Increased reliability of service life extension through field measurements at the individual wind turbine

Introduction

At present, the determination of the potential for service life extension of wind turbines (WT) has to be based in Germany, according to the civil building authority, on a detailed inspection and an in-depth simulation of the WT with the site’s individual wind regime to determine the remaining service life (RSL). However, the individual WT’s configuration and parameterisation has as well a strong impact on the endured loads and the resulting lifetime consumption. Therefore, it is highly recommendable to perform field measurements at the individual WT to determine its parameterisation and/or validate simulation. For WT types with no design data available, RSL determination based on load measurements in combination with a suitable approach for extrapolation of the measured load spectra to the service life seems advisable.

Approach

Based on experience with independent measurements at more than 1100 WT, the need for individual field measurements is identified by statistics for deviations in fatigue-relevant parameterisation or structural properties. The related increased fatigue loads, under-estimated by simulation with ideal parameters, are shown. Examples are given for the fatigue impact of increased operational loads during turbine fault shut down. Reliable and cost efficient measurement methods are described. The benefit of field measurements for a more accurate determination of the individual WT’s RSL is explained.

Main body of abstract

For many older WT types, fatigue load simulation is hindered by missing design data and operational data recordings, required to determine e.g. the individual power-rotor speed characteristics. Furthermore, for one and the same wind conditions the individual WT’s structural properties (e.g. natural frequencies) as well as its configuration and parameterisation (e.g. blade angles, etc.) influence the fatigue load level. Field measurements at the individual WT reliably determine these individual properties, thus increasing the accuracy of the simulation. Wrong control parameterisation causing tower resonance issues or a changed tower’s natural frequency can be detected.
Furthermore, fatigue-relevant rotor issues (imbalance or blade angle deviation) need to be measured and correctly implemented in the simulation. Otherwise, the RSL is over-estimated which would lead to wrong economic conclusions. Additionally, measures taken to reduce lifetime consumption should be verified by measurements.
Finally, load measurements of several weeks may serve for simulation validation or even allow measurement-based determination of the RSL. Results from a two-year load measurement campaign with many sensors are shown to discuss the suitability of different sensors and measurement positions. The design fatigue load spectra include high loads from operational faults, e.g. pitch system failure. Measurements are presented to assess their impact on the RSL. If design data is missing, the comparison with the tower top load spectrum from the type approval is one proposed possibility for the RSL assessment.

Conclusion

Field measurements at the individual WT provide its parameterisation and configuration which are key points for a correct fatigue load simulation and a realistic determination of its RSL. Reliable measurement methods are available and cost-effective since they increase the accuracy of RSL determination. Load measurements are reasonable for simulation verification and for WT types with missing detailed design data a suitable way for RSL assessment.


Learning objectives
The benefit of using the individual WT's real, not ideal, parameters and characteristics for realistic RSL determination is described.
Field statistics and examples explain typical parameter changes or deviations found impacting on the RSL.
Measurement methods suitable during realistic RSL determination are presented.
Results from a two-year's load measurement campaign will give insight in the real WT's vibration behaviour.
The importance of neither under-, nor over-estimating the RSL is high-lighted because of the impact on drawn conclusions concerning economics and safety.