Engineering: performance analysis of wind assets

News - 26 May 2020

A fine and regular analysis of the production data of a wind farm allows to detect possible anomalies and thus to optimize the performance of the power plant. VALEMO’s Engineering Department is developing an analysis tool based on specific performance indicators in order to identify underperforming turbines. The focus is on the operational efficiency of the fleets in operation.


In constant search for expertise and innovation, VALEMO has developed this tool to carry out a scan on an entire fleet of machines.

Filter the data

By filtering the SCADA data, this analysis tool studies only the nominal operation of the wind turbines, thus excluding invalid 10-minute points such as environmental (chiropterian or acoustic), mechanical or grid related restraints.

Previously identified clamping plans, grid limitation periods (at the PDL level for example), 10-minute points containing a shutdown or a transition (such as restarting the turbine) are filtered out. Only operating points that are comparable to the manufacturer’s data and that are comparable between different wind farms are therefore studied.

In order to allow a valid comparison of the wind turbines within a wind farm, the wind turbines are then placed in an iso-production configuration of the wind farm. After this “cleaning” of the data, the analyses begin. VALEMO has built several indicators to compare the performance of wind turbines within a wind farm and also between wind farms.


Develop specific indicators

  • AEP (Annual Energy Production)

Annual Energy Production is an estimate of the annual energy produced. It is calculated from a power curve and a reference wind distribution. The power curve is established with a range of wind speeds every 0.5 m.s-1 and by calculating the average production from the reworked SCADA data. The reference wind distribution is calculated with a Weibull distribution of a standard site. The objective of this indicator is to provide information that is less dependent on the wind speed measured by the nacelle anemometers. Indeed, this data can be imprecise. Therefore, VALEMO recommends an analysis of the production data first.

  • Deviation with respect to the reference power curve of the fleet

From historical and filtered data over a period of 1 year, a reference power curve was constructed for each park with a step of 0.1 m.s-1 for wind speeds. This power curve represents the theoretical expected power as a function of the wind data measured by the nacelle anemometer. VALEMO recommends to use this power curve identified from the data rather than the manufacturer’s one because it is more representative of the actual behaviour of the wind turbines of each wind farm. For the analysis, the deviation from this reference is thus calculated for each 10-minute point, focusing on the rising area of the power curve, before performing an average. This indicator is used to identify the deviation of a wind turbine from the rest of the wind farm.

  • Deviation from average wind deviation

From the filtered data, the sum of the production per machine and the average wind speed is calculated. For each machine, the deviation of the production and wind speed from the reference of the park is thus obtained. As the data are at iso-production, VALEMO establishes the relevance of this indicator to identify an under-performance.

Regularly and simultaneously analyse the results studied to identify any anomalies

These different indicators are thus complementary and their simultaneous analysis makes it possible to highlight underperformances in a fleet of heterogeneous machines due to their model, location and implementation. Each month, VALEMO performs this analysis on the basis of the data of the past month as well as a cumulated data of the last 6 months in order to be able to identify a possible deviation as soon as possible.

Next step for VALEMO: to integrate these different indicators into a real-time monitoring tool, allowing operations managers to receive alerts if abnormal behaviour is identified.

VALEMO is at your disposal to assist you in the optimization of your wind assets.


The VALEMO Engineering team