Prediction of unsaturated hydraulic conductivity using adaptive neuro- fuzzy inference system (ANFIS)
TL;DR: In this article, the unsaturated hydraulic conductivity of soil using Adaptive Neuro- fuzzy inference system (ANFIS), Multi-Linear Regression (MLR), and Artificial Neural Network (ANN) was predicted.
Abstract: This paper aims to predict the unsaturated hydraulic conductivity of soil using Adaptive Neuro- fuzzy inference system (ANFIS), Multi-Linear Regression (MLR), and artificial neural network (ANN). Laboratory experiments carried out on 46 samples of sand, rice husk ash and fly ash (FA) mixture. Out of 46 data-set for modeling of unsaturated hydraulic conductivity 32 random data used for training and remaining 14 to the test. The results suggest improved performance by Gaussian membership function than triangular and generalized bell-shaped membership-based ANFIS. MLR is better than ANN and Gaussian membership function-based ANFIS for unsaturated hydraulic conductivity.
Citations
More filters
01 May 2010
TL;DR: It was found that the constructed RBF exhibited a high performance than MLP, ANFIS and MR for predicting S%.
Abstract: Research highlights? The use of multiple regression (MR), artificial neural network (ANN) and artificial neuro-fuzzy inference system (ANFIS) models, for the prediction of swell percent of soils, was described and compared. ? However the accuracies of ANN and ANFIS models may be evaluated relatively similar, it is shown that the constructed ANN models of RBF and MLP exhibit a high performance than ANFIS and multiple regression for predicting swell percent of clays. ? The performance comparison showed that the soft computing system is a good tool for minimizing the uncertainties in the soil engineering projects. ? The use of soft computing will also may provide new approaches and methodologies, and minimize the potential inconsistency of correlations. In the recent years, new techniques such as; artificial neural networks and fuzzy inference systems were employed for developing of the predictive models to estimate the needed parameters. Soft computing techniques are now being used as alternate statistical tool. Determination of swell potential of soil is difficult, expensive, time consuming and involves destructive tests. In this paper, use of MLP and RBF functions of ANN (artificial neural networks), ANFIS (adaptive neuro-fuzzy inference system) for prediction of S% (swell percent) of soil was described, and compared with the traditional statistical model of MR (multiple regression). However the accuracies of ANN and ANFIS models may be evaluated relatively similar. It was found that the constructed RBF exhibited a high performance than MLP, ANFIS and MR for predicting S%. The performance comparison showed that the soft computing system is a good tool for minimizing the uncertainties in the soil engineering projects. The use of soft computing will also may provide new approaches and methodologies, and minimize the potential inconsistency of correlations.
364 citations
TL;DR: In this article, a new ensemble machine learning model called Extra Tree Regression (ETR) was introduced for predicting monthly WQI values at the Lam Tsuen River in Hong Kong.
Abstract: The Water Quality Index (WQI) is the most common indicator to characterize surface water quality. This study introduces a new ensemble machine learning model called Extra Tree Regression (ETR) for predicting monthly WQI values at the Lam Tsuen River in Hong Kong. The ETR model performance is compared with that of the classic standalone models, Support Vector Regression (SVR) and Decision Tree Regression (DTR). The monthly input water quality data including Biochemical Oxygen Demand (BOD), Chemical Oxygen Demand (COD), Dissolved Oxygen (DO), Electrical Conductivity (EC), Nitrate-Nitrogen ( NO 3 -N), Nitrite-Nitrogen ( NO 2 -N), Phosphate ( P O 4 3 - ), potential for Hydrogen (pH), Temperature (T) and Turbidity (TUR) are used for building the prediction models. Various input data combinations are investigated and assessed in terms of prediction performance, using numerical indices and graphical comparisons. The analysis shows that the ETR model generally produces more accurate WQI predictions for both training and testing phases. Although including all the ten input variables achieves the highest prediction performance ( R 2 t e s t = 0.98 , R M S E t e s t = 2.99 ), a combination of input parameters including only BOD, Turbidity and Phosphate concentration provides the second highest prediction accuracy ( R 2 t e s t = 0.97 , R M S E t e s t = 3.74 ). The uncertainty analysis relative to model structure and input parameters highlights a higher sensitivity of the prediction results to the former. In general, the ETR model represents an improvement on previous approaches for WQI prediction, in terms of prediction performance and reduction in the number of input parameters.
127 citations
TL;DR: Parametric study outcomes suggested that the higher size of medium sand increases the re charging rate, and a higher concentration of impurities in water decreases the recharging rate.
