Monitoring nonstationary and dynamic trends for practical process fault diagnosis

Yuanling Lin, Uwe Kruger, Fengshou Gu, Andrew Ball, Qian Chen

Research output: Contribution to journalArticle

Abstract

This article introduces a revised common trend framework to monitor nonstationary and dynamic trends in industrial processes and shows needs for each improvement on the basis of three application studies. These improvements relate to (i) the extension of the common trend framework to include sets that contain stationary and nonstationary variables, (ii) handling cases where residuals are not drawn from multivariate normal distributions and (iii) the application of the framework to larger variable sets. Existing work does not adequately address these practically important issues. Industrial application studies highlight the needs for (i) the extended framework to model data sets containing stationary and nonstationary variables, (ii) handling statistics that are not based on normally distributed residuals and (iii) the use of Chigira procedure to robustly extract common trends. The extended framework is compared to traditional approaches.

Original languageEnglish
Pages (from-to)139-158
Number of pages20
JournalControl Engineering Practice
Volume84
Early online date6 Dec 2018
DOIs
Publication statusPublished - 1 Mar 2019

Fingerprint

Fault Diagnosis
Failure analysis
Monitoring
Normal distribution
Industrial applications
Statistics
Stationary Set
Multivariate Normal Distribution
Industrial Application
Data Model
Monitor
Framework
Trends

Cite this

@article{9d788f2b1fbe4fbea8d335ae030631d9,
title = "Monitoring nonstationary and dynamic trends for practical process fault diagnosis",
abstract = "This article introduces a revised common trend framework to monitor nonstationary and dynamic trends in industrial processes and shows needs for each improvement on the basis of three application studies. These improvements relate to (i) the extension of the common trend framework to include sets that contain stationary and nonstationary variables, (ii) handling cases where residuals are not drawn from multivariate normal distributions and (iii) the application of the framework to larger variable sets. Existing work does not adequately address these practically important issues. Industrial application studies highlight the needs for (i) the extended framework to model data sets containing stationary and nonstationary variables, (ii) handling statistics that are not based on normally distributed residuals and (iii) the use of Chigira procedure to robustly extract common trends. The extended framework is compared to traditional approaches.",
keywords = "Cointegration, Common trends models, Forecast recovery, Generalized data structure, Kasa decomposition, Latent stationary and nonstationary factors, Multivariate normal distribution",
author = "Yuanling Lin and Uwe Kruger and Fengshou Gu and Andrew Ball and Qian Chen",
year = "2019",
month = "3",
day = "1",
doi = "10.1016/j.conengprac.2018.11.020",
language = "English",
volume = "84",
pages = "139--158",
journal = "Control Engineering Practice",
issn = "0967-0661",
publisher = "Elsevier Limited",

}

Monitoring nonstationary and dynamic trends for practical process fault diagnosis. / Lin, Yuanling; Kruger, Uwe; Gu, Fengshou; Ball, Andrew; Chen, Qian.

In: Control Engineering Practice, Vol. 84, 01.03.2019, p. 139-158.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Monitoring nonstationary and dynamic trends for practical process fault diagnosis

AU - Lin, Yuanling

AU - Kruger, Uwe

AU - Gu, Fengshou

AU - Ball, Andrew

AU - Chen, Qian

PY - 2019/3/1

Y1 - 2019/3/1

N2 - This article introduces a revised common trend framework to monitor nonstationary and dynamic trends in industrial processes and shows needs for each improvement on the basis of three application studies. These improvements relate to (i) the extension of the common trend framework to include sets that contain stationary and nonstationary variables, (ii) handling cases where residuals are not drawn from multivariate normal distributions and (iii) the application of the framework to larger variable sets. Existing work does not adequately address these practically important issues. Industrial application studies highlight the needs for (i) the extended framework to model data sets containing stationary and nonstationary variables, (ii) handling statistics that are not based on normally distributed residuals and (iii) the use of Chigira procedure to robustly extract common trends. The extended framework is compared to traditional approaches.

AB - This article introduces a revised common trend framework to monitor nonstationary and dynamic trends in industrial processes and shows needs for each improvement on the basis of three application studies. These improvements relate to (i) the extension of the common trend framework to include sets that contain stationary and nonstationary variables, (ii) handling cases where residuals are not drawn from multivariate normal distributions and (iii) the application of the framework to larger variable sets. Existing work does not adequately address these practically important issues. Industrial application studies highlight the needs for (i) the extended framework to model data sets containing stationary and nonstationary variables, (ii) handling statistics that are not based on normally distributed residuals and (iii) the use of Chigira procedure to robustly extract common trends. The extended framework is compared to traditional approaches.

KW - Cointegration

KW - Common trends models

KW - Forecast recovery

KW - Generalized data structure

KW - Kasa decomposition

KW - Latent stationary and nonstationary factors

KW - Multivariate normal distribution

UR - http://www.scopus.com/inward/record.url?scp=85057617414&partnerID=8YFLogxK

U2 - 10.1016/j.conengprac.2018.11.020

DO - 10.1016/j.conengprac.2018.11.020

M3 - Article

VL - 84

SP - 139

EP - 158

JO - Control Engineering Practice

JF - Control Engineering Practice

SN - 0967-0661

ER -