EVALUATION OF THE STORAGE FUNCTION MODEL PARAMETERS USING FUZZY LINEAR REGRESSION APPROACH

摘要

The storage function model is a nonlinear rainfall runoff model that has been developed and applied in Japan. This paper studies the model applicability at some catchments outside Japan. The model parameters for each flood event are estimated by using a series of techniques i.e. wavelet transform analysis, numerical filter separation technique and fuzzy linear regression. The wavelet transform analysis is applied to estimate a recession constant of flood hydrograph whereas the numerical separation technique is chosen to separate the flood hydrograph into baseflow and direct flow based on the resulted recession constant. By using the resulted direct flow, the fuzzy linear regression is then used to estimate the model parameters. The study clearly shown that the model parameters could be well estimated by using the techniques. Moreover, the storage parameter K and the delay time parameter T are found to be a function of catchment area.