Machine learning-enabled mapping of techno-economic and environmental performance for passive envelope systems towards low-energy medium office buildings in China

Building envelopes are pivotal in controlling heat transfer, significantly influencing building energy consumption, sustainability, and progress towards carbon neutrality. This study presents a comprehensive, nationwide strategy for enhancing envelope performance in China by integrating advanced passive technologies, including state-of-the-art radiative cooling roofs and walls, as well as thermally insulated smart windows. Leveraging a standard-compliant, three-story medium office building as a benchmark, a machine learning model based on optimized extreme gradient boosting (XGBoost) technique was developed to predict the techno-economic and environmental impacts of these retrofitted envelope systems. High-resolution performance maps are generated, quantifying the seasonal and spatial variation in thermal and energy efficiency gains attributable to envelope upgrades. Results demonstrate substantial cooling electricity savings of 21.3 %, 55.2 %, 20.7 %, and 2.7 % for spring, summer, fall, and winter, respectively. Furthermore, with considering the operational energy saving benefits while omitting the capital and investment costs, tailored optimal retrofit strategies are identified for diverse climatic zones across China, revealing significant annual energy cost reductions in response to different climate regions. This study offers actionable insights and decision-support tools for optimizing passive envelope retrofits, thereby accelerating the transition towards low-energy buildings and supporting China's carbon neutrality ambitions.