This manuscript presents a novel hybrid technique, termed the MRA-SDRN method, for optimizing renewable energy-based electric vehicle (EV) charging within an Internet of Things (IoT) architecture. The key objective is to enhance the efficiency of PV-connected EV systems while concurrently reducing electricity costs and pollutant emissions. The MRA-SDRN approach leverages the Mud Ring Algorithm (MRA) and the Spiking Deep Residual Network (SDRN) to optimize several parameters critical for effective EV charging and renewable energy utilization. Comparative analysis with existing techniques conducted in MATLAB demonstrates notable reductions in computation time, with the MRA-SDRN method achieving a calculation time of 13.04 s, outperforming simulated annealing (SA), multi-objective particle swarm optimization and genetic algorithms (GA).Additionally, significant decreases in pollutant emissions are observed, with levels measured at 472.38 g compared to 570.48 g for GA, 528.35 g for SA, and 502.01 g for MOPSO. Regarding electricity costs, the MRA-SDRN approach achieves a competitive rate of 0.306 yuan/kWh, emphasizing its cost-effectiveness. These numerical results underscore the efficacy of the MRA-SDRN method in optimizing renewable energy management for EV charging, thereby contributing to enhanced reliability, efficiency, and sustainability in the transportation sector within the IoT framework.
Read full abstract