Efficient Quality Diversity Optimization of 3D Buildings through 2D Pre-Optimization

Alexander Hagg; Martin L. Kliemank; Alexander AsterothDominik WildeMario C. BedrunkaHolger FoysiDirk Reith

首页> 外文期刊>Evolutionary computation >Efficient Quality Diversity Optimization of 3D Buildings through 2D Pre-Optimization

【24h】

Efficient Quality Diversity Optimization of 3D Buildings through 2D Pre-Optimization

机译：Efficient Quality Diversity Optimization of 3D Buildings through 2D Pre-Optimization

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相关主题

摘要

Quality diversity algorithms can be used to efficiently create a diverse set of solutionsto inform engineers’ intuition. But quality diversity is not efficient in very expensiveproblems, needing hundreds of thousands of evaluations. Even with the assistance ofsurrogate models, quality diversity needs hundreds or even thousands of evaluations,which can make its use infeasible. In this study, we try to tackle this problem by using apre-optimization strategy on a lower-dimensional optimization problem and thenmapthe solutions to a higher-dimensional case. For a use case to design buildings that minimizewind nuisance, we show that we can predict flow features around 3D buildingsfrom 2D flow features around building footprints. For a diverse set of building designs,by sampling the space of 2D footprints with a quality diversity algorithm, a predictivemodel can be trained that ismore accurate than when trained on a set of footprints thatwere selected with a space-filling algorithm like the Sobol sequence. Simulating only16 buildings in 3D, a set of 1,024 building designs with low predicted wind nuisanceis created. We show that we can produce better machine learning models by producingtraining data with quality diversity instead of using common sampling techniques.The method can bootstrap generative design in a computationally expensive 3D domainand allow engineers to sweep the design space, understanding wind nuisance inearly design phases.

著录项

来源
《Evolutionary computation》 |2023年第3期|287-307|共21页
作者
Alexander Hagg; Martin L. Kliemank; Alexander AsterothDominik WildeMario C. BedrunkaHolger FoysiDirk Reith;
展开▼
作者单位

Institute of Technology, Resource and Energy-efficient Engineering (TREE),Bonn-Rhein-Sieg University of Applied Sciences, Sankt Augustin, 53757, Germany;

Institute of Technology, Resource and Energy-efficient Engineering (TREE),Bonn-Rhein-Sieg University of Applied Sciences, Sankt Augustin, 53757, GermanyDpt. of Mechanical Engineering, University of Siegen, Siegen, 57076, Germany;

Dpt. of Mechanical Engineering, University of Siegen, Siegen, 57076, GermanyInstitute of Technology, Resource and Energy-efficient Engineering (TREE),Bonn-Rhein-Sieg University of Applied Sciences, Sankt Augustin, 53757, GermanyFraunhofer Institute for Algorithms and Scientific Computing (SCAI), Sankt Augustin,53754, Germany;

展开▼
收录信息
原文格式 PDF
正文语种英语
中图分类计算技术、计算机技术;
关键词
Multisolution optimization; quality diversity; efficiency; pre-optimization; wind nuisance; Bayesian optimization; lattice Boltzmann method; design process;

Efficient Quality Diversity Optimization of 3D Buildings through 2D Pre-Optimization

摘要

著录项

相关主题

期刊订阅