Evapotranspiration is an important component in planning and management of water resources. It depends on climatic factors and the influence of these factors on each other makes evapotranspiration estimation difficult. This study attempts to explore the possibility of predicting this important component using three different heuristic methods, Support Vector Machine (SVM), Adaptive Neuro-Fuzzy Inference System (ANFIS) and Gene Expression Programming (GEP). In this regard, according to the FAO Penman-Monteith equation, the monthly potential evapotranspiration in four synoptic stations- Zahedan, Zabol, Iranshahr, and Chabahar- was calculated using the monthly weather data. The weather data were then used as inputs to the SVM, ANFIS and GEP models to estimate potential evapotranspiration. Five different input combinations were tried in the applications. The results of SVM, ANFIS and GEP models were compared based on the coefficient of determination (R2), Mean Absolute Error and root mean square error. Findings showed that the SVM whose inputs are average air temperature, relative humidity, wind speed, and sunny hours of the current and one previous month performed superior to the other models for the Zahedan, Zabol, Iranshahr, and Chabahar stations. Comparison of three heuristic methods indicated that in all stations, SVM, GEP and ANFIS models were take the first, second, and third place in estimation of the monthly potential evapotranspiration, respectively.
5 August 2017 in Neural Computing and Applications
A forecast of runoff hydrograph leads to effective decision making in flood management. In this study, effect of three base flow separation methods- straight line (DRH1), fixed base (DRH2), and variable slope (DRH3) and two methods of excess rainfall estimation (Ø index and SCS (Soil Conservation Service) method) with six different rainfall-runoff events are studied on the flood hydrograph simulation. The 6 rainfall-runoff events recorded at Jong catchment in Iran were used for this study. The Percentage Error in Volume (PEV), Percentage Error in Peak (PEP), Percentage Error in Time to Peak (PETP), Nash-Sutcliffe Coefficient (NSC), R2 and Root Mean Square Error (RMSE), were also used to evaluate the results. The results show that the combination of SCS-DRH2 method is the best combination of base flow separation - excess rainfall for improving the accuracy of the model in predicting the hydrograph shape. Also, in estimating the flood volume, the SCS-DRH3 and SCS-DRH1 are the best combination in estimating peak flood and time to peak the flood respectively. The analysis of the results suggests that the combination of SCS-DRH2 is the best combination in improving the accuracy of Gamma model.
4 March 2024 in Iranian Journal of Science and Technology, Transactions of Civil Engineering