基于EEMD和GA-BP的列车辅助逆变器开路故障诊断研究

地铁列车随着运营年限增加,辅助逆变器故障频发,严重制约轨道交通安全运营,故障及时诊断至关重要。针对列车辅助逆变器典型的大功率器件开路故障,提出了一种基于集合经验模态分解(EEMD)和遗传算法优化的BP神经网络(GA-BP)的故障诊断方法,该方法以辅助逆变器输出的三相半波电压值为监测信号,通过EEMD分解采用能量矩的方法提取故障特征向量,基于GA-BP神经网络实现故障智能诊断。仿真实验结果表明该方法故障诊断准确率能达到95.5%。

Research on Open-circuit Fault Diagnosis of Train Auxiliary Inverter Based on EEMD and GA-BP

As the operation years of metro vehicles increase, the malfunction of auxiliary inverter occurs frequently, which seriously restricts the safe operation of rail transit, and timely diagnosis of faults is essential. Aming at the typical high power device open-circuit fault for train auxiliary inverter, a fault diagnosis method based on ensemble empirical mode decomposition (EEMD) and genetic algorithm optimized BP neural network (GA-BP) is proposed. The method uses the inverter output three-phase half-wave voltage value as the monitoring signal, and extracts the fault feature vector by means of energy moments through EEMD decomposition. The fault intelligent diagnosis is realized based on GA-BP neural network. Simulation results show that the fault diagnosis accuracy of this method can reach 95.5%.