Integration of additional information sources for improved alarm flood detection
Publication date
2017
Document type
Conference paper
Author
Organisational unit
Conference
IEEE 21st International Conference on Intelligent Engineering Systems : INES 2017
Book title
IEEE 21st International Conference on Intelligent Engineering Systems (INES), Larnaca, Cyprus, 20-23 Oct. 2017
First page
19
Last page
26
Peer-reviewed
✅
Part of the university bibliography
✅
Abstract
The aim of alarm flood detection is the identification of similar, frequently occurring sequences of alarm messages in historical alarm data and uses the results for root cause analysis or alarm flood reduction. Various promising approaches for alarm data of automated production systems exist. However, due to the high amount of alarm messages transmitted by industrial alarm systems, floods are often interrupted by alarms stemming from different root causes, leading to non-relevant or invalid results of purely data-driven flood detection approaches. To improve the results of data-driven approaches, this paper suggests considering a process plant's hierarchy to divide historical alarm data into independent sub-datasets. For this reason, the paper discusses necessary plant information to explain a process plant's hierarchy and analyzes existing approaches to extract this hierarchy automatically from information sources. It then discusses whether existing approaches for alarm flood detection consider this hierarchy and how it could improve the approaches' results.
Version
Not applicable (or unknown)
Access right on openHSU
Metadata only access