Abstract
The qualitative analysis of comp lex process systems is an im portant task at the design of control and process monitoring algorithm s. Qualitative models require interpretable description of the operating regimes of the process. This work shows a novel approach to discover and isolate operating regimes of process systems based on process models, time series analysis, and decision tree induction technique. The novelty of this approach is the application of time series segmentat ion algorithms to detect the homogeneous periods of the operation. Advanced sequence alignment algorithm used in bioinformatics is applied for the calculation of the similarity of the process trends described by qualitative variables. Decision tree induction is applied for the transformation of this hidden knowledge into easily in terpretable rule base to represent the operation regions of the process. The whole methodology is applied to detect operating regimes of an industrial fixed bed tube reactor.
Original language | English |
---|---|
Title of host publication | Computer Aided Chemical Engineering |
Pages | 1039-1044 |
Number of pages | 6 |
Volume | 25 |
DOIs | |
Publication status | Published - 2008 |
Publication series
Name | Computer Aided Chemical Engineering |
---|---|
Volume | 25 |
ISSN (Print) | 15707946 |
Fingerprint
Keywords
- decision tree
- operating regime
- qualitative analysis
- sequence alignment
ASJC Scopus subject areas
- Chemical Engineering(all)
- Computer Science Applications
Cite this
Decision tree based qualitative analysis of operating regimes in industrial production processes. / Varga, Tamás; Szeifert, F.; Réti, József; Abonyi, J.
Computer Aided Chemical Engineering. Vol. 25 2008. p. 1039-1044 (Computer Aided Chemical Engineering; Vol. 25).Research output: Chapter in Book/Report/Conference proceeding › Chapter
}
TY - CHAP
T1 - Decision tree based qualitative analysis of operating regimes in industrial production processes
AU - Varga, Tamás
AU - Szeifert, F.
AU - Réti, József
AU - Abonyi, J.
PY - 2008
Y1 - 2008
N2 - The qualitative analysis of comp lex process systems is an im portant task at the design of control and process monitoring algorithm s. Qualitative models require interpretable description of the operating regimes of the process. This work shows a novel approach to discover and isolate operating regimes of process systems based on process models, time series analysis, and decision tree induction technique. The novelty of this approach is the application of time series segmentat ion algorithms to detect the homogeneous periods of the operation. Advanced sequence alignment algorithm used in bioinformatics is applied for the calculation of the similarity of the process trends described by qualitative variables. Decision tree induction is applied for the transformation of this hidden knowledge into easily in terpretable rule base to represent the operation regions of the process. The whole methodology is applied to detect operating regimes of an industrial fixed bed tube reactor.
AB - The qualitative analysis of comp lex process systems is an im portant task at the design of control and process monitoring algorithm s. Qualitative models require interpretable description of the operating regimes of the process. This work shows a novel approach to discover and isolate operating regimes of process systems based on process models, time series analysis, and decision tree induction technique. The novelty of this approach is the application of time series segmentat ion algorithms to detect the homogeneous periods of the operation. Advanced sequence alignment algorithm used in bioinformatics is applied for the calculation of the similarity of the process trends described by qualitative variables. Decision tree induction is applied for the transformation of this hidden knowledge into easily in terpretable rule base to represent the operation regions of the process. The whole methodology is applied to detect operating regimes of an industrial fixed bed tube reactor.
KW - decision tree
KW - operating regime
KW - qualitative analysis
KW - sequence alignment
UR - http://www.scopus.com/inward/record.url?scp=52749098246&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=52749098246&partnerID=8YFLogxK
U2 - 10.1016/S1570-7946(08)80179-4
DO - 10.1016/S1570-7946(08)80179-4
M3 - Chapter
AN - SCOPUS:52749098246
SN - 9780444532275
VL - 25
T3 - Computer Aided Chemical Engineering
SP - 1039
EP - 1044
BT - Computer Aided Chemical Engineering
ER -