Abstract: Dynamic supply current test is very useful for fault diagnosis in analog circuits.In order to improve the defect of wavelet neural network in fault diagnosis in analog circuits I_DDT ，a method of fault diagnosis in analog circuits I_DDT based on multiwavelet neural network is proposed.Multiwavelet transform is used to extract each frequency’s energy of current of power supply.Then the energy is used as input vector of neural network for fault diagnosis.Simulation results indicate that this method is effective，multiwavelet neural network converges more rapidly than wavelet neural network.

摘要： 动态电源电流测试（I_DDT ）对模拟电路故障诊断非常有效。针对小波神经网络在模拟电路I_DDT故障诊断中存在的缺陷，提出了一种基于多小波变换的模拟电路I_DDT故障诊断方法。即利用多小波变换提取电源电流各频段的能量，作为神经网络的输入特征向量进行故障诊断。仿真结果表明，该方法是有效的，而且比小波神经网络方法的收敛速率快。