Validity of Identified Modal Parameters based on Rotor-Stop Tests of Offshore Wind Turbines

Validity of Identified Modal Parameters based on Rotor-Stop Tests of Offshore Wind Turbines

Rhersellah, O.

Metrikine, A.V. (mentor)
Versteijlen, W.G. (mentor)

Civil Engineering and Geosciences

Structural Engineering

Structural Mechanics

2013-07-20

The increasing energy prices together with the uncertainty about long term sufficiency of fossil fuel resources, confirm the increasing need for reliable alternatives based on renewable energy. Offshore wind power, particularly, has a great potential in providing a large share of the energy demand. The sustainable character of this energy fulfills the idealistic aspirations of some, however, the real debate is still concentrated around its economical disadvantage compared to the conventional energy sources. For this reason, great efforts are made to bring the Levelised Cost of Energy down by various research and development projects aiming to increase the efficiency and to reduce the risks related to offshore wind parks. One of the components that keep the costs linked to offshore wind high is the support structure of the turbine. To design this last, the turbine manufacturers rely on Finite Element Models to simulate the dynamic response of the turbine under different load cases. This is done to calculate the damage accumulation and consequently, estimate the fatigue life of the support structure. A particularly relevant parameter therein is the damping experienced by the turbine during its life time. However, this parameter is the most difficult to estimate among all other dynamic properties of the structure. The damping magnitude is quantified with critical damping ratios and their currently used values in design are considered conservative. If higher damping ratios can be proven, significant saving on material can be achieved, or longer fatigue life can be certified. To estimate these damping parameters from experimental data, free decay tests are carried out by means of rotor-stops. The damping estimated values based on these tests are found to have a high scatter leading to uncertainty about their real magnitude. Two important features in dynamic systems must be observed in order for damping to be represented accurately enough in terms of damping ratios. These features consist of low damping and well separated natural frequencies. These conditions lead to small off-diagonal terms in the modal damping matrix that are responsible for coupling the modes of vibrations originally derived for the undamped system. The physical reasons for this modal interaction to occur is the proximity of natural frequencies and non-proportional damping. When testing the structure in order to estimate the damping ratios, caution must be paid to the sensitivity of the structure to the excitation put into the system when nonproportional damping is applicable. In laboratory conditions, such aspects can be controlled. However, during rotor-stops, nothing guarantees that the force vector applied to the structure is so precise such as to excite a single mode of the tower. The interaction of the first bending modes in the fore-aft and side-side is probable under the effects of aerodynamic damping. In this case, estimating the damping based on the assumption that a single mode is excited can give biased results for the damping ratios. For this reason, the applicability of modal coupling during rotor-stop experiments is investigated in this thesis. The possible aerodynamic contribution to the damping experienced by the support structure has other implications on the interest of rotor-stop tests to provide damping ratios the turbine manufacturer is interested in. These damping ratios are needed to account only for soil, hydrodynamic and structural damping. As a first step, a hypothesis is formulated so that when verified, conclusions about the occurrence of modal coupling can be made. This hypothesis is based on assuming that the tower has coupled or decoupled degrees-of-freedom would not make a difference in the identified parameters if modal coupling does not occur. To identify these damping parameters, an identification approach is formulated. This approach uses simplified models to represent the coupling effects of the rotor on the tower. The identification procedure is based on solving an optimisation problem, using as input the main features of the system’s response. These features, in accordance with the linear response theory, are divided between the real and imaginary components of the frequency representation. The imaginary part is of high importance as it contains information about how dissipation occurs in the system. The results of application of this identification procedure to test data have shown that modal coupling is applicable in the specific rotor-stop tests. The aerodynamic damping is found to have a contribution to the damping of the support structure. Compared to the values found in previous research based on similar data within SIEMENS (W.G. Versteijlen), the identified damping for the fore-aft mode is lower in the present thesis. The specific rotor-stop tests used in this thesis are considered not valid to provide the damping ratios SIEMENS is interested in as long as the aerodynamic component is not extractable from the identified ratios so that it would be possible to obtain the damping ratios applicable to the support structure that is primarily linear and has decoupled modes of motion. From this study, it appears that it is of primary importance to study the aerodynamic contribution in detail. The availability of the drag terms for the blades at different configurations and wind speeds can improve considerably the identification of the damping of the support structure.

offshore wind support structures
damping
rotor stop

http://resolver.tudelft.nl/uuid:03d9b08b-84d8-4cc9-8f4a-00d372760ac6

Student theses

master thesis

(c) 2013 Rhersellah, O.