Simulation of PWR's Passive Containment Cooling with an Advanced Water Film Model

摘要

In this paper the physical model describing water film behaviors in German containment code system COCOSYS has been improved by taking consideration of film breakup and subsequent phenomena as well as the effect of film interfacial shear stress created by countercurrent gas flow. The modified model has extended its capability to predict particular water film behaviors including breakup at a critical film thickness according to the MTE (Minimum Total Energy) criterion, the formation of rivulets calculated by taking account of the equilibrium of total energy and a prescribed contact angle between rivulets and steel substrate. The performance of rivulets such as constriction/expansion can also be predicted by the new model with several assumptions. Furthermore, the modification considers the change of cross section velocity distribution of film/rivulets due to shear stress resulted by countercurrent air flow. Simulations predicting containment pressure variation during accidents with operation of PCCS (Passive Containment Cooling System) have been carried out. The results indicate that with the new model, considerably larger peak pressures are obtained by comparing with those predicted by original water film model. This suggests that inadequate cooling water flow rate for PCCS could result in dry patches of steel containment surface and therefore reduction of coverage rate of water film. Additionally, countercurrent gas flow can increase the film thickness and thus reduce the film surface temperature. These facts could restrict the capability of heat removal and lead to higher pressure inside containment. Therefore, consideration of film breakup and its further behavior is of crucial importance for predicting PCCS performance. Simulation results also point out that contact angle between water rivulets and steel substrate plays a significant role in the film cooling. Since the value of contact angle is dependent on numerous factors and hence hard to determine, sensitivity analysis as well as comparisons with experiment data have been presented and discussed.