Fault diagnosis based on radial basis function neural network in analog circuits

The radial basis function (RBF) neural network (NN) is a type of feedforward network. It has many good properties, such as a powerful ability for function approximation, classification and learning rapidly. A sinusoidal input to an analog circuit is simulated with constant amplitude and different frequencies; frequency domain features of the output response are used to build a fault dictionary. The paper proposes an RBF NN method for response analysis and fault diagnosis. Results illustrate that this method is feasible and has many powerful features, such as diagnosing and locating faults quickly and exactly.