The nuclear industry has fully embraced the development of accelerated fuel qualification (AFQ) approaches to speed up the assessment and validation of new fuel designs with respect to performance and safety metrics. To support the AFQ approach to shortening the time to develop and qualify new fuel for higher plant performance, Westinghouse utilizes advanced modeling and simulation technologies as part of their integrated and comprehensive AFQ vision through improved fuel performance prediction under various operating conditions and accident scenarios.

This paper provides example applications, prioritized in Westinghouse using machine learning technology, for fuel thermal-hydraulic applications with methodologies that are under development for the prediction of critical heat flux for pressurized water reactor (PWR) fuel thermal margin assessment and surrogate model development for crud-induced power shift risk prediction to enhance PWR fuel operation performance.