Analog circuit fault diagnosis method for dynamic supply current monitoring

The invention relates to an analog circuit fault diagnosis method for dynamic supply current monitoring. The method includes the steps of building a training set sample, carrying out neural network training, and carrying out fault diagnosis, wherein when corresponding to-be-detected circuits operate normally or under a fault state, supply current signals are respectively collected and discretized, wavelet analysis and N layers of wavelet decomposition are carried out on the collected signals, and the sum of absolute values of each layer of high-frequency wavelet decomposition coefficient sequences serves as elements to constitute an N-dimensional fault feature vector; according to the dimensionality N of the fault feature vector and the sum M-1 of categories of fault modes of the corresponding to-be-detected circuits, the number N of nodes of the input layer, the number M of nodes of the output layer and the number of hidden layer neurons of a neural network are determined, network training is carried out by means of the training set sample according to target conditions, a is a preferential value adjusted and chosen from 1-10 in the network training process, and N, M and a are natural numbers; the fault feature vector of an object to-be-detected circuit is extracted and input into the neural network undergoing training to obtain a diagnosis result.