动态电源电流测试(IDDT)对模拟电路故障诊断非常有效,而小波变换能够有效提取动态电流的故障特性。因此提出一种基于小波-神经网络的模拟电路IDDT故障诊断方法。利用小波变换具有时频局部化特性,分别对模拟电路正常模式和故障模式的IDDT采样信号进行特征向量提取,建立故障字典;然后利用神经网络具有非线性映射优点,对各种状态下的特征向量进行分类决策,实现模拟电路的故障诊断。 Dynamic power supply current test (IDDT) is very effective in fault diagnosis of analog circuits, while wavelet transform can effectively extract the fault characteristics of dynamic current. Therefore, an IDDT fault diagnosis method for analog circuits based on wavelet neural network is proposed. Using time-frequency localization of wavelet transform, the eigenvectors of IDDT samples in normal mode and fault mode of analog circuit are extracted respectively, and the fault dictionary is constructed. Then, by using the neural network with the advantages of nonlinear mapping, the eigenvectors Carry on the classification decision-making, realizes the breakdown diagnosis of the simulation circuit.