Use of Imprecise Probabilities in Severe Accident Assessments considering Material Fatigue

Abstract of the technical paper at:
International Conference on Topical Issues in Nuclear Installation Safety: Learning from the Past to Accelerate the Future- IAEA TIC-2026
June 29 – July 3, 2026

Prepared by:
Mounia Berdaï
Canadian Nuclear Safety Commission

Abstract:

Repeated thermal cycling and prolonged exposure to neutron radiation can significantly alter the microstructure of materials and components within Nuclear Power Plants (NPPs), potentially degrading their mechanical properties over time. Consequently, the design values of these properties cannot be assumed to remain constant throughout the operational life of an NPP. To enhance the accuracy of predictive models, it becomes necessary to update mechanical property data to reflect the effects of aging.

However, due to the high cost of experimental testing and the limited availability of comprehensive data, expert judgment plays a critical role. In such contexts, experts may employ various methodologies to characterize epistemic uncertainty, including probabilistic and imprecise frameworks.

This study evaluates the influence of imprecise probabilities in uncertainty quantification, with a particular focus on their impact on the behavior of Figures of Merit (FOM) during the progression of a severe accident scenario. The analysis employs a multi-variable propagation model to capture the complex interactions and uncertainties inherent in such events.

The research highlights how degraded material properties, resulting from repeated thermal cycling and prolonged neutron exposure, can significantly affect accident progression and containment integrity. This underscores the need to update mechanical property data in safety assessments, directly supporting CNSC requirements for Aging Management and Fitness-for-Service programs aimed at ensuring the continued safety and reliability of structures, systems, and components (SSCs).

The application of imprecise probabilities and uncertainty quantification aligns with CNSC expectations for robust safety analysis methodologies that account for uncertainties in accident evolution and system behavior. Moreover, this study demonstrates how advanced modeling techniques can enhance predictive capabilities and inform future updates to CNSC guidance documents, particularly in areas where expert judgment is essential due to limited empirical data.

To obtain a copy of the abstract’s document, please contact us at info@cnsc-ccsn.gc.ca or call 613-995-5894 or 1-800-668-5284 (in Canada). When contacting us, please provide the title and date of the abstract.