Eco smart charging: Cost forecasting and intelligent energy management for PV-wind-biomass hybrid EV stations with battery backup

Abstract

The rapid expansion of electric vehicles (EVs) is driven by their significant role in reducing fossil fuel consumption and CO2 emissions. However, meeting the changing needs of millions of EVs directly from the grid poses a risk of overloading the network and placing a substantial burden on the power sector. This paper proposes a hybrid method for smart control and energy management of a photovoltaic (PV)-wind-biomass hybrid system with battery backup for electric vehicle charging stations. The proposed hybrid approach integrates both the Zebra Optimization Algorithm (ZOA) and Verifiable Convolutional Neural Network (VCNN). The major objective of the proposed method is to enhance the overall efficiency of the system while minimizing fossil fuel consumption and reducing CO2 emissions. Renewable energy sources include PV, wind, biomass, and battery sources. The ZOA method is to optimize the control signal of the inverter, and the VCNN method is to predict the load demand. The model is implemented in MATLAB/Simulink, and its operational performance is validated under simulated off-grid conditions using standardized metrics. The proposed method efficiency is 98.2%, and the cost value of $12 000 is better than that of other existing methods such as sea-horse optimization, radial basis function neural networks, and convolutional neural networks.