Abstract: In this paper, recharging rate of stormwater filter system is assessed by using predictive models of Gaussian Process (GP) and Support Vector Machines (SVM). Four kernel functions: normalized poly kernel, polynomial kernel, Pearson VII kernel (PUK) and radial basis kernel (RBF) were used with both modelling approaches (GP and SVM). A dataset consists of 678 measurements was collected from the experimental investigations on the infiltration of the storm-water filter system. Out of 678 observations, randomly selected 462 observations were used for training, whereas remaining 216 were used for testing the model. Input variables were comprise of cumulative time (T), the thickness of medium sand bed (B), size of medium sand (S) and concentration of impurities (Conc.), whereas the recharging rate (R) was considered as output. Correlation coefficient (C.C) and root mean square error (RMSE) were used to compare the performance of both modelling approaches. The evaluation of result suggests that Pearson VII based GP regression approach works well as compared to the other kernel functions based on GP and SVM models. Sensitivity analysis suggested that the size of medium sand (S) is an important parameter for predicting the recharging rate of stormwater filter system. Parametric study outcomes suggested that the higher size of medium sand increases the recharging rate, and a higher concentration of impurities in water decreases the recharging rate. Moreover, on expanding the thickness of the medium sand bed the recharging rate of the storm water filter system was observed to be increased.
78 citations
TL;DR: While a GP, GRNN, and GEP model gives a good estimation performance, the ANN model outperforms them and concludes that the parameter time is the most effective parameter for the estimation of infiltration rate.
Abstract: Knowledge of infiltration process is very helpful in designing and planning of irrigation networks. In this study, the Artificial Neural Network (ANN) technique was used to estimate the infiltratio...
54 citations
Cites background or methods from "Prediction of unsaturated hydraulic..."
...Many researchers developed various conventional models for estimating infiltration rate (Kostiakov 1932; Mishra et al. 2003; Philips 1957; Richards 1931; Sihag et al. 2017a; Singh and Yu 1990)....
[...]
...Some researchers used soft-computing in estimating infiltration process (Sihag et al. 2017b; ; Sihag et al. 2018a; Singh et al. 2017; Sy 2006; Tiwari et al. 2017)....
[...]
...…Haghiabi 2015a; Parsaie and Haghiabi 2017a; Parsaie et al. 2017a; Parsaie et al. 2017b; Parsaie et al. 2017c; Roushangar et al. 2014; Rosushangar et al. 2017; Shiri and Kisi 2012; Shiri et al. 2016; Shiri et al. 2017c; Sihag et al. 2017c; Sihag et al. 2018b; Tiwari et al. 2018; Yavari et al. 2017)....
[...]
TL;DR: In this paper, the potential of four conventional infiltration models (Kostiakov, Modified, Novel and Philip's models) were evaluated by least square fitting to observed infiltration data and three statistical comparison criteria including coefficient of correlation (C.C), coefficient of determination (R2) and root mean square error (RMSE) were used to determine the best performing infiltration models.
Abstract: Infiltration models are very helpful in designing and evaluating surface irrigation systems. The main purpose of this study is to compare infiltration models which are used to evaluate infiltration rates of Davood Rashid, Kelat and Honam in Iran. Field infiltration tests were carried out at sixteen different locations comprising of 155 observations by use of double ring infiltrometer. The potential of four conventional infiltration models (Kostiakov, Modified Kostiakov, Novel and Philip’s models) were evaluated by least–square fitting to observed infiltration data. Three statistical comparison criteria including coefficient of correlation (C.C), coefficient of determination (R2) and root mean square error (RMSE) were used to determine the best performing infiltration models. The novel infiltration model suggests improved performance out of other three models. Further a Multi-linear Regression (MLR) equation has been developed using field infiltration data and compare with Support Vector Machine and Gaussian Process based regression with two kernels (Pearson VII and radial basis) modeling. Results suggest that Pearson VII based SVM works well than other modeling approaches in estimating the infiltration rate of soils. Sensitivity analysis concludes that the parameter, time, plays the most significant role in the estimation of infiltration rate. Comparison of results suggests that there is no significant difference between conventional and soft-computing based infiltration models.
49 citations
References
More filters
TL;DR: Van Genuchten et al. as mentioned in this paper proposed a closed-form analytical expression for predicting the hydraulic conductivity of unsaturated soils based on the Mualem theory, which can be used to predict the unsaturated hydraulic flow and mass transport in unsaturated zone.
