Image fragmented learning for data-driven topology design

This paper proposes a data-driven topology design (DDTD) framework, incorporating image fragmented learning that leverages the technique of dividing an image into smaller segments for learning each fragment. This framework is designed to tackle the challenges of high-dimensional, multi-objective opt...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Structural and multidisciplinary optimization Ročník 68; číslo 8; s. 148
Hlavní autori: Yang, Yusibo, Yaji, Kentaro, Yamasaki, Shintaro, Fujita, Kikuo
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Berlin/Heidelberg Springer Berlin Heidelberg 12.08.2025
Springer Nature B.V
Predmet:
Computational Mathematics and Numerical Analysis
Data compression
Datasets
Design
Design optimization
Effectiveness
Engineering
Engineering Design
Generative adversarial networks
Genetic algorithms
Methods
Modules
Multiple objective analysis
Neural networks
Optimization techniques
Probability distribution
Research Paper
Theoretical and Applied Mechanics
Topology
Turbulent heat transfer
Topology optimization
Multi-objective optimization
Image fragmented learning
Data-driven topology design
ISSN:1615-147X, 1615-1488
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:This paper proposes a data-driven topology design (DDTD) framework, incorporating image fragmented learning that leverages the technique of dividing an image into smaller segments for learning each fragment. This framework is designed to tackle the challenges of high-dimensional, multi-objective optimization problems. Original DDTD methods leverage the sensitivity-free nature and high capacity of deep generative models to effectively address strongly nonlinear problems. However, their training effectiveness significantly diminishes as input size exceeds a certain threshold, which poses challenges in maintaining the high degrees of freedom crucial for accurately representing complex structures. To address this limitation, we split a trained conditional generative adversarial network into two interconnected modules: the first performs dimensionality reduction, compressing high-dimensional data into a lower-dimensional representation, which is then fed into a variational autoencoder (VAE) to generate new low-dimensional data. The second module reconstructs the generated low-dimensional data back into the high-dimensional design space. The effectiveness of the proposed approach is demonstrated through two case studies: the optimization of an L-bracket design problem and a turbulent heat transfer design problem, both involving design variables at a scale unattainable by the conventional VAE-based method.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1615-147X
1615-1488
DOI:10.1007/s00158-025-04054-3