INERTIAL RESPONSE FOR GRID STABILITY

Abstract

Provided is a method for controlling wind turbines of a wind park connected to a utility grid, in case of a drop of a grid frequency, the method including controlling each of the wind turbines individually by an individual wind turbine control signal indicating an individual additional wind turbine power to be output by the respective wind turbine, wherein the individual wind turbine control signal is based on: a desired additional wind park power to be supplied from the wind park to the utility grid and an operational characteristic of the respective wind turbine.

Claims

1. Method for controlling one or a plural of wind turbines (103a, 103b, 103c) of a wind park (100, 200) connected to a utility grid (111, 211), in particular in case of a drop of a grid frequency, the method comprising: controlling each of the wind turbines (103a, 103b, 103c) individually by an individual wind turbine control signal (113a, 113b, 113c) indicating an individual additional wind turbine power (107a, 107b, 107c) to be output by the respective wind turbine, wherein the individual wind turbine control signal is based on: a desired additional wind park power (115) to be supplied from the wind park to the utility grid and an operational characteristic (117, 218, 210) of the respective wind turbine.

2. Method according to the preceding claim, wherein the individual wind turbine control signal (113a, 113b, 113c) is further based on the operational characteristics of all other wind turbines of the wind park.

3. Method according to one of the preceding claims, wherein the individual wind turbine control signal (113a, 113b, 113c) is further based on a inertial response profile (123) of the respective wind turbine, the inertial response profile in particular defining a maximally allowed additional power and/or a maximally allowed time range over which the maximal additional power is allowed to be output.

4. Method according to one of the preceding claims, wherein the individual wind turbine control signal (113a, 113b, 113c) is further based on a recovery profile (125) of the respective wind turbine defining recovery parameters of a recovery process re-accelerating the wind turbine rotor, the recovery profile in particular defining a maximally allowed time range over which recovery should be completed or defining maximally allowed power drop.

5. Method according to one of the preceding claims, further comprising: considering individual power losses from an output terminal (105a, 105b, 105c) of the respective wind turbine to a point of common coupling (109) to which the wind turbines are connected such that by controlling the wind turbines with the individual wind turbine control signals the desired additional wind park power (115) is available for the utility grid at the point of common coupling.

6. Method according to one of the preceding claims, wherein the operational characteristic of each wind turbine includes at least one of: an individual capability of inertial response or inertial power, measured or estimated; an individual availability (218) of inertial response or inertial power, measured or estimated; an individual temperature (210) of at least one component, in particular a generator and/or a converter and/or a bearing of a wind turbine rotor, measured or estimated; an individual electrical condition, measured or estimated; an individual wind speed, measured or estimated; an individual rotor speed, measured or estimated of the respective wind turbine.

7. Method according to one of the preceding claims, wherein the individual additional wind turbine power to be output by the respective wind turbine, as governed by individual wind turbine control signal, is the higher: the higher the individual capability of inertial response or inertial power is; and/or the higher the rotor speed is; and/or the lower the temperature of the component is; and/or the higher the individual wind speed is.

8. Method according to one of the preceding claims, wherein based on the operational characteristics of one or a plural of wind turbines and the desired additional wind park power, an optimisation is applied such that a target function is optimized, in particular minimized, the target function including at least one of: park power loss; individual recovery time and/or recovery energy loss of each wind turbine; collective recovery time and/or recovery energy loss of all wind turbines; individual load and/or wear of each wind turbine; individual generated noise.

9. Method according to one of the preceding claims, wherein the optimisation uses electrical and/or mathematical models or a closed loop control.

10. Method according to one of the preceding claims, wherein at least one, in particular at least two, wind turbines are controlled to supply different amounts of power to the utility grid.

11. Method according to one of the preceding claims, wherein the desired additional wind park power (115) comprises desired additional wind park active power, wherein the individual additional wind turbine power (107a, 107b, 107c) comprises an individual additional wind turbine active power.

12. Method according to one of the preceding claims, further comprising: measuring a wind park output power (233), in particular at the point of common coupling (209); summing the desired additional wind park power (215) and a park reference power (229) to obtain a total desired wind park power (231); deriving a difference (235) between the total desired wind park power and the measured wind park output power; supplying the difference (235) or an estimate of the park loss derived using a model of park electrical layout to a closed loop controller (237); supplying an output signal (239) of the controller (237) and the operational characteristics (218, 210) of all wind turbines to a wind park power distribution algorithm (202), that is configured to generate the individual wind turbine control signals (213a, 213b, 213c) based thereon.

13. Wind park controller (101, 201), configured to perform a method according of one of the preceding claims.

14. Wind park (100, 200), comprising: wind turbines (103a, 103b, 103c); and a wind park controller (101) according to the preceding claim.

