Machine learning-based prediction of the structural performance of a long-bolted steel connection

Abstract

In response to the housing shortage in Canada, particularly in northern and remote communities, modular houses have emerged as a viable solution. These prefabricated structures offer speed, cost-efficiency, and flexibility. To enhance the durability and functionality of these modular homes, innovative construction techniques are being explored. A new bolted connection, utilizing high-strength long bolts, has been introduced for hollow structural sections (HSS), which can be designed using regression models trained by an experimentally validated finite element model (FEM). This study employs machine learning techniques, including neural networks, genetic regression, and decision trees, to detect the failure mode and predict the ultimate moment capacity of HSS moment connections under monotonic loading. A nonlinear validated FEM was developed using LS-DYNA software, and a matrix of 240 FEMs was generated to train and test the machine learning models, including a range of various design parameters such as the extended plate thickness, number of bolts, bolt arrangement, and bolt diameter. Five machine learning algorithms were used for classification and regression learning, with hyperparameter optimization applied to enhance their accuracy. Mathematical formulas for predicting the ultimate moment capacity were developed using genetic algorithm-based symbolic regression, trained on 70% of the matrix parameters. These formulas were then validated and tested with the remaining 30%, demonstrating high accuracy. Findings illustrate the efficiency of machine learning approaches for precisely predicting the ultimate capacity and failure patterns of bolted connections, highlighting their promise as reliable tools in design, complementing both experimental and analytical methods.

Keywords

Machine learning / Regression / Artificial neural network / Failure mode / Finite element model / Steel connection

References

[1] Abarkan I, Rabi M, Ferreira FPV, Shamass R, Limbachiya V, Jweihan YS, Pinho Santos LF. Machine learning for optimal design of circular hollow section stainless steel stub columns: A comparative analysis with Eurocode 3 predictions. Engineering Applications of Artificial Intelligence. 2024, 132. 107952

[2] Akbari Hamed A, Charkhtab Basim M. Experimental-numerical study on weakened HSS-to-HSS connections using HBS and RBS approaches. Structures. 2020, 28: 1449-1465.

[3] Asgarkhani N, Kazemi F, Jakubczyk-Gałczyńska A, Mohebi B, Jankowski R. Seismic response and performance prediction of steel buckling-restrained braced frames using machine-learning methods. Engineering Applications of Artificial Intelligence. 2024.

[4] Asgarkhani N, Kazemi F, Jankowski R. Machine learning-based prediction of residual drift and seismic risk assessment of steel moment-resisting frames considering soil-structure interaction. Computers and Structures. 2023, 289. 107181

[5] Chen H, Liu Y, Huang Y, Huang J, Caisong L, Guo Z. Maximum displacement prediction model for steel beams with hexagonal web openings under impact loading based on artificial neural networks. Engineering Applications of Artificial Intelligence. 2024, 136. 108932

[6] Chen Y, Huang Y, Liu H, Liu Y, Zhang T. Ultimate bearing capacity prediction method and sensitivity analysis of PBL. Engineering Applications of Artificial Intelligence. 2023, 123. 106510

[7] Das S, Teweldebrhan BT, Tesfamariam S. High-dimensional multi-objective optimization of coupled cross-laminated timber walls building using deep learning. Engineering Applications of Artificial Intelligence. 2024, 136. 109055

[8] DataRobot. (2024). Eureqa. https://www.datarobot.com

[9] Digital Research Alliance of Canada. (2021). Digital Research Alliance of Canada. https://alliancecan.ca/en

[10] Du W, Zhao Y, Wang Y, Wang H, Yang M. Novel machine learning approach for shape-finding design of tree-like structures. Computers and Structures. 2022, 261–262. 106731

[11] Elhadary, M., Bediwy, A., & Elshaer, A. (2024a). Experimental investigation on a novel steel connection for modular indigenous houses. Canadian Society of Civil Engineering (CSCE) 2024 Annual Conference.

[12] Elhadary M, Bediwy A, Elshaer A. Novel steel connection for modular houses in indigenous communities: An experimental study. Journal of Constructional Steel Research. 2024, 220. 108850

[13] Elhadary, M., Bediwy, A., & Elshaer, A. (2025). Structural performance of novel steel connection for indigenous modular houses. In Lecture Notes in Civil Engineering (pp. 253–266). Springer Singapore. https://doi.org/10.1007/978-3-031-61535-1_21

[14] Fadden, M., Asce, S. M., Mccormick, J., & Asce, A. M. (2012). Cyclic quasi-static testing of hollow structural section beam members. 138, 561–570. https://doi.org/10.1061/(ASCE)ST.1943-541X.0000506

[15] Fadden, M., & McCormick, J. (2013). Evaluation of HSS-to-HSS moment connections for seismic applications. Structures congress 2013: Bridging your passion with your profession—Proceedings of the 2013 structures congress, 2334–2345. https://doi.org/10.1061/9780784412848.204

[16] Fadden M, Wei D, McCormick J. Cyclic testing of welded HSS-to-HSS moment connections for seismic applications. Journal of Structural Engineering. 2015, 141(2): 1-14.

[17] HeuristicLab. (2002). https://dev.heuristiclab.com/trac.fcgi/

[18] Hu J, Jiang L, Hu Y, He J, Cheng X, Yang J. Machine learning-based axial compressive capacity estimation of cold-formed steel build-up sections. Thin-Walled Structures. 2025.

