Optimal energy management strategy for electric vehicles and electricity distribution system based on quantile deep learning with enhanced African vulture optimization

The increasing fluctuation in electricity prices due to unpredictable demand loads on the power grid has underscored the importance of finding optimal scheduling solutions for electric vehicles (EVs). Integrating EVs into the distribution system (DS) is critical for efficiently managing their energy needs. However, current EV scheduling approaches face challenges. Centralized solutions are often unreliable and complex, while existing decentralized methods struggle to efficiently schedule EVs in large networks with smart energy DS. To address these challenges, we propose a hybrid technique called Quantile Deep Learning with Enhanced African Vulture Optimization (Quantum VultureNet) for energy management systems (EMS) involving EVs and DS. This approach combines the African Vultures Optimization algorithm with elite mutation, learning based on dynamic opposition, and chaotic map -based population initialization. By enhancing exploration and exploitation capabilities, this hybrid method aims to improve the scheduling of EVs, particularly in dynamic environments. The primary goal of the Quantum VultureNet technique is to minimize system costs and power losses while optimizing energy management. The proposed system considers factors such as bidirectional energy trading, EV fleet arrival times in driving plans, Photo Voltaic uncertainty impacts on energy management systems, and factors affecting energy selling back to the grid. The model was implemented in MATLAB and compared with conventional methods. Simulation results demonstrate that the Quantum VultureNet technique effectively finds near -global optimum solutions with reduced computational complexity. This indicates the effectiveness of the proposed approach for EMS in EVs integrated with DS.