Prediction of shear capacity for PBL connectors in concrete using visualization-based machine learning

Abstract

Accurately predicting the ultimate shear capacity of perfobond rib (PBL) connectors is of great significance for the design of steel–concrete composite structures. This study predicted the ultimate shear capacity of PBL using machine learning (ML) methods. Initially, a dataset comprising 233 sets of PBL push-out test data was established. To enhance data quality, an Isolation Forest was used to identify and eliminate outliers from the dataset. Subsequently, four ML models—XGBoost, DT, RF, and ANN—were trained on this dataset to predict the ultimate shear capacity of PBL. By comparing and analyzing the prediction results, XGBoost demonstrated the best predictive performance with an R² value of 0.97, outperforming other models. Then, a visual analysis, including SHAP and PDP, was conducted on the XGBoost model, revealing the contribution levels of different features to the predicted values. The analysis found that the number of perforated holes (n) had the greatest impact. Moreover, based on the analysis of visualizations, recommended ranges of values for the input features are provided to maximize the ultimate shear capacity of the PBL connectors. In comparison with traditional formulas, the trained ML models exhibit superior accuracy. The MAE of XGBoost is approximately 10% of that of the traditional formulas, and its RMSE value is less than 20% of those.

Keywords

PBL connectors / Steel–concrete composite structures / Ultimate shear capacity / Machine learning / Visualization

References

[1] Armaghani DJ, Asteris PG. A comparative study of ANN and ANFIS models for the prediction of cement-based mortar materials compressive strength. Neural Comput Appl. 2021, 33(9): 4501-4532.

[2] Asteris PG, Apostolopoulou M, Armaghani DJet al. . On the metaheuristic models for the prediction of cement-metakaolin mortars compressive strength. Metaheur Comput Appl. 2020, 1(1): 63-99

[3] Asteris PG, Skentou AD, Bardhan A, Samui P, Pilakoutas K. Predicting concrete compressive strength using hybrid ensembling of surrogate machine learning models. Cem Concr Res. 2021, 145. 106449

[4] Asteris PG, Tsavdaridis KD, Lemonis ME, Ferreira FPV, Le TT, Gantes CJ, Formisano A. AI-powered GUI for prediction of axial compression capacity in concrete-filled steel tube columns. Neural Comput Appl. 2024, 36(35): 22429-22459.

[5] Asteris PG, Sivenas T, Gkantou M, Formisano A, Le TT. Estimation of axial load-carrying capacity of elliptical concrete filled steel tubular columns using computational intelligence. J Build Eng. 2025, 112: 113738.

[6] Bardhan A, Asteris PG. Application of hybrid ANN paradigms built with nature inspired meta-heuristics for modelling soil compaction parameters. Transp Geotech. 2023, 41. 100995

[7] Cândido - Martins JPS, Costa - Neves LF, Vellasco PCGdaS. Experimental evaluation of the structural response of Perfobond shear connectors. Eng Struct. 2010, 32(8): 1976-1985.

[8] Cao ZP, Li ZW, Deng SW, Wang L, Jiang HB, Xian BX. Experimental study on interfacial shear behavior of PBL shear connector deeply embedded in UHPC. Case Stud Constr Mater. 2023, 18e02192

[9] Chen YX, Huang YK, Liu H, Liu YS, Zhang T. Ultimate bearing capacity prediction method and sensitivity analysis of PBL. Eng Appl Artif Intell. 2023, 123. 106510

[10] EN 1994-1-1:2004 (2005) Eurocode 4: Design of composite steel and concrete structures–Part 1-1: General rules and rules for buildings. European Committee for Standardization, Brussels, Belgium

[11] Fan L, Zhou ZX. New shear connectors based on PBL shear connector for composite arch members. Struct Eng Int. 2014, 242281-284.

