Browsing by Author "Shafiee, Mahmood"
Now showing 1 - 20 of 38
Results Per Page
Sort Options
Item Open Access Adopting machine learning and condition monitoring P-F curves in determining and prioritizing high-value assets for life extension(Elsevier, 2021-03-13) Ochella, Sunday; Shafiee, Mahmood; Sansom, Christopher L.Many machine learning algorithms and models have been proposed in the literature for predicting the remaining useful life (RUL) of systems and components that are subject to condition monitoring (CM). However, in cases where data is ubiquitous, identifying the most suitable equipment for life-extension based on CM data and RUL predictions is a rather challenging task. This paper proposes a technique for determining and prioritizing high-value assets for life-extension treatments when they reach the end of their useful life. The technique exploits the use of key concepts in machine learning (such as data mining and k-means clustering) in combination with an important tool from reliability-centered maintenance (RCM) called the potential-failure (P-F) curve. The RCM process identifies essential equipment within a plant which are worth monitoring, and then derives the P-F curves for equipment using CM and operational data. Afterwards, a new index called the potential failure interval factor (PFIF) is calculated for each equipment or unit, serving as a health indicator. Subsequently, the units are grouped in two ways: (i) a regression model in combination with suitably defined PFIF window boundaries, (ii) a k-means clustering algorithm based on equipment with similar data features. The most suitable equipment for life-extension are identified in groups in order to aid in planning, decision-making and deployment of maintenance resources. Finally, the technique is empirically tested on NASA’s Commercial Modular Aero-Propulsion System Simulation datasets and the results are discussed in detail.Item Open Access Advanced data-driven methods for prognostics and life extension of assets using condition monitoring and sensor data.(Cranfield University, 2021-12) Ochella, Sunday Moses; Sansom, Christopher L.; Shafiee, MahmoodA considerable number of engineering assets are fast reaching and operating beyond their orignal design lives. This is the case across various industrial sectors, including oil and gas, wind energy, nuclear energy, etc. Another interesting evolution is the on-going advancement in cyber-physical systems (CPS), where assets within an industrial plant are now interconnected. Consequently, conventional ways of progressing engineering assets beyond their original design lives would need to change. This is the fundamental research gap that this PhD sets out to address. Due to the complexity of CPS assets, modelling their failure cannot be simplistically or analytically achieved as was the case with older assets. This research is a completely novel attempt at using advanced analytics techniques to address the core aspects of asset life extension (LE). The obvious challenge in a system with several pieces of disparate equipment under condition monitoring is how to identify those that need attention and prioritise them. To address this gap, a technique which combined machine learning algorithms and practices from reliability-centered maintenance was developed, along with the use of a novel health condition index called the potential failure interval factor (PFIF). The PFIF was shown to be a good indicator of asset health states, thus enabling the categorisation of equipment as “healthy”, “good ” or “soon-to-fail”. LE strategies were then devoted to the vulnerable group labelled “good – monitor” and “soon-to-fail”. Furthermore, a class of artificial intelligence (AI) algorithms known as Bayesian Neural Networks (BNNs) were used in predicting the remaining useful life (RUL) for the vulnerable assets. The novelty in this was the implicit modelling of the aleatoric and epistemic uncertainties in the RUL prediction, thus yielding interpretable predictions that were useful for LE decision-making. An advanced analytics approach to LE decision-making was then proposed, with the novelty of implementing LE as an on-going series of activities, similar to operation and maintenance (O&M). LE strategies would therefore be implemented at the system, sub-system or component level, meshing seamlessly with O&M, albeit with the clear goal of extending the useful life of the overall asset. The research findings buttress the need for a paradigm shift, from conventional ways of implementing LE in the form of a project at the end of design life, to a more systematic approach based on advanced analytics.Item Open Access Advanced reliability analysis of complex offshore Energy systems subject to condition based maintenance.