一种推挽式电路故障诊断方法的仿真研究*

提出了基于小波变换和神经网络的推挽式电路故障诊断方法。先仿真得到各种故障状态下的输出电压信号，然后对输出电压信号进行Daubechies小波变换获取多尺度低频系数和高频系数，并对小波系数进行处理提取故障特征量，最后利用故障特征矢量训练神经网络确定了推挽式电路故障诊断的神经网络模型。仿真结果表明基于小波变换和神经网络的推挽式电路故障诊断方法取得了较好的效果。

Simulation Study on a Sort of Fault Diagnosis Method of Push-pull Circuits

A fault diagnosis method based on wavelet transform and neural networks for push-pull circuits is presented. Firstly, output voltage signals of the push-pull circuits under faulty conditions are obtained with simulation, Then approximation and detail coefficients of output voltage signals are gotten by Daubechies wavelet transform, and are disposed to extract faulty features, After training the networks by faulty feature vectors, the neural networks model of the fault diagnosis system for the push-pull circuits is built. The simulation result shows the fault diagnosis method on wavelet transform and neural networks for the push-pull circuits has good effect.