System and method for stabilizing a weak power grid connected to one or more wind farms

Abstract

A method for controlling a wind farm electric power system is provided. The wind farm electric power system includes a controller and a plurality of wind turbines electrically connected to a power grid via an interconnection point. Each wind turbine includes a voltage regulator. The method includes receiving, via the controller, one or more electrical signals associated with the interconnection point in the frequency domain. Furthermore, the method includes estimating, via an estimator of the controller, a voltage sensitivity of the power grid using the one or more electrical signals. Furthermore, the method includes dynamically controlling the voltage of the wind farm electric power system at the interconnection point based on the voltage sensitivity.

Description

Technical Field

[0001] The present disclosure relates generally to systems and methods for controlling a wind farm having one or more wind turbines, and more particularly to systems and methods for stabilizing a weak grid to which one or more wind farms are connected. Background Art

[0002] Wind power is considered one of the cleanest and most environmentally friendly energy sources currently available, and wind turbines are gaining increasing attention in this regard. Existing electrical power distribution systems (e.g., power grids) can be used to distribute power from renewable energy sources (such as wind) by using control systems and methods to coordinate the power generated by renewable energy sources, the power demand on the power distribution system, and the power consumed based on the varying operating conditions inherent in renewable energy sources. For example, the operating conditions of a wind turbine may vary based on wind speed or calmness.

[0003] Wind power does not always have a constant power output, but can include variations; therefore, operators of power distribution systems must account for this. For example, one consequence is that distribution and transmission networks become more difficult to manage. This also involves managing resonances in power distribution systems (including wind turbines). Like conventional power plants, wind turbines or wind farms must be managed or controlled to provide stable power to the grid (e.g., constant voltage and frequency, minimal disturbances, and low harmonic emissions) to ensure reliability and adequate power delivery.

[0004] With respect to renewable energy devices, such as wind farms, since these devices may be located in remote locations, their connection to the power grid may include long, high-voltage transmission lines. Transmission lines (i.e., power cables) and additional electrical infrastructure (e.g., transformers, reactors, capacitors) can cause resonances at low frequencies (e.g., below the second or third harmonics). Furthermore, due to the remoteness or harsh conditions in which many wind farms are located, wind farms are often integrated with weak power grids, which can be negatively impacted by the resonances, causing circuit overloads at the point of interconnection (POI). As a result, oscillations can occur in the POI phase voltages.

[0005] In such cases, grid management agencies require wind farm operators to reduce the power supplied to the grid to ensure an appropriate short circuit ratio (SCR). While this action ensures grid stability, it also means a loss in annual energy production (AEP) for the wind farm. Moreover, integrating inverter-based sources (such as wind farms) into weak grids can bring many challenges, such as the potential for generating resonant conditions in the system, which can be mitigated through different methods (including strengthening transmission lines or integrating additional equipment into the grid to increase strength). These methods of grid strengthening include disadvantages, including additional space requirements, multiple control locations and settings, increased system component costs, increased system installation costs, and increased system maintenance costs.

[0006] In view of the foregoing, systems and methods for stabilizing a weak grid (to which one or more wind farms are connected) without negatively impacting AEP would be welcome in the art.As such, the present disclosure relates to dynamically stabilizing a grid using frequency domain and machine learning. Summary of the Invention

[0007] Aspects and advantages of the invention will be set forth in part in the following description, or may be obvious from the description, or may be learned through practice of exemplary embodiments of the disclosure.

[0008] In one aspect, the present disclosure relates to a method for controlling a wind farm electric power system. The wind farm electric power system includes a controller and a plurality of wind turbines electrically connected to a power grid via an interconnection point. Each wind turbine includes a voltage regulator. The method includes receiving, via the controller, one or more electrical signals associated with the interconnection point in the frequency domain. Furthermore, the method includes estimating, via an estimator of the controller, a voltage sensitivity of the power grid using the one or more electrical signals. Furthermore, the method includes dynamically controlling the voltage of the wind farm electric power system at the interconnection point based on the voltage sensitivity.

[0009] In an embodiment, the electrical signal may include any one or a combination of the following: phase voltage, phase current, active power and / or reactive power at the interconnection point.

[0010] In another embodiment, the method may include measuring the electrical signal via at least one sensor, or determining the electrical signal via a computer-implemented model of the controller.

[0011] In further embodiments, the method may include processing the electrical signal associated with the interconnection point prior to estimating the voltage sensitivity. For example, in embodiments, processing the electrical signal may include filtering, sorting, or the like, or a combination thereof.

[0012] In additional embodiments, using electrical signals to estimate the voltage sensitivity of the power grid may, for example, include modeling the power grid as a linear time-invariant system in the frequency domain via a controller, monitoring the frequency domain for disturbances, determining whether the disturbance corresponds to the wind farm electric power system or to an adjacent wind farm electric power system, and removing the disturbance corresponding to the adjacent wind farm electric power system so as to isolate the effect of the disturbance on the voltage sensitivity.

[0013] More particularly, in an embodiment, determining whether the disturbance corresponds to the wind farm electric power system or to an adjacent wind farm electric power system may include grouping disturbances of at least one of active power or reactive power by phase voltage in the frequency domain and removing ungrouped disturbances.

[0014] In an embodiment, dynamically controlling the voltage of the wind farm electrical power system based on the voltage sensitivity may include varying at least one parameter of one or more voltage regulators to avoid grid instability. In certain embodiments, the parameter may include, for example, a regulator gain, an active power setpoint, a reactive power setpoint, a combination thereof, or any other suitable parameter.

[0015] In several embodiments, the method may include processing electrical signals associated with the interconnection point via a post-processor of the controller to determine an error analysis of parameters of the voltage regulator after dynamically controlling the voltage of the wind farm electrical power system based on the voltage sensitivity.

[0016] In further embodiments, the method may further include receiving feedback from the post-processor via a machine learning algorithm of the controller, and training the feedback via the machine learning algorithm. In such embodiments, the method may include using the output of the machine learning algorithm to generate one or more control commands for an estimator of the controller to continuously update the estimator.

