Thu, Mar
14 New Articles

Operational Energy Yield Assessment - Review of the Real Energy Production Data of Operating Wind Farms in Comparison to Former Energy Yields


The prediction of the site specific energy yield of wind turbines plays an important role for financial agreements especially under consideration that the energy productions differ from site to site.


A validation of energy yield predictions on the basis of real produced energy of the wind farms after a considerable time of production is an important step in order to understand the current performance of the wind farm, for the revenue forecast and financial planning. Furthermore a review of the former EYA in comparison to the real produced energy is imperative for advisors to check for the consistency of applied procedures and uncertainties when performing energy yield assessments and to enhance the quality and accuracy of the same.


For the comparison of former energy yields with the real produced energy of the wind farms, the production data and availability information are necessary. 

To ensure a comparison with energy yield predictions, the real production data has to be corrected to 100% availability and possible outlier ¬ values need to be excluded from the investigation. Furthermore it is necessary to perform a long-term correction of the real production due to the long-term given values in former EYA.

Once the real production is corrected to 100% availability and to a long ¬ term value, this long ¬ term value is then compared to the calculated value from the former EYA. This comparison is done with the long-term corrected energy yield (p50 value) and under consideration of the given uncertainties in the former EYA.

Under consideration of the uncertainty the probability values, the p75 and the p90 values from the former EYA are also to be compared with the results of the performed work.

Case study and results

Since 1998 UL- DEWI performed more than 3600 Energy Yield Assessments worldwide in the form of:

• Feasibility Studies

• Energy Yield Assessments (EYA)

• Review of 3rd Party EYA

• EYA based on production data (Wind farm already in operation)

With the above experience, UL-DEWI performed the first comparison in 2008, followed by a further comparison of the predicted energy yield of 190 wind farms located in France and Germany with the real production of the wind farms. 

The uncertainty of the former assessments was taken into account and a comparison of the probability values with the corrected real production was additionally performed. 

The main focus of the presented work lies within the following key questions:

• Are the calculation results (P50) within the “expectations”, are they over- or underestimating the sites?

• Do the assessed uncertainties correspond to the reality?

• Are there any general trends or rules / guidelines derivable from this data analysis?


The Tab. 1 shows the results of the comparison of the p50 from the original EYA to the corrected real production for the different services considered. The average of all 190 wind farms shows that the corrected real production is 8% below the EYA. The given uncertainty in the former EYA is around 13 % within the service EYA and 3rd party EYA and 5% higher within feasibility studies. 8% uncertainty is given as a mean value for EYA based on production data, what means that initial EYA was performed after the commissioning of the wind farm based on the real production. These EYA based on production data shows an underestimation concerning the prediction versus the corrected real values of only 3%.

Delving deeper into these comparison results, present a well-balanced distribution of hits and overestimation of the real production as well as low grade of underestimation. The results are shown in Fig. 1. For the classification of the comparison, a deviation of +/- 5% is defined as “hit”. The classification over-/underestimation were given for more than 10% deviation compared to the former result.

Due to the fact that performing an EYA is connected with several uncertainty sources, the uncertainty has to be taken into account while performing a comparison of the prediction with the real production. For considering the uncertainty the given p75 and p90 values in the former EYA were compared with the corrected real production.

The following graph (Fig 2) is showing that there is on an average no deviation between the calculated p75 and the real energy production. That is showing that uncertainties are very important to consider and that within the performed investigation the p75 value is showing a confidence match with reality.

 As an important point it has to be viewed more in detail, especially the timing of the EYA performed. As it is shown in the following graph (Fig 3) there is a constant improvement in the assessments concerning the ratio between calculated and real production.

Discussion and Conclusion

Given a large number of possible reasons on why a prediction is fulfilled or not, several aspects need to be considered for a comparison of predicted energy yields with the long¬ term corrected real energy production.

With respect to the real production, aspects technical availability, strong connectivity to O&M, deviations concerning the used power curve for the EYA and the reality, and possible limitations in the operation, e.g. due to high turbulence are possibly not considered in the former EYA. Furthermore limitation of the grid connection and especially later wind farm extensions can lead to lower production than predicted.

Concerning the models used we can see a slight but still not statistically significant tendency of overestimating the production of large wind farms and on the other hand of underestimating the production of turbines with large hub heights. 

Especially considering large hub heights, it is observed that this underestimation mainly is caused by a high difference between turbine hub height and height of the reference (e.g. met mast) and can be reduced by keeping the vertical distance low or covering the vertical distance with remote sensing devices such as LiDAR.

Dealing with former assessments, deviations to the real production could be caused by topographical input data, calculation models, changed working procedures, measurement equipment used or different long-term correlation. 

As these described reasons are not always connected with the accuracy of the former energy yield it is not only for advisors necessary to check their former results to improve quality but as well important for operators to have a closer look on the current performance to update revenue planning with operational assessments and to leverage the IRR of their assets.

As an important part of the continuous improvement program of UL-DEWI, we continue to perform this review in a frequent manner in order to check for the consistency of the applied procedures and uncertainties when performing energy yield assessments and to enhance the quality and accuracy of the same.

For more details, please refer to: Http://www.dewi.de/ 


Combining technical expertise with many years of in-depth industry experience, the DEWI Group (a UL company) offers global, one-stop wind energy services to turbine manufacturers, component manufacturers, project developers, utilities and other companies within the sector. The DEWI Group helps stakeholders – developers, investors and operators – to identify the critical aspects related to wind farm projects through comprehensive one-stop services, individually tailored and flexibly delivered. 

Worldwide we have completed more than 270 Due diligence projects, 1,610 power curve measurements & wind measurements and More than 3,630 energy yield assessments in over 30 countries.


B.S. Nivedh

UL India Pvt Ltd, Bangalore, India

This email address is being protected from spambots. You need JavaScript enabled to view it.


Till Schorer

UL International GmbH, Germany

This email address is being protected from spambots. You need JavaScript enabled to view it.