Rationalization of Energy Consumption in the Smart Cities Using AI Algorithms

Energy optimization and management are crucial in addressing rising energy demands, environmental concerns, and smart technology integration. Traditional homes often experience unscheduled power usage, resulting in high electricity bills and an increased Peak-to-Average Ratio (PAR). Smart homes utilize Demand Side Management (DSM) to optimize appliance scheduling based on Real-Time Pricing (RTP), thereby reducing costs, enhancing grid stability, and improving user comfort. This study applies two meta-heuristic optimization techniques—the Gaining-Sharing Knowledge (GSK) algorithm and the Genetic Algorithm (GA)—to a real power optimization problem using real-time electricity pricing data from the New York Independent System Operator (NYISO). Simulation results demonstrate that the GSK algorithm achieves a 34% cost reduction and a 28.2% PAR reduction for a single smart home, outperforming the GA in cost savings. For a larger network of 30 homes, GSK achieves a 34.96% cost reduction, whereas GA is more effective in minimizing PAR by 52.7%. These findings highlight the efficiency of AI-based optimization in smart city power management.