摘要
藏式民居是我国民居瑰宝的重要组成部分,具有典型的高原特色,蕴含了历史悠久的藏族文化。随着西藏地区现代化进程的推进,当地居民对居住环境的需求也日益提高。传统藏式民居院落空间难以满足居民的热舒适需求,造成了不必要的能源消耗和温室气体排放。为此,本研究选取西藏自治区山南市藏式民居为研究对象,针对其院落空间开展低碳改造研究。以碳排量、室内热舒适和经济性为优化目标,建立建筑性能模拟与反向传播人工神经网络(BPNN)及遗传算法(NSGA-II)结合的多目标优化框架,探讨不同院落封闭改造策略的性能差异并识别特定宅形下的最优改造策略。优化结果显示,与未进行封闭改造的场景相比,最优策略可实现60.76%的碳排放减少以及6.86%的热舒适度提升。研究结果证实了在建筑低碳改造过程中引入人工智能技术的必要性和有效性,为藏式民居的现代化改造提供参考,为促进藏区建筑可持续发展作出贡献。
Tibetan-style residential houses represent a significant aspect of China’s residential heritagè,exhibiting typical plateau characteristics and a long history of Tibetan culture.With the advancement of modernisation in Xizang,the demand of local residents for a comfortable living environment is also increasing.However,the traditional Tibetan residential compound space is unable to meet the thermal comfort needs of residents,resulting in unnecessary energy consumption and greenhouse gas emissions.Consequently,this study selects Tibetan-style residential houses in Shannan City,Xizang Autonomous Region,as the research object and carries out a low-carbon renovation study for their courtyard spaces.A multi-objective optimisa-tion framework combining building performance simulation with back-propagation artificial neural network(BPNN)and genetic algorithm(NSGA-II)was established to explore the performance differences between different courtyard closure retrofit strategies and identify the optimal retrofit strategy for a specific house shape,with the optimisation objectives of carbon emission,indoor thermal comfort and economy.The opti-misation results demonstrate that the optimal strategy achieves a 60.76%reduction in carbon emissions and a 6.86%improvement in thermal comfort compared to the unenclosed scenario.These findings substantiate the necessity and efficacy of integrating AI technology into the process of low-carbon retrofitting of build-ings,providing a reference for the modernisation of Tibetan houses and contributing to the advancement of sustainable development in Tibetan areas.
作者
徐峰
白云峰
杨清欣
黄丽娟
罗喜红
温宝华
XU Feng;BAI Yunfeng;YANG Qingxin;HUANG Lijuan;LUO Xihong;WEN Baohua
出处
《建筑师》
CSSCI
2024年第6期59-65,共7页
The Architect
基金
国家自然科学基金青年项目(52108010)
湖南省重点研发计划(2024AQ2011)。
关键词
人工神经网络
多目标优化
藏式民居
人工智能
低碳改造
Artificial neural network
Multi-objective optimisation
Tibetan houses
Artificial intelligence
Low-car-bon retrofit
