Grid frequency control in an offshore wind farm

Abstract

A method of controlling a wind farm including a plurality of wind turbines installed in an AC subgrid, which is connected via a high-voltage rectifier and a DC line to an energy conversion device, the method including: controlling a power infeed of each wind turbine to maintain a setpoint grid frequency of the subgrid; sensing a grid frequency in the subgrid; determining a reference quantity based on a difference of the sensed grid frequency and the setpoint grid frequency; and controlling the energy conversion device in accordance with the determined reference quantity.

Claims

1. A method of controlling a wind farm including a plurality of wind turbines installed in an AC subgrid, which is connected via a high-voltage rectifier and a DC line to an energy conversion device, the method comprising: controlling a power infeed of each wind turbine to maintain a setpoint grid frequency (.sub.1) of the subgrid; sensing a grid frequency (.sub.g) in the subgrid; determining a reference quantity based on a difference of the sensed grid frequency and the setpoint grid frequency; and controlling the energy conversion device in accordance with the determined reference quantity.

2. The method of claim 1, wherein the reference quantity includes a time integral of the difference of the sensed grid frequencies.

3. The method of claim 1, wherein the reference quantity further includes a term proportional to the difference of the sensed grid frequencies and/or a time derivative of the difference of the sensed grid frequencies.

4. The method of claim 1, wherein the reference quantity is an active output power reference or an active output current reference of the energy conversion device.

5. The method of claim 1, wherein the energy conversion device is a high-voltage DC converter which connects the DC line to an AC transmission grid.

6. The method of claim 1, wherein the energy conversion device is an energy storage device, such as a power-to-hydrogen converter or a battery.

7. The method of claim 1, wherein the power infeed of each wind turbine is controlled by means of a droop control algorithm, in particular by means of active-power/grid-frequency droop.

8. The method of claim 1, wherein the power infeed of each wind turbine is controlled in accordance with a variable active-power reference.

9. The method of claim 1, wherein the grid frequency in the subgrid is sensed in the vicinity of the high-voltage DC rectifier, which is arranged at the subgrid side of the DC line.

10. The method of claim 1, which is implemented in a first controller arranged in the vicinity of the energy conversion device, wherein sensing the grid frequency in the subgrid comprises communicating the sensed grid frequency, or a quantity derived therefrom, to the first controller.

11. The method of claim 1, which is implemented in a second controller (42) arranged in the vicinity of the wind farm, wherein controlling the energy conversion device comprises communicating the determined reference quantity to the energy conversion device.

12. The method of claim 10, wherein said communicating step is performed using a radio link, optical link or wired data link.

13. The method of claim 1, wherein the subgrid is an offshore grid.

14. A controller for controlling a wind farm including a plurality of wind turbines installed in an AC subgrid, which is connected via a high-voltage rectifier and a DC line to an energy conversion device, the controller configured to: control a power infeed of each wind turbine to maintain a setpoint grid frequency (.sub.1) of the subgrid; sense a grid frequency (.sub.g) in the subgrid; determine a reference quantity based on a difference of the sensed grid frequency and the setpoint grid frequency; and control the energy conversion device in accordance with the determined reference quantity.

15. A computer program comprising instructions to cause a controller to execute a method, the controller for controlling a wind farm including a plurality of wind turbines installed in an AC subgrid, which is connected via a high-voltage rectifier and a DC line to an energy conversion device, the controller configured to: control a power infeed of each wind turbine to maintain a setpoint grid frequency (.sub.1) of the subgrid; sense a grid frequency (.sub.g) in the subgrid; determine a reference quantity based on a difference of the sensed grid frequency and the setpoint grid frequency; and control the energy conversion device in accordance with the determined reference quantity, the method including: controlling a power infeed of each wind turbine to maintain a setpoint grid frequency (.sub.1) of the subgrid; sensing a grid frequency (.sub.g) in the subgrid; determining a reference quantity based on a difference of the sensed grid frequency and the setpoint grid frequency; and controlling the energy conversion device in accordance with the determined reference quantity.

