20200602_Predicting of daily Khazir basin flow using SWAT and hybrid SWAT-ANN

 

1. Title, Journal, and Authors

Title: Predicting of daily Khazir basin flow using SWAT and hybrid SWAT-ANN

Journal: Ain Shams Engineering Journal

Authors: Abdulwahd A. Kassema,*, Adil M. Raheemb, Khalid M. Khidirc, Mohammad Alkattand

aDams and Water Resources Engineering Department, College of Engineering Salahaddin University Erbil

bSurveying Engineering Dept., College of Engineering, AI-Kitab University

cWater Resources Engineering Dept., Faculty of the Engineering University of Dohuk.

dDams and Water Resources Engineering Dept, College of Engineering University of Mosul

 

2. Summary

In this paper, SWAT was coupled with an ANN model to predict daily streamflow in the basin. Using daily streamflow data and meteorological data, the SWAT model was conducted. Subtraction values between the SWAT model’s results and observed data are used to the input data of the ANN model. Then, the outputs were added to the SWAT model’s results.

 

As a result, it is proved that the prediction accuracy of the hybrid SWAT-ANN model is better than a single SWAT model. Therefore, this approach would be useful to predict other water fields such as future chlorophyll-a concentration, water level fluctuation, and precipitation. Also, other hydrological models and artificial intelligence model would be adapted to this method.

 

3. Application to research

Like this paper, we can also use a hybrid hydrological and artificial intelligence model to predict water fields. This approach could solve the disadvantages of each model.

 

4. Contact

Dae Seong Jeong / Integrated Ph.D. program

 

Environmental Systems Engineering Lab.

School of Environmental Science & Engineering

Gwangju Institute of Science and Technology

1 Oryong-dong Buk-gu Gwangju, 500-712, Korea

 

Phone : +82-10-2003-7860

E-mail : jeongds92@gist.ac.kr