Session description

This session considers the broad topic of fixed offshore foundation systems and includes papers addressing the primary elements of global or whole-structural-system analysis and assessment. A range of speakers will represent academia and industry with contributions covering different aspects of bottom-fixed support structures and foundations, their design, analysis and optimisation. Topics addressed will include hydrodynamic loads, soil-structure interaction and geotechnical issues, support structure dynamics and simulation technology, field testing and laboratory experiments as well as pile design.

Learning objectives

• Better understand soil-structure interaction mechanisms and analyse methods
• Appreciate how to analyse and assess structural dynamic behaviour
• Examine fatigue damage models applied to offshore wind foundations
• Recognise performance indicators for the whole-structure
• Identify methods to objectively assess optimum foundation configuration

Wout Weijtjens Vrije Universiteit Brussel (VUB), Belgium
Co-authors:
Wout Weijtjens (1) ^{F P} Rasoul Shirzadeh (1) Gert De Sitter (2) Christof Devriendt (2)
(1) Vrije Universiteit Brussel (VUB), Brussel, Belgium (2) Offshore Wind Infrastructure Application Lab (OWI-lab), Brussel, Belgium

Classifying resonant frequencies and damping values of an offshore wind turbine on a monopile foundation for different operational conditions

Introduction

This paper shows the most recent results in the ongoing long-term monitoring campaign on an offshore wind turbine in the Belgian North Sea [1,2]. This contribution will focus on the behavior of the resonant frequencies and damping values of the fundamental modes over a full year, during which all relevant ambient and operational conditions have occurred.

The latest results show that the damping values and resonance frequencies of the support structure are greatly influenced by the operating conditions of the turbine. We will therefore classify the dynamics of the turbine during operating conditions into different reference sets for fatigue-life design.


Approach

The measurement campaign is performed at the Belwind wind farm, which consists of 55 Vestas V90 3MW wind turbines. The wind farm is located in the North Sea on the Bligh Bank, 46 km off the Belgian coast. The campaign started in December 2011 and is currently still generating data.

The structures instrumented in this campaign are the tower and transition piece. Measurements are taken at 4 levels on 9 locations using a total of 10 sensors. The monitoring system measures continuously and sends data every 10 minutes to an onshore server. In order to classify the operating conditions of the wind turbine during the measurements SCADA data (power, rotor speed, pitch angel, nacelle direction) is gathered at a sample rate of 1Hz and the ambient data (wind speed, wind direction, significant wave height, air temperature, ...) is collected at 10 minute intervals.

This paper will analyze the dynamics of the turbine over a period of one year. To process this long period of measurements the monitoring system presented in [2] was used and adapted to automatically track structural parameters even during production. This allows us to process a full year of resonance frequencies and damping values of an operational offshore wind turbine. In this year all relevant operational conditions have occurred, making an investigation into the effect of different operational conditions on the structural parameters possible.

In order to translate these observations into a tool for the cost-effective design of bottom-fixed support structures and foundations, we suggest a classification into different dynamic reference sets. Each reference set is linked to a range of operational conditions and provides the resonance frequencies and damping values representative for those operational conditions. As they are experimental results, the obtained information is complementary to the presently available standards and guidelines related to damping.


Main body of abstract

Many large-scale offshore wind farm projects use monopile foundations to realize a cost effective design. During the design of these monopile structures fatigue-life is a driving design parameter. As such most designed turbines avoid resonant behavior by avoiding that the first Fore-Aft (FA) mode coincides with the wave-period and the first rotor harmonic (1P).
Damping ratios are also crucial for lifetime predictions as the amplitude of vibrations at resonance are inversely proportional to these ratios. The overall damping of the first bending mode of an offshore wind turbine in parked conditions was discussed in [2]. But damping values are highly dependent on operating conditions and might vary significantly over different operational conditions of the turbine. As damping ratios are very difficult to predict by numerical tools and measurements on existing offshore wind turbines are therefore crucial to verify the existing design assumptions [3].

Processing the long-term dataset requires an automated solution to identify and track the resonance frequencies and damping values of the support structure over time. However, the monitoring system designed for parked turbines [2] employs operational modal analysis (OMA) to continuously obtain the structural parameters. Because OMA relies on natural/ambient vibrations to obtain the structural parameters no additional experiment such as e.g. an overspeed stop is necessary. In Figure 1 the four first modes (from left to right: FA1, SS1, SS2 and FA2) identified during parked conditions are presented. As [2] was focused on the dynamics of the parked turbine, the found damping values and resonance frequencies are mostly representative for parked conditions and an extended study into all operational conditions is still preferred.





