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.

Soledad Sanz Sánchez-Luengo Iberdrola Engineering and Construction , Spain
Co-authors:
Sanz Sánchez-Luengo Soledad (1) F P Abascal Mendez Alejandro (2) del Castillo Martin Ana (1) Martin Martin Javier (1)
(1) Iberdrola Engineering and Construction, Madrid, Spain (2) Iberdrola Renewables, Madrid, Spain

Printer friendly version: printer.gif Print

Presenter's biography

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

Bachelor Degree in Atmospheric Physics and Meteorology by Complutense University (Madrid, 2006).
Master Degree in Applied Meteorology by Spanish Meteorology Agency and Complutense University (Madrid, 2007).
Final Master Project about ultrasonic spinner anemometry (published in EWEC 2008, co-author with Cuerva.A, Pedersen.T.F).
Energetic Resources Technician in Acciona Energy (2007- 2011).
Wind Resource Engineer in Iberdrola Engineering (2011-present).
Regular teaching in several university courses.

Abstract

AN INNOVATIVE APPROACH TO CHARACTERIZE COMPLEX TERRAIN PARAMETERS (UPFLOW AND CCT TURBULENCE STRUCTURE CORRECTION)

Introduction

The characterization of complex terrain sites following the requirements of IEC 61400-1 Ed3 (Amendment 1)
standard [1] is a key issue in current site suitability studies, and involves a whole analysis of upflow wind and turbulence intensity components.
An innovative study of measured and modelled values of these parameters and a comparative with their theoretical approximations has been accomplished.
Measured values of vertical wind speed are essential; therefore propeller sensor data registered in different complex sites were analysed.
Relevant conclusions concerning the current vertical wind speed measurement campaigns and the magnitude
of the required complex terrain corrections were obtained.


Approach

The IEC standard [1] requires a turbulence structure correction for complex terrains. The longitudinal
component of turbulence (denoted as σ1 in [1]) should be increased to take into account the distortion of the
turbulent flow.
This effect can be included by an additional multiplication of the representative turbulence intensity with a
correction parameter Cct.
The estimation of Cct values is crucial in site assessment studies because variations in Cct values cause
variations in the representative turbulence intensity, and therefore in effective turbulence intensity (which directly involves the separation of turbines, shut-down strategies or others considered in the design of a wind farm layout).

The standard proposes three different procedures for Cct estimations:
- Measured Cct. Cct is dependent on the longitudinal, lateral and upward standard deviations σi of the wind
speed measured on-site.

- Theoretical Cct. If the terrain is complex but there are no on-site data for these components Cct = 1+0.15•ic
could be used. The complexity index ic takes the 0 value for non-complex sites and 1 value for complex sites;
in between ic varies linearly. Therefore the standard Cct value for complex sites is 1.15.
- Modelled Cct. Also, if there are no on-site data for Cct calculations, this parameter can be modelled. In this
work the software WAsP Engineering (WEng) developed by RISØ- DTU Wind Energy was used for modelling.

On the other hand the site flow inclination or upflow shall be less than 8º (considering this value as the maximum for all directions).
The upflow values can be derived from vertical wind speed (v3 from now on) and horizontal wind speed (v1 from now on) measurements.

If the flow inclination has not been measured on site, it will be assumed that the flow is always parallel to a
terrain fitted plane, within a distance of 5zhub from the wind turbine (considering zhub as the hub height).

The purpose of this work is to analyze the above described IEC standard parameters and provide a comparison
between the different Cct estimation methods for several complex terrain sites.


Main body of abstract

1. Analysis of propeller sensors data
v3 and σ3 data of propeller sensors available in several complex terrain sites were analysed in order to obtain
measured upflow and Cct values.
The sample of selected sites includes five Spanish wind farms with about one full year of propeller measurements, and classified as complex according to IEC methodology.

A new approach was required to the usual data treatment for the analysis of measured v3 and σ3 variables. Some innovative filtering criteria were developed and have been described. The sign criterion followed for v3 is upward flow with positive sign and negative sign for downward flow (seen from the mast position).

The more complex the nearby terrain is, the higher v3 values are found. But even in the less complex positions,
the usual v3 10-min average values range from -0.5 to 1 m/s, therefore a significant amount of instantaneous
data was below the threshold limit of the sensor (±0.3 - ±0.4 m/s depending on the model). In addition, a high
dispersion in the 10-min measurements was found. Therefore, when working with 10-min data it becomes
relevant the analysis of extreme values, trying to avoid the use of data below the threshold limit.

2. Measured upflow analysis
The upflow values were calculated for every 10-min data using measured values of v1 and v3 with Equation 2.
The standard analysis in 30º amplitude sectors is not enough to evaluate the influence of topography and other
factors in the upflow values. A detailed analysis in 5º amplitude sectors was carried out for this particular work.
However, the obtained maximum values tend to be more restrictive in this specific 5º analysis. The upflow value
considered for the comparison with the maximum allowed limit of 8º is the maximum of the directional average
sector values.
The upflow directional behaviour was analysed and compared with the mean slope towards the centre line of
each sector within a distance of 5zhub from the position, in order to analyse the validity of this approach.



