Identification of Wiener-type nonlinear systems in a noisy environment

Abstract

Wiener systems are widely encountered in engineering and science applications. This paper proposes a straightforward approach to the identification of Wiener systems in a noisy environment. The obtained models are represented in terms of the frequency response for the linear subsystem and the inverse of the static nonlinearity. A variety of input excitation signals, including random binary and periodic, can be used with the proposed method. Simulation studies using a pH neutralization continuous stirred tank reactor show that excellent results are obtained with relatively short data lengths.