Abstract
The incorporation of renewable energy sources and smart grids has resulted in the need for intelligent and adaptive energy management systems that are able to manage dynamic residential energy demands, uncertainties in renewable energy sources, and various constraints. Traditional optimization techniques and those based on artificial intelligence techniques have not been successful in addressing the issues of exploration and have been unable to offer a holistic solution for energy prediction, scheduling, energy management from batteries, and renewable energy integration. This study aims to introduce a new technique known as Quantum-Inspired Neural Optimization Framework for Sustainable Energy Management in Smart Grid-Connected Residential Systems (QINE-SGEM), which incorporates neural learning, quantum-inspired optimization, and adaptation techniques for efficient energy management in residential areas. The quantum-inspired neural optimization component carries out feature extraction, demand prediction, and decision-making for appliance scheduling, battery charging and discharging, grid energy distribution, and renewable energy integration. The performance of the framework is tested by using the Pecan Street Austin Smart Grid Energy dataset in a Python simulation environment. The results show that the QINE-SGEM model consumes an average energy of 24.2 kWh/day, produces 26.2% energy savings, enhances the battery utilization efficiency up to 95.8%, and reaches prediction accuracies of 0.084, 0.126, and 3.2% for MAE, RMSE, and MAPE, respectively. The ablation study shows the role of quantum optimization and adaptive feedback components with 36.8% energy cost savings, 82.6% renewable energy utilization, and 28.7% carbon emission savings. The proposed approach represents a sustainable and adaptive solution for future energy management in smart grids.