- 123 views
- 135 downloads
Advanced Machine Learning-based Alarm Flood Monitoring Using Alarm Data
-
- Author / Creator
- Alinezhad, Haniyeh Seyed
-
Monitoring industrial processes is essential to maintain efficient and safe operation. An alarm system is an imperative component of industrial process monitoring, as it alerts operators to abnormal conditions. Alarms indicate when the process is disrupted, which allows operators to react accordingly. A modern industrial plant consisting of many interconnections is susceptible to fault propagation via information and material flow pathways. During fault propagation, a large number of alarms are triggered in control rooms, which results in a phenomenon known as alarm flood. As a result, operators are overwhelmed with a high volume of alarms and may miss out on safety and efficiency measures. It is therefore crucial to deal with the problem of alarm floods, particularly for online applications. The development of efficient online alarm flood analysis can provide operator decision support for root cause analysis of abnormal events and prevent the occurrence of destructive effects. The large amount of data generated by modern computerized processes has recently led to considerable interest in data-based methods. Developing data-driven methods in the field of machine learning can reduce reliance on expert knowledge and human effort in online alarm monitoring. Therefore, this thesis focuses on the development of machine learning-based methods for alarm management and alarm flood monitoring using alarm data.
-
- Subjects / Keywords
-
- Graduation date
- Fall 2023
-
- Type of Item
- Thesis
-
- Degree
- Doctor of Philosophy
-
- License
- This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.