Description

BRIEF DESCRIPTION

[0039] Some of the embodiments will be described in detail, with references to the following Figures, wherein like designations denote like members, wherein:

[0040] FIG. 1 schematically illustrates a wind park according to an embodiment of the present invention comprising a wind park controller according to an embodiment of the present invention which is configured to carry out a method for controlling one or a plural of wind turbines according to an embodiment of the present invention; and

[0041] FIG. 2 schematically illustrates a wind park according to an embodiment of the present invention comprising a wind park controller according to an embodiment of the present invention which is configured to carry out a method for controlling one or a plural of wind turbines according to an embodiment of the present invention.

DETAILED DESCRIPTION

[0042] The illustration in the drawings is in schematic form. It is noted that in different figures, similar or identical elements are provided with the same reference signs or with reference signs, which are different from the corresponding reference signs only within the first digit.

[0043] The wind park 100 schematically illustrated in FIG. 1 has a wind park controller 101 according to an embodiment of the present invention and further wind turbines 103a, 103b, 103c. The wind park may comprise a larger number of wind turbines, such as larger than 10 or larger than 100. At respective output terminals 105a, 105b, 105c, the wind turbines 103a, 103b, 103c, respectively, deliver their output power 107a, 107b, 107c to a point of common coupling 109 which is connected (optionally via one or more transformers) to a utility grid 111. The wind park controller 101 is configured, to carry out a method for controlling the wind turbines 103a, 103b, 103c, in case of a drop of a frequency of the utility grid 111. The drop of the grid frequency may be detected by measuring the grid frequency and/or a derivative of the grid frequency and comparing the measured frequency and/or derivative of the grid frequency with one or more thresholds.

[0044] The wind park controller 101 thereby supplies individual wind turbine control signals 113a, 113b, 113c to the wind turbines 103a, 103b, 103c, respectively, which indicate the individual additional wind turbine power to be output by the respective wind turbine. Thus, the total power 107a, 107b, 107c of the power output by the wind turbines 103a, 103b, 103c, respectively, is or may be a sum of a nominal wind turbine power and the individual additional wind turbine power as indicated by the respective individual wind turbine control signals 113a, 113b, 113c.

[0045] Thereby, the individual wind turbine control signal 113a, 113b, 113c is based on a desired additional wind park power as indicated by an inertial response park request signal 115 and is also further based on operational characteristics 117 of one or more wind turbines and in particular further based on profiles 123, 125. In the illustrated example, the operational characteristics of all wind turbines 103a, 103b, 103c are supplied to the wind park controller 101 and are considered for deriving the individual wind turbine control signals 113a, 113b, 113c.

[0046] The electrical characteristics of the point of common coupling 109, with respect to for example its frequency, may for example be determined by park measurement equipment 119 which may also deliver measurement values 121 to the wind park controller 101. The individual wind turbine control signals 113a, 113b, 113c may be further derived by the wind park controller 101 taking into account an inertial response profile setup 123 and an inertial response recovery profile setup 125 which may define operational borders of the inertial response and the recovery from the inertial response, respectively.

[0047] From the individual wind turbine output terminals 105a, 105b, 105c to the point of common coupling 109, a power loss may occur which may be taken into account for defining the individual wind turbine control signals 113a, 113b, 113c. Thus, the wind turbines may deliver an inertial response that they are individually capable of in response to a frequency derived measurement yielding a full response to the event and consequently an optimized but yet necessary recovery profile in order to recover the lost energy. This is however not the response necessary as what TSO's often require is a predefined response magnitude for a predefined length and time from an entire plant.

[0048] According to an embodiment of the present invention, the park response inertial response may optimize the delivery of primary inertial response as seen from a park level based on the requirements to the park response, yielding two primary beneficial factors: [0049] 1) Ensured level of park response to the requirements needed on park level compensating for park losses to the extent possible (wind generator limitations). [0050] 2) Distributing the needed inertial response on a park level amongst the turbines in an optimized manner, thus limiting the recovery profile to a minimum, in particular minimizing the recovery time. When looking at the inertial response over an entire park, turbines are experiencing different wind conditions and thus possibly different rotor speeds and consequently inertia in the rotor. Some turbines may therefore be able to deliver more power to the inertial response than others and also the recovery profile may look differently depending on the different rotational speeds of the rotors.

[0051] Based on, e.g., an individual estimation of the inertia or potential inertial power, the wind farm or plant controller may optimize the response using electrical/mathematical models or closed loop control. Thereby, some turbines may contribute more to the inertial response than others as they are subject to better conditions to do so than other turbines. The optimization may also take into account the electrical and temperature condition of the turbines in order to optimize the response even further.

[0052] According to embodiments of the present invention, the park level control may ensure that the desired inertial response output on a park level is delivered meaning if the desired inertial response level is less than the collective capability of the turbines, then the wind park controller may control the turbines to produce the requested inertial response on park level in the point of common coupling. If maximum possible inertial response from a turbine is requested, or close to it, the park control cannot necessarily compensate for the park losses.

[0053] The following embodiments are included in the present application: [0054] a) One embodiment would be if each individual wind turbine would be able to compensate for park losses and adapting the response based on mathematical models (upstream compensation). [0055] b) One embodiment would be if the turbines were able to communicate between each other to collectively determine the optimal individual response. [0056] c) One embodiment would be a fixed response based on individual wind speed or inertial estimation, with a compensation based on plant level measurement.

