Optimal day-ahead scheduling of microgrid equipped with electric vehicle and distributed energy resources: SFO-CSGNN approach

Electric Vehicles (EV) and Renewable Energy Sources (RES) hold significant promise for mitigating fossil fuel consumption and environmental effects within the transportation sectors and electricity; however challenges related to energy storage, grid infrastructure, and raw material supply chains must be addressed. This manuscript proposes a hybrid method for optimizing day-ahead Microgrid (MG) scheduling, incorporating EV and energy sources. The proposed hybrid method is the joint execution of the Sunflower optimization algorithm (SFO) and Contrastive Self-Supervised Graph Neural Network (CSGNN). Hence, it is named as SFO-CSGNN method. The principal aim of the proposed technique is to enhance MG reliability, reduce environmental impact, and improve flexibility, thereby contributing to a smart energy system. The proposed energy system integrates EVs with renewable sources including photovoltaic (PV), wind turbines (WT), and geothermal units (GU). The SFO algorithm optimizes EV charging and discharging schedules, while the CSGNN algorithm predicts the performance under high EV penetration levels. On the MATLAB platform, the proposed strategy is evaluated and contrasted with other existing techniques. The proposed s strategy displays superior outcomes in all existing like the Salp Swarm Algorithm (SSA), Particle Swarm Optimization (PSO) and Sandpiper Optimization Algorithm (SOA). Results indicate that the proposed approach achieves a cost reduction of 0.8$ compared to existing methods.