Linear MPC based on data-driven Artificial Neural Networks for large-scale nonlinear distributed parameter systems

Weiguo Xie, Ioannis Bonis, Constantinos Theodoropoulos

Research output: Chapter in Book/Report/Conference proceedingChapter

5 Scopus citations

Abstract

Process controller synthesis with detailed models is a challenging task, which may lead to many advantageous closed-loop features. Model reduction such as Proper Orthogonal Decomposition (POD) and (adaptive) linearization can be applied to tackle with the arising problems, whereas process data can be directly used to build accurate models via training of artificial neural networks (ANN). In this contribution, we present two methodologies we have recently developed, which combine ANN with POD, for use in the context of MPC: the process at hand is represented as a sum of products of time- varying coefficients (computed with ANN) with the POD basis functions computed from plant " snapshots" The resulting accurate model can be used in NMPC, or trajectory piecewise linearization along a reference path can be applied on the ANN, yielding a series of linear models, suitable for linear MPC.

Original languageEnglish (US)
Title of host publicationComputer Aided Chemical Engineering
PublisherElsevier B.V.
Pages1212-1216
Number of pages5
DOIs
StatePublished - 2012
Externally publishedYes

Publication series

NameComputer Aided Chemical Engineering
Volume30
ISSN (Print)1570-7946

Keywords

  • Model predictive control
  • Model reduction for (non-)linear predictive control
  • Neural network training
  • Proper orthogonal decomposition
  • Reduced order NMPC

Fingerprint Dive into the research topics of 'Linear MPC based on data-driven Artificial Neural Networks for large-scale nonlinear distributed parameter systems'. Together they form a unique fingerprint.

Cite this