Filter-based Constraints to Easier Plant Feasibility in Modifier Adaptation Schemes

Aris Papasavvas, Gregory Francois

Research output: Chapter in Book/Report/Conference proceedingChapter


Output modifier adaptation (MAy) is an iterative model-based real-time optimization (RTO) method with the proven ability to reach, upon converge, the unknown plant optimal steady-state operating conditions despite structural and parametric plant-model mismatch and disturbances. However, as such, feasibility of the iterates cannot be guaranteed before convergence is achieved. This issue can be mitigated with additional modelling and/or experimental efforts, which is costly and sometimes cannot be envisaged, especially for large-scale systems. In this article, MAy is compared to one of its most recent extensions, namely “KMAy”, which is shown to have more favorable sufficient conditions for fea- sibility of the iterates. A simulation study, performed on the Williams-Otto reactor – a standard benchmark case study for RTO algorithms – illustrates that KMAy is safer for practical applications.
Original languageEnglish
Title of host publication29th European Symposium on Computer Aided Process Engineering
ISBN (Electronic)978-0-12-819939-8, 978-0-12-818634-3
Publication statusPublished - 25 Jul 2019
EventEuropean Symposium on Computer-Aided Process Engineering: From molecule to entreprise - Eindhoven, Netherlands
Duration: 16 Jun 201919 Jun 2019

Publication series

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


ConferenceEuropean Symposium on Computer-Aided Process Engineering
Abbreviated titleESCAPE 29
Internet address


  • Real-Time Optimization
  • Modifier Adaptation
  • Feasibility


Dive into the research topics of 'Filter-based Constraints to Easier Plant Feasibility in Modifier Adaptation Schemes'. Together they form a unique fingerprint.

Cite this