Skip to main content

I aten't dead

Here's a quick status update; I have been meaning to post these more often.


I'm tinkering away and playing bug whack-a-mole with PlasmaPy #1059, the pull request that's going to become my masters thesis. Basically, it implements neoclassical transport calculations - solves for particle and heat flows - in axisymmetric plasma configurations. It's rather basic, but it's a start.

Example flux surface

Here's a sample flux surface - I will soon-ish write a note that explains this concept. In a nutshell, though - I'm splitting the domain (a 2D slice through a tokamak) into these circular flux surfaces, a radial grid:

Example flux surface grid

And then I calculate how much flow there is between them. I still have a good bunch of problems with it:

  • I'm using astropy.units, from which I infer that my formulae have issues, as each flux comes out with a different physical unit attached. And I'm still not quite sure why.
  • I still have to figure out a decently nice way to input the quantities (density and temperature) that, together with the magnetic field, determine this transport.
  • I don't understand differential geometry yet, and it really would have been useful here. But I'm picking it up, slowly.

The rest

Whoops, I'm writing this on a timer as an exercise and that one's rapidly running away. Okay, here goes.

  • Part 2 of the series on fusion is coming soon; I'm using it as a way to cobble together a first draft of my actual thesis introduction.
  • PlasmaPy's coming along nicely, though I'm mostly knee deep in the neoclassics PR. There have been some cool new developments with documentation and speeding up our formulary functions. We're also trying to figure out how to host essentially a hackathon. Tips appreciated!
  • I've been getting back into reading books, recently, and really enjoying it. I'll do a write up of my current system (because of course I have a system, for it, what did you expect?) in a few weeks.
  • Speaking of systems, we recently sat down together with Paulina, realized our way of consistently keeping the flat clean... wasn't at all consistent, and figured out a system based on a python script and a shared yaml file with dates of when we last did X thing (say, vacuuming) vs how often do we want to do that particular X. The ratio of time since last occurrence to expected period is then a metric than we can actually try to keep down. As you have probably guessed by now - write up coming up! But probably not soon.

All right, that's in for this week! I will try to be back here next week with another status update, if I don't actually manage to get fusion2 out. Time will tell! In the meantime, stay safe!