Consistent comparison of Monte Carlo and whole-core transport solutions for cores with thermal feedback

For systematic and consistent comparison of Monte Carlo and whole-core transport solutions in various core states including power generating conditions, a test problem set that spans from two-dimensional uniform temperature pin cell problems to three-dimensional core problems involving thermal feedback is solved by a continuous energy Monte Carlo code MCCARD and a multigroup whole-core transport code DeCART. The neutron spectra, k-effective, pin-wise power distribution, fuel temperature distribution, and Doppler coefficients obtained from the two solutions are compared taking the MCCARD solution as the reference. For the uniform temperature problems, excellent agreement between the two solutions is observed in every solution aspect. The pin power distribution error is less than 1% and the k-effective error is within 100 pcm in most cases. For the problems with thermal feedback, the discrepancy becomes larger, yet within the tolerable range. In the hot-full-power mini-core calculation, a maximum of 3.7% error in the radial pin power distribution and the k-effective error of about 260 pcm are observed. Through this comparison, it is demonstrated that accurate multigroup direct whole-core calculations are possible even at power generating conditions with a much less computing time than the corresponding Monte Carlo calculations. (authors)