Toward Automated Topology Optimization: Identification of Non-Design Features of CAD Models Using Graph Neural Networks

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Standard

Toward Automated Topology Optimization: Identification of Non-Design Features of CAD Models Using Graph Neural Networks. / Jasinski, Michael; Schöfer, Fabian; Seibel, Arthur.
Industrializing Additive Manufacturing: Proceedings of AMPA2023. ed. / Christoph Klahn; Mirko Meboldt; Julian Ferchow. Cham: Springer International Publishing AG, 2024. p. 267-279 (Springer Tracts in Additive Manufacturing).

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Harvard

Jasinski, M, Schöfer, F & Seibel, A 2024, Toward Automated Topology Optimization: Identification of Non-Design Features of CAD Models Using Graph Neural Networks. in C Klahn, M Meboldt & J Ferchow (eds), Industrializing Additive Manufacturing: Proceedings of AMPA2023. Springer Tracts in Additive Manufacturing, Springer International Publishing AG, Cham, pp. 267-279, 3th International Conference on Additive Manufacturing in Products and Applications - AMPA 2023, Zürich, Switzerland, 12.09.23. https://doi.org/10.1007/978-3-031-42983-5_19

APA

Jasinski, M., Schöfer, F., & Seibel, A. (2024). Toward Automated Topology Optimization: Identification of Non-Design Features of CAD Models Using Graph Neural Networks. In C. Klahn, M. Meboldt, & J. Ferchow (Eds.), Industrializing Additive Manufacturing: Proceedings of AMPA2023 (pp. 267-279). (Springer Tracts in Additive Manufacturing). Springer International Publishing AG. https://doi.org/10.1007/978-3-031-42983-5_19

Vancouver

Jasinski M, Schöfer F, Seibel A. Toward Automated Topology Optimization: Identification of Non-Design Features of CAD Models Using Graph Neural Networks. In Klahn C, Meboldt M, Ferchow J, editors, Industrializing Additive Manufacturing: Proceedings of AMPA2023. Cham: Springer International Publishing AG. 2024. p. 267-279. (Springer Tracts in Additive Manufacturing). doi: 10.1007/978-3-031-42983-5_19

Bibtex

@inbook{e5f9cdeb7fb4429184011b7bcdf3e611,
title = "Toward Automated Topology Optimization: Identification of Non-Design Features of CAD Models Using Graph Neural Networks",
abstract = "This paper presents an automated identification of non-design features of CAD models for topology optimization using learning-based segmentation. The CAD files are taken from a large database of industry-relevant components. Based on the geometry and topology of the components, a graph structure is created and processed by a deep neural network. The results show good match with real cases and can be continuously improved by training with additional data.",
keywords = "Engineering, Design Automation, Topology Optimization, Graph Neural Networks",
author = "Michael Jasinski and Fabian Sch{\"o}fer and Arthur Seibel",
note = "{\textcopyright} 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG; 3th International Conference on Additive Manufacturing in Products and Applications - AMPA 2023, AMPA 2023 ; Conference date: 12-09-2023 Through 14-09-2023",
year = "2024",
doi = "10.1007/978-3-031-42983-5_19",
language = "English",
isbn = "978-3-031-42982-8",
series = "Springer Tracts in Additive Manufacturing",
publisher = "Springer International Publishing AG",
pages = "267--279",
editor = "Christoph Klahn and Mirko Meboldt and Julian Ferchow",
booktitle = "Industrializing Additive Manufacturing",
address = "Switzerland",
url = "https://ampa.ethz.ch/",

}

RIS

TY - CHAP

T1 - Toward Automated Topology Optimization

T2 - 3th International Conference on Additive Manufacturing in Products and Applications - AMPA 2023

AU - Jasinski, Michael

AU - Schöfer, Fabian

AU - Seibel, Arthur

N1 - Conference code: 3

PY - 2024

Y1 - 2024

N2 - This paper presents an automated identification of non-design features of CAD models for topology optimization using learning-based segmentation. The CAD files are taken from a large database of industry-relevant components. Based on the geometry and topology of the components, a graph structure is created and processed by a deep neural network. The results show good match with real cases and can be continuously improved by training with additional data.

AB - This paper presents an automated identification of non-design features of CAD models for topology optimization using learning-based segmentation. The CAD files are taken from a large database of industry-relevant components. Based on the geometry and topology of the components, a graph structure is created and processed by a deep neural network. The results show good match with real cases and can be continuously improved by training with additional data.

KW - Engineering

KW - Design Automation

KW - Topology Optimization

KW - Graph Neural Networks

UR - https://www.mendeley.com/catalogue/8f45c01f-21f5-3101-8f2f-0377a38446d7/

U2 - 10.1007/978-3-031-42983-5_19

DO - 10.1007/978-3-031-42983-5_19

M3 - Article in conference proceedings

SN - 978-3-031-42982-8

T3 - Springer Tracts in Additive Manufacturing

SP - 267

EP - 279

BT - Industrializing Additive Manufacturing

A2 - Klahn, Christoph

A2 - Meboldt, Mirko

A2 - Ferchow, Julian

PB - Springer International Publishing AG

CY - Cham

Y2 - 12 September 2023 through 14 September 2023

ER -