Hacker News Reader

Comment by acidburnNSA | Hacker News Reader

Nuclear engineer here (specialty in core design/simulation). It's fun to see the Monte Carlo method be used in so many other fields now. Even in nuclear, deterministic methods are still orders of magnitude faster for most 'normal' reactor analyses on reactor configurations that are common enough to have all the important deterministic effects known. But with computers so fast, it's quite common for people, especially in conceptual design space, to use Monte Carlo methods since it's a lot easier to believe the answer once you get it.

The code MCNP is still the most common nuclear analysis code, and it's directly descended from these original codes from LANL. [1]

There is also a very powerful research code called OpenMC from ANL that anyone can run on their (powerful) computer. [2]

[1] https://mcnp.lanl.gov/reference_collection.html

[2] https://docs.openmc.org/en/stable/

View on HN

Hello fellow nuclear engineer, MCNP is also my main workhorse. Very much looking forward to the 6.3 release!

As a cool bit of additional history, check out the FERMIAC. An analog Monte Carlo device for doing neutron transport in two dimensions. https://en.m.wikipedia.org/wiki/FERMIAC

>But with computers so fast, it's quite common for people, especially in conceptual design space, to use Monte Carlo methods since it's a lot easier to believe the answer once you get it.

Is it really easier to "believe" an answer on a purely stochastical level? I'm kinda surprised, I would be way more confident (if I were to choose) with answers from deterministic descriptions/equations despite being more abstract and potentially harder to "visualize".

I find more often than not supposed 'comprehensibility' on the surface level to be quite misleading. Of course if one doesn't have clue where to start and enough processing power the Monte Carlo method and alike certainly can help to jumpstart/brute force the process.

Yes, definitely. With deterministic methods you have to make all sorts of approximations to discretize the spatial details of the fuel assemblies, the energy space of the neutrons, the angular directions of the neutrons, and so on. The approximations are complex and sensitive. With monte carlo methods you can treat all those things without approximations. Under the hood, both deterministic and monte carlo nucleonics methods are depending on the same measured/interpolated nuclear interaction probability tables (aka nuclear cross sections).

MC tends to be more physically accurate (in the limit of a large number of particle histories) because it can simulate radiation transport in continuous space, energy, and angular distribution. Deterministic methods can be faster, but the discretization process is somewhat of a dark art because the underlying distributions can be highly nonlinear and rapidly varying. A combination of methods with varying levels of fidelity are typically used in real nuclear engineering applications, and are always referenced back to a common suite of experimental benchmarks for validation.

You can write down an equation for evaluating a ray-traced picture with perfect mathematical precision, you just cannot evaluate the integral for any scene that is non-trivial. MC is the integration technique for it.

Edit: so you know exactly what you get, if you keep it simple - the gotchas start if you try to be clever and use fewer samples (biased MC)

Thanks for clarifying with examples, y'all :) I wasn't thinking concretely but on some vague principle. Now I see, of course, makes sense.

That sounds super interesting, how does one get into simulating reactor cores? I'm very familiar with MC methods from computer graphics.

An Exact MCNP Modeling of Pebble Bed Reactors:

https://www.researchgate.net/publication/264537140_An_Exact_...

OP's link was a rabbit hole (in a v. good way), sent me down some paper on the LCG random number generator used for MCNP modelling, which somehow led to that.

(https://mcnp.lanl.gov/pdf_files/la-ur-07-7961.pdf)

I would definitely start with some of the openmc tutorials.

https://docs.openmc.org/en/stable/usersguide/beginners.html

If you're asking at a higher level, you end up in nuclear engineering school after having a nebulous interest in energy issues.

Reply on news.ycombinator.com