F. Ahmadi; S. Ayashm; K. Khalili; J. Behmanesh
Abstract
Introduction Crop evapotranspiration modeling process mainly performs with empirical methods, aerodynamic and energy balance. In these methods, the evapotranspiration is calculated based on the average values of meteorological parameters at different time steps. The linear models didn’t have a good ...
Read More
Introduction Crop evapotranspiration modeling process mainly performs with empirical methods, aerodynamic and energy balance. In these methods, the evapotranspiration is calculated based on the average values of meteorological parameters at different time steps. The linear models didn’t have a good performance in this field due to high variability of evapotranspiration and the researchers have turned to the use of nonlinear and intelligent models. For accurate estimation of this hydrologic variable, it should be spending much time and money to measure many data (19).
Materials and Methods Recently the new hybrid methods have been developed by combining some of methods such as artificial neural networks, fuzzy logic and evolutionary computation, that called Soft Computing and Intelligent Systems. These soft techniques are used in various fields of engineering.
A fuzzy neurosis is a hybrid system that incorporates the decision ability of fuzzy logic with the computational ability of neural network, which provides a high capability for modeling and estimating. Basically, the Fuzzy part is used to classify the input data set and determines the degree of membership (that each number can be laying between 0 and 1) and decisions for the next activity made based on a set of rules and move to the next stage. Adaptive Neuro-Fuzzy Inference Systems (ANFIS) includes some parts of a typical fuzzy expert system which the calculations at each step is performed by the hidden layer neurons and the learning ability of the neural network has been created to increase the system information (9).
SVM is a one of supervised learning methods which used for classification and regression affairs. This method was developed by Vapink (15) based on statistical learning theory. The SVM is a method for binary classification in an arbitrary characteristic space, so it is suitable for prediction problems (12).
The SVM is originally a two-class Classifier that separates the classes by a linear boundary. In this method, the nearest samples to the decision boundary called support vectors. These vectors define the equation of the decision boundary. The classic intelligent simulation algorithms such as artificial neural network usually minimize the absolute error or sum of square errors of the training data, but the SVM models, used the structural error minimization principle (5).
Results Discussion Based on the results of performance evaluations, and RMSE and R criteria, both of the SVM and ANFIS models had a high accuracy in predicting the reference evapotranspiration of North West of Iran. From the results of Tables 6 and 8, it can be concluded that both of the models had similar performance and they can present high accuracy in modeling with different inputs. As the ANFIS model for achieving the maximum accuracy used the maximum, minimum and average temperature, sunshine (M8) and wind speed. But the SVM model in Urmia and Sanandaj stations with M8 pattern and in other stations with M9 pattern achieves the maximum performance. In all of the stations (apart from Sanandaj station) the SVM model had a high accuracy and less error than the ANFIS model but, this difference is not remarkable and the SVM model used more input parameters (than the ANFIS model) for predicting the evapotranspiration.
Conclusion In this research, in order to predict monthly reference evapotranspiration two ANFIS and SVM models employed using collected data at the six synoptic stations in the period of 38 years (1973-2010) located in the north-west of Iran. At first monthly evapotranspiration of a reference crop estimated by FAO-Penman- Monteith method for selected stations as the output of SVM and ANFIS models. Then a regression equation between effective meteorological parameters on evapotranspiration fitted and different input patterns for model determined. Results showed Relative humidity as the less effective parameter deleted from an input of the model. Also in this paper to investigate the effect of memory on predict of evapotranspiration, one, two, three and four months lag used as the input of model. Results showed both models estimated monthly evapotranspiration with the high accuracy but SVM model was better than ANFIS model. Also using the memory of evapotranspiration time series as the input of model instead of meteorological parameters showed less accuracy.
majid montaseri; sarvin zamanzad ghavidel
Abstract
Introduction: A total dissolved solid (TDS) is an important indicator for water quality assesment. Since the composition of mineral salts and discharge affects the TDS of water, it is important to understand the relationships of mineral salts composition with TDS.
Materials and Methods: In this study, ...
