有源复合围护结构多目标优化设计方法  

Multi-objective Optimization Design Method for Thermal-activated Envelope

作　　者：赵田 庄智[1] 项平 汪铮 顾文 ZHAO Tian;ZHUANG Zhi;XIANG Ping;WANG Zheng;GU Wen(Tongji University,Shanghai 200092,China;Tongji Architectural Design(Group)Co.,Ltd,Shanghai 200093,China)

机构地区：[1]同济大学,上海200092 [2]同济大学建筑设计研究院(集团)有限公司,上海200093

出　　处：《建筑科学》2024年第12期275-284,共10页Building Science

基　　金：国家自然科学基金项目“夏热冬冷地区农宅通风蓄热式表皮耦合自然能源的热调节特性研究”(51608370);中央高校基本科研业务费专项资金“电热协同双稳态可展折光伏太空舱设计理论与方法研究”(2023-2-YB-22);上海市“科技创新行动计划”碳达峰碳中和专项项目“大型公共建筑超低能耗设计建造与运维关键技术研究及示范”(22dz1207100)。

摘　　要：有源复合围护结构是1种利用低品位能源并能够与环境交互实现动态可调的新型高效围护结构,如何最大化发挥其性能潜力需要科学的设计方法。本文以内置水管式有源复合围护结构为例,提出基于多目标优化的有源复合围护结构设计方法。该方法将构造选型、材料物性、热源运行3个方面的设计参数作为优化变量,把围护结构引起的全年冷热负荷最低、全年舒适小时数最多、材料经济成本最低及隐含碳排放量最小设为优化目标,借助非支配排序遗传算法(Non-dominated Sorting Genetic Algorithm II,NSGA-II)求得帕累托前沿优化解集,并利用熵权法与逼近理想解排序法(Technique for Order of Preference by Similarity to Ideal Solution,TOPSIS)对优化解集进行排序和评价,最终确定多目标性能最优的围护结构设计方案。基于此,以典型的内嵌毛细水管网墙体为例进行方法应用,得到了多目标性能综合最优的设计方案,并与单目标性能最优下的设计方案对比发现:对于以舒适小时数最多为目标函数的最优方案,其材料经济成本与隐含碳排放分别是多目标性能最优方案的3.15倍、5.98倍;对于以全年冷热负荷最低为目标函数的最优方案,其材料经济成本与隐含碳排放分别是多目标性能优化方案的2.95倍、4.34倍;对于以材料经济成本或隐含碳排放最低为目标函数的最优方案,其全年冷热负荷值及舒适小时数两项指标明显劣于多目标性能最优方案。可见,基于多目标优化的有源复合围护结构设计方法能综合考虑多方面的性能要求,可为其性能化设计和方案寻优提供科学且有力的指导。The thermal-activated envelope is an innovative structure that efficiently utilizes low-grade energy and can interact with the environment to achieve dynamic adjustability.Achieving its maximum performance potential necessitates a rigorous scientific design approach.This paper proposed a design method exemplified by the pipe-embedded thermal-activated envelope,employing multi-objective optimization.This method considered various design parameters,including structural types,material properties,and operating parameters of the internal heat source,as optimization variables.Four objective functions were defined,encompassing minimum annual heating and cooling loads,maximum indoor comfort hours,lowest material costs,and minimum embodied carbon emissions.The multi-objective genetic algorithm NSGA-II was employed to optimize the envelope design,resulting in a Pareto front solution set.Subsequently,the Entropy-weighted TOPSIS(Technique for Order of Preference by Similarity to Ideal Solution)method was used to evaluate and rank optimal solutions,identifying the most suitable envelope scheme.On this basis,the typical example of multilayer walls with a thermal capillary network was used for the application of the method,resulting in a design solution that achieves the most comprehensive optimal performance in terms of multiple objectives.A comparative analysis with the single-objective optimal design revealed that,when maximum indoor comfort hours was set as the single optimization objective,the material costs and embodied carbon emissions were 3.15 times and 5.98 times higher than those of the multi-objective optimal design,respectively.Similarly,when the minimum annual heating and cooling loads were set as the single optimization objective,the material cost and embodied carbon emissions were 2.95 and 4.34 times those of the multi-objective optimal scheme,respectively.When minimum material costs or embodied carbon emissions were set as the single optimization objective,the two indexes of annual loads and indoor comfort hour

关 键 词：有源复合围护结构 设计方法 性能化设计 多目标优化 遗传算法 

分 类 号：TU831[建筑科学—供热、供燃气、通风及空调工程]