Neuro-Fuzzy Approach for Predicting the Infiltration of Soil

This paper aims to develop an adaptive neuro-fuzzy inference system (ANFIS) approach based model for predicting the infiltration of sandy loam soil. To achieve the objective, cumulative infiltration was observed from prepared soil samples at different moisture contents and different densities with the help of minidisk infiltrometer. Input dataset consists of time, the percentage of sand and clay, moisture content, and bulk density, whereas cumulative infiltration was taken as a target. Different membership functions, i.e., triangular, trapezoidal, Gaussian, and generalized bell-shaped, were implemented in ANFIS-based models. Using statistical performance indices, coefficient of correlation (C.C), root mean square (RMSE), and Nash–Sutcliffe efficiency coefficient (NS) were selected to compare the performance of ANFIS-based models and Kostiakov model. Obtained results suggest that Gaussian-based membership function based on ANFIS model had higher prediction performance than other discussed models.

[1]  V. Singh,et al.  Entropy theory for derivation of infiltration equations , 2010 .

[2]  Varun Singh,et al.  Estimation of models for cumulative infiltration of soil using machine learning methods , 2018, ISH Journal of Hydraulic Engineering.

[3]  D. Stevenson,et al.  Comparing different methods for statistical modeling of particulate matter in Tehran, Iran , 2018, Air Quality, Atmosphere & Health.

[4]  Vijay P. Singh,et al.  Comparison of infiltration models , 2003 .

[5]  Parveen Sihag,et al.  Prediction of unsaturated hydraulic conductivity using fuzzy logic and artificial neural network , 2018, Modeling Earth Systems and Environment.

[6]  N. K. Tiwari,et al.  Support vector regression-based modeling of cumulative infiltration of sandy soil , 2018 .

[7]  Parveen Sihag,et al.  Evaluation of the impact of fly ash on infiltration characteristics using different soft computing techniques , 2018, Applied Water Science.

[8]  Diogo Bolster,et al.  A quick and inexpensive method to quantify spatially variable infiltration capacity for artificial recharge ponds using photographic images , 2012 .

[9]  Parveen Sihag,et al.  Comparison of infiltration models in NIT Kurukshetra campus , 2018, Applied Water Science.

[10]  Parveen Sihag,et al.  Comparative Evaluation of Infiltration Models , 2018, KSCE Journal of Civil Engineering.

[11]  Balraj Singh,et al.  Modeling the infiltration process with soft computing techniques , 2020 .

[12]  Michio Sugeno,et al.  Fuzzy identification of systems and its applications to modeling and control , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[13]  R. Baker,et al.  A DESCRIPTIVE THEORY OF FINGERING DURING INFILTRATION INTO LAYERED SOILS , 1988 .