Tampere University of Technology

TUTCRIS Research Portal

Fast neural predictive control algorithm

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Details

Original languageEnglish
Title of host publicationEuropean Control Conference, ECC 1999 - Conference Proceedings
PublisherThe Institute of Electrical and Electronics Engineers, Inc.
Pages4269-4273
Number of pages5
ISBN (Print)9783952417355
Publication statusPublished - 24 Mar 2015
Publication typeA4 Article in a conference publication
EventEuropean Control Conference -
Duration: 1 Jan 1900 → …

Conference

ConferenceEuropean Control Conference
Period1/01/00 → …

Abstract

This study presents the idea of a fast algorithm that can be used for the synthesis of the control law in a neural predictive control strategy. It is compared, as computation time and regulation performance, with a similar procedure that uses Brent's search combined with the golden section search algorithm. The process considered for simulation is a continuous fermentation. The results show that the same regulation performance can be achieved but calculation time and costs are reduced.

ASJC Scopus subject areas

Keywords

  • bio-process, neural networks, predictive control

Publication forum classification