Using artificial neural network approach for modelling rainfall–runoff due to typhoon

Abstract

In Taiwan, owing to the nonuniform temporal and spatial distribution of rainfall and high mountains all over the country, hydrologic systems are very complex. Therefore, preventing and controlling flood disasters is imperative. Nevertheless, water level and flow records are essential in hydrological analysis for designing related water works of flood management. Due to the complexity of the hydrological process, reliable runoff is hardly predicted by applying linear and non-linear regression methods. Therefore, in this study, a model for estimating runoff by using rainfall data from a river basin is developed and a neural network technique is employed to recover missing data. For achieving the objectives, hourly rainfall and flow data from Nanhe, Taiwu, and Laii rainfall stations and Sinpi flow station in the Linbien basin are used. The data records were of 27 typhoons between the years 2005 and 2009. The feed forward back propagation network (FFBP) and conventional regression analysis (CRA) were employed to study their performances. From the statistical evaluation, it has been found that the performance of FFBP exceeded that of regression analysis as reflected by the determination coefficients R 2, which were 0.969 and 0.284 for FFBP and CRA, respectively.