جستجوی مقالات مرتبط با کلیدواژه « آزمون گاما » در نشریات گروه « کشاورزی »

Flood is one of the most devastating natural disasters, causing financial and human losses each year. At the same time, many rivers in Iran's watersheds lack complete and accurate statistics and information. On the other hand, estimating the flow of floods is one of the most important factors for the design and implementation of water structures. In such cases, one of the appropriate solutions to estimate the maximum flow rate with different return periods is flood analysis. In order to conduct the present study, 55 hydrometric stations with a common statistical period of 20 years were considered to perform the work after the statistical deficiencies were eliminated. Then, based on the distribution of the third type of Pearson logo with the lowest error rate and the highest number of first rank as the most suitable fit function, the amount of discharge in different return periods was estimated. The following information was collected on the types of physiography, land use, climate and geology variables. After collecting information about all independent variables using Gamma test, the most important variables affecting the maximum instantaneous flow, including area, drainage density, maximum 24-hour rainfall and watershed environment, were selected and modeled using methods. Random forest modeling and support vector modeling were performed and their efficiency was determined based on statistical indicators With an efficiency coefficient of 74 to 83%, the error of 3.05 to 32.11 m3 and the coefficient of explanation of 76 to 91 are more accurate than the random forest model.

Using transfer functions to predict soil moisture parameters has been considered strictly a scientific and economical method among researchers. In this research, field capacity (FC) and permanent wilting point (PWP) of soil were predicted using classic regression (linear and non-linear), support vector machine (SVM) algorithm, and gene programming expression (GEP) algorithm based on three performance assessment criteria as determination of coefficient (R2), root mean square error (RMSE), and standardized developed discrepancy Ratio (DDR) in the Arasbaran plain in the northwest of Iran. Independent parameters were determined as clay percent (Cl), silt percent (Si), gravel percent (Sa), organic carbon (OC), bulk density (ρb), and actual density (ρs) which (S, ρb, ρs) and (ρb, ρs) were opted to predict FC and PWP using Gamma test, respectively. The results showed that each three transfer functions are capable to simulate FC and PWP parameters but the SVM algorithm is the superior predictor among the three functions so the values of (R2, RMSE, and DDRmax) of training and testing phases for FC were obtained (0.9908, 0.5517, 17.50), (0.9785, 0.7004, 11.62) and those of PWP were calculated (0.9782, 0.5764, 2.85) and (0.8389, 1.187, 3.09), respectively.

The properties of wood composite products are due to the control variables of factory machines. Identifying these sensitive places is of particular importance, because it allows the operator to be able to adjust the production line and product quality in a completely engineered way. Also, it allows operator to have a more detailed study of material and process behavior. Factory available data are including variables of moisture content of chips, chips density, density Mat, material temperature The amount of glue used in percent, the amount of glue used in kilograms, fiber moisture and press speed, and also properties of density, flexural modulus, modulus of elasticity, internal bonding, board moisture and thickness swelling. The effect of input variables on each of the properties of fiberboard was determined by artificial neural network method. The results show that all of the input variables have affected on the output properties of fiberboard and the order in which the influential variables are depends on the type of properties. The effect of variables such as density Mat is almost greater than other variables. Some input variables, such as the, the amount of glue used in kilograms, have less effect on the properties of fiberboard with medium density, and this variable is considered constant in the production line of some factories. The high correlation between the prediction value and the actual value of the output and its low mean absolute percent error indicates the high validity of the prediction that have a high value of 0.9 and less than 9%, respectively.

