Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost Algorithm

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost Algorithm

by

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost AlgorithmReference Paper

Jianping Lin, Chengwei Qi, Hailang Wan, et al. Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Using Finite Element Simulation and XGBoost Algorithm. Chinese Journal of Mechanical Engineering, 2021 34: 36.

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost AlgorithmResearch Background and Purpose

The use of high-strength aluminum alloys is an important means of lightweighting in automobiles. However, due to the high cost of aluminum alloys, it is still necessary to use high-strength steel in some parts to save costs, leading to the emergence of steel/aluminum hybrid bodies. Self-piercing riveting (SPR) is an effective means of connecting dissimilar materials of steel and aluminum, widely used in lightweight body structure connections. However, the design of the riveting process involves deformation and hardening of the rivets and connected materials, with many factors affecting the joint strength and unclear mechanisms, resulting in significant uncertainty in connection strength and a lack of effective methods to determine the strength of self-piercing riveted joints. This paper focuses on the self-piercing riveted joints of steel/aluminum material combinations, studying the prediction method of cross-tension strength based on finite element simulation and algorithmic approaches.
Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost AlgorithmExperimental Methods

This study uses CR590 steel with a thickness of 1.1mm, GMW-S-6000-IH aluminum alloy with a thickness of 1.2mm, and AA6022-T4 aluminum alloy with a thickness of 2mm as the research objects. Mechanical property experiments are conducted on a universal testing machine MTS E45.105 to obtain the mechanical properties of the materials for finite element simulation.

Self-piercing riveting experiments are performed on a servo lock riveting device ESPR-80, with a riveting speed of 1mm/s controlled by a servo motor and a termination force of 50kN controlled by a force sensor. Plate samples are processed by laser cutting to create self-piercing riveting samples, and the joint cross-section morphology is obtained using a metallographic cutting machine and Keyence VHX-6000 digital microscope system. Cross-tension experiments are conducted on a universal testing machine MTS E45.105 using a peeling experimental fixture, with a peeling speed set at 10mm/min to obtain the joint peeling strength.

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost AlgorithmResults

(1) The self-piercing riveting process and the mechanical properties of the joint cross-tension strength tests are simulated jointly using Simufact forming software. The simulated cross-sectional geometry obtained from the riveting process matches the experimental results, and the simulated joint peeling strength agrees with the experimental results, indicating a certain accuracy of the simulation. It can effectively obtain the cross-tension strength of the steel/aluminum combination joints discussed in this paper. Finite element simulation can serve as an effective means to acquire joint cross-sectional size data and self-piercing riveted joint strength data.

(2) The XGBoost algorithm can effectively achieve regression prediction of the peeling strength of self-piercing riveted joints based on substrate material parameters.

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost AlgorithmConclusions

(1) The finite element model that considers the hardening phenomenon of the connected materials and rivets during the riveting process can better simulate the testing process of the joint cross-tension strength.

(2) By calibrating the finite element model through experiments, the finite element method can obtain self-piercing riveted joint strength data, and using the XGBoost algorithm for regression prediction can effectively predict the strength of self-piercing riveted joints.

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost AlgorithmProspects and Applications

Aluminum alloys, as a lightweight material, have been widely used in the automotive industry. The increasing number of steel/aluminum hybrid body structures makes their connection strength crucial for the structural integrity of the vehicle body. This paper proposes a method for predicting the strength of steel/aluminum self-piercing riveted joints, which can provide guidance for the application of self-piercing riveting technology in vehicle bodies.

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost AlgorithmRelated Articles/Book Recommendations

[1] Lin, J., Sun, C., Min, J., Wan, H., & Wang, S.(2020). Effect of atmospheric pressure plasma treatment on surface physicochemical properties of carbon fiber reinforced polymer and its interfacial bonding strength with adhesive. Composites Part B: Engineering, 108237.

[2] Min, J., Wan, H., Carlson, B. E., Lin, J., & Sun, C. (2020). Application of laser ablation in adhesive bonding of metallic materials: A review. Optics & Laser Technology, 128, 106188.

