A fuzzy-neural hybrid system of simulating typhoon waves

Abstract

This study addresses a fuzzy-neural hybrid system of simulating typhoon waves. A membership function based on the fuzzy theory is expressed by a union Gaussian function to illustrate the rapid wave decaying. Four areas separated by two lines which intersect at the Hua-Lien harbor indicate the case of typhoon's position and propagation. Better simulation performance of the peak wave heights and their occurrence time in both the learning stage and the verification stage simulated by the NF2 model than by the NF1 model is identified. The wave decaying due to land effect is well described by the NF2 model. The NF2 model is applicable for well simulating typhoon waves during the whole period of a typhoon approaching to Taiwan. (c) 2006 Elsevier B.V All rights reserved.