In this paper, a new type of neural network called recurrent fuzzy neural n
etwork (RFNN) is proposed to model the fuzzy dynamical systems (FDS). FDS i
s considered as an order system. The network developed in this paper is bas
ed on recurrent neural networks (RNN) to capture the dynamical properties o
f FDS. The training algorithm is derived based on the tool of order derivat
ive. An example is given to demonstrate the validity of the approach. (C) 2
001 Elsevier Science B.V. All rights reserved.