Geomechanics and Engineering

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Machine learning-based analysis and prediction model on the strengthening mechanism of biopolymer-based soil treatment

  • Haejin Lee (Department of Civil Systems Engineering, Ajou University) ;
  • Jaemin Lee (Department of Mechanical Engineering, Korea Advance Institute of Science and Technology) ;
  • Seunghwa Ryu (Department of Mechanical Engineering, Korea Advance Institute of Science and Technology) ;
  • Ilhan Chang (Department of Civil Systems Engineering, Ajou University)
  • Received : 2023.12.21
  • Accepted : 2024.01.16
  • Published : 2024.02.25
⟨ Previous Next ⟩

Abstract

The introduction of bio-based materials has been recommended in the geotechnical engineering field to reduce environmental pollutants such as heavy metals and greenhouse gases. However, bio-treated soil methods face limitations in field application due to short research periods and insufficient verification of engineering performance, especially when compared to conventional materials like cement. Therefore, this study aimed to develop a machine learning model for predicting the unconfined compressive strength, a representative soil property, of biopolymer-based soil treatment (BPST). Four machine learning algorithms were compared to determine a suitable model, including linear regression (LR), support vector regression (SVR), random forest (RF), and neural network (NN). Except for LR, the SVR, RF, and NN algorithms exhibited high predictive performance with an R2 value of 0.98 or higher. The permutation feature importance technique was used to identify the main factors affecting the strength enhancement of BPST. The results indicated that the unconfined compressive strength of BPST is affected by mean particle size, followed by biopolymer content and water content. With a reliable prediction model, the proposed model can present guidelines prior to laboratory testing and field application, thereby saving a significant amount of time and money.

Keywords

Acknowledgement

This study was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. 2022R1A2C2091517).

