IDENTIFICATION OF HAMMERSTEIN OUTPUT-ERROR SYSTEMS WITH TWO-SEGMENT NONLINEARITIES: ALGORITHM AND APPLICATIONS1

Bo Yu, Huazhen Fang, Yang Lin, and Yang Shi

Keywords

System identification, Hammerstein systems with twosegment nonlinearities, stochastic gradient algorithm, distillation columns

Abstract

This paper studies the parameter identification problem of Hammerstein output-error systems with two-segment nonlinearities. New switching sequences are proposed to reduce the number of parameters to be estimated. A stochastic gradient identification algorithm together with the convergence analysis is further developed. A varying forgetting factor scheme is used to make a tradeoff between the convergence rate and estimation accuracy. Both numerical simulation results and applications to distillation columns are provided to show the effectiveness of the proposed method.

Important Links:



Go Back