Probabilistic Assessments: Principles and Computational Methods
3rd International Seminar on Probabilistic Methods for Nuclear Applications
October 22-24, 2019
Canadian Nuclear Safety Commission
Professor, NSERC-UNENE Chair
University of Waterloo
The risk-informed decision making approach has been receiving increasing consideration by the nuclear industry and the regulatory authorities worldwide. In Canada, the observed advantages motivated additional activities towards introducing probabilistic methodologies into the evaluations relating to fitness-for-service of CANDU reactor components including pressure tubes, steam generator tubing and feeder piping.
A commonly observed “intuitive” approach in developing of a new methodology with probabilistic sampling is to adopt already existing and widely accepted backbone of deterministic methodology and enrich it with a set of distributed variables. Consequently, the outputs also become of a distributed quantity viewed as the results from a probabilistic evaluation, and the best estimate is typically selected to conform whether component condition is acceptable. Nevertheless, it is prudent to further study in details the approach of converting of an existing deterministic methodology into a probabilistic one. The faced challenges may relate to the interpretations of probabilistic outputs, and relating them to a suitable measure of reliability.
While more realistic mechanical responses are usually obtained through numerical finite element modeling, more realistic evaluations of reactor components condition could be envisioned as obtained from a simulation-based probabilistic platform. Some existing examples of the probabilistic evaluations of pressure tubes relate to deformation with operating time due to creep, probabilistic leak-before-break and probabilistic fracture protection.
The assurance of safe operation of the reactor components with intended reliability over an evaluation period requires that the concept of time-dependent reliability framework be properly utilized while considering the intent of the evaluation. The underlying principles are presented for a probabilistic framework that is in harmony with the reliability theory that has been developed and enhanced over a number of decades in engineering literature. The approaches in which these principles could be incorporated into probabilistic methodologies are highlighted.
To obtain a copy of the abstract’s document, contact us at email@example.com or call 613-995-5894 or 1-800-668-5284 (in Canada). When contacting us, please provide the title and date of the abstract.
- Date modified: