Estimation of hydraulic conductivity of porous media using data-driven techniques

Abstract Knowledge of hydraulic conductivity (K) is inevitable for sub-surface flow and aquifer studies. Hydrologists groundwater researchers are employing data-driven techniques to indirectly evaluate K using porous media characteristics as an alternative direct measurement. The study examines the ability Adaptive Neuro-Fuzzy Inference System (ANFIS) predict two membership functions (MFs), i.e., triangular Gaussian, support vector machine (SVM) via four kernel functions, linear, quadratic, cubic, Gaussian. used easily measurable parameters namely effective mean grain size, uniformity coefficient, porosity input variables. A 70 30% dataset training testing models, respectively. correlation coefficient (R) root square error (RMSE) were models. Gaussian MF-based ANFIS model outperformed having R RMSE values 0.9661 & 0.0010 0.9532 0.0015, respectively, whereas quadratic kernel-based SVM with 0.9520 0.0015 performs better than other Based on evaluation establishes efficacy in estimating media.