Recursive partial least squares algorithms for monitoring complex industrial processes

Xun Wang; Uwe Kruger; Barry Lennox

doi:10.1016/S0967-0661(02)00096-5

Recursive partial least squares algorithms for monitoring complex industrial processes

Xun Wang, Uwe Kruger, Barry Lennox

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

Abstract

The monitoring of processes that exhibit non-stationary and/or time varying behaviour is discussed in this paper. It is shown that the application of recursive partial least squares (RPLS) algorithms together with adaptive confidence limits can lead to a considerable reduction in the number of false alarms. The integration of these algorithms into the multivariate statistical process control (MSPC) framework is introduced and its extensions to multi-block approaches is discussed. Example studies are given with respect to a simulation of a fluid catalytic cracking unit and the analysis of data obtained from an industrial distillation process. © 2003 Elsevier Science Ltd. All rights reserved.

Original language	English
Pages (from-to)	613-632
Number of pages	19
Journal	Control Engineering Practice
Volume	11
Issue number	6
DOIs	https://doi.org/10.1016/S0967-0661(02)00096-5
Publication status	Published - Jun 2003

Keywords

Adaptive algorithms
Data reduction
Fault detection
Process models
Statistical process control

Access to Document

10.1016/S0967-0661(02)00096-5

Financial and environmental benefits through the development and transfer of control and monitoring technology in the process industries
Lennox, B. (Participant), Marjanovic, O. (Participant), (Participant) & (Participant)
Impact: Policy, Economic, Environmental

Cite this

@article{b31ef23dcf514b8c8f0747b18d68ce2f,

title = "Recursive partial least squares algorithms for monitoring complex industrial processes",

abstract = "The monitoring of processes that exhibit non-stationary and/or time varying behaviour is discussed in this paper. It is shown that the application of recursive partial least squares (RPLS) algorithms together with adaptive confidence limits can lead to a considerable reduction in the number of false alarms. The integration of these algorithms into the multivariate statistical process control (MSPC) framework is introduced and its extensions to multi-block approaches is discussed. Example studies are given with respect to a simulation of a fluid catalytic cracking unit and the analysis of data obtained from an industrial distillation process. {\textcopyright} 2003 Elsevier Science Ltd. All rights reserved.",

keywords = "Adaptive algorithms, Data reduction, Fault detection, Process models, Statistical process control",

author = "Xun Wang and Uwe Kruger and Barry Lennox",

note = "Times Cited: 39",

year = "2003",

month = jun,

doi = "10.1016/S0967-0661(02)00096-5",

language = "English",

volume = "11",

pages = "613--632",

journal = "Control Engineering Practice",

publisher = "Elsevier BV",

number = "6",

}

TY - JOUR

T1 - Recursive partial least squares algorithms for monitoring complex industrial processes

AU - Wang, Xun

AU - Kruger, Uwe

AU - Lennox, Barry

N1 - Times Cited: 39

PY - 2003/6

Y1 - 2003/6

N2 - The monitoring of processes that exhibit non-stationary and/or time varying behaviour is discussed in this paper. It is shown that the application of recursive partial least squares (RPLS) algorithms together with adaptive confidence limits can lead to a considerable reduction in the number of false alarms. The integration of these algorithms into the multivariate statistical process control (MSPC) framework is introduced and its extensions to multi-block approaches is discussed. Example studies are given with respect to a simulation of a fluid catalytic cracking unit and the analysis of data obtained from an industrial distillation process. © 2003 Elsevier Science Ltd. All rights reserved.

AB - The monitoring of processes that exhibit non-stationary and/or time varying behaviour is discussed in this paper. It is shown that the application of recursive partial least squares (RPLS) algorithms together with adaptive confidence limits can lead to a considerable reduction in the number of false alarms. The integration of these algorithms into the multivariate statistical process control (MSPC) framework is introduced and its extensions to multi-block approaches is discussed. Example studies are given with respect to a simulation of a fluid catalytic cracking unit and the analysis of data obtained from an industrial distillation process. © 2003 Elsevier Science Ltd. All rights reserved.

KW - Adaptive algorithms

KW - Data reduction

KW - Fault detection

KW - Process models

KW - Statistical process control

U2 - 10.1016/S0967-0661(02)00096-5

DO - 10.1016/S0967-0661(02)00096-5

M3 - Article

VL - 11

SP - 613

EP - 632

JO - Control Engineering Practice

JF - Control Engineering Practice

IS - 6

ER -

Recursive partial least squares algorithms for monitoring complex industrial processes

Abstract

Keywords

Access to Document

Fingerprint

Impacts

Financial and environmental benefits through the development and transfer of control and monitoring technology in the process industries

Cite this