Fault Diagnosis Method of Analog Circuit Based on Radial Base Function Neural Network

Abstract

Neural network is competent for fault diagnosis and pattern classification such as poorly defined model system, noisy input environment and nonlinearity in analog circuit. One of the most known type of neural network used to identify and classify the method of faulty diagnosis of analog circuit is presented based on the radial basis function (RBF) neural network. The proposed method introduces the fault features and classify the fault classes of the given two circuits. The experiment and simulations of the fault diagnosis for the two circuits (Sallen-Key and four OP-AMP band-pass filter) establish the diagnosing process of our method that proposed in this paper, locating faults effectively and be successful in classifying the fault values. Its performance analysis show that the fault diagnosability learned by RBF method accomplishes a better satisfactory diagnostic accurateness for the fault diagnosis of analog circuit.