噪声与振动控制2019，Vol.39Issue(4)：12-19,8.DOI:10.3969/j.issn.1006-1355.2019.04.003
利用故障因果信息的汽轮机故障智能诊断研究
Study on Intelligent Fault Diagnosis of Steam Turbines using Fault Causality Information
摘要
Abstract
Due to the high-level coupling of the turbine structure, some failure modes have similar characteristics in the form of shafting vibration and are difficult to distinguish. In order to diagnose the failure modes accurately, this paper proposes a method which integrates the fault cause information into the diagnosis model so as to realize the fault causal chain reasoning. First of all, based on FTA and FMEA analyses, the fault cause and effect nets (FCEN) are defined to describe the diagnostic knowledge. Then, using the Leaky Noisy-Or/And model, the FCEN is transformed into the Bayesian network (BN) model, and the physical meaning of the uncertainty relationship in the model is analyzed. In the two cases of the rub-impact fault, the model reasoning is performed repeatedly according to the troubleshooting results to obtain a more accurate fault causal chain. This paper innovatively proposes a networked knowledge expression form and the corresponding intelligent process. It provides a new solution for the diagnosis of steam turbines.关键词
振动与波/汽轮机/智能诊断/因果信息/FTA/FMEA/贝叶斯网络Key words
vibration and wave/ steam turbine/ intelligent diagnosis/ causal information/ FTA/ FMEA/ Bayesian network分类
信息技术与安全科学引用本文复制引用
顾煜炯,杨楠,陈东超,宋磊..利用故障因果信息的汽轮机故障智能诊断研究[J].噪声与振动控制,2019,39(4):12-19,8.基金项目
国家重点研发计划资助项目(2017YFB0603904-4) （2017YFB0603904-4）
中央高校基本科研业务费专项资金资助项目(2016XS35) （2016XS35）
中国科学院太空应用重点实验室开放基金资助项目(LSU-2016-04-02) （LSU-2016-04-02）
