Prediction of Nonlinear Stress-strain Behaviors with Artificial Neural Networks and Its Application for Automotive Rubber Parts

被引:0
|
作者
Junye Park
Cheol Kim
Hyung-seok Lee
机构
[1] Kyungpook National University,Department of Mechanical Engineering
[2] Sanyang Rubber & Chemical Co.,Research Center
[3] LTD,undefined
关键词
Artificial neural network; Rubber properties; Nonlinear stress-strain; Optimum design; Dust covers;
D O I
暂无
中图分类号
学科分类号
摘要
This study presents a new method to predict the stress-strain curves of rubber materials using artificial neural networks in order to reduce the numbers of tensile tests and shows its application. Various stress-strain curves used for the machine learning are obtained by uniaxial, biaxial, planar tension tests on the chloroprene rubber specimens. Tests are carried out at a rate of 0.01 strain/s at 23 °C, and the Mullins effect is reflected through five load-unload processes in the strain range of 0 ∼ 20 %, 0 ∼ 50 %, 0 ∼ 70 %, and 0 ∼ 100 %. After training, the stress-strain relationships in untrained ranges are predicted. The predictions are compared with the experimental data in the strain range of 0 ∼ 100 %, which was previously reserved to confirm the prediction performance. It was predicted with errors within 0.04, 0.08, and 0.01 MPa for the uniaxial, biaxial, and planar tests, respectively. These small errors indicate predictions are reliable. For optimization of rubber parts, material constants of Ogden model are obtained using the predicted data in the strain of 0 ∼ 60 % and 0 ∼ 80 %. Dust covers are optimized to reduce stresses by the Taguchi method. The maximum von Mises stresses in the optimal designs are reduced by approximately 8 % and 14 %, compared to the initial ones.
引用
收藏
页码:1481 / 1491
页数:10
相关论文
共 50 条
  • [1] PREDICTION OF NONLINEAR STRESS-STRAIN BEHAVIORS WITH ARTIFICIAL NEURAL NETWORKS AND ITS APPLICATION FOR AUTOMOTIVE RUBBER PARTS
    Park, Junye
    Kim, Cheol
    Lee, Hyung-seok
    INTERNATIONAL JOURNAL OF AUTOMOTIVE TECHNOLOGY, 2023, 24 (06) : 1481 - 1491
  • [2] Stress-strain evaluation of structural parts using artificial neural networks
    Ribeiro, Joao P. A.
    Tavares, Sergio M. O.
    Parente, Marco
    PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART L-JOURNAL OF MATERIALS-DESIGN AND APPLICATIONS, 2021, 235 (06) : 1271 - 1286
  • [3] STRESS-STRAIN MODELING OF SANDS USING ARTIFICIAL NEURAL NETWORKS
    ELLIS, GW
    YAO, C
    ZHAO, R
    PENUMADU, D
    JOURNAL OF GEOTECHNICAL ENGINEERING-ASCE, 1995, 121 (05): : 429 - 435
  • [4] Stress-strain modeling of sands using artificial neural networks - Discussion
    Najjar, YM
    Basheer, IA
    JOURNAL OF GEOTECHNICAL ENGINEERING-ASCE, 1996, 122 (11): : 949 - 951
  • [5] Stress-strain state of rubber-cord machine parts
    Sokolov S.L.
    Ushakov B.N.
    Russian Engineering Research, 2009, 29 (05) : 463 - 468
  • [6] Artificial Neural Network Approach for Prediction of Titanium alloy Stress-Strain Curve
    Srinivasu, G.
    Rao, R. N.
    Nandy, T. K.
    Bhattacharjee, Amit
    INTERNATIONAL CONFERENCE ON MODELLING OPTIMIZATION AND COMPUTING, 2012, 38 : 3709 - 3714
  • [7] Stress-strain identification for cottonseed and castor bean based on artificial neural networks
    Zheng Xiao
    He Dongping
    Lin Guoxiang
    Wang Jingzhou
    ADVANCED COMPUTER TECHNOLOGY, NEW EDUCATION, PROCEEDINGS, 2007, : 3 - 6
  • [8] Stress-Strain Behaviors and Crosslinked Networks Studies of Natural Rubber-Zinc Dimethacrylate Composites
    Chen, Yukun
    Xu, Chuanhui
    JOURNAL OF MACROMOLECULAR SCIENCE PART B-PHYSICS, 2012, 51 (07): : 1384 - 1400
  • [9] Approximation Of Stress-Strain Curve Of Rubber-Like Material Using An Artificial Neural Network
    Vodka, Oleksii
    Pogrebnyak, Sergii
    2020 IEEE KHPI WEEK ON ADVANCED TECHNOLOGY (KHPI WEEK), 2020, : 221 - 226
  • [10] Prediction of composite microstructure stress-strain curves using convolutional neural networks
    Yang, Charles
    Kim, Youngsoo
    Ryu, Seunghwa
    Gu, Grace X.
    MATERIALS & DESIGN, 2020, 189