We’ve got the Navier-Stokes equations, but solving them for turbulence feels like taming a hurricane with a napkin. It’s chaotic as hell, and we only describe it statistically, not deterministically.I ask:Is turbulence fundamentally unsolved, or do we just lack the computational power to brute-force it?Why does laminar flow break into turbulence at certain Reynolds numbers? What's actually happening at the transition?Is there a clear boundary for turbulence, or is it just a messy spectrum?
>>16554881>Is turbulence fundamentally unsolvedYes, my fluid dynamics prof said his prof told him to not to bother trying to solve it. You'll spend your entire life and still understand nothing.>do we just lack the computational power to brute-force it?Kind of, but no. What would you brute-force anyway?>Why does laminar flow break into turbulence at certain Reynolds numbers?It's not really at certain Reynolds numbers. Dimensionless numbers are kinda hand-wavy arguments when you can ignore certain terms in your equations.>What's actually happening at the transition?The nonlinear term can't be neglected anymore.>Is there a clear boundary for turbulence, or is it just a messy spectrum?Spectrum. You can get eddies at low Re, even in relatively simple geometries.
>>16554881>A fuck ton of particles moving in different directions, at different speeds, some with different masses, influencing each other differently with the added fact that a single small change can cause a butterfly effect for the whole system.Physicists: >"Yeah, I think I might be able to find a nice solution to this."Why are physicists deluding themselves into thinking this shit has a nice and neat mathematical model that will always have good solutions? Ffs I think this is so obvious that if physicists literally cant see the writing on the wall, we should cut their funding. Turbulence is unconquerable, the best you can probably do is rely on some statistical methods.
>>16555004>Turbulence is unconquerableStill it's everywhere, so you need a model.If you have a model, might as well improve it.
>>16555004>Why are physicists deluding themselves into thinking this shit has a nice and neat mathematical model that will always have good solutions?We already do. It’s called the Navier-Stokes equations. I don’t know what you mean by a “good” solution, but existence and smoothness is one of the Millennium problems. And it’s more of a mathematical technicality because numerical solutions converge for pretty much any wild geometry you can imagine.
>>16554881Messy spectrum. Chaos theory in general seems to be that way. There was an old blog on Lorenz attractors that showed how as one coefficient increased, the system would oscillate periodically, semi-periodically, then increasingly chaotically. Similarly, three body systems are famously chaotic, yet the sun-earth-moon system is mostly predictable. And with fluids, there's obviously the Reynolds number.There's clearly a threshhold that can be calculated.
>>16554891>What would you brute-force anyway?aerodynamics for videogames like Kerbal Space Program. and no, the state of the art is not "good enough"
>>16555247Good luck with that
>>16555247Ferram does a decent job. The main problem with vanilla aero in KSP is that it just assigns coefficients of lift and drag to each part independently, which leads to cheesy bullshit like clipping wings together. Ferram calculates the drag and lift of the entire vehicle and makes it a function of Mach number.
>>16554881>Why does turbulence exist?Because it likes to!
>>16555640The right hand rule.
