Wind farm performance assessment through hybrid modelling

The Context

How can I get a complete understanding of the performance of my wind farm?

Production optimisation is one of the key factors to ensure a sustainable wind farm. The production of wind turbines is mainly influenced by:

  • Availability: The percentage of time the turbine is ready to operate
  • Performance: The active power produced with respect to the wind speed presented through the power curve

Often, SCADA data is used to monitor these 2 factors. If these data can be considered of good consistency it may be used to accurately evaluate the availability, but the situation is quite different when it comes to the evaluation of the performance.

In most cases, it can indeed be observed that some wind turbines show an inconsistency such as an over/under estimated wind speed because of incorrect calibration or a wrongly referenced wind and nacelle direction, which may lead to misinterpretations.

Download a summary of this case study in English / French / Spanish

Our Solution

To understand the difficulty of setting the correct wind speed, you have to consider 2 elements:

  • The anemometer of the turbine is behind the rotor and is affected by its turbulences
  • The recalculation of the controller (i.e the transfer function) is not always correctly done leading to a biased value

The procedure for a power curve measurement as described in the standard IEC 61400-12-1 requires the use of a met mast measuring the wind speed ideally at hub height.

The new concept of rotor equivalent wind speed (REWS) is often proposed as an option for an accurate measurement of the power curve. REWS methodology combines wind speed measurement from a met mast and a ground-based LIDAR. The LIDAR gives the wind speeds at different height along the rotor area. Such systems and analysis however come with a serious investment.

P4A therefore proposes a different methodology to achieve an accurate measurement of the actual power curve and evaluation of the wind speed distribution. Instead of installing expensive measuring tools, we propose a new approach based on our unique hybrid model for the implementation of the REWS concept. The P4A hybrid model combines SCADA data and weather data generated by a local meteorological model (short-term forecast and historical re-analysis).

Our hybrid model diagnosed an absolute error of nacelle alignment of approximately 10 degrees, resulting in a prognosed potential production gain increase of approximately 5% per year.

The Benefits

In our hybrid model, the weather data is used as a guideline to reconcile the observed SCADA data, resulting in a more robust model with a more absolute view.

Our P4A solution is cheaper and quicker then LIDAR and allows easy monitoring of the result and trend.

The P4A hybrid model can be implemented for a minimal investment with an average gain of € 21.000*.

*Result depending of electricity price & wind conditions

Contact Us

Contact us for more information at +32 (0) 81 71 99 82 or send us a message via the contact page.