Uncertainty from non-CHF closure models

Quantify and mitigate the uncertainty introduced by non-CHF closure models in CTF-based rod bundle critical heat flux prediction.

Background

The study evaluates machine-learning-based critical heat flux models trained on tube data and deployed in the CTF subchannel code for square rod bundle simulations. Its results show that model performance depends not only on the CHF formulation but also on the thermal-hydraulic conditions supplied by CTF, including quantities affected by closure relations such as turbulent mixing, wall friction, boiling, and void-drift models.

The authors identify uncertainty from these non-CHF closure models as an unresolved issue. Addressing it would help distinguish errors caused by the CHF prediction method from errors introduced by the underlying subchannel thermal-hydraulic representation, thereby improving the reliability of machine-learning and conventional CHF predictions in rod bundle applications.