Logical processing of sequential alarms for safer plant operation

• Published in: Computer Aided Chemical Engineering Vol. 53; pp. 1669 - 1674
• Main Authors: Tanaka, Ryuto, Noda, Masaru
• Format: Book Chapter
• Language: English
• Published: 2024
• Subjects: Dot Matrix Analysis, Plant Operation Data; Logical Alarm Processing; Sequential Alarms; Logical Alarm Processing; Dot Matrix Analysis, Plant Operation Data; Sequential Alarms
• ISBN: 9780443288241; 0443288240
• ISSN: 1570-7946
• DOI: 10.1016/B978-0-443-28824-1.50279-9

Logical alarm processing is a technique for processing signals from alarm sensors to generate more meaningful alarms for plant operators. This technique involves grouping and suppressing redundant alarms, and eclipsing of alarms on the same variable etc. While it can enhance the operational value of alarms, inappropriate deployment of logical alarm processing can result in safety hazards during plant operations. In this study, we propose a novel approach for logical alarm processing that focuses on suppressing sequential alarms. Sequential alarms refer to sets of alarms that occur in succession within a short period of time after the first waring alarm indicating an abnormality. These types of alarms reduce the ability of plant operators to cope with operation abnormalities as critical alarms are often lost in numerous other correlated ones. The proposed method comprises three parts: identifying sequential alarms in plant operation data through dot matrix analysis, detecting the occurrence of sequential alarms in real-time by matching patterns of alarms during plant operation, and suppressing unnecessary alarms in the detected sequential alarms. Application of the method to simulated operational data for a distillation column demonstrated that its capability to decrease the number of unnecessary alarms displayed to a plant operator.