한국수자원학회:학술대회논문집 (Proceedings of the Korea Water Resources Association Conference)
- 한국수자원학회 2012년도 학술발표회
- /
- Pages.336-339
- /
- 2012
Application of Soft Computing Model for Hydrologic Forecasting
- Kim, Sung-Won (Department of Railroad and Civil Engineering, Dongyang University) ;
- Park, Ki-Bum (Department of Construction Information, Andong Science College)
- 발행 : 2012.05.16
초록
Accurate forecasting of pan evaporation (PE) is very important for monitoring, survey, and management of water resources. The purpose of this study is to develop and apply Kohonen self-organizing feature maps neural networks model (KSOFM-NNM) to forecast the daily PE for the dry climate region in south western Iran. KSOFM-NNM for Ahwaz station was used to forecast daily PE on the basis of temperature-based, radiation-based, and sunshine duration-based input combinations. The measurements at Ahwaz station in south western Iran, for the period of January 2002 - December 2008, were used for training, cross-validation and testing data of KSOFM-NNM. The results obtained by TEM 1 produced the best results among other combinations for Ahwaz station. Based on the comparisons, it was found that KSOFM-NNM can be employed successfully for forecasting the daily PE from the limited climatic data in south western Iran.