[0017] In certain embodiments, the machine learning algorithm may include a trained neural network, a simple linear regression model, a random forest regression model, or a support vector machine. More particularly, in embodiments, the method may include embedding reinforcement learning techniques into the machine learning algorithm.

[0018] In certain embodiments, the response time of the estimator may be faster than that of the controller, and the response time of the controller may be faster than generating control commands from a machine learning algorithm.

[0019] On the other hand, the present disclosure relates to a system for controlling a wind farm electric power system. The wind farm electric power system includes a plurality of wind turbines electrically connected to a power grid via an interconnection point. Each wind turbine includes a voltage regulator. The system includes a controller having a plurality of processors, the plurality of processors including at least an estimator. The estimator is configured to perform operations that, for example, include receiving one or more electrical signals associated with the interconnection point in the frequency domain and using the one or more electrical signals to estimate the voltage sensitivity of the power grid. As such, the controller is configured to dynamically control the voltage of the wind farm electric power system at the interconnection point based on the voltage sensitivity to avoid power grid instability by changing at least one parameter of one or more of the voltage regulators. It should be understood that the system may also include any of the additional features described herein.

[0020] Technical Solution 1. A method for controlling a wind farm electric power system, wherein the wind farm electric power system includes a controller, a plurality of wind turbines electrically connected to a power grid via an interconnection point, and each wind turbine includes a voltage regulator, the method comprising:

[0021] receiving, via the controller, one or more electrical signals associated with the interconnection point in the frequency domain;

[0022] estimating a voltage sensitivity of the electrical grid using the one or more electrical signals via an estimator of the controller; and

[0023] The voltage of the wind farm electric power system at the interconnection point is dynamically controlled based on the voltage sensitivity.

[0024] Technical Solution 2. The method according to Technical Solution 1, wherein the one or more electrical signals include at least one of phase voltage, phase current, active power and reactive power at the interconnection point.

[0025] Technical Solution 3. The method according to Technical Solution 1, wherein the method further comprises at least one of the following: measuring the one or more electrical signals via at least one sensor, or determining the one or more electrical signals via a computer-implemented model of the controller.

[0026] Technical Solution 4. The method according to Technical Solution 1, wherein the method further comprises processing the one or more electrical signals associated with the interconnection point before estimating the voltage sensitivity, wherein processing the one or more electrical signals comprises at least one of filtering or classification.

[0027] Technical Solution 5. The method according to Technical Solution 2, wherein using the one or more electrical signals to estimate the voltage sensitivity of the power grid further comprises:

[0028] modeling the power grid as a linear time-invariant system in the frequency domain via the controller;

[0029] monitoring the frequency domain for disturbances;

[0030] determining whether the disturbance corresponds to the wind farm electric power system or to an adjacent wind farm electric power system; and

[0031] A disturbance corresponding to an adjacent wind farm electric power system is removed to isolate the influence of the disturbance on the voltage sensitivity.

[0032] Technical Solution 6. The method according to Technical Solution 5, wherein determining whether the disturbance corresponds to the wind farm electric power system or to an adjacent wind farm electric power system further comprises:

[0033] causing disturbances of at least one of the active power or the reactive power to be grouped by the phase voltages in the frequency domain; and

[0034] Remove ungrouped disturbances.

[0035] Technical Solution 7. The method according to Technical Solution 1, wherein dynamically controlling the voltage of the wind farm electric power system based on the voltage sensitivity further comprises:

[0036] At least one parameter of one or more of the voltage regulators is varied to avoid instability of the grid.

[0037] Technical Solution 8. The method according to Technical Solution 7, wherein the at least one parameter includes at least one of a regulator gain, an active power set point, or a reactive power set point.

[0038] Technical Solution 9. A method according to Technical Solution 7, wherein the method further includes processing the one or more electrical signals associated with the interconnection point via a post-processor of the controller to determine an error analysis of one or more parameters of the voltage regulator after dynamically controlling the voltage of the wind farm electric power system based on the voltage sensitivity.

[0039] Technical Solution 10. The method according to Technical Solution 9, wherein the method further comprises:

[0040] receiving feedback from the postprocessor via a machine learning algorithm of the controller; and

[0041] The feedback is trained via the machine learning algorithm.

[0042] Technical Solution 11. The method according to Technical Solution 10, wherein the method further comprises using the output of the machine learning algorithm to generate one or more control commands for the estimator of the controller to continuously update the estimator.

[0043] Technical Solution 12. The method according to Technical Solution 11, wherein the response time of the estimator is faster than that of the controller, and the response time of the controller is faster than generating the one or more control commands from the machine learning algorithm.

[0044] Technical Solution 13. The method according to Technical Solution 10, wherein the machine learning algorithm includes a trained neural network, a simple linear regression model, a random forest regression model or a support vector machine.

[0045] Technical Solution 14. The method according to Technical Solution 10, wherein the method further includes embedding reinforcement learning technology into the machine learning algorithm.

[0046] Technical Solution 15. A system for controlling a wind farm electric power system, wherein the wind farm electric power system includes a plurality of wind turbines electrically connected to a power grid via an interconnection point, and wherein each wind turbine includes a voltage regulator, the system comprising:

[0047] A controller comprising a plurality of processors, the plurality of processors including at least an estimator, the estimator configured to perform operations comprising:

[0048] receiving one or more electrical signals associated with the interconnection point in the frequency domain; and

[0049] estimating a voltage sensitivity of the electrical grid using the one or more electrical signals; and

[0050] The controller dynamically controls the voltage of the wind farm electric power system at the interconnection point based on the voltage sensitivity by varying at least one parameter of one or more of the voltage regulators to avoid grid instability.

[0051] Technical Solution 16. The system according to Technical Solution 15, wherein the one or more electrical signals include at least one of phase voltage, phase current, active power and reactive power at the interconnection point.