16. The method of claim 2, wherein the reference quantity is an active output power reference or an active output current reference of the energy conversion device.

17. The method of claim 2, wherein the reference quantity is an active output power reference or an active output current reference of the energy conversion device.

18. The method of claim 2, wherein the energy conversion device is a high-voltage DC converter which connects the DC line to an AC transmission grid.

19. The method of claim 2, wherein the energy conversion device is an energy storage device, such as a power-to-hydrogen converter or a battery.

20. The method of claim 2, wherein the power infeed of each wind turbine is controlled by means of a droop control algorithm, in particular by means of active-power/grid-frequency droop.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Aspects and embodiments are now described, by way of example, with reference to the accompanying drawings, on which:

(2) FIG. 1 is a schematic depiction of a wind farm, in which multiple wind turbines are installed in an AC subgrid, wherein electric power is transferred away from the subgrid via a DC line;

(3) FIG. 2 is a flowchart of a method for controlling a wind farm according to embodiments herein; and

(4) FIG. 3 is a more detailed illustration of the wind farm with associated control equipment.

DETAILED DESCRIPTION

(5) The aspects of the present disclosure will now be described more fully hereinafter with reference to the accompanying drawings, on which certain embodiments of the invention are shown. These aspects may, however, be embodied in many different forms and should not be construed as limiting; rather, these embodiments are provided by way of example so that this disclosure will be thorough and complete, and to fully convey the scope of all aspects of invention to those skilled in the art. Like numbers refer to like elements throughout the description.

(6) FIG. 1 shows, in its left-hand part, a wind farm 10 including a plurality of wind turbines 11, which are installed in an AC subgrid 12. In the terminology of the present disclosure, a wind farm (or wind power station, or wind park) refers to a group of wind turbines installed on one site. Usually, the horizontal spacing of the wind turbines in a wind farm is minimal or near-minimal, in the sense that a closer spacing would imply significant energy losses due to blockage effects. The acceptable level of energy losses may be related to the applicable monetary and/or environmental cost of using the land that the wind farm occupies. In a conventional installation at the time of this disclosure, the spacing of identical wind turbines may typically be between 3 and 4 times the rotor diameter. The wind farm 10 may include tens of wind turbines 11; their total number may for example be N=50.

(7) The present disclosure addresses in particular such wind farms where the power infeed of each wind turbine 11 is controlled by means of a droop control algorithm, in particular by means of active-power/grid-frequency droop. This is to say, the difference between a setpoint grid frequency .sub.1 [rad/s] and the actual grid frequency .sub.g,n at the nth wind turbine 11 is linked to its respective output active-power reference P.sub.ref,n and actual output active power P.sub.n as follows:

(8) $\begin{array}{cc}{}_1-{}_{g,n}=-K_p\left(P_{ref,n}-P_n\right),& \left(1\right)\end{array}$
where K.sub.p is a positive control gain. The power reference P.sub.ref,n can be fixed or variable. A variable power reference P.sub.ref,n can be set by as a function of the windspeed, with a view to maintaining an equilibrium of input power and output power (and losses) in an HVDC converter; in particular, the power reference P.sub.ref,n may need to be set higher when the windspeed increases, or else the voltage on the DC line 20 will increase. The setpoint grid frequency may be a nominal grid frequency, such as exactly 50 Hz or another specified value. Instead of the control law (1), one may use a filter with a static gain of K.sub.p; by means of the filter, a desired amount of virtual inertia and/or virtual damping can be introduced into the droop control algorithm.

(9) One way of implementing such control is by means of power-synchronization control (PSC), as described in [4], where the control law (1) is solved for .sub.g,n and integrated into the voltage angle at the converter terminals.

