Prediction of hydraulic conductivity of sodium bentonite GCLs by machine learning approaches

Six machine learning methods (Linear Regression, Logistic Regression, XGBoost, SVM, KNN,and ANN) were used to predict/classify hydraulic conductivity of conventional sodium-bentonite geosynthetic clay liners (Na-B GCLs) to saline solutions or leachates. Data were collected from literature and randomly divided into two groups, i.e., 80% of the data were used to train machine learning models and the rest 20% were applied to evaluate model performance. Features, that are known to affect the hydraulic conductivity of Na-B GCLs (e.g., mass per unit area of GCLs, monovalent and divalent cation, ionic strength (I), relative abundance of monovalent and divalent cations (RMD), swell index, and effective stress), were employed to predict/classify hydraulic conductivity of Na-B GCLs. Comparative analyses were conducted with seven Subsets corresponding to the combination of different features and the best model was determined via cross-validation. The results showed that XGBoost consistently had the best performance among all methods over all Subsets of feature for both regression and classification analyses. Subset 4, using swell index, I, RMD, I2·RMD, monovalent cation, divalent cation, effective stress, and mass per unit area as features, outperformed all other six Subsets in both regression analysis (R2=0.826) and classification analysis (Accuracy=0.887) in the out-of-sample tests.