[0052] Technical Solution 17. The system according to Technical Solution 15, wherein the plurality of processors further comprises a pre-processor configured to process the one or more electrical signals associated with the interconnection point before estimating the voltage sensitivity.

[0053] Technical Solution 18. The system according to Technical Solution 16, wherein using the one or more electrical signals to estimate the voltage sensitivity of the power grid further comprises:

[0054] modeling the power grid as a linear time-invariant system in the frequency domain;

[0055] monitoring the frequency domain for disturbances;

[0056] determining whether the disturbance corresponds to the wind farm electric power system or to an adjacent wind farm electric power system; and

[0057] A disturbance corresponding to an adjacent wind farm electric power system is removed to isolate the influence of the disturbance on the voltage sensitivity.

[0058] Technical Solution 19. A system according to Technical Solution 15, wherein the multiple processors further include a post-processor, which is used to process the one or more electrical signals associated with the interconnection point to determine an error analysis of one or more parameters of the voltage regulator after dynamically controlling the voltage of the wind farm electric power system based on the voltage sensitivity.

[0059] Technical Solution 20. A system according to Technical Solution 19, wherein the multiple processors further include a machine learning algorithm, which is configured to receive and train feedback from the post-processor and generate one or more control commands for the estimator to continuously update the estimator.

[0060] Variations and modifications may be made to these exemplary aspects of the present disclosure. These and other features, aspects, and advantages of various embodiments will become better understood with reference to the following description and appended claims. The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the relevant principles. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] A detailed discussion of embodiments for those skilled in the art is set forth in the specification with reference to the accompanying drawings, in which:

[0062] Figure 1 A perspective view showing a portion of a wind turbine according to an exemplary embodiment of the present disclosure;

[0063] Figure 2 Shown suitable for Figure 1 A schematic diagram of a wind turbine electrical power system according to an exemplary embodiment of the present disclosure used by a wind turbine shown in FIG;

[0064] Figure 3 A schematic diagram illustrating a wind farm electric power system according to an exemplary embodiment of the present disclosure is shown;

[0065] Figure 4 A block diagram illustrating a controller according to an exemplary embodiment of the present disclosure;

[0066] Figure 5 A flow chart showing an embodiment of a method for controlling a wind farm electric power system according to the present disclosure;

[0067] Figure 6 A schematic diagram illustrating one embodiment of a system for controlling a wind farm electric power system according to the present disclosure;

[0068] Figure 7 A diagram illustrating one embodiment of a frequency domain with respect to voltage and reactive power according to an exemplary embodiment of the present disclosure;

[0069] Figure 8 A diagram illustrating one embodiment of the frequency domain regarding voltage and active power according to an exemplary embodiment of the present disclosure;

[0070] Figure 9 a schematic control diagram showing one embodiment of an estimator of a system for controlling a wind farm electric power system according to the present disclosure; and

[0071] Figure 10 A schematic control diagram showing another embodiment of an estimator of a system for controlling a wind farm electric power system according to the present disclosure. DETAILED DESCRIPTION

[0072] Reference will now be made in detail to the embodiments of the present disclosure, one or more examples of which are illustrated in the drawings. Each example is provided as an explanation of the present disclosure and is not a limitation of the present disclosure. In fact, it will be apparent to those skilled in the art that various modifications and variations can be made in the present disclosure without departing from the scope or spirit of the present disclosure. For example, a feature shown or described as part of one embodiment can be used with another embodiment to produce still further embodiments. Therefore, it is intended that the present disclosure covers such modifications and variations as fall within the scope of the appended claims and their equivalents.

[0073] Now refer to the figure, Figure 1A perspective view of one embodiment of a wind turbine 10 according to the present disclosure is depicted. As shown, the wind turbine 10 generally includes a tower 12 extending from a supporting surface (not shown), a nacelle 14 mounted on the tower 12, and a rotor 16 coupled to the nacelle 14. The rotor 16 includes a rotatable hub 18 and at least one rotor blade 20 coupled to the hub 18 and extending outwardly from the hub 18. For example, in the illustrated embodiment, the rotor 16 includes three rotor blades 20. However, in alternative embodiments, the rotor 16 may include more or less than three rotor blades 20. Each rotor blade 20 may be spaced about the hub 18 to facilitate rotating the rotor 16, enabling kinetic energy from the wind to be converted into usable mechanical energy and subsequently into electrical energy. For example, the hub 18 may be rotatably coupled to a generator 28 ( Figure 2 ), to allow the generation of electrical energy.

[0074] Now refer to Figure 2 , shows a wind turbine power system 100 including a wind turbine 10 and an associated power system 102. When wind strikes rotor blades 20, the blades 20 convert wind energy into a mechanical rotational torque that rotatably drives a low-speed shaft 22. Low-speed shaft 22 is configured to drive a gearbox 24 (if present), which then increases the low rotational speed of low-speed shaft 22 to drive a high-speed shaft 26 at an increased rotational speed. High-speed shaft 26 is generally rotatably coupled to a generator 28 (such as a doubly-fed induction generator, or DFIG) to rotatably drive a generator rotor 30. As such, a rotating magnetic field can be induced by generator rotor 30, and a voltage can be induced within generator stator 32, which is magnetically coupled to generator rotor 30. The associated electrical power can be transmitted from generator stator 32 to a main three-winding transformer 34, which is connected to the grid at POI 56 via grid breaker 36. Therefore, the main transformer 34 increases the voltage amplitude of the electric power so that the transformed electric power can be further transmitted to the grid.

[0075] In addition, as shown, the generator 28 can be electrically coupled to a bidirectional power converter 38, which includes a rotor-side converter 40 that is connected to a line-side converter 42 via a regulated DC link 44. The rotor-side converter 40 converts the AC power provided from the generator rotor 30 into DC power and provides DC power for the DC link 44. The line-side converter 42 converts the DC power on the DC link 44 into AC output power suitable for the grid. Thus, the AC power from the power converter 38 can be combined with the power from the generator stator 32 to provide multi-phase power (e.g., three-phase power) having a frequency that is substantially maintained at the grid frequency (e.g., 50 Hz / 60 Hz).

