Qualitative Representation of Trends (QRT): Extended method for identification of consecutive inflection points

Kris Villez; Christian Rosén; François Anctil; Carl Duchesne; Peter A. Vanrolleghem

doi:10.1016/j.compchemeng.2012.08.010

Qualitative Representation of Trends (QRT): Extended method for identification of consecutive inflection points

Kris Villez, Christian Rosén, François Anctil, Carl Duchesne, Peter A. Vanrolleghem

Research output: Contribution to journal › Article › peer-review

27 Scopus citations

Abstract

In this paper a methodology for extraction of qualitative descriptions of data series is extended for proper inflection point recognition. The original method allows the extraction of qualitative information such as the sign of the first and second derivatives of the assumed signals underlying univariate noisy time series. However, it is not able to handle consecutive inflection points, i.e. inflection points which follow each other in time with no minimum or maximum in between them. To improve this method for proper assessment of inflection points, the original method is compared with three modifications of the original method. One of the alternatives based on repeated application of Witkin's stability criterion delivers better results for both identification of the qualitative descriptions and the locations in time of extrema and inflection points. Furthermore, the same modified method is shown to deliver the best fault diagnosis performance in a benchmark batch fermentation study. Furthermore, the same modified method is shown to deliver the best fault diagnosis performance in a benchmark batch fermentation study.

Original language	English
Pages (from-to)	187-199
Number of pages	13
Journal	Computers and Chemical Engineering
Volume	48
DOIs	https://doi.org/10.1016/j.compchemeng.2012.08.010
State	Published - Jan 2013
Externally published	Yes

Funding

This work was originally supported by the Institute for Encouragement of Innovation by means of Science and Technology in Flanders (IWT) and has benefited from additional support by the ICIS Distinctive Signature at Idaho National Laboratory (INL). Peter Vanrolleghem holds the Canada Research Chair in Water Quality Modelling. The first author wishes to thank Prof. V. Venkatasubramanian for the provided opportunity to continue this research.

Funders	Funder number
Institute for Encouragement of Innovation by means of Science and Technology in Flanders
International Cancer Imaging Society
Agentschap voor Innovatie door Wetenschap en Technologie

Keywords

Data mining
Qualitative Representation of Trends (QRT)
Qualitative Trend Analysis (QTA)
Qualitative analysis

Access to Document

10.1016/j.compchemeng.2012.08.010

Cite this

@article{9f3511dae20443e79ebbab876291af2b,

title = "Qualitative Representation of Trends (QRT): Extended method for identification of consecutive inflection points",

abstract = "In this paper a methodology for extraction of qualitative descriptions of data series is extended for proper inflection point recognition. The original method allows the extraction of qualitative information such as the sign of the first and second derivatives of the assumed signals underlying univariate noisy time series. However, it is not able to handle consecutive inflection points, i.e. inflection points which follow each other in time with no minimum or maximum in between them. To improve this method for proper assessment of inflection points, the original method is compared with three modifications of the original method. One of the alternatives based on repeated application of Witkin's stability criterion delivers better results for both identification of the qualitative descriptions and the locations in time of extrema and inflection points. Furthermore, the same modified method is shown to deliver the best fault diagnosis performance in a benchmark batch fermentation study. Furthermore, the same modified method is shown to deliver the best fault diagnosis performance in a benchmark batch fermentation study.",

keywords = "Data mining, Qualitative Representation of Trends (QRT), Qualitative Trend Analysis (QTA), Qualitative analysis",

author = "Kris Villez and Christian Ros{\'e}n and Fran{\c c}ois Anctil and Carl Duchesne and Vanrolleghem, {Peter A.}",

year = "2013",

month = jan,

doi = "10.1016/j.compchemeng.2012.08.010",

language = "English",

volume = "48",

pages = "187--199",

journal = "Computers and Chemical Engineering",

issn = "0098-1354",

publisher = "Elsevier",

}

TY - JOUR

T1 - Qualitative Representation of Trends (QRT)

T2 - Extended method for identification of consecutive inflection points

AU - Villez, Kris

AU - Rosén, Christian

AU - Anctil, François

AU - Duchesne, Carl

AU - Vanrolleghem, Peter A.

PY - 2013/1

Y1 - 2013/1

N2 - In this paper a methodology for extraction of qualitative descriptions of data series is extended for proper inflection point recognition. The original method allows the extraction of qualitative information such as the sign of the first and second derivatives of the assumed signals underlying univariate noisy time series. However, it is not able to handle consecutive inflection points, i.e. inflection points which follow each other in time with no minimum or maximum in between them. To improve this method for proper assessment of inflection points, the original method is compared with three modifications of the original method. One of the alternatives based on repeated application of Witkin's stability criterion delivers better results for both identification of the qualitative descriptions and the locations in time of extrema and inflection points. Furthermore, the same modified method is shown to deliver the best fault diagnosis performance in a benchmark batch fermentation study. Furthermore, the same modified method is shown to deliver the best fault diagnosis performance in a benchmark batch fermentation study.

AB - In this paper a methodology for extraction of qualitative descriptions of data series is extended for proper inflection point recognition. The original method allows the extraction of qualitative information such as the sign of the first and second derivatives of the assumed signals underlying univariate noisy time series. However, it is not able to handle consecutive inflection points, i.e. inflection points which follow each other in time with no minimum or maximum in between them. To improve this method for proper assessment of inflection points, the original method is compared with three modifications of the original method. One of the alternatives based on repeated application of Witkin's stability criterion delivers better results for both identification of the qualitative descriptions and the locations in time of extrema and inflection points. Furthermore, the same modified method is shown to deliver the best fault diagnosis performance in a benchmark batch fermentation study. Furthermore, the same modified method is shown to deliver the best fault diagnosis performance in a benchmark batch fermentation study.

KW - Data mining

KW - Qualitative Representation of Trends (QRT)

KW - Qualitative Trend Analysis (QTA)

KW - Qualitative analysis

UR - http://www.scopus.com/inward/record.url?scp=84866995240&partnerID=8YFLogxK

U2 - 10.1016/j.compchemeng.2012.08.010

DO - 10.1016/j.compchemeng.2012.08.010

M3 - Article

AN - SCOPUS:84866995240

SN - 0098-1354

VL - 48

SP - 187

EP - 199

JO - Computers and Chemical Engineering

JF - Computers and Chemical Engineering

ER -

Qualitative Representation of Trends (QRT): Extended method for identification of consecutive inflection points

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this