Slashdot Mirror
Ask Slashdot: Can FOSS Help In the Fight Against Climate Change?
dryriver writes: Before I ask my question, there already is free and open-source software (FOSS) for wind turbine design and simulation called QBlade. It lets you calculate turbine blade performance using nothing more than a computer and appears compatible with Xfoil as well. But consider this: the ultimate, most efficient and most real-world usable and widely deployable wind turbine rotor may not have traditional "blades" or "foils" at all, but may be a non-propeller-like, complex and possibly rather strange looking three-dimensional rotor of the sort that only a 3D printer could prototype easily. It may be on a vertical or horizontal axis. It may have air flowing through canals in its non-traditional structure, rather than just around it. Nobody really knows what this "ultimate wind turbine rotor" may look like.
The easiest way to find such a rotor might be through machine-learning. You get an algorithm to create complex non-traditional 3D rotor shapes, simulate their behavior in wind, and then mutate the design, simulate again, and get a machine learning algorithm to learn what sort of mutations lead to a better performing 3D rotor. In theory, enough iterations -- perhaps millions or more -- should eventually lead to the "ultimate rotor" or something closer to it than what is used in wind turbines today. Is this something FOSS developers could tackle, or is this task too complex for non-commercial software? The real world impact of such a FOSS project could be that far better wind turbines can be designed, manufactured and deployed than currently exist, and the fight against climate change becomes more effective; the better your wind turbines perform, and the more usable they are, the more of a fighting chance humanity has to do something against climate change. Could FOSS achieve this?
The easiest way to find such a rotor might be through machine-learning. You get an algorithm to create complex non-traditional 3D rotor shapes, simulate their behavior in wind, and then mutate the design, simulate again, and get a machine learning algorithm to learn what sort of mutations lead to a better performing 3D rotor. In theory, enough iterations -- perhaps millions or more -- should eventually lead to the "ultimate rotor" or something closer to it than what is used in wind turbines today. Is this something FOSS developers could tackle, or is this task too complex for non-commercial software? The real world impact of such a FOSS project could be that far better wind turbines can be designed, manufactured and deployed than currently exist, and the fight against climate change becomes more effective; the better your wind turbines perform, and the more usable they are, the more of a fighting chance humanity has to do something against climate change. Could FOSS achieve this?
You get an algorithm to create complex non-traditional 3D rotor shapes, simulate their behavior in wind, and then mutate the design, simulate again, and get a machine learning algorithm to learn what sort of mutations lead to a better performing 3D rotor. In theory, enough iterations -- perhaps millions or more -- should eventually lead to the "ultimate rotor"
You're describing Genetic Algorithms. It's a fairly old technique. It shouldn't be too hard to implement it. The problem here is not FOSS, it's computational power. You need quite a lot of CPU time to run all the simulations and evolve the solution.
Some sort of distributed computing framework like INSERT_PROJECT_NAME@home would work. But then you'd have to convince everyone to use it....
"Traditional" turbine designs are already up to 80% theoretical maximum efficiency. Trying to eke that last 20% is not really going to save the planet since we're nowhere near using that much wind in the first place.
That is - if you want to get FOSS to improve tech adoption, direct it to making things more affordable or accessible, not toward having more expensive higher-efficiency, higher-complexity devices.
"There are a dozen opinions on a matter until you know the truth. Then there is only one." - CS Lewis (paraprhase)
If you want to save the world through free and open source software, there's an easy way to do this: stop building systems that waste resources.
Don't use programming languages that spend 10 CPU cycles to do 1 cycle of work. Don't arrange things so a program is recompiled every time it is run. Write software that uses less RAM. Write replacements for spyware-laden crap. Do not support battery-burning DRM and tell them why. Encourage wired rather than wireless connections.
Stop thinking like a coder and start thinking like an engineer.
sub f{($f)=@_;print"$f(q{$f});";}f(q{sub f{($f)=@_;print"$f(q{$f});";}f});
We know how to make cheap plentiful power, we lack the will to do so. The same people that are running around yelling global warming also hate it. Fission works it's got far too many regulations to do cheaply. It gots far to many court delays to finance. Everyone in the US is a bespoke design but all 70's level tech. This is all by design you need far to much political capital to get one put in.
Build them in factories with a design that's not 40+ years old. Hells we can use the spent rods as feedstock for modern designs.
Carbon taxes are crap it's just a tax and a regressive one at that. Want change put in sensible PV incentives and billing. The entire concept of taxing things you don't like people doing is broken, an end run around the constitution to allow nearly unlimited federal power.
No sir I dont like it.