Vertical velocity distribution at hydraulic jump is one of the challenging and significant issues among researchers because of the complexity of measurement and calculations. In this research, the application of the SVM and GEP intelligent models has been considered to determine the vertical velocity profile at the submerged hydraulic jump downstream of a sluice gate. Laboratory measured data of number 312 has been used in the simulation. Using dimensional analysis, dimensionless input parameters were introduced to models including  upstream Froud number (Fr1), Tail water Froud number (Fr3), the ratio of upstream flow depth to the tailwater depth , the ratio of the gate opening to the channel width , and the ratio of vertical distance from the channel bed to the channel width . Using the gama test, all five parameters were determined as the optimum combination to simulate velocity profile. Of two Nu-SVM and C-SVM classification models, the first one was opted as optimum model of the SVM algorithm with RBF Kernel function with the setting parameters γ and Nu of values 1.2 and 0.486, respectively. The performance of the Nu-SVM and the GEP intelligent,models were assessed using statistical criteria. The results showed that the values of (RMSE, R2, ) indices for the test phase of the Nu-SVM and the GEP algorithms are (0.09588,0.9770,0.4489) and (0.1161,0.9718,0.3588) respectively; illustrating the superiority of the Nu-SVM algorithm. Also, according to the gana test, the arrangment of the effective dimensionless parameters on the velocity profile is , Fr1, , Fr3 and .

Soil erosion and sediment transportation decrease water resources, and cause many social and economic problems. On the other hand, sediment transportation by rivers causes problems such as water quality degradation, reservoirs sedimentation, redirect of rivers, or decrease in their transportability. Therefore, finding the proper methods in sediment yield study in watersheds is essential in planning and management of land and water resources.  Climatic characteristics, physiography, geology, and hydrology of basins are the most effective factors in producing and transporting sediments according to several sources, but the role and impact of some factors are more pronounced than the others in different areas. As a result, the objective of this study was to investigate and identify the most important climatic, physiographic, geological, and hydrological factors in several watersheds of the northeastern part of Iran, by applying Gamma Test (GT) and principal component analysis (PCA) techniques. In this study, the data of discharge flow and suspended sediment concentration, and daily flow discharge recorded in 15 hydrometric stations in Mashhad and Neyshbour restricts and required maps were provided from the Regional Water Company of Khorasan Razavi, Iran. After drawing statistical bar graph period of suspended sediment, daily discharge, annual precipitation, and relatively adequate data, stations with the longest period and with the lowest deficit data were selected to determine the common statistical periods. Therefore, in this study, the time period of 1983-1984 to 2011-2012 was selected, and the run test was applied to control data quality and homogeneity. Then, the most effective factors of sediment yield were determined by principal component analysis (PCA) and Gamma Test (GT). The results of the principal component analysis showed that 90 percent of the first five components justify the changes. Among the factors, area and gross gradient of the mainstream from the first component, the average annual flow rate of mainstream, meandering waterways of the mainstream from second component, and drainage density of third component were identified as the most important influencing factors on suspended sediment production. Ninety superior combinations of 1500 proposed combinations were obtained by Gamma Test to evaluate the effects of each parameter on suspended sediment yield. To determine the order of importance of the entered parameters, first, Gamma Test was performed on all 12 parameters. Gamma values of all cases for each proposed combination were compared. The results showed that the impact of these statistics was lowered by eliminating high gamma parameters and the removal of low values. The data analysis revealed that the low levels of gamma and high accuracy of ratio to find the desired outputs from entries. By lowering the gradient, the complexity of the model was lowered and more suitable model was provided. As a result, high levels of gradient represented the complexity of the final model. The results of the percentage values of each of the 12 variables were considered among the superior equations for estimating the suspended sediment composition. In this regard, the mean annual discharge, main channel length, area, average annual rainfall, and percentage of the outcrop of erosion sensitive rocks with a total of 63 percent of the proposed equations were the most important factors affecting the sediment yield in the study area. The average height parameter of area, the average and gross slope of the mainstream had the lowest presence among the optimized compounds. Based on the results of the principal component analysis, the two factors of basin area and gross slope of the mainstream were selected as the most important factors affecting the amount of annual suspended sediment load, respectively. Based on the results of the Gamma Test, 12 main variables affecting suspended sediment load were identified and the effect of each of them on the production and transport of suspended sediment was determined. Based on the comparison of the results of the two methods of PCA and GT, it can be concluded that if the purpose of research or study is to prepare a model with the highest accuracy in estimating suspended sediment load, the 12-variable model of GT includes factors related to physiographical, geological, climatic and hydrological factors are suggested. However, if the preparation of a model with appropriate accuracy and a limited number of input variables is considered, a 5-variable model derived from the PCA method is proposed. At the same time, if the purpose is to prepare a model with the least input variables and their easy access and calculation and initial estimation of suspended sediments, a bivariate model (based on basin area and gross slope of the mainstream factors) resulting from PCA is proposed. According to the results of the present study, it can be concluded that the study of more parameters has provided grounds for evaluating their importance in sediment yield. Finally, due to the correlation of many parameters with each other, a limited number of parameters that have a more important role in suspended sediment estimation, were selected. Another finding of this study is the increase in the accuracy of the sediment model’s preparation due to achieving more important and effective parameters in sediment yield and identifying them in order to investigate the best sediment management measures in watersheds. It is suggested that similar research should be done in other watersheds with different conditions in terms of climatic conditions, topography, geology, and so on.

