MexNext unveils the secrets of the wind
“In terms of aerodynamics wind turbines are far more complicated than aeroplanes”, says Gerard Schepers of ECN Wind Energy. “So complicated in fact, designers are largely still very much in the dark and sometimes have to rely on intuition and experience
Exact mathematical definition of the air currents round the blades and other components of a wind turbine has been possible since the nineteenth century, thanks to Navier and Stokes. So why not just do the maths? “Well”, Schepers says, “That’s one of the ‘Millennium Problems’. The American Clay Mathematics Institute would give you a million dollars if you could make a good start at it.”
It is not even certain that any usable solutions of the Navier-Stokes equations actually exist for the air currents concerned here. Mathematicians don’t see any light at the end of the tunnel yet and the darkness could, in this case, even be eternal. Exact maths has let us down. The aerodynamics of wind turbines is a question of approximation. Now this still means an uncertain compromise; computational fluid dynamics (CFD) can simulate the air currents very accurately. But the computing time for a design soon amounts to years, even using supercomputers.
Designers therefore have to rely on simplifications such as the blade element momentum (BEM) method. This divides the rotor blade into sections and assumes that the air current around each section can be calculated separately, independently of the adjacent blade sections. Moreover, a BEM approach only works if the current remains fairly constant.
The experimental turbine of the Mexico project in DNW’s large wind tunnel in the Noordoostpolder.
Computational errors of 50 percent
Schepers: “The current approximations predict the reality very nicely, provided the wind is blowing head on and the rotor does not use too much capacity. The error margin quickly increases with off-design conditions, greater load and tilt. It can easily mount up to something like 50 per cent.”
Wind consists of gusts and whirls. Direction and force can easily change and not all parts of the rotor will feel the same force at any one time. The deviation from the clean, constant wind the design software bases its calculations on is often considerable, resulting in nasty consequences for the wind turbines. Overload, instability, disappointing performance, unexpected defects; the cause is usually a lack of control on the aerodynamics.
Certainty about air
In 2000 the American National Renewable Energy Laboratory (NREL) set up a wind turbine with two rotor blades and a span of ten metres in the largest wind tunnel in the world at NASA’s Ames Research Center. Six years later ECN Wind Energy coordinated the European project Mexico (Model rotor EXperIments under COntrolled conditions). A slightly smaller three-blade rotor was installed in the largest European wind tunnel, owned by the German-Dutch Wind Tunnels (DNW) and situated in the Noordoostpolder. “In 2006 a new measuring technique offered new possibilities”, Schepers explains.
“Six years ago NREL could only measure the turbine itself. We also had particle image velocimetry (PIV).” Small soap bubbles floated with the air current. Every 200 nanoseconds they were flashed by lasers and photographed by two cameras. This resulted each time in a detailed definition of the air current as a vector field of velocities, a form which corresponds very closely with the results of approximations based on Navier-Stokes.
A blade of the Mexico experimental turbine. Three different profiles were used, connected by two transition regions. The three blades were fitted with a total of 148 pressure sensors.
Wonderful data
Schepers is now coordinating the follow-on to Mexico: MexNext. Seventeen institutes in eleven countries are collaborating to compare the Mexico data with the results of the calculations. “Mexico has produced wonderful data”, Gerard Schepers says. “Including data relating to tilts of up to 45 degrees and various blade angles. Now we can see the results of the first analyses. Very exciting! They demonstrate that the CFD methods work well at a fundamental level. And it is also becoming clear where simpler models made major errors.”
MexNext will run until 1st of June 2011. A follow-on project is being considered; Gerard Schepers would like to send the turbine that was tested in Noordoostpolder in 2006 to the US for a series of measurements in the wind tunnel at NASA’s Ames, to be able to see exactly what difference the tunnels make. The effect of the new control techniques, such as the fast adjustment of the blade angle, has not yet been examined. Schepers would also like to combine the aerodynamic and acoustic measurements in order to identify the aerodynamic effect on noise emission.
Benefit per kWh?
The efforts will certainly lead to better and faster calculation methods, with the addition of CFD where necessary and the addition of new correction factors where possible. What difference will it make in the long run, in terms of kWh? “It is too soon to name a figure”, Schepers says. “And it will also depend on how manufacturers will implement our work. But this is certainly the basis. The quality of the aerodynamic calculation models will make or break the success of the larger wind turbines and complex control systems the designers are currently working on.”
Above: A result of particle image velocimetry (PIV), the subject of which was a tip vortex of the Mexico experimental turbine. Such detailed information about the interaction between the rotor blades and the air current will, in the future, enable far more accurate wind turbine designs. A significant parameter is the Betz limit, an aerodynamic limit to the possible energy that could be derived from the ideal wind turbine. A maximum of 59 percent of the energy in the air current through the rotor disc can be extracted from the wind. It goes without saying that we aim to realise that limit as closely as possible. But that also entails a heavy mechanical load, with a negative impact on the cost per kWh.
The objective is the lowest price for energy – that optimum comes within reach if the designer has the tools to make better, yet fast calculations.
Contact
Gerard Schepers
ECN Wind Energy
Phone +31 (0)22 456 4894
E-mail: Gerard Schepers
Info
ECN - Wind energy presentations of the Dutch Wind Workshops 2008
A short history of wind turbine aerodynamics, or: From Betz to Better
The MEXICO project: Analysis of yaw measurements and comparison with existing models
Model Experiments in Controlled Conditions
Website
www.mexnext.org
Text: Steven Bolt
This ECN Newsletter article may be published without permission provided reference is made to the source: www.ecn.nl/nl/nieuws/newsletter-en/