[0076] In some configurations, the power system 102 may include a turbine stage controller 224 (in Figure 3 ). The turbine stage controller 224 may be a control such as Figure 4 The controller shown and described in FIG.

[0077] The illustrated three-winding transformer 34 may have (1) a 33 kilovolt (kV) medium voltage (MV) primary winding 33 connected to the grid, (2) a 6 to 13.8 kV MV secondary winding 35 connected to the generator stator 32, and (3) a 690 to 900 volt (V) low voltage (LV) tertiary winding 37 connected to the line-side power converter 42.

[0078] With particular reference Figure 3 , a schematic diagram illustrating one embodiment of a wind farm electric power system 200 according to an exemplary embodiment of the present disclosure is shown. More particularly, as shown, the wind farm electric power system 200 may include a plurality of wind turbine power systems 100 connected to a power grid via a POI 56. The wind farm electric power system 200 may include at least two clusters 204 to form the electric power system 200. The individual wind turbine power systems 100 (including a plurality of wind turbines 10) may be arranged in a predetermined geographic location and electrically connected together to form a wind farm 202.

[0079] The electrical power associated with each wind turbine power system 100 may be transmitted to the main line 206 via one or more cluster lines 220. Each wind turbine power system 100 may be connected to or disconnected from the one or more cluster lines 220 via one or more switches or circuit breakers 222. The wind turbine power systems 100 may be arranged into a plurality of groups (or clusters) 204, each of which is individually connected to the main line 206 via switches 208, 210, 212, respectively. Thus, as shown, each cluster 204 may be connected to a separate transformer 214, 216, 218, respectively, via switches 208, 210, 212, for increasing the voltage amplitude of the electrical power from each cluster 204 so that the transformed electrical power can be further transmitted to the grid. Additionally, as shown, transformers 214, 216, 218 are connected to the main line 206, which combines the voltage from each cluster 204 before transmitting the power to the grid via the POI 56. The POI 56 may be a circuit breaker, a switch, or other known method of connecting to the electrical grid.

[0080] Each wind turbine power system 100 may include a voltage regulator 228 (i.e., a wind turbine terminal voltage regulator). As such, the voltage regulator 228 regulates the voltage output by each wind turbine power system 100. Furthermore, the voltage regulator 228 may be in electrical communication with the turbine controller 224 or the central master controller 226. Thus, the turbine stage controller 224 or the central master controller 226 may control the voltage regulator gain command (VCMD ) is transmitted to one or more of the voltage regulators 228 , which in turn dictate the amount of power to distribute to the POI 56 via the cluster line 220 .

[0081] Each wind turbine power system 100 may include one or more controllers, such as a turbine controller 224. The turbine controller 224 may be configured to control components of the wind turbine power system 100 (including the switch 222 or the voltage regulator 228) and / or implement some or all of the method steps described herein. The turbine controller 224 may be located on or within each wind turbine 10, or may be located remotely from each wind turbine 10. The turbine controller 224 may be part of or included within one or more other controllers associated with the wind turbine power system 100 and / or the wind farm electrical power system 200. Turbine controller 224 may operate switch 222 to connect or disconnect one or more wind turbine power systems 100 from cluster line 220 and control voltage regulator 228, such as voltage regulator gain, based at least in part on the power required at POI 56 and / or at least in part on characteristics of wind turbine power system 100, wind farm electrical power system 200, and / or characteristics of wind turbine 10 (e.g., size, location, age, maintenance status of wind turbine), characteristics of the grid (e.g., strength or condition of the grid, strength or condition of the wind farm or wind turbine connection to the grid, grid architecture, grid location), loads on the grid (e.g., heavy or variable loads), and / or environmental conditions (e.g., wind conditions with respect to one or more wind turbines).

[0082] The wind farm electric power system 200 may include one or more controllers, such as a central master controller 226. The central master controller 226 may be configured to control components of the wind farm electric power system 200 (including switches 208, 210, and 212, voltage regulator 228), communicate with one or more other controllers (such as turbine stage controller 224), and / or implement some or all of the method steps described herein. The central master controller 226 may be located within the geographic area of ​​the wind farm electric power system 200 or any portion thereof, or may be located remotely from the wind farm electric power system 200 or any portion thereof. The central master controller 226 may be part of or included within one or more of the other controllers associated with the wind farm electric power system 200 and / or one or more of the wind turbine power systems 100. Each of the cluster 204, the wind turbine power system 100, or the turbine stage controller 224 may be communicatively coupled to the central master controller 226.

[0083] Central master controller 226 may generate control signals based at least in part on the power required at POI 56 and send the control signals to turbine controller 224 to operate switch 222 to connect or disconnect one or more wind turbine power systems 100 from cluster line 220. Central master controller 226 may generate control signals based at least in part on the power required at POI 56 and send the control signals to voltage regulator 228 to operate or control voltage regulator 228 and control the amount of power delivered from one or more wind turbine power systems 100 to the POI through cluster line 220. Central master controller 226 may generate and send control signals to switches 208, 210, and / or 212 and / or voltage regulator 228 to regulate the power delivered to POI 56 based at least in part on the power required at POI 56 and / or at least in part on characteristics of wind turbine power system 100, wind farm electrical power system 200, and / or characteristics of wind turbine 10 (e.g., size, location, age, maintenance status of wind turbine), characteristics of the grid (e.g., strength or condition of the grid, strength or condition of the wind farm or wind turbine connection to the grid, grid architecture, grid location), loads on the grid (e.g., heavy or variable loads), and / or environmental conditions (e.g., wind conditions with respect to one or more wind turbines).

