Nuclear data uncertainty propagation to the main physical parameters of NUR research reactor

Abstract

An accurate nuclear reactor model with less statistical uncertainty is not enough to have an accurate assessment of the calculated physical parameter value. It is worthwhile to note that another source of uncertainties originating from nuclear data uncertainties must be quantified. To realize such an assessment, sensitivity and uncertainty analysis based on first-order perturbation theory implemented in the SCALE code system was applied to the IV.N (an optimized one in series of configurations) core configuration of the Algerian light water Nuclear Uranium Reactor (NUR). The impact of nuclear data uncertainties on the effective multiplication factor was analyzed. Criticality calculation was performed by using the KENO V.a Monte Carlo code of the SCALE code system and the ENDF/B libraries with 238 and 252 energy group structures. The results were compared to the available measurement and validated. The accuracy of this criticality calculation was judged sufficient to do a sensitivity and uncertainty calculation for the same model.First, sensitivity and uncertainty calculations were performed using the TSUNAMI3D module of the SCALE code system and ENDF/B-VII.0 library with the covariance data in 44 and 56 energy group structures. Next, the model was updated to the latest library used in SCALE code system version 6.2.3; ENDF/B-VII.1 with 252 energy group and the corresponding 56 group covariance data.The uncertainty results of NUR effective multiplication factor due to uncertainties in nuclear data were determined. The top contributor is ν‾ (the average number of neutrons per fission) of 235U for all cases. The total uncertainty in keff due to uncertainties in the cross-section data was determined to be around 0.84% using 56 group covariance data. This uncertainty was higher than the uncertainty when using 44 group covariance data.