Sensitivity and uncertainty assessment associated to burnup calculations - 064

The theoretical background of sensitivity and uncertainty calculations associated with burnup calculations has been studied in depth. In general a burnup calculation scheme consists of consecutive sets ('steps') of spectrum-, cross section collaps- and one-group burnup step calculations. It was found that in order to be able to calculate uncertainties in final nuclide densities for such a scheme it is necessary to calculate, for each single step, the sensitivity of the nuclide densities at the end of this step with respect to uncertainties in the microscopic (one-group) cross sections and the nuclide densities at the beginning of the step. Hereby it should be noted that the flux/spectrum may be different for different steps. The sensitivity matrices for the consecutive steps are then combined into a single 'one-group cross section-to-density' sensitivity matrix of the entire scheme, which can then be used to calculate the uncertainties and covariance matrix of the final nuclide density vector. The applicability of the method is demonstrated by calculations of BR3 MOX pin irradiation experiments employing multi-group cross section uncertainty data from the EAF4 data library. Although not all differences between measured and calculated densities can be explained by the propagated cross section uncertainties, the comparison still demonstrates the feasibility of the approach presented here. (author)