A Multi-Objective Optimization of Passive Strategies for Atrium Spaces in Retrofitting Old Industrial Buildings： a Case Study of Hohhot

In the context of rapid urban expansion and the “dual carbon” goals， the functions， performance， energy consumption， and emissions of old industrial buildings are incompatible with the trend of contemporary architectural development. The introduction of atrium spaces provides the conditions for old industrial buildings to adapt to new functions and optimize performance. Taking the atrium spaces of renovated old industrial buildings in Hohhot as the research object and aiming to optimize indoor comfort and energy saving， this study proposed a multi-objective optimization method that integrated parametric performance simulations， BP neural networks， and NSGA-Ⅱ non-dominated genetic algorithm. Based on this， passive strategies suitable for atrium spaces of renovated old industrial buildings in Hohhot were proposed， including selecting a larger atrium floor area ratio， adopting an open interface form， opting for a narrow-top-wide-bottom “A”-shaped sectional form， and choosing a larger skylight area ratio， etc. Practical cases were selected to verify the effectiveness of these strategies.