Abstract: A new and relatively simple equation for the soil-water content-pressure head curve, 8(h), is described in this paper. The particular form of the equation enables one to derive closedform analytical expressions for the relative hydraulic conductivity, Kr, when substituted in the predictive conductivity models of N.T. Burdine or Y. Mualem. The resulting expressions for Kr(h) contain three independent parameters which may be obtained by fitting the proposed soil-water retention model to experimental data. Results obtained with the closed-form analytical expressions based on the Mualem theory are compared with observed hydraulic conductivity data for five soils with a wide range of hydraulic properties. The unsaturated hydraulic conductivity is predicted well in four out of five cases. It is found that a reasonable description of the soil-water retention curve at low water contents is important for an accurate prediction of the unsaturated hydraulic conductivity. Additional Index Words: soil-water diffusivity, soil-water retention curve. van Genuchten, M. Th. 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44:892-898. T USE OF NUMERICAL MODELS for simulating fluid flow and mass transport in the unsaturated zone has become increasingly popular the last few years. Recent literature indeed demonstrates that much effort is put into the development of such models (Reeves and Duguid, 1975; Segol, 1976; Vauclin et al., 1979). Unfortunately, it appears that the ability to fully characterize the simulated system has not kept pace with the numerical and modeling expertise. Probably the single most important factor limiting the successful application of unsaturated flow theory to actual field problems is the lack of information regarding the parameters entering the governing transfer equations. Reliable estimates of the unsaturated hydraulic conductivity are especially difficult to obtain, partly because of its extensive variability in the field, and partly because measuring this parameter is time-consuming and expensive. Several investigators have, for these reasons, used models for calculating the unsaturated conductivity from the more easily measured soil-water retention curve. Very popular among these models has been the Millington-Quirk method (Millington and Quirk, 1961), various forms of which have been applied with some success in a number of studies (cf. Jackson et al., 1965; Jackson, 1972; Green and Corey, 1971; Bruce, 1972). Unfortunately, this method has the disadvantage of producing tabular results which, for example when applied to nonhomogeneous soils in multidimensional unsaturated flow models, are quite tedious to use. Closed-form analytical expressions for predicting 1 Contribution from the U. S. Salinity Laboratory, AR-SEA, USDA, Riverside, CA 92501. Received 29 June 1979. Approved 19 May I960. 'Soil Scientist, Dep. of Soil and Environmental Sciences, University of California, Riverside, CA 92521. The author is located at the U. S. Salinity Lab., 4500 Glenwood Dr., Riverside, CA 92502. the unsaturated hydraulic conductivity have also been developed. For example, Brooks and Corey (1964) and Jeppson (1974) each used an analytical expression for the conductivity based on the Burdine theory (Burdine, 1953). Brooks and Corey (1964, 1966) obtained fairly accurate predictions with their equations, even though a discontinuity is present in the slope of both the soil-water retention curve and the unsaturated hydraulic conductivity curve at some negative value of the pressure head (this point is often referred to as the bubbling pressure). Such a discontinuity sometimes prevents rapid convergence in numerical saturated-unsaturated flow problems. It also appears that predictions based on the Brooks and Corey equations are somewhat less accurate than those obtained with various forms of the (modified) Millington-Quirk method. Recently Mualem (1976a) derived a new model for predicting the hydraulic conductivity from knowledge of the soil-water retention curve and the conductivity at saturation. Mualem's derivation leads to a simple integral formula for the unsaturated hydraulic conductivity which enables one to derive closed-form analytical expressions, provided suitable equations for the soil-water retention curves are available. It is the purpose of this paper to derive such expressions using an equation for the soil-water retention curve which is both continuous and has a continuous slope. The resulting conductivity models generally contain three independent parameters which may be obtained by matching the proposed soil-water retention curve to experimental data. Results obtained with the closedform equations based on the Mualem theory will be compared with observed data for a few soils having widely varying hydraulic properties. THEORETICAL Equations Based on Mualem's Model The following equation was derived by Mualem (1976a) for predicting the relative hydraulic conductivity (Kr) from knowledge of the soil-water retention curve
22,781 citations
"Prediction of unsaturated hydraulic..." refers methods in this paper
...…obtain the net output (zj) to the unit: (18)zj = ∑ i Wij × yi where C1 is the slope of the curve of the cumulative infiltration vs. the square root of time and (A) is a value relating the Van Genuchten (1980) parameters for a given soil type to the suction rate and radius of the infiltrometer disk....
[...]
01 Jan 1985
TL;DR: A mathematical tool to build a fuzzy model of a system where fuzzy implications and reasoning are used is presented and two applications of the method to industrial processes are discussed: a water cleaning process and a converter in a steel-making process.
