建筑设计领域人工智能探索——从生成式设计到智能决策

周子骞; 高雯; 贺秋时; 林波荣; 韩雨乔

doi:10.13204/j.gyjzG21090801

建筑设计领域人工智能探索——从生成式设计到智能决策

doi: 10.13204/j.gyjzG21090801

1. 同济大学建筑与城市规划学院, 上海 200092;
2. 清华大学建筑学院, 北京 100084;
3. 生态规划与绿色建筑教育部重点实验室, 北京 100084

基金项目:

国家杰出青年科学基金项目（51825802）。

详细信息

作者简介:
周子骞,男,1996年出生,博士研究生。

通讯作者:
贺秋时,女,1995年出生,博士研究生,hqs18@mails.tsinghua.edu.cn。

计量
- 文章访问数: 2127
- HTML全文浏览量: 158
- PDF下载量: 50
- 被引次数: 0
出版历程
- 收稿日期: 2021-09-08
- 网络出版日期: 2022-10-28

Artificial Intelligence Exploration in Architectural Design-From Generative Design to Intelligent Decision-Making

1. College of Architecture and Urban Planning, Tongji University, Shanghai 200092, China;
2. School of Architecture, Tsinghua University, Beijing 100084, China;
3. Key Laboratory of Eco-Planning & Green Building, Ministry of Education, Beijing 100084, China

摘要

摘要: 建筑设计是一个复杂的过程，一方面需要知识和经验，另一方面又需要想象力和创造力。人工智能技术能够将建筑理论和设计经验数字化，提高设计过程的工作效率，同时也为设计提供了更多的思路。针对建筑设计中的不同问题，学者们尝试应用多种人工智能技术，并提出了多种方法。通过选取近5年的相关研究，从设计问题、人工智能技术和解决方案的角度进行梳理和归类，并对研究的现状和发展趋势进行分析。总结揭示了近年来相关研究的分布特点：在设计任务中，最热门的是平面布局和形体，分别达到42%和26%；在人工智能技术中，应用最多的是优化算法和神经网络，分别占41%和22%。然而，人工智能在建筑设计领域的应用尚未形成统一的理论体系，面临着通用性和标准性等关键挑战。
- 建筑设计 /
- 人工智能 /
- 生成式设计 /
- 优化算法
Abstract: Architectural design is a complex process, which not only needs knowledge and experience of architects but also their imagination and creativity. Artificial intelligence (AI) technology can digitize the architectural theory and design experience, improve the efficiency of the design process, and also provide more ideas for design. To solve different problems in architectural design, a variety of AI technologies have been applied and many methods have been proposed. Relevant achievements in the past five years were summarized and sorted out from the perspective of design problems, AI technologies and solutions. And the research status and development trends were analyzed. The summarization revealed distribution features of relevant researches in recent years. In the design task, the most popular issues were the layout and shape, which reached 42% and 26% of the reports respectively, and in AI technology, the most extensive application was optimization algorithms and neural networks, which accounted for 41% and 22% of the reports respectively. However, the application of AI in the field of architectural design has not formed a unified theoretical system yet, facing a crucial challenge of generality and standardization.
- architectural design /
- artificial intelligence /
- generative design /
- optimization algorithm

HTML全文

参考文献(17)


[70]	BOMFIM K, TAVARES F. Building facade optimization for maximizing the incident solar radiation[C]//Proceedings of the 37th eCAADe and 23rd SIGraDi Conference. 2019:171-180.
[71]	STEINØ N. Mapping the architectural genome:a preliminary study of facade syntax[C]//Proceedings of the 35th eCAADe Conference. 2017:453-462.
[72]	DE LUCA F, WORTMANN T. Multi-objective optimization for daylight retrofit[C]//Proceedings of the 38th eCAADe Conference. 2020:57-66.
[73]	李煜, 李玲玲, 刘滢. 光舒适导向的体育运动训练馆顶界面采光口参数化设计研究[C]//2020计算性设计国际学术论坛. 2020:413-429.
[74]	丁炜豪, 俞天琦, 吴玉冬, 等. 地下交通枢纽天窗采光模拟与形式优化研究[C]//2020计算性设计国际学术论坛. 2020:382-392.
[75]	GERBER D, PANTAZIS E. Design exploring complexity in architectural shells:interactive form finding of reciprocal frames through a multi-agent system[C]//Proceedings of the 34th eCAADe Conference. 2016:455-464.
[76]	孙澄,韩昀松.基于计算性思维的建筑绿色性能智能优化设计探索[J].建筑学报,2020(10):88-94.
[77]	SEBESTYEN A, TYC J. Machine learning methods in energy simulations for architects and designers:the implementation of supervised machine learning in the context of the computational design process[C]//Proceedings of the 38th eCAADe Conference. 2020:613-622.
[78]	SINGH M M, SCHNEIDER-MARIN P, HARTER H, et al. Applying Deep learning and databases for energy-efficient architectural design[C]//Proceedings of the 38th eCAADe Conference.2020:79-87.
[79]	TAKIZAWA A. Estimating potential event occurrence areas in small space based on semi-supervised learning[C]//Proceedings of the 34th eCAADe Conference.2016:169-178.
[80]	TARABISHY S, PSARRAS S, KOSICKI M, et al. Deep learning surrogate models for spatial and visual connectivity[J]. International Journal of Architectural Computing, 2020, 18(1):53-66.
[81]	BROWN L, YIP M, GARDNER N, et al. Drawing recognition-integrating machine learning systems into architectural design workflows[C]//Proceedings of the 38th eCAADe Conference. 2020:289-298.
[82]	HUANG W, HAO Z. Architectural drawings recognition and generation through machine learning[C]//Proceedings of the 38th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA). 2018.
[83]	UZUN C, ÇOLAKOGLU M B. Architectural drawing recognition:a case study for training the learning algorithm with architectural plan and section drawing images[C]//Proceedings of the 37th eCAADe and 23rd SIGraDi Conference.2019:29-34.
[84]	张荷花,顾明.BIM模型智能检查工具在审查平台及消防审查中的应用[J].土木建筑工程信息技术,2021,13(1):1-7.
[85]	张荷花,顾明.BIM模型智能检查工具研究与应用[J].土木建筑工程信息技术,2018,10(2):1-6.