Computers and Concrete

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

A generalized explainable approach to predict the hardened properties of self-compacting geopolymer concrete using machine learning techniques

  • Endow Ayar Mazumder (Department of Civil Engineering, National Institute of Technology Silchar) ;
  • Sanjog Chhetri Sapkota (Department of Civil Engineering, Sharda University) ;
  • Sourav Das (Department of Civil Engineering, Barak Valley Engineering College) ;
  • Prasenjit Saha (Department of Civil Engineering, ICFAI University) ;
  • Pijush Samui (Department of Civil Engineering, National Institute of Technology)
  • Received : 2023.09.05
  • Accepted : 2024.02.05
  • Published : 2024.09.25
⟨ Previous Next ⟩

Abstract

In this study, ensemble machine learning (ML) models are employed to estimate the hardened properties of Self-Compacting Geopolymer Concrete (SCGC). The input variables affecting model development include the content of the SCGC such as the binder material, the age of the specimen, and the ratio of alkaline solution. On the other hand, the output parameters examined includes compressive strength, flexural strength, and split tensile strength. The ensemble machine learning models are trained and validated using a database comprising 396 records compiled from 132 unique mix trials performed in the laboratory. Diverse machine learning techniques, notably K-nearest neighbours (KNN), Random Forest, and Extreme Gradient Boosting (XGBoost), have been employed to construct the models coupled with Bayesian optimisation (BO) for the purpose of hyperparameter tuning. Furthermore, the application of nested cross-validation has been employed in order to mitigate the risk of overfitting. The findings of this study reveal that the BO-XGBoost hybrid model confirms better predictive accuracy in comparison to other models. The R2 values for compressive strength, flexural strength, and split tensile strength are 0.9974, 0.9978, and 0.9937, respectively. Additionally, the BO-XGBoost hybrid model exhibits the lowest RMSE values of 0.8712, 0.0773, and 0.0799 for compressive strength, flexural strength, and split tensile strength, respectively. Furthermore, a SHAP dependency analysis was conducted to ascertain the significance of each parameter. It is observed from this study that GGBS, Flyash, and the age of specimens exhibit a substantial level of influence when predicting the strengths of geopolymers.

