Validation of a new MCNP-ORIGEN linkage program for burnup analysis

摘要

The analysis of core composition changes is complicated by the fact that the time and spatial variation in isotopic composition depend on the neutron flux distribution and vice versa. Fortunately, changes in core composition occur relatively slowly and hence the burnup analysis can be performed by dividing the burnup period into some burnup spans and assuming that the averaged flux and cross sections are constant during each step. The burnup span sensitivity analysis attempts to find that how much the burnup spans could be increased without any significant deviation in results. This goal has been achieved by developing a new MCNP-ORIGEN linkage program named as MOBC (MCNP-ORIGEN Burnup Calculation). Three kinds of coupling schemes have been implemented in MOBC. Two of which are based on second order predictor-corrector technique and enable us to choose larger time steps, whilst the third one is based on Euler explicit first order method and is faster than the other two. The validity of the developed program has been evaluated by code vs. code comparison technique. Two different types of codes are employed. The first one is based on deterministic two dimensional transport method, like CASMO-4 and HELIOS codes and the second one is based on Monte Carlo method, like MCODE code. Only one coupling technique is employed in each of these state of the art codes, while the MOBC excels in its ability to consider all of the schemes. Calculations were made for a well known high burnup UO_2 lattice benchmark. The results showed that the average absolute differences in eigenvalue are about 1.2, 1.0 and 0.5 in comparison with CASMO, HELIOS and MCODE respectively. The sensitivity analysis showed that the standard predictor-corrector approach better accounts for nonlinear behaviour of burnup effect. In summary, it is sufficiently good to use the first order Euler method at burnup spans smaller than 10 MWd/kg but for larger spans use of the second order scheme is mandatory.