A machine learning-based calibration method for strength simulation of self-piercing riveted joints

Abstract

This paper presents a new machine learning-based calibration framework for strength simulation models of self-piercing riveted (SPR) joints. Strength simulations were conducted through the integrated modeling of SPR joints from process to performance, while physical quasi-static tensile tests were performed on combinations of DP600 high-strength steel and 5754 aluminum alloy sheets under lap-shear loading conditions. A sensitivity study of the critical simulation parameters (e.g., friction coefficient and scaling factor) was conducted using the controlled variables method and Sobol sensitivity analysis for feature selection. Subsequently, machine-learning-based surrogate models were used to train and accurately represent the mapping between the detailed joint profile and its load-displacement curve. Calibration of the simulation model is defined as a dual-objective optimization task to minimize errors in key load displacement features between simulations and experiments. A multi-objective genetic algorithm (MOGA) was chosen for optimization. The three combinations of SPR joints illustrated the effectiveness of the proposed framework, and good agreement was achieved between the calibrated models and experiments.

Keywords

Machine learning / Self-piercing riveting (SPR) / Sensitivity analysis / Multi-objective optimization

References

[1] Zhang W, Xu J. Advanced lightweight materials for automobiles: a review. Mater Design, 2022, 221: 110994.

[2] Li D, Chrysanthou A, Patel I, et al. Self-piercing riveting‒a review. Int J Adv Manuf Tech, 2017, 92: 1777-1824.

[3] Li M, Liu Z, Huang L, et al. Automatic identification framework of the geometric parameters on self-piercing riveting cross-section using deep learning. J Manuf Process, 2022, 83: 427-437.

[4] Li M, Liu Z, Huang L, et al. Multi-fidelity data-driven optimization design framework for self-piercing riveting process parameters. J Manuf Process, 2023, 99: 812-824.

[5] Sui B, Du D, Chang B, et al. Simulation and analysis of self-piercing riveting process in aluminum sheets. Mater Sci Tech-Lond, 2007, 5: 713-717.

[6] Wan S, Hu S, Li SY, et al. Process parameters and joint evaluation of self-piercing riveting with half-hollow rivets. J Tianjin Univ, 2007, 04: 494-498.

[7] Liu XQ (2007) Experimental investigation and finite element numerical simulation of self-piercing riveted process. Dissertation, Tianjin University

[8] Bouchard PO, Laurent T, Tollier L. Numerical modeling of self-pierce riveting—from riveting process modeling down to structural analysis. J Mater Process Tech, 2008, 202(1/3): 290-300.

[9] Casalino G, Rotondo A, Ludovico A. On the numerical modelling of the multiphysics self piercing riveting process based on the finite element technique. Adv Eng Softw, 2008, 39(9): 787-795.

[10] Kang SW, Huang ZC, Chen XM. Introduction of fretting in self-piercing riveting. Coal Mine Machinery, 2008, 06: 7-9.

[11] Grujicic M, Snipes J, Ramaswami S, et al. Process modeling, joint-property characterization and construction of joint connectors for mechanical fastening by self-piercing riveting. Multidiscip Model Ma terials and structures, 2014, 10(4): 631-658.

[12] Hanssen A, Olovsson L, Porcaro R, et al. A large-scale finite element point-connector model for self-piercing rivet connections. Eur J Mech A-Solid, 2010, 29(4): 484-495.

[13] Chung CS, Kim HK. Fatigue strength of self-piercing riveted joints in lap-shear specimens of aluminium and steel sheets. Fatigue Fract Eng M, 2016, 39(9): 1105-1114.

[14] Yan W, Xie Z, Yu C, et al. Experimental investigation and design method for the shear strength of self-piercing rivet connections in thin-walled steel structures. J Constr Steel Res, 2017, 133: 231-240.

[15] Bock FE, Blaga LA, Klusemann B. Mechanical performance prediction for friction riveting joints of dissimilar materials via machine learning. Proc Manuf, 2020, 47: 615-622.

[16] Zhao H, Han L, Liu Y, et al. Quality prediction and rivet/die selection for SPR joints with artificial neural network and genetic algorithm. J Manuf Process, 2021, 66: 574-594.

