Full Content is available to subscribers

Subscribe/Learn More  >

Neural-Network Based On-Line Adaptation of Model Predictive Controller for Dynamic Systems With Uncertain Behavior

[+] Author Affiliations
Marco E. Sanjuan

Universidad del Norte

Paper No. IMECE2005-82609, pp. 1033-1040; 8 pages
  • ASME 2005 International Mechanical Engineering Congress and Exposition
  • Dynamic Systems and Control, Parts A and B
  • Orlando, Florida, USA, November 5 – 11, 2005
  • Conference Sponsors: Dynamic Systems and Control Division
  • ISBN: 0-7918-4216-9 | eISBN: 0-7918-3769-6
  • Copyright © 2005 by ASME


Model-based controllers have positioned themselves in industrial applications, working mainly on top of a layer of PID controllers. Their implementation takes an important amount of time because of the required PID tuning and the model characterization/identification. This paper presents a strategy to perform on-line adaptation for the dynamic matrix coefficients in a DMC controller. Based on the observed PH (Prediction Horizon) elements of the response and controller signal vectors, and based on a non-residing control horizon controller design, the direct control problem is reformulated using the full-effect dynamic matrix (PHxPH) as an unknown. Data is collected and used in two directions: training a RAWN Network (Random Allocation Weight Neural Network), to describe recently observed process behavior, and to solve a least-squares problem for a set of linear equations where the unknowns are the characteristic response coefficients. The paper presents the effect of both approaches, illustrating the adaptation algorithms operation in a highly nonlinear process where the controller is designed in a low-gain region. Then the process operating condition is shifted so that it moves to a high-gain region to observe controller response.

Copyright © 2005 by ASME



Interactive Graphics


Country-Specific Mortality and Growth Failure in Infancy and Yound Children and Association With Material Stature

Use interactive graphics and maps to view and sort country-specific infant and early dhildhood mortality and growth failure data and their association with maternal

Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In