基于变分模态分解色散熵和SVM的轨道车辆受电弓故障诊断

RAILWAY VEHICLES PANTOGRAPH FAULT DIAGNOSIS BASED ON VARIATIONAL MODE DECOMPOSITION DISPERSION ENTROPY AND SVM

为提高轨道车辆受电弓的故障诊断精度,提出了基于变分模态分解(VMD)色散熵(DE)和支持向量机(SVM)的受电弓故障诊断方法。首先,以VMD为受电弓各运行状态信号的分解方法实现信号的多尺度自适应分解,得到若干个本征模态函数(IMF)分量。其次,以DE为特征提取方法计算各IMF分量的DE值,并组成故障特征向量。最后,以SVM为模式识别方法对故障特征进行识别,得到诊断结果。受电弓故障诊断实例验证了方法的有效性。

In order to improve the fault diagnosis accuracy of railway vehicles pantograph, a fault diagnosis method of pantograph based on variational mode decomposition(VMD) dispersion entropy(DE) and support vector machine(SVM) was proposed in this paper. Firstly, VMD was used as the fault signal decomposition method to realized pantograph signal in different states adaptively decomposition, and obtained several intrinsic mode functions(IMF). Secondly, DE was used as feature extration method to calculated each IMF DE value and obtained the fault feature vectors. Finally, SVM was used as pattern recgnition method to identified the fault features and the diagnosis results are obtained. An example of pantograph fault diagnosis shows the effectiveness of the method.