Abstract:This paper proposes a BP neural network algorithm based on chaos particle?swarm?optimization, in order to solve power transformer fault diagnosis problem.This algorithm combined chaos ,particle swarm and the BP neural network,through chaos particle swarm optimization algorithm to obtain the optimal weights and the initial value of the threshold value of BP neural network, then take the network training and testing.The algorithm takes advantage of chaos's characteristic of ergodicity and sensitive to the initial values, to optimize the parameters of particle swarm algorithm,and imported early judgment mechanism,and in order to avoid the algorithm easily falling into local optimum,the algorithm took chaotic disturbance at the precocious state.The training and testing examples suggest that CPSO-BP neural network algorithm has better effect in transformer fault diagnosis.
