小波包分析和最小二乘支持向量机的电机故障诊断

Motor Fault Diagnosis Based on Least Squares Support Vector Machine and Wavelet Packet Analysis

为了提高电机故障诊断的准确性,提出一种小波包分析和最小二乘支持向量机的电机故障诊断模型。首先,采集电机不同故障状态下输出信号,采用小波包对信号进行分解提取能量特征,然后,能量特征输入最小二乘支持向量机中进行训练,建立故障状态分类器,最后,采用仿真实验对模型的性能进行测试。实验结果表明,相对于其它电机故障诊断模型,本文模型获得更优的电机故障诊断结果,具有更高的实际应用价值。

In order to improve the accuracy of motor fault diagnosis, this paper proposed a motor fault diagnosis model based on least squares support vector machine and wavelet packet analysis. Firstly, output signals of motor fault in different types are collected, and wavelet packet decomposition is used to extract energy features of signal. Secondly, the energy features are input to least squares support vector machine to establish the fault state classifier. Finally the simulation experiment is carried out to test the performance of the model.The experimental results show that, compared with other motor fault diagnosis models, the proposed model can obtain better results of the motor fault diagnosis, and it has higher practical application value.