[0084] Now refer to Figure 4 , a block diagram of a controller 400 according to an exemplary embodiment of the present disclosure is shown. As shown, the controller 400 may be a turbine-level controller 224 or a central master controller 226. In addition, as shown, the controller 400 may include one or more processors 402 and associated memory devices 404 configured to perform various computer-implemented functions (e.g., perform methods, steps, calculations, etc., and store related data as disclosed herein). The memory devices 404 may also store data related to: certain characteristics of the wind turbine power system 100, the wind farm electrical power system 200, and / or characteristics of the wind turbine 10 (e.g., size, location, age, maintenance status of the wind turbine), characteristics of the power grid (e.g., strength or condition of the power grid, strength or condition of the wind farm or wind turbine connection to the power grid, power grid architecture, power grid location), loads on the power grid (e.g., heavy or variable loads), and / or environmental conditions (e.g., wind conditions with respect to one or more wind turbines).

[0085] Additionally, the controller 400 may include a communication module 406 to facilitate communication between the controller and various components of the wind turbine power system 100, the wind farm electric power system 200, and / or the central master controller 226, including communication between the central master controller 226 and the turbine stage controller 224. Furthermore, the communication module 406 may include a sensor interface 408 (e.g., one or more analog-to-digital converters) to allow signals transmitted from one or more sensors 410, 412, and 414 to be converted into signals that can be understood and processed by the processor 402. Sensors 410, 412, and 414 may be used to measure, determine, or collect data regarding: characteristics of wind turbine power system 100, wind farm electrical power system 200, and / or characteristics of wind turbine 10 (e.g., size, location, age, maintenance status of wind turbines), characteristics of the electrical grid (e.g., strength or condition of the grid, strength or condition of the wind farm or wind turbine connection to the grid, grid architecture, grid location), loads on the grid (e.g., heavy or variable loads), and / or environmental conditions (e.g., wind conditions regarding one or more wind turbines).

[0086] Still refer to Figure 4 , the controller 400 may also include a user interface 416. The user interface 416 may have various configurations, and controls may be installed in the user interface 416. The user interface 416 may also be located within the geographic area of ​​the wind farm electric power system 200 or any portion thereof, or may be located remotely from the wind farm electric power system 200 or any portion thereof. The user interface 416 may include an input component 418. For example, the input component 418 may be a capacitive touch screen. The input component 418 may allow selective activation, adjustment, or control of the wind farm controller 226 and the turbine controller 224, as well as any timer features or other user-adjustable inputs. One or more of a variety of electrical, mechanical, or electromechanical input devices (including rotary dials, buttons, and touchpads) may also be used, alone or in combination, as the input component 418. The user interface 416 may include a display component, such as a digital or analog display device designed to provide operational feedback to the user.

[0087] It should be appreciated that sensors 410, 412, and 414 can be communicatively coupled to communication module 406 using any suitable means. For example, sensors 410, 412, and 414 can be coupled to sensor interface 408 via a wired connection. However, in other embodiments, sensors 410, 412, and 414 can be coupled to sensor interface 408 via a wireless connection (such as by using any suitable wireless communication protocol known in the art). As such, processor 402 can be configured to receive one or more signals from sensors 410, 412, and 414. Sensors 410, 412, and 414 can be part of or included in one or more other controllers associated with wind farm electric power system 200 and / or one or more of wind turbine power systems 100. Sensors 410, 412, and 414 can also be located within the geographic area of ​​wind farm electric power system 200 or any portion thereof, or can be located remotely from wind farm electric power system 200 or any portion thereof.

[0088] It should also be understood that sensors 410, 412, and 414 may be any number or type of voltage and / or current sensors that may be employed at any location within wind turbine power system 100. For example, the sensors may be current transformers, shunt sensors, Rogowski coils, Hall effect current sensors, micro inertial measurement units (MIMUs), or the like, and / or any other suitable voltage or current sensors now known or later developed in the art. Accordingly, one or more controllers (such as wind farm controller 226 and turbine controller 224) are configured to receive one or more voltage and / or current feedback signals from sensors 410, 412, and 414.

[0089] As used herein, the term "processor" refers not only to integrated circuits that are considered in the art to be included in computers, but also to controllers, microcontrollers, microcomputers, programmable logic controllers (PLCs), application-specific integrated circuits, and other programmable circuits. The processor 402 is also configured to calculate advanced control algorithms and communicate using a variety of Ethernet or serial-based protocols (Modbus, OPC, CAN, etc.). In addition, the memory device 404 may generally include memory elements, including but not limited to computer-readable media (e.g., random access memory (RAM), computer-readable non-volatile media (e.g., flash memory), floppy disks, compact disc-read only memory (CD-ROM), magneto-optical discs (MODs), digital versatile discs (DVDs), and / or other suitable memory elements). Such memory devices 140 may generally be configured to store suitable computer-readable instructions that, when executed by the processor 402, configure the controller to perform various functions as described herein.

[0090] Now refer to Figure 5 and Figure 6 , respectively showing a method 500 and a system 600 for controlling a wind farm electric power system according to the present disclosure. More particularly, Figure 5 A flow chart showing an embodiment of a method 500 for controlling a wind farm electric power system according to the present disclosure is shown. Figure 6 A schematic diagram illustrating one embodiment of a system 600 for controlling a wind farm electric power system according to the present disclosure is shown.

[0091] In general, this article will refer to Figure 1-4 However, it should be appreciated that the disclosed method 500 and system 600 may be implemented with wind turbines and wind farms having any other suitable configurations.

[0092] In addition, although Figure 5 For purposes of illustration and discussion, the steps are depicted as being performed in a particular order, and the methods discussed herein are not limited to any particular order or arrangement. Those skilled in the art, using the disclosure provided herein, will appreciate that the various steps of the methods disclosed herein may be omitted, rearranged, combined, and / or adapted in various ways without departing from the scope of the present disclosure. Furthermore, it should be understood that the method 500 may be performed by one or more controllers (such as the central master controller 226 and / or the turbine stage controller 224) and by other devices included in the wind turbine power system 100 and / or the wind farm electrical power system 200.

