Airborne wind turbines

magenn power

Figure 1: Airborne Wind turbines could be a major game changer…..if they work! [Credit: Magenn power, 2008]

I recall someone asking a while ago for a review of airborne wind turbines (often known by the acronym AWES as in Airborne Wind Energy Systems), which I never got around to doing. But no time like the present.

So, what is the point of AWES? After all wind energy is already the cheapest of renewables (perhaps according to some the cheapest overall means of electricity generation), so why re-invent the wheel? Well because the capacity factor of conventional wind turbines is around about 33%. Capacity factors can be higher for some wind turbines offshore, but it can also be lower for those further inland (and not all countries have coastlines!). By placing a wind turbine higher up we gain the benefits of more constant winds. Its suggested that capacity factors closer to 60% would be possible, perhaps even 80% in the jet streams. Also wind speeds tend to increase with altitude. Go up a mountain and the strong breeze at the bottom can be a howling gale you couldn’t walk against at 3,000 ft. So a higher wind speed means more power for a given turbine size.


Figure 2: Wind Speeds with altitude

Also by putting turbines in the atmosphere we get to stack them in three dimensions. There is much discussion as to how much energy can be usefully derived from the Earth’s winds. Some estimates are relatively low, others much higher. AWES’s however change the game by allowing us to tap into much more atmosphere than simply the thin crust of the lower atmosphere.

Its worth pausing to define the capacity factor as the rated capacity of the turbine over time e.g. a powerplant that gets turned on at full power for 3 months each year (typical for a peaking load plant) would have a capacity factor of 25%. This does not mean the turbine generates nothing 70% of the time, actually it will usually be generating some power, just not working at its rated capacity. Also the grid will not be drawing on its full power load all of the time (hence why peaking and intermediate load plants have much lower capacity factors).


Figure 3: Various types of AWES designs have been proposed [Credit: Gizmag,2013]

Various different types of AWES have been proposed, a recent review of the technology is available from Cherubini et al (2015). These generally break down into two general categories, Fly-Gen systems (where the generator is carried aloft) and Ground-Gen (the generator stays on the ground). Fly-Gen systems can involve large balloons with the turbine either mounted inside (to create a funnel effect) or on the outside (larger diameter, hence larger swept area). Others operate more like a giant kite, with an attached turbine. Ground-Gen systems typically involve the use of kites (rigid or otherwise) flying at altitude while connected by a tether to the ground, which creates a rotational torque creating power. A variation on this idea is to use such kites as a means of propulsion for ships (so called Sky sails).


Figure 4: The operation of a Fly-Gen type AWES

There are several factors which determine the power output of a wind turbine. The wind velocity (to the cube, so a doubling of wind speed will increase 8 fold the power output), the swept area (more area, means a larger mass of air is capture, hence more power output) and the air density. The denser the air, the more power can be extracted by it. Unfortunately as you go up in the atmosphere the density of air falls. At 1,000 metres the air pressure is 90% that at sea level (so a 3 MW turbine at this height will only generate 2.7 MW’s) and at 5,000 metres its about 50% sea level pressure (so our 3MW turbine now only generates 1.5 MW’s). Also any tether will be creating drag and the higher the operating altitude the greater the drag.


Figure 5: Increasing wind speed leads to higher energy yields…..[Credit: Kitves (ND)]

So there’s a trade off in terms of how high a turbine is positioned and falling air pressure. This is why most current designs plan on operating in the hundred’s of metres or perhaps just over 1,000 metres. Of course this goes back to the argument about maximum wind energy extraction, as it means that we can only tap a small portion of the atmosphere’s wind energy. Certainly an order of magnitude larger volume would likely be possible than with conventional turbines, but not quite as large a reservoir as previously suggested. Even so even the most conservative estimates (Marvel et al 2012) significantly exceed current global power demand.


Figure 6: …..But higher altitude means lower air pressure [Credit: Engineer’s toolbox (ND)]

Average wind speeds more or less double between sea level and 1,000 metres (so the density of power output from the turbine increases almost 8 fold, as discussed). It should be noted however this doesn’t mean that the rated speed of a turbine for a given diameter falls (this factor is determined by physics), its just the same size turbine will be exposed to a higher average wind speed and generate much more power and hence should have a higher capacity factor than if we left it on the ground.

But what about down time? All of these AWES include a tether mechanism. That will need regular maintenance, which might well mean landing the turbine. And as Cherubini et al (2015) discuss, there may be issues with landing and take off procedures, particularly in incremental weather.

Also one has to question whether the tether will be designed to withstand stormy conditions, e.g. hurricane strength winds. Normal turbines simply feather themselves out of the wind, if the wind speed is too high, then restart once it drops to a safe level (and its worth remembering that this is generally only needed for the period of peak wind speed, during the bulk of the storm they can keep generating). AWES would have to land and stay down until winds dropped to a safe launch speed. Also how would they ride out the storm on the ground? Many strike me as not being very sturdy. Likely they’d need to be hangared on the ground (normal for airships and kite like aircraft), which adds an additional cost.


