基于融合模糊推理的关键设备运行安全评价

Running Safety Performance Evaluation for Key Equipment Based on Multi-information Fusion and Fuzzy Inference Method

针对生产中关键设备的运行安全问题,提出一种基于多元信息模糊推理的安全性能评价方法。评价方法从设备的运行状态中提取出多维故障敏感特征,应用主成分分析方法进行特征融合;利用故障特征的模糊不确定性进行模糊化处理,构造模糊推理的mass函数,应用推理准则对融合的故障特征进行安全性能评价。并以工业现场往复压缩机为例,进行安全评价测试,验证了基于多元信息融合的安全性能评价方法的有效性。

In view of the running seeurity issues of key equipment in produetion,a method of safety performance evaluation based on muhi-information fusion and fuzzy inference method is proposed. Multi-dimensional fault sensitive eharaeteristies are extracted from the running conditions,using principal component analysis to make fusion of the characteristics. Aeeording to the fuzzy uncertainty of the fault eharaeteristies,fuzzy processing is done to the charac- teristics and then mass functions in DS evidence theory is built and DS inference rule is used to get the safety evaluation. Reciprocating compressor industrial example verified the affeetivity of the proposed safety performance evaluation method.