高铁牵引系统三相逆变器IGBT和电机速度传感器的复合故障诊断

Composite fault diagnosis of IGBT in three-phase inverter and speed sensor of motor for high-speed railway traction system

主要提出了针对高铁牵引系统中三相逆变器IGBT开路故障和牵引电机速度传感器增益故障的复合故障诊断方法,解决了复合故障发生后传感器故障难以从中分离的问题,并实现了复合故障的在线诊断。首先利用改进的递归最小二乘(RLS)算法的自适应滤波对信号进行降噪,以减小噪声对复合故障的故障特征提取的干扰;然后利用小波包分解放大复合故障特征,检测出传感器故障;最后利用提取的复合故障特征对L-M算法优化后的BP神经网络进行训练,实现对复合故障的实时在线诊断。仿真部分验证了该诊断方法的可行性。

It proposes a diagnosis method for composite fault of three-phase inverter IGBT open-circuit fault and the speed sensor gain failure of motor. This diagnosis method solves the problem that the sensor fault is difficult to separate and realizes the on-line diagnosis of the composite fault. Firstly it develops an adaptive filter and uses the improved RLS algorithm to reduce noise. Then,it proposes wavelet packet decomposition to amplify composite fault characteristics and detect the sensor fault. Finally,it uses Levenberg-Marquardt algorithm to realize optimized BP neural network and achieve real-time on-line diagnosis of the composite fault. Numerical simulation results verify the validity and availability of the composite fault detection method.