Operating state identification of hybrid power system based on back propagation neural network

Abstract This paper presents a system state identification method based on back propagation (BP) neural network. First, the working modes of the hybrid power system are given. Besides, the method of normalizing training parameters is provided to facilitate the training of neural network. Then, a neural network method is applied to state identification of hybrid power system. Finally, the weights and thresholds of the structure are obtained by the normalized training parameters, and the test results indicate that the trained neural network structure satisfies the system identification requirements.