Innovative Analysis Techniques for Canadian SCWR Neutronics

Abstract

Knowledge of the effects of nuclear data uncertainties and physics approximations is crucial for the development, design, operation, and accident mitigation, of nuclear power plants. A framework to create a simulated fuel bundle, based on sensitivities and similarities, has been developed. The methodology allows safe-to-handle fuel to be manufactured such that it mimics irradiated fuel and can be used to reduce simulation uncertainties and better predict an application’s response. In this work, similarity values of ck = 0.967, E = 0.992, and G = 0.891 were found between between the irradiated fuel, and non-irradiated simulated fuel. In addition, a set of ZED-2 experiments has been analyzed that are applicable to an SCWR nuclear data adjustment and simulation bias determination. This was shown through high sensitivity coverage of many important nuclides, however, a low completeness value of R=0.24 indicates the set of 39 experiments alone is not sufficient for an accurate bias determination. Lastly, a technique has been presented that reduces diffusion calculation errors through the use of novel and practical mean discontinuity factors. The discontinuity factors have shown to reduce maximum channel power errors by up to 6.7%, and reactivity errors by 2.6 mk, compared to conventional analysis techniques.