In order to predict the charging energy of electric vehicle battery, two kinds of piecewise linear prediction models are established. Through charging mode identification, the electric vehicle battery charging energy prediction is realized. To further improve the prediction accuracy, three nonlinear prediction models are adopted. Through correlation analysis, the variables with strong correlation with charging energy are selected, and the input of nonlinear models of electric vehicle battery charging energy prediction is determined. BP neural network model, genetic algorithm optimized BP neural network model and support vector machine model are established. By comparing the prediction effects of different models, it is shown that the prediction stability and accuracy of the nonlinear models is better than the linear models. In the nonlinear models, the support vector machine model has the best prediction effect.

CICTP