Fault diagnosis in high voltage breakes based on IRBF neural network

According to the questions of Radial Basis Function (RBF) neural network in the mechanical failure diagnose of high voltage breakers, which can extremely affect convergence speed and precision of the RBF neural networks. This paper develops an improved RBF neural network learning algorithm based on immune algorithm. In the algorithm, the input data are regarded as antigens and the compression mapping of antigens as antibodies, i.e., the concealed layer center point, which also avoid network concealed layer center point hard problem. The weights of the output layer are determined by adopting the gradient descent algorithm. Then it imposes discipline good network on the mechanical failure diagnose of high voltage breakers. The simulation results indicate that this method has preferable application value in the mechanical vibration signal of high voltage breakers.