Conference programme

Back to the programme printer.gif Print

Wednesday, 18 November 2015
14:30 - 16:00 Power curves in the real world
Resource assessment  
Onshore      Offshore    

Room: Montparnasse

Investors acknowledge low risk through accurate predictions of their projects performance. To improve our predictions we need to understand real turbine behavior – real sites and real wind. We will look at new methods for better predicting and reviewing of power curve performance

Learning objectives

New methods for more advanced power analysis are presented

•             Determine the extent to which power curve degradation is an issue in your windfarm

•             Evaluate first power curve measurement results from a spinner measurement as an alternative to the industry standard met mast based

Lead Session Chair:
Tomas Blodau, Senvion, Germany
Jørgen Højstrup ROMO Wind/Højstrup Wind Energy, Denmark
Henrik S. Pedersen (2) F P Eduardo Gil Marin (2)
(1) ROMO Wind/Højstrup Wind Energy, Maarslet, Denmark (2) ROMO Wind, Aarhus, Denmark

Share this presentation on:

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2015 and are published here unedited

Jørgen Højstrup has been working in wind energy for more than 30 years in research (Risø National Laboratory) and wind industry (NEGMicon, Suzlon Energy and ROMO Wind). He holds a MSc in Electrical Engineering and a PhD in Meteorology. He was Executive Vice President in ROMO Wind, responsible for wind technology and optimization. Presently he is CEO and owner of the consultancy “Højstrup Wind Energy”, CTO of "Wind Solutions" and Senior Advisor for ROMO Wind.


Spinner anemometer power curves compared with IEC measurements


The spinner anemometer has demonstrated its capabilities for measurement of yaw misalignment and verification of the relative improvement of power curves by correction of yaw misalignment (or other changes in the turbine configuration). Now through an intercomparison experiment with a conventional IEC 61400-12-1 measurement, it has been shown that the spinner anemometer can also measure absolute power curves.

The power curve is influenced by a number of external parameters, such as turbulence intensity, yaw misalignment, inflow angle and vertical wind shear, which can all be measured reliably by the spinner anemometer.

The latest addition to the toolbox has been the capability of measuring vertical wind shear by analysis of the 3D measurement of turbulent wind speeds and temperature fluctuations.


A windfarm consisting of 13 Siemens 2.3 MW windturbines was instrumented with spinner anemometers. A meteorology mast instrumented in accordance with IEC 61400-12-1 was positioned three rotor diameters from one of the turbines. Additionally the metmast was equipped with further anemometers for the measurement of the wind profile. Also a 3D ultrasonic anemometer was installed on the mast near hubheight.

The spinner anemometer was calibrated on a nearby turbine (equipped with a nacelle mounted Lidar), and the same calibration was used for the measurements of the power by the spinner anemometer on all turbines. Power curves were derived from three months of measurements. Additionally power curves were produced using the nacelle Lidar and the nacelle anemometers on the turbines.

24 hours of measurements from the spinner anemometer (10Hz sampling), the ultrasonic anemometer on the metmast and the metmast profile instrumentation were used to develop and test the procedures to derive wind shear from the spinner anemometer. Firstly we verified that turbulence (both statistics and spectral distributions) measurements of the spinner anemometer was consistent with measurements by the ultrasonic anemometer and the cups and vanes on the metmast. Secondly further analysis (through the Turbulent Kinetic Energy equation) was applied to derive the vertical wind shear which was then compared with the measured shear on the metmast.

Main body of abstract

1. The IEC power curve was reproduced very well using the spinner anemometer instead of the metmast cup.
2. The power curve measured by the spinner anemometer shows considerably less scatter than the IEC power curve.
3. The power curve measured by the Lidar shows scatter similar to the IEC curve, but the measured windspeeds were too high (no explanation found).
4. All 13 power curves measured by the spinner anemometers could be collapsed into one curve.
5. It was demonstrated that wind shear could be measured by the spinner anemometer.
6. All three components of the turbulent velocity fluctuations measured by the spinner anemometer compared well with the mast mounted ultrasonic anemometer. The spinner anemometer was less affected by precipitation than the mast mounted ultrasonic, and the overall quality of the spinner anemometer signal was less noisy.


Power curves were measured with IEC instrumentation, spinner anemometer, nacelle mounted Lidar and nacelle anemometers. The power curve measured with the spinner anemometer compared very well with the IEC measurement. Furthermore the capabilities of the spinner anemometer for measurements of turbulence and wind shear were verified. The 13 power curves measured by the spinner anemometer could be collapsed into one curve.

Learning objectives
Power curves can be measured with the spinner anemometer easily, at any time, at any turbine in a wind farm, much cheaper than an IEC measurement. More importantly the measurements are not restricted to one or two turbines, but can readily be performed for a larger sample of turbines. Added advantages are that the influence of any improvements or upgrades of the turbines on the energy production can be checked using the capability of the spinner anemometer to measure relative power curves accurately and yaw misalignment can also be corrected.