基于小波包奇异谱熵和IWOA-ELM的列车轴承故障诊断

Fault Diagnosis of Train Bearing Based on Wavelet Packet Singular Spectrum Entropy and IWOA-ELM

为了有效提高多工况下地铁列车滚动轴承故障诊断精度,基于轴承振动数据,提出一种基于小波包奇异谱熵和改进鲸鱼优化算法(IWOA)优化极限学习机(ELM)的故障诊断方法。针对轴承振动信号的非平稳性和非线性特点,采用小波包提取样本特征,使用奇异值分解提取小波包数据集的样本信息熵,获得样本特征集。其次,针对模型参数难以确定,优化速度慢且容易陷入局部最优问题,采用变异算子和混沌动态权重因子改进鲸鱼优化算法(WOA),使用IWOA优化ELM参数获得故障诊断模型。最后,使用美国凯斯西储大学的轴承故障数据验证了模型的可靠性和稳定性,在多工况下不同类型组合的300组测试样本中,模型诊断准确率为99.33%。同时与同一数据源的其他诊断模型进行对比验证模型的优越性。结果表明,基于小波包奇异谱熵和IWOA-ELM的轴承故障诊断模型诊断可靠性强、准确率高。

In order to effectively improve the fault diagnosis accuracy of rolling bearing of metro train under multiple working conditions,a fault diagnosis method based on wavelet packet singular spectral entropy and improved Whale Optimization Algorithm(IWOA)is proposed based on bearing vibration data.In view of the non-stationary and nonlinear characteristics of bearing vibration signals,wavelet packets are used to extract sample features,and singular value decomposition is used to extract sample information entropy of wavelet packet data sets to obtain sample feature sets.Secondly,as the model parameters are difficult to determine,the optimization speed is slow and easy to fall into local optimization problems,the Whale Optimization Algorithm(WOA)is improved by mutation operator and chaotic dynamic weight factor,and the fault diagnosis model is obtained by optimizing ELM parameters using IWOA.Finally,the reliability and stability of the model were verified with the bearing fault data of Case Western Reverse University.In 300 groups of test samples with different types of combinations under multiple working conditions,the diagnostic accuracy of the model was 99.33%.At the same time,the model is compared with other diagnostic models of the same data source to verify the superiority of the model.The results show that the bearing fault diagnosis model based on wavelet packet singular spectrum entropy and IWOA-ELM has high reliability and accuracy.