The Analysis Of Influencing Factors In Neutron Transport Problem Calculation Using MCNP

With MCNP applied increasingly and broadly in the aspects of the nuclear science and engineering, analyses of the influencing factors in the computed results using MCNP have recently received more attention from areas such as computations of reactivity changes caused by the minor changes of the reactor state and benchmark computations of the neutron flux of the reactor pressure vessel. To serve these needs, analyses of the influencing factors in the computed results using MCNP are present in the thesis.After comprehending the physics of neutron interaction being the essence of MCNP code and the method about how to use MCNP, this paper sets up computation model basing on two typical benchmark problems: the criticality computation of the MOX fuel research reactor and the flux computation of the reactor pressure vessel. During the process, there are the operative skills advanced in the paper. Finally, the most important work that analyzes the influencing factors in the computed results has been finished.The criticality computation of the MOX fuel research reactors firstly calculates the effective multiplication factors and the relative power distributions. And secondly it proves that the computed results rerunning the problems with a different random number sequence always have certain statistical fluctuation that reduces gradually along with the number of history increasing by the way of changing the initial fission source (SRCTP) from a previous MCNP criticality calculation. As following, under the condition of the statistical fluctuation weakly enough, the difference between the computed results using different data libraries is discussed, it pointed out that using S(α,β) thermal scatting treatment is the essential condition to obtain the appropriate effective multiplication factors.In the neutron flux computation of the reactor pressure vessel, the statistical fluctuation existing in computed results and the influence of the data libraries being on the computed result are further confirmed. There is the discussion about the reasons of difference between the computed results using different variance reduction techniques and the influence of procedure edition on the computed results. Finally, it indicates that the computation process is correctly according to the fast flux relative distribution of the pressure vessel internal surface presenting in the benchmark problem.