(Cranfield University, 2021-04) Elusakin, Tobiloba; Simms, Nigel J.; Shafiee, MahmoodAs the demand for energy in our world today continues to increase and conventional reserves become less available, energy companies find themselves moving further offshore and into more remote locations for the promise of higher recoverable reserves. This has been accompanied by increased technical, safety and economic risks as the unpredictable and dynamic conditions provide a challenge for the reliable and safe operation of both oil and gas (O&G) and offshore wind energy assets. Condition-based maintenance (CBM) is growing in popularity and application in offshore energy production, and its integration into the reliability analysis process allows for more accurate representation of system performance. Advanced reliability analysis while taking condition-based maintenance (CBM) into account can be employed by researchers and practitioners to develop a better understanding of complex system behaviour in order to improve reliability allocation as well as operation and maintenance (O&M). The aim of this study is therefore to develop models for reliability analysis which take into account dynamic offshore conditions as well as condition-based maintenance (CBM) for improved reliability and O&M. To achieve this aim, models based on the stochastic petri net (SPN) and dynamic Bayesian network (DBN) techniques are developed to analyse the reliability and optimise the O&M of complex offshore energy assets. These models are built to take into account the non-binary nature, maintenance regime and repairability of most offshore energy systems. The models are then tested using benchmark case studies such as a subsea blowout preventer, a floating offshore wind turbine (FOWT), an offshore wind turbine (OWT) gearbox and an OWT monopile. Results from these analyses reveal that the incorporation of degradation and CBM can indeed be done and significantly influence the reliability analysis and O&M planning of offshore energy assets.Item Open Access Advanced structural health monitoring strategies for condition-based maintenance planning of offshore wind turbine support structures(2019-04) Martinez Luengo, Maria; Shafiee, Mahmood; Kolios, Athanasios; Engineering and Physical Sciences (EPSRC)Condition-based maintenance strategies need to be adopted as distance-to-shore and water depth increase in the offshore wind industry. The aim of the research presented herein is to develop advance structural health monitoring strategies that enhance the condition-based maintenance of offshore wind turbine support structures. The focus is on the selection of technologies, the implementation process, the analysis of the asset’s structural response under complex loading, the economic justification for structural health monitoring implementation and the effective structural health monitoring data analysis. Research activities consist of the provision of a comprehensive study for structural health monitoring technologies’ utilisation in the offshore wind industry. This is followed by parametric structural modelling, simulation and validation of an operational offshore wind turbine tower, support structure and soil-structure interaction, using commercial software. The evaluation of the asset’s response under complex loading subject to design changes and failure mechanisms is also undertaken. A combination of existing and newly developed methodologies is deployed for the effective data management of structural health monitoring systems and validated with industrial data for the case of strain monitoring. These include unsupervised learning algorithms (neural networks), deterministic and probabilistic methods for noise cleansing and missing data imputation. Guidelines for the structural health monitoring implementation from design stage of a wind farm are proposed and applied to a baseline scenario. This is utilised to assess the economic impact that structural health monitoring has in the lifecycle of the assets. The achieved results show that the implementation of structural health monitoring in offshore wind turbine following the Statistical Pattern Recognition paradigm and the proposed guidelines has the potential to reduce the Operational Expenditure. This reduction is much greater than the cost associated with the implementation of these systems. Monitoring from the commissioning of the assets is crucial for the system’s calibration and establishing thresholds. The developed noise cleansing and missing data imputation methodologies can successfully be employed together to produce more complete low-disturbed datasets.Item Open Access Artificial intelligence in prognostics and health management of engineering systems(Elsevier, 2021-12-08) Ochella, Sunday; Shafiee, Mahmood; Dinmohammadi, FatemePrognostics and health management (PHM) has become a crucial aspect of the management of engineering systems and structures, where sensor hardware and decision support tools are deployed to detect anomalies, diagnose faults and predict remaining useful lifetime (RUL). Methodologies for PHM are either model-driven, data-driven or a fusion of both approaches. Data-driven approaches make extensive use of large-scale datasets collected from physical assets to identify underlying failure mechanisms and root causes. In recent years, many data-driven PHM models have been developed to evaluate system’s health conditions using artificial intelligence (AI) and machine learning (ML) algorithms applied to condition monitoring data. The field of AI is fast gaining acceptance in various areas of applications such as robotics, autonomous vehicles and smart devices. With advancements in the use of AI technologies in Industry 4.0, where systems consist of multiple interconnected components in a cyber–physical space, there is