特征提取在工况监测与故障诊断中的应用

随着信息技术和计算机网络技术、虚拟仪器等在企业的普及和应用，基于网络技术的未来企业全球制造化模式的建立，使日趋大型化、复杂化集成化生产设备的远程工况监视，远程故障诊断成为重要的研究课题。生产现场工况信息的自动检测、数据采集和存储是故障诊断系统的基础和起点，而对检测信息进行加工、变换和特征分析，提取敏感的故障征兆是诊断系统实现故障可靠诊断的关键之一。阐述了机械故障诊断的理论方法及特点，分析了信号的特征分析方法——时域分析方法，频域分析方法，联合时—频分析方法等多种信号处理方法的特征分析特点及其在机械工况监到与故障诊断分析过程中的工程应用。简要介绍了工况监测中特殊的时变性随机机械故障信号的分析与处理方法。

Application on Extracting Feature in State Monitoring and Fault Diagnosis

With the development and application of information and internet and virtual instrument technology ,the virtual globular company based on internet arises .To study remote state monitoring, remote fault detection and diagnosis about large scale, complicated and integrative equipment become very important. In the whole fault diagnosis system , the detecting ,data acquisition is original ,processing; transform and extracting features with the signal detected is a key factor. The theories and methods used in mechanical fault diagnosis is stated. The application of signal process and its feature extracting methods is introduced which are time domain, frequency domain and time frequency domain analysis, in state monitoring and fault diagnosis with its signal analysis. The processing method of random time variant special signal is given also.