In the context of emergency conditions in Nuclear Power Plants, FLEX strategies aimed at preventing fuel damage or mitigating severe accidents may be required. The scale and complexity of this problem increase when such strategies are analysed for multi-unit sites.
The objective of this paper is to present a probabilistic modelling approach of human actions involved in the management of emergency conditions that could affect a hypothetical SMR multi-unit site and that may generate post-initiating human failure events (HFEs). In general, SMR designs are characterized by a reduced need for operator actions to maintain operability, as well as by the incorporation of passive safety systems for emergency conditions. However, in scenarios involving the failure or unavailability of passive systems, FLEX strategies may be required. The scope focuses on the human actions required to implement FLEX-type strategies (mobile equipment deployment), under conditions that may lead to core damage in one or both units.
The methodology begins with a bibliographic review aimed at identifying human actions from the early design stage, based on concepts of operations, and their associated HFEs. A bottom-up scheme of human actions has been considered, ranging from the requirement for FLEX system or equipment connection/actuation at the unit level to decision-making processes related to resource allocation at the site level. Subsequently, taking advantage of the flexibility of the SPAR-H method, different HFEs have been evaluated, including the analysis of dependencies among human actions. These methodological aspects have been applied to a hypothetical case study. A conceptual modelling approach using EMRALD as a dynamic Probabilistic Risk Assessment tool is presented.
As a result of this work, a modelling approach for human actions in the management of emergency conditions has been developed, including the quantification of HFEs. The approach encompasses the development of a conceptual modelling scheme in EMRALD, the quantification of HFEs using the SPAR-H method, and the modelling of dependency relationships that influence HFE probabilities among unit operators, shift supervisor responsible for site-level monitoring and decision-making, and their coordination with the Technical Support Center (TSC). This approach is being used in the development of a dynamic probabilistic model intended to support the analysis of alternative on-site emergency management strategies and the development of emergency operating documentation.