Figure 7: Wind speeds worldwide [Credit: Kitves (ND)]

One solution to the above is to only mount these turbines in regions of the world where we don’t see such severe storms on a regular basis. These tend to be around the coasts of the major continents. Moving them inland would thus reduce down time due to storms, as we’d see less severe weather in terms of damaging high wind speeds (perhaps negating the need for hangars). However, overland in the middle of continents tends to mean lower average wind speeds. While we’ll get higher wind speeds at altitude (and again doubling speeds at altitude means an eight fold increase in power output), but it won’t be quite as great a pay off as we get on the coasts. And again if we only mount them in such locations, we greatly reduce the effective area where such turbines can operate, so on a global scale we are reducing the total amount of power we can extract from them.


Figure 8: AWES may be transported by truck…until they reach a low bridge!

Another factor is that of average wind speeds. Increasing it certainly does mean higher power output (for a given altitude) however this also means the turbine needs to be designed to withstand these more intensive operating conditions. Modern wind turbines are largely overdesigned (one of the reasons for recent price drops being engineers finding ways of reducing the level of overdesign, yet still retaining a reasonable level of reliability) but they don’t have to fly! Designing anything to fly often means paring down the mass to the bone, which in simple terms, will generally push up the costs for the same size of turbine.

Some costs savings are possible ,on the lack of a tower for example, a more constant wind speed would allow for some savings in terms of the gearbox and generator design, but none of this is going to produce a huge saving. An airborne turbine with a given rated power output is, I suspect, going to inevitably have a higher installation cost than a comparable ground based wind turbine. Maintenance costs are also likely to be higher. Although in both these cases, its difficult to be specific as the designs are not sufficiently advanced to give a firm idea of what the costs will be. As regards the overall levelised energy cost, that should work out lower for an airborne wind turbine given the higher capacity factor.


Figure 9: The cost breakdown of a conventional wind turbine [Credit: GTMS, 2013]

Exactly how much lower however is difficult to assess. This brings us back to the debate regarding wind energy and its low capacity factor. The critics (often nuclear energy supporters) of wind energy will tend to pile on masses of costs, often insisting on building an entire fossil fuel plant and all its associated costs to compensate for the intermittency of output. However, such simplistic analysis often ignores the fact that all energy sources require some level of backup.

Furthermore the load they are supplying isn’t necessarily drawing on power all of the time, so in any fossil fuel free grid there will be a need to funnel some significant quantity of energy away and store it, regardless of what energy sources we rely on (be in nuclear, CSP, geothermal or intermittent sources such as wind and PV). Certainly a grid which is heavily dependant on intermittent wind or PV will have a larger storage requirement, although exactly how much more is a matter of some debate, as I’ve discussed before.

Certainly, with regard to AWES, one has to comment that research is mostly still at a non-commercial stage so it is very difficult to speculate as to the exact costs, but of course that means they could go either way. Even if airborne wind turbines work out as somewhat more expensive than conventional wind, the advantages they bring of being able to tap into a much larger wind resource and a more even power output with greater capacity factors will mean they offer some benefits. But, like I said, I would argue that the balance of probability is that airborne wind turbines will have higher install costs, but a lower life cycle cost (although like I said, that’s unproven), although how much lower (assuming they do have a lower LOCE) will depend on how much they reduce the need for energy storage.

As with any technology there will be some level of overshoot with AWES. One topic that often comes up is the idea of placing them within the jet streams. Here they would benefit from a very high wind regime and a capacity factor that is higher still, perhaps as high as 90%. However this would mean operating AWES at a significantly higher attitude 9-12 km’s height in fact (thus a much lower air density, i.e. less power output for a given wind speed) but a higher wind speed (+40 m/s v’s the normal rated speed of 14 m/s for a ground based turbine). Needless to say, operating turbines in this regime would be even more of a challenge than simply a few hundred metres off the ground. Such a turbine would represent a significant engineering challenge to design, although certainly not impossible (aircraft do this all the time remember).


Figure 10: The jet streams [Credits: NASA & NOAA (2013)]

Also the jet streams only cover a relatively narrow corridor of air. If we apply the conditions above (i.e. only tapping into them over land in central continental areas) this greatly reduces the locations where such turbines could be mounted. So only a handful of countries, most notably the US, Canada, China and several central countries, would be able to benefit from them. Even so its worth noting that the theoretical power densities of the jet streams are in the order of 3,000 – 4,000 W/m2 well above anything achievable for conventional wind or even proposed AWES. So you won’t need to cover a very large area with them to extract a lot of power. Although its also worth noting that the jet streams do move, so the turbines would need to do so also (yes I’m saying wind turbines will need to be migratory….so no African swallows then). I would argue that such projects are somewhat speculative at the moment, as its simply too early to say whether it would be possible. Certainly I don’t see anyone such projects (on a commercial scale) within the next decade or two.

