Computational & Technology Resources
an online resource for computational,
engineering & technology publications
engineering & technology publications
Civil-Comp Conferences
ISSN 2753-3239
ISSN 2753-3239
CCC: 11
PROCEEDINGS OF THE SEVENTH INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE, SOFT COMPUTING, MACHINE LEARNING AND OPTIMIZATION IN ENGINEERING
PROCEEDINGS OF THE SEVENTH INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE, SOFT COMPUTING, MACHINE LEARNING AND OPTIMIZATION IN ENGINEERING
Edited by: P. Iványi, J. Kruis and B.H.V. Topping
Paper 5.2
Forward and Inverse Topology Optimization via Deep Rank-Reduction Autoencoders
I. Ben-Yelun1, M. El-Fallaki Idrissi2, J. Mounayer2, S. Rodríguez2, F.J. Montans1 and F. Chinesta2
1Escuela Técnica Superior de Ingeniería Aeronáutica y del Espacio, Universidad Politécnica de Madrid, Spain
2PIMM, ParisTech Arts et Métiers Institute of Technology, France
Full Bibliographic Reference for this paper
I. Ben-Yelun, M. El-Fallaki Idrissi, J. Mounayer, S. Rodríguez, F.J. Montans, F. Chinesta, "Forward and Inverse Topology Optimization via Deep Rank-Reduction Autoencoders", in P. Iványi, J. Kruis, B.H.V. Topping, (Editors), "Proceedings of the Seventh International Conference on
Artificial Intelligence, Soft Computing, Machine Learning and Optimization in Engineering", Civil-Comp Press, Edinburgh, UK,
Online volume: CCC 11, Paper 5.2, 2025,
Keywords: autoencoder, topology optimization, fi nite element method, machine learning, model order reduction, structural design.
Abstract
Topology optimization has gained renewed attention thanks to the improvement of 3D printing techniques, especially for metals. While it is currently mainly limited to small-scale production and non-critical components, improvements in printing quality and mass production will make topology optimization a key engineering design technology for enhancing products. However, designing components through topology optimization is computationally expensive, as many finite element simulations are required. This is particularly important when performing inverse analysis. The purpose of this paper is to explore the use of deep rank-reduction autoencoders to accelerate both the forward and the inverse topology optimization schemes, solving also the well-known ill-posedness of the inverse problem.
download the full-text of this paper (PDF, 2611 Kb)
go to the previous paper
return to the table of contents
return to the volume description