References

  1. Abraham, A. (2005), "Artificial neural networks", Handbook of Measuring System Design. 
  2. Agbulut, U,, Gurel, A.E. and Bicen, Y. (2021), "Prediction of daily global solar radiation using different machine learning algorithms: Evaluation and comparison", Renew. Sustain. Energ. Rev., 135, 110114. https://doi.org/10.1016/j.rser.2020.110114. 
  3. Almajed, A., Lateef, M.A., Moghal, A.A.B. and Lemboye, K. (2021), "State-of-the-art review of the applicability and challenges of microbial-induced calcite precipitation (MICP) and enzyme-induced calcite precipitation (EICP) techniques for geotechnical and geoenvironmental applications", Crystals, 11(4), 370. https://doi.org/10.3390/cryst11040370. 
  4. Altmann, A., Tolosi, L., Sander, O. and Lengauer, T. (2010), "Permutation importance: a corrected feature importance measure", Bioinformatics, 26(10), 1340-1347. https://doi.org/10.1093/bioinformatics/btq134. 
  5. Benayoun, F., Boumezerane, D., Bekkouche, S.R.and Bendada, L. (2020), "Application of genetic algorithm method for soil nailing parameters optimization", Proceedings of the IOP Conference Series: Materials Science and Engineering. 
  6. Bharti, S.N. and Swetha, G. (2016), "Need for bioplastics and role of biopolymer PHB: a short review", J. Pet. Environ. Biotechnol., 7(272), 2. https://doi.org/10.4172/2157-7463.1000272. 
  7. Bhattacharya, B. and Solomatine, D.P. (2006), "Machine learning in soil classification", Neural Networks, 19(2), 186-195. https://doi.org/10.1016/j.neunet.2006.01.005. 
  8. Bobbo, T., Biffani, S., Taccioli, C., Penasa, M. and Cassandro, M. (2021), "Comparison of machine learning methods to predict udder health status based on somatic cell counts in dairy cows", Scientific Reports, 11(1), 1-10. https://doi.org/10.1038/s41598-021-93056-4. 
  9. Breiman, L. (2001), "Random forests", Mach.Learning, 45, 5-32. https://doi.org/10.1023/A:1010933404324. 
  10. Breuel, T.M. (2015), "The effects of hyperparameters on SGD training of neural networks", arXiv Preprint arXiv:1508.02788. 
  11. Bullard, J.W., Jennings, H.M., Livingston, R.A., Nonat, A., Scherer, G.W., Schweitzer, J.S., Scrivener, K.L. and Thomas, J.J. (2011), "Mechanisms of cement hydration", Cem.Concr.Res., 41(12), 1208-1223. https://doi.org/10.1016/j.cemconres.2010.09.011. 
  12. Cabalar, A.F., Wiszniewski, M. and Skutnik, Z. (2017), "Effects of xanthan gum biopolymer on the permeability, odometer, unconfined compressive and triaxial shear behavior of a sand", Soil Mech. Found. Eng., 54(5), 356-361. https://doi.org/10.1007/s11204-017-9481-1. 
  13. Cameron, A.C. and Windmeijer, F.A. (1997), "An R-squared measure of goodness of fit for some common nonlinear regression models", J. Econ., 77(2), 329-342. https://doi.org/10.1016/S0304-4076(96)01818-0. 
  14. Chang, I. and Cho, G. (2012), "Strengthening of Korean residual soil with -l,3/1,6-glucan biopolymer", Constr. Build. Mater., 30(1), 30. https://doi.org/10.1016/j.conbuildmat.2011.11.030. 
  15. Chang, I., Im, J., Prasidhi, A.K. and Cho, G. (2015a), "Effects of Xanthan gum biopolymer on soil strengthening", Constr. Build. Mater., 74, 65-72. https://doi.org/10.1016/j.conbuildmat.2014.10.026. 
  16. Chang, I., Jeon, M. and Cho, G. (2015b), "Application of microbial biopolymers as an alternative construction binder for earth buildings in underdeveloped countries", Int. J. Polymer Sci., https://doi.org/10.1155/2015/326745. 
  17. Chang, I., Prasidhi, A.K., Im, J. and Cho, G. (2015c), "Soil strengthening using thermo-gelation biopolymers", Constr. Build. Mater., 77, 430-438. https://doi.org/10.1016/j.conbuildmat.2014.12.116. 
  18. Chang, I., Prasidhi, A.K., Im, J., Shin, H. and Cho, G. (2015d), "Soil treatment using microbial biopolymers for anti-desertification purposes", Geoderma, 253-254, 39-47. https://doi.org/10.1016/j.geoderma.2015.04.006. 
  19. Chang, I., Im, J. and Cho, G. (2016), "Introduction of microbial biopolymers in soil treatment for future environmentally-friendly and sustainable geotechnical engineering", Sustainability, 8(3), 251. https://doi.org/10.3390/su8030251. 
