Innovative research on building energy consumption and comfort optimization based on BIM-BECS-AI collaboration

Against the backdrop of the continuous growth of global building energy consumption, achieving a balance between building energy conservation and indoor comfort has become a research hotspot. Traditional optimization methods mostly rely on a single simulation tool or empirical formula, which makes it difficult to deal with the nonlinear relationships and multi-objective conflicts of complex building systems. This study aims to provide a scientific and reasonable decision-making basis for building design and operation management by integrating domain knowledge enhancement technology and using the Pareto optimal solution of energy consumption intensity (EUI) and thermal comfort (PPD) based on the NSGA-II genetic algorithm and the hybrid superposition model (FNN XGB), so as to achieve the goal of energy saving while ensuring indoor thermal comfort. Artificial intelligence algorithms have significantly improved the scientific nature of green building design and the iterative efficiency of energy-saving solutions. Empirical studies based on public data sets (such as the London building data set) have verified the advantages of artificial intelligence algorithms in key indicators such as energy optimization efficiency.