Deep learning for determining a near-optimal topological design without any iteration

In this study, we propose a novel deep learning-based method to predict an optimized structure for a given boundary condition and optimization setting without using any iterative scheme. For this purpose, first, using open-source topology optimization code, datasets of the optimized structures paire...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Structural and multidisciplinary optimization Jg. 59; H. 3; S. 787 - 799
Hauptverfasser: Yu, Yonggyun, Hur, Taeil, Jung, Jaeho, Jang, In Gwun
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Berlin/Heidelberg Springer Berlin Heidelberg 01.03.2019
Springer Nature B.V
Schlagworte:
Artificial neural networks
Boundary conditions
Coders
Computational Mathematics and Numerical Analysis
Computing time
Datasets
Deep learning
Engineering
Engineering Design
Generative adversarial networks
Iterative methods
Machine learning
Neural networks
Optimization
Performance evaluation
Research Paper
Source code
Theoretical and Applied Mechanics
Topology optimization
Deep learning
Topology optimization
Convolutional neural network
Machine learning
Generative adversarial network
Generative model
ISSN:1615-147X, 1615-1488
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this study, we propose a novel deep learning-based method to predict an optimized structure for a given boundary condition and optimization setting without using any iterative scheme. For this purpose, first, using open-source topology optimization code, datasets of the optimized structures paired with the corresponding information on boundary conditions and optimization settings are generated at low (32 × 32) and high (128 × 128) resolutions. To construct the artificial neural network for the proposed method, a convolutional neural network (CNN)-based encoder and decoder network is trained using the training dataset generated at low resolution. Then, as a two-stage refinement, the conditional generative adversarial network (cGAN) is trained with the optimized structures paired at both low and high resolutions and is connected to the trained CNN-based encoder and decoder network. The performance evaluation results of the integrated network demonstrate that the proposed method can determine a near-optimal structure in terms of pixel values and compliance with negligible computational time.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1615-147X
1615-1488
DOI:10.1007/s00158-018-2101-5