this study aims to determine the most effective features on the cost of drip irrigation systems in four parts, including the Cost of pumping station and central control system (TCP), Cost of on-farm equipment (TCF), Cost of installation and operation on-farm and pumping station (TCI) and Total cost (TCT). First, data of 100 drip irrigation projects implemented in different parts of the country were collected and it was prepared a database containing 39 important and influential variables in the cost of the mentioned parts. Based on the sensitivity analysis, the best evaluation criteria were obtained in TCP and the numerical amount of gamma statistic, Expected Absolute Error, Gradient statistic, Standard Error of Γ, coefficient of determination (R2), and V-Ratio index were recorded as 0.048, 0.219, 0.008, 0.024, 0.87 and 0.192, respectively, which indicate the high correlation between the experimental variables and the cost of the corresponded sector. To find the optimal combination of data for cost modeling, we used Genetic Algorithm (GA), Hill Climbing (HC), and Full Embedding (FE). The results showed that the number of required variables and the optimal input combination, which covered 40 and 90% of the variables (16 and 35 variables, respectively) in GA and HC method reached 20% in the FE method and only eight variables were selected for cost modeling and also the results of this method were selected as the superior model. Moreover, the result of the hybrid model revealed the simplest and most optimal model was obtained when QT (l/s) (total amount of available water flow), SR (m) (plant row spacing), QE (l/s) (emitter flow), T (h) (number of working hours per day) and NIT (n) (number of irrigation shifts) were used as the optimal input combination to modeling the cost of drip irrigation systems.

The simulation of the rainfall-runoff process in the watershed has particular importance for a better understanding of hydrologic issues, water resources management, river engineering, flood control structures, and flood storage. In this study, to simulate the rainfall-runoff process, rainfall and discharge data were used in the period 1997-2017. After data qualitative control, rainfall, and discharge delays were determined using the coefficients of autocorrelation, partial autocorrelation, and cross-correlation in R Studio software. Then, the effective parameters and the optimum combination were determined by the Gamma test method and used to implement the model under three different scenarios in MATLAB software. Gamma test results showed that todaychr('39')s precipitation parameters, precipitation of the previous day, discharge of the previous day, and discharge of two days ago have the greatest effect on the outflow of the basin. Also, the Pt Qt-1 and Pt Pt-1 Qt-1 Qt-2 Qt-3 combinations were selected as the most suitable input combinations for modeling. The results of the modeling showed that in the support vector machine model, the Radial Base kernel Function (RBF) has a better performance than multiple and linear kernels. Also, the performance of the Artificial Neural Network model (ANN) is better than the Support Vector Machine model (SVM) with Radial Base kernel Function (RBF).

