Method of selecting process signals for creating diagnostic machines optimised to detect abnormalities in a plant using a support vector machine

Hirotsugu Minowa, Yoshiomi Munesawa, Yuichiro Furuta, Akio Gofuku

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In industrial plants, a multi-agent diagnostic system using machine learning with high discrimination performance to prevent the recurrence of accidents are desired because the damages to the company, industrial world, and humans are serious if an accident occurs. However, machine learning needs to be optimised according to the target, e.g. a plant operating condition diagnosis. Thus, we propose a method that selects process signals to optimise the performance of a diagnostic machine for a plant using factor analysis and decision-tree analysis. A feature of our optimisation method is that it accounts for combinations of process signals. Further, an advantage of our method is that the time to create an optimised diagnostic machine is short. The diagnostic machines need to be updated in a limited period, e.g. plant equipment repair, which changes regular process values. Another advantage is that our method can be applied to various learning machines to improve performance. This advantage allows the designer of the diagnostic system to use the best machine-learning method on each diagnostic machine. This paper reports our methodology, our proposed method, and the experimental results where a diagnostic machine was improved by 5.3% to 98.8% from 93.5% abnormality detection accuracy when our method which implemented a support vector machine was applied to the diagnostic machine to detect the blockage of a pipe in the desulfurisation process in a chemical plant simulator.

Original languageEnglish
Pages (from-to)205-210
Number of pages6
JournalChemical Engineering Transactions
Volume36
DOIs
Publication statusPublished - 2014

Fingerprint

Support vector machines
Learning systems
Accidents
Chemical plants
Factor analysis
Decision trees
Industrial plants
Repair
Simulators
Pipe
Industry

ASJC Scopus subject areas

  • Chemical Engineering(all)

Cite this

Method of selecting process signals for creating diagnostic machines optimised to detect abnormalities in a plant using a support vector machine. / Minowa, Hirotsugu; Munesawa, Yoshiomi; Furuta, Yuichiro; Gofuku, Akio.

In: Chemical Engineering Transactions, Vol. 36, 2014, p. 205-210.

Research output: Contribution to journalArticle

@article{951a38a0896e4301acc876646312cd8a,
title = "Method of selecting process signals for creating diagnostic machines optimised to detect abnormalities in a plant using a support vector machine",
abstract = "In industrial plants, a multi-agent diagnostic system using machine learning with high discrimination performance to prevent the recurrence of accidents are desired because the damages to the company, industrial world, and humans are serious if an accident occurs. However, machine learning needs to be optimised according to the target, e.g. a plant operating condition diagnosis. Thus, we propose a method that selects process signals to optimise the performance of a diagnostic machine for a plant using factor analysis and decision-tree analysis. A feature of our optimisation method is that it accounts for combinations of process signals. Further, an advantage of our method is that the time to create an optimised diagnostic machine is short. The diagnostic machines need to be updated in a limited period, e.g. plant equipment repair, which changes regular process values. Another advantage is that our method can be applied to various learning machines to improve performance. This advantage allows the designer of the diagnostic system to use the best machine-learning method on each diagnostic machine. This paper reports our methodology, our proposed method, and the experimental results where a diagnostic machine was improved by 5.3{\%} to 98.8{\%} from 93.5{\%} abnormality detection accuracy when our method which implemented a support vector machine was applied to the diagnostic machine to detect the blockage of a pipe in the desulfurisation process in a chemical plant simulator.",
author = "Hirotsugu Minowa and Yoshiomi Munesawa and Yuichiro Furuta and Akio Gofuku",
year = "2014",
doi = "10.3303/CET1436035",
language = "English",
volume = "36",
pages = "205--210",
journal = "Chemical Engineering Transactions",
issn = "1974-9791",
publisher = "AIDIC-Italian Association of Chemical Engineering",

}

TY - JOUR

T1 - Method of selecting process signals for creating diagnostic machines optimised to detect abnormalities in a plant using a support vector machine

AU - Minowa, Hirotsugu

AU - Munesawa, Yoshiomi

AU - Furuta, Yuichiro

AU - Gofuku, Akio

PY - 2014

Y1 - 2014

N2 - In industrial plants, a multi-agent diagnostic system using machine learning with high discrimination performance to prevent the recurrence of accidents are desired because the damages to the company, industrial world, and humans are serious if an accident occurs. However, machine learning needs to be optimised according to the target, e.g. a plant operating condition diagnosis. Thus, we propose a method that selects process signals to optimise the performance of a diagnostic machine for a plant using factor analysis and decision-tree analysis. A feature of our optimisation method is that it accounts for combinations of process signals. Further, an advantage of our method is that the time to create an optimised diagnostic machine is short. The diagnostic machines need to be updated in a limited period, e.g. plant equipment repair, which changes regular process values. Another advantage is that our method can be applied to various learning machines to improve performance. This advantage allows the designer of the diagnostic system to use the best machine-learning method on each diagnostic machine. This paper reports our methodology, our proposed method, and the experimental results where a diagnostic machine was improved by 5.3% to 98.8% from 93.5% abnormality detection accuracy when our method which implemented a support vector machine was applied to the diagnostic machine to detect the blockage of a pipe in the desulfurisation process in a chemical plant simulator.

AB - In industrial plants, a multi-agent diagnostic system using machine learning with high discrimination performance to prevent the recurrence of accidents are desired because the damages to the company, industrial world, and humans are serious if an accident occurs. However, machine learning needs to be optimised according to the target, e.g. a plant operating condition diagnosis. Thus, we propose a method that selects process signals to optimise the performance of a diagnostic machine for a plant using factor analysis and decision-tree analysis. A feature of our optimisation method is that it accounts for combinations of process signals. Further, an advantage of our method is that the time to create an optimised diagnostic machine is short. The diagnostic machines need to be updated in a limited period, e.g. plant equipment repair, which changes regular process values. Another advantage is that our method can be applied to various learning machines to improve performance. This advantage allows the designer of the diagnostic system to use the best machine-learning method on each diagnostic machine. This paper reports our methodology, our proposed method, and the experimental results where a diagnostic machine was improved by 5.3% to 98.8% from 93.5% abnormality detection accuracy when our method which implemented a support vector machine was applied to the diagnostic machine to detect the blockage of a pipe in the desulfurisation process in a chemical plant simulator.

UR - http://www.scopus.com/inward/record.url?scp=84899442626&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899442626&partnerID=8YFLogxK

U2 - 10.3303/CET1436035

DO - 10.3303/CET1436035

M3 - Article

VL - 36

SP - 205

EP - 210

JO - Chemical Engineering Transactions

JF - Chemical Engineering Transactions

SN - 1974-9791

ER -