  20. Chang, I., Im, J., Lee, S. and Cho, G. (2017), "Strength durability of gellan gum biopolymer-treated Korean sand with cyclic wetting and drying", Constr. Build. Mater., 143, 210-221. https://doi.org/10.1016/j.conbuildmat.2017.02.061. 
  21. Chang, I. and Cho, G. (2019), "Shear strength behavior and parameters of microbial gellan gum-treated soils: From sand to clay", Acta Geotechnica, 14(2), 361-375. https://doi.org/10.1007/s11440-018-0641-x. 
  22. Chang, I., Lee, M., Tran, A.T.P., Lee, S., Kwon, Y., Im, J. and Cho, G. (2020), "Review on biopolymer-based soil treatment (BPST) technology in geotechnical engineering practices", Transport. Geotech., 24, 100385. https://doi.org/10.1016/j.trgeo.2020.100385. 
  23. Chen, J., de Hoogh, K., Gulliver, J., Hoffmann, B., Hertel, O., Ketzel, M., Bauwelinck, M., Van Donkelaar, A., Hvidtfeldt, U.A. and Katsouyanni, K. (2019), "A comparison of linear regression, regularization, and machine learning algorithms to develop Europe-wide spatial models of fine particles and nitrogen dioxide", Environ. Int., 130, 104934. https://doi.org/10.1016/j.envint.2019.104934. 
  24. Choi, S., Chang, I., Lee, M., Lee, J., Han, J. and Kwon, T. (2020), "Review on geotechnical engineering properties of sands treated by microbially induced calcium carbonate precipitation (MICP) and biopolymers", Constr. Build. Mater., 246, 118415. https://doi.org/10.1016/j.conbuildmat.2020.118415. 
  25. Cohen, I., Huang, Y., Chen, J., Benesty, J., Benesty, J., Chen, J., Huang, Y. and Cohen, I. (2009), "Pearson correlation coefficient", Noise Reduction in Speech Processing, 1-4. https://doi.org/10.1007/978-3-642-00296-0_5. 
  26. Cortes, C. and Vapnik, V. (1995), "Support-vector networks", Mach. Learning, 20(3), 273-297. https://doi.org/10.1007/BF00994018. 
  27. Drucker, H., Burges, C.J., Kaufman, L., Smola, A. and Vapnik, V. (1996), "Support vector regression machines", Adv. Neural Inform. Process. Syst., 9. 
  28. Fatehi, H., Ong, D.E., Yu, J. and Chang, I. (2021), "Biopolymers as green binders for soil improvement in geotechnical applications: A review", Geosciences, 11(7), 291. https://doi.org/10.3390/geosciences11070291. 
  29. Fatehi, H., Ong, D.E., Yu, J. and Chang, I. (2023), "The effects of particle size distribution and moisture variation on mechanical strength of biopolymer-treated soil", Polymers, 15(6), 1549. https://doi.org/10.3390/polym15061549. 
  30. Garcia-Ochoa, F., Santos, V.E., Casas, J.A. and Gomez, E. (2000), "Xanthan gum: production, recovery, and properties", Biotechnol. Adv., 18(7), 549-579. https://doi.org/10.1016/S0734-9750(00)00050-1. 
  31. Goh, A.T. and Goh, S.H. (2007), "Support vector machines: their use in geotechnical engineering as illustrated using seismic liquefaction data", Comput. Geotech., 34(5), 410-421. https://doi.org/10.1016/j.compgeo.2007.06.001. 
  32. Gonos, I.F. and Stathopulos, I.A. (2005), "Estimation of multilayer soil parameters using genetic algorithms", IEEE Trans.Power Del., 20(1), 100-106. https://doi.org/10.1109/TPWRD.2004.836833. 
  33. Kim, M. and Gilley, J.E. (2008), "Artificial Neural Network estimation of soil erosion and nutrient concentrations in runoff from land application areas", Comput. Electron. Agric., 64(2), 268-275. https://doi.org/10.1016/j.compag.2008.05.021. 
  34. Kiran, S., Lal, B. and Tripathy, S.S. (2016), "Shear strength prediction of soil based on probabilistic neural network", Indian J. Sci. Technol., 9(41), 1-6. https://doi.org/10.17485/ijst/2016/v9i41/124740. 
  35. Konstantinou, C., Wang, Y. and Biscontin, G. (2023), "A systematic study on the influence of grain characteristics on hydraulic and mechanical performance of MICP-treated porous media", Transport. Porous Media, 147(2), 305-330. https://doi.org/10.1007/s11242-023-01909-5. 
  36. Kwon, Y., Chang, I., Lee, M. and Cho, G. (2019), "Geotechnical engineering behavior of biopolymer-treated soft marine soil", Geomech. Eng., 17(5), 453-464. https://doi.org/10.12989/gae.2019.17.5.453. 