increasing pressure on industries to move towards more predictive and proactive maintenance practices. In this paper, a thorough state-of-the-art review of the AI techniques adopted for PHM of engineering systems is conducted. Furthermore, given that the future of inspection and maintenance will be predominantly AI-driven, the paper discusses the soft issues relating to manpower, cyber-security, standards and regulations under such a regime. The review concludes that the current systems and methodologies for maintenance will inevitably become incompatible with future designs and systems; as such, continued research into AI-driven prognostics systems is expedient as it offers the best promise of bridging the potential gap.Item Open Access A Bayesian network model for the probabilistic safety assessment of offshore wind decommissioning(SAGE, 2022-09-02) Shafiee, Mahmood; Adedipe, TosinWith increasing the number of wind turbines approaching the end of their service life, it has become crucial for businesses to understand and assess safety and security issues related to the decommissioning phase of wind farm asset lifecycle. This paper aims to develop, for the first time, a Bayesian Network (BN) model for the safety assessment of offshore wind farm decommissioning operations. The most critical safety incidents are identified and their corresponding risk-influencing factors (RIF) are determined. The impacts of human errors as well as procedural and mechanical/electrical failures on the safety and efficiency of decommissioning operations are thoroughly analysed. The findings of the study revealed that the most critical RIFs during offshore wind decommissioning operations include: visibility, crew fatigue, number of personnel per operation, proper safety procedures, crane integrity, number of lifts available in the wind farm, inspection frequency, as well as equipment design.Item Open Access Bayesian network modelling for the wind energy industry: an overview(Elsevier, 2020-05-29) Adedipe, Tosin; Shafiee, Mahmood; Zio, EnricoWind energy farms are moving into deeper and more remote waters to benefit from availability of more space for the installation of wind turbines as well as higher wind speed for the production of electricity. Wind farm asset managers must ensure availability of adequate power supply as well as reliability of wind turbines throughout their lifetime. The environmental conditions in deep waters often change very rapidly, and therefore the performance metrics used in different life cycle phases of a wind energy project will need to be updated on a frequent basis so as to ensure that the wind energy systems operate at the highest reliability. For this reason, there is a crucial need for the wind energy industry to adopt advanced computational tools/techniques that are capable of modelling the risk scenarios in near real-time as well as providing a prompt response to any emergency situation. Bayesian network (BN) is a popular probabilistic method that can be used for system reliability modelling and decision-making under uncertainty. This paper provides a systematic review and evaluation of existing research on the use of BN models in the wind energy sector. To conduct this literature review, all relevant databases from inception to date were searched, and a total of 70 sources (including journal publications, conference proceedings, PhD dissertations, industry reports, best practice documents and software user guides) which met the inclusion criteria were identified. Our review findings reveal that the applications of BNs in the wind energy industry are quite diverse, ranging from wind power and weather forecasting to risk management, fault diagnosis and prognosis, structural analysis, reliability assessment, and maintenance planning and updating. Furthermore, a number of case studies are presented to illustrate the applicability of BNs in practice. Although the paper details information applicable to the wind energy industry, the knowledge gained can be transferred to many other sectors.Item Open Access Condition assessment, remaining useful life prediction and life extension decision making for offshore oil and gas assets(Elsevier, 2017-05-04) Animah, Isaac; Shafiee, MahmoodOffshore oil and gas assets are highly complex structures comprising of several components, designed to have a lifecycle of about 20 to 30 years of working under harsh operational and environmental conditions. These assets, during their operational lifetime, are subjected to various degradation mechanisms such as corrosion, erosion, wear, creep and fatigue cracks. In order to improve economic viability and increase profitability, many operators are looking at extending the lifespan of their assets beyond the original design life, thereby making life extension (LE) an increasingly critical and highly-discussed topic in the offshore oil and gas industry. In order to manage asset aging and meet the LE requirements, offshore oil and gas operators have adopted various approaches such as following maintenance procedures as advised by the original equipment manufacturer (OEM), or using the experience and expertise of engineers and inspectors. However, performing these activities often provides very limited value addition to operators during the LE period