基于LCD基本尺度熵的齿轮故障RVM识别

GEAR FAULT IDENTIFICATION OF RVM BASED ON LCD BASE-SCALE ENTROPY

针对齿轮故障振动信号非线性、非平稳性等特点,以及其故障特征提取较为困难的实际,提出了基于LCD基本尺度熵的齿轮故障特征提取方法。该方法利用局部特征尺度分解(LCD)对齿轮振动信号进行自适应分解,获取原始信号不同尺度分量;根据基本尺度熵能有效区分不同故障信号的复杂度,计算LCD分解所得内禀尺度分量(ISC)基本尺度熵,获得原始信号多个尺度的复杂度特征作为齿轮不同故障下的特征参数;将该特征参数输入相关向量机(RVM)分类器中判断齿轮故障,实现故障诊断。齿轮故障诊断实验结果表明,所提方法能够有效地识别齿轮的典型故障,相比其他一些方法,具有一定的优势。

Aiming at the fact that the gear vibration signal would exactly display non-stationary characteristics and fault features is hard to extracted,a fault extraction method of gear based on multiscale base-scale entropy of LCD was proposed.The vibration signal was decomposed adaptively with local characteristic-scale decomposition(LCD)to obtain the components in different scales of the original signal.Considering the ability of the base-scale entropy in distinguishing the complexity of different signals effectively,the base-scale entropy of intrinsic scale components(ISCs)by LCD was calculated.Thus the complexity metric in different scales of the original signal was gained,which was consequently taken as the feature parameter to describe different gear states.The feature parameters were then put into relevance vector machine(RVM)for diagnosing the gear faults.Experiment results of gear show that the proposed method can classify typical fault of gear exactly and has certain superiority when compared with some other methods.