小波阈值算法与包络谱分析结合的电机故障诊断方法∗

摘要

针对采集的电机故障信号中噪声干扰的问题，提出一种基于贝叶斯估计的小波收缩新阈值与包络谱分析结合的电机故障诊断方法。新阈值的选取考虑了故障信号经小波变换后在不同尺度上的去噪特性，更符合噪声在各层中的分布情况；改进阈值函数对故障信号进行降噪处理，并基于包络谱分析处理故障信号，可提取电机故障信号的特征信息。通过对仿真信号分析与实例分析，结果表明该方法能够有效地降低噪声干扰并识别出电机故障类型。%According to the problem of noise interference in the motor fault signal acquisition, a novel motor fault diagnosis method of wavelet shrinkage threshold based on Bayesian estimation combined with envelope spectrum was proposed. The fault signal denoising characteristics of different scales were considered in proposed method. The new threshold was suitable for the situation of noise distribution. Noise reduction could be gotten by improving the threshold function,and processing the fault signal based on the envelope spectrum analysis. The motor fault signal feature information could be extracted. The results of analysis applied to simulated signal and the measured signal showed that the noise interference could be reduced effectively and the motor fault type could be identified.