Natural Circulation in the Upper Plenum of a Scaled Model of a Very High Temperature Reactor in the Event of Loss-of-Coolant Accident

Abstract

The very high temperature reactor (VHTR) is one of the most promising next generation reactors which will be commercialized in 2030. A loss-of-coolant accident (LOCA) is a major accident scenario in which the primary coolant loop is broken, resulting in a loss of forced circulation of helium into the reactor vessel. With the onset of natural circulation, coolant flow reverses and is driven by buoyancy forces. The goal of the research is to simulate this accident condition on a 1/16th scaled model and visualize the flow behavior in the upper plenum of the VHTR. The facility was designed and constructed from a set of scaling parameters and outfitted with various instrumentation to characterize the depressurized conduction cooldown (DCC) event. Particle image velocimetry (PIV) is a nonintrusive optical laser technique used to obtain an instantaneous velocity field and was successfully applied to this system. Throughout the preliminary tests, the number of frames to be averaged to reach a statistically steady state was obtained from 1,000 images. The performance of the PIV method is validated with a flowmeter and analytic flowrate equation. The uncertainty of PIV system was also quantified.

Single jet tests are performed to provide a basic understanding of the simplest turbulent buoyant jet mixing in the upper plenum. By the Morton length scale, it was observed that the buoyant jet behaves like a plume and self-similarity is obtained for the axial velocity profiles. Q criterion is applied to identify the eddy structures of the turbulent jet mixing as a way to characterize the mechanism of vortex-pair mixing on the dome surface. Subsequent triple jet experiments are performed and compared with the results from single jet tests. Velocity distributions along the concave wall show that higher wall shear stress is obtained in single jet tests. The experiment results will provide the benchmark data for the PIV validation.