All in all I think that we can conclude that AWES might be able to provide a source of wind energy that is more regular and less intermittent. However they are likely to come with a number of strings attached (if you’ll pardon the pun). They may not be suitable in all locations, hence they will likely compliment rather than replace existing wind energy technology. The install costs will probably be higher than with existing wind energy, although the levelised cost of energy might well prove to be lower (but that remains to be proven). However the technology isn’t sufficiently mature for us to assess that at present. They may well reduce the levels of energy storage needed, but as noted, a substantial level of energy storage will be needed in any future low carbon grid, regardless of what options we take.

In terms of the overall energy picture AWES will increase the amount of renewable energy that can be extracted from the winds, although exactly how much is difficult to access at present, as this the main constraint on the extraction rate of renewable energy is likely to be economics.


About daryan12

Engineer, expertise: Energy, Sustainablity, Computer Aided Engineering, Renewables technology
This entry was posted in clean energy, energy, future, power, renewables, sustainability, sustainable, technology and tagged , , , , , , , . Bookmark the permalink.

10 Responses to Airborne wind turbines

  1. John Harding says:

    Not a word on how these devices will interfere with air traffic.

    • daryan12 says:

      Good point. Presumably you’d need to put a controlled air space box around the turbines and keep aircraft out. Keep in mind that they would generally be flying at a maximum of 3,000 ft (and often lower than this) while commercial aviation tends to operate much higher, typically 30,000 ft.

      There were some concerns and objections to conventional wind farms by the military in some European countries (during the early days), as their aircraft do fly low enough sufficiently often that a collision with a turbine was seen as a realistic concern. I have yet to hear of an actual collision occurring and that issue seems to have gone away.

      In short, so long as pilots obey the rules and read their maps there should be nothing to worry about. Of course perhaps there’s the problem, what if they don’t?

  2. It seems to me that the majority of these suggested designs rely on a wind generating device on the end of a string. Has anyone considered using the tether element as the generator? I don’t mean the ones with surface based generators, but a system that generates energy using the actual tether.
    I visualise a carbon fibre ribbon say a metre wide, hollow but naturally flat, but with “bubbles” of air being allowed in at intervals. These bubbles would deform the ribbon and the shape formed would be naturally aerodynamic. The wind would draw on the outer surface of the bubble and propel it up inside the ribbon. As it reached higher altitudes it would accelerate and form a vacuum behind it which would accelerate the following bubble. Once primed in this way the air input for the ribbon could be passed through a turbine to generate electricity. Base stations could be large buoys or redundant rigs, so the ribbons could be easily deployed using small barges. Since there’s nothing on the end of the tether, other than a small balloon or kite to provide the first lift, complicated maintenance of airborne components is avoided. The only new technology is the design of the ribbon.

    • daryan12 says:

      Would have to do a calculation on how long the tether would have to be, what worries me is its likely to be very long indeed. Not sure if you’ve heard of solar updraft towers, as they are a similar idea for an induced stack effect.

      Another option with long tethers is the electromagnetic tether effect, which basically draws some power from the earth’s magnetic field. NASA have done some experiments with the idea of using them as a form of reactionless propulsion in LEO. However, the size of the tethers needed is rather large, many km’s in fact, so not sure it would be practical for bulk power generation.

      • Yes, I remember solar updraught towers as a concept. I was well into low impact energy systems in my days as an architectural student back in the 1960s. I started considering the ribbon system in the 1980s as an idea for raising sea water skyward and evaporating it to encourage precipitation. – Parking a few in the Atlantic along the Skeleton Cost of Africa could have helped to increase moisture transfer to the southern Sahara. Unfortunately I had no time to pursue the idea at a practical level (wife & kids!).
        Incidentally the tether ribbon ideas are extremely scalable and could start small and evolve.

  3. neilrieck says:

    Is wind the cheapest? I thought there was a better ROI from water (hydro) because dams allowed for the buildup of water in a reservoir. (of course there are many primary and secondary consequences to building a dam)

    • daryan12 says:

      Again, this depends on whether we are thinking in terms of LOCE, the timelines involved (the CAPEX of a dam is quite high but lasts a long time) and to what extent are we factoring in the costs of energy storage. I don’t have the relevant stat’s to hand but my understanding is that existing hydro is cheap yes, but new build hydro has a higher LOCE cost than onshore wind energy.

  4. pendantry says:

    Interesting article! It reminded me of a thought experiment I had back in 2009 (I had to go hunting for the relevant post, but I finally found it). I’m interested to find that my ‘ergbird’ idea (thrown out as an afterthought in a comment on that article) resonates with your migratory turbines thinking. Good to hear that I might not be as nuts as I sometimes suspect that I might be! 🙂

    • daryan12 says:


      Your article does mention the NIMBY factor, which I probably should have addressed. On the one hand, AWES will look fairly small from the ground and not quite as imposing as, say a conventional turbine does up close. On the other hand, fields of these “beanstalks” of tethers might be seen as unsightly. And on a clear day an AWES will probably be visible from quite a long distance away.

      In short, someone’s probably going to object to it, but someone’s going to object to anything we do!

      • pendantry says:

        Ain’t that the truth… :/ Unfortunately, someone’s alway going to object to anything not done, too. Unless they’re dead because nothing was done…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s