Multi-objective optimization design method for thermal performance of office building envelopes based on stochastic operation

The goal of building optimization design for thermal performance is to achieve a balance between indoor thermal comfort and operational energy consumption. However, the current thermal design of building envelope is usually based on fixed occupancy schedules, neglecting the impact of the randomness of building operation, which makes it difficult to accurately estimate the building performance under the thermal design scheme, thereby affecting the rationality of the thermal design scheme. This study established a stochastic operation prediction model of office building air-conditioning(AC) and exterior windows to determine the annual hourly random sequences of AC and exterior windows. Based on the stochastic operation schedule, a building performance simulation was conducted, and an artificial neural network was used to construct a rapid prediction model for building heating load and cooling load as well as thermal comfort. Furthermore, with the aims of enhancing building thermal comfort and reducing heating/cooling loads, a multi-objective optimization design method for thermal performance of office building envelope based on stochastic operation was established using genetic algorithm. Finally, taking office buildings in hot-summer-cold-winter region of China as an example, considering the stochastic operation characteristics of buildings and the thermal comfort needs of occupants, this study explores the optimization design strategies for thermal performance of office building envelopes that meet dual objectives of energy efficiency and thermal comfort. The current provides a methodology for the thermal design of low-energy building envelope, which satisfies the comfortable thermal environment under random building operation.