Daily Discharge Forecasting using artificial neural network and Support Vector Machine model
Mahdi Moharrampour1, Mohammad Kherad Ranjbar2, Abdulhamid Mehrabi3
1 Department of civil engineering, Islamic Azad University Buinzahra Branch, Buinzahra,Iran
2 Sama Technical and Vocational Training College Islamic Azad University, Karaj Branch, Alborz, Iran
3 Department of civil engineering, Islamic Azad UniversityBuinzahra Branch, Buinzahra, Iran
Abstract: Successful management of water resources requires directional, comprehensive and systematic approaches in order to remove consumers need considering accelerated process of water-related problems and increased demands. In this regard, utilization of modern methods of water resources modeling is so important. Efficiency of statistical learning models in many issues related to management such as water resource modeling or controlling has been proved. On the other hand, advances in the information processing methods have increased data-driven methods in comparison with behavior-driven (physical) methods. For modeling, these methods use minimum information from physical processes and more based on data to describe the characteristics of input and output variables. Therefore, in cases limited information of effective processes are available in the system and physical models results are not very satisfied, data-driven models can be used. This paper compares two expert models in daily flow forecasting. The artificial neural networks (ANN) and Support Vector Machine (SVM) model, are used to forecast daily river flow in north of Iran and the results of these models are compared with Observed daily values .Daily river flow data On Ghara-soo river in north of Iran are used in This study. The RMSE values of ANN model in testing Step was 0.02423 and SVM model in testing step was 030401/0. The comparison results show that the ANN model have better performances in forecasting of river flow from SVM.
[Mahdi Moharrampour, Mohammad Kherad Ranjbar, Abdulhamid Mehrabi. Daily Discharge Forecasting using artificial neural network and Support Vector Machine model. Life Sci J 2013;10(1):914-919]. (ISSN: 1097-8135). http://www.lifesciencesite.com. 2
Keywords: Water Resource Management, Flow Forecasting, Support Vector Machine (SVM), Ghara-soo River,artificial neural network (ANN)