure control valve/surface choke control will become erratic as rapid changes in its position are required to control the surging of the liquid and the subsequent two phase flow. Applied back pressure (34.5 bar/500 psi) may not be enough to offset the loss in hydrostatic pressure exerted by the WBM as the influx expands during circulation. Constant ASBP will often be compromised at larger gas volumes, resulting in large variations in surface pressure. If this happens, the gas handling circulation method would be changed to the CSBP method. There is presently no agreement as to whether, under the circumstances described in this section, the accuracy of Coriolis flowmeter outputs is better with the flowmeter positioned upstream or downstream of the pressure control valves.

7.3. Current status quo on empirical gas handling For rigs that use a high-capacity RGH system without an MPD subsea system installed, there is a precautionary mandate regarding the pressure-tight integrity of the subsea diverter closure device. If, at any time, there is any doubt of the pressure-tight integrity of the subsea closure device (trip tank level increase during RGH), the rig’s diverter system should be activated to the ‘divert’. This will shut the diverter packer and flowline valve and divert overboard to the lee side of the installation. Rigs with RGH/MPD deployed may consider using the MPD RCD as the secondary barrier in the event of the subsea diverter packer leaking. The procedures described here have been formulated and tested using software simulation packages. The real-world behaviour of gas cut drilling fluids captured in a closed-boundary riser annulus in ultradeep water is, to a grear extent, speculative. Much research has been conducted over the past decade or so on the migrational behaviour of different

126

04-SUT252_34(3).indd 126

28/07/17 7:17 pm

Underwater Technology Vol. 34, No. 3, 2017

gases in different liquids at different pressures and temperatures. The industry is acutely aware of the solubility of gases in oil-based liquids and bubble and break-out points are significantly retarded in oil-based or synthetic-based muds. However, it is now accepted throughout the industry that the application of ASBP retards the bubble and break out point for dissolved gases.

7.4. Empirical testing of a high-capacity riser gas handling system A number of different organisations in the drilling industry have, to date, made efforts and proposals to conduct a practical experiment to circulate free gas from extended riser strings in a controlled manner. No published test results exist as yet but such a test programme would achieve a number of important aspects of RGH, including: empirical proof of the capability of the system in-line degasser; empirical correlation between all prior modelling software simulations and calculations; physical performance data collection; absolute safety assurance; enhanced confidence levels of equipment; and practical system data in respect of system component design and system dynamic behaviour.

8. Summary and conclusions With the greater reservoir potential hydrocarbon reserves identified at deep and ultra-deep water offshore locations, together with the considerable areas of offshore ultra-deep water ocean margin territory yet to be exploited, the industry will undoubtedly continue ultra-deep water drilling activities in the foreseeable future. Therefore, the industry has no option but to consider the steadfast validity of installing high-capacity RGH systems on drilling installations built for deep and ultra-deep water exploration. The wake-up calls have been vivid and, in some cases, distressing. RGH systems, previously considered expensive luxuries and/or optional features, should now be categorised as safety critical

systems. Although, the integration of a working RGH system into the architecture of a managed pressure drilling system is virtually transparent, there remains a safety critical case to install RGH for those drilling installations that continue to use a conventional open riser drilling system. Early and deep water kick detection systems are now being offered to the industry. These contribute to the effectiveness of using a high-capacity RGH system and increase the achievable safety margins. Further, this can be extrapolated to conventional riser drilling systems, where RGH and early kick detection systems used together in a conventional riser drilling system can elevate the safety level for drilling rigs exploring in deep and ultradeep water offshore locations.

Acknowledgements The author wishes to express his gratitude to those vendors featured throughout this paper, in particular AFGlobal and Seadrill.

References American Petroleum Institute (API). (2001). RP64: Recommended practice for diverter systems equipment and operations, second edition. Washington: American Petroleum Institute. 76pp. International Association of Drilling Contractors (IADC). (2015). Deepwater well control guidelines, second edition. Houston: IADC. Kozicz, JR. (2012). Development of a marine riser gas management system. In: The Proceedings of the IADC/SPE Asia Pacific Drilling Technology Conference and Exhibition, 9â&#x20AC;&#x201C;11 July, Tianjin, China. doi: 10.2118/156399-MS. MacDougall, GR. (1991). Mud/gas separator sizing and evaluation. SPE Drilling Engineering 6: 279â&#x20AC;&#x201C;284. doi: 10.2118/20430-PA. Norsok (2012). D-001 Drilling facilities. Edition 3. Norsok. 72pp. Sutherland VA, Ehrlich M, Engler R and Kulinowski K. (2016). Drilling rig explosion and fire at the Macondo Well: Investigation report executive summary. Report No. 2010-10-I-OS. U.S. Chemical Safety and Hazard Investigation Board. 24pp.

127

04-SUT252_34(3).indd 127

28/07/17 7:17 pm

Underwater Technology is now

Open Access Open Access means that you can view and download Underwater Technology for free!

Underwater Technology is published in March, July and November each year.

All published papers from Volume 20 (1995) onwards* are available for Open Access at www.ingentaconnect.com/content/sut/unwt Abstracting and indexing services covering Underwater Technology include: American Academy of Underwater Sciences (AAUS) E-Slate; Aquatic Sciences and Fisheries Abstracts (Biological Sciences and Living Resources; Ocean Technology, Policy and Non-Living Resources; Aquatic Pollution and Environmental Policy); Compendex; EBSCO Discovery Service; Fluidex; Geobase; Marine Technology Abstracts; Oceanic Abstracts; and Scopus. * Volumes 1 to 19 are available in print version only.

Now Open Access

It is currently still free to get your paper published in Underwater Technology.

04-SUT252_34(3).indd 128

28/07/17 7:17 pm

www.sut.org

Technical Brieﬁng

doi:10.3723/ut.34.129 Underwater Technology, Vol. 34, No. 3, pp. 129–134, 2017

Application of the threat matrix to improving the efﬁciency of risk assessments for the integrity management of subsea pipeline systems JHA Baker* Baker Marine Technology Ltd, 7 Langton Court, Ponteland, Newcastle upon Tyne, NE20 9AT, UK Received 4 March 2017; Accepted 2 May 2017

Abstract The failure modes, effects and criticality analysis (FMECA) is commonly used to assess and manage risks in subsea pipeline systems but it can become large and cumbersome. Approaches are suggested, both utilising the threat matrix, for reducing the size of a FMECA and for reducing the manpower effort required for its application, so as to increase the efﬁciency of the risk assessment process. Keywords: risk assessment, failure modes, effects and criticality analysis, FMECA, threat matrix, segmentation, integrity, subsea, pipeline