As shown in the figure above the upflow values generally followed the most significant slopes around the masts, but a high dispersion of upflow measurements was found. This, together with the significant number of measurements under the sensor threshold has led to consider the analysis of upflow in terms of the percentage of 10-min data exceeding the limit value of 8º. The upflow results summarized in the next table show that the mean directional upflow values did not generally exceed the 8º value; however the 10-min values exceed the 8º with a frequency higher than 20% in the most complex sites.

3. Measured Cct analysis
An analysis of Cct and its horizontal (σ2/σ1) and vertical (σ3/σ1) components was accomplished. σ1 is the standard deviation of the horizontal wind speed v1, and σ3 is the standard deviation of the vertical wind speed v3. The transversal component σ2 is calculated as v1•σθ, being σθ the standard deviation of wind direction.
The Cct values were calculated also for every 10-min data using the measured values of σ1, σ2 and σ3 with Equation 1.
In the case of Cct and its components, the most significant analysis is the one in terms of v1. When analyzing
Cct values in terms of v1 a great dispersion can also be observed, mainly for v1< 3m/s approximately as it is represented in the next figure.

Cct values for v1> 3m/s are usually a 3-5% lower than the corresponding ones for the whole range of v1. But again the general dispersion found, together with the significant percentage of data under the threshold limit of the sensor may compromise the reliability of measured Cct values. The results obtained are summarized in the next table.

4. Comparative of Cct estimation methods
The WEng model estimates the longitudinal, lateral and upward standard deviations of velocity perturbations
with its own turbulence model [2], [3].
When comparing modelled, measured and theoretical Cct values in Wind Farm A and Wind Farm E, it was
observed that measured values of Cct were higher than the theoretical value of 1.15, with discrepancies
between 3-17% in the most complex sites. But WEng modelled values were lower than theoretical ones, with
discrepancies between 3-16% in absolute value. The theoretical Cct value of 1.15 results in an intermediate value between measured and modelled results for the analysed sites. The next table show the results obtained.


Conclusion

A detailed study of the complex terrain characterization parameters proposed by the IEC standard [1] was carried out. Measured, modelled and theoretical values of upflow and Cct variables were analysed and
compared.
The main results regarding the analysis of measured v3 data with propeller sensors is that a significant amount
of instantaneous data was below the threshold limit of the sensor and a high dispersion in the 10-min
measurements was found. Therefore, when working with 10-min data it is necessary the analysis of extreme
values, trying to avoid the use of data below the threshold limits.
The analysis of measured directional upflow values concludes that most of them did not generally
exceed the 8º value. However the 10-min data values of upflow exceed the 8º with a significant frequency in
the most complex sites, which seems to be a reasonable result in the analysed sites.
Regarding the analysis of Cct measured values, a first result is the high dispersion encountered when
considering horizontal wind speeds below 3 m/s approximately. Thermal and turbulent effects are very
significant at low wind speeds so it is recommended the calculation of Cct avoiding low wind speed ranges.
The high general dispersion found in measured Cct values, together with the significant percentage of data under the threshold limit of the sensor may compromise the reliability of measured Cct values.
Finally, the comparison of measured and theoretical Cct values with modelled ones showed that theoretical
Cct value of 1.15 is an intermediate result, between the highly scattered Cct measured values and the
modelled ones (with all the usual modelling limitations in complex terrain sites).
As a future working guideline it would be necessary to make comparisons of the propeller measurements with
lidar and sonic 3D data trying to avoid the effects due to the measurement range.
On the other hand, the horizontal extrapolation of complex terrain corrections from mast to wind turbines
positions is crucial due to the very local behaviour of these variables, and needs to be analysed. Models like
WEng or CFD models could be used for this purpose.



Learning objectives
Improve the knowledge about critical parameters for the complex terrain characterization proposed by the
IEC 61400-1 Ed3 A1 standard.
Identify the main problems when dealing with measured vertical wind speed data.
Gain an appreciation of the standard measured values of upflow and Cct found in several complex sites.
Analyze the differences between the allowed Cct estimation methods (theoretical, measured and modelled).
Evaluate future working guidelines and requirements relative to complex terrain characterization.



References
[1]International Standard IEC 61400-1 Edition 3.0 2010-10 Amendment 1, “Wind Turbines- Part 1: Design
Requirements”, 2010
[2] WAsP Engineering 3.0 program (WEng), “User's guide”, Risø National Laboratory, 2011
[3] Wind Farm Assessment Tool (WAT) Version 3.0, “User´s guide”, Wind Energy Division, Risø National Laboratory, 2011
[4] Kristensen L., “Cups, Props and Vanes”, Risø National Laboratory, August 1994