Uncertainty analysis of suppression pool heating during an ATWS in a BWR-5 plant. An application of the CSAU methodology using the BNL engineering plant analyzer

Abstract

The uncertainty has been estimated of predicting the peak temperature in the suppression pool of a BWR power plant, which undergoes an NRC-postulated Anticipated Transient Without Scram (ATWS). The ATWS is initiated by recirculation-pump trips, and then leads to power and flow oscillations as they had occurred at the LaSalle-2 Power Station in March of 1988. After limit-cycle oscillations have been established, the turbines are tripped, but without MSIV closure, allowing steam discharge through the turbine bypass into the condenser. Postulated operator actions, namely to lower the reactor vessel pressure and the level elevation in the downcomer, are simulated by a robot model which accounts for operator uncertainty. All balance of plant and control systems modeling uncertainties were part of the statistical uncertainty analysis that was patterned after the Code Scaling, Applicability and Uncertainty (CSAU) evaluation methodology. The analysis showed that the predicted suppression-pool peak temperature of 329.3 K (133{degrees}F) has a 95-percentile uncertainty of 14.4 K (26{degrees}F), and that the size of this uncertainty bracket is dominated by the experimental uncertainty of measuring Safety and Relief Valve mass flow rates under critical-flow conditions. The analysis showed also that the probability of exceeding the suppression-pool temperature limit of 352.6 Kmore » (175{degrees}F) is most likely zero (it is estimated as < 5-104). The square root of the sum of the squares of all the computed peak pool temperatures is 350.7 K (171.6{degrees}F).« less