A spatial rehomogenization method in nodal calculations

Infinite medium flux weighted cross-sections used in nodal calculations enable equivalence with the corresponding fine configuration if the following condition is satisfied: the flux shape inside the assembly in the core is close to the infinite medium flux shape (computed in lattice calculations). In presence of big flux gradients this condition is not satisfied and the absence of information about cross-sections distributions inside a node does not permit to predict the reaction rates with the same accuracy attained in ordinary situations. This tendency is amplified in case of high heterogeneous regions where tilting the flux causes big changes in reaction rates. The method presented here uses information coming from the lattice calculations that produced the homogenized cross-sections, in order to predict the right reaction rate even in presence of high tilted flux shapes. This is done in evaluating a variation of the cross-sections equivalent to the variation in reaction rate, but no variation is applied to the discontinuity factors. The accuracy of the method and its limitations are shown in several significant configurations. Its implementation in the Areva NP reactor core simulation system SCIENCE has shown better evaluation of control rod worth in comparisons with experimental results.