Artificial neural-fuzzy inference networks as flood forecasting models
Journal
the World Water and Environmental Resources Congress 2001
Journal Volume
111
ISBN
0784405697; 9780784405697
Date Issued
2004
Author(s)
Abstract
The present paper promotes a novel structure and reasoning processes of flood forecasting models using an artificial neural-fuzzy inference network (ANFIN). ANFIN has a hybrid learning scheme. The unsupervised learning scheme employs fuzzy min-max clustering to extract the information from the input data. The supervised learning scheme uses linear regression method to determine the weights of ANFIN. Compared with the conventional rainfall-runoff models, ANFIN, which learns from the examples, is a model free estimator. Most of the parameters, weights of the network, will be adjusted automatically during the network training. The one-hour-ahead floods of the Chingshui River during tropical storms are forecasted by the constructed models. Our results show that the simple but reliable model is capable of forecasting floods from rainfall by ANFIN.
SDGs
Type
conference paper