基于小波包Tsallis奇异熵的磁浮列车辅助逆变器故障诊断研究

Research on Fault Diagnosis of Auxiliary Inverter in Maglev Train Based on Wavelet Packet Tsallis Singular Entropy

针对中低速磁浮列车辅助逆变器的绝缘栅双极性晶体管(insulated gate bipolar transistor,IGBT)开路故障,提出了一种基于小波包Tsallis奇异熵的故障诊断方法。首先,利用快速傅里叶算法对辅助变流器输出电流故障前后的频率成分进行分析,获取故障的基本特征;其次,对输出电流信号进行小波包分解,并对各个节点进行重构,计算相应的小波包Tsallis奇异熵,获取可用于诊断的故障特征;最后,通过计算故障电流的直流分量来区分上下桥臂的故障,实现故障的定位。在MATLAB/Simulink中搭建了相应的仿真模型,对单个IGBT故障以及两个IGBT故障的情况进行了仿真。结果表明,所提算法可以有效地对故障IGBT进行检测与定位。

Aiming at the open-circuit fault of insulated gate bipolar transistor(IGBT) of the auxiliary inverter of medium and low speed maglev train,a fault diagnosis method based on wavelet packet Tsallis singular entropy was proposed.The whole idea is as follows:firstly,use the fast Fourier algorithm to analyze the frequency components of the auxiliary converter output current before and after the fault to obtain the basic characteristics of the fault;secondly,perform wavelet packet decomposition on the output current signal,and reconstruct each node,and calculate the corresponding wavelet packet Tsallis singular entropy to obtain the fault characteristics that can be used for diagnosis;finally,by calculating the DC component of the fault current,distinguish the faults of the upper and lower bridge arms,and realize the fault localization.The corresponding simulation models were built in MATLAB/Simulink to simulate a single IGBT failure and two IGBT failures.The results show that the proposed algorithm can effectively detect and locate the faulty IGBT.