Analog Circuit Fault Diagnosis Based On Pattern Recognition, Neural Network Methods

The analog fault diagnosis research is one of the forefront of the testing arena, one of the reason is that reliability of electrical depends on that of analog of circuits and systems. This paper presents the fault diagnosis of analog circuits with tolerances using the artificial neural networks method. The main aim of this paper is to provide a mechanism to deal with the problem of elcmcnl. tolcranccs and rcducc thc tcsting time. Applying this classic pattern recognition theory and artificial networks method, this paper proposes the analog fault diagnosis principles with backward-propagation neural network (BPNN) algorithm implementation. The robustness and associated memory of ANN make the method more advantageous than tradition method. The simulation results show us that the proposed method can perform correct diagnosis in the linear analog circuit with tolerances.