An Adaptive Meta-Modelling Approach for Multi-Dimensional Correlated Flow Field Responses

被引:0
|
作者
Chen, Jiangtao [1 ]
Xiao, Wei [1 ]
Lv, Luogeng [1 ]
Zhao, Jiao [1 ]
Zhao, Wei [1 ]
Wu, Xiaojun [1 ]
机构
[1] China Aerodynam Res & Dev Ctr, Computat Aerodynam Inst, Mianyang, Peoples R China
关键词
Adaptive sampling; surrogate model; multi-dimensional correlated responses; flow field reduction; machine learning;
D O I
10.1080/10618562.2024.2373200
中图分类号
O3 [力学];
学科分类号
08 ; 0801 ;
摘要
It has become common practice to build inexpensive surrogate models to supplant time-consuming numerical simulations. By gradually increasing the number of training samples, the efficiency of model construction can be significantly improved. This study proposes an adaptive meta-modelling approach for multi-dimensional correlated responses within the framework of proper orthogonal decomposition (POD) and the Kriging model. The algorithm begins with the adaptive sampling algorithm for each reduced-dimension response, which integrates the prediction variance, distances between samples, and sensitivity indicator of each parameter. The adaptive sampling criterion for each reduced-dimension response is weighted by the energy of the modal, forming the adaptive sampling algorithm for multi-dimensional correlated responses. Tests on an analytical function and M6 wing simulation show that, under the same number of training samples, the proposed adaptive algorithm results in a model with lower prediction error than the random sampling algorithm, offering a more efficient model for flow field prediction.
引用
收藏
页码:791 / 801
页数:11
相关论文
共 50 条
  • [41] A Multi-Dimensional Approach to Suicide Bombing
    Gill, Paul
    INTERNATIONAL JOURNAL OF CONFLICT AND VIOLENCE, 2007, 1 (02): : 143 - 159
  • [42] A multi-dimensional approach to major depression
    Tacchini, G
    Fumagalli, S
    Fusi, O
    Altamura, AC
    EUROPEAN NEUROPSYCHOPHARMACOLOGY, 2003, 13 : S241 - S242
  • [43] Hamiltonian approach to multi-dimensional screening
    Basov, S
    JOURNAL OF MATHEMATICAL ECONOMICS, 2001, 36 (01) : 77 - 94
  • [44] Nutritional Immunology: A Multi-Dimensional Approach
    Ponton, Fleur
    Wilson, Kenneth
    Cotter, Sheena C.
    Raubenheimer, David
    Simpson, Stephen J.
    PLOS PATHOGENS, 2011, 7 (12)
  • [45] Multi-Dimensional Modelling of Bioinspired Flow Channels Based on Plant Leaves for PEM Electrolyser
    Alobeid, Mohammad
    Celik, Selahattin
    Ozcan, Hasan
    Horri, Bahman Amini
    ENERGIES, 2024, 17 (17)
  • [46] Fused inverse regression with multi-dimensional responses
    Cho, Youyoung
    Han, Hyoseon
    Yoo, Jae Keun
    COMMUNICATIONS FOR STATISTICAL APPLICATIONS AND METHODS, 2021, 28 (03) : 267 - 279
  • [47] Gene expression programming based meta-modelling approach to production line design
    Baykasoglu, A.
    INTERNATIONAL JOURNAL OF COMPUTER INTEGRATED MANUFACTURING, 2008, 21 (06) : 657 - 665
  • [48] Optimizing chute-flip bucket system based on meta-modelling approach
    Bananmah, Mohammad
    Nikoo, Mohammad Reza
    Nematollahi, Banafsheh
    Sadegh, Mojtaba
    CANADIAN JOURNAL OF CIVIL ENGINEERING, 2020, 47 (05) : 584 - 595
  • [49] An approach to robust optimization of impact problems using random samples and meta-modelling
    Lonn, David
    Fyllingen, Orjan
    Nilssona, Larsgunnar
    INTERNATIONAL JOURNAL OF IMPACT ENGINEERING, 2010, 37 (06) : 723 - 734
  • [50] Multi-dimensional modelling of the solar wind - LISM interaction including neutrals: A Boltzmann approach
    Zank, GP
    Pauls, HL
    Williams, LL
    Hall, DT
    SOLAR WIND EIGHT - PROCEEDINGS OF THE EIGHTH INTERNATIONAL SOLAR WIND CONFERENCE, 1996, (382): : 654 - 657