Back to the programme printer.gif Print




Delegates are invited to meet and discuss with the poster presenters in this topic directly after the session 'How does the wind blow behind wind turbines and in wind farms?' taking place on Tuesday, 11 March 2014 at 16:30-18:00. The meet-the-authors will take place in the poster area.

Giorgio Demurtas DTU Wind Energy, Denmark
Co-authors:
Giorgio Demurtas (1) F P Troels Friis Pedersen (1)
(1) DTU Wind Energy, Roskilde, Denmark

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2014 and are published here unedited

Giorgio Demurtas is currently Phd student in the Test and Measurements section of DTU Wind Energy. His main research project for the past year has been test and development of spinner anemometer calibration procedures to measure inflow inclination and yaw misalignment with extreme accuracy. He holds a BSc in Electrical Engineering from the University of Cagliari (2009), and a MSc in Wind energy (2012) from the Technical University of Denmark (DTU).

Abstract

New methods to calibrate a spinner anemometer

Introduction

Instead of traditional wind sensors mounted on the nacelle top, the spinner anemometer utilizes the flow over the wind turbine spinner to determine the local wind conditions. Not being disturbed by the flow distortion due to blades and nacelle, it can provide very accurate measurements of flow inclination, yaw misalignment, wind speed and turbulence, which might be used to better control a wind turbine in order to improve energy production and reduce the loads.
The calibration of the spinner anemometer, which is important to measure accurately, has never been presented before and it is described here.

Approach

This wind sensor is based on three 1D sonic sensors mounted on the spinner and an algorithm to convert the wind speeds measured by the three sonic sensors to horizontal wind speed, yaw error and flow inclination angle. This conversion algorithm utilizes two constants that are specific to the spinner and blade root design and to the mounting position of the sonic sensors on the spinner.
The two constants must be calibrated as an integral calibration of the spinner anemometer in order to measure accurately. Because of the size of a wind turbine spinner, the calibration in a wind tunnel is not feasible. A method for calibrate a spinner anemometer to accurately measure flow inclination, direction and, off-course, speed, has been developed at DTU Wind Energy and tested on a 500 kW stall regulated wind turbine.
In this method, the two constants are calibrated independently by means of two different tests and instruments set-up. For both methods the rotor of the wind turbine is stopped. The calibration of the first constant, ka, uses the yaw position measurement as reference and fast sampled measurements (at least 1 Hz, 20 Hz in this case). For the calibration of the second constant, k1, requires a hub height wind speed measurement, therefore an averaging time is needed to improve the correlation between the two wind speeds, measured at different points in space.
For a good calibration, it is important that the wind direction during the test is as stable as possible and the terrain flat. A wind speed of at least six meter per second should guarantee a sufficient stability of the wind direction.

Main body of abstract

The measurement set-up consist of a wind turbine equipped with spinner anemometer, yaw position measurement and a hub heigh horizontal wind speed measurement, which in this case is a cup anemometer on a nearby met-mast. The distance between the hub-heigh measurement of the wind speed and the wind turbine is not really important, since the calibration is performed in stopped condition, when there is not induction due to the rotor. Another option is to use a nacelle based or a ground based lidar. The time needed to acquire the measurements for the calibration, in good wind conditions, is a couple of hours.
Following the installation of the spinner anemometer on the wind turbine spinner, the procedure for the "internal calibration" must be executed, as explained in the spinner anemometer user manual. This procedure compensate for differences in the mechanical mounting and characteristics of the three sonic probes, so that they measure the same mean wind speed over several rotor rotations.

The spinner anemometer constants are initially set to default value, one. In this conditions the spinner anemometer can still measure with good precision (but not so good accuracy) flow inclination, direction and speed, as well as a cup anemometer can measure wind speed with default calibration values. Measurements can be corrected afterwards, once the correct calibration constants are known. The calibration constant ka mainly relates to angular measurements, while the constant k1 mainly relates to the wind speed measurements. The procedure to calibrate measurements made with default calibration constants requires to back calculate the speed at the sonic probes, and then apply the (non linear) conversion algorithm with the new constants.

To calculate the first constant, ka, the wind turbine, in stopped condition, is yawed several times plus minus approximately 80 degrees respect to the wind direction and fast sampled measurements of the spinner anemometer and nacelle yaw position are recorded. The measurements are then corrected with the appropriate ka that better the relation between the yaw error measured by the spinner anemometer and the wilfully generated yaw error by yawing the wind turbine. The calibration of ka shows to have very little influence on the wind speed measurements.
A simple sensitivity analysis indicates that the accuracy of yaw error measurements is directly dependent of the accuracy on the calibration constant ka, and that the uncertainty is a relative value, being quite small when the yaw error is small.

In order to calculate the second constant, k1, the wind speed measured by the spinner anemometer should be compared to the wind speed just in front of the spinner. A measurement right in front of the spinner is practically difficult to implement, therefore a hub-height wind speed measurement, located at some distance from the spinner is used instead.
The wind turbine induction makes the two wind speeds different during operation. This problem, however, can be easily overcome by stopping the wind turbine until a sufficient number of ten minutes averaged measurements are recorded. The wind turbine should be free to into the wind. The value of k1 is then calculated from equations derived from the conversion algorithm and should be constant respect to the wind speed.


Conclusion

Thanks to this procedures to calibrate the spinner anemometer constants, it is possible to measure very accurately the yaw misalignment, the flow inclination, and the wind speed at the centre of a wind turbine rotor.
Being the sensor based on sonic probes, the sampling frequency can be quite high, in the order of 20 Hz. Thanks to the rotation of the instrument, there is not offset error in the yaw misalignment and flow inclination measurements. This is a unique advantage respect to traditional direction sensors (vanes and sonic anemometers) which are always subjected to a mounting offset uncertainty and, worst, operates in a region of flow disturbed by blades and nacelle.
The flow inclination angle and the yaw misalignment have not yet being a big concern in the design and operation of a wind turbine, while now, the possibility of measure them accurately and at high sampling rate, opens the possibility of study their effect on the structure and power performance of a wind turbine.
For example, they could be used by the control system to adopt the appropriate control strategy to improve the energy production and reduce the loads.
Together with those useful informations about the flow at the spinner, the spinner anemometer can replace or be redundant to other sensors, such as the rotor position, and the air temperature.

Another important advantage is that for the measurements of flow angle, the uncertainty is a relative value, being quite small when the flow angle is small, which makes the spinner anemometer ideal to measure the wind turbine misalignment.



Learning objectives
You will learn how to perform the test to calculate the calibration coefficient of a spinner anemometer. A basic description of the algorithms involved is also given. You will be introduced on the possible use of such accurate measurements of flow inclination and yaw misalignment.



References
1. Højstrup J., Nielsen D., Hansen K., Lauritzen L., Maximise energy production by minimazing yaw misalignment. Large scale field deployment of spinner anemometer, Poster 0162, EWEC 2013, Vienna 2013.
2. Frandsen S., Sørensen N., Mikkelsen R., Pedersen T.F., Antoniou I., Hansen K., The generics of wind turbine nacelle anemometry, EWEC 2009, Bruxelles 2009.
3. Pedersen T.F., Spinner anemometer - an innovative wind measurement concept, EWEC 2007 Milan 2007.
4. Pedersen T.F., Sørensen N., Evevoldsen P., Aerodynamics and characteristics of a spinner anemometer, Journal of physics: Conference Series 75 012018, 2007.
5. Frandsen S., et.al.,Nacelle anemometry, EWEA2006, Bruxelles 2006.