基于分数阶傅里叶变换的无刷直流电机逆变器故障诊断技术研究

Brushless DC Motor Inverter Fault Diagnosis Technology Research Based on Fractional Fourier Transform

研究了基于分数阶傅里叶变换(FRFT)的无刷直流电机逆变器开路故障诊断技术。首先,对待测电路的输出电流信号进行分数阶信息分析;其次以类内类间距离为判据,寻取最优分数阶傅里叶变换p值,并提取故障特征;最后,采用基于支持向量机方法进行故障诊断。另外,对几种常用的信号处理方法得到的结果进行比较,仿真结果证明,基于分数阶傅里叶变换的故障诊断技术是可行的。

This paper studies the open-circuit fault diagnosis technology of Brushless dc motor（ BLDCM） inverter. Firstly,the output current signal of the circuit which has yet to be tested is analyzed based on fractional Fourier transform（ FRFT） .Secondly, the best FRFT value of p and the fault features are obtained based on internal and interval distance criterion. Final y,the support vector machine（ SVM） is utilized for identifying the fault. In addition,the results based on several frequently-used signal processing methods are compared in this paper. The results of the simulation prove that the fault diagnosis technology based on fractional Fourier transform is feasible.