Earthquakes and Structures

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Study on failure mode prediction of reinforced concrete columns based on class imbalanced dataset

  • Mingyi Cai (College of Civil Engineering, Nanjing Tech University) ;
  • Guangjun Sun (College of Civil Engineering, Nanjing Tech University) ;
  • Bo Chen (College of Civil Engineering, Nanjing Tech University)
  • Received : 2024.02.03
  • Accepted : 2024.06.12
  • Published : 2024.09.25
⟨ Previous Next ⟩

Abstract

Accurately predicting the failure modes of reinforced concrete (RC) columns is essential for structural design and assessment. In this study, the challenges of imbalanced datasets and complex feature selection in machine learning (ML) methods were addressed through an optimized ML approach. By combining feature selection and oversampling techniques, the prediction of seismic failure modes in rectangular RC columns was improved. Two feature selection methods were used to identify six input parameters. To tackle class imbalance, the Borderline-SMOTE1 algorithm was employed, enhancing the learning capabilities of the models for minority classes. Eight ML algorithms were trained and fine-tuned using k-fold shuffle split cross-validation and grid search. The results showed that the artificial neural network model achieved 96.77% accuracy, while k-nearest neighbor, support vector machine, and random forest models each achieved 95.16% accuracy. The balanced dataset led to significant improvements, particularly in predicting the flexure-shear failure mode, with accuracy increasing by 6%, recall by 8%, and F1 scores by 7%. The use of the Borderline-SMOTE1 algorithm significantly improved the recognition of samples at failure mode boundaries, enhancing the classification performance of models like k-nearest neighbor and decision tree, which are highly sensitive to data distribution and decision boundaries. This method effectively addressed class imbalance and selected relevant features without requiring complex simulations like traditional methods, proving applicable for discerning failure modes in various concrete members under seismic action.

Keywords

Acknowledgement

The research described in this paper was financially supported by the National Natural Science Foundation of China (Grant No. 51878347).