Considering the natural conditions of Iran, not paying attention to floods can cause irreparable damages, among which flood estimation and zoning of floodplain areas are very significant in controlling hazards, so zoning of climate change is necessary. The present study aims to investigate the risk of floods in selected basins of Khorasan Razavi using the VIKOR, L-THIA, and ANT models. Then, fourteen variables affecting the occurrence of floods including climate, land use, altitude, drainage density, geomorphological units, lithology, runoff height, permeability, slope and direction, distance to rivers/waterways, precipitation, temperature, and soil were used. The results showed that among the mentioned variables, climate parameters, land use, slope, drainage density, distance to rivers/waterways, precipitation, soil, and geomorphological units have greater effects on the occurrence of floods according to statistical calculations. Quantitative and qualitative evaluation of the results using various statistics showed that the L-THIA model, with a γ=0.8, had the highest correlation with the primary layers and was more accurate and efficient than the two VIKOR and ANT models in flood prediction.

One of the important and complex steps for nonlinear modeling is pre-processing of input data in order to select the appropriate combination of them in the model. The gamma test was used to select the optimal combination of input variables for available potassium modeling in this study. The M test was used for determining the optimal number of data needed for modeling. Eight input variables were used for modeling. Modeling the available potassium was done by the number of optimum points and selected variables with subtractive clustering in the fuzzy neural system. The results showed that six variables (clay percentage, silt, organic matter, electrical conductivity, saturation moisture and pH) are the optimal combination of variables in modeling the available potassium in Mian-Kangi region. Also, 112 of measured data (60%) were considered as suitable data for the modeling training section using the M test results. The results indicated that the M method has better accuracy and speed than the trial and error method for finding the appropriate number of input data in training section. The results of modeling also indicated that the fuzzy neural method has high capability and performance in estimating the amount of available potassium in the soil of Mian-Kangi region (R2 = 0.90 and RMSE = 4.27). Also, organic carbon percentage was the most important input for modeling and predicting the amount of available potassium.

Accurate and reliable forecasts of river flow are required for proper management of watershed systems. In recent years, data-driven models and especially artificial intelligent based models have been successfully used in various areas related to water resources. However, uncertainty analysis of these models has been less appreciated in prior studies. In the present study, the output uncertainty of five data-driven models including modular, PCA (Principle Component Analysis), TLRN (Time-Lagged Recurrent Network), ANFIS (Adaptive-Network-based Fuzzy Inference System) and SVM (Support Vector Machine) type models in forecasting river flow has been investigated using 95PPU, p-factor and d-factor quantities. Using the observed meteorological and flow data during 1998-2012 in Hablehroud Basin, different structures of the proposed models were trained and tested. The final values of p-factor and d-factor for each model type were obtained. The results showed that SVM with a p-factor of 82% produces the most reliable forecasts in the present study.

Analysis of suspended sediment load data in rivers is the basis for understanding the trend of erosion and sediment in the management and planning of soil and water resources. Due to lack of access to daily suspended sediment loading data with direct measurement, it is important to use methods for modeling and estimating it in watersheds. One of the best methods used in this field is the use of artificial neural networks. To evaluate daily suspended sediment load, Sira hydrometric station was studied in Karaj River watershed. The number of data used in this study included 624 information records of 31 years (1981–2011) statistical period .Input data to the artificial neural network models included instantaneous flow discharge, average daily flow discharge, average daily flow discharge with a delay of three days, average daily precipitation and average daily precipitation with a delay of three days. Output data to models was daily suspended sediment load. In this research, gamma test and genetic algorithm were used to obtain optimal variables and best combination of variables for entering the model. Then, these combinations with some combination of test and error variables were entered to artificial neural network models. The self-organizing map neural network was used for data clustering and all data were divided into three homogeneous groups: 70 percentage training data, 15 percentage validation data and 15 percentage test data. Then, the combination of variables entered to neural network models with activation functions log sigmoid and tangent sigmoid. The results showed that the neural networks using the optimal variable combinations in comparison with manual combinations have a more accurate estimate for suspended sediment load. In all combinations of inputs to neural network models, a model with tangent sigmoid activation function, with input variables combination including, instantaneous flow discharge (Q), average daily flow discharge (Qi), average daily flow discharge for two day ago (Qi-2), average daily flow discharge for three day ago (Qi-3), average daily precipitation (Pi), average daily precipitation for two day ago (Pi-2) and average daily precipitation for three day ago (Pi-3), was the best model for estimating daily suspended sediment load. This model has the lowest of error (MAE=500.05 (ton/day), RMSE=1995.33(ton/day) and Erel=7%), the highest accuracy (R2=0.96), the highest performance model (NSE=0.96) and has the lowest general standard deviation (GSD=0.97) compared to other models. Also, this model is the best combination with the most influential input variables derived from gamma test and genetic algorithm for estimating SSL.

