Structural and thermal fluid dynamics analyses of the ITER Pressure Suppression System considering no stable steam condensation regimes

At the University of Pisa, an experimental campaign was performed with the purpose of qualifying the ITER Vacuum Vessel Pressure Suppression System (VVPSS) and studying the Direct Contact Condensation at sub-atmospheric conditions. With financial support from ITER Organization, a Large-Scale Experimental Facility was designed and built to investigate the steam condensation in the operation conditions of the ITER VVPSS. The experimental tests were performed by injecting steam, produced by an electric steam generator (1.7 MW of power), through a multi-hole sparger into a condensation tank called Experimental Test Tank (ETT). On the sparger support, strain gauges and an accelerometer were installed to analyse the structural behaviour during the experimental tests. Under certain thermal hydraulic conditions (low subcooling and low steam mass flow rate) a large steam pocket can form and collapse causing high loads on the sparger structure. The aim of this study was comparing numerical fluid dynamics and structural analyses with the experimental results in order to estimate the pressure impulse caused by bubble collapse. The bubble collapse was simulated by means of ANSYS Fluent code in order to analyse the bubble dynamics and to determine the pressure impulse. The shape, the dimension and the collapse time of steam pocket were experimentally determined by means of an image analysis of video recorded during the tests. The numerical and experimental maximum steam bubble volume at the final stage of the transient before the detachment shows a good agreement with an error of 6 %. The pressure load was estimated at 5 different points of the simulation domain; the maximum pressure impulse on the sparger is 1.15 bar and has a duration of 40 ms. A dynamic structural FEM analysis was performed applying the pressure impulses on the sparger in order to evaluate the accelerations. The average value of RMS, maximum, minimum and duration of the experimental acceleration signals was obtained and compared to the numerical one. The errors of the RMS, the maximum and the minimum acceleration of the sparger are 16 %, 35 % and 51 %, respectively. The time duration of the oscillations is in good agreement (the error is only 2.54 %).