465078 Delay-Timer Alarm Design for Uncertain Chemical Systems

Monday, November 14, 2016
Grand Ballroom B (Hilton San Francisco Union Square)
Aditya Tulsyan, Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA and Bhushan Gopaluni, University of British Columbia, Vancouver, BC, Canada

Modern process industries are equipped with highly automated fault detection and diagnosis

(FDD) systems to maintain process safety and reliability standards. The primary objective

of an alarm system is to alert the operator every time a fault orccurs or the process deviates

from its normal behaviour. According to the Abnormal Situation Management Consortium

(ASM), the US petrochemical plants alone lose more than $10 billion per year due to ab-

normal plant situations and lack of ecient alarm management strategies [1]. This is due to

the complex tasks performed by the plant operators, which include presentation, processing,

and handing of hundreds and thousands of alarms from dierent units in a process plant.

Without an ecient alarm management system, alarm monitoring becomes a daunting task.

The alarm design is a critical rst step in setting up an ecient alarm management system.

The design and performance evaluation of alarm systems have received considerable attention

from both academic and industrial communities [2, 3, 4]. Although it has been extensively

studied and applied primarily to linear, Gaussian, and stationary systems with uncorrelated

signals [5], the real systems often tend to be nonlinear, non-Gaussian and exhibit complex

correlations in the dynamics. This is particularly true for closed-loop systems, since they

introduce strong system correlations even if the open-loop system has uncorrelated states.

We consider a limit-checking based alarm design problem for complex chemical processes

represented by discrete-time nonlinear stochastic state space models (SSMs). Given a sto-

chastic model representation of a process, we are interested in designing a univariate delay

timer alarm conguration for the process states. Once a delay-timer is designed, we evaluate

its performance based on the false and missed alarm rates. The ecacy of the proposed

alarm design is illustrated on a continuous stirred tank reactor (CSTR) reaction system.

References

[1] E. Cochran, C. Miller, and P. Bullemer, "Abnormal situation management in petrochemical plants: can

a pilot's associate crack crude?" in Proceedings of the IEEE 1996 National Aerospace and Electronics

Conference, Dayton, Ohio, May 20-23 1996.

[2] J. Ahnlund, T. Bergquist, and L. Spaanenburg, "Rule-based reduction of alarm signals in industrial

control," Journal of Intelligent and Fuzzy Systems, vol. 14, no. 9, pp. 73-84, 2003.

[3] R. Brooks, R. Thorpe, and J. Wilson, "A new method for dening and managing process alarms and for

correcting process operation when an alarm occurs," Journal of Hazardous Materials, vol. 115, no. 1, pp.

169-174, 2004.

[4] D. Rothenberg, Alarm Management for Process Control: A Best-Practice Guide for Design, Implemen-

tation, and Use of Industrial Alarm Systems. Momentum Press, 2009.

[5] N. A. Adnan, Y. Cheng, I. Izadi, and T. Chen, "Study of generalized delay-timers in alarm conguration,"

Journal of Process Control, vol. 23, no. 3, pp. 382-395, 2013.


Extended Abstract: File Not Uploaded
See more of this Session: Interactive Session: Systems and Process Control
See more of this Group/Topical: Computing and Systems Technology Division