However, in the presence of harmonic forces due the rotation of the turbine OMA should be used with special care to avoid error-nous estimates and a poor overall monitoring result. This paper will briefly discuss the cause of these potential errors and suggests a strategy to assure a reliable monitoring result. By comparison the original algorithm applied to operational data yields the estimated resonance frequencies of the second (SS2) Side-Side mode in Figure 2. While minor alterations dramatically improve the results and clearly show the influence of the tides on the resonance frequencies, Figure 3.




The altered algorithm is used to process one year of measurements, considering both a parked as well as a rotating turbine. This will result in a better understanding in the fundamental differences between the dynamics, resonance frequencies and damping ratios of a producing and a parked turbine. The obtained resonance frequencies and damping ratios can also be evaluated with regard to the environmental conditions (e.g. wind speed, temperature, …). For example Figure 3 already clearly shows the impact of the tidal level on the resonance frequencies of the second Side-to-Side mode.
However, even larger changes in the structural parameters can be related directly to the operational conditions of the turbine itself. Figure 4 shows that the structural parameters of the entire support structure are affected by a changed operational condition.



In this figure the data, for simplicity, was divided into just two different operational conditions, namely operating and Parked. From this simple example we can already see that some modes switch resonance frequency (e.g. FA3 and SS3) and that for a parked turbine the damping in Side-Side direction is greater that in the Fore-Aft direction, while during rotation this is the other way around.

While the previous example is only considering two (broad) operational conditions, the full paper will consider a larger number of different loading conditions including also environmental conditions and different rotating speeds of the turbine. With each operational condition we can then associate a set of representative resonance frequencies and damping values that define the dynamic behavior of the turbine in that operational condition. In a cost-effective design of the support structure and foundation these reference sets can be used in addition to the current guidelines and standards to optimize for fatigue-life.


Conclusion

The results of the paper show that the dynamics of a turbine support structure are greatly influenced by a change in the operational conditions of the supported turbine. In this abstract we have already shown the major differences in resonance frequencies and damping ratios between a rotating and a parked turbine. As real-life turbines are not fixed to a single operational condition during their full lifetime a design engineer cannot accurately predict fatigue life when these changed dynamics are not properly accounted for.

We will give a better insight in the influence of different operational conditions to the dynamics of an off shore wind turbine on a monopile foundation at the Belwind wind farm, by considering one year of experimental data. In order to process the long-term data have used the monitoring algorithm, based on operational modal analysis (OMA), first presented in [2]. We show the pitfalls of OMA driven monitoring solutions in the presence of rotor harmonics (1P-3P-6P-…) and modified the monitoring algorithm to circumvent these issues. From the observed resonance frequencies and damping values, together with the available SCADAS and meteorological data we’ve identified several relevant dynamic conditions of the turbine each associated with a range of operational and environmental conditions.

This paper suggests to use these experimentally obtained reference sets of damping values and resonance frequencies as a tool for design engineers to predict fatigue-life more accurately for future projects. The presented reference sets can also be considered as training sets for O&M solutions to monitor for scour or loss of stiffness in the foundation/tower structure.



Learning objectives
The goal of our research is to derive all relevant information considering resonance frequencies and damping values from the long-term measurement campaign currently in progress on an offshore wind turbine on a monopile foundation. These findings should help design engineers to achieve a more cost-effective and correct fatigue-life assessment in the construction of future offshore wind infracstructure.


References
[1] C.Devriendt, P.J. Jordaens, G. De Sitter and P. Guillaume Damping estimation of an offshore wind turbine on a monopile foundation, Renewable Power Generation, IET, 2013.
[2] C.Devriendt, P.J. Jordaens, I. Van Ingelgem, G. De Sitter and P. Guillaume, Monitoring of resonant frequencies and damping values of an offshore wind turbine on a monopile foundation, Proceedings of EWEA, Vienna, 2013
[3] J.van der Tempel, Design of Support Structures for Offshore Wind Turbines, PhD Thesis T.U. Delft, 2006