Latent Crossover for Data-Driven Multifidelity Topology Design

Kii, Taisei; Yaji, Kentaro; Fujita, Kikuo; Sha, Zhenghui; Seepersad, Carolyn Conner

doi:https://doi.org/10.1115/1.4064979

Latent Crossover for Data-Driven Multifidelity Topology Design

Kii, Taisei; Yaji, Kentaro; Fujita, Kikuo et al.

Journal of Mechanical Design, 2024, 146(5), 051713

Number of Access:407（2025-08-14 15:06 Counts）

Identifier to cite or link to this item: https://hdl.handle.net/11094/95639

Link to primary information

File	Format	Terms of use	Size	Views	Date.Available	Description	information
JMechDes_146_5_051713	pdf	None	1.53 MB	165	2024.05.17

Creative Commons : Attribution

Item Information

Output File Export EndNote Basic Export Mendeley

Title	Latent Crossover for Data-Driven Multifidelity Topology Design
Creator	CreatorName Kii, Taisei
	NRID 1000090779373 1000090779373 ORCID 0000-0002-4309-2043 0000-0002-4309-2043 researchmap permalink yajiken0609 yajiken0609 Scopus Author ID 55787947500 55787947500 CreatorName Yaji, Kentaro
	NRID 1000010228992 1000010228992 researchmap permalink read0014111 read0014111 Scopus Author ID 7404057902 7404057902 CreatorName Fujita, Kikuo
	CreatorName Sha, Zhenghui
	CreatorName Seepersad, Carolyn Conner
Subject	data-driven design
	deep generative model
	latent crossover
	machine learning
	maximum stress minimization
	multiobjective optimization
	topology optimization
Description	Kii T., Yaji K., Fujita K., et al. Latent Crossover for Data-Driven Multifidelity Topology Design. Journal of Mechanical Design 146, 051713 (2024); https://doi.org/10.1115/1.4064979.
Abstract	Topology optimization is one of the most flexible structural optimization methodologies. However, in exchange for its high level of design freedom, typical topology optimization cannot avoid multimodality, where multiple local optima exist. This study focuses on developing a gradient-free topology optimization framework to avoid being trapped in undesirable local optima. Its core is a data-driven multifidelity topology design (MFTD) method, in which the design candidates generated by solving low-fidelity topology optimization problems are updated through a deep generative model and high-fidelity evaluation. As its key component, the deep generative model compresses the original data into a low-dimensional manifold, i.e., the latent space, and randomly arranges new design candidates over the space. Although the original framework is gradient free, its randomness may lead to convergence variability and premature convergence. Inspired by a popular crossover operation of evolutionary algorithms (EAs), this study merges the data-driven MFTD framework and proposes a new crossover operation called latent crossover. We apply the proposed method to a maximum stress minimization problem in 2D structural mechanics. The results demonstrate that the latent crossover improves convergence stability compared to the original data-driven MFTD method. Furthermore, the optimized designs exhibit performance comparable to or better than that in conventional gradient-based topology optimization using the P-norm measure.
Publisher	American Society of Mechanical Engineers (ASME)
Source Title	Journal of Mechanical Design
Volume (Issue)	146(5)
Page	051713
Date of Issued	2024-05-01
Language	English
Handle URL	https://hdl.handle.net/11094/95639
PISSN	10500472
Relation.isIdenticalTo	DOI (Publisher Version) https://doi.org/10.1115/1.4064979
Access Rights	open access
CopyRight	This article is licensed under a Creative Commons Attribution 4.0 International License.
oaire:version	Version of Record
Category	学術雑誌論文 Journal Article

Resource Type	journal article
Local Type	Journal Article
DCMI Type	Text
DCTERMS.bibliographicCitation	Journal of Mechanical Design. 2024, 146(5), p. 051713
DC.title	Latent Crossover for Data-Driven Multifidelity Topology Design
DC.creator	Kii, Taisei
	Yaji, Kentaro
	Fujita, Kikuo
	Sha, Zhenghui
	Seepersad, Carolyn Conner
DC.publisher	American Society of Mechanical Engineers (ASME)
DC.language' scheme='DCTERMS.RFC1766	English
DC.type' scheme='DCTERMS.DCMIType	Text
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2024-05-01
DC.identifier	https://doi.org/10.1115/1.4064979
DC.subject	data-driven design
	deep generative model
	latent crossover
	machine learning
	maximum stress minimization
	multiobjective optimization
	topology optimization
DC.description	Kii T., Yaji K., Fujita K., et al. Latent Crossover for Data-Driven Multifidelity Topology Design. Journal of Mechanical Design 146, 051713 (2024); https://doi.org/10.1115/1.4064979.
DCTERMS.abstract	Topology optimization is one of the most flexible structural optimization methodologies. However, in exchange for its high level of design freedom, typical topology optimization cannot avoid multimodality, where multiple local optima exist. This study focuses on developing a gradient-free topology optimization framework to avoid being trapped in undesirable local optima. Its core is a data-driven multifidelity topology design (MFTD) method, in which the design candidates generated by solving low-fidelity topology optimization problems are updated through a deep generative model and high-fidelity evaluation. As its key component, the deep generative model compresses the original data into a low-dimensional manifold, i.e., the latent space, and randomly arranges new design candidates over the space. Although the original framework is gradient free, its randomness may lead to convergence variability and premature convergence. Inspired by a popular crossover operation of evolutionary algorithms (EAs), this study merges the data-driven MFTD framework and proposes a new crossover operation called latent crossover. We apply the proposed method to a maximum stress minimization problem in 2D structural mechanics. The results demonstrate that the latent crossover improves convergence stability compared to the original data-driven MFTD method. Furthermore, the optimized designs exhibit performance comparable to or better than that in conventional gradient-based topology optimization using the P-norm measure.
DC.rights	This article is licensed under a Creative Commons Attribution 4.0 International License.
DC.format	95639
citation_title	Latent Crossover for Data-Driven Multifidelity Topology Design
citation_author	Kii, Taisei
	Yaji, Kentaro
	Fujita, Kikuo
	Sha, Zhenghui
	Seepersad, Carolyn Conner
citation_publisher	American Society of Mechanical Engineers (ASME)
citation_language	English
citation_date	2024-05-01
citation_journal_title	Journal of Mechanical Design
citation_volume	146
citation_issue	5
citation_firstpage	051713
citation_doi	https://doi.org/10.1115/1.4064979
citation_public_url	https://hdl.handle.net/11094/95639

Latent Crossover for Data-Driven Multifidelity Topology Design

Link to primary information

Item Information

Show details