Please use this identifier to cite or link to this item:
|Title:||An adjoined multi-model approach for monitoring batch and transient operations|
|Authors:||Ng, Y.S. |
|Citation:||Ng, Y.S., Srinivasan, R. (2009-04-21). An adjoined multi-model approach for monitoring batch and transient operations. Computers and Chemical Engineering 33 (4) : 887-902. ScholarBank@NUS Repository. https://doi.org/10.1016/j.compchemeng.2008.11.014|
|Abstract:||Most process monitoring techniques are suitable for steady-state operation but inadequate for these multiphase transient operations with complex dynamics. Specifically, statistical approaches do not function adequately since the basic assumptions that the statistics are developed upon - normal distribution, stationarity - are violated. Consequently, they become prone to false positives and false negatives. Multi-model approaches overcome this by using several local models; however these perform inadequately in the interregnum between models. In this paper, we propose a method, called adjoined principal component analysis that overcomes this. The key characteristic of AdPCA is that the different models are not disjoint; rather they overlap at the edges of their regime and thus ensure smooth evolution of the monitoring. A fuzzy c-means algorithm is used to identify suitable regimes for the constituent models. The applications of the proposed methodology to a distillation unit startup and a fed-batch penicillin cultivation process illustrate the method's efficacy. © 2008 Elsevier Ltd. All rights reserved.|
|Source Title:||Computers and Chemical Engineering|
|Appears in Collections:||Staff Publications|
Show full item record
Files in This Item:
There are no files associated with this item.
checked on Jun 16, 2018
WEB OF SCIENCETM
checked on May 30, 2018
checked on May 5, 2018
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.