Data-driven topology design using a deep generative model

被引:16
|
作者
Yamasaki, Shintaro [1 ]
Yaji, Kentaro [1 ]
Fujita, Kikuo [1 ]
机构
[1] Osaka Univ, Dept Mech Engn, 2-1 Yamadaoka, Suita, Osaka 5650871, Japan
关键词
Data-driven design; Topology optimization; Deep generative model; Sensitivity-free methodology; Multi-objective methodology; Estimation of distribution algorithm; LEVEL SET METHOD; GENETIC ALGORITHMS; OPTIMIZATION;
D O I
10.1007/s00158-021-02926-y
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
In this paper, we propose a sensitivity-free and multi-objective structural design methodology called data-driven topology design. It is schemed to obtain high-performance material distributions from initially given material distributions in a given design domain. Its basic idea is to iterate the following processes: (i) selecting material distributions from a dataset of material distributions according to eliteness, (ii) generating new material distributions using a deep generative model trained with the selected elite material distributions, and (iii) merging the generated material distributions with the dataset. Because of the nature of a deep generative model, the generated material distributions are diverse and inherit features of the training data, that is, the elite material distributions. Therefore, it is expected that some of the generated material distributions are superior to the current elite material distributions, and by merging the generated material distributions with the dataset, the performances of the newly selected elite material distributions are improved. The performances are further improved by iterating the above processes. The usefulness of data-driven topology design is demonstrated through numerical examples.
引用
收藏
页码:1401 / 1420
页数:20
相关论文
共 50 条
  • [41] Topology Aware Data-Driven Inverse Kinematics
    Ho, Edmond S. L.
    Shum, Hubert P. H.
    Cheung, Yiu-ming
    Yuen, P. C.
    COMPUTER GRAPHICS FORUM, 2013, 32 (07) : 61 - 70
  • [42] A Data-Driven Topology Estimation for Distribution Grid
    Liang, Haiwei
    Tong, Li
    Zou, Xudong
    Proceedings of the 16th IEEE Conference on Industrial Electronics and Applications, ICIEA 2021, 2021, : 1020 - 1023
  • [43] Data-Driven Analysis of Pareto Set Topology
    Hamada, Naoki
    Goto, Keisuke
    GECCO'18: PROCEEDINGS OF THE 2018 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, 2018, : 657 - 664
  • [44] A Deep Learning Model for Data-Driven Discovery of Functional Connectivity
    Mahmood, Usman
    Fu, Zening
    Calhoun, Vince D.
    Plis, Sergey
    ALGORITHMS, 2021, 14 (03)
  • [45] Fast Data-Driven Simulation of Cherenkov Detectors Using Generative Adversarial Networks
    Maevskiy, A.
    Derkach, D.
    Kazeev, N.
    Ustyuzhanin, A.
    Artemev, M.
    Anderlini, L.
    19TH INTERNATIONAL WORKSHOP ON ADVANCED COMPUTING AND ANALYSIS TECHNIQUES IN PHYSICS RESEARCH, 2020, 1525
  • [46] DATA-DRIVEN SOFTWARE-DESIGN USING INVERSION
    STORER, R
    INFORMATION AND SOFTWARE TECHNOLOGY, 1988, 30 (02) : 99 - 107
  • [47] Data-driven prediction of αIIbβ3integrin activation paths using manifold learning and deep generative modeling
    Dasetty, Siva
    Bidone, Tamara C.
    Ferguson, Andrew L.
    BIOPHYSICAL JOURNAL, 2024, 123 (17) : 2716 - 2729
  • [48] Generative Adversarial Networks-Based Synthetic Microstructures for Data-Driven Materials Design
    Narikawa, Ryuichi
    Fukatsu, Yoshihito
    Wang, Zhi-Lei
    Ogawa, Toshio
    Adachi, Yoshitaka
    Tanaka, Yuji
    Ishikawa, Shin
    ADVANCED THEORY AND SIMULATIONS, 2022, 5 (05)
  • [49] Data-driven design of thin-film optical systems using deep active learning
    Hong, Youngjoon
    Nicholls, David P.
    OPTICS EXPRESS, 2022, 30 (13) : 22901 - 22910
  • [50] Data-Driven Crowd Simulation with Generative Adversarial Networks
    Amirian, Javad
    van Toll, Wouter
    Hayet, Jean-Bernard
    Pettre, Julien
    PROCEEDINGS OF THE 32ND INTERNATIONAL CONFERENCE ON COMPUTER ANIMATION AND SOCIAL AGENTS (CASA 2019), 2019, : 7 - 10