Electric Vehicle Remaining Range Prediction with a Three-Layer Weighted Stacking Model
To achieve accurate prediction of electric vehicle remaining range,a method based on a three-layer weighted stacking model for predicting remaining range of electric vehicles is proposed in this paper.By com-bining the maximal information coefficient and Spearman correlation coefficient as criteria for variable evaluation,the minimum redundancy maximum relevance algorithm is employed to optimize and obtain the input feature set from the candidate features.A three-layer stacking model that incorporates the original training features is then con-structed,and Bayesian optimization algorithm is used to determine the weights of the base models within the stack-ing model.Finally,the input feature set is used to train the three-layer weighted stacking model and realize electric vehicle remaining range prediction.The results show that the proposed three-layer weighted stacking model has high prediction accuracy and,compared to other models,with stronger generalization capabilities.
electric vehiclemRMR algorithmStacking modelremaining range
万方数据
维普
