Browse by author
Lookup NU author(s): Nicola Fletcher,
Emeritus Professor Julian Morris,
Professor Elaine Martin
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
The first stage in the development of a performance monitoring tool is a representative model of the process. Traditionally batch processes are associated with non-linear, dynamic behaviour. One method of developing a non-linear model is through a local modelling approach. These local models can be combined to provide a global non-linear representation of the process. The paper develops and compares local models formed using linear Partial Least Squares (PLS), dynamic linear PLS and ARX structures with global models based on dynamic linear and non-linear PLS, as well as non-linear PLS using a simulation of a fed-batch fermentation process. Finally the use of a local linear dynamic PLS model for process performance monitoring representation was evaluated. © 2002 Elsevier B.V. All rights reserved.
Author(s): Fletcher N, Morris A, Martin E
Editor(s): Johan Grievink and Jan van Schijndel
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Published
Conference Name: 12th European Symposium on Computer Aided Process Engineering (ESCAPE-12)
Year of Conference: 2002
Publisher: Computer Aided Chemical Engineering: Elsevier BV
Library holdings: Search Newcastle University Library for this item