@phdthesis{1103,
  author = {Arne Gloe},
  title = {Influence of grid support functionalities on the mechanical loads and the energy yield of wind turbines},
  abstract = {<p>As the share of inverter connected generation increases in power systems, grid operators worldwide<br />
increase their demand for system services by wind turbines. One of these services, grid frequency<br />
control, is very challenging for wind turbines, as it requires a change of the active power of the wind<br />
turbine. Furthermore, wind turbines are curtailed when the grid capacity is not sufficient to transport<br />
the produced power: the so-called feed-in management. Hence, the control of the wind turbines has<br />
to cope with the needs of the grid. Thus, the wind turbines are exposed to an additional external<br />
excitation in addition to the prevailing wind conditions.<br />
In this thesis, the influence of the providing these services on the wind turbine is analysed. The analysis<br />
is based on time domain simulations of the wind turbine. The models were partly developed in this<br />
project. Additionally, a focus is laid on analysing data to derive realistic scenarios of the grid signals. As<br />
some of these signals were not available, several measurement had to be set up and maintained during<br />
the project.<br />
In a research project with the wind turbine manufacturer Suzlon Energy, a controller for providing<br />
inertial response continuously was developed. In contrast to the state-of-the-art, the magnitude of the<br />
inertial response is scaled with the operating point of the wind turbine in order to provide this service<br />
reliably and at the same time to use the full potential of the wind turbine in strong wind conditions.<br />
This controller is tested with grid frequency scenarios from Europe and India to identify the<br />
consequences for the energy yield and the mechanical loads of the wind turbine. It is shown, that<br />
neither the energy yield nor the loads are affected significantly.<br />
During the research project, a grid islanding and eventually a blackout occurred in Flensburg. The local<br />
grid operator provided data of the grid situation for the day of the blackout. This data allowed to<br />
analyse the cause of the blackout and to develop fictive scenarios how wind turbines could have helped<br />
to stabilize the islanded grid. In the analysed scenarios it was shown, that wind turbines equipped with<br />
an inertial response and a fast frequency response controller could have stabilized the islanded grid<br />
making a blackout less likely. However, the wind turbines were also at risk of running in overspeed, as<br />
they had to reduce their power drastically when stabilizing the grid.<br />
As a consequence of the overspeed problems, a feedforward loop for the pitch control was added to<br />
the grid frequency support controller. It modifies the pitch angle signal from the feedback controller<br />
by a signal depending on the pitch sensitivity and the magnitude of the power change for grid<br />
frequency support. The proposed controller achieved to reduce the overspeed significantly.<br />
In addition to the grid frequency support, a controller for continuous feed-in management was<br />
developed at the institute. It was tested with a fictive case of a weak grid connection, which was<br />
designed based on measurements at the campus in Flensburg. In comparison to the state-of-the-art<br />
feed-in management, the controller allowed a higher energy yield of the wind turbine and a better<br />
utilization of the grid components. The loads of the wind turbine were only slightly increased. Finally,<br />
it is also shown, that feed-in management and inertial response can be provided by a wind turbine at<br />
the same time, as the excitations and the wind turbine response are largely decoupled in the frequency<br />
domain.<br />
It is concluded that wind turbines can provide system services to a larger extend than today without<br />
suffering from unduly harm for its mechanical loads and its energy yield.</p>
},
  year = {2023},
  booktitle = {Department of Energy and Environmental Management},
  journal = {Department of Energy and Environmental Management},
  volume = {Ph.D.},
  pages = {199},
  month = {12/2023},
  publisher = {EUROPA-UNIVERSITÄT },
  address = {Flensburg},
}