Methodology for the use of experimental data to enhance model output uncertainty assessment in thermal hydraulics codes

Uncertainty analysis methodology in complex system models requires a comprehensive treatment of many different types and sources of uncertainties. This paper develops and demonstrates the application of a methodology for updating the results of uncertainty quantification of TH system codes when additional data are available from tests and experiments that are fully or partially applicable, or when they can be paired with corresponding code calculations, as well as cases where such pairing is not possible. The techniques discussed in this paper are a part of a more comprehensive methodology (IMTHUA) for the integrated assessment of TH system code uncertainties, covering both the model (structural) and parameter uncertainties. The Bayesian updating techniques presented enable one to further refine and adjust the uncertainty distribution that is obtained by propagation of various inputs and sub-model uncertainty contributors to TH code prediction. This additional refinement accounts for the residual uncertainties (model and parameter) that are not explicitly accounted for, including those resulting from missed/screened out models and known sources of uncertainties that cannot be easily and explicitly accounted for. Data from LOFT facility LBLOCA is used to demonstrate how the proposed Bayesian uncertainty adjustment methodology can be applied.