Performance Comparison of ANN and ANFIS Techniques to Rainfall Prediction and Using MIKE11/NAM Hydrological Model to Runoff Simulation

Kakaei Lafdai, Elham (2011) Performance Comparison of ANN and ANFIS Techniques to Rainfall Prediction and Using MIKE11/NAM Hydrological Model to Runoff Simulation. Masters thesis, University of Zabol.

[img]
Preview
Text
Performance Comparison of ANN and.pdf

Download (58kB) | Preview

Abstract

Rainfall-runoff modeling as one of the most important hydrological processes has key issues in flood forecasting and water resources studies. In addition, for prevention and control the flood damages, management and early warning of flood, the rainfall prediction is necessary. The variety of models that have been developed to simulate these processes can be classified into physically based models, conceptual models and black-box models. In this study, rainfall of the Skandari and Qalehshahrokh sub basins located in upper of Zayandehrood dam watershed was predicted using Artificial Neural Fuzzy Inference System (ANFIS) and Artificial Neural Networks (ANNs) models by using historical rainfall data for a 10-year period (1999-2009). Also, the best input combination to rainfall prediction identify by genetic algorithm. Then, runoff discharge produced by the predicted rainfall and the observed rainfall were simulated by a conceptual hydrological model called MIKE11/NAM model and the results were compared together. Nine main parameters of the NAM model were calibrated for Skandari and Qalehshahrokh sub basins. The results indicated that ANFIS model can reasonably predict rainfall. The NAM model simulates the base flows more accurately in both sub basins. The Efficiency Index (EI) during this study in Skandari sub basin is 0.7 for simulating runoff based on the observed rainfall and 0.58 for simulating runoff based on the predicted rainfall. Also the Efficiency Index of this model in Qalehshahrokh sub basin for runoff simulation based on observed and predicted rainfall is 0.86 and 0.84 respectively.

Item Type: Thesis (Masters)
Uncontrolled Keywords: ANFIS, ANN, NAM Rainfall-Runoff Model, Rainfall prediction, Zayandehrood dam watershed
Subjects: S Agriculture > S Agriculture (General)
Depositing User: admin admin1 admin2
Date Deposited: 01 Nov 2016 07:48
Last Modified: 01 Nov 2016 07:48
URI: http://eprints.uoz.ac.ir/id/eprint/956

Actions (login required)

View Item View Item