电机驱动系统PWM逆变器故障诊断

电机变频调速系统中,逆变器是故障高发的薄弱环节。设计一种基于小波包分解和RBF神经网络的三相电机驱动系统PWM逆变器故障诊断模型,利用小波包变换提取三相PWM逆变器故障信号特征向量,并将其作为RBF神经网络的输入量;采用狼群—模拟退火算法优化RBF神经网络的结构和参数,利用32组学习样本和6组测试样本分别训练和检验RBF神经网络。仿真实验分析表明,该方法用于三相电机驱动系统PWM逆变器开路故障的诊断,速度快、准确率高。

Fault Diagnosis for PWM Inverter in Motor Driving System

In the motor frequency conversion system, the inverter is the weak link of high incidence of failure. A fault diagnosis model for PWM inverter in three phase motor drive system based on wavelet packet decomposition and RBF neural network is designed. The fault signal eigenvectors of three phase PWM inverters is extracted by wavelet packet transform, and use it as the input of the RBF neural network. The structure and parameters of RBF neural network are optimized by using the wolves-simulated annealing algorithm. The RBF neural network is trained and tested by 32 groups of learning samples and 6 sets of test samples respectively. The simulation results show that this method is used to diagnose the fault of PWM inverter in three-phase motor drive system, which is fast and accurate.