摘要
根据柴油机表面振动信号非平稳时变的特性 ,采用小波对柴油机缸体表面的振动信号进行分解得到特征信息量 ,并以此建立 AR模型 ,计算得到 K L信息量。提出了基于小波 K L信息量的模式识别方法 ,实现故障诊断技术的量化。对柴油机活塞和缸套间的磨损状态进行了故障诊断实例分析 ,诊断结果说明基于小波 K L信息量的模式识别方法对柴油机的故障诊断具有很好的实用性。
According to non-stationary, time-varying characteristics of a diesel engine surface vibration signal, this work made use of wavelet to decompose the diesel engine surface vibration signal and obtained the characteristic information based on which the AR model was built and the KL information quantum was computed. The method of pattern recognition is presented based on wavelet KL information quantum, which makes the fault diagnosis technology get qualification. Taking the diesel engine piston-linear wear as an example of fault diagnosis, the method of diesel engine fault diagnosis based on KL information quantum was put forward. The diagnosis result proves that the pattern recognition method based on wavelet KL information quantum has a practicability in diesel engine fault diagnosis.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2004年第2期138-141,共4页
Transactions of the Chinese Society for Agricultural Machinery
