On Machine-Learning-Driven Surrogates for Sound Transmission Loss Simulations

被引:2
|
作者
Cunha, Barbara Zaparoli [1 ,2 ]
Zine, Abdel-Malek [3 ]
Ichchou, Mohamed [1 ]
Droz, Christophe [4 ]
Foulard, Stephane [2 ]
机构
[1] Ecole Cent Lyon, Lab Tribol & Dynam Syst, F-69134 Ecully, France
[2] Compredict GmbH, D-64283 Darmstadt, Germany
[3] Ecole Cent Lyon, Inst Camille Jordan, F-69134 Ecully, France
[4] Univ Gustave Eiffel, INRIA, I4S Team, COSYS SII, F-35042 Rennes, France
来源
APPLIED SCIENCES-BASEL | 2022年 / 12卷 / 21期
关键词
surrogate; machine learning; sound transmission loss; vibroacoustics; sensitivity analysis; physics-guided features; GAUSSIAN-PROCESSES; ELEMENT; OPTIMIZATION; WAVE;
D O I
10.3390/app122110727
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
Surrogate models are data-based approximations of computationally expensive simulations that enable efficient exploration of the model's design space and informed decision making in many physical domains. The usage of surrogate models in the vibroacoustic domain, however, is challenging due to the non-smooth, complex behavior of wave phenomena. This paper investigates four machine learning (ML) approaches in the modelling of surrogates of sound transmission loss (STL). Feature importance and feature engineering are used to improve the models' accuracy while increasing their interpretability and physical consistency. The transfer of the proposed techniques to other problems in the vibroacoustic domain and possible limitations of the models are discussed. Experiments show that neural network surrogates with physics-guided features have better accuracy than other ML models across different STL models. Furthermore, sensitivity analysis methods are used to assess how physically coherent the analyzed surrogates are.
引用
收藏
页数:21
相关论文
共 50 条
  • [1] Modelling and understanding battery materials with machine-learning-driven atomistic simulations
    Deringer, Volker L.
    [J]. JOURNAL OF PHYSICS-ENERGY, 2020, 2 (04):
  • [2] A Machine-Learning-Driven Sky Model
    Satilmis, Pinar
    Bashford-Rogers, Thomas
    Chalmers, Alan
    Debattista, Kurt
    [J]. IEEE COMPUTER GRAPHICS AND APPLICATIONS, 2017, 37 (01) : 80 - 91
  • [3] Machine-Learning-Driven Simulations on Microstructure and Thermophysical Properties of MgCl2-KCl Eutectic
    Liang, Wenshuo
    Lu, Guimin
    Yu, Jianguo
    [J]. ACS APPLIED MATERIALS & INTERFACES, 2021, 13 (03) : 4034 - 4042
  • [4] Machine-Learning-Driven Advanced Characterization of Battery Electrodes
    Finegan, Donal P.
    Squires, Isaac
    Dahari, Amir
    Kench, Steve
    Jungjohann, Katherine L.
    Cooper, Samuel J.
    [J]. ACS ENERGY LETTERS, 2022, 7 (12) : 4368 - 4378
  • [5] The absorption and multiplication of uncertainty in machine-learning-driven finance
    Hansen, Kristian Bondo
    Borch, Christian
    [J]. BRITISH JOURNAL OF SOCIOLOGY, 2021, 72 (04): : 1015 - 1029
  • [6] Machine-Learning-Driven Matrix Ordering for Power Grid Analysis
    Cui, Ganqu
    Yu, Wenjian
    Li, Xin
    Zeng, Zhiyu
    Gu, Ben
    [J]. 2019 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION (DATE), 2019, : 984 - 987
  • [7] Machine-learning-driven on-demand design of phononic beams
    He, Liangshu
    Guo, Hongwei
    Jin, Yabin
    Zhuang, Xiaoying
    Rabczuk, Timon
    Li, Yan
    [J]. SCIENCE CHINA-PHYSICS MECHANICS & ASTRONOMY, 2022, 65 (01)
  • [8] Machine-learning-driven on-demand design of phononic beams
    Liangshu He
    Hongwei Guo
    Yabin Jin
    Xiaoying Zhuang
    Timon Rabczuk
    Yan Li
    [J]. Science China(Physics,Mechanics & Astronomy), 2022, 65 (01) : 37 - 48
  • [9] Machine-learning-driven on-demand design of phononic beams
    Liangshu He
    Hongwei Guo
    Yabin Jin
    Xiaoying Zhuang
    Timon Rabczuk
    Yan Li
    [J]. Science China Physics, Mechanics & Astronomy, 2022, 65
  • [10] A Machine-Learning-Driven Evolutionary Approach for Testing Web Application Firewalls
    Appelt, Dennis
    Nguyen, Cu D.
    Panichella, Annibale
    Briand, Lionel C.
    [J]. IEEE TRANSACTIONS ON RELIABILITY, 2018, 67 (03) : 733 - 757