Disturbance compensation for servo-control applications using a discrete adaptive neural network feedforward method

G Herrmann, FL Lewis, SS Ge, J Zhang

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

5 Citations (Scopus)

Abstract

This paper introduces a novel adaptive neural network compensator for feedforward compensation of external disturbances affecting a closed loop system. The neural network scheme is posed so that the nonlinear disturbance model for a measurable disturbance can be adapted for rejection of the disturbance affecting a closed loop system. The non-linear neural network approach has been particularly developed for mobile applications where the adaptation algorithm has to remain simple. For that reason, the theoretical framework justifies a very simple least-mean-square approach suggested in a mobile hard disk drive context. This approach is generalized to a non-linear adaptive neural network compensation scheme. In addition, usual assumptions are relaxed, so that it is sufficient to model the nonlinear disturbance model as a stable system avoiding strictly positive real assumptions. The output of the estimated disturbance model is assumed to be matched to the compensation signal for effectiveness, although for stability this is not necessary. Simulation examples show different features of the adaptation algorithm also considering a realistic hard disk drive simulation.
Translated title of the contributionDisturbance compensation for servo-control applications using a discrete adaptive neural network feedforward method
Original languageEnglish
Title of host publication46th IEEE Conference on Decision and Control, New Orleans, USA
Pages5965 - 5972
Number of pages8
DOIs
Publication statusPublished - Dec 2007

Bibliographical note

Conference Organiser: IEEE

Fingerprint Dive into the research topics of 'Disturbance compensation for servo-control applications using a discrete adaptive neural network feedforward method'. Together they form a unique fingerprint.

Cite this