Support vector regression for Hammerstein-Wiener model identification

Abstract

Support vector regression is applied to identify nonlinear systems represented by Hammerstein-Wiener models, proceeded by a static nonlinear block with linear dynamic system and followed by a static nonlinear blocks. The linear block is expanded in terms of basis functions, and the static nonlinear block is determined using support vector machine regression. The fitness of the system is to be compared with conventional identification methods and improved the efficiency of the model which is present in the paper making process.