[17] Jäckel M, Coppieters S, Hofmann M, et al. (2017) Mechanical joining of materials with limited ductility: analysis of process-induced defects. AIP Conf Proc, 1896, 1: 110009.

[18] Kang H, Lee JH, Choe Y, et al. Prediction of lap shear strength and impact peel strength of epoxy adhesive by machine learning approach. Nanomaterials, 2021, 11(4): 872.

[19] Qi BF (2020) Crashworthiness analysis of heterogeneous metal automobile front rail. Dissertation, Dalian University of Technology.

[20] Xie Y (2019) Research on the performance of self-piercing riveting and its application on front longero simulation. Dissertation, Hefei University of Technology

[21] Xu JS (2018) Simulation study on failure of self-piercing riveted joints between steel and aluminum. Dissertation, Jilin University

[22] Zhang ZY, Shi BJ, Zhong JB. Optimisation of parameter matching of semi-hollow self-pierce riveted joints in aluminium alloy car bodies. Mach Design Manuf Eng, 2019, 48(3): 64-68.

[23] Fang Y, Huang L, Zhan Z, et al. A framework for calibration of self-piercing riveting process simulation model. J Manuf Process, 2022, 76: 223-235.

[24] Haque R. Quality of self-piercing riveting (SPR) joints from cross-sectional perspective: a review. Arch Civ Mech Eng, 2018, 18(1): 83-93.

[25] Ma Y, Lou M, Li Y, et al. Effect of rivet and die on self-piercing rivetability of AA6061-T6 and mild steel CR4 of different gauges. J Mater Process Tech, 2018, 251: 282-294.

[26] Huang L, Lasecki JV, Guo H, et al. Finite element modeling of dissimilar metal self-piercing riveting process. SAE Int J Mater Manu, 2014, 7(3): 698-705.

[27] Huang L, Wu Y, Huff G et al (2020) Sensitivity study of self-piercing rivet insertion process using smoothed particle Galerkin method. In: The 16th international LS-DYNA conference, 10–11 June, pp 1–11

[28] Yue Z, Chen Q, Huang L, et al. A surrogate model based calibration method for structural adhesive joint progressive failure simulations. J Adhesion, 2023, 99(10): 1579-1606.

[29] Bhadeshia H. Neural networks and information in materials science. Stat Anal Data Min, 2009, 1(5): 296-305.

[30] Li M, Liu Z, Huang L, et al. A new multi-fidelity surrogate modelling method for engineering design based on neural network and transfer learning. Eng Computation, 2022, 39(6): 2209-2230.

[31] Efron B, Stein C. The Jackknife estimate of variance. Ann Stat, 1981, 9(3): 586-596.

[32] Soboĺ I. Sensitivity estimates for nonlinear mathematical models. Math Model Comput Exp, 1993, 1: 407.

[33] Wan H, Ren W, Wang N. Parameter selection and sampling methods for global sensitivity analysis of Gaussian process models. J Vib Eng, 2015, 28(05): 714-720.

[34] McKay MD, Beckman RJ, Conover WJ. A comparison of three methods for selecting values of input variables in the analysis of output from a computer code. Technometrics, 2000, 42(1): 55-61.

[35] Gunantara N. A review of multi-objective optimization: methods and its applications. Cogent Eng, 2018, 5(1): 1502242.

[36] Marler RT, Arora JS. Survey of multi-objective optimization methods for engineering. Struct Multidiscip O, 2004, 26: 369-395.

[37] Gao Y, Shi L, Yao P (2000) Study on multi-objective genetic algorithm. In: Proceedings of the 3rd world congress on intelligent control and automation, IEEE, 28 June–2 July, Hefei

[38] Huang L, Guo H, Shi Y, et al. Fatigue behavior and modeling of self-piercing riveted joints in aluminum alloy 6111. Int J Fatigue, 2017, 100: 274-284.

[39] Jiang P, Zhou Q, Shao X, et al. Surrogate-model-based design and optimization, 2020, Singapore: Springer.

Funding

National Natural Science Foundation of China http://dx.doi.org/10.13039/501100001809(52205377)

Key Technologies Research and Development Program http://dx.doi.org/10.13039/501100012165(2022YFB4601804)

Key Basic Research Project of Suzhou(#SJC2022031)