[0093] With particular reference Figure 5 As shown at (502), method 500 includes receiving one or more electrical signals associated with interconnection point 56 in the frequency domain, for example, via main controller 226. For example, in an embodiment, as Figure 6 As shown in FIG, pre-processor 602 of controller 226 may receive electrical signal 603 from interconnection point 56. Moreover, in embodiments, the electrical signal may include any one or a combination of the following: phase voltage, phase current, active power, and / or reactive power at interconnection point 56. Additionally, in certain embodiments, method 500 may include measuring the electrical signal via at least one sensor (such as one of sensors 410, 412, 414). Alternatively, method 500 may include determining or estimating the electrical signal via a computer-implemented model of controller 226.

[0094] As shown at (504), the method 500 includes estimating the voltage sensitivity of the grid (e.g., grid strength) using the electrical signal via the estimator 604 of the controller 226. In further embodiments, the method 500 may include processing the electrical signal associated with the interconnection point 56 before estimating the voltage sensitivity. For example, Figure 6 As shown in FIG. 1 , the pre-processor 602 may be capable of filtering, classifying, or the like, or a combination thereof, the electrical signal.

[0095] In certain embodiments, controller 226 is configured to estimate the voltage sensitivity of the power grid by modeling the power grid as a linear time-invariant system in the frequency domain (e.g., with a short measurement window and small linear signal perturbations). Figure 7 and Figure 8 As shown in FIG, the controller 226 may be configured to generate frequency domains 700, 800 for the electrical signal. More particularly, as shown, Figure 7 shows the frequency domain of voltage and reactive power, and Figure 8 The frequency domain for voltage and active power is shown. Because wind farm electric power system 200 has disturbances in active power due to wind speed variations, and because the frequencies of disturbances in active power and reactive power exist for each wind farm, controller 226 can easily monitor the frequency domain for the disturbances. More specifically, a disturbance in active power or reactive power at a specific frequency is converted into a voltage response at the same frequency.

[0096] Therefore, if Figure 7 As shown in FIG, the controller 226 is configured to identify corresponding response pairs of voltage and reactive power (as indicated via arrows 702, 704). Figure 8 As shown in FIG, the controller 226 is configured to identify corresponding response pairs of voltage and active power (as indicated by arrows 802 and 804). In other words, the controller 226 is configured to group the disturbances of active power and / or reactive power in the frequency domain with phase voltages. Thus, the controller 226 can determine whether the disturbance corresponds to the wind farm electric power system 200 or to an adjacent wind farm electric power system. For example, in an embodiment, the controller 226 can identify disturbances that are not grouped or paired (as those corresponding to disturbances from an adjacent wind farm electric power system). Accordingly, the controller 226 can remove disturbances corresponding to the adjacent wind farm electric power system (such as disturbances 706 and 806) to separate the effects of disturbances 706 and 806 on voltage sensitivity.

[0097] Back reference Figure 6 The system 600 may further include a post-processor 608 for processing electrical signals associated with the interconnection point 56 after estimating the voltage sensitivity of the grid, such as to determine an error analysis of parameters of the voltage regulator 228 .

[0098] Accordingly, if Figure 5 As shown at (506), the method 500 includes dynamically controlling the voltage of the wind farm electric power system 200 at the interconnection point 56 based on the voltage sensitivity. For example, in an embodiment, as Figure 6As shown in FIG, the controller 226 may include a voltage controller 606 configured to dynamically control the voltage of the wind farm electric power system 200 to avoid grid instability by varying at least one parameter of one or more of the voltage regulators 228 of the respective wind turbine power systems 100 of the wind farm 200. In certain embodiments, for example, the parameter may include a regulator gain, an active power set point, a reactive power set point, a combination thereof, and any other suitable parameter.

[0099] Now refer to Figure 9 and Figure 10 , showing the reactive power set point and the active power set point (Q SP and P SP ) changes to avoid grid instability in various embodiments of the control diagram 900, 1000. In addition, as Figure 9 and Figure 10 As shown in FIG, the control diagrams 900, 1000 may include a proportional-integral (PI) controller 906, 1006. As such, the control diagrams 900, 100 may be capable of dynamically adjusting the regulator gain associated with the PI controller 906, 1006. More particularly, as shown, the control diagrams illustrate an implementation of feed-forward voltage control based on the output 902, 1002 of the estimator 604. Thus, as shown at blocks 904 and 1004, respectively, ΔQ and ΔP may be calculated based on the voltage sensitivity (e.g., dQ / dV, dP / dV) from the estimator 604. More particularly, as shown, Figure 9 As shown in block 904, ΔQ may be determined based on at least the gain (α1) that is adjustable for a particular wind farm, the voltage set point (V sp ), actual voltage (V act ) and the voltage sensitivity (dQ / dV) from the estimator 604. Similarly, Figure 10 As shown in block 1004, ΔP may be determined based on at least a gain (α2) that is adjustable for a particular wind farm, a voltage set point (V sp ), actual voltage (V act ) and the voltage sensitivity (dP / dV) from the estimator 604. Furthermore, it should be understood that α1 and α2 may be dynamically changed by a machine learning algorithm in closed-loop operation depending on the post-processor error analysis.

[0100] Therefore, in certain embodiments, the controller 226 is configured to distribute adjusted set points (Q SP and P SP) to achieve the desired voltage response. Like this, the response of the wind farm can be recorded to process the data and provide feedback to the machine learning based algorithm to make further decisions. For example, Figure 6 As shown in FIG, the system 600 may include a machine learning algorithm 610 that receives feedback 612 from the post-processor 608 and trains the feedback 612. In such an embodiment, as shown, the machine learning algorithm 610 is configured to generate one or more control commands 614 for the estimator 604 of the controller 226 to continuously update the estimator 604.

