The assessment of precipitation_runoff model by using of artificial neural network and regression methods (Case study: Minab Basin)
2009
Zoratipour, A., Scientific Member of Ahvaz University | Salageghe, A., Scientific Member of TehranUniversity | Almaali, N., M.Sc. in Natural Resources Faculty Tehran University | Askari, H. M., Scientific Member of oceanic and Marine Sciences of Khoramshahr University
One of the most important event in hydrology and flood management is a daily flow predication water different characteristic .There are several methods for estimating amount river discharge. One of these methods is nonlinear mathematic model of neural network. Artificial neural network is a useful technique which allows the user to detect nonlinear complex interaction between output and input data without considering natural phenomena. In addition, due to spiral changes in precipitation, complexity is great. Artificial neural network is a flexible method which helps us to distinguish nonlinear relationship between input and output data. The aim of this research was to investigate application of ANN in prediction of daily discharge. Then, estimated data of this method were compared with estimated data of regression method algorithm was used in ANN was Back Propagation and function was sigmoid results showed that estimation of ANN was accurate that regression method. Determination Coefficient between data estimated with ANN and observed data was 62.94 percent and Residual Mean Square Error (RMSE) and Mean Absolute Error (MAE) of estimated data of this method were 11.88 , 3.7 respectively. Thus, the neural network model is recommended because its structure is simple, the speed of presses is high and the required data are available.
Afficher plus [+] Moins [-]Mots clés AGROVOC
Informations bibliographiques
Cette notice bibliographique a été fournie par Agricultural Research and Education Organization
Découvrez la collection de ce fournisseur de données dans AGRIS