Nuclear Science and Engineering / Volume 180 / Number 3 / July 2015 / Pages 286-300
Technical Paper / dx.doi.org/10.13182/NSE14-92
Four schemes for coupling the neutronics and depletion in burnup calculations are compared in four assembly segment test cases with various step lengths. Three of the coupling schemes use only one transport solution per step. While none of the methods was superior in every test case or in every respect, there are significant differences that can make one or the other preferable in different applications. The fourth method included in the comparison is the one dubbed CE/CM in our previous study, which compares schemes that use two transport solutions per step. The methods using only one transport solution per step were found to be more accurate than CE/CM but less accurate than the newer LE/LI and LE/QI methods. In cases where desired output intervals, rather than accuracy, are the limiting factor for step lengths, methods using only one transport solution per step can still provide a major advantage even when compared to LE/LI and LE/QI. Significant differences were also observed in the propagation of the statistical uncertainty from Monte Carlo neutronics through the different methods. While this topic was not studied further, it seems that differences in error propagation may in some cases be as significant as those in accuracy.