Modeling nonlinear dynamic using multilayer neural networks

Certain deterministic nonlinear systems may show chaotic behaviour. Time series derives from such systems seem stochastic when analysed with linear technique. However, uncovering the deterministic structure is important because it allows constructing more realistic and better models and thus improved predictive capabilities. This paper provides a new approach for features of chaotic systems definition. Proposed method provides the calculation of Lyapunov exponents using multilayer neural networks trained by modified backpropagation error (BPE) algorithm. In this paper we compare the proposed technique and one of the widespread method for largest Lyapunov exponent definition for Henon and Lorenz chaotic processes.