Abstract:

In order to classify power quality disturbance signals more effectively,the wavelet transform and neural network are organically combined to build a four-layer wavelet neural network model in this paper.Besides,the chaotic is embedded into particle swarm optimization algorithm to improve the convergence speed and accuracy of the network model training based on the property of chaotic.Then the trained network model is used to classify the normal voltage and several common power quality disturbances.The classification result shows that the wavelet neural network optimized by chaotic particle swarm optimization algorithm can effectively classify power quality disturbances,and has the advantage of strong interference resistance and good stability.Meanwhile,compared with particle swarm optimization algorithm and BP algorithm,classifying the power quality disturbances based on chaotic particle swarm optimization algorithm has the higher accuracy rate of classification.

Key words: power quality disturbances, wavelet neural network, chaotic, particle swarm optimization algorithm, classification