| The hydraulic system is an indispensable part of the subsea control module(SCM),which is mainly used to control the opening and closing of hydraulic output and return passages,so as to control the opening and closing of equipment such as trees and manifolds.Once a fault occurs,it will cause incalculable consequences such as paralysis and production stoppage of the SCM system.It is of great engineering significance to carry out the research on the fault diagnosis method of the SCM hydraulic system to ensure its health management in the long-term service process.Aiming at the first domestic compound electro-hydraulic control system in a gas field in Bohai Sea,combined with the fault characteristics and failure mechanism impact analysis of the hydraulic control system,this paper constructs the common fault failure table of each valve in the hydraulic system.Based on the wavelet transform method,the data preprocessing process such as noise reduction is carried out on the field collected test sensor signals.The process is simulated and analyzed by MATLAB,and the Bayesian network(BN)fault diagnosis model of SCM hydraulic system is established by using the processed data.The probability parameters of each node of BN are determined by maximum likelihood algorithm,the transition probability distribution between time slices is determined according to expert experience,the static BN is dynamically extended in the direction of time series modeling,and a dynamic Bayesian network(DBN)fault diagnosis model of SCM hydraulic system with multiple time slices is established.Finally,the posterior probability of fault nodes in SCM hydraulic system is calculated by DBN model,and the diagnosis results of fault examples are obtained.Through analysis,the order of parameter sensitivity of faulty nodes in different fault cases is obtained.It can provide certain theoretical support and engineering reference for the fault diagnosis research of the subsea control module in the long-term service. |
PDF Full Text Request