Sensor fault diagnosis based on on-line random forests

Li Yang; Jia Zhang; Fang Deng; Jie Chen

doi:10.1109/ChiCC.2016.7553992

Sensor fault diagnosis based on on-line random forests

Li Yang
, Jia Zhang
, Fang Deng
, Jie Chen

Beijing Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

7 Scopus citations

Abstract

In order to reduce the memory requirement, and obtain real-time status of equipment, the paper proposed to use the the on-line random forests (ORFs) algorithm to identify sensor fault. The sample set is derived from Tennessee Eastman (TE) process. The models are updated by a group of sensor data, which are collected in each interval. As models are real-time and dynamic, the equipment could be tested at any time. Moreover, the samples obtained at previous intervals are not need to store. The results of experiments show that the accuracies of ORFs and Random Forests (RFs) are similar in sensor fault diagnosis process. And in some fast changing process, ORFs distinguishes fault types with higher accuracy, better adaptable and faster than RFs.

Original language	English
Title of host publication	Proceedings of the 35th Chinese Control Conference, CCC 2016
Editors	Jie Chen, Qianchuan Zhao, Jie Chen
Publisher	IEEE Computer Society
Pages	4089-4093
Number of pages	5
ISBN (Electronic)	9789881563910
DOIs	https://doi.org/10.1109/ChiCC.2016.7553992
State	Published - 26 Aug 2016
Externally published	Yes
Event	35th Chinese Control Conference, CCC 2016 - Chengdu, China Duration: 27 Jul 2016 → 29 Jul 2016

Publication series

Name	Chinese Control Conference, CCC
Volume	2016-August
ISSN (Print)	1934-1768
ISSN (Electronic)	2161-2927

Conference

Conference	35th Chinese Control Conference, CCC 2016
Country/Territory	China
City	Chengdu
Period	27/07/16 → 29/07/16

Keywords

On-line Random Forests
Sensor fault
Tennessee Eastman process

Access to Document

10.1109/ChiCC.2016.7553992

Cite this

@inproceedings{7fd1dc7870b2497bbd86bfeca5a543df,

title = "Sensor fault diagnosis based on on-line random forests",

abstract = "In order to reduce the memory requirement, and obtain real-time status of equipment, the paper proposed to use the the on-line random forests (ORFs) algorithm to identify sensor fault. The sample set is derived from Tennessee Eastman (TE) process. The models are updated by a group of sensor data, which are collected in each interval. As models are real-time and dynamic, the equipment could be tested at any time. Moreover, the samples obtained at previous intervals are not need to store. The results of experiments show that the accuracies of ORFs and Random Forests (RFs) are similar in sensor fault diagnosis process. And in some fast changing process, ORFs distinguishes fault types with higher accuracy, better adaptable and faster than RFs.",

keywords = "On-line Random Forests, Sensor fault, Tennessee Eastman process",

author = "Li Yang and Jia Zhang and Fang Deng and Jie Chen",

note = "Publisher Copyright: {\textcopyright} 2016 TCCT.; 35th Chinese Control Conference, CCC 2016 ; Conference date: 27-07-2016 Through 29-07-2016",

year = "2016",

month = aug,

day = "26",

doi = "10.1109/ChiCC.2016.7553992",

language = "英语",

series = "Chinese Control Conference, CCC",

publisher = "IEEE Computer Society",

pages = "4089--4093",

editor = "Jie Chen and Qianchuan Zhao and Jie Chen",

booktitle = "Proceedings of the 35th Chinese Control Conference, CCC 2016",

address = "美国",

}

Yang, L, Zhang, J, Deng, F & Chen, J 2016, Sensor fault diagnosis based on on-line random forests. in J Chen, Q Zhao & J Chen (eds), Proceedings of the 35th Chinese Control Conference, CCC 2016., 7553992, Chinese Control Conference, CCC, vol. 2016-August, IEEE Computer Society, pp. 4089-4093, 35th Chinese Control Conference, CCC 2016, Chengdu, China, 27/07/16. https://doi.org/10.1109/ChiCC.2016.7553992

Sensor fault diagnosis based on on-line random forests. / Yang, Li; Zhang, Jia; Deng, Fang et al.
Proceedings of the 35th Chinese Control Conference, CCC 2016. ed. / Jie Chen; Qianchuan Zhao; Jie Chen. IEEE Computer Society, 2016. p. 4089-4093 7553992 (Chinese Control Conference, CCC; Vol. 2016-August).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Sensor fault diagnosis based on on-line random forests

AU - Yang, Li

AU - Zhang, Jia

AU - Deng, Fang

AU - Chen, Jie

PY - 2016/8/26

Y1 - 2016/8/26

N2 - In order to reduce the memory requirement, and obtain real-time status of equipment, the paper proposed to use the the on-line random forests (ORFs) algorithm to identify sensor fault. The sample set is derived from Tennessee Eastman (TE) process. The models are updated by a group of sensor data, which are collected in each interval. As models are real-time and dynamic, the equipment could be tested at any time. Moreover, the samples obtained at previous intervals are not need to store. The results of experiments show that the accuracies of ORFs and Random Forests (RFs) are similar in sensor fault diagnosis process. And in some fast changing process, ORFs distinguishes fault types with higher accuracy, better adaptable and faster than RFs.

AB - In order to reduce the memory requirement, and obtain real-time status of equipment, the paper proposed to use the the on-line random forests (ORFs) algorithm to identify sensor fault. The sample set is derived from Tennessee Eastman (TE) process. The models are updated by a group of sensor data, which are collected in each interval. As models are real-time and dynamic, the equipment could be tested at any time. Moreover, the samples obtained at previous intervals are not need to store. The results of experiments show that the accuracies of ORFs and Random Forests (RFs) are similar in sensor fault diagnosis process. And in some fast changing process, ORFs distinguishes fault types with higher accuracy, better adaptable and faster than RFs.

KW - On-line Random Forests

KW - Sensor fault

KW - Tennessee Eastman process

UR - https://www.scopus.com/pages/publications/84987905638

U2 - 10.1109/ChiCC.2016.7553992

DO - 10.1109/ChiCC.2016.7553992

M3 - 会议稿件

AN - SCOPUS:84987905638

T3 - Chinese Control Conference, CCC

SP - 4089

EP - 4093

BT - Proceedings of the 35th Chinese Control Conference, CCC 2016

A2 - Chen, Jie

A2 - Zhao, Qianchuan

A2 - Chen, Jie

PB - IEEE Computer Society

T2 - 35th Chinese Control Conference, CCC 2016

Y2 - 27 July 2016 through 29 July 2016

ER -

Sensor fault diagnosis based on on-line random forests

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this