Advanced methodological framework in energy conservation calculation for evaluating retrofitted and operational buildings via artificial intelligence and machine learning

Energy efficiency is crucial for sustainability, particularly in urban environments where buildings contribute significantly to energy consumption and carbon emissions. Singapore’s Green Plan 2030 emphasizes the need for robust methodologies to assess energy conservation. However, existing models often lack standardization and fail to integrate renewable energy considerations, especially in retrofitted buildings. This study presents a data-driven approach leveraging AI and ML to enhance energy performance evaluation for commercial and institutional buildings. Aligned with Green Mark certification, our model accurately establishes energy baselines and quantifies savings by incorporating Energy Use Intensity (EUI), operational variances, and meteorological factors. A case study on a retrofitted hotel demonstrated substantial savings of 2,813 MWh and a CO_₂ reduction of 1,161 tons over four years. The methodology not only improves energy assessments but also supports green finance mechanisms like sustainability-linked loans and energy conservation certificates, advancing energy management and global decarbonization efforts. By filling the critical gap in standardized evaluation frameworks, this research advances energy management practices and fosters sustainable industrial applications, contributing significantly to global decarbonization efforts. [<i>Li, Y., Jin, G., Jiang, D., and Lan, L. (2024)</i>. <i>Methodology of energy conservation calculation for commercial and institutional buildings. Project NovA! and Project SLEB Smart Hub. Retrieved from slebhub.sg</i>]