1. Introduction A risk-based approach to integrity management has become standard practice for the offshore oil and gas industry on the UK Continental Shelf, both because the requirement to assess risks and manage them is defined in the Pipelines Safety Regulations 1996 (SI 1996 No. 825) and because it has become well proven as an effective management process. It is, in fact, common practice around much of the world. A common engine of risk-based integrity management, in which the maintenance of integrity is based upon a risk-based goal-setting rather than a prescriptive regime, is the failure modes, effects and criticality analysis (FMECA) which provides a structured approach to identifying risks and establishing the requirements for controlling them. This analysis, which is usually carried out via a spreadsheet, examines all threats to system integrity, the mechanisms generated by those threats that might lead to failure, and the failure modes (i.e. what * Email address: jerry.baker@bmartec.com

failure ‘would look like’) for every component (down to a suitable level of detail). Each row of the spreadsheet is dedicated to a single root cause of potential failure. The FMECA cannot handle joint causes and probabilities. In the row, the probability and consequence severity of the potential failure cause are assessed and, hence, the risk of that cause is determined. In addition, one or more activities are defined to either reduce the probability or alleviate the consequence in each case, together described as mitigations. Once complete, the rows can be ranked according to risk and these mitigations, prioritised by risk, then form the basis of the integrity management plan. One disadvantage of the FMECA is that, because every root cause requires its own row, its application can be time-consuming. Since workshops to initially establish, and subsequently review, a FMECA should involve representatives from several disciplines (e.g. production, well, corrosion, reliability, subsea and pipeline engineering, and production chemistry), the overall manpower demand and cost can prove significant. It is therefore of interest to seek ways to reduce the effort required while, at the same time, improving its effectiveness if possible. An understanding of the threats to the components is fundamental to risk-based integrity management. Recommended practices such as PD8010-4 (British Standards Institution (BSI), 2012) or DNVRP-F116 (Det Norske Veritas (DNV), 2009) present lists of generic threats that can be used to structure an integrity management FMECA. The integrity management guidelines published by the Energy Institute (EI, 2009) take this a step further by introducing the threat matrix. In that example, the threats and failure causes are laid out in the columns 129

05-SUT285_34(3).indd 129

28/07/17 7:16 pm

2. System segmentation Segmentation of a subsea system is explained in PD8010-4 (BSI, 2012) and the Energy Instituteâ&#x20AC;&#x2122;s integrity management guidelines (EI, 2009), with the latter including a discussion of the various generic segments. In simple terms, segmentation should reflect the commonality of threats and failure mechanisms, as well as commonality of the means of identifying influences or degradation associated with those threats. For example, consider the threat of internal corrosion as applied to a rigid production flowline system (i.e. no flexible pipe components). Starting at the subsea field, every tie-in spool between tree and manifold may be subject to different corrosivity. In a number of fields, the wells associated with one manifold may tap into different reservoirs so that levels of CO2 and H2S may vary significantly, as may fluid temperatures. Even for a single-reservoir field, where water injection is used for pressure support the degree of water breakthrough may be different at different wells. The export tie-in spool will carry the same commingled product as the riser tie-in spool, but temperatures may be very different and water may have dropped out along the length so that corrosivity may be quite different. If the pipeline can be pigged, the method for identifying corrosion will be the same in the export and riser tie-in spools, but not in the tree tie-in spools. Hence tree, export and riser tie-in spools all have to be treated as different segments. Within the manifold, the threat of corrosion will vary according to the commingling within the pipework, and detection of corrosion is complex, so this has to be a separate segment. The main flowline may be one segment unless topographic variation along its length leads to step changes in temperature profile or water pooling (e.g. if part of the length is buried, while another part is exposed and subject to spanning). The riser will be a further segment. This is illustrated in Fig 1, noting that erosion follows a broadly similar logic. In this very simple example, there is a requirement for ten segments. Even if the tree tie-in spools can be grouped (where the product is very similar

n zo 0m

Riser

ne zo

J-tube

0m 50

while the segments of the system are laid out in the rows. Within this relatively simple structure it is possible to identify those segments that are subject to each threat and, hence, within which each failure cause might be active. The aim of this paper is to consider two potential benefits from the threat matrix approach, namely reducing the effort required, while improving the effectiveness of the FMECA in reliability and integrity management.

JHA Baker. Application of the threat matrix to improving the efficiency of risk assessments for the integrity management of subsea pipeline systems

Fig 1: Example segmentation for internal corrosion and erosion

from all wells, as may be the case in some FMECAs) there are still six segments. Commonly, these segments will then be used for every threat and failure mechanism. An argument of this paper is that this is not optimal. Note that, in some risk-based integrity management systems, the corrosion management FMECA is separate to the integrity management FMECA, but the example would remain valid.

3. Construction of the FMECA In practice, one segmentation of the system is usually applied to all failure mechanisms. This will inevitably contain the greatest number of segments, and this has the disadvantage that for many failure mechanisms, the system is excessively divided. Since this subdivision is reflected in the construction of the FMECA, it results in more rows being created than are necessary. Those who have participated in risk assessment workshops will be familiar with the copying down of inputs from previous rows because â&#x20AC;&#x2DC;everything is the sameâ&#x20AC;&#x2122;. Apart from taking up valuable time unnecessarily, this also has a stultifying effect on the participants, dulling their acuity for those rows where it should be recognised that changes have occurred and may have altered the risk. Consider, for example, the external corrosion threat. It is likely that the tree and export tie-in spools will have been fabricated to the same specification, with the same coatings, such that the probability of coating breakdown or application-related failures is equal for all, and the same anodes are attached. All the spools will usually be protected by mattresses, so all would be equally difficult to inspect. In this case, all tie-in spools can be treated as a single segment so that the whole system divides into just four segments. It is possible that corrosion-resistant alloy will have been used for tree tie-in spools or, if the main flowline is piggable, it may be possible to identify external corrosion in the export tie-in spool via in-line inspection. Either of these considerations may result in the spools being sub-divided into two

130

05-SUT285_34(3).indd 130

28/07/17 7:16 pm

n zo m 0 50

Riser

J-tube

0m 50

ne zo

All tie-in spools

Riser

n zo

Fig 2: Example segmentation for external corrosion, impact, mechanical overstress and fatigue

0 50

J-tube

ne zo

All tree spools

0m 50

segments but this is still several fewer than required for internal corrosion. If the impact threat is considered, the tie-in spools in the subsea field (both tree and export) will be at risk mainly from fishing interactions and hence lateral impact, whereas the riser base spool will be at risk from vertical impact (from dropped and falling objects). This may be another basis for segmentation, noting that external corrosion can be a consequence of damage. Fig 2 illustrates segmentation where the main flowline is not piggable, all spools in the subsea field are of the same material but the variation between lateral and vertical impact is assumed significant. It requires five segments (compared to as many as ten in Fig 1). Mechanical overstress is associated with failure mechanisms such as upheaval buckling, lateral buckling, or excessive loads due to lack of support (e.g. spanning). The former tend predominantly to affect the main flowline, either throughout its length (single segment) or in discrete areas (e.g. due to topographical variation requiring more than one segment). Where tie-in spools are not protected by covering with mattresses or other (typically concrete) systems, they too may be subject to spanning but, with a few long exceptions, not to buckling. In these circumstances, the tie-in spools in the subsea field may be different to the riser tiein spool. Mechanical overstress may also result from constrained expansion which applies only to the tie-in spools since these are designed to accommodate flowline thermal expansion. Fatigue is most commonly associated with vortexinduced vibration affecting pipeline freespans, hence correlating with span-related mechanical overstress. It may also affect risers. Internal slugging can lead to vibration and hence fatigue in risers. Thermal and pressure cycling can lead to fatigue where damage has occurred (e.g. dents) and this is closely correlated with impact. Thus the segmentation in Fig 2 is typically applicable to these threats too. It is interesting to note that in relation to overpressurisation the system could require only three segments if all tree tie-in spools have a suitable pressure margin by design, as is generally the case (see Fig 3). While this is a generic review of a simple system, it illustrates that segmentation for external corrosion, impact, mechanical overstress and fatigue may well be the same in each case. Even with small variations in particular systems, the number of segments required is likely to be fewer than for internal corrosion, and this is even more so for over-pressurisation. Hence, if the maximum segmentation is still applied, there will be many rows that are effectively duplicates. These are not necessary, and lead to the copying down mentioned earlier.