(10) Another way of implementing (1) is by use of a so-called grid-forming phase-locked loop (PLL), as described in [5]. Because all turbines must operate in synchronism, it holds that .sub.g,n=.sub.g in the steady state, where .sub.g can be described as a global subgrid frequency. This way, the effect of a nonzero deviation in power, P.sub.ref,nP.sub.n0, will be a deviation of .sub.g from .sub.1. Specifically, an increase of P.sub.ref,n without a corresponding matching of P.sub.n will normally lead to an increase in .sub.g,n and thus to an increase in the global subgrid frequency .sub.g.

(11) The subgrid 12 is connected via a DC line 20 to an energy conversion device 21. As an interface between the subgrid 12 and the DC line 20, there is installed an HVDC rectifier 22. The HVDC rectifier 22 may be a passive device built using series-connected diodes. The DC line 20 may be a sea cable that connects an offshore wind farm 10 to an onshore energy conversion device 21. It could have a total length of one or more kilometers. The DC line 20 shown in FIG. 1 may as well correspond an onshore-onshore link or an offshore-offshore link.

(12) In the example use case shown in FIG. 1, the energy conversion device 21 is an HVDC converter, which connects the DC line 20 to an AC transmission grid (or mains grid) 30. In alternative use cases, the energy conversion device 21 may be an energy storage device configured converts incoming DC energy into chemical energy in some periods and re-generates electric energy in other periods. For example, the re-generated electric energy may be fed into a transmission grid. The energy storage device may include a battery. The energy storage device may further be a power-to-hydrogen converter, including an electrolytic cell, a gas storage tank and a fuel cell for electricity generation.

(13) As will be described in greater detail below, the energy conversion device 21 and the HVDC rectifier 22 may be provided with respective controllers 41, 42, which are operable to exchange information over a communication link 50, such as a radio link, optical link or wired data link. The wired data link may in particular include a modulated signal that is superimposed on the current flowing on the DC line 20 (power-line communication.)

(14) The inventors are proposing a method 200 for controlling a wind farm 10 comprising the steps depicted in the flowchart of FIG. 2. It is understood that some of these steps are optional (i.e., absent from some embodiments of the method 200). Further, the steps need not be disjoint in time; rather, their execution can overlap in time.

(15) The step 210 of controlling a power infeed of each wind turbine to maintain a setpoint grid frequency has been described above, e.g., by a droop control algorithm. It is understood that this control can continue in parallel to the execution of the further steps of the method 200. This is suggested graphically by the loop in FIG. 2.

(16) In a step 212, a grid frequency .sub.g in the subgrid is sensed. The grid frequency .sub.g may be measured by a PLL. The sensing or measuring may be performed. For example, the sensing or measuring may take place in the vicinity of the HVDC rectifier 22, e.g., at a point in the subgrid that is separated by a low or negligible impedance from the HVDC rectifier 22.

(17) In a next step 214, a reference quantity based on a difference of the sensed grid frequency and the setpoint grid frequency .sub.1 is determined. The reference quantity may be an active output power reference p.sub.ref or an active output power-producing current reference i.sub.p,ref of the onshore HVDC converter 21. A pure integrator can be used, which can be expressed as:

(18) $\begin{array}{cc}{p}_{ref}=-K_i\left({}_1-{}_g\right)\mathrm{dt}& \left(2a\right)\end{array}$
or, in terms of operator corresponding to the Laplace transform variable s=d/dt, as

(19) $\begin{array}{cc}{p}_{ref}=-\frac{K_i}{s}\left({}_1-{}_g\right).& \left(2b\right)\end{array}$

(20) In some embodiments of the control method 200, the reference quantity further includes a term proportional to the difference of the sensed grid frequencies (P regulator) and/or a time derivative of the difference of the sensed grid frequencies (D regulator). This can be used to fine-tune the behavior of the control loop. Instead of using a constant setpoint frequency .sub.1 in (2a)-(2b), such as the nominal grid frequency, it is possible to apply a reference frequency that transiently deviates from the nominal grid frequency. This enables, in the active powers from the wind turbines 11, a transient response to frequency variations in the main grid (a so-called frequency response) and to similar perturbations.