[12] He SH, Fang Z, Zhang L, Li G, Liu M. Research on mechanical performance of PBL shear connectors for steel - concrete joint section of hybrid girder bridge. J China Railw Soc. 2015, 37(10): 100-109In Chinese

[13] Hosaka T, Mitsuki K, Hiragi Het al. . An experimental study on shear characteristics of perfobond strip and its rational strength equations. J. Struct. Eng. JSCE. 2000, 46: 1593-1604

[14] Hosaka T, Mitsugi K, Hiragi H, Ushijima H. Study on shear strength and design method of perfobond strip. Jap J Struct Eng. 2002, 48: 1265-1272

[15] JTG D64 - 2015 (2015) Specifications for design of highway steel bridge, JTG D64 - 2015 specifications for design of highway steel bridge, China Communications Press; Beijing, China. (In Chinese)

[16] Klaiber, F.W., Wipf, T.J., Nauman, J.C. and others (2000), Investigation of two bridge alternatives for low volume road - Phase II: beam in slab bridge, Research Report; USA: Iowa State University of Science and Technology

[17] Le TT, Skentou AD, Mamou A, Asteris PG. Correlating the unconfined compressive strength of rock with the compressional wave velocity, effective porosity and Schmidt hammer rebound number using artificial neural networks. Rock Mech Rock Eng. 2022, 55(11): 6805-6840.

[18] Li, S. W. (2013), Research on PBL shear connectors for steel - concrete composite segment of hybrid girder cable - stayed bridges, Ph.D. Dissertation, Huazhong University of Science and Technology, Wuhan

[19] Liao X (2022) Research on static and fatigue properties of PBL shear connectors in UHPC, Ph.D. Dissertation, Southwest Jiaotong University, Chengdu

[20] Liu FT, Ting KM, Zhou ZH (2008) Isolation forest. In: Proceedings of the 8th IEEE International Conference on Data Mining (ICDM 2008), Pisa, Italy, IEEE, Piscataway, NJ, USA, pp. 413–422.

[21] Machacek J, Studnicka J. Perforated shear connectors. Steel Compos Struct. 2002, 2(1): 51-66.

[22] Mahmood W, Mohammed AS, Sihag P, Asteris PG, Ahmed H. Interpreting the experimental results of compressive strength of hand-mixed cement-grouted sands using various mathematical approaches. Arch Civil Mech Eng. 2021, 22(1): 19.

[23] Mahmood W, Mohammed AS, Asteris PG, Kurda R, Armaghani DJ. Modeling flexural and compressive strengths behaviour of cement-grouted sands modified with water reducer polymer. Appl Sci. 2022, 1231016.

[24] Medberry SB, Shahrooz BM. Perfobond shear connector for composite construction. Eng J. 2002, 3912-12.

[25] Sarir P, Chen J, Asteris PG, Armaghani DJ, Tahir MM. Developing GEP tree-based, neuro-swarm, and whale optimization models for evaluation of bearing capacity of concrete-filled steel tube columns. Eng Comput. 2021, 37(1): 1-19.

[26] Shariati M, Ramli Sulong NH, Shariati A, Kueh ABH. Comparative performance of channel and angle shear connectors in high strength concrete composites: An experimental study. Construct Build Mater. 2016, 120: 382-392.

[27] Shi Z, Qin BC. Study on the failure mechanism and bearing capacity influence factors of two kinds of push - out tests of PBL. J Railw Sci Eng. 2019, 16(04): 943-952In Chinese

[28] Shi Q, Yang F, Guo J. Study on bearing capacity influence factors of the PBL shear connector, in IOP Conference Series: Earth and Environmental Science. IOP Publishing. 2020, 560(1012038

[29] Su QT, Wang W, Luan HW, Yang GT. Experimental research on bearing mechanism of perfobond rib shear connectors. J Constr Steel Res. 2014, 95: 22-31.

[30] Tabachnick BG, Fidell LS (2013) Using multivariate statistics (6th edn.). Pearson, Boston, MA, USA.

[31] Valente I, Cruz PJS. Experimental analysis of Perfobond shear connection between steel and lightweight concrete. J Constr Steel Res. 2004, 603–5): 465-479.

[32] Wang ZH, Li Q, Zhao CH. Ultimate shear resistance of perfobond rib shear connectors based on a modified push - out test. Adv Struct Eng. 2013, 16(4): 667-680.

[33] Wang WA, Zhao CH, Li Q, Zhuang WL. Study on load-slip characteristic curves of perfobond shear connectors in hybrid structures. J Adv Concr Technol. 2014, 12(10): 413-424.

[34] Wang C, Kou LY, Hu XM. Study of formula for the bearing capacity of perfobond connectors for steel - concrete composite beams. J Railw Sci Eng. 2015, 12(4): 892-899In Chinese

[35] Wang XW, Zhu B, Cui SA, Lui EM. Experimental research on PBL connectors considering the effects of concrete stress state and other connection parameters. J Bridge Eng. 2018, 23(1): 04017125.

