Subgroup decomposition method

This paper presents a new method which is an improvement over the consistent generalized energy condensation theory in which the energy dependence of the angular flux and cross sections is expanded in continuous or discrete orthogonal basis functions. The method extends the cross section condensation process, preserving spectral accuracy in condensed-group transport calculations in a simpler and more direct manner, without the need for this expansion in energy. The new “group decomposition” method directly couples a consistent coarse-group criticality calculation with a set of fixed-source “decomposition sweeps” to obtain the fine-group spectrum without the need to solve for higher energy moments of the flux. The method is derived in general geometry and verified with a 1D discrete ordinates lattice cell problem. In addition, one significant application of the method in whole-core reactor analysis, cross section recondensation, is explored in detail and shown to be highly effective at correcting the cross sections on-the-fly for spectral core environment effects. This is demonstrated with two 1D benchmark problems characteristic of LWR and VHTR systems.

► Method provides fine-group accuracy within coarse-group calculations. ► Method couples a coarse-group calculation to fixed-source fine-group decomposition sweeps. ► Allows on-the-fly cross section recondensation to correct for spectral effects. ► Verified in 1D discrete ordinates with BWR and VHTR core benchmark problems.