[3] Min, J., Zhang, K., Wang, S., et al. Research on the effect of rivet structure on the penetration force of stir friction single-sided riveting of aluminum alloy plates. Journal of Mechanical Engineering, 2020, 56(6): 159-168.

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost AlgorithmTeam Leader Introduction

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost Algorithm

Professor Jianping Lin, Ph.D., doctoral supervisor, reviewer for multiple SCI international journals. He has undertaken over 100 key projects for the nation, Shanghai, and enterprises, receiving several awards for scientific and technological progress, the first prize for national teaching achievements, excellent teacher awards, and awards for guiding outstanding doctoral theses. He has applied for 26 national patents and published over 300 papers in domestic and international journals, including over 170 indexed by SCI/EI. The main research directions of his team include: vehicle lightweight design and manufacturing technology, modern forming technology, and lightweight material connection technology.

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost AlgorithmAuthor Introduction

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost Algorithm

Professor Junying Min, Humboldt Scholar, Ph.D., doctoral supervisor. Selected for the overseas high-level talent introduction program’s youth project, CIRP young member (2015-present), guest editor/editorial board member of Automotive Innovation, expert of the China Automotive Lightweight Technology Innovation Strategic Alliance, deputy director of the Automotive Body Mould and Equipment Committee of the China Die and Mould Industry Association, reviewer for over ten SCI international journals, and has published more than 60 SCI papers. The main research directions of his team include: advanced forming manufacturing technology, plastic forming mechanics and advanced characterization technology, and advanced connection technology.

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost AlgorithmTeam Publications in the Last Two Years

2020:

[1] Hou, Y., Min, J., Stoughton, T. B., Lin, J., Carsley, J. E., & Carlson, B. E.(2020). A non-quadratic pressure-sensitive constitutive model under non-associated flow rule with anisotropic hardening: Modeling and validation. International Journal of Plasticity, 135, 102808.

[2] Lin, J., Sun, C., Min, J., Wan, H., & Wang, S.(2020). Effect of atmospheric pressure plasma treatment on surface physicochemical properties of carbon fiber reinforced polymer and its interfacial bonding strength with adhesive. Composites Part B: Engineering, 108237.

[3] Min, J., Wan, H., Carlson, B. E., Lin, J., & Sun, C. (2020). Application of laser ablation in adhesive bonding of metallic materials: A review. Optics & Laser Technology, 128, 106188.

[4] Ni, J., Min, J., Wan, H., Lin, J., Wang, S., & Wan, Q. (2020). Effect of adhesive type on mechanical properties of galvanized steel/SMC adhesive-bonded joints. International Journal of Adhesion and Adhesives, 97, 102482.

[5] Wan, H., Min, J., Zhang, J., Lin, J., & Sun, C.(2020). Effect of adherend deflection on lap-shear tensile strength of laser-treated adhesive-bonded joints. International Journal of Adhesion and Adhesives, 97, 102481.

[6] Min, J., Guo, N., Hou, Y., Jiang, K., Chen, X., Carsley, J. E., & Lin, J.(2020). Effect of tension-compression testing strategy on kinematic model calibration and springback simulation of advanced high strength steels. International Journal of Material Forming, 1-14.

[7] Lin, Y., Min, J., Teng, H., Lin, J., Hu, J., & Xu, N.(2020). Flexural Performance of Steel–FRP Composites for Automotive Applications. Automotive Innovation, 1-16.

[8] Tekkaya, A. E., & Min, J. (2020). Special Issue on Automotive Lightweight. Automotive Innovation, 1-2.

[9] Zhang, W., Cai, W., Min, J., Fleischer, J., Ehrmann, C., Prinz, C., & Kreimeier, D. (2020). 5G and AI Technology Application in the AMTC Learning Factory. Procedia Manufacturing, 45, 66-71.

[10] Min, J., Zhang, K., Wang, S., et al. Research on the effect of rivet structure on the penetration force of stir friction single-sided riveting of aluminum alloy plates. Journal of Mechanical Engineering, 2020, 56(6): 159-168.

2019:

[11] Hou, Y., Min, J., Guo, N., Lin, J., et al. Investigation of evolving yield surfaces of dual-phase steels. Journal of Materials Processing Technology, 2019.

