Application of BP neural network in analog circuits diagnosis

The theories of using neural network to construct the fault dictionary of analog circuits were studied The response signatures of fault and fault-free circuit under test were generated during a simulation of the circuit before the diagnosis phase and are used to train an neural network. If we have obtained the response signature of the circuit under test on line, we can identify the fault by inputting the signature to the neural network. Experiment on a continuous-time state-variable filter demonstrates that the method of this paper has high diagnose sensitivity and fast fault identification and deducibility