Power Battery Charging State-of-Charge Prediction Based on Genetic Neural Network

The problem of power battery state of charge estimation for hybrid vehicle directly affects the vehicle performance and driving distance. Considering there exists nonlinear relationship between the battery state of charge and the observable external characteristics, this paper presents a kind of algorithm which is based on the combination of genetic algorithm and back-propagation neural network namely GA-BP algorithm, taking full advantage of the strong capability of global search of genetic algorithm and the remarkable generalization performance of back-propagation neural network,the hybrid vehicle Ni-MH power battery GA-BP charging model is designed for the charging process.The simulation analysis shows that the network training speed is superior to the traditional BP network, after training the optimal solutions can be approximated in short time on the basis of real-time battery external characteristics being collected.