"Calculations indicate that we can reduce material fatigue by 70 percent," says Herman Snel of ECN Wind Energy. "And that would allow us to build much larger wind turbines."

It can happen quickly or it may take a long while, but every piece of metal or composite material which is subjected to cyclic stress will break eventually. Making critical parts sufficiently strong, which means heavy, can be a solution for relatively small constructions. "But doubling the size of a wind turbine will then make it eight times as heavy, says Snel, "while the amount of power it supplies increases by only half as much." So far, improving technology has kept weight and power in somewhat closer step. As rotor diameter went from 50 to 100 meters, wind turbines gained sixfold in weight and their power increased 4.5 times, making the larger turbines economically viable. Especially at sea, because the cost of building an offshore wind farm rises with the number of foundations.

Stall to pitch control
Older wind turbines depend on the aerodynamic 'stall' of their rotor blades to control power. When the airflow hits the chord line of a rotor blade exactly from the front, the angle of attack is zero and the blade does little or no work. Ideally, the relative wind approaches a bit more from below, with an angle of attack of around 8 degrees. This results in a lift force much larger than de air resistance (drag) of the blade, with a component in the rotor's direction of rotation. At higher wind speed and roughly constant revolutions per minute, both angle of attack and lift force increase, causing a higher power output.
Beyond around 15 degrees the blade will stall. The drag continues to increase, but the lift force drops, which limits the power to a safe value.
Snel: "An operating point close to the stall is not too bad in some respects. Variations in the angle of attack don't alter the forces on the rotor very much, in that region. But those forces will already be very large. And the way in which the airflow detaches during the stall is a rather unpredictable process, with variations in time, causing cyclic loads." So far, stall power control has only been used for wind turbines up to about two megawatts. Larger turbines are pitch controlled; the blades are continuously adjusted to keep the power within narrow limits, and the angle of attack remains well outside the stall region.

  The angle of attack is the angle between the relative wind and the chord line of the rotor blade.

The forces of lift and drag on a rotor blade vary with the angle of attack. The exact shape of the curves depends on the blade profile.

Fatigue loads
This means less turbulence reduces the load on the rotor blades, at least on average. But there is also a disadvantage. The slender blades of a large wind turbine are extremely sensitive to variations in the airflow, especially at small angles of attack. A change of one degree may alter the lift force by several tons, causing severe bending stresses on the blade roots.
Snel: "Pitch control per individual blade is enough to counter differences which occur once per revolution, like the dip in the bending load caused by passing in front of the tower. But the faster variations also contain a large amount of energy. And the current pitch control systems aren't fast enough to handle them."

Synthetic jets
What's needed is a continuous adjustment of the lift force, without having to change the angle of incidence for the entire blade. Flaps on the trailing edge (often called ailerons) could do it, but would be rather vulnerable and difficult to maintain. Together with the University of Twente ECN is working on synthetic jets; each being a cavity containing an oscillating diaphragm and having small openings in the upper and lower skin of the blade, near the trailing edge. Loudspeakers are used in the prototype which is currently being tested in a wind tunnel. They operate at a resonant frequency of the cavity, causing air to be sucked in on one side while being expelled at the other side, about 100 times per second. To decrease the lift force, air is blown out on the upper side of the blade; directing the jet downwards causes an increase. The flow must of course be accurately controlled, and that requires sensors.

Cheap and effective
"There may be an elegant solution: pressure sensors in the leading edge of the blade, to measure the pressure difference between the lower and upper sides," says Snel. "Changes in the angle of attack could be accurately derived from their data, allowing the software controlling the jets to compensate almost immediately."
Synthetic jets are mechanically simple and require very little energy. In this application their openings would be close to the trailing edge of the rotor blade, where ice is unlikely to cause trouble; they should work very reliably. The (very small) openings for the pressure sensors are more likely to suffer from ice and dirt, but: "The software would notice that an could easily clean them using air pressurised by the rotor itself - the hollow blades allow it to work as a gigantic centrifugal pump."
The system's accuracy and reliability can be further improved by putting affordable accelerometers into the tips of the blades, to measure the bending caused by airflow variations.

 
Wind turbines rated at 10 megawatts will have blades which are more than 70 meters long. Within that length, a turbulent wind may vary considerably (source: Erich Hau, 1988, Springer Verlag).

Lighter by ten percent
Calculations and experiments must first prove that the application of synthetic jets is a viable concept. If so, the information gathered during this phase will allow the design of a complete installation, and large benefits can be expected.
Snel: "Our mathematical models have come a long way. They can predict fatigue rather well, by simulating both the atmosphere and a wind turbine in detail. The cyclic loads and the damage resulting from them can be estimated with a 20 percent margin of error."
It appears likely that the use of synthetic jets can reduce the rate at which fatigue damage increases by 70 percent. "In that case a construction with the same fatigue life can be ten percent lighter, allowing much larger wind turbines to be built. Think of blades exceeding 70 meters in length and 10 megawatts of power!"

Text: Steven Bolt

Contact
Herman Snel
ECN Windenergie
Tel.: +31 (0)22 456 4170
E-mail: Herman Snel 

This ECN Newsletter article may be published without permission provided reference is made to the source: www.ecn.nl/nl/nieuws/newsletter-en/