[12] Lin, J., Hou, Y., Min, J., Tang, H., Carsley, J. E., Stoughton, T. B. Effect of constitutive model on springback prediction of MP980 and AA6022-T4. International Journal of Material Forming, 2019.

[13] Lin, Y., Min, J., Li, Y., Lin, J. A thin-walled structure with tailored properties for axial crushing. International Journal of Mechanical Sciences, 2019:157, 119-135.

[14] Sun, C., Min, J., Lin, J., Wan, H. (2018). Effect of Atmospheric Pressure Plasma Treatment on Adhesive Bonding of Carbon Fiber Reinforced Polymer. Polymers, 11(1), 139.

[15] Wang, S., Min, J., Lin, J., Wu, Y. Effect of neutral salt spray (NSS) exposure on the lap-shear strength of adhesive-bonded 5052 aluminum alloy (AA5052) joints. Journal of Adhesion Science and Technology, 2019, 33(5): 549-560.

[16] Zhu, C., Wan, H., Min, J., Mei, Y., Lin, J., Carlson, B. E., Maddela, S. Application of pulsed Yb:Fiber laser to surface treatment of Al alloys for improved adhesive bonded performance. Optics and Lasers in Engineering, 2019, 119: 65-76.

[17] Volk, W., Groche, P., Brosius, A., Ghiotti, A., Kinsey, B. L., Liewald, M., Min, J., Yanagimoto, J. (2019). Models and modelling for process limits in metal forming. CIRP Annals.

[18] Wang, S., Min, J., Lin, J., Wan, H., Wang, Y. Flow drill riveting of carbon fiber-reinforced polymer and aluminum alloy sheets. Welding in the World, 2019, 63(4): 1013-1024.

[19] Nan, Guo, Junying Min, Yong Hou, Jianping Lin, John E. Carsley, Thomas B. Stoughton. An experimental study on yield surface evolution of a trip-assisted steel. NUMIFORM 2019.

[20] Zhang, K., Min, J., Lin, J., Wang, Y., Zhu, F., Wu, Y. Study on the mechanical properties of self-piercing riveted joints of CFRP and AA6022-T4 aluminum alloy. Forging Technology, 2019, 44(1): 150-156.

For more detailed content, please click Read the Original at the end of the article.

Editor: Jin Cheng Proofreader: Zhang Tong

Seeking Partners

There is a kind of cooperation called sincerity, a kind of development that can be infinite, and a kind of partnership that can achieve mutual benefit and win-win results. May our cooperation be enjoyable and our development promising. For paper recommendations, team introductions, book publications, academic live broadcasts, recruitment information, conference promotions, etc., please contact us.

Thank you for following us! The editorial department of the Chinese Journal of Mechanical Engineering will strive to create an academic media that isattitudinal, profound, and warm!

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost Algorithm

Recommended Reading

CJME latest SCI impact factor 1.824, 70/130

Latest | Overview of SCI journals in the field of mechanical engineering (2019 JCR)

Normal force and sag resistance of pipe conveyor | CJME Paper Recommendation

Data monitoring and analysis in data-driven manufacturing processes | CJME Paper Recommendation

Research progress on bionic anti-solid particle erosion materials: Learning from nature, surpassing nature | CJME Paper Recommendation

Efficient preparation of nickel-based composite coatings with high (220) preferred orientation | CJME Paper Recommendation

New impact bulging forming technology of cavitation jet | CJME Paper Recommendation

Copyright Statement:

This article is an original content of the editorial department of the Chinese Journal of Mechanical Engineering. Reproduction is welcome, please contact for authorization!

For reprinting, please leave the article title and the public account ID in the background of the public account to obtain authorization!

Contact Us:

Contact: Nuan Nuan

Phone: 010-88379909

E-mail: jme@cmes.org

Website: http://www.cjmenet.com.cn

Official WeChat: jmewechat

Long press the image to identify the QR code, and follow me!

Prediction of Cross-Tension Strength of Self-Piercing Riveted Joints Based on Finite Element Simulation and XGBoost Algorithm

Leave a Comment