[0057] FIG. 2 schematically illustrates a wind park 200 comprising a wind park controller 201 according to an embodiment of the present invention which is configured to carry out a method according to an embodiment of the present invention. Features similar in structure and/or function in FIGS. 1 and 2 are labelled with the same reference sign differing only in the first digit.

[0058] Embodiments of the present invention ensure that a specified inertial response at the park level is achieved compensating for losses in the park electrical network to the extent possible in the actuators (turbines) and the minimization of the impact (recovery period) of inertial response on a park level.

[0059] FIG. 2 thereby is a representation of a possible concept that may implement an embodiment of the present invention. There may be several other implementations that may yield the same closed loop response with e.g. with a feed forward.

[0060] The wind turbines 203a, 203b, 203c output their respective wind turbine powers 207a, 207b, 207c to the point of common coupling 209 which is connected to the utility grid 211. The turbines output the available inertia indicating signal 218a, 218b, 218c to the wind farm power distribution algorithm 202 comprised in the wind park controller 201. Further, the wind turbines 203a, 203b, 203c output operational characteristics, in particular temperatures of one or more components 210a, 210b, 210c to the wind farm power distribution algorithm 202.

[0061] Furthermore, the inertial response park request 215 is added using an additional element 227 with a park reference 229 to obtain a total requested park power signal 231. The actually delivered power to the utility grid at the point of common coupling 209 is measured as a signal 233 which is subtracted from the total requested park power output 231 to obtain an error signal 235 which is provided and supplied to a wind farm closed loop controller 237. The wind farm closed loop controller 237 for example comprises a P controller and/or an I controller and/or D controller component (or any other form of controller like e.g. linear-quadratic regulator (LQR)) in order to derive a control signal 239 at its output which is configured such as to decrease the error signal 235. The error signal is supplied to the wind farm power distribution algorithm 202 which also takes this control signal 239 into account to derive the individual wind turbine control signals 213a, 213b, 213c which are supplied to the turbines 203a, 203b, 203c, respectively.

[0062] When taking into account the requested inertial response in the reference for the park or plant controller, it may be ensured that the requested inertial response is delivered in the point of common coupling 209. This may eliminate the effects of the park electrical losses to yield a more definable and uniform response from the plant. This is of course dependent on the individual turbine's ability to deliver the necessary inertial response to compensate for those losses, i.e. turbines not in actuator limitation.

[0063] Embodiments of the present invention may introduce a smart distribution of the desired park response within the wind park. Each individual wind turbine in the park may in many cases see different wind and temperature conditions or simply there may be different turbine types with different capabilities which for example could be limited in some cases by noise restrictions (less available inertia). Each turbine may deliver an estimate of the available inertia in the turbine based on measurements of the rotor speed and the turbine setup. Along with one or several measurements of temperature, the turbine (or an aggregation thereof) the two may be used to distribute the requested park level response amongst the turbines in an optimum manner to reduce the recovery energy loss and/or recovery time in the park. Thereby, some turbines may contribute more to the inertial response than others as they are subject to better conditions to do so than other turbines.

[0064] The following example may illustrate embodiments of the present invention. The turbine 203a may have an available inertia of 6 and a temperature (may be several temperatures or an aggregation) of 10, The turbine 203b may have an available inertia of 4 and a temperature of 29. The turbine 203c may have an available inertia of 4 and a temperature of 89 Which is near the high limit of that turbine.

[0065] The park is requested to deliver an inertial response of 8 and therefore the closed loop may in turn compensate for a park loss of 2 yielding a need of 10 from the turbines as the controller converges. The distribution algorithm may now control turbine 203a to contribute the most as it has the best conditions to contribute the most and the least from turbine 203c as it has the worst conditions to contribute the least. One such distribution could then be to request 7 from turbine 203a to request 2 from turbine 203b and to request 1 from turbine 203c.

[0066] The exact amounts of additional wind turbine power requested from the different wind turbines may be derived due to an algorithm which may depend on the particular application and constitution of the wind park.

[0067] In the example as illustrated in FIG. 1, turbine 103a may indicate an inertial response availability of 3 and a low temperature, turbine 103b may indicate an inertial response availability of 2 and a medium temperature and the wind turbine 103c may indicate an inertial response availability of 1 and a high temperature. Accordingly, as an exemplary scenario, the park control 101 may request inertial response of 3 from turbine 103a, may request an inertial response of 1 from turbine 103b and may request an inertial response of 0 from turbine 103c.

[0068] Although the invention has been illustrated and described in greater detail with reference to the preferred exemplary embodiment, the invention is not limited to the examples disclosed, and further variations can be inferred by a person skilled in the art, without departing from the scope of protection of the invention.

[0069] For the sake of clarity, it is to be understood that the use of a or an throughout this application does not exclude a plurality, and comprising does not exclude other steps or elements.