AI-Assisted Multidimensional Optimization of Thermal and Morphological Performance in Small-to-Medium Sports Buildings

With the advancement of China’s “dual-carbon” strategy, optimizing the thermal performance of small-to-medium-sized sports buildings—key contributors to urban energy consumption and carbon emissions—has become a critical area of green building research. This study conducts a systematic literature review following the PRISMA framework, analyzing 96 high-relevance articles sourced from Web of Science, ScienceDirect, and CNKI. The review focuses on four key dimensions: building morphology, envelope thermal performance, eco-friendly material application, and thermal comfort strategies. Findings indicate that building geometry significantly influences natural ventilation and solar gain; optimizing the envelope system can enhance energy efficiency by 12–18%; and incorporating sustainable materials contributes to lifecycle carbon reduction. Furthermore, effective thermal comfort regulation requires the integration of climate-responsive strategies with intelligent control systems. The growing use of AI-assisted technologies—such as fuzzy logic, reinforcement learning, and real-time environmental feedback—is facilitating a shift from single-dimensional energy-saving approaches to multidimensional coupled optimization. This review establishes a comprehensive theoretical and practical framework for low-carbon design in small-to-medium sports buildings and highlights the urgent need for empirical validation and integrated design approaches across diverse climate zones.