The theory of analog circuit fault diagnosis is very important and significative,now it has become a hot studied project. The traditional methods of fault diagnosis areperformed only if the faults of the circuits are those hard faults, such as open-circuit,short-circuit, etc. Those soft faults aroused by the tolerance of circuit componentscannotbeeasilydiscovered.Moreover,thetraditionalmethodscannotlocatethefaultinsome circuits which only have one test node. However, a method based on neuralnetworkcaneasilysolvetheseproblems.In this paper, it is principally studied applying neural network to analog circuitfault diagnosis. The main contents are as follows. First, the intact diagnosis techniqueand its implement scheme are proposed. The paper investigates the design andalgorithm of BP neural network, and its feasibility applying to fault diagnosis isanalyzed. And some fault feature extraction methods are studied, including effectivesamplingpoints extraction and wavelete analysis. At last, an actual circuit is selected totest these methods and process. The simulation results of the example show that thefault diagnosis methods based on neural network have good diagnosis effect andfeasibilityintolerantcircuit. |