Share this page on:

Conference programme 

Back to the programme printer.gif Print

Poster session

Lead Session Chair:
Stephan Barth, Managing Director, ForWind - Center for Wind Energy Research, Germany
Christof Devriendt OWI-lab / VUB, Belgium
Co-authors:
Christof Devriendt (1) F P Wout Weijtjens (1) Tim Verbelen (1) Gert De Sitter (1)
(1) OWI-lab / VUB, Brussels, Belgium

Printer friendly version: printer.gif Print

Poster
Download poster(1.16 MB)

Presenter's biography

Biographies are supplied directly by presenters at OFFSHORE 2015 and are published here unedited

Christof Devriendt is scientific coordinator of “ The Offshore Wind Infrastructure Lab” or OWI-lab (www.owi-lab.eu). His main research-objective is cost reduction by increasing the availability and reliability of wind turbines, by optimizing the operation and maintenance management of wind farms by using state-of-the are monitoring solutions.

Abstract

Monitoring the structural health of a monopile foundation: keeping an eye on the resonance frequencies.

Introduction

Monopile foundations are currently the most common type of offshore substructures for wind turbines. This contribution introduces a method to monitor the overall health of the substructure through a constant monitoring of the tower resonance frequencies. In particular changes in the soil conditions, e.g. scour, passing sand waves, soil densification will alter these resonance frequencies. A method relying on resonance frequencies allows detecting these changes without the necessity for bathymetry. Of special interest is to guarantee that tower resonances do not shift towards wave frequencies and rotor harmonics, as this will greatly affect fatigue life.

Approach

The current contribution uses measurements conducted at the Belwind offshore wind farm 46km outside the Belgian coast. One of the 55 Vestas V90-3.0MW turbines on monopile foundations was equipped with an array of accelerometers in 2011. The obtained measurements are translated to resonance frequencies through operational modal analysis. As these parameters are affected by the environmental and operational condition of the wind turbine (EOC) a data-normalization has to be performed to compensate for this variability. Once compensated the resonance frequency can be used as a health indicator for the offshore wind turbine.

Main body of abstract

For this contribution three years of measurements were processed with state-of-the-art operational modal analysis. This resulted in a database of more than 100000 instantaneous resonance frequencies and damping ratios. While the damping ratios are a great input to design, the resonance frequencies can be tracked to detect changes in the overall health of the built structure.
However, the resonance frequencies can shift significantly between different operating conditions. By dividing the data into different operational cases the resonance frequencies for each operating condition are obtained. A further analysis of the found frequencies showed that they also are affected by the environmental conditions. The proposed method allowed detecting the effect of the tidal level upon the resonance frequencies. Moreover, it is so sensitive even a dependency of the resonance frequencies to secondary structures was detected, e.g. an overall stiffening of the turbine when it was aligned with the J-Tube. After a training period of one year, a health index, which is unaffected by the environmental conditions, is now continuously monitored over time.
First results show a gradual stiffening of the turbine over time. A stiffening that could only be detected through the data-normalization, as it is smaller than the natural variability of the resonance frequency. Future monitoring should confirm whether this stiffening continuous over time or converges to a final condition.


Conclusion

In this contribution we showed the potential for a resonance frequency based health monitoring strategy for offshore wind turbines. It allows detecting changes in the soil without the necessity to perform underwater measurements. The proposed method was sensitive enough to detect the effect of secondary structures on the resonance frequency. Over recent years an overall stiffening of the turbine was observed. Widespread deployment of the proposed method will allow to monitor existing structures and cut O&M costs and reduce the risks and insurance-fees of currently installed wind turbines.



Learning objectives
The main goal of this contribution is to demonstrate the feasibility a foundation monitoring strategy for offshore substructures. Ultimately, reducing the cost of O&M through early detection and reduce the insurance fees of operational wind farms.