Computers and Concrete

  • 제32권3호
  • /
  • Pages.233-246
  • /
  • 2023
  • /
  • 1598-8198(pISSN)
  • /
  • 1598-818X(eISSN)
테크노프레스 (Techno-Press)
권호 표지
DOI QR코드

DOI QR Code

Enhancing prediction accuracy of concrete compressive strength using stacking ensemble machine learning

  • Yunpeng Zhao (Department of Civil & Environmental Engineering, University of Maryland) ;
  • Dimitrios Goulias (Department of Civil & Environmental Engineering, University of Maryland) ;
  • Setare Saremi (Department of Civil & Environmental Engineering, University of Maryland)
  • 투고 : 2022.06.23
  • 심사 : 2023.02.21
  • 발행 : 2023.09.25
⟨ 이전 논문 다음 논문 ⟩

초록

Accurate prediction of concrete compressive strength can minimize the need for extensive, time-consuming, and costly mixture optimization testing and analysis. This study attempts to enhance the prediction accuracy of compressive strength using stacking ensemble machine learning (ML) with feature engineering techniques. Seven alternative ML models of increasing complexity were implemented and compared, including linear regression, SVM, decision tree, multiple layer perceptron, random forest, Xgboost and Adaboost. To further improve the prediction accuracy, a ML pipeline was proposed in which the feature engineering technique was implemented, and a two-layer stacked model was developed. The k-fold cross-validation approach was employed to optimize model parameters and train the stacked model. The stacked model showed superior performance in predicting concrete compressive strength with a correlation of determination (R2) of 0.985. Feature (i.e., variable) importance was determined to demonstrate how useful the synthetic features are in prediction and provide better interpretability of the data and the model. The methodology in this study promotes a more thorough assessment of alternative ML algorithms and rather than focusing on any single ML model type for concrete compressive strength prediction.

키워드

과제정보

This research is funded by the Maryland Transportation Institute from the University of Maryland, College Park. The authors are thankful to Dr. Mark Fuge for giving insightful suggestions for the research.

