Standard

A multi-feature extraction technique based on principal component analysis for nonlinear dynamic process monitoring. / Guo, Lingling; Wu, Ping; Lou, Siwei; Gao, Jinfeng; Liu, Yichao.

In: Journal of Process Control, Vol. 85, 2020, p. 159-172.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

APA

Vancouver

Author

Guo, Lingling ; Wu, Ping ; Lou, Siwei ; Gao, Jinfeng ; Liu, Yichao. / A multi-feature extraction technique based on principal component analysis for nonlinear dynamic process monitoring. In: Journal of Process Control. 2020 ; Vol. 85. pp. 159-172.

BibTeX

@article{6f91766211b44275a1e203df640e6a2c,
title = "A multi-feature extraction technique based on principal component analysis for nonlinear dynamic process monitoring",
abstract = "Principal component analysis (PCA) and its modified methods have been widely applied in industrial process monitoring. In practice, industrial processes are with disparate characteristics, the process monitoring system should consider as many process characteristics as possible, such as dynamic and nonlinear characteristics. In this paper, a multi-feature extraction technique based on PCA is proposed for nonlinear dynamic process monitoring. The proposed method integrates dynamic inner PCA (DiPCA), PCA and kernel PCA (KPCA) methods through a serial structure to extract the dynamic, linear and nonlinear features among the process data. Along with the proposed method, the original data space is decomposed into several orthogonal subspaces, in which abnormal variations of different features can be monitored. For real-time process monitoring, a combined Hotelling's T2 statistic based on the extracted multi-feature and a squared prediction error (SPE or Q) statistic are established. Case studies on a numerical example and the Tennessee Eastman process are carried out to demonstrate the superior process monitoring performance of the proposed method compared with other relevant methods.",
keywords = "Multi-feature extraction, Nonlinear dynamic process, Principal component analysis, Process monitoring",
author = "Lingling Guo and Ping Wu and Siwei Lou and Jinfeng Gao and Yichao Liu",
year = "2020",
doi = "10.1016/j.jprocont.2019.11.010",
language = "English",
volume = "85",
pages = "159--172",
journal = "Journal of Process Control",
issn = "0959-1524",
publisher = "Elsevier",

}

RIS

TY - JOUR

T1 - A multi-feature extraction technique based on principal component analysis for nonlinear dynamic process monitoring

AU - Guo, Lingling

AU - Wu, Ping

AU - Lou, Siwei

AU - Gao, Jinfeng

AU - Liu, Yichao

PY - 2020

Y1 - 2020

N2 - Principal component analysis (PCA) and its modified methods have been widely applied in industrial process monitoring. In practice, industrial processes are with disparate characteristics, the process monitoring system should consider as many process characteristics as possible, such as dynamic and nonlinear characteristics. In this paper, a multi-feature extraction technique based on PCA is proposed for nonlinear dynamic process monitoring. The proposed method integrates dynamic inner PCA (DiPCA), PCA and kernel PCA (KPCA) methods through a serial structure to extract the dynamic, linear and nonlinear features among the process data. Along with the proposed method, the original data space is decomposed into several orthogonal subspaces, in which abnormal variations of different features can be monitored. For real-time process monitoring, a combined Hotelling's T2 statistic based on the extracted multi-feature and a squared prediction error (SPE or Q) statistic are established. Case studies on a numerical example and the Tennessee Eastman process are carried out to demonstrate the superior process monitoring performance of the proposed method compared with other relevant methods.

AB - Principal component analysis (PCA) and its modified methods have been widely applied in industrial process monitoring. In practice, industrial processes are with disparate characteristics, the process monitoring system should consider as many process characteristics as possible, such as dynamic and nonlinear characteristics. In this paper, a multi-feature extraction technique based on PCA is proposed for nonlinear dynamic process monitoring. The proposed method integrates dynamic inner PCA (DiPCA), PCA and kernel PCA (KPCA) methods through a serial structure to extract the dynamic, linear and nonlinear features among the process data. Along with the proposed method, the original data space is decomposed into several orthogonal subspaces, in which abnormal variations of different features can be monitored. For real-time process monitoring, a combined Hotelling's T2 statistic based on the extracted multi-feature and a squared prediction error (SPE or Q) statistic are established. Case studies on a numerical example and the Tennessee Eastman process are carried out to demonstrate the superior process monitoring performance of the proposed method compared with other relevant methods.

KW - Multi-feature extraction

KW - Nonlinear dynamic process

KW - Principal component analysis

KW - Process monitoring

UR - http://www.scopus.com/inward/record.url?scp=85076245705&partnerID=8YFLogxK

U2 - 10.1016/j.jprocont.2019.11.010

DO - 10.1016/j.jprocont.2019.11.010

M3 - Article

VL - 85

SP - 159

EP - 172

JO - Journal of Process Control

T2 - Journal of Process Control

JF - Journal of Process Control

SN - 0959-1524

ER -

ID: 68361262