Keywords

References

  1. Afridi, S., Sikandar, M.A., Waseem, M., Nasir, H. and Naseer, A. (2019), "Chemical durability of superabsorbent polymer (SAP) based geopolymer mortars (GPMs)", Constr. Build. Mater., 217, 530-542. https://doi.org/10.1016/j.conbuildmat.2019.05.101. 
  2. Akcaoglu, T., Cubukcuoglu, B. and Awad, A. (2019), "A critical review of slag and fly-ash based geopolymer concrete", Comput. Concrete, 24(5), 453-458. https://doi.org/10.12989/cac.2019.24.5.453. 
  3. Aliabdo, A.A., Abd Elmoaty, M. and Salem, H.A. (2016), "Effect of water addition, plasticizer and alkaline solution constitution on fly ash based geopolymer concrete performance", Constr. Build. Mater., 121, 694-703. https://doi.org/10.1016/j.conbuildmat.2016.06.062. 
  4. Alkroosh, I.S. and Sarker, P.K. (2019), "Prediction of the compressive strength of fly ash geopolymer concrete using gene expression programming", Comput. Concrete, 24(4), 295-302. https://doi.org/10.12989/cac.2019.24.4.295. 
  5. Awoyera, P.O., Kirgiz, M.S., Viloria, A. and Ovallos-Gazabon, D. (2020), "Estimating strength properties of geopolymer self-compacting concrete using machine learning techniques", J. Mater. Res. Technol., 9(4), 9016-9028. https://doi.org/10.1016/j.jmrt.2020.06.008. 
  6. Bentejac, C., Csorgo, A. and Martinez-Munoz, G. (2019), "A comparative analysis of XGBoost", Artif. Intell. Rev., 54, 1937-1967. https://doi.org/10.1007/s10462-020-09896-5. 
  7. Chen, T. and Guestrin, C. (2016), "XGBoost: A scalable tree boosting system", Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, August. 
  8. Davidovits, J. and Cordi, S.A. (1979), "Synthesis of new high-temperature geo-polymers for reinforced plastics composites", PACTEC '79 Society of Plastics Engineers, Costa Mesa, CA, USA, January-February. 
  9. EFNARC (2002), Specification and Guidelines for Self-Compacting Concrete, EFNARC, Farnham, UK.
  10. Gao, X. and Lin, C. (2021), "Prediction model of the failure mode of beam-column joints using machine learning methods", Eng. Fail. Anal., 120, 105072. https://doi.org/10.1016/j.engfailanal.2020.105072. 
  11. Hutagi, A., Khadiranaikar, R.B. and Zende, A.A. (2020), "Behavior of geopolymer concrete under cyclic loading", Constr. Build. Mater., 246, 118430. https://doi.org/10.1016/j.conbuildmat.2020.118430. 
  12. Jindal, B.B., Jangra, P. and Garg, A. (2020), "Effects of ultra fine slag as mineral admixture on the compressive strength, water absorption and permeability of rice husk ash based geopolymer concrete", Mater. Today Proc., 32, 871-877. https://doi.org/10.1016/j.matpr.2020.04.219. 
  13. Kabir, M.A.B., Hasan, A.S. and Billah, A.M. (2021), "Failure mode identification of column base plate connection using data-driven machine learning techniques", Eng. Struct., 240, 112389. https://doi.org/10.1016/j.engstruct.2021.112389. 
  14. Kurtoglu, A.E., Hussein, A.K., Gulsan, M.E. and Cevik, A. (2021), "Flexural behavior of HDPE tubular beams filled with self-compacting geopolymer concrete", Thin Wall. Struct., 167, 108096. https://doi.org/10.1016/j.tws.2021.108096. 
  15. Liu, G. and Sun, B. (2023), "Concrete compressive strength prediction using an explainable boosting machine model", Case Stud. Constr. Mater., 18, e01845. https://doi.org/10.1016/j.cscm.2023.e01845. 
  16. Mangalathu, S., Jang, H., Hwang, S.H. and Jeon, J.S. (2020), "Data-driven machine-learning-based seismic failure mode identification of reinforced concrete shear walls", Eng. Struct., 208, 110331. https://doi.org/10.1016/j.engstruct.2020.110331. 
  17. Mazumder, E.A. and Prasad Meesaraganda, L.V. (2023), "Probabilistic estimation for mechanical properties of self-compacting geopolymer concrete using machine learning technique", Arab. J. Sci. Eng., 48(10), 13591-13604. https://doi.org/10.1007/s13369-023-07866-x. 
  18. Nguyen, K.T., Nguyen, Q.D., Le, T.A., Shin, J. and Lee, K. (2020a), "Analyzing the compressive strength of green fly ash based geopolymer concrete using experiment and machine learning approaches", Constr. Build. Mater., 247, 118581. https://doi.org/10.1016/j.conbuildmat.2020.118581. 
  19. Nguyen, K.T., Nguyen, Q.D., Le, T.A., Shin, J. and Lee, K. (2020b), "Analyzing the compressive strength of green fly ash based geopolymer concrete using experiment and machine learning approaches", Constr. Build. Mater., 247, 118581. https://doi.org/10.1016/j.conbuildmat.2020.118581. 
  20. Rahman, J., Ahmed, K.S., Khan, N.I., Islam, K. and Mangalathu, S. (2021), "Data-driven shear strength prediction of steel fiber reinforced concrete beams using machine learning approach", Eng. Struct., 233, 111743. https://doi.org/10.1016/j.engstruct.2020.111743. 
  21. Saha, P., Debnath, P. and Thomas, P. (2020), "Prediction of fresh and hardened properties of self-compacting concrete using support vector regression approach", Neural Comput. Appl., 32(12), 7995-8010. https://doi.org/10.1007/s00521-019-04267-w. 
  22. Sastry, K.G.K., Sahitya, P. and Ravitheja, A. (2021), "Influence of nano TiO2 on strength and durability properties of geopolymer concrete", Mater. Today Proc., 45, 1017-1025. https://doi.org/10.1016/j.matpr.2020.03.139. 
  23. Shahmansouri, A.A., Bengar, H.A. and Jahani, E. (2019), "Predicting compressive strength and electrical resistivity of eco-friendly concrete containing natural zeolite via GEP algorithm", Constr. Build. Mater., 229, 116883. https://doi.org/10.1016/j.conbuildmat.2019.116883. 
  24. Shahmansouri, A.A., Yazdani, M., Ghanbari, S., Bengar, H.A., Jafari, A. and Ghatte, H.F. (2021), "Artificial neural network model to predict the compressive strength of eco-friendly geopolymer concrete incorporating silica fume and natural zeolite", J. Clean. Prod., 279, 123697. https://doi.org/10.1016/j.jclepro.2020.123697. 
  25. Shamayleh, O. and Far, H. (2023), "Utilising artificial neural networks for prediction of properties of geopolymer concrete", Comput. Concrete, 31(4), 327-335. https://doi.org/10.12989/cac.2023.31.4.327. 
  26. Shamayleh, O. and Far, H. (2023), "Utilising artificial neural networks for prediction of properties of geopolymer concrete", Comput. Concrete, 31(4), 327-335. https://doi.org/10.12989/cac.2023.31.4.327. 
  27. Shekhar, S., A. Bansode, and A. Salim. (2022), "A comparative study of hyper-parameter optimization tools", 2021 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE), Brisbane, Australia, August. 
  28. Singh, N.B., Saxena, S.K. and Kumar, M. (2018), "Effect of nanomaterials on the properties of geopolymer mortars and concrete", Mater. Today Proc., 5(3), 9035-9040. https://doi.org/10.1016/j.matpr.2017.10.018. 
  29. Su, M., Zhong, Q., Peng, H. and Li, S. (2021), "Selected machine learning approaches for predicting the interfacial bond strength between FRPs and concrete", Constr. Build. Mater., 270, 121456. https://doi.org/10.1016/j.conbuildmat.2020.121456. 
  30. Wainer, J. and Cawley, G. (2021), "Nested cross-validation when selecting classifiers is overzealous for most practical applications", Expert Syst. Appl., 182, 115222. https://doi.org/10.1016/j.eswa.2021.115222.