of operation. This paper aims to propose a systematic framework to help operators meet LE requirements while optimizing their cost structure. This framework establishes an integration between three individual life assessment modules, namely: condition assessment, remaining useful life (RUL) prediction and LE decision-making. The benefits of the proposed framework are illustrated through a case study involving a three-phase separator system on a platform which was constructed in the mid-1970s in West Africa. The results of this study affirm the effectiveness of this framework in minimizing catastrophic failures during the LE phase of operations, whilst ensuring compliance to regulatory requirements.Item Open Access Data management for structural integrity assessment of offshore wind turbine support structures: data cleansing and missing data imputation(Elsevier, 2019-02-05) Martinez Luengo, Maria; Shafiee, Mahmood; Kolios, AthanasiosStructural Health Monitoring (SHM) and Condition Monitoring (CM) Systems are currently utilised to collect data from offshore wind turbines (OWTs), to enhance the accurate estimation of their operational performance. However, industry accepted practices for effectively managing the information that these systems provide have not been widely established yet. This paper presents a four-step methodological framework for the effective data management of SHM systems of OWTs and illustrates its applicability in real-time continuous data collected from three operational units, with the aim of utilising more complete and accurate datasets for fatigue life assessment of support structures. Firstly, a time-efficient synchronisation method that enables the continuous monitoring of these systems is presented, followed by a novel approach to noise cleansing and the posterior missing data imputation (MDI). By the implementation of these techniques those data-points containing excessive noise are removed from the dataset (Step 2), advanced numerical tools are employed to regenerate missing data (Step 3) and fatigue is estimated for the results of these two methodologies (Step 4). Results show that after cleansing, missing data can be imputed with an average absolute error of 2.1%, while this error is kept within the [+ 15.2%−11.0%] range in 95% of cases. Furthermore, only 0.15% of the imputed data fell outside the noise thresholds. Fatigue is found to be underestimated both, when data cleansing does not take place and when it takes place but MDI does not. This makes this novel methodology an enhancement to conventional structural integrity assessment techniques that do not employ continuous datasets in their analyses.Item Open Access Decision support methods and applications in the upstream oil and gas sector(Elsevier, 2018-10-17) Shafiee, Mahmood; Animah, Isaac; Alkali, Babakalli; Baglee, DavidDecision-making support (DMS) methods are widely used for technical, economic, social and environmental assessments within different energy sectors, including upstream oil and gas, refining and distribution, petrochemical, power generation, nuclear power, solar, biofuels, and wind. The main aim of this paper is to present a comprehensive literature review and classification framework for the latest scholarly research on the application of DMS methods in the upstream oil and gas industry. To achieve this aim, a systematic review is conducted on the current state-of-the-art and future perspectives of various DMS methods applied to different upstream operations (such as exploration, development and production) which take place prior to shipping of crude oil and natural gas to the refineries for processing. Journal and conference proceeding sources that contain literature on the subject are identified, and based on a set of inclusion criteria the related papers are selected and reviewed carefully. A framework is then proposed to classify the literature according to the year and source of publications, type of fossil fuel sources, stages of oil and gas field lifecycle, data collection techniques, decision-making methods, and geographical distribution and location of case studies. The proposed literature classification and content analysis can help upstream oil and gas industry stakeholders such as field owners, asset managers, service providers, policy makers, environmentalist, financial analyst, and regulatory agencies to gain better insight about their business activities with well-informed decision-making processes.Item Open Access Decommissioning strategy to reduce the cost and risk-driving factors in the offshore wind industry.