Application of radial basis function neural network on fault diagnosis of electric vehicle

An electric vehicle's fault diagnosis method based on radial basis function neural network was proposed to overcome the traditional back propagation neural networks' slow computing speed. Comparison of convergence speed between the radial basis function (RBF) method and the BP method is made. The Matlab module of the neural networks was adopted to generate and train a RBF neural network. Training and testing samples were collected from field data and processed to the form which can be identified by RBF network. The results suggested that RBF based fault diagnosis procedures showed a fast and robust response to fault signals.