基于贝叶斯网络的催化燃烧式瓦斯传感器故障诊断  

Fault Diagnosis of Catalytic Combustion Type Gas Sensor Based on Bayesian Network

作　　者：张新建 张玉芝 李贤功[2] Zhang Xinjian;Zhang Yuzhi;Li Xiangong(Chensilou Coal Mine,Yongmei Company,Henan Energy and Chemical Group,Yongcheng 476600,China;School of Mining Engineering,China University of Mining and Technology,Xuzhou 221000,China)

机构地区：[1]河南能源化工集团永煤公司陈四楼煤矿,河南永城476600 [2]中国矿业大学矿业工程学院,江苏徐州221000

出　　处：《煤矿机械》2020年第5期168-170,共3页Coal Mine Machinery

基　　金：国家重点研发计划(2017YFC0804408)。

摘　　要：为了有效诊断瓦斯传感器故障,提出一种基于贝叶斯网络的瓦斯传感器故障诊断模型。在对瓦斯传感器故障的各种原因、事件及内在逻辑关系进行致因分析的基础上建立瓦斯传感器故障树,并依据相关规则建立贝叶斯网络模型。其次,在基于故障树确定基本事件风险率基础上,利用贝叶斯网络模型找到最为关键的瓦斯传感器关键故障因素。最后,选取开采工作面回风隅角催化燃烧式瓦斯传感器进行故障诊断案例分析。结果表明:该瓦斯传感器故障诊断结果为存在事故风险,该结果与工程实际较为一致。同时提出了对电源及开关、催化剂有效性等进行着重检修与排查,对人为因素、其他技术因素、环境因素进行控制的建议。In order to effectively diagnose the fault of gas sensor, a fault diagnosis model of gas sensor based on Bayesian network was proposed. The fault tree of gas sensor was established on the basis of cause analysis of various causes, events and internal logic relations of gas sensor fault, and the Bayesian network model was established according to relevant rules. Secondly, on the basis of determining the basic event risk rate based on fault tree, the key fault factors of gas sensor were found by Bayesian network model. At last, the case of fault diagnosis of catalytic combustion type gas sensor in the upper corner of mining face was analyzed. The results show that the fault diagnosis result of the gas sensor is accident risk, which is more consistent with the engineering practice. At the same time,put forward the suggestions of focusing on the maintenance and troubleshooting of power supply, switch,catalyst effectiveness, etc., as well as the control of human factors, other technical factors and environmental factors.

关 键 词：贝叶斯网络 瓦斯传感器 故障树 催化燃烧式 故障诊断 

分 类 号：TD712[矿业工程—矿井通风与安全]