The need for optimized exploitation of water resources has been increased due to the limited water resources and the different section´s competition. For this purpose, data of six selected synoptic stations including Mashhad, Shiraz, Tabriz, Kermanshah, Khorramabad and Urmia stations were used. Input variables consist of mean temperature (T), relative humidity (RH), sunshine hour (S) and wind speed at 2 m elevation (U2). The M test method was used to determine the length of test period. Since, both Gama index and Standard Error are closed to the axis at the end of figures, the last five-year results were used to test the models. According to the gamma test results, the best input parameters for Mashhad, Shiraz, Tabriz, Kermanshah, Khorramabad and Urmia are respectively (S, U2, RH), (T, U2, RH, S), (T, U2, RH, S), (T, U2, RH), (T, RH, S), (RH, S) under the combined conditions and in a same way, the lowest gamma are 0.005, -0.01, 0.001, -0.002, 0.008, 0.009. Local Linear Regression (LLR), Dynamic Local Linear Regression (DLLR), ANN (conjugate gradient) and ANN (BFGS) models were used to estimate the annual evapotranspiration. The R, MAE, RMSE, MBE, Jakouvidiez (t) and Sabagh (R2/t) criteria were used to evaluate the proposed models. The results showed that the best performance was obtained for the stations; Mashhad, Kermanshah, Tabriz and Shiraz using the best inputs, so that the correlation coefficients for neural network model conjugate gradient were 0.91, 0.98, 0.96 and 0.97, respectively. The general results showed that the non-parametric methods are able to estimate the annual ET, properly.

In this research, the data of the machine hydrometric station was used in the Rood Zard watershed with a statistical period of 36 years (1977-2012). In order to reduce time and cost, pre-processing of input data into the GEP model was obtained using gamma test method and entered the GEP model along with non-preprocessing combinations of the test and error method. The results of comparison between all models showed that the best combination of input variable from gamma test with the lowest standard error is zero, gamma statistic is 0.000092 and Vratio statistic is 0.012 and the combination of variables including average daily flow discharge with two steps of time delay and average daily precipitation with three steps of time delay, had the most accurate and correct estimate for suspended sediment load. This model had the lowest value of RMSE=1671.90 (ton/day) and MAE=475.68 (ton/day) and the highest value of R2=0.99 and NSE=0.99 compared to other models. Therefore, the use of gamma test method as a data preprocessing method, by selecting combinations of appropriate input variables to models, an average of up to 40% of the estimated error (RMSE) of daily suspended sediment load compared to the inputs from the test and reduce the error and increase the performance of the GEP model in estimating the suspended sediment load by increasing the similarity between the values of observational data with computational data.

