Intelligent generative structural design methods for shear wall buildings: From data-driven to physics-enhanced
Digital Twins and Intelligent Construction|更新时间:2023-11-28
|
Intelligent generative structural design methods for shear wall buildings: From data-driven to physics-enhanced
“In the field of intelligent construction, researchers have compared the design capabilities of data-driven and physics enhanced data-driven methods in detail through deep generative algorithms and case analysis, and verified the positive correlation between computer vision based evaluation and mechanical analysis based evaluation methods. The results indicate that the physical enhanced data-driven method has more stable design performance and basically eliminates the constraints of data quality and quantity; The rationality threshold based on the computer vision comprehensive evaluation index SCV is 0.5, corresponding to a mechanical performance difference of about 10%.”
Journal of Civil and Environmental EngineeringVol. 46, Issue 1, Pages: 82-92(2024)
作者机构:
清华大学 土木工程安全与耐久教育部重点实验室
土木工程系,北京 100084
作者简介:
brief: LIAO Wenjie (1995- ), assistant researcher, main research interest: intelligent structural design of buildings, E-mail: liaowj17@tsinghua.org.cn.
LU Xinzheng (corresponding author), professor, doctorial supervisor, E-mail: luxz@tsinghua.edu.cn.
基金信息:
National Key R & D Program of China(2019YFE0112800);Tencent Foundation (XPLORER PRIZE);Shuimu Tsinghua Scholar Program(2022SM005)
LIAO Wenjie,LU Xinzheng,HUANG Yuli,et al.Intelligent generative structural design methods for shear wall buildings: From data-driven to physics-enhanced[J].Journal of Civil and Environmental Engineering,2024,46(01):82-92.
LIAO Wenjie,LU Xinzheng,HUANG Yuli,et al.Intelligent generative structural design methods for shear wall buildings: From data-driven to physics-enhanced[J].Journal of Civil and Environmental Engineering,2024,46(01):82-92. DOI: 10.11835/j.issn.2096-6717.2022.078.
Intelligent generative structural design methods for shear wall buildings: From data-driven to physics-enhanced