Three dimensional effects in analysis of PWR steam line break accident

Abstract

A steam line break accident is one of the possible severe abnormal transients in a pressurized water reactor. It is required to present an analysis of a steam line break accident in the Final Safety Analysis Report (FSAR) of a nuclear power reactor. Only simple one-dimensional core simulations are currently applied to such accident analysis. However, the asymmetric characteristics of a steam break accident may require more detailed local information in order to determine the potential fuel damage resulting from the transient. TITAN, a coupled (neutronics and thermal-hydraulics) code with state-of-the-art neutronics and thermal-hydraulics models, is therefore modified and applied to steam line break accident simulations.

The capabilities that are added to the code for a steam line break analysis include multiregion core inlet temperature forcing function, total inlet coolant flow rate boundary condition, total inlet coolant flow rate transient simulation capability, boron tracking equations, flow/coolant temperature transient plus control rod transient option, and one-dimensional, fully implicit numerical scheme.for thermal-hydraulics calculations. The modifications to TITAN are tested with a ten-channel PWR model. For inlet coolant temperature transients (one of the transients involved in a steam line break accident) test calculations lead to the conclusion that there is no significant difference between the results of closed- and open-channel calculations until boiling occurs.

A ten-channel model with two partially inserted control rods is employed for the transient simulations. Steady state conditions are obtained first by both open- and closed-channel calculations. Results show that cross flow between channels is insignificant. Thus, the onedimensional, fully implicit numerical scheme for the thermal-hydraulics equations is useful to speed up the calculations. More than half of the computational effort is saved by using this scheme compared to the semi-implicit numerical scheme.

Two extreme situations relevant to a steam line break accident are investigated: (1) Significant boiling due to severe depressurization when no return to power exists. (2) Return to power with no boiling because of high coolant temperature feedback coefficients. It is concluded that even after boiling occurs, the global parameters, such as total power and assembly power, still show no significant difference between the closed- and open-channel calcualtions. However, the local parameters, such as nodal power, void fraction and MDNBR, reveal differences between the two calculations. Results show that the open-channel calculation predicts lower MDNBR values as compared with the results of closed-channel calculation in a vapor generation process, since the coolant is driven away from of the hot spots. On the other hand, during a vapor condensation process, closed-channel calculations predict lower MDNBR results, since no cross flow is allowed to accelerated the condensation process.

A closed-channel, uniform inlet coolant temperature transient calculation is performed. The results verify the necessity for a three-dimensional calculation of the accident simulations, since no boiling was predicted by-the one-dimensional calculation throughout the simulation period.