460837 Comparison of Two Identification Methods for Identifying Sequential Alarms in Plant Operation Data

Monday, November 14, 2016
Grand Ballroom B (Hilton San Francisco Union Square)
Zhexing Wang, Fukuoka University, Fukuoka, Japan and Masaru Noda, Department of Chemical Engineering, Fukuoka University, Fukuoka, Japan

The advance of distributed control systems (DCS) in the chemical industry has made it possible to inexpensively and easily install numerous alarms in DCS. While most alarms help operators detect an abnormality and identify its cause, some are unnecessary. A poor alarm system might cause sequential alarms, which are a collection of numerous alarms that almost always occur simultaneously with specific time lags within a short time (EEMUA, 2007). Sequential alarms reduce the ability of operators to cope with plant abnormalities because critical alarms are buried under many unnecessary ones. Sequential alarms are usually caused by poor alarm rationalization (Hollifield, et al., 2007). Therefore, it is very important to identify sequential alarms in a plant operation data for improving the alarm system and process operation.

The grouping correlated sequential alarms in accordance with their degree of similarity helps to reduce sequential alarms more effectively than by analyzing individual sequential alarms. Event correlation analysis (Nishiguchi, et al., 2010) was proposed to identify sequential alarms in noisy plant operation data. Event correlation analysis was applied to the operation data of an industrial ethylene plant and was able to correctly identify similarities between correlated sequential alarms (Higuchi, et al., 2010). However, event correlation analysis occasionally failed to detect similarities between two physically related sequential alarms when deletions, substitutions, and/or transpositions occurred in the alarm sequence.

In previous study (Wang, et al., 2015), we proposed a new identification method of sequential alarms by applying the dot matrix analysis to a plant operation data (Kurata et al., 2011). Dot matrix analysis (Gibbs, et al., 1970) is one of the sequence alignment methods for identifying similar regions in DNA or RNA. Similar regions in DNA or RNA may be a consequence of functional, structural, or evolutionary relationships between the sequences. In this research, in order to demonstrate the usefulness of the previous study, we compare it with the event correlation analysis. Two methods were applied to the plant operation data of an azeotropic distillation column. The results revealed that the dot matrix analysis is able to identify similar sequential alarms in plant operation data, when deletions, substitutions, and/or transpositions occurred in the alarm sequence..



The Engineering Equipment and Material Users' Association (EEMUA): Alarm Systems A Guide to Design, Management and Procurement Publication No. 191 Edition 2,EEMUA, London (2007)

Gibbs, A. J., McIntyre, G. A.: The diagram method for Comparing Sequences. Its Use with Amino Acid and Nucleotide Sequences, Eur. J. Biochem., 16, 1-11 (1970)

Hollifield, B. R., Habibi, E.: Alarm Management: Seven Effective Methods for Optimum Performance, ISA, Research Triangle Park (2007)

Higuchi, F., Noda, M., Nishitani, H.: Alarm Reduction of Ethylene Plant using Event Correlation Analysis (in Japanese), Kagaku Kogaku Ronbunshu, 36(6), 576-581 (2010)

Kurata, K., Noda, M., Kikuchi, Y., and Hirao, M.: Extension of Event Correlation Analysis for Rationalization of Plant Alarm Systems (in Japanese), Kagaku Kogaku Ronbunshu, 37, 338-343 (2011)

Nishiguchi, J., and Takai, T: IPL2&3 Performance Improvement Method for Process Safety Using the Event Correlation Analysis, Computers & Chemical Engineering, 34, 2007-2013 (2010)

Wang, Z., Noda, M. : Identification of Sequential Alarms in Plant Operation Data by using Dot Matrix Analysis (in Japanese), Kagaku Kogaku Ronbunshu, 41, 333-339 (2015)

Extended Abstract: File Not Uploaded