MCNP6 UNCERTAINTY QUANTIFICATION APPLIED TO UAM-EXSERCISE I-1

Uncertainty quantification has proved itself to be a more and more important aspect of the nuclear society for best estimate predictions to be provided with their confidence bounds. Besides, the ability to compute the multiplication factor sensitivity coefficients for uncertainty estimation is also useful for code validation, the development of benchmark suites applicable to specific sets of applications and the design of critical (integral) experiments. In the context of the OECD/NEA project on Benchmark for Uncertainty Analysis in Modelling (UAM), the MCNP 6.1 code has been used for sensitivity and uncertainty analysis in criticality calculations for the Exercise 1 “Cell physics”, Phase I. MCNP 6.1 is the first version of MCNP with the possibility to calculate the sensitivity coefficients of the multiplication factor k for nuclear data. The methods employed are based upon linear-perturbation theory using adjoint weighting, performed in a single forward calculation by means of the Iterated Fission Probability method. The test case under consideration is the Three Mile Island Pressurized Water Reactor fuel pin in Hot Zero Power and Hot Full Power conditions. The results of the MCNP 6.1 code are presented and compared with the ones already available obtained using Serpent and SCALE 6.0 codes. In particular, the results used for the present comparison had been produced by the TSUNAMI-1D module of SCALE and Serpent version 2.1.22. It was found a good agreement between the aforementioned codes in k uncertainty, though there are differences when the sensitivity profiles are analysed. The discrepancy obtained with the three codes in the k values are also presented.