Optimal Scheduling of Microgrid with Electric Vehicle Integration in Smart Grid using Progressive Graph Convolutional Network

Electric vehicles (EVs) integrated into microgrids (MGs) provide multiple benefits but create complex optimal scheduling problems which emerge from energetic unpredictability and fluctuating renewable energy sources (RESs) as well as dissimilar EV charging behaviors. Standard scheduling approaches show poor results when trying to handle complex energy systems because they do not apply effective adaptations to changes. The paper introduces a Progressive Graph Convolutional Network (PGCN) to perform optimal scheduling for MGs with embedded EVs. The proposed method uses Levelized Cost of Energy minimization (LCOE) as its main objective and increases grid reliability performance. The PGCN model delivers predictions of energy needs together with renewable power output and EV charging activity patterns and electric grid distribution patterns to enable better planning decisions. By then, the proposed method is implemented on MATLAB platform and compared with various existing methods such as Hybrid Genetic Ant Colony (HGAC), Artificial Neural Network and Particle Swarm Optimization (ANN-PSO), Opposition-based Competitive Swarm Optimizer (OCSO), Gray Wolf Algorithm (GWA), and Genetic Algorithm (GA).With a LCOE of 0.048 $/kWh the proposed PGCN method demonstrates superior cost-effectiveness as it beats all other methods for energy scheduling and management.