Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
The Birmingham Blade: geographically tailored urban wind turbine designed by AI (birmingham.ac.uk)
28 points by taubek on Dec 7, 2024 | hide | past | favorite | 47 comments


"... a solution to one of the most pressing issues in the green energy landscape – how to produce small-scale, affordable, generators of clean wind energy."

Wrong problem, unless you're in an isolated area. There are already good wind turbines for isolated plants. That's more a ruggedness problem than an efficiency problem.

There are good reasons wind turbines have become huge. The big ones are far more cost-effective, partly because wind speeds are higher around 100-200m from the ground. 1970s wind turbines were in the 10KW-20KW range. Pacheco Pass in California had row after row of them. Now, 1MW is small, and the little ones have been removed.

Here's a video of the thing.[1] No info about how much power they get out, but it's probably under a kilowatt. Three standard solar planels will produce more than that, and would fit in about the same ground space.

This is a decorative object, not a power source. It's in the same category as energy-generating sidewalks.[2] These are Rule of Cool devices, not useful energy sources.

[1] https://www.youtube.com/watch?v=StLyTZazgVE

[2] https://www.archdaily.com/911965/sidewalks-that-generate-ene...


I agree, however our 5kW solar roof in southern England has barely generated 1kWh/day since October. Apart from that still period in November we've got wind, generally 10-15mph at roof height.

A feasible home wind turbine would be a good addition in winter.

Or buying a share of a massive wind turbine, but the maths of Ripple Energy have never stacked up when I've looked at it.


Looking at urban landscapes in Europe in general I expect diminishing returns with everyone placing a useful windmill. And it would be ugly as hell no matter what (imagine every roof having one of these monstrosities).


Pop culture and mainstream politics affect people’s opinions in a weird way. Only in the USA and some part of the western world, people see wind turbines as “ugly”. Coming from another part where wind turbines are picking up very fast, the words I have heard are “amazing” and “elegant”.


Until you're living near one of them, and constantly hear the low thump, thump, thump their blades make when passing lowest point, and as such their very own mast, by necessity.

At varying speeds.


Indeed. I regard wind turbines as striking kinetic sculptures in the landscape.


Hello from 50-55 degrees latitude where we get our heads blown off by the jet stream all. the. time.


Yours might be HN's first comment from South Georgia!


I’ll learn to not mix those two sometime in the next decade. Pinky promise


From what I hear the showstopper for home wind turbines is the noise. Even the quietest ones produce some sound, which at your house when you're trying to sleep is annoying.

I'd imagine things with moving parts need alot of maintenance too.


Make them sparkle and glitter. Can't be more ugly than the forests of countless antennas for terrestial TV, which were not that uncommon not too long ago, or the satellite dishes pointing south to GEO.


Agree on Ripple Energy especially for the way the tax worked on it.

Their solar offer seemingly made even less financial sense, due to the very high cost of paying for your slice of the huge costs of running the national grid. The only reason for going with it would be environmental (effectively subsidising sustainable energy on the grid to displace other sources)


> It's in the same category as energy-generating sidewalks.

I met a guy who did one of those. Huge scam artist vibes, but people absolutely threw greenwashing money at him and could not stop writing articles.

Obviously no concrete numbers in his flashy "decks" because they were all completely terrible and made no practical sense if you actually did the numbers (watt-minutes per day if I recall).


Thanks for including a video. The original article fails to even add a picture of the thing.

I'm most sceptical of the storm rating of this rig. Putting large rotating sheets on city roofs seems like a hazard nobody would want to take up for a few watts.


Fortunately it's usually cheap enough to tie them down when the time for the second round of maintenance comes. The rooftops of high status building projects in regions prone to greenwashing already are a cemetery of abandoned microturbines of various designs.


It's includes the picture with the creators. Turbine is in the background. But it is of such unconventional form that i guess no one understands that this is the turbine (including me)


I saw that picture and thought it was a test size model, not the real thing.

It's conventional. There are lots of "how I built a small vertical wind turbine from scrap parts" articles and videos.[1][2] This is easy to do and worth it for off-grid locations. Most people just cut blades out of some cylinder such as a metal or plastic barrel. The Birmingham people just optimized the blades a bit.

Note the lack of a direct comparison with a simple blade form vertical wind turbine.

Nothing wrong with doing this, but the hype is excessive.

[1] https://www.youtube.com/watch?v=G-WkJI5TtmQ

[2] https://www.youtube.com/watch?v=ti9FkEg8G_Y


Why should we grant the editor any slack? Out of focus and partially occluded is of no use and does not count as picture of the thing.


Looked like just another https://en.wikipedia.org/wiki/Vertical-axis_wind_turbine (VAWT) to me, the blades not that different from https://en.wikipedia.org/wiki/Darrieus_wind_turbine in so called 'helical' form. But hey, such AI, much WOW!


I'm not sure the product is the real story here. It's probably just a way to sell their design process.

I have no way of judging how valuable it is, but achieveng 7x improvement in efficiency of this wind turbine, and the team this company has[1] may hint their process could be quite valuable.

