Frequent Alarm Pattern Mining of Industrial Alarm Flood Sequences by an Improved PrefixSpan Algorithm

Alarm systems are essential to the process safety and efficiency of complex industrial facilities. However, with increasing size plants growing complexity processes, alarm flooding is becoming a serious problem posing challenges systems. Extracting patterns from an flood database can assist root cause analysis, decision support, configuration suppression model. due large sequence ambiguity in sequence, existing algorithms suffer excessive computational overhead, incomplete patterns, redundant outputs. In order solve these problems, we propose pattern extraction method based on improved PrefixSpan algorithm. Firstly, priority-based pre-matching strategy proposed cluster similar sequences advance. Secondly, by considering timestamps tolerate short-term sequences. Thirdly, compression for further distillation information output representative patterns. Finally, evaluated effectiveness applicability using real diesel hydrogenation unit.