IDENTIFICATION OF NONLINEAR DYNAMIC SYSTEMS WITH ASYMMETRIC INPUT NONLINEARITIES

Jozef Vörös

   The paper deals with the identification of nonlinear dynamic systems having strongly asymmetric input nonlinearities. A special form of Hammerstein model with two-segment polynomials enables an iterative estimation of the model parameters on the basis of input/output records. The accuracy of resulting model is better, compared to the single polynomial models of higher degrees. The results of Monte Carlo analysis are also presented for the noisy systems.

Keywords: nonlinear systems, polynomials, Hammerstein model, identification

[full-paper]