[0101] For example, in an embodiment, the control commands 614 may include gains α1 and α2 for clustering / grouping the frequency domain signals of the measured electrical signals, as well as thresholds set to filter out unwanted frequency domain signals to be grouped. Furthermore, the control commands 614 may include some adjustments to pre-processor parameters (such as the sampling window for analysis, the number of samples considered at a time for estimation, etc.). In other words, the machine learning algorithm 610 is configured to adjust any of the parameters described herein in order to reduce the error calculated by the post-processor (which is the difference between the actual voltage measured and the voltage estimated using the voltage sensitivity). As such, the goal of the machine learning algorithm 610 is to make the system 600 adaptable and robust in weak grid conditions.

[0102] In certain embodiments, the response time of the estimator 604 may be faster than the controller 226, which in turn may be faster than the commands 614 from the machine learning algorithm 610. Furthermore, in certain embodiments, the machine learning algorithm 610 may be a trained neural network, a simple linear regression model, a random forest regression model, a support vector machine, or any suitable type of supervised learning model based on the quality and quantity of the received data. More particularly, in certain embodiments, the method may include embedding reinforcement learning techniques into the machine learning algorithm 610.

[0103] Various aspects and embodiments of the invention are defined by the following numbered clauses:

[0104] Clause 1. A method for controlling a wind farm electric power system, wherein the wind farm electric power system comprises a controller, a plurality of wind turbines electrically connected to an electric grid via an interconnection point, and wherein each wind turbine comprises a voltage regulator, the method comprising:

[0105] receiving, via the controller, one or more electrical signals associated with the interconnection point in the frequency domain;

[0106] estimating, via an estimator of the controller, a voltage sensitivity of the grid using the one or more electrical signals; and

[0107] The voltage of the wind farm electric power system at the interconnection point is dynamically controlled based on the voltage sensitivity.

[0108] Clause 2. The method of clause 1, wherein the one or more electrical signals include at least one of a phase voltage, a phase current, active power, and reactive power at the point of interconnection.

[0109] Clause 3. The method of any of the preceding clauses, further comprising at least one of: measuring the one or more electrical signals via at least one sensor, or determining the one or more electrical signals via a computer-implemented model of the controller.

[0110] Clause 4. The method of any of the preceding clauses, further comprising processing one or more electrical signals associated with the interconnection point prior to estimating the voltage sensitivity, wherein processing the one or more electrical signals comprises at least one of filtering or classifying.

[0111] Clause 5. The method of any preceding clause, wherein estimating the voltage sensitivity of the power grid using the one or more electrical signals further comprises:

[0112] The power grid is modeled as a linear time-invariant system in the frequency domain via a controller;

[0113] Monitoring the frequency domain regarding disturbances;

[0114] determining whether the disturbance corresponds to the wind farm electric power system or to an adjacent wind farm electric power system; and

[0115] The disturbance corresponding to the adjacent wind farm electric power system is removed to isolate the impact of the disturbance on voltage sensitivity.

[0116] Clause 6. The method of any preceding clause, wherein determining whether the disturbance corresponds to the wind farm electric power system or to a neighboring wind farm electric power system further comprises:

[0117] grouping disturbances of at least one of active power or reactive power by phase voltage in the frequency domain; and

[0118] Remove ungrouped disturbances.

[0119] Clause 7. The method of any preceding clause, wherein dynamically controlling the voltage of the wind farm electric power system based on voltage sensitivity further comprises:

[0120] At least one parameter of one or more of the voltage regulators is varied to avoid grid instability.

[0121] Clause 8. The method of any of the preceding clauses, wherein the at least one parameter comprises at least one of a regulator gain, an active power set point, or a reactive power set point.

[0122] Clause 9. The method of any preceding clause, further comprising processing via a post-processor of the controller one or more electrical signals associated with the interconnection point to determine an error analysis of one or more parameters of the voltage regulator after dynamically controlling the voltage of the wind farm electrical power system based on the voltage sensitivity.

[0123] Clause 10. The method of any of the preceding clauses, further comprising:

[0124] receiving feedback from a post-processor via a machine learning algorithm of the controller; and

[0125] Feedback is trained via machine learning algorithms.

[0126] Clause 11. The method of any of the preceding clauses, further comprising using an output of the machine learning algorithm to generate one or more control commands for an estimator of the controller to continuously update the estimator.

[0127] Clause 12. The method of any of the preceding clauses, wherein the response time of the estimator is faster than that of the controller, and the response time of the controller is faster than generating the one or more control commands from the machine learning algorithm.

[0128] Clause 13. The method of any of the preceding clauses, wherein the machine learning algorithm comprises a trained neural network, a simple linear regression model, a random forest regression model, or a support vector machine.

[0129] Clause 14. The method of any of the preceding clauses, further comprising embedding reinforcement learning techniques into the machine learning algorithm.

[0130] Clause 15. A system for controlling a wind farm electric power system, wherein the wind farm electric power system comprises a plurality of wind turbines electrically connected to an electric grid via an interconnection point, and wherein each wind turbine comprises a voltage regulator, the system comprising:

[0131] A controller comprising a plurality of processors including at least an estimator configured to perform operations comprising:

[0132] receiving one or more electrical signals associated with the interconnection point in the frequency domain; and

[0133] estimating a voltage sensitivity of the electrical grid using the one or more electrical signals; and

[0134] The controller dynamically controls the voltage of the wind farm electric power system at the interconnection point based on the voltage sensitivity by changing at least one parameter of one or more of the voltage regulators to avoid grid instability.

[0135] Clause 16. The system of clause 15, wherein the one or more electrical signals include at least one of a phase voltage, a phase current, active power, and reactive power at the point of interconnection.

[0136] Clause 17. The system of clauses 15-16, wherein the plurality of processors further comprises a pre-processor for processing one or more electrical signals associated with the interconnect point prior to estimating the voltage sensitivity.

