Seasonal electric vehicle forecasting model based on machine learning and deep learning techniques

In this paper,multiple featured machine learning algorithms and deep learning algorithms are applied in fore-casting the electric vehicles charging load profile from real datasets of Spain’s electrical grid.The study aims to provide realistic datasets of electric vehicle load profiles to cope with the potential increase in electric vehicle penetration taking into consideration the seasonality effects.Technical issues are caused by the distribution network of the electricity grid;such as the huge charging power and stochastic charging behaviors of the drivers of electric vehicles due to the mass rollout of electric vehicles.Forecasting electric vehicles’load profile is necessary to face challenges to solve the problem of the potential mass rollout of electric vehicles penetration.However,Electric vehicle is considered one of the most promising solutions that develops faster than other stochastic renewable solution to reduce greenhouse emissions.The seasonality effect is one of the huge chal-lenges on electrical loads,so it is investigated by creating four separate forecasting networks to increase system accuracy and studying the effect of seasonal factors such as temperature fluctuation in the four seasons affecting the electric vehicles’battery in charging and draining modes.These factors are affecting the accuracy of the forecasting model.Four featured algorithms are investigated.Artificial Neural Networks and Adaptive Neuro-Fuzzy Inference Systems are applied as machine learning algorithms,and Long Short-Term Memory and the Gated Recurrent Units are also utilized as deep learning algorithms.The Gated Recurrent Units model performs slightly better than the long short-term memory employed on the hourly average daily historical data of charging electric vehicles.While the Adaptive Neuro-Fuzzy Inference System gathers both Artificial Neural Network and Fuzzy Inference System advantages.