Fault diagnosis of Electric Power Grid Based on Improved RBF Neural Network

This paper introduces a novel clustering algorithm that combines crisp and fuzzy clustering. It not only has the high accuracy of fuzzy clustering, but also reduces the dependency on initialization. Specifically, it constitutes a fast learning process and therefore, the convergence rate and the accuracy of the RBFNN are greatly improved. The simulation results show that this strategy is successfully applied to the fault diagnosis of electric power grid. The training speed and the fault-tolerance of information aberrance, which comes from the maloperation of the protections and breakers, are superior to the traditional RBFNN