(Cranfield University, 2021-05) Adedipe, Tosin; Hart, Phil; Shafiee, MahmoodWith the increasing number of wind turbines approaching their end of life, there has to be a decommissioning strategy in place as the removal of these assets is not as direct as reverse installation. Offshore asset decommissioning involves technical, financial, operational, safety, policy, and environmental considerations on handling offshore marine assets at their end-of-life, with phases from the planning to site clean-up and monitoring. Offshore decommissioning activities cost significantly more than onshore; thus, adequate financial and safety provisions are essential, and more research required in this area. Decommissioning projects have hitherto been performed on a small scale, but with large-scale aging structures, they must be optimised for lowered costs and risks. In terms of planning, execution and costs, there have been significant cost overruns on decommissioning projects, which are not profit-generating projects. These forecasted large-scale decommissioning activities also have associated risks. Although risk management is a well-researched area, there is limited literature on offshore wind decommissioning risk management. This research thus, applies risk management methods and strategies to develop a robust decommissioning risk framework. In addition, to improve decommissioning processes and technologies, there is a need to develop new protocols for decommissioning. This research identifies potentials for computational simulations and automations that need to be developed to identify and manage the highest cost and risk-drivers. This study seeks to close the research gap in understanding how to decrease decommissioning costs and risks. This research addresses potential opportunities in cost and risk estimation research, impact analysis and reduction frameworks that can be adapted to decommissioning activities specific to the offshore wind industry.Item Open Access Designing a framework for materials flow by integrating circular economy principles with end-of-life management strategies(MDPI, 2022-04-02) Huang, Yuan; Shafiee, Mahmood; Charnley, Fiona; Encinas-Oropesa, AdrianaCircular economy is an upward trending notion that has drawn worldwide attention of policymakers, industry administrators, environmentalist as well as academic researchers. Though there are several tools developed for monitoring the material recovery, a very few number of research have been conducted to integrate circular economy principles with end-of-life (EOL) management strategies. This paper proposes an EOL-driven circular economy framework for the management of materials flow so as to extend the lifetime of materials through improved durability as well as to provide more social, economic and environmental benefits through less material waste. A case study from the agricultural waste industry is presented in order to test the model and validate its performance. The results show that the proposed framework has a good potential for small and medium enterprises (SME) advances.Item Open Access Determination of the most suitable technology transfer strategy for wind turbines using an integrated AHP-TOPSIS decision model(MDPI, 2017-05-06) Dinmohammadi, A.; Shafiee, MahmoodThe high-speed development of industrial products and goods in the world has caused “technology” to be considered as a crucial competitive advantage for most large organizations. In recent years, developing countries have considerably tended to promote their technological and innovative capabilities through importing high-tech equipment owned and operated by developed countries. There are currently a variety of solutions to transfer a particular technology from a developed country. The selection of the most profitable technology transfer strategy is a very complex decision-making problem for technology importers as it involves different technical, environmental, social, and economic aspects. In this study, a hybrid multiple-criteria decision making (MCDM) model based on the analytic hierarchy process (AHP) and the technique for order of preference by similarity to ideal solution (TOPSIS) is proposed to evaluate and prioritise various technology transfer strategies for wind turbine systems. For this purpose, a number of criteria and sub-criteria are defined from the viewpoint of wind energy investors, wind turbine manufacturers, and wind farm operators. The relative importance of criteria and sub-criteria with respect to the ultimate goal are computed using the eigenvalue method and then, the technology transfer alternatives are ranked based on their relative closeness to the ideal solution. The model is finally applied to determine the most suitable wind turbine technology transfer strategy among four options of reverse engineering, technology skills training, turn-key contracts, and technology licensing for the renewable energy sector of Iran, and the results are compared with those obtained by classical decision-making models.Item Open Access Development of a techno-economic framework for life extension decision making of safety critical installations(Elsevier, 2016-09-22) Shafiee, Mahmood; Animah, Isaac; Simms, Nigel J.One of the major decisions in management of the industrial assets is to ensure the feasibility of life extension process for safety critical components when they reach end-of-life. Most of the existing life extension decision-making models are restricted solely to either “technical” or “economic” feasibility analyses that may lead to inaccurate results or incorrect conclusions. In this paper, a comprehensive life extension feasibility assessment framework by taking into account both the technical and economic considerations is developed. The proposed techno-economic model for life extension of safety critical elements consists of three phases: preparation, assessment, and implementation. The technical assessment part of the framework incorporates all aspects of data collection and