Nile river flow forecasting based Takagi-Sugeno fuzzy model

Abstract

In this study, a fuzzy model for forecasting the Nile river flow is developed. The fuzzy model is represented by a set of rules based on the Takagi-Sugeno type. The Gustanfson-Kessel (GK) algorithm was applied to determine the antecedent membership functions and least-square estimation was used to determine the consequence parameters. The performance of the fuzzy model was tested using a set of measurements recorded at Dongola station in Egypt. The readings span over the period from 1975 to 1993. These measurements were split in to two groups one for training and another for testing. The performance of the developed proposed fuzzy model was checked in both training and testing cases. The developed fuzzy logic model showed a better modeling capability compared to traditional modeling. © 2010 Asian Network for Scientific Information.