Modeling unsaturated hydraulic conductivity of compacted bentonite using a constrained CatBoost with bootstrap analysis
Source abstract
Abstract
Accurately determining the hydraulic conductivity of unsaturated bentonite is important for modeling subsurface thermo-hydro-mechanical and chemical processes. This study introduced a new hybrid model that employs a constrained categorial boosting (CatBoost) algorithm, combined with a genetic algorithm for hyperparameter tuning, to estimate the hydraulic conductivity of unsaturated compacted bentonite The performance of the constrained CatBoost model was benchmarked against a diverse set of data-driven baseline regression models, including lasso, elastic net, polynomial, k-nearest neighbors, decision tree, bagging tree, random forest, and CatBoost. The results indicated that the constrained CatBoost model offers a superior balance between model robustness and predictive accuracy in estimating the hydraulic conductivity of compacted bentonite-based materials during the wetting phase. The model effectively captured the U-shape relationship between hydraulic conductivity and suction, a key characteristic of bentonite behavior. Additionally, bootstrapping analyses confirmed the model's reliability under data variability, further validating its applicability in environmental and geotechnical applications.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Soil and Unsaturated Flow · Groundwater flow and contamination studies · Dam Engineering and Safety
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (48)
Show source-provided references
1. An introduction to kernel and nearest-neighbor nonparametric regression10.1080/00031305.1992.10475879 ↗
2. Bäck · 2018
3. Bagging predictors
4. Random forests
5. On over-fitting in model selection and subsequent selection bias in performance evaluation
6. Unsaturated hydraulic conductivity of compacted bentonite: revisit of microstructure effects
7. Chernick · 2011
8. Determining the unsaturated hydraulic conductivity of a compacted sand–bentonite mixture under constant-volume and free-swell conditions10.1016/j.pce.2008.10.017 ↗
9. Determining the unsaturated hydraulic conductivity of a compacted sand–bentonite mixture under constant-volume and free-swell conditions10.1016/j.pce.2008.10.017 ↗
10. Applying machine learning methods to estimate the thermal conductivity of bentonite for a high-level radioactive waste repository10.1016/j.nucengdes.2022.111765 ↗
11. Catboost: gradient boosting with categorical features support
12. Efron · 1994
13. Eiben · 2015
14. Modeling gas breakthrough and flow phenomena through engineered barrier systems using a discrete fracture approach10.1016/j.compgeo.2022.105148 ↗
15. Physics-guided, physics-informed, and physics-encoded neural networks in scientific computing
16. Evaluation of equivalent thermal conductivity for carbon fiber-reinforced bentonite through experimental and numerical analysis10.1016/j.compgeo.2023.105880 ↗
17. Hall · 2013
18. Random forests
19. Genetic algorithms
20. A differentiable programming system to bridge machine learning and scientific computing
21. James · 2013 · vol. 112
22. Cross-validation pitfalls when selecting and assessing regression and classification models
23. Classification and regression by randomforest
24. Roulette-wheel selection via stochastic acceptance10.1016/j.physa.2011.12.004 ↗
25. Investigation of the hydro-mechanical behaviour of GMZ bentonite pellet mixtures10.1007/s11440-020-00976-y ↗
26. Laboratory hydraulic testing in unsaturated soils10.1007/s10706-008-9202-7 ↗
27. Hydraulic conductivity, microstructure and texture of compacted claystone/bentonite mixtures saturated with different solutions10.1016/j.clay.2023.106982 ↗
28. The use of feed-forward and cascade-forward neural networks to determine swelling potential of clayey soils10.1016/j.compgeo.2023.105319 ↗
29. A dual-porosity model for analysis of temperature effects on hydro-mechanical behaviour of GMZ bentonite under unconfined conditions10.1016/j.compgeo.2022.105127 ↗
30. Machine learning approaches for prediction of fine-grained soils liquefaction10.1016/j.compgeo.2022.105014 ↗
31. Applying knowledge-guided machine learning to slope stability prediction10.1061/JGGEFK.GTENG-11053 ↗
32. Catboost: unbiased boosting with categorical features
33. Induction of decision trees
34. Deep hidden physics models: deep learning of nonlinear partial differential equations
35. A microstructural insight into compacted clayey soils and their hydraulic properties10.1016/j.enggeo.2013.05.024 ↗
36. Hydraulic conductivity and x-ray diffraction tests of unsaturated bentonites with a multi-ring and their predictions by pores distributions10.1016/j.enggeo.2022.106738 ↗
37. Towards reliable barrier systems: a constrained xgboost model coupled with gray wolf optimization for maximum swelling pressure of bentonite10.1016/j.compgeo.2024.106132 ↗
38. Coupled hydro-mechanical modeling of swelling processes in clay–sulfate rocks10.1007/s00603-022-03039-8 ↗
39. A hydro-mechanical approach to model swelling tests of clay-sulfate rocks10.1007/s00603-023-03343-x ↗
40. An efficient neural network model to determine maximum swelling pressure of clayey soils10.1016/j.compgeo.2023.105693 ↗
41. Hydraulic conductivity and microstructure changes of compacted bentonite/sand mixture during hydration10.1016/j.enggeo.2013.06.013 ↗
42. Microstructure of unsaturated loess and its influence on strength characteristics10.1038/s41598-022-05464-9 ↗
43. An experimental study of the water transfer through confined compacted GMZ bentonite10.1016/j.enggeo.2009.08.003 ↗
44. An experimental study of the water transfer through confined compacted GMZ bentonite10.1016/j.enggeo.2009.08.003 ↗
45. Advances on the knowledge of the buffer/backfill properties of heavily-compacted GMZ bentonite10.1016/j.enggeo.2010.06.002 ↗
46. Temperature effects on the unsaturated permeability of the densely compacted GMZ01 bentonite under confined conditions10.1016/j.enggeo.2011.10.011 ↗
47. An unsaturated hydraulic conductivity model for compacted GMZ01 bentonite with consideration of temperature10.1007/s12665-013-2599-1 ↗
48. Role of different suction components on swelling behavior of compacted bentonites10.1016/j.clay.2015.11.022 ↗