Uncertainty evaluation of RBMK NPP DGH blockage calculation by the Statistical and the CIAU Methodologies

The best-estimate calculation results from complex system codes are affected by approximations that are un-predictable without the use of computational tools that account for the various sources of uncertainty. Two uncertainty methodologies (the statistical methodology based on GRS approach and the CIAU - Code with the capability of Internal Assessment of Uncertainty) have been applied to the RBMK NPP distribution group header (DGH) blockage simulation. The statistical methodology allows building with the given reliability the confidence interval for the estimated parameter both in the case of unknown distribution and in the case of the normal distribution. It also allows finding the factors with the most essential influence on the result of calculation. In the framework of the statistical methodology the uncertainty analysis of the simulation of the damaged DGH parameters response is carried out in two stages: • base mode calculation (without uncertainty variations), definition of boundary conditions for a small subsystem (blocked DGH): pressure and water enthalpy in pressure header (PH); pressure and void fraction in the point of connection of steam-water lines to drum separator (DS); • performance of variant calculations for the small subsystem: damaged DGH from PH to DS at constant (from calculation to calculation) boundary conditions. The distribution of the maximal power channel break times is received. This event leads to the scram initiation, and the upper boundary for the maximal fuel rod cladding and pressure tube temperatures of the channels at lower power is calculated. All calculations have been done with the help of RELAP5\MOD3.2 code. The Code with (the capability of) Internal Assessment of Uncertainty (CIAU) has been proposed by the University of Pisa to realize the integration between a qualified best-estimate thermal-hydraulic system code (RELAP5\MOD3.2) and a suitable uncertainty methodology (UMAE, Uncertainty Method based on Accuracy Extrapolation) and to supply proper uncertainty bands each time a nuclear power plant (NPP) transient scenario is calculated. S6-37-2 The idea at the basis of the CIAU is connected with the status approach. Each status is characterized by the numerical value of relevant quantities and by the value of the time interval from transient start. The selected quantities characterize in a multidimensional space (i.e. a hypercube) the status of a LWR during any transient (LBLOCA, SBLOCA, operational transient, etc.). Then, uncertainty values are calculated and stored in each hypercube (Uq) and time interval (Ut). A Quantity Uncertainty Matrix (QUM), formed by hypercubes, and a Time Uncertainty Vector (TUV), formed by time intervals, can be set up. The application of CIAU is straightforward once QUM and TUV are available. The “error matrixes” and the “error vector” are currently used as post-processor of a CIAU calculation and their application to the RBMK NPP DGH Blockage calculation has been performed. Uncertainty evaluations for upper plenum pressure, mass inventory of the affected DGH Core channels and hottest rod clad temperature have been derived and continuous uncertainty bands have been generated.