Flood discharge estimation with different return periods is one of important factors for water structures design and installation. On the other hand, a lot of rivers existing in Iran watersheds have no complete and accurate hydrometric data. In these cases, one of the suitable solutions to estimate peak discharges with different return periods is the regional flood analysis. In this research, 55 hydrometric stations were used. For this purpose, at first, peak discharges in different return periods were estimated using the EasyFit software. Then, the effective variables on the peak discharges were collected and the input variables of the models were selected by using gamma test with the help of the WinGamma software. Finally, data modeling was performed using the support vector machine, artificial neural networks and nonlinear multivariate regression techniques. Quantitative and qualitative assessment of the results using various indices including Nash-Sutcliffe Efficiency Coefficient (NSC) showed that SVM modeling method had the most accuracy in comparison to the other two modeling methods to predict the peak discharges in the Namak Lake Watershed.

This research uses the multilayer perceptron (MLP) model to predict daily evaporation at two synoptic stations located in Rasht and Manjil, Guilan province, in north-west of Iran. Initially the most important combinations of climatic parameters for both of the stations were identified using the gamma test; and daily evaporation were modeled based on the obtained optimal combination. The results of the artificial neural network- Gamma Test (ANN-GT) model are evaluated using the root mean square errors (RMSE), correlation coefficient and Nash-Sutcliffe (NS) criteria. The results showed that the ANN-GT model for Rasht station with a correlation coefficient 0.86, root mean square error 0.95 and Nash-Sutcliffe criteria 0.74 and for Manjil station with correlation coefficient 0.94, root mean square error1.58 and Nash-Sutcliffe criteria 0.89 has an acceptable performance in predicting daily evaporation. To evaluate the uncertainty, we considered a percentage of data which were included in 95 percent of uncertainty (p-factor) and the average width of the 95ppu band (d-factor). Regarding the uncertainty results, the average with of 95PPU bound were obtained as 0.33 and 0.3 for the Manjil and Rasht stations, respectively. This shows the low uncertainty level of the ANN-GT model for predicting daily evaporation at both of the stations. Furthermore, the percentage of the observed data at 95PPU band was low and equal to %25 and %45 for the Rasht and Manjil stations, respectively. The reason for these low values can be due to low uncertainty in the parameters.

Soil saturated hydraulic conductivity is one of the most important physical characteristics of soils which affects water movement in soil. Knowledge of this parameter can help to understand and solve environmental problems. But measurement of this parameter by direct laboratory and field methods is hard, time consuming and expensive. Thus there is need to use alternative methods based on conveniently available soil properties to estimate it with less effort, time and cost. Nonparametric methods are new indirect methods to estimate hydraulic properties of soil, including soil saturated hydraulic conductivity (ks). The aim of this study was to use two methods such as M5P decision tree and an IBk instance-based learning method, which is a classifier with k nearest neighbors to estimate ks from conveniently available properties of soil. In this study a dataset of 151 soil samples which was collected from a site in Bojnord province was used. Conveniently available soil properties included sand, silt and clay percentage, bulk density, particle density, EC, OC, TNV, saturated moisture and pH. Saturated hydraulic conductivity was measured with the Guelph permeameter. The Gamma test was used to determine important parameters for predicting and the modeling procedure of ks. Then various combinations of parameters of the data set were compared to each other based on their Gamma value, to determine the optimum combination of parameters for modeling ks. Using the optimum combination which had the least Gamma value, the M5P decision tree and the IBk instance-based learning methods were performed. To improve the IBk, two different distance weighting systems were used. Finally, evaluation statistics of each model including R2, RMSE, MAE, and MAPE were calculated. The optimum combination determined by the Gamma test which was then used for modeling, included sand, silt and clay percent, TNV percent, EC, and bulk density. The tree selected bulk density as the most important discriminative parameter, and constructed 3 linear equations for predicting ks, based on the bulk density value. Evaluation criteria calculated for this model with RMSE= 23.89 cm/d and MAPE= 20.50% it didn’t predict ks accurately. Different weighting systems didn’t improve IBk performance. Also the IBk model with RMSE= 31.23 cm/d and MAPE= 23.24% didn't estimate ks accurately. The decision tree model performed better than the instance-based learning model to estimate ks. Also the tree showed some information about the structure of the studied soil.

