eCite Digital Repository

Detecting significant alarms using outlier detection algorithms

Citation

Kang, BH and Kim, YS and Chen, Z and Kim, T, Detecting significant alarms using outlier detection algorithms, Proceedings of Interdisciplinary Research Theory and Technology conference (IRRT 2013), 21-23 November 2013, Jeju Island, Korea, pp. 1-8. ISSN 2287-1233 (2013) [Refereed Conference Paper]


Preview
PDF
Restricted - Request a copy
2Mb
  

Copyright Statement

Copyright 2013 Science and Engineering Research Support soCiety (SERSC)

Official URL: http://onlinepresent.org/proceedings/vol29_2013/84...

Abstract

Although alarms in plants are designed to notify any anomaly or faults in order to prevent accidents or to improve process, it is very difficult for the operators to identify meaningful alarms, since there are large volumes of false and nuisance alarms. Outlier detection algorithms are used to identify anomaly in data, and thus they can be used to suggest abnormal alarms. In this research, we analysed real world alarm data collected from an iron processing company and constructed data features for algorithmic outlier detection. With the data we identified outlier alarms and compared their run length with nonoutlier alarms. Our results demonstrated that outlier alarms detected by the algorithms have significantly difference patterns in average run length compared to the normal alarms.

Item Details

Item Type:Refereed Conference Paper
Keywords:alarm management, outlier detection
Research Division:Information and Computing Sciences
Research Group:Artificial Intelligence and Image Processing
Research Field:Pattern Recognition and Data Mining
Objective Division:Expanding Knowledge
Objective Group:Expanding Knowledge
Objective Field:Expanding Knowledge in the Information and Computing Sciences
Author:Kang, BH (Professor Byeong Kang)
Author:Kim, YS (Dr Yang Kim)
Author:Chen, Z (Mr Zhao Chen)
ID Code:89327
Year Published:2013
Deposited By:Information and Communication Technology
Deposited On:2014-03-02
Last Modified:2017-11-13
Downloads:0

Repository Staff Only: item control page