참고문헌

  1. ACI Committee 318 (2019), Building Code Requirements for Reinforced Concrete, American Concrete Institute, Farmington Hills, MI, USA.
  2. Armaghani, D.J., Hatzigeorgiou, G.D., Karamani, C., Skentou, A., Zoumpoulaki, I. and Asteris, P.G. (2019), "Soft computing-based techniques for concrete beams shear strength", Procedia Struct. Integr., 17, 924-933. https://doi.org/10.1016/j.prostr.2019.08.123.
  3. Asteris, P.G., Argyropoulos, I., Cavaleri, L., Rodrigues, H., Varum, H., Thomas, J. and Lourenco, P.B. (2018), "Masonry compressive strength prediction using artificial neural networks", International Conference on Transdisciplinary Multispectral Modeling and Cooperation for the Preservation of Cultural Heritage, Athens, Greece, October.
  4. Asteris, P.G., Lourenco, P.B., Adami, C.A., Roussis, P.C., Armaghani, D.J., Cavaleri, L., Chalioris, C.E., Hajihassani, M., Lemonis, M.E., Mohammed, A.S. and Pilakoutas, K. (2021), "Revealing the nature of metakaolin-based concrete materials using artificial intelligence techniques", Constr. Build. Mater., 322, 126500. https://doi.org/10.1016/j.conbuildmat.2022.126500.
  5. Asteris, P.G., Nozhati, S., Nikoo, M., Cavaleri, L. and Nikoo, M. (2019), "Krill herd algorithm-based neural network in structural seismic reliability evaluation", Mech. Adv. Mater. Struct., 26(13), 1146-1153. https://doi.org/10.1080/15376494.2018.1430874.
  6. Asteris, P.G., Rizal, F.I.M., Koopialipoor, M., Roussis, P.C., Ferentinou, M., Armaghani, D.J. and Gordan, B. (2022), "Slope stability classification under seismic conditions using several tree-based intelligent techniques", Appl. Sci., 12, 1753. https://doi.org/10.3390/app12031753.
  7. Box, G.E.P. and Cox, D.R. (1964), "An analysis of transformations", J. Royal Stat. Soc. Ser. B: Stat. Methodol., 26(2), 211-243. https://doi.org/10.1111/j.2517-6161.1964.tb00553.x.
  8. Breiman, L. (1996), "Bagging predictors", Mach. Learn., 24, 123-140. https://doi.org/10.1007/BF00058655.
  9. Breiman, L., Friedman, J., Stone, C.J. and Olshen, R. (1984), Classification and Regression Trees, Taylor & Francis, New York, NY, USA.
  10. Buhlmann, P. and Yu, B. (2002), "Analyzing bagging", Ann. Stat., 30(4), 927-961. https://doi.org/10.1214/aos/1031689014.
  11. Chakraborty, D., Awolusi, I. and Gutierrez, L. (2021), "An explainable machine learning model to predict and elucidate the compressive behavior of high-performance concrete", Result. Eng., 11, 100246. https://doi.org/10.1016/j.rineng.2021.100245.
  12. Chou, J.S., Chiu, C.K., Farfoura, F. and Al-Taharwa, I. (2011), "Optimizing the prediction accuracy of concrete compressive strength based on a comparison of data-mining techniques", J. Comput. Civil Eng., 25(3), 242-253. https://doi.org/10.1061/(ASCE)CP.1943-5487.0000088.
  13. Combrisson, E. and Jerbi, K. (2015), "Exceeding chance level by chance: The caveat of theoretical chance levels in brain signal classification and statistical assessment of decoding accuracy", J. Neurosci. Method., 250, 126-136. https://doi.org/10.1016/j.jneumeth.2015.01.010.
  14. Cortes, C. and Vapnik, V. (1995), "Support-vector networks", Mach. Learn., 20, 273-297. https://doi.org/10.1007/BF00994018.
  15. Duan, J., Asteris, P.G., Nguyen, H. Bui, X.N. and Moayedi, H. (2020), "A novel artificial intelligence technique to predict compressive strength of recycled aggregate concrete using ICA-XGBoost model", Eng. Comput., 37, 3329-3346. https://doi.org/10.1007/s00366-020-01003-0.
  16. Duan, Z.H., Kou, S.C. and Poon, C.S. (2013), "Prediction of compressive strength of recycled aggregate concrete using artificial neural networks", Constr. Build. Mater., 40, 1200-1206. https://doi.org/10.1016/j.conbuildmat.2012.04.063.
  17. Erdal, H.I., Karakurt, O. and Namli, E. (2013), "High performance concrete compressive strength forecasting using ensemble models based on discrete wavelet transform", Eng. Appl. Artif. Intell., 26, 1246-1254. https://doi.org/10.1016/j.engappai.2012.10.014.
  18. Farooq, F., Ahmed, W., Akbar, A., Aslam, F. and Alyousef, R. (2021), "Predictive modeling for sustainable high-performance concrete from industrial wastes: A comparison and optimization of models using ensemble learners", J. Clean. Prod., 292, 126032. https://doi.org/10.1016/j.jclepro.2021.126032.
  19. Feng, D.C., Liu, Z.T., Wang, X.D., Chen, Y., Chang., J.Q., Wei., D.F. and Jiang, Z.M. (2020), "Machine learning-based compressive strength prediction for concrete: An adaptive boosting approach", Constr. Build. Mater., 230, 117000. https://doi.org/10.1016/j.conbuildmat.2019.117000.
  20. Fisher, A., Rudin, C. and Dominici, F. (2019), "All models are wrong, but many are useful: Learning a variable's importance by studying an entire class of prediction models simultaneously", J. Mach. Learn. Res., 20(177), 1-81.
  21. Freund, Y. and Schapire, R.E. (1996), "Experiments with a new boosting algorithm", Machine Learning: Proceedings of the Thirteenth International Conference, Bari, Italy, July.
  22. Grolemund, G. (2014), Hands-On Programming with R, O'Reilly Media, Inc., Sebastopol, CA, USA.
  23. Gupta, S. (2008), "Support vector machines based modeling of concrete strength", World Acad. Sci. Eng. Technol., 36, 305-311.
  24. Han, Q., Gui, C., Xu, J. and Lacidogna, G. (2019), "A generalized method to predict the compressive strength of high-performance concrete by improved random forest algorithm", Constr. Build. Mater., 226, 734-742. https://doi.org/10.1016/j.conbuildmat.2019.07.315.