Accurate simulation runoff process can have a significant role in water resources management and related issues. The inherent complexity of this process makes difficult the use of physical and numerical models. In recent years, application of intelligent models is increased a powerful tool in hydrological modeling. The aim of this study was the application of the Gamma test to select the optimal combination of input variables for runoff modeling in Sofi Chay. Streamflow modeling was performed based on the optimum number of the selected variables using the artificial neural network (ANN) and Support vector machine (SVM) methods .Gamma test results showed that monthly runoff with six antecedent runoff values provide better results to predict. Runoff simulation using support vector machines and artificial neural network models also showed that the best input structure will be delayed until six to predict of next month runoff. Among to models with the same input structure, support vector machine have relatively high efficiency compared to artificial neural network .

This research assessment ability of entropy theory and Gamma test for input variable of Artificial Network and Support Vector Machine as evaporation estimation for Rasht, Astara and Anzali in Guilan province. According the results, for Rasht synoptic, Astara and Anzali, Entropy Theory is determined that existence all variables are effective for modelling. Gamma test, for Rasht station two variables contain maximum and average humidity, for Anzali station three variables, contain minimum and average temperature and average humidity and for Astara station one variable contain wind speed eliminated form optimal composition. According to results two model’s Artificial Neural Network and Support Vector Machine performance have been acceptable. For determination of single input data in Rasht station Entropy Theory method and in the Anzali station Gamma test method have had good performance. For Astara station both of them have had good performance. Generally, according to the results, it can be said that Entropy theory had the stronger performance than Gamma test. But at the look of managerial, because the a few input variable selection than the Entropy theory, the Gamma performance has been acceptable than Entropy theory.

Overview, drought is effected an unusual dry period which is enough continued and causes imbalance in the hydrologic status, as depletion of surface water and groundwater resources. The purpose of this research is modeling meteorological drought prediction using Neural Network- Multi layer Perceptron, parameters and climatic signals in three time scales include short, middle and long term in a rain-gauge station located at south plain of Qazvin Province. Three different scenarios were tested as inputs model. Optimal combination of variables was determinate by Gamma-Test after identification of input variables using cross-correlation. Results showed, influence of climatic signals increased and against the influence of meteorological parameters decreased when time scale were increased from short-term to long-term. MEI (Multivariate ENSO Index) and rainfall were introduced as the most effective climatic signals and meteorological parameter for each scale, respectively. Neural Network modeling which has hidden layer with enough neurons, Sigmoid Function in middle layer and linear function at output layer was used. The most appropriate of the number neurons was determined in each scenario and wasn’t observed significant correlation between increasing or decreasing the error and number of neurons. Finally, the most appropriate network structure was determined based on evaluation indexes in three scenarios and each time scale.

The importance of water resources planning and management, the fast growing population, and the limited surface water resources, have made the application of the new technology to forecasting of river flow. A necessity, various methods have been presented in recent years to forecast the river flow, and the data-based models are considered the most reliable for this purpose. The river flow in the Karaj Catchment has been simulated using the data based models (KNN and M5). Hydroclimatological data (discharge, precipitation, temperature and evaporation) for the 2002 to 2009 duration have been collected to carry out the simulation processes. The performance and accuracy of the models were examined and compared. The Gamma test was used to select appropriate compositions. Suitable compositions were determined as the model inputs (KNN and M5). These features were entered in to the two data-based models. Results showed that both models simulated reliable flow predictions, if the discharge had been entered as an input. The M5 model showed a better precision as compared with the KNN model. The Coefficient of determination (R2) for the KNN and M5 models were 0.97 and 0.93, respectively. The RMSE were 0.55 and 0.87, for the same two models, respectively, and the value of the KGE were 0.99, 0.96, respectively