Underwater Technology Vol. 34, No. 3, 2017

Fig 3: Example segmentation for over-pressurisation

The threats discussed above lend themselves to a spatial segmentation of the system. Some other parts of the system, notably valves, meters, sensors and the subsea production control system itself, may be better served by a sub-system segmentation. For example, poor cleanliness of hydraulic fluid is likely to impact the whole control system, including topsides units, whereas failure of electrical power distribution may affect only certain sub-systems, which may not be spatially orientated. It is argued that using a threat matrix, which focusses on segmentation by threat without moving to component level (i.e. prior to constructing the FMECA), facilitates this elimination of duplicate rows from the FMECA in a way that is not possible, or at best very difficult, if working only with a threat list since this becomes component-orientated. The old adage about ‘not being able to see the wood for the trees’ comes to mind. The discussion in this section illustrates this. It is in this way that starting with a threat matrix results in a more streamlined FMECA.

4. Reduction of review effort In principle, at the start of every integrity management cycle, one of the first activities is to review the risk assessment. This is frequently interpreted as organising a workshop to work through every row of the FMECA (including any duplicated rows) and ask the question, “Has anything changed?” This is hugely time-consuming and, frequently very little has changed, exacerbating the stultifying effect of

131

05-SUT285_34(3).indd 131

28/07/17 7:16 pm

05-SUT285_34(3).indd 132

Change in nearby impressed current system

CP survey results

Visual inspection results

Change in composition, concentration, ﬂow regime

Testing; probe, coupon, sand detector results

Sampling: CO2, H2S, O2, SRB, solids, wax, etc

(1)

Notes Span lengths, gaps change slowly (2) Upheaval or lateral buckling (3) Constrained expansion (4) Vortex-induced vibration, spans, risers

Bacteria in soil

High internal temperature

Zero volts, amps High anode depletion rate

Visual inspection results

Change in ﬁeld/ platform activity (maintenance, drilling, etc)

High platform or ﬁshing activity Low levels of protection

External corrosion, Impact cathodic protection

Impurities

Low temperature

High temperature

Low pressure

Zero ﬂow, empty High pressure

Low ﬂow, low level

High ﬂow, high level

Internal corrosion, erosion, ﬂow assurance

Visual inspection results

Change in behaviours (ﬂow, wave ht, soils, etc)

Soil deposition(3)

High internal temperature(2)

High internal pressure(2)

High level of spanning Low transience(1)

Mechanical overstress (spanning, buckling, etc)

Table 1: Possible guide words for identifying changes via a threat matrix

Natural frequencies of spans(4)

Damage (dents, gouges) found Riser support failures(4)

Lack of support High probability of buckling Low structural damping(4)

High tidal/ current ﬂows(4)

Fatigue

Condition of pressure sensors

Change in isolations, procedures, etc

High pressure

Overpressurisation

Sensor calibration

Valve testing

Change in constraint (soil deposition, drill cuttings)(3)

High temperature Low temperature

Low pressure

High pressure

Fittings (ﬂanges, connectors, valves, sensors)

Fluid sampling, MCS data assessment

System test results

Cleanliness of hydraulic ﬂuid Change of control ﬂuids (umbilical blockage)

Low temperature

High temperature

Low pressure

High pressure

Subsea production control system

Fluid sampling

Cleanliness of chemicals Change of chemicals (reactions, non-metallics, seals) Test results: valves, sensors, etc

Low temperature

High temperature

Low pressure

Zero ﬂow, empty High pressure

Low ﬂow, low level

High ﬂow, high level

Chemical injection system JHA Baker. Application of the threat matrix to improving the efficiency of risk assessments for the integrity management of subsea pipeline systems

132

28/07/17 7:16 pm

Underwater Technology Vol. 34, No. 3, 2017

the duplicate rows. This is a significant downside of the approach since the process requires participants to be alert to the levels of risk. It is argued that a better approach is to use the threat matrix for the initial screening. In most cases, having the necessary disciplines represented and asking, “Has anything changed?” for each cell of the threat matrix (where merged segments are reflected in merged cells) should be sufficient. For those segments where change is noted, it is then necessary to enter the FMECA for a row by row appraisal. But where no change has occurred, those parts of the FMECA itself can be by-passed. This should of course be documented. In practice, the questioning has to be more detailed than a simply asking if anything has changed. It is recommended that guide words are developed, similar to those used in a hazard and operability (HAZOP) study, to ensure a robust review. The HAZOP is a well-established, structured process that harnesses the experience within a multi-disciplinary team to identify, via a set of guide words, the possibility that a system in operation may deviate from the design intent, thus infringing safe or allowable operating limits (SOL/AOL). Some suggestions for guide words for this application are given in Table 1. Since HAZOPs generally deal with production systems and plant, the typical guide words, when subsea pipeline system integrity and reliability are considered, are best suited to internal corrosion, erosion and flow assurance. Therefore in Table 1 a typical set of guide words (created from several of the sets available on the internet), has been inserted in the left hand column, under this title. The remaining columns refer to different failure mechanisms, or system failures (fittings, subsea production control system, chemical injection system). These could no doubt be expanded by discipline experts. The guide words in these columns have been selected by taking the typical HAZOP guide words and seeking an equivalence, leaving the cell blank when no words appear relevant. Thus the guide words in Table 1 are only examples, but this process can be used by integrity, reliability or corrosion engineers to select agreed guide words to suit the system in question and the corporate approach to risk management. The guide words should be interpreted broadly. For example, ‘change’ may be physical, organisational or procedural, or may refer to trends in monitored data or inspection results (e.g. rates of degradation updated in the light of additional data). Likewise, ‘high’ and ‘low’ may be treated as comparative as well as absolute (i.e. ‘higher’ or ‘lower’ than typical previous values or rates of degradation).

Incorporating these data-related aspects constitutes the feedback into the system required at the end of each integrity cycle. At the commencement of every integrity cycle, it is necessary to investigate every segment in the threat matrix with the resulting set of guide words, with all discipline engineers present. Following this, the FMECA need be reassessed only for those parts associated with any segment where change is identified and with only the relevant discipline engineers present. This approach should require many fewer man-hours than a full row by row reappraisal of the FMECA without compromising the intent.

5. Discussion The effectiveness of an integrity management process depends upon its user-friendliness. No matter how well defined and constructed, if the process leads to ennui such that engineers lose concentration or skip steps, risks will be missed and failures are likely to occur. Hence it is important to remove all duplication and streamline the FMECA. Clearly, by-passing those sections of the FMECA where no changes have been identified also helps in this regard. In section 3, the riser is defined as a single segment. In practice, a rigid riser should be divided into a number of segments vertically (e.g. fully submerged, splash zone and dry), because external threats are different in these zones. Depending upon the arrangement, the cellar deck, and base of the riser may also require separate segments, and riser caissons or J-tubes may result in even more segments being required. However, the principle of minimising segments according to common threats and common means of identification still applies. The same principle also applies to flexible pipes and risers. These are subject to different failure mechanisms (e.g. impact can lead to breaches of the outer sheath, permeation from the bore into the annulus can lead to corrosion of the pressure and tensile armour wires, and certain zones are particularly prone to fatigue or over-bending). Therefore, the basis of segmentation will be different, but segmentation should be minimised for different failure mechanisms in the same way, using the threat matrix.