References

  1. Al-Shboul, K.F., Almasabha, G., Shehadeh, A. and Alshboul, O. (2023), "Exploring the efficacy of machine learning models for predicting soil radon exhalation rates", Stoch. Env. Res. Risk. A., 37(11), 4307-4321. https://doi.org/10.1007/s00477-023-02509-x. 
  2. Alcantara, P.A. and Imai, H. (2000), "Failure mode classification of reinforced concrete columns by the analysis of the strain distribution in the main reinforcement", Proceedings of the 12th World Conference on Earthquake Engineering, Auckland, New Zealand, January-February. 
  3. Almasabha, G., Al-Shboul, K.F., Shehadeh, A. and Alshboul, O. (2023), "Machine learning-based models for predicting the shear strength of synthetic fiber reinforced concrete beams without stirrups", Struct., 52, 299-311. https://doi.org/10.1016/j.istruc.2023.03.170. 
  4. Almasabha, G., Shehadeh, A., Alshboul, O. and Al Hattamleh, O. (2023), "Structural performance of buried reinforced concrete pipelines under deep embankment soil", Constr. Innov., 2023, 1. https://doi.org/10.1108/CI-10-2021-0196. 
  5. Alshboul, O., Almasabha, G., Al-Shboul, K.F. and Shehadeh, A. (2023), "A comparative study of shear strength prediction models for SFRC deep beams without stirrups using Machine learning algorithms", Struct., 55, 97-111. https://doi.org/10.1016/j.istruc.2023.06.026. 
  6. Alshboul, O., Almasabha, G., Shehadeh, A. and Al-Shboul, K. (2024), "A comparative study of LightGBM, XGBoost, and GEP models in shear strength management of SFRC-SBWS", Struct., 61, 106009. https://doi.org/10.1016/j.istruc.2024.106009. 
  7. Alshboul, O., Almasabha, G., Shehadeh, A., Al Hattamleh, O. and Almuflih, A.S. (2022), "Optimization of the structural performance of buried reinforced concrete pipelines in cohesionless soils", Mater., 15(12), 4051. https://doi.org/10.3390/ma15124051. 
  8. Alshboul, O., Shehadeh, A., Almasabha, G., Mamlook, R.E.A. and Almuflih, A.S. (2022), "Evaluating the impact of external support on green building construction cost: A hybrid mathematical and machine learning prediction approach", Build., 12(8), 1256. https://doi.org/10.3390/buildings12081256. 
  9. Alshboul, O., Shehadeh, A., Tatari, O., Almasabha, G. and Saleh, E. (2024), "Multiobjective and multivariable optimization for earthmoving equipment", J. Facil. Manag., 22(1), 21-48. https://doi.org/10.1108/JFM-10-2021-0129. 
  10. Babajanian Bisheh, H., Ghodrati Amiri, G., Nekooei, M. and Darvishan, E. (2019), "Damage detection of a cable-stayed bridge using feature extraction and selection methods", Struct. Infrastruct. Eng., 15(9), 1165-1177. https://doi.org/10.1080/15732479.2019.1599964. 
  11. Berry, M., Parrish, M. and Eberhard, M. (2004), PEER Structural Performance Database User's Manual (Version 1.0), University of California, Berkeley, Berkeley, CA, USA. 
  12. Bolon-Canedo, V. and Alonso-Betanzos, A. (2019), "Ensembles for feature selection: A review and future trends", Inf. Fusion., 52, 1-12. https://doi.org/10.1016/j.inffus.2018.11.008. 
  13. Bugata, P. and Drotar, P. (2019), "On some aspects of minimum redundancy maximum relevance feature selection", Sci. Chin. Inf. Sci., 63(1), 112103. https://doi.org/10.1007/s11432-019-2633-y. 
  14. Cai, M., Gu, X., Hua, J. and Lin, F. (2011), "Seismic response analysis of reinforced concrete columns considering shear effects", J. Build. Struct., 32, 97-108. 
  15. Chawla, N.V., Bowyer, K.W., Hall, L.O. and Kegelmeyer, W.P. (2002), "SMOTE: synthetic minority over-sampling technique", J. Artif. Intell. Res., 16, 321-357. https://doi.org/10.1613/jair.953. 
  16. Cortes, C. and Vapnik, V. (1995), "Support-vector networks", Mach. Learn., 20(3), 273-297. https://doi.org/10.1007/BF00994018. 
  17. Cover, T. and Hart, P. (1967), "Nearest neighbor pattern classification", IEEE Trans. Inf. Theory., 13(1), 21-27. https://doi.org/10.1109/TIT.1967.1053964. 
  18. Feng, D.C., Liu, Z.T., Wang, X.D., Jiang, Z.M. and Liang, S.X. (2020), "Failure mode classification and bearing capacity prediction for reinforced concrete columns based on ensemble machine learning algorithm", Adv. Eng. Inform., 45, 101126. https://doi.org/10.1016/j.aei.2020.101126. 
  19. Freund, Y. and Schapire, R.E. (1997), "A decision-theoretic generalization of on-line learning and an application to boosting", J. Comput. Syst. Sci., 55(1), 119-139. https://doi.org/10.1006/jcss.1997.1504. 
  20. Friedman, J.H. (2001), "Greedy function approximation: A gradient boosting machine", Ann. Stat., 29, 1189-1232. https://doi.org/10.1214/AOS/1013203451. 
  21. Ghasemi, A. and Sobhani, M. (2023), "Modeling the impact of corrosion rate of stirrups on seismic performance of reinforced concrete columns", Earthq. Struct., 24(3), 183-192. https://doi.org/10.12989/eas.2023.24.3.183. 
  22. Ghee, A.B., Priestley, M.J.N. and Paulay, T. (1989), "Seismic shear strength of circular reinforced concrete columns", ACI Struct. J., 86(1), 45-59.  https://doi.org/10.14359/2634
  23. Gu, X.L., Hua, J.J. and Cai, M. (2020), "Seismic responses of reinforced concrete intermediate short columns failed in different modes", Eng. Struct., 206, 110173. https://doi.org/10.1016/j.engstruct.2020.110173. 
  24. Han, H., Wang, W.Y. and Mao, B.H. (2005), "Borderline-SMOTE: A new over-sampling method in imbalanced data sets learning", International Conference on Intelligent Computing, Hefei, China, August. 
  25. Han, Q., Zhou, Y., Du, X., Huang, C. and Lee, G.C. (2014), "Experimental and numerical studies on seismic performance of hollow RC bridge columns", Earthq. Struct., 7(3), 251-269. https://doi.org/10.12989/eas.2014.7.3.251. 
  26. Hao, X., Zhu, A.P., Zhai, W. and Jiang, B. (2019), "Research on seismic shear behavior of concrete columns with CRB600H stirrups", Build. Struct., 49(9), 99-106. https://doi.org/10.19701/j.jzjg.2019.09.017. 
  27. Ke, G.L., Meng, Q., Finley, T., Wang, D.S., Chen, W., Ma, W.D., Ye, Q.W. and Liu, T.Y. (2017), "LightGBM: A highly efficient gradient boosting decision tree", Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA, December. 
  28. Kohonen, T. (1988), "An introduction to neural computing", Neural Networks, 1(1), 3-16. https://doi.org/10.1016/0893-6080(88)90020-2. 
  29. Krzywinski, M. and Altman, N. (2017), "Classification and regression trees", Nat. Methods, 14(8), 757-758. https://doi.org/10.1038/nmeth.4370. 
  30. Li, Y.A., Huang, Y.T. and Hwang, S.J. (2014), "Seismic response of reinforced concrete short columns failed in shear", ACI Struct. J., 111(4), 945. https://doi.org/10.14359/51686780. 
  31. Ma, Y. and Gong, J.X. (2018), "Probability identification of seismic failure modes of reinforced concrete columns based on experimental observations", J. Earthq. Eng., 22(10), 1881-1899. https://doi.org/10.1080/13632469.2017.1309603. 
  32. Maekawa, K. and An, X. (2000), "Shear failure and ductility of RC columns after yielding of main reinforcement", Eng. Fract. Mech., 65(2), 335-368. https://doi.org/10.1016/S0013-7944(99)00119-8. 
  33. Mangalathu, S., Hwang, S.H. and Jeon, J.S. (2020), "Failure mode and effects analysis of RC members based on machine-learning-based SHapley Additive exPlanations (SHAP) approach", Eng. Struct., 219, 110927. https://doi.org/10.1016/j.engstruct.2020.110927. 
  34. Mangalathu, S. and Jeon, J.S. (2019), "Machine learning-based failure mode recognition of circular reinforced concrete bridge columns: Comparative study", J. Struct. Eng., 145(10), 04019104. https://doi.org/10.1061/(ASCE)ST.1943-541X.0002402. 
  35. Matamoros, A.B., Matchulat, L. and Woods, C. (2008), "Axial load failure of shear critical columns subjected to high levels of axial load", Proceedings of the 14th World Conference on Earthquake Engineering, Beijing, China, October. 
  36. Moretti, M. and Tassios, T.P. (2007), "Behaviour of short columns subjected to cyclic shear displacements: Experimental results", Eng. Struct., 29(8), 2018-2029. https://doi.org/10.1016/j.engstruct.2006.11.001. 
  37. Naderpour, H., Mirrashid, M. and Parsa, P. (2021), "Failure mode prediction of reinforced concrete columns using machine learning methods", Eng. Struct., 248, 113263. https://doi.org/10.1016/j.engstruct.2021.113263. 
  38. Ousalem, H., Kabeyasawa, T., Tasai, A. and Iwamoto, J. (2003), "Effect of hysteretic reversals on lateral and axial capacities of reinforced concrete columns", 5th US-Japan Workshop on Performance-Based Earthquake Engineering Methodology for Reinforced Concrete Building Structures, Hakone, Japan, September. 
  39. Phan, H.D. and Lin, K.C. (2020), "Seismic behavior of full-scale square concrete filled steel tubular columns under high and varied axial compressions", Earthq. Struct., 18(6), 677-689. https://doi.org/10.12989/eas.2020.18.6.677. 
  40. Qi, Y.L., Han, X.L. and Ji, J. (2013), "Failure mode classification of reinforced concrete column using Fisher method", J. Central South Univ., 20(10), 2863-2869. https://doi.org/10.1007/s11771-013-1807-1. 
  41. Qiu, J.L. and Gong, J.X. (2019), "Load-deformation analysis of reinforced concrete columns considering axial-flexure-shear interaction", Eng. Mech., 36(10), 189-201. https://doi.org/10.6052/j.issn.1000-4750.2018.11.0584. 
  42. Ramirez, H. and Jirsa, J.O. (1980), "Effect of axial load on shear behavior of short RC columns under cyclic lateral deformations", Research Report No. 80-1; University of Texas at Austin, Austin, TX, USA. 