[36] Wang XW, He QX, An ZW, Liu GJ, Wen XK, Wang YQ, Zhong ZX. Experimental study of perfobond rib shear connector under lateral force. Appl Sci. 2021, 11199088.

[37] Wang CX, Zou XX, Sneed LH, Zhang F, Zheng KQ, Xu H, Li GF. Shear strength prediction of FRP - strengthened concrete beams using interpretable machine learning. Constr Build Mater. 2023, 407. 133553

[38] Wang HR, Song YS, Xiao G (2024) Research on prediction method of shear capacity of lead bolt based on machine learning. In: Proceedings of the 21st Shenyang Science and Academic Annual Conference-Natural Sciences, Shenyang Jianzhu University, Shenyang, China, pp. 50-56. https://doi.org/10.26914/c.cnkihy.2024.045759. (In Chinese)

[39] Xiao L, Qiang SZ, Li XZ, Wei X. Research on mechanical performance of PBL shear connectors considering the perforated plate’s thickness. Eng Mech. 2012, 298282-288+296In Chinese

[40] Xiao L, Wei X, Qiang SZ. Comparative study on two kinds of push - out tests of PBL shear connectors. Chin Civil Eng J. 2013, 46(11): 70-80In Chinese

[41] Xiao L, Li XZ, Wei X, Qiang SZ. Experimental study on fatigue performance of PBL shear connectors. Chin Civil Eng J. 2015, 48(7): 93-101In Chinese

[42] Xue WS, Dai Y, Zhou L, Lu YC. Experimental studies on shear behavior of perfohond connectors. J Build Struct. 2009, 30(5): 103-111In Chinese

[43] Yang Y, Chen Y. Experimental study on mechanical behavior of PBL shear connectors. J Bridge Eng. 2018, 23(9): 04018062.

[44] Yang Y, Chen Y. Experimental study on the shear capacity of PBL shear connectors. Eng Mech. 2018, 35989-96In Chinese

[45] Yang Y, Chen Y, Cai JW. Experiment on static performance of perforated steel plate shear connectors. China J Highw Transp. 2017, 30(3): 255-263In Chinese

[46] Zhang F, Wang CX, Zou XX, Wei Y, Chen DD, Wang QD, Wang LB. Prediction of the shear resistance of headed studs embedded in precast steel-concrete structures based on an interpretable machine learning method. Buildings. 2023, 13(2): 496.

[47] Zhang F, Wang CX, Liu J, Zou XX, Sneed LH, Bao Y, Wang LB. Prediction of FRP - concrete interfacial bond strength based on machine learning. Eng Struct. 2023, 274. 115156

[48] Zhao C, Li Z, Deng K, Wang W. Experimental investigation on the bearing mechanism of perfobond rib shear connectors. Eng Struct. 2018, 159: 172-184.

[49] Zhou J, Asteris PG, Armaghani DJ, Pham BT. Prediction of ground vibration induced by blasting operations through the use of the Bayesian network and random forest models. Soil Dyn Earthq Eng. 2020, 139. 106390

[50] Zhou Y, Gan LY, Di SQ, He WY, Li NB. Bridge model modification experiment based on strain influence line. J Zhejiang Univ (Engineering Science). 2024, 583537-546In Chinese

[51] Zhou Y, Yu M, Li H, Chen X. Cable-stayed bridge model updating using a novel approach for deflection influence line identification: theoretical basis and field validation. Int J Struct Stab Dyn. 2025, 25. 2650227

[52] Zhou Y, Gan LY, Di SQ, Chen D, Fang DJ. Existing continuous beam bridges seismic vulnerability assessment method considering with model modification based on influence line. J Vib Eng. 2025, 38(4): 838-848In Chinese

Funding

National Natural Science Foundation of China(No.52308129)

Anhui Provincial Department of Science and Technology(2308085QE185)

Anhui Provincial Department of Education(2023AH050175)

State Key Laboratory Breeding Base of Mountain Bridge and Tunnel Engineering(No. SKLBT-ZD2301)

Doctoral Startup Foundation of Anhui Jianzhu University(2023QDZ01)

RIGHTS & PERMISSIONS

The Author(s)