6. Conclusions The FMECA has been established as an effective tool for carrying out integrity risk assessments as part of risk-based integrity management, but it can

133

05-SUT285_34(3).indd 133

28/07/17 7:16 pm

JHA Baker. Application of the threat matrix to improving the efficiency of risk assessments for the integrity management of subsea pipeline systems

become cumbersome. The more that can be done to streamline the process the better. In this paper, two approaches to this are considered. Firstly, creating a threat matrix prior to developing the FMECA can significantly reduce the number of rows required, because the clarity it provides enables some segments to be merged. Secondly, working through the threat matrix and asking â&#x20AC;&#x153;Has anything changed?â&#x20AC;? using a suitable set of guide words based upon those applied during a HAZOP can reduce the manpower required when reappraising the FMECA. Both these approaches will reduce the ennui associated with many FMECA workshops in

practice, thus resulting in a more user-friendly and effective process.

References British Standards Institution (BSI). (2012). PD 8010-4:2012: Pipeline systems, Part 4: Steel pipelines on land and subsea pipelines. Code of practice for integrity management. London: BSI. 50pp. Det Norske Veritas (DNV). (2009). DNV-RP-F116: Integrity management of submarine pipeline systems. Oslo: DNV. 149pp. The Energy Institute (EI). (2009). Guidelines for the management of integrity of subsea facilities. London: Energy Institute. 82pp.

134

05-SUT285_34(3).indd 134

28/07/17 7:16 pm

doi:10.3723/ut.34.135 Underwater Technology, Vol. 34, No. 3, pp. 135–139, 2017

Technical Brieﬁng

www.sut.org

Underwater monitoring system for body temperature and ECG recordings Andreas Schuster*1,2, Olivier Castagna3, Bruno Schmid3, Tobias Cibis4 and Arne Sieber1,5 1 SEABEAR GmbH, Leoben, Austria 2 ACREO AB, Gothenburg, Sweden 3 IRBA Institut de Recherche Biomedicale des Armées, Toulon, France 4 Digital Sports Group, Friedrich-Alexander University Erlangen-Nürnberg, Germany 5 Chalmers University of Technology, MC2, Gothenburg, Sweden Received March 2017; Accepted June 2017

Abstract A new device was developed and tested in a series of diving experiments investigating the physiological effects of immersion on military divers for long periods (8 h to 12 h). During these experiments, the body temperature (core and skin) and electrocardiogram (ECG) of the divers were recorded and monitored in real time. The system developed for this purpose comprised a modiﬁed VitalSense temperature monitoring device from Philips Respironics and a one-channel ECG housed in a pressure-proof case. Recorded data were transmitted wirelessly to a PC. The recording and visualisation software was developed under National Instruments LabWindows.

1. Introduction The French Navy wanted to investigate the physiological effects of long-term (8 h to 12 h) immersion on military divers in moderate water temperatures at depths of up to 20 m (Castagna et al., 2013; Desruelle et al., 2014). The investigation was focused on the divers’ electrocardiogram (ECG), skin and core temperature, and required a system capable of monitoring these physiological effects. In order to identify a suitable design, a number of different approaches in ECG and temperature measurement were evaluated (Baig et al., 2013; Bosco et al., 2014; Sieber et al., 2010). From the evaluation results, the authors derived a new modified device suitable for application in hyperbaric and underwater environments. The system developed was a one-channel ECG device and several temperature measurement sensors. This technical briefing focuses on the technical details * Contact author. Email: andreas.michael.schuster@gmail.com

of the ECG and temperature recording, as well as the data processing and visualisation.

2. Underwater electrocardiogram To monitor a one-channel ECG during diving and to access recorded ECG data, a small-sized circuit board was designed and assembled. Fig 1 shows the circuit board (referred to as the monitoring board). The circuit was designed using an instrumentation amplifier (AD632, Analog Devices) to pre-amplify the ECG signal obtained from the electrodes and additional operational amplifiers in series to filter the signal. The filters contained a band pass filter with cut-off frequencies at 1.5 Hz and 100 Hz, and a notch filter to reduce 50 Hz interferences. A microcontroller (ATxmega32A4, Atmel) digitalised the signal, sampling the ECG with a frequency of 250 Hz and 12 bit resolution. EL502 long-term electrodes from BIOPAC were used as ECG electrodes. Before each dive, a new set

Fig 1: Monitoring board

135

06-SUT281_34(3).indd 135

28/07/17 7:16 pm

Schuster et al. Underwater monitoring system for body temperature and ECG recordings

of electrodes was soldered to the ECG cables and the contacts were covered with hot glue and further isolated with a self-adhesive waterproof film (Tegaderm, 3M). A rechargeable Li-ion 18500 battery (Trustfire) with a voltage of 3.7 V and 1800 mAh of capacity powered the monitoring board.

3. Underwater temperature monitoring In addition to the ECG, the French Navy wanted to monitor the divers’ body temperature while wearing 7.5 mm thick neoprene wetsuits. Hence, the skin and core temperature in eight different locations needed to be measured for each diver to determine the temperature distribution in different areas of their body. Therefore, a total of nine temperature sensors were applied per diver. To identify the most suitable temperature measurement, different methods were considered. Wired temperature sensors. One solution to measure the skin and core temperature of divers is to use wired temperature sensors. However, a high number of cables would have been required to be routed along the diver. All the cables would have posed a safety hazard, such as possible entanglement, and seemed to be uncomfortable for the diver. Ingestible sensor. Another method considered was using HQinc CorTemp® ingestible core temperature sensors, which are little pills (length: 22.4 mm; diameter: 10.9 mm) that can be swallowed. These pills measure the temperature inside the digestive tract and transmit it out of the body via a low-frequency electromagnetic signal (262 kHz to 300 kHz). The advantage of this system is the relatively low cost of the pills. However, only one type of temperature sensor is available. There are no sensors that can be used to measure the skin temperature within this system. In addition, the way the system transmits the data does not allow a discrimination of data sent by the different sensors that have been swallowed. VitalSense sensors. Philips Respironics provides two different types of body temperature sensor: an ingestible pill to measure the core temperature; and patches that can be attached to the skin to measure the skin temperature. The VitalSense device is capable of tracking ten sensors simultaneously, displaying the data on a screen, and storing the data for later analysis. The VitalSense system has its own internal rechargeable battery. The VitalSense system from Philips Respironics proved to be most suitable to measure the body temperature of the divers, since it provides a sufficient number of temperature sensors for both skin and core temperature measurement, without the need for a lot of cables. The authors designed a new serial interface on the VitalSense system by modifying

the internal transceiver chip’s serial connection. This way the system could be read in real time.

4. Data transmission Several options were evaluated to determine the most suitable system for data transmission between the monitoring device, a PC in the laboratory and open water dive settings. The following options were taken into consideration. Wired connection. In the hyperbaric chamber, a pressure-sealed cable could be fed through the wall of the chamber to transmit the data through the cable. However, this solution presents two issues: a potential separation between the computer outside of the chamber; and the fact that ECG electrodes were required to be attached to the diver’s body. Given the PC is connected to a power outlet, if a malfunction occurred the diver might be directly connected to the power outlet via the data transmission cable and the ECG electrodes. In addition, having a cable from the monitoring system to a fixed connector in the wall of the chamber would be a safety hazard if the divers have to be evacuated. Electromagnetic signal transmission. Technology such as Wlan, Bluetooth or ZigBee based communication operate on a carrier frequency of 2.4 GHz. Signals of that frequency travel only a few centimetres in fresh water (Lloret et al., 2010) and even less in sea water because of the increased conductivity. Therefore, this type of technology is not a suitable solution for transmitting the data from the diver through the water, out of the chamber and to the laptop. Other wireless communication systems that have previously been employed underwater, for example for the communication between tank pressure sensors and dive computers, rely on low-frequency electromagnetic carriers of 5 kHz to 32 kHz. However, the data rate is only a few bytes/s (Sieber et al., 2010) and therefore too low to transmit an ECG signal. Acoustic/ultrasonic communication. Ultrasonic communication systems can be used to communicate up to 20 km in water. Even though the transmission range and data rate are sufficient, it does not work well in pressure chambers because of echoes. In addition, ultrasonic modems are bulky, disturbing during experimental dives and very expensive. Infrared. One option the authors tested used an infrared (IR) transmitter placed close to one of the windows in the hyperbaric chamber and a cable connecting the monitoring system to the IR transmitter. Outside of the chamber, an IR receiver was used for IR remote controls that received and decoded the IR signal. This method was able to transmit the ECG signal and the temperature information, but

