TY - GEN
T1 - Design concept generation with variational deep embedding over comprehensive optimization
AU - Fujita, Kikuo
AU - Minowa, Kazuki
AU - Nomaguchi, Yutaka
AU - Yamasaki, Shintaro
AU - Yaji, Kentaro
N1 - Funding Information:
The initial study on this topic was performed a decade ago [45]. Many students of our laboratory, Masatoshi Muraoka, Atsushi Mitsunaka, Keisuke Ikuta, Eigo Tanaka, Mai Onizuka, and Kentaro Nishimi, have worked toward this paper since it. The authors acknowledge that their challenges brought us to this point through various trials.
Publisher Copyright:
© 2021 by ASME
PY - 2021
Y1 - 2021
N2 - This paper proposes a framework for generating design concepts through the loop of comprehensive exploitation and consequent exploration. The former is by any sophisticated optimization such as topology optimization with diversely different. The latter realization is due to the variational deep embedding (VaDE), a deep learning technique with classification capability. In the process of design concept generation first, exploitation through computational optimization generates various possibilities of design entities. Second, VaDE learns them. This learning encodes the clusters of similar entities over the latent space with smaller dimensions. The clustering result reveals some design concepts and identifies voids where as-yet-unrecognized design concepts are prospective. Third, the decoder of the learned VaDE generates some possibilities for new design entities. Forth such new entities are examined, and relevant new conditions will trigger further exploitation by the optimization. In this paper, this framework is implemented for and applied to the conceptual design problem of bridge structures. This application demonstrates that the framework can identify voids over the latent space and explore the possibility of new concepts. This paper brings up some discussion on the promises and possibilities of the proposed framework.
AB - This paper proposes a framework for generating design concepts through the loop of comprehensive exploitation and consequent exploration. The former is by any sophisticated optimization such as topology optimization with diversely different. The latter realization is due to the variational deep embedding (VaDE), a deep learning technique with classification capability. In the process of design concept generation first, exploitation through computational optimization generates various possibilities of design entities. Second, VaDE learns them. This learning encodes the clusters of similar entities over the latent space with smaller dimensions. The clustering result reveals some design concepts and identifies voids where as-yet-unrecognized design concepts are prospective. Third, the decoder of the learned VaDE generates some possibilities for new design entities. Forth such new entities are examined, and relevant new conditions will trigger further exploitation by the optimization. In this paper, this framework is implemented for and applied to the conceptual design problem of bridge structures. This application demonstrates that the framework can identify voids over the latent space and explore the possibility of new concepts. This paper brings up some discussion on the promises and possibilities of the proposed framework.
UR - http://www.scopus.com/inward/record.url?scp=85118300668&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85118300668&partnerID=8YFLogxK
U2 - 10.1115/DETC2021-69544
DO - 10.1115/DETC2021-69544
M3 - Conference contribution
AN - SCOPUS:85118300668
T3 - Proceedings of the ASME Design Engineering Technical Conference
BT - 47th Design Automation Conference (DAC)
PB - American Society of Mechanical Engineers (ASME)
T2 - 47th Design Automation Conference, DAC 2021, Held as Part of the ASME 2021 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC-CIE 2021
Y2 - 17 August 2021 through 19 August 2021
ER -