基于小波包分析的电动汽车直流充电桩功率器件开路故障诊断研究

Research on Fault Diagnosis of DC Charger Power Device Based on Wavelet Packet Analysis

针对电动汽车直流充电桩中的PWM整流器模块,提出了一种基于小波包分析的功率器件实时故障诊断方法。对直流充电桩的直流母线电压信号进行4层小波包分解与重构,利用小波包重构系数计算出频带能量并归一化,比较对应频段的功率谱,得到不同故障状态下的集中频带、特征频率和功率谱数据。通过建立基于双闭环SVPWM控制的PWM整流器模型,进行仿真验证,得到了不同故障状态下的集中频带和特征频率数据,并将集中频带和特征频率作为故障诊断的依据,最终实现了对直流充电桩功率器件开路故障诊断。仿真结果验证了该故障诊断方法的有效性和正确性。

Aiming at the PWM rectifier module in DC charger,a real-time fault diagnosis method of DC charger for electric vehicle based on wavelet packet analysis was proposed. The method decomposes and reconstructs the DC bus voltage signal of DC charger by four-layer wavelet packet. The wavelet packet reconstruction coefficient is used to calculate and normalize the frequency band energy. The power spectrum of the corresponding frequency band is compared to obtain the concentration frequency band,eigenfrequency and power spectrum data in different fault conditions. The PWM rectifier model based on double closed-loop SVPWM control was established. By simulation,the concentration frequency bands and eigenfrequency data of different fault conditions were obtained.The concentration frequency band and eigenfrequency are taken as the basis for fault diagnosis,and it eventually realizes the DC charger power device open circuit fault diagnosis. The simulation results verify the validity and correctness of the fault diagnosis method.