136

06-SUT281_34(3).indd 136

28/07/17 7:16 pm

Underwater Technology Vol. 34, No. 3, 2017

it relied on the IR transmitter being in line of sight with the IR receiver. This could not be guaranteed in open water conditions, even if the IR transmitter was placed on top of a buoy floating at the surface. Bluetooth. Another approach to solve the data transmission problem that was tested involved a Bluetooth module that was put into a small buoy that floated at the water surface. A cable connected the Bluetooth module to the monitoring system. The Bluetooth module received the data through that cable and transmitted it to a Bluetooth module inside a laptop outside of the hyperbaric chamber or on a support vessel in the open water setting. Worries arose that the walls of the hyperbaric chamber would block the Bluetooth signal if the module was not close enough to one of the windows. However, in all tests in the hyperbaric chamber of the French Navy in Toulon the system worked without interruption of the Bluetooth link. In conclusion, the Bluetooth module seemed most likely to possess all requirements for successful application in both the hyperbaric chamber and open water dive settings.

5. Data processing For supervision, real-time monitoring of all measured parameters on a laptop was requested. Since the measurement system was newly developed, there was no software that could use the serial output of the system to display the data in real time and also save it. This meant a program needed to be developed capable of fulfilling the following requirements: i) Distinguish between different temperature sensors by selecting and assigning each sensor to specific body parts.

ii) Produce a graphical representation of the temperature data to visualise trends. iii) Plot the ECG signal in real time and save short samples of the signal at different instances during the dive to identify changes that might occur. iv) Store all measured data in a format that would allow in-depth analysis after the dives. A graphical user interface (GUI) was implemented using LabWindowsTM/CVI from National Instruments to fulfill all those tasks (Fig 2). The user could select the serial numbers of the temperature sensors and assign them to different predefined parts of the body. Furthermore, the user could open the serial ports to the matching Bluetooth receiver. The user interactions with the interface were programmed in C, the same programming language that was used to program the firmware on the microcontroller of the monitoring board. The monitoring board transmitted blocks of 50 ECG samples five times per second. Temperature data from the VitalSense system were distinguished from the ECG data by additional serial string identifiers. Each temperature sensor sent the measured temperature in 30 s intervals. Once a stable transmission was initiated, the program displayed the ECG signal in the lower graph of the main window. The ECG signal could be analysed in depth by opening a window showing a graph of 5 s intervals. Any of those 5 s intervals that were displayed could be stored and compared to previously saved intervals. The table in the upper left part of the window listed all received temperature data. In addition, the core temperature was plotted in the graph in the upper right part of the window, and the user could open another window to plot

Fig 2: Main window of the data processing software

137

06-SUT281_34(3).indd 137

28/07/17 7:16 pm

Schuster et al. Underwater monitoring system for body temperature and ECG recordings

Laptop Temperature sensors

Bluetooth module

Vital sense device Wireless data transmission

ECG electrodes

Bluetooth transmission

OtterBox: DryCase 2000

Diver

Analog signal

Monitoring board

Serial transmission (USART)

Bluetooth module Buoy

Fig 3: Block diagram of the monitoring system

the temperature curve of any previously selected temperature sensors. This enabled a visual representation of the temperature trend. All data were stored in different csv files. Those files could be accessed with different spreadsheet programs to analyse the data in more detail and plot the different parameters.

6. Underwater monitoring system The microcontroller on the monitoring board has a total of five universal synchronous/asynchronous receiver/transmitter (USART) interfaces available. One of these was connected to the newly designed serial interface of the VitalSense system. This way the temperature data were also available on the monitoring board and could be sent to a laptop for further data processing. The monitoring board and a VitalSense device were connected by a short cable and placed in a waterproof DryCase 2000 from OtterBox. Externally the case was equipped with two waterproof connectors, one of which was used for the three ECG electrodes; and the other for data transmission. The Bluetooth module in the buoy was connected to this data transmission connector by a cable. Depending on the dive setting the cable was 2 m or 30 m long. The Bluetooth module received the ECG and temperature data from the monitoring board and transmitted it to a laptop for supervision. Fig 3 shows a block diagram of this setup. Fig 4 shows the two monitoring systems that were built; one has the ECG cables and Bluetooth buoy attached to it. The outer dimensions of the DryCase 2000 are 14.6 cm by 7.9 cm by 2.5 cm, and it weighs a total of 450 g.

7. Proof of concept and discussion The French Navy performed a series of dive experiments in a hyperbaric chamber and in the ocean testing the underwater monitoring system. It proved

Fig 4: Two assembled monitoring systems

to be capable of collecting the body temperature and ECG data continuously for up to 12 h. The data were transmitted to a laptop outside the hyperbaric chamber or on a support vessel and could be inspected in real time for supervision. When the infrared data transmission link to transmit the data out of the hyperbaric chamber was tested, the transmitter was connected with rubber bands to a stationary object in the chamber. This was not considered a safety issue in case of an emergency, because this attachment could easily be removed just by pulling on the cable. However, the attachment to that stationary object dramatically decreased the ability to move around freely inside the chamber. Regarding the open water trials, the cable running from the DryCase 2000 to the Bluetooth module in the buoy was a limitation of the monitoring system. Special attention had to be paid to avoid entanglement. The case from OtterBox used to house the monitoring system was tested to a depth of 30 m. At this depth it was significantly deformed, but did not break. It is unclear to which depth it can withstand the pressure, but alternative housings can be manufactured

138

06-SUT281_34(3).indd 138

28/07/17 7:16 pm

Underwater Technology Vol. 34, No. 3, 2017

easily and cheaply. In cases where only the ECG is needed, the whole device could be assembled to be much smaller since only the monitoring board and a battery are needed inside the waterproof housing. For a more specific analysis of the cardiac activity, upgrading the ECG to more than just one channel could be considered (Cibis et al., 2015).

Acknowledgements This project was supported by the French Navy and the EU-FP7-Marie Curie Initial Training Network Phypode.

References Baig MM, Gholamhosseini H and Connolly, MJ. (2013). A comprehensive survey of wearable and wireless ECG monitoring system for older adults. Medical & Biological Engineering & Computing 51: 485–495. Bosco G, De Marzi E, Michieli P, Omar HR, Camporesi EM, Padulo J, Paoli A, Mangar D and Schiavon M. (2014). 12-lead Holter monitoring in diving and water sports: a preliminary investigation. Diving and Hyperbaric Medicine 44: 202–207.

