Mousavi, Farzaneh (2015) Evaluation of Empirical and Artificial Intelligence Methods for Estimating Daily Evaporation in Dehloran Synoptic Station. Masters thesis, University of Zabol.
|
Text
Evaluation of Empirical and Artificial Intelligence Methods for Estimating.pdf Download (266kB) | Preview |
Abstract
Nowadays in the world, water resources are the basis for sustainable development of the environment. By considering the limitation of water resources in arid and semiarid regions, the prevention of water loss through evaporation as one of the principal components of hydrological cycle plays an important role in the development and management of water resource evaporation is a complex and nonlinear phenomenon which depends on several interacting climate logical factors. Multi-Layer Perception Artificial Neural Network (MLPANN) and Adaptive Neuro – Fuzzy Inference System (ANFIS) are from those new methods that have been developed for evaluate and forecasting parameters by inherent relation between data. In this study daily evaporation values of Dehloran synoptic station was estimated using empirical relations (Penman and Stephens-Stewart), MLP-ANN and ANFIS. For this purpose the data of Dehloran synoptic station in 20 years (1994 September - 2014 September) was used. The best combination of model input was chosen by using Gamma test including the daily data of mean air temperature, mean relative humidity, air pressure, and mean wind speed and sunshine hours. The four statistical performance evaluation criteria of the coefficient of determination (R2), root means squared error (RMSE), mean bias error (MBE) and efficiency factor (EF) were employed to evaluate the performances of various model developed. The obtained results indicate the best performance of ANFIS by employing two Gaussian membership function for Dehloran station with R2=0.854, RMSE=3.347, MBE=- 0.225 and EF=0.839. The obtained results indicate the best performance of MLP-ANN with R2=0.785, RMSE=3.379, MBE=0.112 and EF=0.785. The obtained results indicate the best performance of Stephens-Stewart empirical relation with R2=0.737, RMSE=3.77, MBE=-0.106 and EF=0.736. The obtained results indicate the best performance of Penman empirical relation with R2=0.675, RMSE=6.46, MBE=4.602 and EF=0.226. Comparison of the estimated values and measured data show that the ANFIS model has a better performance for estimation of daily pan evaporation than the empirical relations and MLP-ANN Model.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Evaporation, ANN, ANFIS, Gamma Test, Dehloran Station |
| Subjects: | G Geography. Anthropology. Recreation > GE Environmental Sciences |
| Depositing User: | admin admin1 admin2 |
| Date Deposited: | 05 Jun 2016 04:46 |
| Last Modified: | 05 Jun 2016 04:46 |
| URI: | http://eprints.uoz.ac.ir/id/eprint/709 |
Actions (login required)
 |
View Item |
