Prediction model of electric vehicle driving range based on support vector regression

All status data related to electric vehicle driving generally are recorded. Making full use of those status data to predict electric vehicle driving can relax power anxiety caused by insufficient power of electric vehicles during driving to some extent. Aiming at this problem, a driving range prediction model of electric vehicles based on support vector regression is proposed in this paper. Firstly, the proposed model adopts Random forest for feature extraction, which selects influential factors that are closely related to State of charge (SOC) characteristics. Secondly, combined with the actual driving data of EVs, the driving range prediction model of EVs based on support vector regression was constructed. Simulation results show that the proposed prediction model is better than the prediction model using the original support vector progression in terms of Mean Square Error.