Castagna O, Desruelle AV, Blatteau JE, Vallée N, Schmid B, Barrot L, Bellon F, Chopard R, Brocq FX, Valero B, Sieber A, Schuster A and Regnard J. (2013). Physiological Effects of Long-Duration Diving (10H) in Professional Divers. In: Abstract and conference book for the DAN Tricontinental Scientific Meeting on Diving and Hyperbaric Medicine, 22–29 September, La Reunion, France, p 22. Cibis T, Groh BH, Gatermann H, Leutheuser H and Eskofier BM. (2015). Wearable real-time ECG monitoring with emergency alert system for scuba diving. In: Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 25–29 August, Milan, Italy, 6074–6077. Desruelle AV, Schmid B and Castagna O. (2014). Thermal aspects of long-term immersion at 18°C in combat swimmers. Caission 29: 43. Lloret J, Sendra S, Ardid M and Rodrigues J. (2012). Underwater Wireless Sensor Communications in the 2.4 GHz ISM Frequency Band. Sensors 12: 4237–4264. Sieber A, L’Abbate A, Kuch B, Wagner M, Benassi A, Passera M and Bedini R. (2010). Advanced instrumentation for research in diving and hyperbaric medicine. Undersea Hyperbaric Medicine 37: 259–269. Sieber A, Schuster A, Reif S, Madden D and Enoksson P. (2013). Head-Up Display System for Closed Circuit Rebreathers with Animagnetic Wireless Data Transmission. Marine Technology Society Journal 47: 42–51.

139

06-SUT281_34(3).indd 139

28/07/17 7:16 pm

Society for Underwater Technology

Educational Support Fund Sponsorship for Gifted Students in Marine Science, Technology and Engineering to meet industry’s critical shortage of suitably qualiﬁed entrants.

SUT sponsors UK and overseas students (studying in the UK and abroad) at undergraduate and MSc level who have an interest in marine science, technology and engineering. Students are supported who are studying subjects such as:

Offshore and Ocean Technology Subsea Engineering Oceanography Marine Biology Ship Science and Naval Architecture Meteorology and Oceanography The SUT annual awards are £2,000 per annum for an undergraduate, and £4,000 for a one-year postgraduate course. (Part-time postgraduate studies funding available.) As one of the largest non-governmental sources of sponsorship, the SUT has donated grants totaling almost half a million pounds to over 270 students since the launch of the fund in 1990.

For further information please contact Society for Underwater Technology, Unit LG7, 1 Quality Court, London WC2A 1HR UK t +44 (0)20 3440 5535 e info@sut.org or please visit our website

www.sut.org 06-SUT281_34(3).indd 140

28/07/17 7:16 pm

Modern Observational Physical Oceonography: Understanding the Global Ocean By Carl Wunsch Published by Princeton University Press

Hardcover, 2015 ISBN 9780691158822 512 pages This is a remarkable book. Carl Wunsch sets out his stand right at the beginning. “This book is directed primarily to beginning graduate students in physical oceanography and to working scientists in allied fields seeking some understanding of what the science teaches us about the behavior of the fluid ocean.” Significantly he adds, “My main motivation has been to produce an introduction conveying what the observational revolution of the past thirty years has taught us – a revolution that is primarily about the ocean as a time-varying system.” He goes on to state that while many good theoretical textbooks exist, there are none which depict the qualitative behaviour of ocean water as now perceived. Elsewhere he states very clearly his view that observations have revolutionised our picture of the ocean, bringing home its extreme, rapid variability, yet at the same time we are doing fewer

and fewer observations (possibly for economic reasons), placing our faith in models which often fail to represent vital aspects of complex oceanic systems. Wunsch seeks to offset this problem with a book that is deliberately biased towards observations and their immediate interpretation over models. This book succeeds magnificently. The first main chapter, ‘Observing the oceans’, is a tour de force in which he surveys the entire history of ocean observation, uniting past and present observations through the inexorable questions of accuracy and adequacy in capturing time and space dependence. He does not ignore past methods, in fact he seems to have a soft spot for Nansen bottle casts, but stresses the value of new methods and what they tell us. The second chapter, ‘What does the ocean look like?’ is brilliant in that he has assembled a comprehensive collection of coloured maps of the ocean world. They display not just the standard ocean parameters of yesteryear but also the results of comprehensive surveys of quantities such as geoid topography, kinetic energy in the ocean, buoyancy frequencies, sea surface elevation, wind stress, precipitation, and heat transfer. Thus sobered by the complexity of the ocean in space and time, we are presented with a mainly theoretical chapter on ‘Linear Wave Dynamics’ dealing with waves on every scale. In the next chapter, internal and inertial waves are singled out for attention. ‘The tide disturbing potential and the Milankovitch forcing’ is the title of the next chapter, followed by a chapter on observations

www.sut.org

Book Review

doi:10.3723/ut.34.141 Underwater Technology, Vol. 34, No. 3, pp. 141–142, 2017

of tides and related phenomena. I have never seen a clearer account of the tide generating forces and their impact on the oceans. Such material is often omitted from oceanographic textbooks, but Wunsch justifies his extensive treatment from the fact that our complacent view that we understood tidal forcing completely was shaken by the 1992 TOPEX/POSEIDON altimeter mission. This mission showed the global presence of tidal-period internal waves, and hence the unexpected contribution of tides to ocean mixing. The book then proceeds to larger-scale ocean physics, which Wunsch calls ‘Balanced (i.e. geostrophic) motions’. Again, clear and beautiful diagrams are included from many sources to display functions such as power law behavior of velocity data. The time-mean ocean circulation is investigated, with global inverse calculations invoked to derive the full complexity of the thermohaline circulation. Convective regions are described, though with the inexplicable exception of the Greenland Sea. Global heat and freshwater fluxes are explored, with more excellent maps to show the vertical velocity in the flow fields. The large-scale analyses proceed to consider the thermocline, the role of eddies, the Sverdrup balance and kindred topics. Finally, the physical theory of the large scale circulation is developed and applied to outstanding problems of global energetics and mixing. Useful appendices deal with analysis methods and inverse and state estimation methods. My only complaint about the book is a limitation freely admitted by Wunsch: that the book

141

07-SUT264_34(3).indd 141

28/07/17 7:16 pm

Carl Wunsch. Modern Observational Physical Oceonography: Understanding the Global Ocean

does not consider the role of the Arctic Ocean. The book’s beautiful and informative global maps of every kind of ocean process and parameter extend to the shores of the Antarctic continent but usually do not extend north of the Siberian mainland, so that the Arctic appears, if at all, as a narrow smear. It would cost little to have a small circular polar stereographic map alongside the bigger maps to give an indication

of what happens in the Arctic, a region which cannot be ignored and for which much data are available. Direct discussion of the oceanic role of sea ice is confined to six lines on p. 346. For anyone who seeks to know, understand, and especially measure the wondrous phenomena of our liquid oceans, I cannot think of a better book than this one. It will inspire new generations of ocean observers, a

species that is increasingly in short supply in a world of modelling. In defence of his unabashed focus on observation, Wunsch states “The temptation to conflate a model calculation with the real world has rarely been resisted, and the results now haunt climate studies.” (Reviewed by Professor Peter Wadhams, University of Cambridge)

142

07-SUT264_34(3).indd 142

28/07/17 7:16 pm

Robot Fish: Bio-inspired Fishlike Underwater Robots By R Du, Z Li, K YoucefToumi and P Valdivia y Alvarado (Eds) Published by: Springer-Verlag Berlin Heidelberg