[1]https://evophase.co.uk/team/


I am deeply sceptical of the 7x claim. It's easy to find a great improvement when you chose a bad baseline. For example if you just plonk a traditional fan-on-a-stick in the same place, it will probably be terrible, but there are non-AI upright designs that might be in the same ballpark. I don't doubt they have a way to make substantial site-specific design adjustments, but I do doubt that much improvement over the most suitable but not customised design you can already buy today.

Also the AI angle is pretty silly hype because multidimensional optimisation techniques have been part of the industrial design toolbox for decades. Wind turbine companies are very keenly aware of optimisation algorithms for squeezing everything out of a blade geometry. CFD is an entire field for a reason.

Random 2006 article in numerical optimisation of vertical wind turbines from a careless search: https://iawe.org/Proceedings/CWE2006/MC3-01.pdf


Thanks for the video link. I attended Birmingham University (long ago). That looks like the Muirhead Tower in the background. I mention that only because the passage underneath that building was a well known wind tunnel, presumably due to poor 60s building design. But it would actually be a good place to put that wind turbine given the gales that the building generated.


> No info about how much power they get out, but it's probably under a kilowatt. Three standard solar planels will produce more than that

Not in Birmingham though…


Why would you release a press write up of a new blade design and not include a picture of the blade design?


It's designed to be invisible and include itself into the geographic conditions :)

You even didn't notice it. Job well done!


I think the thing behind the group photo at the top is it.


First I thought your comment was a joke about the clock tower, but then I realized the blade was the 6 dark gray rectangles just behind the team. (I initially thought the rectangles were a fountain or sculpture or something.)


Urban wind turbines seem like a terrible way of producing energy. Low wind speeds, lots of people to disturb with noise and flicker, turbulent flows due to buildings. Hard to see how a bunch of these could ever compete with a single gigantic 6MW turbine out in the countryside


They are terrible. Source: I worked for one of the UKs highest profile and now bankrupt micro wind generation companies for years.

Pros: Less need for moving power a long way. (This is a lie because you can't generate enough power in the urban environment anyway.)

Cons: Maintenance is much more expensive. Less wind. More turbulence.


Hah, thanks for confirming my suspicions with professional experience.


> turbine designed by AI

I wonder what definition of AI they are using. It seems anything that uses mathematics is now AI.


Probably some sort of optimization algorithm that's been defined for centuries, and comes standard with any mathematical computing tool. The kind of thing used every day by engineers as routine, but I guess that's "AI" now.

> import scipy.optimize

Look ma! I made an AI!


Some sort of genetic algorithm would be my guess.


Seems like it - From the article:

> Our evolutionary simulations have confirmed the Birmingham Blade is up to seven times more efficient than existing designs in Birmingham’s wind speeds and urban environment.

Pretty good idea.

From what I've read, they iterated through 2000 simulations over the course of a few weeks.


Cool stuff for sure! But cheeky comments like this (and others I personally have made in this thread) are to do with the "AI" branding aspect of a thing that is just bog standard mechanical engineering work. Hooking up optimizers to simulation work is far from new. But that's being called "AI" now to drive news article clicks (and presumably to fish for investor capital).


> Hooking up optimizers to simulation work is far from new. But that's being called "AI" now

It that being called "AI" just now, or has it been called "AI" since the beginning? A lot of the optimization algorithms we take for granted nowadays originated from AI research.


Genetic algorithms were being called AI back in the 90s.


As per the article, it needs to be called AI to qualify for a pot of government funding.


It's buzzword bingo, gets clicks.


What's the right definition of AI?


The definition of AI changes depending on what I am trying to sell you today.


All wind turbines must meet the criteria of bieng extream service 100% duty cycle machines. The competing technology is solar pv, which by bieng a solid state generator with no moving parts is winning ,hands down in all small scale instalations, other than places where there is not enough light, and very very plentiful wind, artic, ant artic, certain mountain research stations and some marine/island/extream rugged coast areas. Whatever scam is in the wrks in Birmingham, should cost someone, there tenure....


"Birmingham’s relatively low wind speeds "

Is there a clue in there somewhere?

IMHO we should carefully pick our climate/energy fights, and install whatever tech is most appropriate, even if that's 'nothing, here, build a transmission line to the irish sea offshore wind farms instead'.

I do like the cleverness of custom parametric solutions, but also hate them for their specific, non-reusable, design-overhead aspects. Just deploy a few more cheap, modular, repairable, replaceable, generic units, instead of the costly custom designs.


Ah Birmingham. A place full of simple well meaning folk, where little is done, and a few misplaced geniuses trying to make ends meet. Much lovelier people than in london, but a backward place full of well intentioned mistakes. Sort of like the play area of an old fashioned rest home. Really relaxed, but nothing going on. A purgatory if you will.


The article didn't do enough literature review, this isn't a new idea: https://www.sciencedirect.com/science/article/pii/S096014811...


Small scale wind would be ideal for winter in our remote northern place. We have the space. Except the energy returns aren't there, and maintaining the moving parts is not trivial.


What does AI mean in this context? Is it more than running scikit-optimizer on the model?




Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: