Abstract :
In many neutronic computational systems functions exist permitting the conversion of core models from 3D to 2 or 1D. Usually, as in collapsing the core model one or two dimensions are lost, the equivalence process to pass from the 3D model to the reduced one, evaluates a transverse buckling distribution that takes into account for the neutron leakage in the lost direction. The most used method to compute this distribution is based on the 3D leakage conservation. In multigroup models this method could give poor results in trying to match two models with a strongly non-uniform composition distribution. The method presented here permits the computation of a transverse buckling distribution that gives a good equivalence in these difficult configurations. It will be also shown that this method can be used to make an equivalence between two 3D models. This capability is useful in resetting computer models on measurements.