Hardcover 2015 ISBN 978-3-662-46869-2 377 pages Nature has often provided inspiration to engineers and designers to develop new or enhanced systems. In the marine environment many fish exhibit characteristics or performances that exceed the current generation of manned or unmanned underwater vehicles, thus providing the potential to improve bioinspired devices. In this book the editors bring together 12 chapters from a range of authors that provide a snapshot into current research and development in robotic fish. The book’s objectives are to: 1) summarise the state of the art in design, build and control of robot fish; 2) analyse major obstacles for further development; and 3) point out future research and development directions. While it broadly succeeds in the first objective, a more rigorous focus on the closed-loop control of freeswimming robot fish and the transition from the laboratory to

real world applications would have been desirable. Chapter 1 provides a broad introduction to the topic discussing the sheer range of fish propulsion methods. It then provides a mathematical basis for body caudal fin propulsion using Lighthill’s elongated body theory as well as a review of robot fish and a methodology for robot fish design. While the authors’ enthusiasm for fish propulsion is clear, readers should always treat claims such as “It is known that the propulsion efficiency of fish exceeds 90 %” with caution. High hydrodynamic efficiency of certain fish propulsion methods has been shown. However, the evolution of fish has led to a plethora of specialisms, where some species have evolved body shapes and propulsion modes that lead to high efficiency. However many others have evolved alternative specialties at the expense of propulsive efficiency. This focus on hydrodynamic efficiency risks missing the broader range of capabilities shown by fish species. In Chapter 2 Lauder and Tangorra discuss the complexity of fish propulsion and manoeuvring, which is achieved through the use of both a flexible body exhibiting undulatory motion as well as numerous flexible control surfaces. These enable the fish to both generate forward propulsion as well as execute rapid manoeuvres in roll, pitch and yaw. Based on detailed fundamental observations of biology and swimming kinematics of bluegill sunfish, the authors have been able to produce complex multi degree of freedom articulated replicas of the dorsal,

www.sut.org

Book Review

doi:10.3723/ut.34.143 Underwater Technology, Vol. 34, No. 3, pp. 143–145, 2017

anal and caudal fin which produce good approximations of the complex fin kinematics used by sunfish. In this chapter the authors demonstrate a deep understanding of the biology of fish swimming which has enabled them to develop truly elegant biomimetic devices. This deep grounding in biological understanding is perhaps missing in some of the following chapters. Generating the required actuation to mimic undulatory fish propulsion presents a challenge to engineers, who are best equipped to generate rotary motion. Chapter 3 gives an overview of approaches developed to replicate this motion and then concentrates on discussing three variants of a wire driven approach which can be used to replicate the muscles and tendons of a fish. Chapter 4 describes a multi joint fish design. The design has seven actuators (one pectoral fin, three body, two pectoral and one pelvic) enabling a range of gaits to be replicated including body caudal fin and median and or paired fin. Two control approaches are presented: wave-based control and central pattern generator based control (CPG). CPG is highlighted as a suitable approach to enable representative body kinematics for robots with multiple joints. However, enhanced closed-loop sensory feedback would be required for true fish-like operation in realistic environments. Chapter 5 moves beyond the instantly recognisable fish propulsion modes to consider the oscillatory pectoral fin based locomotion of rays, skates and

143

08-SUT256_34(3).indd 143

28/07/17 7:16 pm

Du R et al. Robot Fish: Bio-inspired Fishlike Underwater Robots

mantas; rajiform propulsion exemplifies the need for flexible mechanisms and materials able to produce complex three dimensional deformations. After presenting video gait analysis of a cownose ray this chapter presents the design for the bionic fish. Each pectoral fin is articulated with three ray fins which enable the fin to mimic a kinematic wave passing down the pectoral fin. Each ray fin is constructed of a rocker slide mechanism with two, three and one joints respectively moving from the nose to the tail. Combinations of towing tank and free running experiments show the ability of the actuation to generate thrust and enable the bionic fish to cruise, turn, dive and surface using pre-programmed fin actuation. Closed-loop swimming control is highlighted as a future requirement. Conventional marine robots are all based on a rigid body paradigm. The majority of the bionic fish presented in the earlier chapters are an extension of this approach where a flexible skin is placed over rigid discrete assemblies of gears, pulleys, cables or linkages to produce the desired deformation. Each of those linkages must be actively controlled either individually or in groups. Chapter 6 considers how the natural dynamics of a soft flexible body can be exploited to generate complex deformations using an underactuated approach. The design objective is to find the material distributions along the viscoelastic body such that forced vibrations match the required body motions. To generate the forced vibrations a simple actuation system based around a conventional RC servo is embedded in the soft body. Using this approach a single actuator is required to replicate thunniform or carangiform propulsion.

Additional actuators are required to enable control of pitch. The total propulsion efficiency is of the order of 0.1 % which for practical applications is unacceptably low. However the efficiency of the soft body to transmit the propulsive forces is approximately 50 % and the robustness of the design is encouraging, given reliability and robustness are key considerations when designing long term deployments in harsh ocean environments. Much research has focused on developing bionic fish to match the swimming efficiency of pelagic fish species. The iSplash project described in Chapter 7 explores how bionic fish can be designed to outperform their biological contemporaries in terms of swimming speed when measured in body lengths per second. Using the common carp as inspiration the authors develop a 32 cm untethered carangiform swimmer capable of swimming at 3.7 m/s or 11.6 body lengths per second. Nektonic biological systems rely on the cyclic contraction of combinations of muscle tissues to provide actuation. Chapters 8 and 9 argue that novel actuation approaches based on smart materials that undergo a change of shape when simulated by an external stimulus provide a simpler route to fish-like propulsion. Chapter 8 presents three different underwater robots actuated with ionic polymermetal composites: a bionic fish with an ionic polymer-metal composite (IPMC) caudal fin, a cownose ray with IPMC actuated pectoral fins and finally a buoyancy control device actuated using an IPMC-enhanced electrolysis. Chapter 9 develops a bionic fish with a macro fibre composite piezoelectric laminate actuating a caudal fin. For both implementations,

the approach is still at a low technology readiness level and is not yet ready to truly challenge conventional motor based systems in terms of speed or efficiency but the simplicity and ease of fabrication is highly desirable. The last three chapters move beyond fish inspired propulsion. Chapter 10 looks more broadly across nature to develop a multifunctional underwater microrobot, which at 35 mm long is able to walk, float, swim and grasp using a combination of eleven IPMC actuators and two shape-memory alloy (SMA) actuators. Multi-functionality is a common theme through nature and where applied in robots it provides the opportunity to extend the capabilities of new robot systems. Many species of fish gain significant benefits by living as part of a shoal. Chapter 11 considers how robot fish can collaborate to perform complex tasks. Methodologies and experimental results are presented for small robot fish equipped with cameras undertaking: vision-based target tracking collision avoidance, leader follower formation control and cooperative box pushing. The challenges inherent in underwater localisation and communication make this form of behaviour significantly challenging, and scaling from the laboratory to the open ocean is not readily achievable. Chapter 12 considers how fish and robot fish interact. Two experimental studies highlight the ability of using robot fish to modify live zebra fish behaviour. The first experiment illustrates both single and shoals of live zebra fish are attracted to a compartment containing a robot fish in preference to an empty container. In the second set of experiments the average

144

08-SUT256_34(3).indd 144

28/07/17 7:16 pm

Underwater Technology Vol. 34, No. 3, 2017

distance between neighbours was smaller between the live fish than with the robot fish. This book provides a starting point for students and researchers studying robot fish. This field is still in its infancy hence the range of approaches to replicate fish propulsion presented in this book. However, there is

still a way to go before fish inspired propulsion will overtake the conventional propeller for normal submarine or autonomous underwater vehicle (AUV) applications. It is in the later chapters where the focus moves away from fish propulsion that there are hints of some of the potential new and exciting

technologies that will develop from continued research into bio-inspired robot fish. (Reviewed by Dr Alexander Phillips, Head of Marine Autonomous Systems Development National Oceanography Centre, Southampton)

145

08-SUT256_34(3).indd 145

28/07/17 7:16 pm

SUT Publications The SUT publishes a peer-reviewed technical journal Underwater Technology; a quarterly magazine UT2 and e–magazine UT3; a series of conference proceedings Advances in Underwater Technology, Ocean Science and Offshore Engineering and The Operation of Autonomous Underwater Vehicles; and in–house conference proceedings and collected papers from seminars. All SUT books and conference proceedings are available to purchase from the SUT website www.sut.org/publications/books-and-conference-proceedings/ This is a selection of the larger collection of the Society’s books and conference proceedings available to purchase online.

