Flood water level modeling and prediction using radial basis function neural network: Case study kedah

Abstract

Natural disasters are common nowadays and a major adverse event resulting from natural process of Earth. Most of the natural disaster are beyond control of human beings and cannot be predicted accurately when it occurs. For instance, prediction of a river water level is essential for flood mitigation in order to save people’s lives and property. However, it is very difficult to predict river water level accurately since it is influenced by many factors and the fluctuations are highly non-linear. To address this problem, a river water level predictor utilizing the Radial Basis Function Network (RBFN) is proposed in this study. The goal of this project is to design a neural prediction algorithm that can forecast river water level prediction 7 h ahead with lower error. Result shows Best Fit value of 82.43% and Root Mean Square Error (RMSE) of 1.571.