COMSOL MULTIPHYSICS MODEL FOR DWPF CANISTER FILLING, REVISION 1

Kesterson, M.

Savannah River Site (S.C.)

关键词: Heat Flux;    Thermocouples;    Flow Rate;    Waste Processing;    Geometry;   

DOI  :  10.2172/1024868
RP-ID  :  SRNL-STI-2011-00209R
RP-ID  :  DE-AC09-08SR22470
RP-ID  :  1024868

美国|英语

来源: UNT Digital Library

【 摘 要 】

This revision is an extension of the COMSOL Multiphysics model previously developed and documented to simulate the temperatures of the glass during pouring a Defense Waste Processing Facility (DWPF) canister. In that report the COMSOL Multiphysics model used a lumped heat loss term derived from experimental thermocouple data based on a nominal pour rate of 228 lbs./hr. As such, the model developed using the lumped heat loss term had limited application without additional experimental data. Therefore, the COMSOL Multiphysics model was modified to simulate glass pouring and subsequent heat input which, replaced the heat loss term in the initial model. This new model allowed for changes in flow geometry based on pour rate as well as the ability to increase and decrease flow and stop and restart flow to simulate varying process conditions. A revised COMSOL Multiphysics model was developed to predict temperatures of the glass within DWPF canisters during filling and cooldown. The model simulations and experimental data were in good agreement. The largest temperature deviations were {approx} 40 C for the 87 inch thermocouple location at 3000 minutes and during the initial cool down at the 51 inch location occurring at approximately 600 minutes. Additionally, the model described in this report predicts the general temperature trends during filling and cooling as observed experimentally. The revised model incorporates a heat flow region corresponding to the glass pouring down the centerline of the canister. The geometry of this region is dependent on the flow rate of the glass and can therefore be used to see temperature variations for various pour rates. The equations used for this model were developed by comparing simulation output to experimental data from a single pour rate. Use of the model will predict temperature profiles for other pour rates but the accuracy of the simulations is unknown due to only a single flow rate comparison.