Browsing by Author "More, S.B."

Now showing 1 - 3 of 3

Estimation of saturated hydraulic conductivity using fuzzy neural network in a semi-arid basin scale for murum soils of India
(2018) More, S.B.; Deka, P.C.
Saturated hydraulic conductivity, Ks is an important input parameter in modeling flow process in soil. Measurement of Ks in field is time consuming and costly. Also, due to inherent temporal and spatial variability of this parameter, large number of samples are required to characterize the areas of site. In this study, a hybrid approach consists of Fuzzy Neural Network (FNN), has been proposed to estimate Ks from limited number of field measurements using Guelph permeameter. The various soil properties such as bulk density, porosity, specific gravity, sand, clay, silt and organic matter were used as input variables and Ks was kept as output. In this study, 175 field measurements and soil samples were collected in a grid of 40 m 200 m with uniform spacing along the slope of barren land in the site of Punanaka (Solapur city), India. To quantify the prediction accuracy, this FNN approach is compared with regression, Fuzzy Mamdani approach and artificial neural network with BP algorithm. The various statistical performance indices like root mean square error, coefficient of determination (R2), and Mean relative error were used for evaluation of model performance. It was found that the hybrid FNN approach in comparison with others could more accurately predict saturated hydraulic conductivity. 2017 Indian Society for Hydraulics.
Estimation of saturated hydraulic conductivity using fuzzy neural network in a semi-arid basin scale for murum soils of India
(Taylor and Francis Ltd. michael.wagreich@univie.ac.at, 2018) More, S.B.; Deka, P.C.
Saturated hydraulic conductivity, Ks is an important input parameter in modeling flow process in soil. Measurement of Ks in field is time consuming and costly. Also, due to inherent temporal and spatial variability of this parameter, large number of samples are required to characterize the areas of site. In this study, a hybrid approach consists of Fuzzy Neural Network (FNN), has been proposed to estimate Ks from limited number of field measurements using Guelph permeameter. The various soil properties such as bulk density, porosity, specific gravity, sand, clay, silt and organic matter were used as input variables and Ks was kept as output. In this study, 175 field measurements and soil samples were collected in a grid of 40 m × 200 m with uniform spacing along the slope of barren land in the site of Punanaka (Solapur city), India. To quantify the prediction accuracy, this FNN approach is compared with regression, Fuzzy Mamdani approach and artificial neural network with BP algorithm. The various statistical performance indices like root mean square error, coefficient of determination (R2), and Mean relative error were used for evaluation of model performance. It was found that the hybrid FNN approach in comparison with others could more accurately predict saturated hydraulic conductivity. © 2017 Indian Society for Hydraulics.
Machine learning-based modeling of saturated hydraulic conductivity in soils of tropical semi-arid zone of India
(Springer, 2022) More, S.B.; Deka, P.C.; Patil, A.P.; Naganna, S.R.
Saturated hydraulic conductivity (Kfs) is the major parameter that affects the movement of water and solutes in soil strata. Although one can estimate the Kfs directly by using various field or laboratory methods, they turn out to be more time-consuming and painstaking while characterizing the spatial variability of Kfs. For this reason, some recent researches employ indirect approaches such as pedotransfer functions (PTF) and surface modeling methods for estimating Kfs of several scales. Pedotransfer functions are often developed by relating the Kfs with readily available soil properties such as bulk density, porosity, sand content, silt content, and organic material. The present research explores the suitability of Extreme Learning Machine (ELM) in developing PTF's for Kfs by using basic soil properties. In-situ field tests and laboratory experiments on collected samples were performed to acquire the datasets necessary for the analysis. Three competitive soft computing approaches, namely the ELM, Support Vector Machine (SVM), and Adaptive Neuro-Fuzzy Inference System (ANFIS) based on Fuzzy C-means Clustering optimized by Genetic Algorithm were exercised for developing the Kfs models. Further, the performance of these approaches in modeling Kfs was evaluated using various statistical mertics. The performance of ELM was found to be good in comparison to the other two models, with sufficiently good NSE values. The ELM model provided Kfs predictions at the Murarji Peth and Punanaka sites with an NSE of 0.90 and 0.83, respectively, while at the Mulegoan site, the ANFIS model was better with R = 0.80 and NSE = 0.64. © 2022, Indian Academy of Sciences.