论文成果
A dynamic bayesian network based methodology for fault diagnosis of subsea control module hydraulic system
摘要:A subsea control module (SCM) hydraulic system is an important element of subsea control system. Fault diagnosis of SCM hydraulic system is challenged due to the complex system structure. To identify the faulty components and distinguish the fault types, this study develops a dynamic Bayesian networks (DBN)-based fault diagnosis methodology of SCM hydraulic system. The methodology generates evidence from SCM internal sensors and uses the information to update the process knowledge. The dynamic degradation process of the valve is simulated and the conversion relationship between time slices are determined by Markov models. Based on EM algorithm, the probability parameters of BN nodes are calculated, and transfer probability distribution between time slices is determined. A multi-time slices DBN fault diagnosis model of SCM hydraulic system based on reverse analysis was established. Thirty-four fault diagnosis cases including high-pressure and low-pressure hydraulic system of SCM are investigated to illustrate the methodology. The results show that the posterior probability of all cases has changed from 10 % to more than 50 % when failure occurs, and DBN model can correctly diagnose the faults that occurred, with an accuracy rate of 100 %, and failure rate of DCV valve is related to the hydraulic oil circuit flow and pressure. The fault diagnosis cases validate the accuracy and effectiveness of the proposed methodology. ? 2025 The Institution of Chemical Engineers
ISSN号:0957-5820
卷、期、页:卷202
发表日期:2025-10-01
影响因子:0.000000
期刊分区(SCI为中科院分区):二区
收录情况:SCI(科学引文索引印刷版),EI(工程索引),SCIE(科学引文索引网络版)
发表期刊名称:Process Safety and Environmental Protection
参与作者:Yang, Xuguang,Estefen, Segen F.
通讯作者:同武军,祝鸿山,田璐冉
第一作者:王莹莹
论文类型:期刊论文
论文概要:王莹莹,Yang, Xuguang,同武军,祝鸿山,田璐冉,Estefen, Segen F.,A dynamic bayesian network based methodology for fault diagnosis of subsea control module hydraulic system,Process Safety and Environmental Protection,2025,卷202
论文题目:A dynamic bayesian network based methodology for fault diagnosis of subsea control module hydraulic system
分享到:


