Steel and Composite Structures

  • 제51권6호
  • /
  • Pages.679-695
  • /
  • 2024
  • /
  • 1229-9367(pISSN)
  • /
  • 1598-6233(eISSN)
테크노프레스 (Techno-Press)
권호 표지
DOI QR코드

DOI QR Code

Hybrid machine learning with moth-flame optimization methods for strength prediction of CFDST columns under compression

  • Quang-Viet Vu (Laboratory for Computational Civil Engineering, Institute for Computational Science and Artificial Intelligence, Van Lang University) ;
  • Dai-Nhan Le (Faculty of Building and Industrial Construction, Hanoi University of Civil Engineering) ;
  • Thai-Hoan Pham (Faculty of Building and Industrial Construction, Hanoi University of Civil Engineering) ;
  • Wei Gao (Centre for Infrastructure Engineering and Safety, School of Civil and Environmental Engineering, The University of New South Wales) ;
  • Sawekchai Tangaramvong (Center of Excellence in Applied Mechanics and Structures, Department of Civil Engineering, Chulalongkorn University)
  • 투고 : 2023.08.24
  • 심사 : 2024.05.20
  • 발행 : 2024.06.25
⟨ 이전 논문 다음 논문 ⟩

초록

This paper presents a novel technique that combines machine learning (ML) with moth-flame optimization (MFO) methods to predict the axial compressive strength (ACS) of concrete filled double skin steel tubes (CFDST) columns. The proposed model is trained and tested with a dataset containing 125 tests of the CFDST column subjected to compressive loading. Five ML models, including extreme gradient boosting (XGBoost), gradient tree boosting (GBT), categorical gradient boosting (CAT), support vector machines (SVM), and decision tree (DT) algorithms, are utilized in this work. The MFO algorithm is applied to find optimal hyperparameters of these ML models and to determine the most effective model in predicting the ACS of CFDST columns. Predictive results given by some performance metrics reveal that the MFO-CAT model provides superior accuracy compared to other considered models. The accuracy of the MFO-CAT model is validated by comparing its predictive results with existing design codes and formulae. Moreover, the significance and contribution of each feature in the dataset are examined by employing the SHapley Additive exPlanations (SHAP) method. A comprehensive uncertainty quantification on probabilistic characteristics of the ACS of CFDST columns is conducted for the first time to examine the models' responses to variations of input variables in the stochastic environments. Finally, a web-based application is developed to predict ACS of the CFDST column, enabling rapid practical utilization without requesting any programing or machine learning expertise.

키워드

과제정보

The authors sincerely thank the reviewers for their constructive comments on the earlier version of the manuscript. This research is supported by Thailand Science research and Innovation Fund Chulalongkorn University (IND66210025). The support from Ratchadapisek Somphot Fund for Postdoctoral Fellowship and Second Century Fund under Chulalongkorn University is also acknowledged.

참고문헌

  1. Abderazek, H., Hamza, F., Yildiz, A.R. and Sait., S.M. (2021), "Comparative investigation of the moth flame algorithm and whale optimization algorithm for optimal spur gear design", Mater. Test., 63(3), 266-271. https://doi.org/10.1515/mt-2020-0039.
  2. ACI 318-19 (2019), Building Code Requirements for Structural Concrete and Commentary, Building Code Requirements for Structural Concrete and Commentary. American Concrete Institute.
  3. AISC (2016), Specification for Structural Steel Buildings Supersedes the Specification for Structural Steel Buildings dated and all previous versions Approved by the Committee on Specifications.
  4. Ding, F.X., Yu, Z.W., Bai, Y. and Gong, Y.Z. (2011), "Elastoplastic analysis of circular concrete-filled steel tube stub columns", J. Constr. Steel Res., 67(10), 1567-1577. https://doi.org/10.1016/j.jcsr.2011.04.001.
  5. Erdas, M.U., Kopar, M., Yildiz, B.S. and Yildiz, A.R. (2023), "Optimum design of a seat bracket using artificial neural networks and dandelion optimization algorithm", Mater. Test., 65(12), 1767-1775. https://doi.org/10.1515/mt-2023-0201.
  6. Essopjee, Y. and Dundu, M. (2015), "Performance of concrete-filled double-skin circular tubes in compression", Compos. Struct., 133, 1276-1283. https://doi.org/10.1016/j.compstruct.2015.08.033.
  7. Europeenne, N. (2004), EUROPEA STA DARD Eurocode 4: Design of composite steel and concrete structures-Part 1-1: General rules and rules for buildings.
  8. Han, L.H., Ren, Q.X. and Li, W. (2011), "Tests on stub stainless steel concrete carbon steel double-skin tubular (DST) columns", J. Constr. Steel Res., 67(3), 437-452. https://doi.org/10.1016/j.jcsr.2010.09.010.
  9. Hassanein, M.F. and Kharoob, O.F. (2014), "Compressive strength of circular concrete-filled double skin tubular short columns", Thin-Wall. Struct., 77, 165-173. https://doi.org/10.1016/j.tws.2013.10.004.
  10. Hong, Z.T., Wang, W.D., Zheng, L. and Shi, Y.L. (2023), "Machine learning models for predicting axial compressive capacity of circular CFDST columns", Struct., 57, 105285. https://doi.org/10.1016/j.istruc.2023.105285.
  11. Ipek, S. and Guneyisi, E.M. (2019), "Ultimate axial strength of concrete-filled double skin steel tubular column sections", Adv. Civ. Eng., 2019. https://doi.org/10.1155/2019/6493037.
  12. Kim, S.-E., Choi, J.-H., Pham, T.-H., Truong, V.-H., Kong, Z., Duong, N.-T. and Vu, Q.V. (2020), "Behavior of composite CFST beam-concrete column joints", Steel Compos. Struct., 37(1), 75-90. https://doi.org/10.12989/SCS.2020.37.1.075.
  13. Kim, S.-E., Vu, Q.-V., Papazafeiropoulos, G., Kong, Z. and Truong, V.-H. (2020), "Comparison of machine learning algorithms for regression and classification of ultimate load-carrying capacity of steel frames", Steel Compos. Struct., 37(2), 193-209. https://doi.org/10.12989/SCS.2020.37.2.193.
  14. Kong, Z., Le, D. N., Pham, T.H., Keerthan, P., George, P. and Vu, Q.V. (2024), "Hybrid machine learning with optimization algorithm and resampling methods for patch load resistance prediction of unstiffened and stiffened plate girders", Expert Syst. Appl., 249(C), 123806. https://doi.org/10.1016/j.eswa.2024.123806.
  15. Le, D. N., Pham, T.H., George, P., Kong, Z., Tran, V.L. and Vu, Q.V. (2024), "Hybrid machine learning with Bayesian optimization methods for prediction of patch load resistance of unstiffened plate girders", Probabilistic Eng. Mech., 76, 103624. https://doi.org/10.1016/j.probengmech.2024.103624.
  16. Lee, S., Vo, T. P., Thai, H. T., Lee, J. and Patel, V. (2021), "Strength prediction of concrete-filled steel tubular columns using Categorical Gradient Boosting algorithm", Eng. Struct., 238. https://doi.org/10.1016/j.engstruct.2021.112109.
  17. Li, Y.-W., Li, G.-Q., Xiao, L., Yam, M. C. H., Zhang, J.-Z., Li, Y.-W. (2023), "Shear resistance of corrugated web steel beams with circular web openings: Test and machine learning-based prediction", Steel Compos. Struct., 47(1), 103. https://doi.org/10.12989/SCS.2023.47.1.103.
  18. Liu, X.L., Luo, Y., Lu, Y., Jin, Y., Vu, Q.V. and Kong, Z. (2023), "A dual attention network for automatic metallic corrosion detection in natural environment", J. Build. Eng., 75, 107014. https://doi.org/10.1016/j.jobe.2023.107014.
  19. Luat, N.V., Han, S.W. and Lee, K. (2021), "Genetic algorithm hybridized with eXtreme gradient boosting to predict axial compressive capacity of CCFST columns", Compos. Struct., 278. https://doi.org/10.1016/j.compstruct.2021.114733.
  20. Lundberg, S.M., Allen, P.G. and Lee, S.-I. (2017), "A unified Approach to Interpreting Model Predictions. Adv. Neural Inf. Process. Syst., 30.
  21. Lyu, F., Fan, X., Ding, F. and Chen, Z. (2021), "Prediction of the axial compressive strength of circular concrete-filled steel tube columns using sine cosine algorithm-support vector regression. Compos. Struct., 273. https://doi.org/10.1016/j.compstruct.2021.114282.
  22. Mehta, P., Sait, S.M., Yildiz, B.S., Erdas, M.U., Kopar, M. and Yildiz, A.R. (2024), "A new enhanced mountain gazelle optimizer and artificial neural network for global optimization of mechanical design problems", Mater. Test., 66(4), 544-552. https://doi.org/10.1515/mt-2023-0332.
  23. Mirjalili, S. (2015), "Moth-flame optimization algorithm: A novel nature-inspired heuristic paradigm", Knowl Based Syst., 89, 228-249. https://doi.org/10.1016/J.K OSYS.2015.07.006.
  24. Maser, M.Z., Thai, S. and Thai, H.T. (2021), "Evaluating structural response of concrete-filled steel tubular columns through machine learning", J. Build. Eng., 34, 101888. https://doi.org/10.1016/j.jobe.2020.101888.
  25. Ngo, N.T., Pham, T.P.T., Le, H.A., guyen, Q.T. and guyen, T.T.N. (2022), "Axial strength prediction of steel tube confined concrete columns using a hybrid machine learning model", Struct., 36, 765-780. https://doi.org/10.1016/j.istruc.2021.12.054.
  26. Nguyen, T.A. and Ly, H.B. (2024), "Predicting axial compression capacity of CFDST columns and design optimization using advanced machine learning techniques", Struct., 59, 105724. https://doi.org/10.1016/j.istruc.2023.105724.
  27. Nguyen, V.Q., Tran, V.L., guyen, D.D., Sadiq, S. and Park, D. (2022), " ovel hybrid MFO-XGBoost model for predicting the racking ratio of the rectangular tunnels subjected to seismic loading", Transp. Geotech., 37, 100878. https://doi.org/10.1016/j.trgeo.2022.100878.
  28. Pham, V.T. and Kim, S.E. (2023), "A robust approach in prediction of RCFST columns using machine learning algorithm", Steel Compos. Struct., 46(2), 153-173. https://doi.org/10.12989/SCS.2023.46.2.153.
  29. Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A.V. and Gulin, A. (2018), "CatBoost: unbiased boosting with categorical features. Adv. Neural Inf. Process. Syst., 31. https://doi.org/10.48550/arXiv.1706.09516.
  30. Sun, D., Yang, Y., Xue, Y., Yu, Y., An, K., Chen, Y. (2021), "Seismic performance of RC columns with encased prefabricated high-strength CFST core", Steel Compos. Struct., 39(6), 723-736. https://doi.org/10.12989/SCS.2021.39.6.723.
  31. Tao, Z., Han, L.H. and Zhao, X.L. (2004), "Behaviour of concrete-filled double skin (CHS inner and CHS outer) steel tubular stub columns and beam-columns", J. Constr. Steel Res., 60(8), 1129-1158. https://doi.org/10.1016/j.jcsr.2003.11.008.
  32. Tran, V.L. and Kim, J.K. (2024), "Hybrid machine learning models for classifying failure modes of unstiffened steel plate girders subjected to patch loading", Struct., 59. 105742. https://doi.org/10.1016/j.istruc.2023.105742.
  33. Tran, V.L. and Kim, S.E. (2020), "Efficiency of three advanced data-driven models for predicting axial compression capacity of CFDST columns", Thin-Wall. Struct., 152. https://doi.org/10.1016/j.tws.2020.106744.
  34. Tran, V.L., Lee, T.H., guyen, D.D., guyen, T.H., Vu, Q.V. and Phan, H.T. (2023), "Failure mode identification and shear strength prediction of rectangular hollow RC columns using novel hybrid machine learning Models", Bldg., 13(12), 2914. https://doi.org/10.3390/buildings13122914.
  35. Truong, V.H., Sawekchai, T. and George, P. (2024), "An efficient LightGBM-based differential evolution method for nonlinear inelastic truss optimization", Expert Syst. Appl., 237, 121530. https://doi.org/10.1016/j.eswa.2023.121530.
  36. Truong, V.H., Vu, Q.V., Thai, H.T. and Ha, M.H. (2020), "A robust method for safety evaluation of steel trusses using Gradient Tree Boosting algorithm", Adv. Eng. Softw. 147, 102825. https://doi.org/10.1016/j.advengsoft.2020.102825.
  37. Uenaka, K., Kitoh, H. and Sonoda, K. (2010), "Concrete filled double skin circular stub columns under compression", ThinWall. Struct., 48(1), 19-24. https://doi.org/10.1016/j.tws.2009.08.001.
  38. Vu, Q.V., Sawekchai, T., Thu, H.V. and George, P. (2023), "Hybrid GA-A and PSO-A methods for accurate prediction of uniaxial compression capacity of CFDST columns", Steel Compos. Struct., 47(6), 759-779. https://doi.org/10.12989/scs.2023.47.6.759.
  39. Vu, Q.V., Truong, V. H. and Thai, H.T. (2021), "Machine learning-based prediction of CFST columns using gradient tree boosting algorithm", Compos. Struct., 259, 113505. https://doi.org/10.1016/j.compstruct.2020.113505.
  40. Yildiz, B.S. and Yildiz, A.R. (2017), "Moth-flame optimization algorithm to determine optimal machining parameters in manufacturing processes", Mater. Test., 59(5), 425-429. https://doi.org/10.3139/120.111024.
  41. Yu, M., Zha, X., Ye, J. and Li, Y. (2013), "A unified formulation for circle and polygon concrete-filled steel tube columns under axial compression", Eng. Struct., 49, 1-10. https://doi.org/10.1016/j.engstruct.2012.10.018.
  42. Zarringol, M., Patel, V.I., Liang, Q.Q., Hassanein, M.F. and Ahmed, M. (2024), "Machine-learning-based predictive models for concrete-filled double skin tubular columns", Eng. Struct., 304, 117593. https://doi.org/10.1016/j.engstruct.2024.117593.