Research On Reliability And Fault Diagnosis Methodology Of Subsea Blowout Preventer System

Subsea blowout preventer(BOP)is one of the most critical well control devices in offshore oil and gas drilling activities.It has difficult design and manufacturing technology,high price(up to 300 million RMB for a single device)and high reliability requirement.Once it fails,inestimable loss will be caused to the staff on the drilling platform and the ocean environment and so on.At present,only three American companies own the design and manufacturing technology of subsea BOP and they launch an blockade on the these technologies.As the exploitation fields of marine oil and gas resources are extended,emphasis on safety of marine drilling activities and protection of marine ecological environment has been increased.Therefore,research on technology and theory of subsea BOP system is attracting more and more attentions.With the support of national 863 plan project,this paper performs research on reliability and fault diagnosis of subsea BOP system and has obtained some innovative achievements that have important theoretical and practical values.The related research results have been applied to the subsea BOP for 3,000 meters deep water,which is developed by Rongsheng Machinery Manufacture Ltd.of Huabei Oilfield,Hebei.It won the only Innovation Gold Medal on the 13 th China International Petroleum & Petrochemical Technology and Equipment Exhibition.The method for constructing dynamic Bayesian network of a series-parallel system of subsea blowout preventer is presented.Reliability models of the subsea BOP stacks are developed.The influence rules of fault coverage factor and failure rates of main components on reliability and availability of the two kinds of subsea BOP stacks are discovered.Stochastic Petri nets of subsea annular BOP has been studied and developed.The influence rules of failure rates,testing rates and other factors on mean time to failure have been discovered.The method to improve reliability of the subsea annular BOP is proposed.Markov models of wellhead connector in subsea BOP stack are developed.Effects of failure rates on stable availability and mean time to failure of subsea wellhead connector are researched.Markov models of the electronic control system of subsea BOP with 3-2-1-0 and 3-2-0 input voting schemes are developed.The effects of failure rates and repair time of the electronic modules on reliability,availability,stable availability and mean time to failure of the electronic control system are researched.A novel method to construct Bayesian network based on GO models is presented,using which Bayesian network of the electronic control system has been proposed.Reliability value of the electronic control system is obtained.The stochastic Petri nets of the whole subsea BOP system are proposed.Reliability and availability of the electronic control system,mechanical system and the whole system is obtained.Sensitivity analysis of the failure rates and repair time of the main system components is performed and the weakness of influencing reliability and availability is found.A new method to construct Bayesian network based on GO-FLOW models is proposed,using which Bayesian network of closing the BOP is developed.Key factors of influencing the closing function are found out.A novel method to develop Bayesian network for fault diagnosis based on operation procedures is presented.With this method,diagnostic Bayesian network of closing subsea BOP has been developed.Real-time fault diagnosis in the closing process is realized with the developed models.In order to overcome the problems of insufficient information for fault diagnosis and improve the diagnostic accuracy,a fault diagnosis method based on Ensemble Empirical Mode Decomposition and Bayesian network is proposed.Effects of evidences on fault diagnosis results are researched and the influencing rules are discovered.The formation mechanism and characteristics of faults about subsea BOP system is researched.Diagnostic Bayesian networks of the hydraulic control function of subsea BOP are developed and uncertainty and fuzziness in fault diagnosis process is solved.Integrating the online monitoring data of subsea BOP,a fault diagnosis system has been developed and diagnostic performance is verified.Real-time fault diagnosis functions of subsea BOP system have been realized.