Wind turbines have been increasing in size ever since the wind power industry took off.

Wind turbines have been increasing in size ever since the wind energy industry took off. These days, offshore wind farm plants are seeing wind turbines of 5 MW or more with rotor diameters that are longer than a football pitch. But is it economically viable to keep increasing in size? And how much bigger can wind turbines realistically get? The EU-funded UpWind project has been researching the answers to such questions.

20 MW wind turbines are technically feasible and could be the most cost efficient option for expanding Europe’s offshore wind energy capacity. That’s according to the just-published results of UpWind, the largest EU-funded wind energy project ever.

The project, financed by the EU’s sixth Framework Programme, had over 40 partners and ran for five years, and explored the design limits of the upscaling and integration of wind turbines. This included a design of a 20 MW wind turbine.

“UpWind was about seeing what is technically feasible in terms of bigger wind turbines, and incorporating all the different aspects into one model so that researchers know what to work towards”, explains Jos Beurskens from the Netherlands’ Energy Research Centre (ECN), who led UpWind along with Peter Hjuler from Danish research institute Risø DTU.

The thinking behind the wind power project was the large amounts of offshore wind power expected in Europe in the near future. In order to meet EWEA’s target of 33% electricity from wind energy by 2030, 400 GW of installed capacity will be needed, including 150 GW offshore wind farm. By 2050, this should increase to 400 GW, with 350 GW offshore.

To meet the required amounts, offshore wind energy needs to become more cost effective, and one way of doing this is through “upscaling” – that is, taking a current wind turbine design and increasing all the measurements proportionally.

The enlargement question UpWind initially created a computer model of a 20 MW wind turbine by taking a 5 MW wind turbine design from the US’s national renewable energy laboratory (NREL) and scaling it up to 20 MW. However, this was quickly revealed not to be the best option.

“The extrapolated virtual model was unanimously assessed as almost impossible to manufacture, and uneconomic”, explains the project report. This is because increasing the size of the wind turbine has an impact on more than just its height and rotor diameter. “Upscaling the rotor by a factor of two gives a rotor area of four times the original, which means a 5 MW wind turbine becomes a 20 MW wind turbine”, explains Bernard Bulder from ECN, who led the upscaling work package for UpWind. “However, scaling laws show that if you scale dimensions to the power of two, the weight goes up to the power of eight. This means you’ve got four times as much energy but eight times as much weight, and therefore, cost!”

The project partners therefore began investigating innovations that could potentially make the larger wind turbine more cost effi cient. One thing they tried to tackle, for example, was the overly strong ‘load effect’ – the resistance the blade encounters in the air as it moves – on the larger blades.

“Rather than using full pitch control, as we do now, where the whole blade can be controlled and moved depending on the wind, we found that with the really large wind turbines we have to mimic an aeroplane or a bird, using smart control on a blade that has a ‘flap’ on its trailing edge that moves separately”, explains Beurskens. It was found that this could reduce the weight of the blade by up to 25%.

While this model seems to work in the lab, the difficulty is in knowing in reality what would happen, and above all, the effect on the cost and weight of the blade. “With innovations like the rotor blade flap we’re adding complexity, so we need to study the effect on reliability and also on costs of maintenance”,  explains Bulder. “With an aeroplane, the wings get checked before every fl ight. But once this device is on a wind turbine rotor, it is very difficult to carry out such checks – especially when it’s offshore.”

UpWind looked at many different components of a wind turbine and how they should be designed for a bigger machine. It examined support structures (this refers to everything below the rotor blade), wind farm layouts and ways of monitoring how the wind turbine is doing (“condition monitoring technologies”) and whether it is likely to encounter problems (“fault prediction systems”).

However, one issue the project groups (“Work Packages”) came across was how to work out which of their design innovations are compatible with which others, and which would add unnecessary complexity. Bulder gives the example of a smart controlled blade and LIDAR, the laser-based measuring system also investigated by UpWind that, installed on a wind turbine, could give information on the characteristics of the wind two or three seconds before the gust arrives at the wind turbine so that the rotor can be adapted. Will these two systems work together and what will be the combined effect? The next stage would be to investigate such questions of compatibility.

A complicated business

Overall, many elements of the design and costs of the 20 MW wind turbine turned out to be more complex than expected at the start, explains EWEA’s Dorina Iuga, in charge of communication and dissemination for the project. “We are now working on a model that will
integrate all the work of the different researchers in order to find the optimal combination for cost effectiveness”, says Beurskens. “This way, the researchers will be able to see, for example, that a certain type of material – lighter than the current fibreglass – is needed for the blades, and work towards it”.

The 20 MW wind turbine design may well be feasible, but it sounds as though it is not about to go on the market. Realistically, when can we expect to see these giants rising out of the waves? While Bulder believes the current trend of increasing sizes will become less dramatic, and the smaller 10 MW machines will only become commercially available by the end of the decade, Beurskens is more
optimistic about the prospects for the 20 MW wind turbine.

“I can only give an intuitive answer, but I believe we’ll see the 20 MW wind turbines used within 10 years,” he says. “That is, providing they are the cheapest option.”

And now…

Overall, what UpWind showed is that there are various ways in which a 20 MW wind turbine could be designed and made cost effective. The next stage is to try and build up the integrated cost model by taking the information out of the lab, applying it to real turbines to see what happens and work out the way to produce the most efficient 20 MW machine at the least cost.

UpWind was such a wide-ranging project that it is impossible to provide an overview of all its results in just two pages (the full report is available on But for Beurskens some of the key developments, aside from those mentioned above, include:

Wind farm design: it was discovered that is may be more effi cient to place wind turbines in a farm in a stoastic, that is, ‘random’ way, as this lessens the effect a wind turbine has on the others nearby. This may mean today’s regularly spaced wind farms will disappear.

Support structure: UpWind also paved the way for potentially significant cost reductions concerning the support structure – that means all of the wind turbine below the nacelle, including the foundation – of larger machines. Deeper water and larger wind turbines require a bigger, heavier support structure which increases in complexity and costs. Because UpWind integrated the different elements into one overall turbine design, it should be feasible to design a more cost efficient support system for the turbine.

Cost of electricity: the UpWind team has also been working on an instrument that will show the change to the cost of electricity when parts of the turbine are modified.

By Sarah Azau,