Advances in concrete construction

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Optimized machine learning algorithms for predicting the punching shear capacity of RC flat slabs

  • Huajun Yan (School of Civil Engineering, Beijing Jiaotong University) ;
  • Nan Xie (School of Civil Engineering, Beijing Jiaotong University) ;
  • Dandan Shen (SANY Heavy Industry Co., LTD)
  • Received : 2023.07.30
  • Accepted : 2024.06.21
  • Published : 2024.01.25
⟨ Previous Next ⟩

Abstract

Reinforced concrete (RC) flat slabs should be designed based on punching shear strength. As part of this study, machine learning (ML) algorithms were developed to accurately predict the punching shear strength of RC flat slabs without shear reinforcement. It is based on Bayesian optimization (BO), combined with four standard algorithms (Support vector regression, Decision trees, Random forests, Extreme gradient boosting) on 446 datasets that contain six design parameters. Furthermore, an analysis of feature importance is carried out by Shapley additive explanation (SHAP), in order to quantify the effect of design parameters on punching shear strength. According to the results, the BO method produces high prediction accuracy by selecting the optimal hyperparameters for each model. With R2 = 0.985, MAE = 0.0155 MN, RMSE = 0.0244 MN, the BO-XGBoost model performed better than the original XGBoost prediction, which had R2 = 0.917, MAE = 0.064 MN, RMSE = 0.121 MN in total dataset. Additionally, recommendations are provided on how to select factors that will influence punching shear resistance of RC flat slabs without shear reinforcement.

Keywords

References

  1. Abuodeh, O.R., Abdalla, J.A. and Hawileh, R.A. (2020), "Assessment of compressive strength of Ultra-high Performance Concrete using deep machine learning techniques", Appl. Soft. Comput., 95, 106552. https://doi.org/10.1016/j.asoc.2020.106552
  2. ACI Committee (2019), Building Code Requirements for Structural Concrete (ACI 318-19) and Commentary, American Concrete Institute, Farmington Hills, MI, USA.
  3. Alkaissy, M., Arashpour, M., Golafshani, E.M., Hosseini, M.R., Khanmohammadi, S., Bai, Y. and Feng, H.M. (2023), "Enhancing construction safety: Machine learning-based classification of injury types", Safety. Sci., 162, 106102. https://doi.org/10.1016/j.ssci.2023.106102
  4. Bakouregui, A.S., Mohamed, H.M., Yahia, A. and Benmokrane, B. (2021), "Explainable extreme gradient boosting tree-based prediction of load-carrying capacity of FRP-RC columns", Eng. Struct., 245, 112836. https://doi.org/10.1016/j.engstruct.2021.112836
  5. Cao, Y., Qiu, R.Z. and Qi, W. (2023), "Simulating the performance of the reinforced concrete beam using artificial intelligence", Adv. Concrete Constr., Int. J., 15(4), 269-286. https://doi.org/10.12989/acc.2023.15.4.269
  6. CEB-FIP (2001), Punching of structural concrete slabs, CEB-Bull, 12, pp. 215-288.
  7. Chalupka, K., Williams, C.K.I. and Murray, I. (2013), "A Framework for Evaluating Approximation Methods for Gaussian Process Regression", J. Mach. Learn. Res., 14, 333-350.
  8. Christopher, C.G., Pachaivannan, P. and Elamparithi, P.N. (2023), "Study on self-compacting polyester fiber reinforced concrete and strength prediction using ANN", Adv. Concrete Constr., Int. J., 15(2), 85-96. https://doi.org/10.12989/acc.2023.15.2.085
  9. Derogar, S., Ince, C., Yatbaz, H.Y. and Ever, E. (2022), "Prediction of punching shear strength of slab-column connections: A comprehensive evaluation of machine learning and deep learning based approaches", Mech. Adv. Mater. Struct. https://doi.org/10.1080/15376494.2022.2134950
  10. European Committee for Standardization (2004), Design of Concrete Structures (Eurocode 2): Part 1-1: General Rules and Rules for Buildings, British Standards Institution, Brussels, Belgium.
  11. Faridmehr, I., Nehdi, M.L. and Baghban, M. (2022), "Novel informational bat-ANN model for predicting punching shear of RC flat slabs without shear reinforcement", Eng. Struct., 256, 114030. https://doi.org/10.1016/j.engstruct.2022.114030
  12. Federation International du Beton (2010), FIB Model Code for Concrete Structures (Model Code), Wiley-Blackwell, Berlin, Germany.
  13. Feng, D.C., Wang, W.J., Mangalathu, S., Hu, G. and Wu, T. (2021), "Implementing ensemble learning methods to predict the shear strength of RC deep beams with/without web reinforcements", Eng. Struct., 235, 111979. https://doi.org/10.1016/j.engstruct.2021.111979
  14. Feng, J.P., Zhang, H.W., Gao, K., Liao, Y.C., Yang, J. and Wu, G. (2023), "A machine learning and game theory-based approach for predicting creep behavior of recycled aggregate concrete", Case Stud. Constr. Mater., 17, e01653. https://doi.org/10.1016/j.cscm.2022.e01653
  15. Gupta, M., Raj, R. and Sahu, A.K. (2022), "EDNN based prediction of strength and durability properties of HPC using fibres & copper slag", Adv. Concrete Constr., Int. J., 14(3), 185-194. https://doi.org/10.12989/acc.2022.14.3.185
  16. Inacio, M.M.G., Ramos, A.P. and Faria, D.M.V. (2012), "Strengthening of flat slabs with transverse reinforcement by introduction of steel bolts using different anchorage approaches", Eng. Struct., 44, 63-77. https://doi.org/10.1016/j.engstruct.2012.05.043
  17. Jian, G., Wen, S. and Wei, L. (2022), "Use of multi-hybrid machine learning and deep artificial intelligence in the prediction of compressive strength of concrete containing admixtures", Adv. Concrete Constr., Int. J., 13(1), 11-23. https://doi.org/10.12989/acc.2022.13.1.011
  18. Jiao, Z.Y., Li, Y., Guan, H., Diao, M.Z., Yang, Z., Wang, J.K. and Li, Y.C. (2022), "Pre- and post-punching failure performances of flat slab-column joints with drop panels and shear studs", Eng. Fail. Anal., 140, 106604. https://doi.org/10.1016/j.engfailanal.2022.106604
  19. Kang, S.M., Na, S.J., Hwang, H.J. and Kim, S.I. (2022), "Punching shear strength improved by upward panel in reinforced concrete transfer slabs", J. Build. Eng., 46, 103753. https://doi.org/10.1016/j.jobe.2021.103753
  20. Kaya, M., Komur, M.A. and Gursel, E. (2022), "Maturation effect on strength of high-strength concretes which produced with different origin aggregates", Adv. Concrete Constr., Int. J., 14(2), 115-130. https://doi.org/10.12989/acc.2022.14.2.115
  21. Latif, I., Banerjee, A. and Surana, M. (2022), "Explainable machine learning aided optimization of masonry infilled reinforced concrete frames", Structures, 44, 1751-1766. https://doi.org/10.1016/j.istruc.2022.08.115
  22. Li, K., Pan, L. and Wang, Y.F. (2022), "Random forest-based modelling of parameters of fractional derivative concrete creep model with Bayesian optimization", Mater. Struct., 55(8), 215. https://doi.org/10.1617/s11527-022-02054-z
  23. Liang, S.X., Shen, Y.X. and Ren, X.D. (2022), "Comparative study of influential factors for punching shear resistance/failure of RC slab-column joints using machine-learning models", Structures, 45, 1333-1349. https://doi.org/10.1016/j.istruc.2022.09.110
  24. Lovrovich, J.S. and Mclean, D.I. (1990), "Punching shear behavior of slabs with varying span-depth ratios", ACI. Struct. J., 87(5), 507-511.
  25. Mangalathu, S., Karthikeyan, K., Feng, D.C. and Jeon, J.S. (2021a), "Machine-learning interpretability techniques for seismic performance assessment of infrastructure systems", Eng. Struct., 250, 112883. https://doi.org/10.1016/j.engstruct.2021.112883
  26. Mangalathu, S., Shin, H., Choi, E. and Jeon, J.S. (2021b), "Explainable machine learning models for punching shear strength estimation of flat slabs without transverse reinforcement", J. Build. Eng., 39, 102300. https://doi.org/10.1016/j.jobe.2021.102300
  27. Marzouk, H. and Hussein, A. (1991), "Expreimental investigation on the behavior of high-strength concrete slabs", ACI. Struct. J., 88(6), 701-713. https://doi.org/10.12989/cac.2012.91.4.1301
  28. Mellios, N., Uz, O. and Spyridis, P. (2022), "Data-based modeling of the punching shear capacity of concrete structures", Eng. Struct., 275(A), 115195. https://doi.org/10.1016/j.engstruct.2022.115195
  29. Mousavi, M., Gandomi, A.H., Holloway, D. and Chen, F. (2022), "Machine learning analysis of features extracted from time-frequency domain of ultrasonic testing results for wood material assessment", Constr. Build. Mater., 342(A), 127761. https://doi.org/10.1016/j.conbuildmat.2022.127761
  30. Qiu, Y.G., Zhou, J., Khandelwal, W., Yang, H.T., Yang, P.X. and Li, C.Q. (2021), "Performance evaluation of hybrid WOA-XGBoost, GWO-XGBoost and BO-XGBoost models to predict blast-induced ground vibration", Eng. Comput., 38(SUPPL 5), 4145-4162. https://doi.org/10.1007/s00366-021-01393-9
  31. Rad, M.H.G., Nayeban, M.M. and Kazemi, M. (2022), "Externally bonded FRP and mechanically fastened by nailing: A new technique to postpone debonding of FRP sheets in unreinforced concrete slabs", Adv. Concrete Constr., Int. J., 14(6), 413-425. https://doi.org/10.12989/acc.2022.14.6.413
  32. Schulz, E., Speekenbrink, M. and Krause, A. (2018), "A tutorial on Gaussian process regression: Modelling, exploring, and exploiting functions", J. Math. Psychol., 85, 1-16. https://doi.org/10.1016/j.jmp.2018.03.001
  33. Shen, L.L., Shen, Y.X. and Liang, S.X. (2022a), "Reliability analysis of RC slab-column joints under punching shear load using a machine learning-based surrogate model", Buildings, 12(10), 1750. https://doi.org/10.3390/buildings12101750
  34. Shen, Y.X., Wu, L.F. and Liang, S.X. (2022b), "Explainable machine learning-based model for failure mode identification of RC flat slabs without transverse reinforcement", Eng. Fail. Anal., 141, 106647. https://doi.org/10.1016/j.engfailanal.2022.106647
  35. Shi, Q.X., Ma, G., Guo, J.R. and Ma, C.C. (2022), "Punching performance of RC slab-column connections with inner steel truss", Adv. Concrete Constr., Int. J., 14(3), 195-204. https://doi.org/10.12989/acc.2022.14.3.195
  36. Tian, Y., Jirsa, J.O. and Bayrak, O. (2012), "Strength evaluation of interior slab-column connections", ACI Struct. J., 105(6), 692-700.
  37. Truong, G.T., Choi, K.K, Nguyen, T.H. and Kim, C.S. (2023), "Prediction of shear strength of RC deep beams using XGBoost regression with Bayesian optimization", Eur. J. Environ. Civil Eng., 27(14), 4046-4066. https://doi.org/10.1080/19648189.2023.2169357
  38. Walkner, R. (2014), "Critical review of EC2 regarding punching and improving the design approach", Ph.D. Dissertation; University of Innsbruck, Innsbruck, Austria.
  39. Wang, X.W., Mazumder, R.K., Salarieh, B., Salman, A.M., Shafieezadeh, A. and Li, Y. (2022), "Machine learning for risk and resilience assessment in structural engineering: Progress and Future Trends", J. Struct. Eng., 148(8), 03122003. https://doi.org/10.1061/(ASCE)ST.1943-541X.0003392
  40. Wang, Z., Liu, T.X., Long, Z.L., Wang, J.Q. and Zhang, J. (2023), "Predicting the drift capacity of precast concrete columns using explainable machine learning approach", Eng. Struct., 282, 115771. https://doi.org/10.1016/j.engstruct.2023.115771
  41. Wu, Y.Q. and Zhou, Y.S. (2022), "Prediction and feature analysis of punching shear strength of two-way reinforced concrete slabs using optimized machine learning algorithm and Shapley additive explanations", Mech. Adv. Mater. Struct., 30(15), 3086-3096. https://doi.org/10.1080/15376494.2022.2068209
  42. Wu, Y.F., Chen, H., Peng, F. and Yi, W.J. (2021), "Experimental investigation on punching shear mechanism of concrete interior slab-column connections without shear reinforcement", J. Struct. Eng., 148(2), 04021250. https://doi.org/10.1061/(ASCE)ST.1943-541X.0003222
  43. Yin, H., Liu, S.X., Lu, S.S., Nie, W. and Jia, B.X. (2021), "Prediction of the compressive and tensile strength of HPC concrete with fly ash and micro-silica using hybrid algorithms", Adv. Concrete Constr., Int. J., 12(4), 339-354. https://doi.org/10.12989/acc.2021.12.4.339
  44. Zhang, W., Xue, J.Y., Xu, J.J., Li, B.X. and Wei, J.H. (2022), "Bearing capacity assessment of reinforced concrete slab - special-shaped column connections", Adv. Concrete. Constr., Int. J., 14(6), 401-412. https://doi.org/10.12989/acc.2022.14.6.401