>> "Basically a lot of this work is master's or PhD-level research."
Without taking out from this big accomplishment, there's no way for this to be true. In general, when they teach simulation to university students , they tell them to use all the math/physics they've learn to validate their simulation model(to avoid the known garbage-in garbage-out problem - GIGO).
I'm sure he didn't learn all those tools just to validate the simulation - it takes a really long time to acquire them, and it's really hard.
So he probably used conceptual understanding of fluid dynamics +learning the simulation tool + a bit of guessing/forums to build his simulation model in order to design his invention.
And maybe ,educationally that's a good way to go ? maybe we should ,at high school, teach people the conceptual skills, and the use of simulation tools(the higher level simulation tools they find useful) and let them play ? could generate lots of excitement for the sciences and probably good ideas.
Same goes for creating easier simulation tools.