  43. Saccenti, E., Hendriks, M.H.W.B. and Smilde, A.K. (2020), "Corruption of the Pearson correlation coefficient by measurement error and its estimation, bias, and correction under different error models", Sci. Rep., 10(1), 438. https://doi.org/10.1038/s41598-019-57247-4. 
  44. Schonlau, M. and Zou, R.Y. (2020), "The random forest algorithm for statistical learning", Stata J., 20(1), 3-29. https://doi.org/10.1177/1536867X20909688. 
  45. Shehadeh, A., Alshboul, O. and Almasabha, G. (2024), "Slope displacement detection in construction: An automated management algorithm for disaster prevention", Expert. Syst. Appl., 237, 121505. https://doi.org/10.1016/j.eswa.2023.121505. 
  46. Shi, Q.X., Yang, W.X., Wang, Q.W., Tian, Y., Zhang, X.H., Jiang, W.S., Bai, L.G. and Zhao, Q.C. (2012), "Experimental research on seismic behavior of high-strength concrete short columns with high-strength stirrups", J. Build. Struct., 33(9), 49-58. https://doi.org/10.14006/j.jzjgxb.2012.09.016. 
  47. Sun, Z.G., Si, B.J., Wang, D.S., Guo, X. and YU, D.H. (2010), "Research on the seismic performance of high-strength concrete columns with high-strength stirrups", Eng. Mech., 27(5), 128-136. https://doi.org/10.19701/j.jzjg.2014.01.008. 
  48. Tran, C.T.N. (2010), "Experimental and analytical studies on the seismic behavior of reinforced concrete columns with light transverse reinforcement", Ph.D. Dissertation, Nanyang Technological University, Singapore. 
  49. Umehara, H. and Jirsa, J.O. (1984), "Short rectangular RC columns under bidirectional loadings", J. Struct. Eng., 110(3), 605-618.  https://doi.org/10.1061/(ASCE)0733-9445(1984)110:3(605)
  50. Venkatesh, B. and Anuradha, J. (2019), "A review of feature selection and its methods", Cybern. Inf. Technol., 19, 3. https://doi.org/10.2478/cait-2019-0001. 
  51. Wan, H., Qi, Y., Zhao, T., Ren, W. and Fu, X. (2023), "Prediction of column failure modes based on artificial neural network", Earthq. Eng. Eng. Vib., 22(2), 481-493. https://doi.org/10.1007/s11803-023-2179-7. 
  52. Wang, P., Shi, Q.X., Wang, F. and Wang, Q.W. (2020), "Seismic behaviour of concrete columns with high-strength stirrups", Earthq. Struct., 18(1), 15-25. https://doi.org/10.12989/eas.2020.18.1.015. 
  53. Wang, Q.F., Shen, Z.C. and Yang, Y.X. (2008), "Seismic behavior of HRB400 reinforcement concrete short columns", J. Build. Struct., 29(2), 114-117. https://doi.org/10.14006/j.jzjgxb.2008.02.017. 
  54. Wang, Q.F., Zhou, B., Zheng, J.K., Xu, Y.Y. and Wang, L.Y. (2014), "Experimental study on seismic behavior of HRBF500 reinforced concrete short columns", Build. Struct., 44(1), 38-41. https://doi.org/10.19701/j.jzjg.2014.01.008. 
  55. Xu, J.G., Hong, W., Zhang, J., Hou, S.T. and Wu, G. (2022), "Seismic performance assessment of corroded RC columns based on data-driven machine-learning approach", Eng. Struct., 255, 113936. https://doi.org/10.1016/j.engstruct.2022.113936. 
  56. Yoshimura, K., Kikcuri, K. and Kuroki, M. (1991), "Seismic shear strengthening method for existing reinforced concrete short columns", Evaluation and Rehabilitation of Concrete Structures and Innovations in Design: Proceedings of ACI International Conference, Hong Kong, American Concrete Institute, Farmington Hills, MI, USA. 
  57. Yoshimura, M. and Nakamura, T. (2002), "Axial collapse of reinforced concrete short columns", PEER Rep., 21, 187-198.
  58. Yoshimura, M., Takaine, Y. and Nakamura, T. (2003), "Collapse drift of reinforced concrete columns", PEER Rep., 11, 239-253. 
  59. Yu, B., Cheng, H., Yu, Z.C., Li, B. and Zhang, Q. (2023), "Physics-supervised ensemble learning model for predicting failure modes of reinforced concrete columns", Eng. Struct., 292, 116560. https://doi.org/10.1016/j.engstruct.2023.116560. 
  60. Yuan, Z., Niu, M.Q., Ma, H.T., Gao, T., Zang, J., Zhang, Y.W. and Chen, L.Q. (2023), "Predicting mechanical behaviors of rubber materials with artificial neural networks", Int. J. Mech. Sci., 249, 108265. https://doi.org/10.1016/j.ijmecsci.2023.108265. 
  61. Zeng, L., Ren, W.T., Zou, Z.T., Chen, Y.G., Xie, W. and Li, X.J. (2019), "Experimental study on seismic behavior of frame structures composed of concrete encased columns with L-shaped steel section and steel beams", Earthq. Struct., 16(1), 97-107. https://doi.org/10.12989/eas.2019.16.1.097. 
  62. Zhu, L., Elwood, K.J. and Haukaas, T. (2007), "Classification and seismic safety evaluation of existing reinforced concrete columns", J. Struct. Eng., 133(9), 1316-1330.  https://doi.org/10.1061/(ASCE)0733-9445(2007)133:9(1316)