Abstract
Renewable energy forms have a comparatively less impact on the environment when compared with non-renewable sources. Renewable energies like wind power, biomass, hydropower, solar power, and geothermal energies are preferred for production as they do not run out quickly. Owing to the stochastic nature of renewable energy sources (RES) and Plug-in hybrid electric vehicles (PHEVs) load demand, large-scale penetration of these resources from the power system affects the network performance like reduce power quality, increase power loss, and voltage deviation. These issues can be addressed by the optimum planning depending upon the variables outcome in RES to satisfy the extra demand affected by PHEV charging. At the same time, designing a proper and effective energy management strategy in PHEV can be considered an optimization problem and can be resolved by the use of metaheuristic algorithms. With this motivation, this paper presents a novel deep learning with metaheuristic optimization based allocation of RES and charging stations for PHEVs. The proposed model depends upon model predictive control (MPC) where the actual battery state of charge (SOC) is concerned with real time energy management. Moreover, the black widow optimization (BWO) algorithm is used for the optimal allocation of RES and charging stations. The BWO algorithm derives objective functions using different parameters of the charging stations. In addition, based on the MPC model, deep stacked autoencoder (DSAE) is applied for the prediction of near-future velocity. The experimental results stated that the presented model has resulted in improved efficiency of the proposed model over the other methods.