Shannon Entropy

User Support
1

I have modelled a full-core Enhanced Candu 6 reactor, and measured the Shannon entropy in the following figure,

Shannon Entropy

As you can see it converged within about 4 batches around 10.7. I was wondering if this is correct as this seems too fast for a full core reactor, especially since I decided to use a point source at the origin to see how long a bad source would take. I used the bounding box method, with a matrix of size 22 X 22 X 12, with setting, 500 batches, 10,000 particles/gen, and 10 gens/batch. The boundary conditions were vacuum, set at the outside of the Calandria vessel.

As an aside questions, I have seen some discussions around k-eff for depletion calculations in how they should not be too close to critical, but I was wondering what the exact reasoning, along with how close should you get. This was in older discussions on openmc, so this might be outdated.

2

What is your keff value?

3

My keff is basically critical, but I increased my matrix size by double, so 44 X 44 X 24, and increased the particles/gen to 1,000,000 while reducing gens/batch to 1 since I realized gens were essentially batches when doing Shannon entropy. This resulted in around ~20 particles per matrix, and seems to have fixed my issue. Now it took about 80-100 batches, but I think I’m going to run the simulation overnight with 500 batches to double check, because I’m not certain is has completely converged. Thanks for your comment though.

Shannon Entropy-1000000

If anyone still happens to know the answer to my second question that would be helpful, as I still don’t know if I should do depletion while close to critcal/supercritical.