Nuclear data uncertainty analysis using the coupling of DRAGON with SUSD3D

In this paper we present the work performed to make DRAGON (Ecole Polytechnique de Montreal's transport code) and SUSD3D (the nuclear data sensitivity and uncertainty code maintained by OECD) compatible in such a way that extensive nuclear data sensitivities can be evaluated using Generalized Perturbation Theory (GPT) methods. The qualification of this coupling on the Uncertainty Analysis in Modelling (UAM) benchmark is the aim of this paper. We compare the results obtained with this route with TSUNAMI-1D and 3D and with other routes such as the coupling of XSDRNPM to SUSD3D for keff calculations at the cell and assembly levels. The good consistency between the uncertainty results obtained by very different tools tends to demonstrate that the coupling is effective for keff nuclear data uncertainty calculations. The route proposed here gives uncertainties for criticality that are very compatible with TSUNAMI results. The capabilities of DRAGON concerning Generalized Adjoint Function open the field for very broad application range of S/U analysis in the future. (authors)