Preliminary validation of the 1D modeling of the DYNASTY natural circulation loop against results from water experimental campaign

In the continuous improvements of the safety of nuclear power plants, the study of natural circulation is of primary importance for developing passive safety systems and Generation-IV reactors, such as the Molten Salt Fast Reactor (MSFR). In particular, removing the decay heat during the shutdown of the MSFR involves dealing with a circulating fuel driven by natural circulation; as such, the development and validation of natural circulation models are of primary importance. To validate models and provide experimental data on natural circulation with a distributed heat source, the DYNASTY (DYnamics of NAtural circulation for molten SalT internallY heated) natural circulation loop has been built on the premises of Politecnico di Milano. This paper focuses on improving the 1D modeling of natural circulation phenomena in the DYNASTY facility and validating them against the results (also reported in this work) found in the first experimental campaign, carried out with water as the working fluid. The modeling of DYNASTY is based on a 1D approach using DYMOLA®, a modeling environment based on the MODELICA simulation language. The models used are improved versions of the ones available in the literature, which better model the experimental facility as they now include the mass flow rate meter, the heat loss between the facility and the environment, and a more realistic cooler. The improved model was then used to study the outcomes of the model simulations and verify the influence of the different numerical integration algorithms on the simulation results. The first DYNASTY experimental campaign provides in-depth testing of its behavior, with different cooling fan speeds and considering all possible heating configurations. The collected results for the heating and cooling transients are the mass flow rate and four fluid temperatures measured by thermocouples placed at the beginning and the end of the four DYNASTY legs. The experimental results were compared with the model simulations, showing the good predicting capabilities of the latter one following the tuning of the heat exchange model. In particular, the comparison showed an excellent match between the model and the experimental results for the heating transient while showing improving margins for the cooling transient.