往复压缩机故障诊断技术中若干问题研究  

Study on Several Solutions of Failures Diagnoses Technology for Reciprocating Compressor

作　　者：刘卫华[1] 昂海松[1] 

机构地区：[1]南京航空航天大学,江苏南京210016

出　　处：《流体机械》2003年第2期27-30,共4页Fluid Machinery

摘　　要：介绍了往复压缩机故障诊断技术中的若干问题 ,包括 :故障模式与诊断方法、数学模型与故障样本、分级诊断策略与人工神经网络、智能诊断系统构造等的研究结果。提出应根据故障模式来选择诊断方法 ;对于多级压缩机应采用两级诊断方案等观点。文中还简述了基于人工神经网络的智能诊断系统构造 。Several solutions of failure diagnosis technology for reciprocating compressor are introduced, which include failure modes and diagnosing methods, mathematical model and failure samples, diagnosis with strategy of classification and ANN and structure of intellectual diagnostic system. The conclusions that corresponding diagnosis methods should be adopted according to different failure modes are proffered; The method of two step diagnosis should be adopted for the multistage compressor. Artificial intellectual neural network diagnostic system is intrduced and general chart is given.

关 键 词：往复压缩机 故障论断 数学模型 智能诊断 分级诊断 人工神经网络 

分 类 号：TH45[机械工程—机械制造及自动化] TP277[自动化与计算机技术—检测技术与自动化装置]