How OPENMC generates accurate multi-group constants for use in deterministic programs

I am also using OPENMC to generate multigroup constants for use in deterministic programs, but after I generate multigroup constants for different components of a core, I find that the results of the multigroup constant calculations and the results of the continuous energy monteka calculations differ significantly. I did the calculations with a KUCA stack and used different components inside KUCA to generate the few-group constants, but the results were not good. I guess it could be because the same components are in different locations in the core, which makes the spatial effect of the components different, but I am not sure if this is the reason, if it is because of the spatial effect, then we have to generate a set of multigroup constants for each component, so the input file of the determinism will be very complicated.

How to divide the few group constants in different regions of a complex core like this one to be consistent with the calculations of the continuous energy montages?

Best regards

Hello, I have done some work related to this. One major factor is choosing the right group structure. Are you sure that the group structure you are using is good for the reactor you are modeling? Another aspect is which Legendre scattering expansion order you use.

You can take a look at my project’s code and the associated conference paper for further examples of generating multi-group parameters: GitHub - lanl/TOGA: Tool for Optimization and Group-structure Analysis of nuclear multi-group cross sections

By the way, if your deterministic method is diffusion, you should keep in mind that Monte Carlo does not do a great job of tallying diffusion coefficients. In OpenMC, you should tally the diffusion coefficient using a very fine energy grid, and then condense the diffusion coefficient to the energy grid you want to use. You can read more about this method, compared to the state-of-the-art cumulative migration method here: https://doi.org/10.1016/j.anucene.2017.10.039

3 Likes