(21) The energy conversion device 21 is controlled, in a step 216, in accordance with the reference quantity. Also step 216 can last for the remaining duration of the method 200.

(22) Now, if .sub.g increases due to an increase of one or more of the P.sub.ref,n values (step 218), then p.sub.ref starts to increase as well. Summing over N wind turbines yields:

(23) $\begin{array}{cc}\left({}_1-{}_g\right)N=-K_p\left(P_{ref}-P\right){}_1-{}_g=-\frac{K_p}{N}\left(P_{ref}-P\right),& \left(3\right)\end{array}$
where P=.sub.nP.sub.n is the total active power from the wind farm and P.sub.ref=.sub.nP.sub.ref,n its reference. Neglecting any communication time delay, influence of dynamics (e.g., inertia, rotating masses and other energy storage in the running system) and energy dissipation, so that an instantaneous power balance P=p.sub.ref will establish, and substituting (3) into (2b), one obtains:

(24) $P=\frac{K_i{K}_p}{sN}\left(P_{ref}-P\right)P=\frac{1}{sT+1}{P}_{ref}.$
In other words, P in the controlled system is expected to respond to P.sub.ref as a first-order system with a time constant of T=N/K.sub.iK.sub.p.

(25) In one group of embodiments, the described method 200 is implemented in a first controller 41 arranged in the vicinity of the energy conversion device 21 (e.g., the first controller 41 has a communication link to the energy conversion device 21 with acceptable reliability and capacity, preferably a wired communication link). In particular, the first controller 41 may be arranged at the energy conversion device 21 or may be an integral component thereof. In this case, the sensing of the grid frequency in the subgrid (step 212) comprises a substep 212.1 of communicating the sensed grid frequency to the first controller 41. Alternatively, a quantity derived from the sensed grid frequency, such as the difference .sub.1-.sub.g or a time integral thereof, is communicated to the first controller 41. For this purpose, the communication link 50 may be utilized.

(26) FIG. 3 relates to such embodiments where the method 200 is implemented in the first controller in the vicinity of the energy conversion device 21. In detail, a PLL 14 is arranged to estimate the subgrid frequency .sub.g from a time evolution of a grid voltage vg. The subgrid frequency .sub.g is forwarded over a communication link 50, which is in this example maintained by two wireless interfaces 51, 52. A subtractor component 41.1 of the first controller computes the difference .sub.1.sub.g and feeds it to an integral controller 41.2. The output of the integral controller 41.2this is the reference quantity p.sub.ref or i.sub.p,refis used as one control signal to the energy conversion device 21.

(27) FIG. 3 further illustrates in some detail an example structure of the HVDC converter, which converts a three-phase AC voltage into a DC voltage by means of an YYD connection and two diodes. The unit 13 in FIG. 3 represents harmonic filters and/or a reactive compensation unit. As suggested, the unit 13 is an optional placement for the second controller 42 (see next paragraph) if no first controller 41 is provided.

(28) In another group of embodiments, the method 200 is implemented in a second controller 42, which is arranged in the vicinity of the wind farm 10, e.g., where it is convenient to sense the subgrid frequency .sub.g. In particular, the second controller 42 may be installed at or in the HVDC rectifier 22 or at the unit 13. In this group of embodiments, the controlling of the energy conversion device 21 (step 216) comprises a substep 216.1 of communicating the determined reference quantity to the energy conversion device 21. Substep 216.1 may be executed by means of a communication link 50 as illustrated in FIG. 1.

(29) Furthermore, the method 200 can be implemented in a distributed manner, e.g., with functionality in both the first and the second controllers 41, 42, and optionally with the support of networked (cloud) processing resources.

(30) The aspects of the present disclosure have mainly been described above with reference to a few embodiments. However, as is readily appreciated by a person skilled in the art, other embodiments than the ones disclosed above are equally possible within the scope of the invention, as defined by the appended patent claims.