Can a Lobster be an Archaeologist? Quirky Questions and Fascinating Facts about the Underwater World From exploring lost treasure to sea monsters, ocean rubbish and how to build your own ROV, the book is packed with factual and fun illustrated stories.

Offshore Site Investigation and Geotechnics: Integrated Geotechnologies – Present and Future Proceedings of the international conference held in September 2012

Price: £220

Price: £12.99

Order Ref. C42

ISBN 978 0 906940 55 6

Hardback; 2012

Paperback; 2015

674 Pages

ISBN 978 0906940532

152 Pages

Subsea Control and Data Acquisition 2010: Future Technology, Availability and Through Life Changes Guidance Notes for the Planning and Execution of Geophysical and Geotechnical Ground Investigations for Offshore Renewable Energy Developments

Price: £15 ISBN 978 0 906940 54 9 Paperback; 2014

Proceedings of the international conference held in Newcastle, UK, 2-3 June 2010 Proceedings of the International Conference

Price: £95

2–3 June 2010 Newcastle, UK

Order Ref. C41

SUBSEA CONTROL AND DATA ACQUISITION 2010

ISBN 978 0906940525

Future Technology, Availability and Through Life Challenges

Hardback, 2010 176 Pages

48 Pages

The Operation of Autonomous Underwater Vehicles, Volume One: Recommended Code of Practice for the Operation of Autonomous Marine Vehicles, Second Edition

Price: £75 Order Ref. C40 ISBN 978 0906940518 Paperback, 2009 78 Pages

The Collaborative Autosub Science in Extreme Environments: Workshop on AUV Science in Extreme Environments Proceedings for the international science workshop held at the Scott Polar Research Institute, University of Cambridge, 11-13 April 2007

Price: £95 Order Ref. C39 ISBN 978 0906940501 Hardback, 2008 202 Pages

For orders and enquiries, please contact: Cheryl Ince, Society for Underwater Technology, Unit LG7, 1 Quality Court, London WC2A 1HR t +44 (0)20 3440 5535 e cheryl.ince@sut.org

08-SUT256_34(3).indd 146

28/07/17 7:16 pm

Can a Lobster be an Archaeologist? The Society for Underwater Technology has published an exciting, fun and informative, illustrated book for 10â&#x20AC;&#x201C;14 year olds aimed at growing interest in the wonders of underwater technology.

Can a Lobster be an Archaeologist? Quirky Questions and Fascinating Facts about the Underwater World

Written by members and friends of the Society for Underwater Technology

Can a Lobster be an Archaeologist? was released in November 2015 and has been written by past and present SUT members, as well as friends of the Society. From exploring lost treasure to sea monsters, ocean rubbish and how to build your own ROV, the book is packed with factual and fun stories brought to life by quirky illustrations by artist Rachel Hathaway. The book has been funded by the SUT, and all proceeds will go towards its agenda of supporting educational development and facilitating learning and networking opportunities. The title of the book originates from a story about an 8,000 year old settlement near the Isle of Wight, rediscovered by a lobster digging to create a burrow which was then found by divers.

To ďŹ nd out more or to order the book contact emily.boddy@sut.org 08-SUT256_34(3).indd 147

28/07/17 7:16 pm

UT2 and UT3 The magazines of the Society for Underwater Technology

UT2

Underwater Vehicles Oceanography

UT2

February March 2015

August September 2015

June July 2015 2105

Subsea Power Distribution Underwater Vehicles Sonar

T H E M A G A Z I N E O F T H E S O C I E T Y F O R U N D E R W A T EUT2 R T February E C H N O LMarch O G Y 2015

T H E M A G A Z I N E O F T H E S O C I E T Y F O R U N D E R W A T E R T E CUT2 H N April O L OMay G Y 2015

Subsea Engineering Underwater Vehicles

T H E M A G A Z I N E O F T H E S O C I E T Y F O R U N D E R W A TUT2 E R August T E C H September N O L O G Y 2015

UT2 covers a focused range of underwater subjects including offshore, marine renewables, subsea engineering, ocean resources, diving and manned submersibles, underwater science and robotics. The magazine is represented at all the many exhibitions around the world at which the Society both co-organises and attends. Furthermore, the magazine is distributed at the many subsea training courses that are organised by the Society, ensuring it reaches tomorrowâ&#x20AC;&#x2122;s engineers and technologists.

December 2012

UT3 The magazine of the Society for Underwater Technology

UT3

UT3 February March 2014

January 2013

Excavation and Trenching Underwater Intervention

UT2 December 2012 mber 2012

Communications Underwater Vehicles

T H E M A G A Z I N E O F T H E S O C I E T Y F O R U N D E R WAT E R T E C H N O LO G Y

UT3 January 2013

T H E M A G A Z I N E O F T H E S O C I E T Y F O R U N D E R W A T E RUT2 T EFebruary C H N O LMarch O G Y 2014

UT3 is the online magazine of the Society for Underwater Technology, and covers the subsea industry. It consists of the content of the print magazine UT2, greatly expanded with other information.

UT2 and UT3 are available online at http://issuu.com/ut-2_publication

08-SUT256_34(3).indd 148

28/07/17 7:16 pm

Society for Underwater Technology International multidisciplinary learned society This non-aligned membership-based organisation seeks to further the dissemination of knowledge and lessons learned in the underwater environment through networking, events and publications

Its membership covers the following activity areas:

diving and manned submersibles environmental forces marine policy marine renewable energies ocean resources offshore site investigation and geotechnics salvage and decommissioning subsea engineering and operations

For further information For membership, publications or general enquires, contact SUT Head Ofﬁce Unit LG7, 1 Quality Court, London WC2A 1HR t +44 (0)20 3440 5535 e info@sut.org For events, contact SUT Aberdeen Ofﬁce Enerprise Centre Exploration Drive Bridge of Don Aberdeen AB23 8GX UK t +44 (0)1224 823 637 e events@sut.org

underwater robotics underwater science

www.sut.org

underwater vehicles

09C-SUT-34(3)-IBC.indd 1

28/07/17 7:15 pm

5-8 SEPT 2017

Find SOluTiOnS TO all yOur OFFShOrE TEChnOlOgy and buSinESS nEEdS • 56,000 attendees from 100+ countries • 1,500+ exhibitors offering live demos, consultations and interactive sessions • 20 International pavilions • Free to attend technical conference and keynotes: discover game-changing technologies and industry developments

“SPE Offshore Europe is an important event to be at and see the latest technology and thinking in our industry.” ChiEF OPEraTing OFFiCEr, XCiTE EnErgy rESOurCES

• MyEvent online planner and networking opportunities on the day: it is now even easier to make the connections that matter • The only industry event worth attending: 13,000 of our visitors don’t attend any other exhibition

NEw for 2017 Organised by