Computational Models for Forecasting Electric Vehicle Energy Demand

Electric vehicles (EV) are fast becoming an integral part of our evolving society. There is a growing movement in advanced countries to replace gas-driven with EVs towards cutting down pollution from emissions. When fully integrated into society, electric will share energy available on the grid; therefore, it important understand consumption profiles for EVs. In this study, some computation models developed predicting day-ahead city Barcelona. Five different machine learning algorithms namely support vector regression (SVR), Gaussian process (GPR), artificial neural networks (ANN), decision tree (DT), and ensemble learners were used train forecasting models. The hyperparameters each ML tuned by Bayesian optimization algorithm. order propose efficient features modeling EV demand, two model structures investigated, named Type-I Type-II model. instance model, seven regressors representing previous days considered as input features. only day same week. Based results we find that performance was good across all although less considered. Overall, employed study gave about 75-80% accuracy based R 2 criterion. formulated may prove useful planning unit commitment functions management functions.