Read More
Introduction: A total dissolved solid (TDS) is an important indicator for water quality assesment. Since the composition of mineral salts and discharge affects the TDS of water, it is important to understand the relationships of mineral salts composition with TDS.
Materials and Methods: In this study, methods of artificial neural networks with five different training algorithm,Levenberg-Marquardt (LM), Scaled Conjugate Gradient (SCG), Fletcher Conjugate Gradient (CGF), One Step Secant (OSS) and Gradient descent with adaptive learning rate backpropagation(GDA)algorithm and adaptive Neurofuzzy inference system based on Subtractive Clustering were used to model water quality properties of Zarrineh River Basin, to be developed in total dissolved solids prediction. ANN and ANFIS program code were written in MATLAB language. Here, the ANN with one hidden layer was used and the hidden nodes’ number was determined using trial and error. Different activation functions (logarithm sigmoid, tangent sigmoid and linear) were tried for the hidden and output nodes. Therefore, water quality data from seven hydrometer stationswere used during the statistical period of 18years (1993-2010).In this research, the study period was divided into two periods of dry and wet flow, and then in a preliminary statistical analysis, the main parameters affecting the estimation of the TDS are determined and isused for modeling. 75% of data are used for remaining and 25% of the data are used for evaluation of the model, randomly. In this paper, three statistical evaluation criteria, correlation coefficient (R), the root mean square error (RMSE) and mean absolute error (MAE) were used to assess models’ performances.
Results and Discussion: By applying correlation coefficients method between the parameters of water quality and discharge with total dissolved solid in two periods, wet and dry periods, the significant (at 95% level) variables entered into the model were Q, HCO3., Cl, So4, Ca, Na and Mg. The optimal ANN (LM) architecture used in this study consists of an input layer with seven inputs, one hidden and output layer with two and five neurons for dry and wet periods, respectively. Similar ANN(LM), ANFIS-SC model had the best performance. It is clear that the ANFIS with 0/72 and 0/58 radii value has the highest R and the lowest RMSE for dry and wet periods, respectively. Comparing the ANFIS-SC estimations with the measured data for the test stage demonstrates a high generalization capacity of the model, with relatively low error and high correlation. From the scatter plots it is obviously seen that the ANFIS-SC predictions are closer to the corresponding measured TDS than other models in two periods. As seen from the best straight line equations (assume the equation as y=ax) in the scatter plots that the coefficient for ANFIS-SC is closer to 1 than other models. In addition ANFIS-SC performancedwith the correlation coefficients in dry and wet periods, respectively 0.975 , 0.969 and with Root-mean-square errors, respectively 34.41 , 23.85 in order to predict dissolved solids (TDS) in the rivers of Zarrineh River Basin. The obtained results showed the efficiency of the applied models in simulating the nonlinear behavior of TDS variations in terms of performance indices. The results are also tested by using t test for verifying the robustness of the models at 99% significance level. Comparison results indicated that the poorest model in TDS simulation was ANN-GDAin dry and wet periods, especially in test period. The observed relationship between residuals and model computed TDS values shows complete independence and random distribution. It is further supported by the respective correlations for ANFIS-SC models (R2 = 0.0012 for dry period and R2 = 0.0214 for wet period) which are negligible small. Plots of the residuals versus model computed values can be more informative regarding model fitting to a data set. If the residuals appear to behave randomly it suggests that the model fits the data well. On the other hand, if non- random distribution is evident in the residuals, the model does not fit the data adequately. On the base of these results, we propose ANFIS-SC and ANN (LM) methods as effective tools for the computation of total dissolved solids in river water, respectively.
Conclusion: It can be concluded that the ANN with Levenberg-Marquardt training algorithm and ANFIS-SC models can be considered as promising tools for forecasting TDS values, based on water quality parameters. With attention to the aim of current research that is presenting the feasibility of artificial intelligence techniques for modeling TDS values, it is notable that the results presented in this paper are for research purpose and applying the abstained results for real-world needs some complicated steps and building artificial intelligences methods, based on complete data and parameters maybe affected the TDS values
