Monthly runoff estimation in the Orontes basin using artificial intelligence models

• Main Authors: Slieman, Alaa Ali, Kozlov, Dmitry V.
• Format: Conference Proceeding
• Language: English
• Subjects: Artificial intelligence; Artificial neural networks; Estimation; Fuzzy logic; Hydrology; Inference; Mean square errors; Neural networks; Regression coefficients; Root-mean-square errors; Runoff; Water balance
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0143527

The presence of complete hydrological time series without interruptions in any hydrological Basin is essential for conducting all hydrological studies and water balance studies in the studied area and runoff is one of the most important elements for conducting these studies. The objective of this study was to using Artificial Intelligence Models for estimation of the runoff data at Al-Jawadiyah station and Al-Amiri station in the Orontes basin in Syria, and to comparison between the results.A large number of artificial intelligence models were built, including artificial neural networks and fuzzy inference models, with all possible parameters changed to get the most accurate results. The results were compared using regression coefficients R and root mean square errors RMSE. The results gave high values for regression coefficients and low for root mean square errors. Overall, the results showed the high ability of artificial neural network models and fuzzy inference models to predict runoff in the studied area. This study recommends expanding the study of the possibilities of using artificial intelligence models in estimating and predicting hydrological factors and components of water balance.