Generalized Pattern Matching of Industrial Alarm Flood Sequences via Word Processing and Sequence Alignment
Boyuan Zhou, Wenkai Hu, Kevin Brown, Tongwen Chen
Abstract
Analyzing alarm floods in a large-scale industrial facility is a sophisticated task, because of too many fault types and their associated consequential alarms. However, there exist similar alarm floods in different processes, where the alarms are associated with the same fault types, but configured with different tag names. If these alarm floods are discovered and grouped, the obtained results could facilitate root cause analysis and lead to generalized solutions. Motivated by this practical problem, a systematic pattern-matching method to compare alarm floods across different processes is proposed in this article. The contributions are twofold: 1) A word processing approach is proposed to generalize alarm representations; 2) a pattern-matching approach is developed to compare alarm floods across different processes. The effectiveness of the proposed method is demonstrated by a case study using alarm data from an industrial facility.