[0137] Clause 18. The system of clauses 15-17, wherein estimating the voltage sensitivity of the power grid using the one or more electrical signals further comprises:

[0138] Model the power grid as a linear time-invariant system in the frequency domain;

[0139] Monitoring the frequency domain regarding disturbances;

[0140] determining whether the disturbance corresponds to the wind farm electric power system or to an adjacent wind farm electric power system; and

[0141] The disturbance corresponding to the adjacent wind farm electric power system is removed to isolate the impact of the disturbance on voltage sensitivity.

[0142] Clause 19. The system of clauses 15-18, wherein the plurality of processors further comprises a post-processor for processing one or more electrical signals associated with the interconnection point to determine an error analysis of one or more parameters of the voltage regulator after dynamically controlling the voltage of the wind farm electrical power system based on the voltage sensitivity.

[0143] Clause 20. The system of clauses 15-19, wherein the plurality of processors further comprises a machine learning algorithm configured to receive and train feedback from the post-processor and generate one or more control commands for the estimator to continuously update the estimator.

[0144] This written description uses examples to disclose the invention, including the best mode, and also to enable any person skilled in the art to practice the invention, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the invention is defined by the claims and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they include structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal language of the claims.

Claims

1. A method for controlling a wind farm electric power system, wherein the wind farm electric power system comprises a controller, a plurality of wind turbines electrically connected to a power grid via an interconnection point, and wherein each wind turbine comprises a voltage regulator, the method comprising: receiving, via the controller, one or more electrical signals associated with the interconnection point in the frequency domain; estimating a voltage sensitivity of the electrical grid using the one or more electrical signals via an estimator of the controller; and dynamically controlling the voltage of the wind farm electric power system at the interconnection point based on the voltage sensitivity; wherein the one or more electrical signals include at least one of a phase voltage, a phase current, active power, and reactive power at the interconnection point; Wherein estimating the voltage sensitivity of the power grid using the one or more electrical signals further comprises: modeling the power grid as a linear time-invariant system in the frequency domain via the controller; monitoring the frequency domain for disturbances; determining whether the disturbance corresponds to the wind farm electric power system or to an adjacent wind farm electric power system; and A disturbance corresponding to an adjacent wind farm electric power system is removed to isolate the influence of the disturbance on the voltage sensitivity.

2. The method according to claim 1, characterized in that The method further includes at least one of measuring the one or more electrical signals via at least one sensor or determining the one or more electrical signals via a computer-implemented model of the controller.

3. The method according to claim 1, characterized in that The method further includes processing the one or more electrical signals associated with the interconnection point prior to estimating the voltage sensitivity, wherein processing the one or more electrical signals includes at least one of filtering or classifying.

4. The method according to claim 1, wherein Determining whether the disturbance corresponds to the wind farm electric power system or to an adjacent wind farm electric power system further includes: causing disturbances of at least one of the active power or the reactive power to be grouped by the phase voltages in the frequency domain; and Remove ungrouped disturbances.

5. The method according to claim 1, wherein Dynamically controlling the voltage of the wind farm electric power system based on the voltage sensitivity further includes: At least one parameter of one or more of the voltage regulators is varied to avoid instability of the grid.

6. The method according to claim 5, characterized in that The at least one parameter includes at least one of a regulator gain, an active power set point, or a reactive power set point.

7. The method according to claim 5, characterized in that The method also includes processing, via a post-processor of the controller, the one or more electrical signals associated with the interconnection point to determine an error analysis of one or more parameters of the voltage regulator after dynamically controlling the voltage of the wind farm electric power system based on the voltage sensitivity.

8. The method according to claim 7, characterized in that The method further comprises: receiving feedback from the postprocessor via a machine learning algorithm of the controller; and The feedback is trained via the machine learning algorithm.

9. The method according to claim 8, characterized in that The method also includes using the output of the machine learning algorithm to generate one or more control commands for an estimator of the controller to continuously update the estimator.

10. The method according to claim 9, characterized in that The estimator has a faster response time than the controller, and the controller has a faster response time than generating the one or more control commands from the machine learning algorithm.

11. The method according to claim 8, characterized in that The machine learning algorithm includes a trained neural network, a simple linear regression model, a random forest regression model, or a support vector machine.

12. The method according to claim 8, characterized in that The method also includes embedding reinforcement learning techniques into the machine learning algorithm.

13. A system for controlling a wind farm electric power system, wherein the wind farm electric power system comprises a plurality of wind turbines electrically connected to a power grid via an interconnection point, and wherein each wind turbine comprises a voltage regulator, the system comprising: A controller comprising a plurality of processors, the plurality of processors including at least an estimator, the estimator configured to perform operations comprising: receiving one or more electrical signals associated with the interconnection point in the frequency domain; and estimating a voltage sensitivity of the electrical grid using the one or more electrical signals; and wherein the controller dynamically controls the voltage of the wind farm electric power system at the interconnection point based on the voltage sensitivity by varying at least one parameter of one or more of the voltage regulators to avoid instability of the grid; wherein the one or more electrical signals include at least one of a phase voltage, a phase current, active power, and reactive power at the interconnection point; Wherein estimating the voltage sensitivity of the power grid using the one or more electrical signals further comprises: modeling the power grid as a linear time-invariant system in the frequency domain; monitoring the frequency domain for disturbances; determining whether the disturbance corresponds to the wind farm electric power system or to an adjacent wind farm electric power system; and A disturbance corresponding to an adjacent wind farm electric power system is removed to isolate the influence of the disturbance on the voltage sensitivity.

14. The system according to claim 13, wherein: The plurality of processors further includes a pre-processor for processing the one or more electrical signals associated with the interconnection point prior to estimating the voltage sensitivity.

15. The system according to claim 13, wherein: The plurality of processors further includes a post-processor for processing the one or more electrical signals associated with the interconnection point to determine an error analysis of one or more parameters of the voltage regulator after dynamically controlling the voltage of the wind farm electric power system based on the voltage sensitivity.

16. The system according to claim 15, wherein: The plurality of processors further includes a machine learning algorithm configured to receive and train feedback from the post-processor and generate one or more control commands for the estimator to continuously update the estimator.

Images