  37. Kwon, Y., Moon, J., Cho, G., Kim, Y. and Chang, I. (2023), "Xanthan gum biopolymer-based soil treatment as a construction material to mitigate internal erosion of earthen embankment: A field-scale", Constr. Build. Mater., 389, 131716. https://doi.org/10.1016/j.conbuildmat.2023.131716. 
  38. Latifi, N., Horpibulsuk, S., Meehan, C.L., Abd Majid, M.Z., Tahir, M.M. and Mohamad, E.T. (2017), "Improvement of problematic soils with biopolymer-an environmentally friendly soil stabilizer", J. Mater. Civ. Eng., 29(2), 04016204. https://doi.org/10.1061/(ASCE)MT.1943-5533.0001706. 
  39. Lee, S., Chung, M., Park, H.M., Song, K. and Chang, I. (2019), "Xanthan Gum Biopolymer as Soil-Stabilization Binder for Road Construction Using Local Soil in Sri Lanka", J. Mater. Civ. Eng., 31(11), 06019012. https://doi.org/10.1061/(ASCE)MT.1943-5533.0002909. 
  40. Ma, J., Xia, D., Guo, H., Wang, Y., Niu, X., Liu, Z. and Jiang, S. (2022), "Metaheuristic-based support vector regression for landslide displacement prediction: A comparative study", Landslides, 19(10), 2489-2511. https://doi.org/10.1007/s10346-022-01923-6. 
  41. Mekonnen, E., Amdie, Y., Etefa, H., Tefera, N. and Tafesse, M. (2022), "Stabilization of expansive black cotton soil using bioenzymes produced by ureolytic bacteria", Int. J. Geo-Eng., 13(1), 10. https://doi.org/10.1186/s40703-022-00175-6. 
  42. Molnar, C. (2020), Interpretable machine learning, Lulu. com.
  43. Nikou, M., Mansourfar, G. and Bagherzadeh, J. (2019), "Stock price prediction using DEEP learning algorithm and its comparison with machine learning algorithms", Intelligent Systems in Accounting, Finance and Management, 26(4), 164-174.  https://doi.org/10.1002/isaf.1459
  44. Njock, P.G.A., Shen, S., Zhou, A. and Lyu, H. (2020), "Evaluation of soil liquefaction using AI technology incorporating a coupled ENN/t-SNE model", Soil Dyn. Earthq. Eng., 130, 105988. 
  45. Pham, B.T., Hoang, T., Nguyen, D. and Bui, D.T. (2018), "Prediction of shear strength of soft soil using machine learning methods", Catena, 166, 181-191.  https://doi.org/10.1016/j.catena.2018.04.004
  46. Potdar, K., Pardawala, T.S. and Pai, C.D. (2017), "A comparative study of categorical variable encoding techniques for neural network classifiers", Int. J. Comput. Appl., 175(4), 7-9.  https://doi.org/10.5120/ijca2017915495
  47. Reed, M. and Montoya, B.M. (2023), "Influence of the coefficient of uniformity on bio-cemented sands: a microscale investigation", Proceedings of the 8th International Symposium on DEFORMATION CHARACTERISTICS OF GEOMATERIALS, . 
  48. Seo, S., Lee, M., Im, J., Kwon, Y., Chung, M., Cho, G. and Chang, I. (2021), "Site application of biopolymer-based soil treatment (BPST) for slope surface protection: in-situ wet-spraying method and strengthening effect verification", Constr. Build. Mater., 307, 124983. https://doi.org/10.1016/j.conbuildmat.2021.124983. 
  49. Shao, Y. and Lunetta, R.S. (2012), "Comparison of support vector machine, neural network, and CART algorithms for the land-cover classification using limited training data points", ISPRS J. Photogramm. Remote Sens., 70, 78-87. https://doi.org/10.1016/j.isprsjprs.2012.04.001. 
  50. Willmott, C.J. and Matsuura, K. (2005), "Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance", Climate Res., 30(1), 79-82.  https://doi.org/10.3354/cr030079
  51. Wiszniewski, M., Skutnik, Z., Biliniak, M. and Cabalar, A.F. (2017), "Some geomechanical properties of a biopolymer treated medium sand", Annals of Warsaw University of Life Sciences-SGGW Land Reclamation, 49(3), 201-212.  https://doi.org/10.1515/sggw-2017-0016
  52. Worrell, E., Price, L., Martin, N., Hendriks, C. and Meida, L.O. (2001), "Carbon dioxide emissions from the global cement industry", Annu. Rev. Energ. Environ., 26(1), 303-329.  https://doi.org/10.1146/annurev.energy.26.1.303
  53. Zhang, X., Yao, L., Huang, C., Sheng, Q.Z.and Wang, X. (2017), "Intent recognition in smart living through deep recurrent neural networks", Proceedings of the Neural Information Processing: 24th International Conference, ICONIP 2017, Guangzhou, China, November 14-18, 2017, Proceedings, Part II 24.