L-GDP: An Integrated Approach for Energy Optimization in Net Zero Energy Buildings using LSTM, GWO, DE, and PSO Techniques

In the field of sustainable energy, the idea of Net Zero Energy Buildings (NZEB) has attracted a lot of interest, highlighting the necessity of maximizing energy use without sacrificing occupant comfort. However, by ignoring user convenience and depending on small datasets, current state-of-the-art optimization models frequently fall short, underscoring the need for more thorough and well-rounded methods. In order to reduce switching losses and preserve smooth integration within a smart home network, this study suggests an integrated energy optimization framework for smart home appliances. The sequential characteristics of appliance usage data are efficiently managed using Long Short-Term Memory (LSTM) networks. Furthermore, three evolutionary optimization algorithms: Grey Wolf Optimization (GWO), Particle Swarm Optimization (PSO), and Differential Evolution (DE)—are applied for appliance scheduling to enhance energy efficiency while boosting user experience. A robust and varied dataset comprising a range of household equipment, including refrigerators, TVs, and dishwashers, is used in simulations to verify the technique. PSO outperforms the other examined algorithms in producing the optimal scheduling solutions. Key performance measures such as Mean Absolute Percentage Error (MAPE), Mean Square Error (MSE), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE) are used to assess its effectiveness. The findings confirm PSO’s accuracy, resilience, and applicability for managing energy in real-world smart homes, demonstrating that it yields the lowest error rates.