Improved High-Fidelity Multiphysics Modeling of Pulsed Operation of the Annular Core Research Reactor

Emory Colvin^*Todd S Palmer

• Oregon State University, Corvallis, United States

Sandia National Laboratories' Annular Core Research Reactor (ACRR) is a unique research reactor, using UO2-BeO fuel and operating primarily in pulsed mode. To better understand the physical characteristics of the fuel, the distribution of heat generation must be understood. Previous work developed a Serpent 2 model of the ACRR and Python coupling script to provide multiphysics feedback. Simulations of $1.50 and $2.00 pulses were compared to experimental results. This paper expands this work to $2.50 and $3.00 pulses. It further explores potential improvements to the model: dividing the fuel into two radial regions for feedback purposes, allowing additional iterations of the multiphysics coupling and checking for convergence, and the development of alternate specific heat capacity values.The use of two radial fuel regions improved agreement with experimental results for the simulations using the original function for specific heat capacity as a function of temperature but did not consistently improve results with the constant value for specific heat capacity. Allowing additional multiphysics iterations until the power distribution field converges also produced little change for reactor power prediction, though it improved maximum fuel temperature prediction slightly. The new values for specific heat capacity provided the most significant improvements to the models. A third-order polynomial developed from experimental data results in a significant improvement in fuel temperature prediction over the constant value with only a small loss of performance in reactor power prediction.