Fault Diagnosis of Distribution System Based on Distributed Chaotic Fuzzy RBF Neural Networks

Distributed chaotic fuzzy RBF neural networks is described in the paper against the problems of slow convergence and easy trapping into local minimum of BP neural networks. For introducing chaotic noise to choose the initial value of the network, it can avoid the local minimum, while the fuzzy character improves the capability of generalization. The simulation results show that the suggested approach for RBF NN works successfully and effectively in FSE.