  25. Harandizadeh, H., Armaghani, D.J., Asteris, P.G. and Gandomi, A.H. (2021), "TBM performance prediction developing a hybrid ANFIS-PNN predictive model optimized by ICA", Neural Comput. Appl., 33, 16149-16179. https://doi.org/10.1007/s00521-021-06217-x.
  26. Kardani, N., Bardhan, A., Samui, P., Nazem, M., Asteris, P.G. and Zhou, A. (2022), "Predicting the thermal conductivity of soils using integrated approach of ANN and PSO with adaptive and time-varying acceleration coefficients", Int. J. Therm. Sci., 173, 107427. https://doi.org/10.1016/j.ijthermalsci.2021.107427.
  27. Kosmatka, S.H. and Wilson, M.L. (2011), Design and Control of Concrete Mixtures, Portland Cement Association, Skokie, IL, USA.
  28. Lemonis, M.E., Daramara, A.G., Georgiadou, A.G., Siorikis, V.G., Tsavdaridis, K.D. and Asteris, P.G. (2022), "Ultimate axial load of rectangular concrete-filled steel tubes using multiple ANN activation functions", Steel Compos. Struct., 2(4), 459-475. https://doi.org/10.12989/scs.2022.42.4.459.
  29. Ling, H., Qian, C., Kang, W., Liang, C. and Chen, H. (2019), "Combination of support vector machine and K-Fold cross-validation to predict compressive strength of concrete in marine environment", Constr. Build. Mater., 206, 355-363. https://doi.org/10.1016/j.conbuildmat.2019.02.071.
  30. Lu, S., Koopialipoor, M., Asteris, P.G., Bahri, M. and Armaghani, D.J. (2020), "A novel feature selection approach based on tree models for evaluating the punching shear capacity of steel fiber-reinforced concrete flat slabs", Mater., 13, 3902. https://doi.org/10.3390/ma13173902.
  31. Miladirad, K., Golafshani, E.M., Safehian, M. and Sarkar, A. (2021), "Modeling the mechanical properties of rubberized concrete using machine learning methods", Comput. Concrete, 28(6), 567-583. https://doi.org/10.12989/cac.2021.28.6.567.
  32. Nguyen, N.H., Vo, T.P., Lee, S. and Asteris, P.G. (2021), "Heuristic algorithm-based semi-empirical formulas for estimating the compressive strength of the normal and high-performance concrete", Constr. Build. Mater., 304, 124467. https://doi.org/10.1016/j.conbuildmat.2021.124467.
  33. Nguyen-Sy, T., Wakim, J., To, Q.D., Vu, M.N., Nguyen, T.D. and Nguyen, T.T. (2020), "Predicting the compressive strength of concrete from its compositions and age using the extreme gradient boosting method", Constr. Build. Mater., 260, 119757. https://doi.org/10.1016/j.conbuildmat.2020.119757.
  34. Ni, H.G. and Wang, J.Z. (2000), "Prediction of compressive strength of concrete by neural networks", Concrete Cement Res., 30, 1245-1250. https://doi.org/10.1016/S0008-8846(00)00345-8.
  35. Oner, E. (2021), "Frictionless contact mechanics of an orthotropic coating/isotropic substrate system", Comput. Concrete, 28(2), 209-220. https://doi.org/10.12989/cac.2021.28.2.209.
  36. Phillips, R.V., van der Laan, M.J., Lee, H. and Gruber, S. (2022), "Practical considerations for specifying a super learner", arXiv preprint arXiv, 2204, 06139. https://doi.org/10.48550/arXiv.2204.06139.
  37. Popovics, S. and Ujhelyi, J. (2008), "Contribution to the concrete strength versus water cement ratio relationship", J. Mater. Civil Eng., 20(7), 459-463. https://doi.org/10.1061/(ASCE)0899-1561(2008)20:7(459).
  38. Psyllaki, P., Stamatiou, K., Iliadis, I., Mourlas, A., Asteris, P. and Vaxevanidis, N. (2018), "Surface treatment of tool steels against galling failure", MATEC Web Conf., 188, 04024. https://doi.org/10.1051/matecconf/201818804024.
  39. Salehi, H. and Burgueno, R. (2018), "Emerging artificial intelligence methods in structural engineering", Eng. Struct., 171, 170-189. https://doi.org/10.1016/j.engstruct.2018.05.084.
  40. Tahwia, A.M., Heniegal, A., Elgamal, M.S. and Tayeh, B.A. (2021), "The prediction of compressive strength and nondestructive tests of sustainable concrete by using artificial neural networks", Comput. Concrete, 27(1), 21-28. https://doi.org/10.12989/cac.2021.27.1.021.
  41. Tanyildizi, H. (2022), "Predicting bond strength of corroded reinforcement by deep learning", Comput. Concrete, 29(3), 145-159. https://doi.org/10.12989/cac.2022.29.3.145.
  42. van der Laan, M.J., Polley, E.C. and Hubbard, A.E. (2007), "Super Learner", Stat. Appl. Genet. Molecul. Biol., 6(1), 1. https://doi.org/10.2202/1544-6115.1309.
  43. Yeh, I.C. (1998), "Modeling of strength of high-performance concrete using artificial neural networks", Cement Concrete Res., 28(12), 1797-1808. https://doi.org/10.1016/S0008-8846(98)00165-3.
  44. Young, B.A., Hall, A., Pilon, L., Gupta, P. and Sant, G. (2019), "Can the compressive strength of concrete be estimated from knowledge of the mixture proportions? New insights from statistical analysis and machine learning methods", Cement Concrete Res., 115, 379-388. https://doi.org/10.1016/j.cemconres.2018.09.006.
  45. Zhang, J., Ma, G., Huang, Y., Sun, J., Aslani, F. and Nener, B. (2019), "Modelling uniaxial compressive strength of lightweight self-compacting concrete using random forest regression", Constr. Build. Mater., 210, 713-719. https://doi.org/10.1016/j.conbuildmat.2019.03.189.
  46. Zhang, X., Akber, M.Z. and Zhang, W. (2021), "Prediction of seven-day compressive strength of field concrete", Constr. Build. Mater., 305, 124604. https://doi.org/10.1016/j.conbuildmat.2021.124604.