小波包分解在电机故障诊断中的应用

针对常用的时域和频域分析在诊断电机故障时存在不能同时诊断出故障时间和类型的问题，在分析电机故障特征的基础上，利用时频两域都具有表征信号特征能力的小波，对采集来的电机振动信号进行小波包分解，利用分解的小波系数，在各个频段上进行小波信号重构，并计算信号各个频段的能量特征值，提取故障特征，诊断故障发生的时间和故障类型。经仿真验证，小波包分解能将故障信号有效划分到不同的频段内，而且时域和频域局部化特性好，能有效地诊断出电机故障，具有良好的理论意义与工程应用价值。

Application of Wavelet Packet Decomposition in Motor Fault Diagnosis

As to the bottleneck question to diagnose the time and type of motor fault at the same time existing in traditional time domain and frequency domain, a kind of method of the fault diagnosis of motor based on wavelet packet decompositon is proposed in this paper. On basis of analyzing the motor fault characters, the vibration signal is decomposed by wavelet packet which indicates signal character both in time domain and frequency domain, the wavelet coefficient is obtained and energy eigenvalue in each frequency branch is calculated so as to diagnose the fault time and type. Proved by experiment that wavelet packet can effectively divide signal into different frequency domains and diagnose the motor fault, which has good theory meaning and project application value.