Alarm management via temporal pattern learning

Abstract

Industrial plant safety involves integrated management of all the factors that may cause accidents. Process alarm management can be formulated as a pattern recognition problem in which temporal patterns are used to characterize different typical situations, particularly at startup and shutdown stages. In this paper we propose a new approach of alarm management based on a diagnosis process. Assuming the alarms and the actions of the standard operating procedure as discrete events, the diagnosis step relies on situation recognition to provide the operators with relevant information on the failures inducing the alarm flows. The situation recognition is based on chronicle recognition where we propose to use the hybrid causal model of the system and simulations to generate the representative event sequences from which the chronicles are learned using the Heuristic Chronicle Discovery Algorithm Modified (HCDAM). An extension of this algorithm is presented in this article where the expertise knowledge is included as temporal restrictions which are a new input to HCDAM. An illustrative example in the field of petrochemical plants is presented.

An additional layer of protection through superalarms with diagnosis capability

“…The principle of Chronicle Based Alarm Management -CBAMis to consider several process situations (normal or abnormal) during start-up and shutdown stages and to model each of these situations through a learning chronicle. Thus, given the situation to be modeled, the algorithm HCDAM (Heuristic Chronicle Discovery Algorithm Modified) is fed by a set of event sequences structured from simulations and expert knowledge, giving us the respective chronicle of each situation [36]. Finally, with the chronicle built, a super-alarm can be generated giving to the operator's relevant information and assuming it as a new layer of protection to reduce accident occurrences because in many situations of alarm flood, hazard scenarios exist.…”

Decision making under uncertainty in the alarm systems response

“…A new resilience measure is suggested in Yoon et al (2017), which considers false alarm rates and reliability and formulated in a probabilistic framework. Furthermore, many alarm management strategies were also proposed in Zhu et al (2014), Yi (2016) andVasquez Capacho et al (2017). Aven and Zio (2011) argued the need for considering uncertainties in decision making on the grounds of the challenges involved in the treatment of uncertainties in risk assessment.…”

“…A new resilience measure is suggested in Yoon et al (2017), which considers false alarm rates and reliability and formulated in a probabilistic framework. Furthermore, many alarm management strategies were also proposed in Zhu et al (2014), Hu and Yi (2016) and Vasquez Capacho et al (2017).…”