review, screening and prioritization of safety critical elements, condition assessment, estimation of remaining useful life, and risk analysis, while the economic assessment part deals with cost-benefit analysis. The decision to qualify a safety critical element for continuous operation beyond its service life is made based on a “life extension measure (LEM)” which is calculated by combining two indexes of “equipment health condition” and “economic added-value” obtained respectively from the technical and economic assessments. The model is applied to support the life extension decision-making procedure for water deluge systems in offshore oil installations. The results of the study show that the model is highly capable of assisting asset owners to evaluate the technical and economic benefits of extending the service life of components.Item Open Access An economic assessment framework for decommissioning of offshore wind farms using a cost breakdown structure(Springer, 2021-01-04) Adedipe, Tosin; Shafiee, MahmoodPurpose As wind power generation increases globally, there will be a substantial number of wind turbines that need to be decommissioned in the coming years. It is crucial for wind farm developers to design safe and cost-effective decommissioning plans and procedures for assets before they reach the end of their useful life. Adequate financial provisions for decommissioning operations are essential, not only for wind farm owners but also for national governments. Economic analysis approaches and cost estimation models therefore need to be accurate and computationally efficient. Thus, this paper aims to develop an economic assessment framework for decommissioning of offshore wind farms using a cost breakdown structure (CBS) approach. Methods In the development of the models, all the cost elements and their key influencing factors are identified from literature and expert interviews. Similar activities within the decommissioning process are aggregated to form four cost groups including: planning and regulatory approval, execution, logistics and waste management, and post-decommissioning. Some mathematical models are proposed to estimate the costs associated with decommissioning activities as well as to identify the most critical cost drivers in each activity group. The proposed models incorporate all cost parameters involved in each decommissioning phase for more robust cost assessment. Results and discussion A case study of a 500 MW baseline offshore wind farm is proposed to illustrate the models’ applicability. The results show that the removal of wind turbines and foundation structures is the most costly and lengthy stage of the decommissioning process due to many requirements involved in carrying out the operations. Although inherent uncertainties are taken into account, cost estimates can be easily updated when new information becomes available. Additionally, further decommissioning cost elements can be captured allowing for sensitivity analysis to be easily performed. Conclusions Using the CBS approach, cost drivers can be clearly identified, revealing critical areas that require attention for each unique offshore wind decommissioning project. The CBS approach promotes adequate management and optimisation of identified key cost drivers, which will enable all stakeholders involved in offshore wind farm decommissioning projects to achieve cost reduction and optimal schedule, especially for safety-critical tasks.Item Open Access Evaluation of wind resources and the effect of market price components on wind-farm income: a case study of Ørland in Norway(MDPI, 2019-10-29) Marjan, Ali; Shafiee, MahmoodThis paper aims to present a detailed analysis of the performance of a wind-farm using the wind turbine power measurement standard IEC61400-12-1 (2017). Ten minutes averaged wind data are obtained from LIDAR over the period of twelve months and it is compared with the 38 years’ data from weather station with the objective of determining the wind resources at the wind-farm. The performance of one of the wind turbines located in the wind-farm is assessed by comparing the wind power potential of the wind turbine with its actual power production. Our analysis shows that the wind farm under study is rated as ‘good’ in terms of wind power production and has wind power density of 479 W/m2. The annual wind-farm’s income is estimated based on the real-data collected from the wind turbines. The effect of price of electricity and the spot prices of Norwegian-Swedish green certificate on the income will be illustrated by means of a Monte-Carlo Simulation (MCS) approach. Our study provides a different perspective of wind resource evaluation by analyzing LIDAR measurements using Windographer and combines it with the lesser explored effects of price components on the income using statistical tools.Item Open Access Failure mode and effects analysis using a fuzzy-TOPSIS method: a case study of subsea control module(Inderscience, 2017-07-03) Kolios, Athanasios J.; Umofia, Anietie; Shafiee, MahmoodFailure mode and effects analysis (FMEA) is one of the most common reliability engineering techniques used for identifying, evaluating and mitigating the engineering risks. In this paper, the potential failure modes of a subsea control module (SCM) are