Dynamic modeling of chaotic systems using neural networks
A. Erfanian Omidvar, Reza Hashemi, Carrie L. Lucas, Kambiz Badie · Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society · 1992
In this paper we investigate the dynamic modeling of chaotic systems by using neural networks. It is possible for a neural network to approximate a continous fuction f(x1,...,xn) enabling us to construct a static chaotic system with precision e >; 0. It is shown that the dynamic model of a chaotic system can also be costructed with a precision e >; 0 as well as a limited prediction cabability, means the long-term prediction of system evolution from known initial condition is limited. This limitation depends on the precision of our dynamic model and also the degree of sensitivity of chaotic system behavior toward the initial conditions.