基于时序分析和K-L信息距离的柴油机气阀机构故障诊断

柴油机表面振动信号包含着设备运行的大量特征信息,是一种典型的非平稳时变信号,因此信号分析方法是柴油机故障诊断技术的难点。作者通过模拟柴油机气阀机构的气阀漏气和气门间隙异常两种主要故障,采集气缸缸盖表面的振动信号,应用时间序列分析方法建立缸盖振动信号的AR模型,然后利用欧氏距离判别函数进行故障状态识别。结果表明该方法是可行的,诊断效果比较好,同时对往复机械设备的在线实时监测与故障诊断做了有益研究。

Fault Diagnosis of Diesel Engine Valve Train by Using Time-series Analysis and Kullback-Leiber Information Distance

Vibration signals of diesel have much feature information of equipment runing,and vibration signals of diesel are typical unstable time-series signals,so the method of signal analysis is difficulty of diesel fault diagnosis technology.By simulating the two main faults of the valve train:gas leak and abnormal lash,the vibration signals of the cylinder head have been measured.Based on the time-series analysis method,the AR models of cylinder head vibration signals are set up.At last,Kullback-Leiber information distance is used to diagnose the valve train faults.The results show that it is feasible and effective to diagnose the valve train faults by using time-series analysis method and Euclid distance.In this paper,the on-line monitoring and fault diagnosis to mechanism is carried out.