[19] Junda E, Málaga-chuquitaype C. Seismic acceleration demands in tall CLT buildings, predictive models and intensity measures. Engineering Structures. 2024, 298. 117024

[20] Kazemi F, Asgarkhani N, Jankowski R. Machine learning-based seismic response and performance assessment of reinforced concrete buildings. Archives of Civil and Mechanical Engineering. 2023, 23(2): 1-21.

[21] Kazemi F, Jankowski R. Machine learning-based prediction of seismic limit-state capacity of steel moment-resisting frames considering soil-structure interaction. Computers and Structures. 2023, 274. 106886

[22] Kosteski N, Packer JA. Longitudinal plate and through plate-to-hollow structural section welded connections. Journal of Structural Engineering. 2003, 129(4): 478-486.

[23] Kosteski N, Packer JA. Welded Tee-to-HSS Connections. 2003, 129: 151-159

[24] Lai D, Demartino C, Xiao Y. Probabilistic machine leaning models for predicting the maximum displacements of concrete-filled steel tubular columns subjected to lateral impact loading. Engineering Applications of Artificial Intelligence. 2024, 135(May. 108704

[25] Lawson, R. M., Prewer, J., & Trebilcock, P. J. (1999). Modular construction using light steel framing. Steel Construction Institute.

[26] Li Y, Lou J, Feng Q, Wang G, Xu R. Prediction of the load–moment interaction curve for rectangular concrete-filled steel tube columns using composite multi-fidelity neural network. Engineering Applications of Artificial Intelligence. 2025, 142. 109904

[27] Liu XC, Cui FY, Zhan XX, Yu C, Jiang ZQ. Seismic performance of bolted connection of H-beam to HSS-column with web end-plate. Journal of Constructional Steel Research. 2019, 156: 167-181.

[28] Liu XC, Wang Y, Cui XX, Yu C, Bai ZX. Seismic performance of bolted beam-to-column connection with rib-stiffened splicing plate. Journal of Constructional Steel Research. 2020, 174. 106300

[29] Liu XC, Yang ZW, Wang HX, Zhang AL, Pu SH, Chai ST, Wu L. Seismic performance of H-section beam to HSS column connection in prefabricated structures. Journal of Constructional Steel Research. 2017, 138: 1-16.

[30] Liu Y, Ji W, Li J, Liu SB, Yang W. Shear performance prediction for corrugated steel web girders based on machine-learning algorithms. Thin-Walled Structures. 2025, 206. 112668

[31] Livermore software technology corporation (LSTC). (2007). Livermore software technology corporation (LSTC). (2007). LS-DYNA keyword user’s manual, Vol. 1, Livemore, California, USA, R13.0.

[32] Lu Y, Wu B, Li W, Zhou T, Li Y. Regression-classification ensemble machine learning model for loading capacity and bucking mode prediction of cold-formed steel built-up I-section columns. Thin-Walled Structures. 2024, 205. 112427

[33] MATLAB version: 9.13.0. (2022). The MathWorks Inc. https://www.mathworks.com/products/matlab.html

[34] Modular construction reports & industry analysis. (2023). https://greenbuildingelements.com/modular-construction-reports-industry-analysis/

[35] Mustafa R. Prediction of crippling load of I—shaped steel columns by using soft computing techniques. AI in Civil Engineering. 2024.

[36] Nguyen HD, Lee YJ, LaFave JM, Shin M. Seismic fragility analysis of steel moment frames using machine learning models. Engineering Applications of Artificial Intelligence. 2023, 126. 106976

[37] Ni X, Duan K. Machine learning-based models for shear strength prediction of UHPFRC beams. Mathematics. 2022, 1016): 1-26.

[38] Park K, Jiansinlapadamrong C, Chao S-H. Double-HSS seismic resistant beam-to-column moment connections. Journal of Structural Engineering. 2021, 147(7): 1-17.

[39] Rabi M, Ferreira FPV, Abarkan I, Limbachiya V, Shamass R. Prediction of the cross-sectional capacity of cold-formed CHS using numerical modelling and machine learning. Results in Engineering. 2023, 17. 100902

[40] Rahman T, Momin F, Provasha AA. Comprehensive analysis of structural parameters influencing the fundamental period of steel—Braced RC buildings using machine learning interpretability. AI in Civil Engineering. 2025.

[41] Su A, Cheng J, Li X, Zhong Y, Li S, Zhao O, Jiang K. Unified machine-learning-based design method for normal and high strength steel I-section beam–columns. Thin-Walled Structures. 2024, 199. 111835

[42] United Nations. (2019). Adequate housing as a component of the right to an adequate standard of living, and the right to non-discrimination in this context.

[43] Xu Y, Zhang M, Zheng B. Design of cold-formed stainless steel circular hollow section columns using machine learning methods. Structures. 2021, 33(May): 2755-2770.

[44] Xu Y, Zheng B, Zhang M. Capacity prediction of cold-formed stainless steel tubular columns using machine learning methods. Journal of Constructional Steel Research. 2021, 182. 106682

[45] Zhang B, Xue X. Ultimate axial strength prediction of concrete-filled double-skin steel tube columns using soft computing methods. Engineering Applications of Artificial Intelligence. 2024, 1292023. 107676

[46] Zhang J, Huang Y, Ma G, Nener B. Mixture optimization for environmental, economical and mechanical objectives in silica fume concrete: A novel frame-work based on machine learning and a new meta-heuristic algorithm. Resources, Conservation and Recycling. 2021, 167. 105395

Funding

NSERC-DG(RGPIN-2022-04755)

RIGHTS & PERMISSIONS

The Author(s)