Comparative analysis of energy consumption patterns in residential and commercial buildings for sustainable resource management

The rapid growth of urban populations, coupled with increasing technological dependence, has led to a substantial rise in global energy demand, with residential and commercial buildings accounting for a significant share of energy consumption and carbon emissions. Accurate forecasting of building energy usage is therefore essential for enhancing energy efficiency, reducing environmental impact, and supporting sustainable urban development. This study presents a comparative investigation of energy consumption patterns in residential and commercial buildings using advanced deep learning approaches. A comprehensive dataset incorporating structural, operational, and environmental variables—including room area, appliance utilization, insulation thickness, HVAC system characteristics, and ambient temperature conditions—was analysed using Deep Neural Networks (DNNs) and Recurrent Neural Networks (RNNs). The results reveal distinct consumption behaviours across building types: residential buildings exhibit highly dynamic, occupant-driven energy usage, whereas commercial buildings demonstrate more stable and predictable demand profiles. Comparative performance analysis indicates that the RNN model consistently outperforms the DNN model, achieving lower Root Mean Square Error (RMSE) and higher coefficient of determination (R²) values, thereby providing more accurate energy consumption forecasts. These findings underscore the effectiveness of temporal deep learning architectures in capturing complex energy usage patterns and highlight the transformative role of artificial intelligence in intelligent energy management. The proposed approach offers valuable insights for policymakers, building managers, and energy providers seeking to optimize energy utilization, enhance demand-side management, and accelerate the transition toward smart and sustainable building ecosystems.