Abstract: A mathematical tool to build a fuzzy model of a system where fuzzy implications and reasoning are used is presented. The premise of an implication is the description of fuzzy subspace of inputs and its consequence is a linear input-output relation. The method of identification of a system using its input-output data is then shown. Two applications of the method to industrial processes are also discussed: a water cleaning process and a converter in a steel-making process.
18,803 citations
"Prediction of unsaturated hydraulic..." refers methods in this paper
...First-order Sugeno fuzzy model (Sugeno and Takagi 1985) have four fuzzy if-then rule, given as:...
[...]
...…45 5 1 1.34 10 of first-order Sugeno fuzzy model of ANFIS having two inputs, a and b, four rules and one output c. First-order Sugeno fuzzy model (Sugeno and Takagi 1985) have four fuzzy if-then rule, given as: where X1, X2, Y1, and Y2 are fuzzy sets of input a and b, fij (i, j = 1, 2) are the…...
[...]
01 May 1993
TL;DR: The architecture and learning procedure underlying ANFIS (adaptive-network-based fuzzy inference system) is presented, which is a fuzzy inference System implemented in the framework of adaptive networks.
Abstract: The architecture and learning procedure underlying ANFIS (adaptive-network-based fuzzy inference system) is presented, which is a fuzzy inference system implemented in the framework of adaptive networks. By using a hybrid learning procedure, the proposed ANFIS can construct an input-output mapping based on both human knowledge (in the form of fuzzy if-then rules) and stipulated input-output data pairs. In the simulation, the ANFIS architecture is employed to model nonlinear functions, identify nonlinear components on-line in a control system, and predict a chaotic time series, all yielding remarkable results. Comparisons with artificial neural networks and earlier work on fuzzy modeling are listed and discussed. Other extensions of the proposed ANFIS and promising applications to automatic control and signal processing are also suggested. >
15,085 citations
"Prediction of unsaturated hydraulic..." refers background or methods in this paper
...This process finishes at desired epochs (Jang 1993)....
[...]
...The detail of hybrid learning algorithm can be found in Jang (1993)....
[...]
...It learns features of the data-set and adjusts the system characteristics according to a given error criterion (Jang 1993)....
[...]
TL;DR: In this paper, a method is presented for developing probability density functions for parameters of soil moisture relationships of capillary head [h(θ)] and hydraulic conductivity [K(α), which are required for the assessment of water flow and solute transport in unsaturated media.
Abstract: A method is presented for developing probability density functions for parameters of soil moisture relationships of capillary head [h(θ)] and hydraulic conductivity [K(θ)]. These soil moisture parameters are required for the assessment of water flow and solute transport in unsaturated media. The method employs a statistical multiple regression equation proposed in the literature for estimating [h(θ)] or [K(θ)] relationships using the soil saturated water content and the percentages of sand and clay. In the absence of known statistical distributions for either [h(θ)] or [K(θ)] relationships, the method facilitates modeling by providing variability estimates that can be used to examine the uncertainty associated with water flow or solute transport in unsaturated media.
2,050 citations
"Prediction of unsaturated hydraulic..." refers methods in this paper
...Figure 3 shows parameters for the 12 texture classes were obtained from Carsel and Parrish (1988)....
[...]
TL;DR: In this paper, a simple two-term infiltration equation was used to determine soil sorptivity and hydraulic conductivity from cumulative infiltration data from the disk infiltrometer, and the relative error of estimation of soil hydraulic properties was within 5% for most cases.
Abstract: A simple two-term infiltration equation was used to determine soil sorptivity and hydraulic conductivity from cumulative infiltration data from the disk infiltrometer. Parameters of the equation were obtained by fitting the equation to cumulative infiltration data. The parameters of the first and second terms in the equation were related to soil sorptivity and hydraulic conductivity, respectively. By using the two-term infiltration equation, sorptivity and hydraulic conductivity were estimated for various soils, radii and tensions of the disk infiltrometer, and initial infiltration conditions. Sorptivity and hydraulic conductivity values calculated using the method and simulated cumulative infiltration data resulted in excellent agreement with the theoretical results. The relative error of estimation of sorptivity and hydraulic conductivity was within 5% for most cases. The method can be used to determine the soil hydraulic properties from infiltrometer infiltration for a wide range of soils, having a retention function of the type of either van Genuchten, Russo, or Zhang and van Genuchten.
450 citations
"Prediction of unsaturated hydraulic..." refers methods in this paper
...The method proposed by Zhang (1997) is quite simple and works well for measurements of infiltration into unsaturated soil from the recorded data by mini disk infiltrometer (Devices 2014)....
[...]