Best estimate calculation and uncertainty quantification of sodium-cooled fast reactor using MARS-LMR code

Abstract

The safety analysis of nuclear power plants has been performed using conservative approach based on conservative assumptions and boundary conditions to evaluate the safety margin of plant operation. However, this conservative approach could lead to unrealistic behavior predictions and eventually distort some phenomena in reactor systems. Therefore, the nuclear field moved towards an alternative best-estimate approach with uncertainty quantification in order to improve the phenomena prediction and to decrease the excessive conservatism in safety margins. In this study, the best estimate methodology is applied to improve the accuracy and reliability in safety analysis of an SFR. The applied best estimate methodology is based on the CSAU. This methodology is composed of three unique steps for evaluation of code capability, assessment and range of parameters, and sensitivity and uncertainty analysis. The primary purpose of this study is to evaluate the appropriateness of sensitivity parameters and its ranges, which have been determined through intensive experts’ panel discussion, by use of the data obtained from the EBR-II Unprotected Loss of Flow (ULOF) experiment. The MARS-LMR thermal–hydraulic code and the parallel computing platform integrated for uncertainty and sensitivity analysis (PAPIRUS) become the basic calculation tools in the study. Confirmation of data coverage is performed through the evaluation of coolant temperature in the instrumented subassemblies XX09. The appropriateness of parameters and its ranges are evaluated for three different cases: original parameters and ranges suggested in the MIRT, ±10% increased parameter ranges, and 200% increased axial reactivity feedback coefficient only. The case with the original parameters and ranges does not result in a valid data coverage, which means inadequate modeling accuracy for the ULOF scenario. The other two cases give complete coverage of EBR-II temperature data measured at the core top, which suggest the need of further refinement of reactivity models. The relative importance of the parameters is confirmed through the sensitivity analysis with respect to the Figures of Merit (FoM). The selected dominant parameters are the sodium density reactivity, above core load pad strain coefficient, core radial expansion reactivity coefficient and fuel axial expansion reactivity coefficient. The pump coastdown curve and the core inlet form loss are also found to be significant parameters during the transient.