identified based on industry experts' opinions and experiences. This is followed by a comprehensive component based FMEA study using the risk-priority-number (RPN) where the most critical failure modes in the SCM are revealed. A fuzzy TOPSIS-based multiple criteria decision making methodology is then proposed to analyse and prioritise the most critical failure modes identified by the FMEA study. To this aim, a distinct ten-parameter criticality model is developed and, for the first time, is applied to evaluate the risks associated with SCM failures. The results indicate that the proposed fuzzy TOPSIS model can significantly improve the performance and applicability of the conventional FMEA technique in offshore oil and gas industry.Item Open Access A Fuzzy-FMEA risk assessment approach for offshore wind turbines(Prognostics and Health Management Society, 2013-07-23) Dinmohammadi, Fateme; Shafiee, MahmoodFailure Mode and Effects Analysis (FMEA) has been extensively used by wind turbine assembly manufacturers for risk and reliability analysis. However, several limitations are associated with its implementation in offshore wind farms: (i) the failure data gathered from SCADA system is often missing or unreliable, and hence, the assessment information of the three risk factors (i.e., severity, occurrence, and fault detection) are mainly based on experts' knowledge; (ii) it is rather difficult for experts to precisely evaluate the risk factors; (iii) the relative importance among the risk factors is not taken into consideration, and hence, the results may not necessarily represent the true risk priorities; and etc. To overcome these drawbacks and improve the effectiveness of the traditional FMEA, we develop a fuzzy-FMEA approach for risk and failure mode analysis in offshore wind turbine systems. The information obtained from the experts is expressed using fuzzy linguistics terms, and a grey theory analysis is proposed to incorporate the relative importance of the risk factors into the determination of risk priority of failure modes. The proposed approach is applied to an offshore wind turbine system with sixteen mechanical, electrical and auxiliary assemblies, and the results are compared with the traditional FMEA.Item Open Access Identification of gas-liquid flow regimes using a non-intrusive Doppler ultrasonic sensor and virtual flow regime maps(Elsevier, 2019-05-17) Nnabuife, Godfrey; Pilario, Karl Ezra; Lao, Liyun; Cao, Yi; Shafiee, MahmoodThe accurate prediction of flow regimes is vital for the analysis of behaviour and operation of gas/liquid two-phase systems in industrial processes. This paper investigates the feasibility of a non-radioactive and non-intrusive method for the objective identification of two-phase gas/liquid flow regimes using a Doppler ultrasonic sensor and machine learning approaches. The experimental data is acquired from a 16.2-m long S-shaped riser, connected to a 40-m horizontal pipe with an internal diameter of 50.4 mm. The tests cover the bubbly, slug, churn and annular flow regimes. The power spectral density (PSD) method is applied to the flow modulated ultrasound signals in order to extract frequency-domain features of the two-phase flow. Principal Component Analysis (PCA) is then used to reduce the dimensionality of the data so as to enable visualisation in the form of a virtual flow regime map. Finally, a support vector machine (SVM) is deployed to develop an objective classifier in the reduced space. The classifier attained 85.7% accuracy on training samples and 84.6% accuracy on test samples. Our approach has shown the success of the ultrasound sensor, PCA-SVM, and virtual flow regime maps for objective two-phase flow regime classification on pipeline-riser systems, which is beneficial to operators in industrial practice. The use of a non-radioactive and non-intrusive sensor also makes it more favorable than other existing techniques.Item Open Access A kernel design approach to improve kernel subspace identification(IEEE, 2020-05-27) Salgado Pilario, Karl Ezra; Cao, Yi; Shafiee, MahmoodSubspace identification methods, such as canonical variate analysis (CVA), are non-iterative tools suitable for the state-space modelling of multi-input, multi-output (MIMO) processes, e.g. industrial processes, using input-output data. To learn nonlinear system behavior, kernel subspace techniques are commonly used. However, the issue of kernel design must be given more attention because the type of kernel can influence the kind of nonlinearities that the model can capture. In this paper, a new kernel design is proposed for CVA based identification, which is a mixture of a global and local kernel to enhance generalization ability and includes a mechanism to vary the influence of each process variable into the model response. During validation, model hyper-parameters were tuned using random search. The overall method is called Feature-Relevant Mixed Kernel Canonical Variate Analysis (FR-MKCVA). Using an evaporator case study, the trained FR-MKCVA models show a better fit to observed data than those of single-kernel CVA, linear CVA, and neural net models under both interpolation and extrapolation scenarios. This work provides a basis for future exploration of deep and diverse kernel designs for system identification.