Method for operating a wind turbine

Abstract

The invention relates to a method for operating a wind turbine, the method comprising determining a correction model associated with the wind turbine, determining a corrected turbulence intensity parameter associated with the wind turbine based on the correction model, and operating the wind turbine based on the corrected turbulence intensity parameter.

Description

Technical Field

[0001] The present subject matter generally relates to a method for operating a wind turbine, a method for siting a wind turbine, a wind turbine, a system for operating a wind turbine, and a wind park including at least one wind turbine and a wind park controller. Background Art

[0002] Wind power is considered one of the cleanest and most environmentally friendly energy sources currently available, and wind turbines have attracted increasing attention in this regard. A modern wind turbine typically consists of a tower, a generator, a gearbox, a nacelle, and one or more rotor blades. The rotor blades capture kinetic energy from the wind using known aerodynamic principles and transfer this energy through rotational energy to rotate a shaft, which connects the rotor blades to the gearbox or, if a gearbox is not used, directly to the generator. The generator then converts the mechanical energy into electrical energy that can be distributed to the utility grid.

[0003] There is an objective to improve wind turbine operation and wind turbine siting. Turbulence intensity is an important wind characterization parameter for wind turbine operation and wind turbine siting. Turbulence intensity estimated using wind speed data (e.g., rotor effective wind speed) can be used as a basis for wind turbine control. Turbulence intensity estimated based on reference wind speed information (e.g., meteorological mast wind sensor data) can be used as a basis for wind turbine siting. Therefore, there is a challenge to improve turbulence intensity estimation for efficient wind turbine operation and optimal wind turbine siting. Summary of the Invention

[0004] 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 the invention.

[0005] In one aspect, the present disclosure is directed to a method of operating a wind turbine, the method comprising determining a correction model associated with the wind turbine, determining a corrected turbulence intensity parameter associated with the wind turbine based on the correction model, and operating the wind turbine based on the corrected turbulence intensity parameter.

[0006] In one aspect, the present disclosure is directed to a method of siting a wind turbine at a new site, the method comprising determining a site-corrected model, determining an adjusted turbulence intensity parameter associated with the wind turbine at the new site, and siting the wind turbine at the new site based on the adjusted turbulence intensity parameter.

[0007] In one aspect, the present disclosure is directed to a method of siting a new wind turbine, the method comprising determining a site-corrected model, determining an adjusted turbulence intensity parameter associated with the new wind turbine based on the site-corrected model, and siting the new wind turbine based on the adjusted turbulence intensity parameter.

[0008] In one aspect, the present disclosure is directed to an apparatus comprising: a wind turbine including at least two rotor blades; and a model-based controller configured to determine a rotor average wind speed, wherein the wind turbine is operated according to one aspect or is sited according to aspects described herein.

[0009] In one aspect, the present disclosure is directed to a wind farm comprising at least one wind turbine and a wind farm controller comprising a processor configured to perform a method according to aspects described herein.

[0010] In one aspect, the present disclosure is directed to a system for operating a wind turbine, the system comprising a processor configured to perform a method according to aspects described herein.

[0011] These and other features, aspects and advantages of the present invention will be further supported and described 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 invention and, together with the description, serve to explain the principles of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] A complete and open disclosure of the present invention including the best mode thereof for those skilled in the art is set forth in the specification with reference to the accompanying drawings, in which:

[0013] Figure 1 A perspective view of a wind turbine is shown;

[0014] Figure 2 A simplified interior view of a nacelle of a wind turbine is shown;

[0015] Figure 3A A method of operating a wind turbine according to the present disclosure is shown;

[0016] Figure 3B A method of siting a wind turbine according to the present disclosure is shown;

[0017] Figure 4A A method of calibrating a correction model according to the present disclosure is shown;

[0018] Figure 4B A method of calibrating a location correction model according to the present disclosure is shown;

[0019] 5A to 5D A method of calibrating a correction model according to the present disclosure is shown; and

[0020] Figure 6 An example according to the present disclosure is shown. DETAILED DESCRIPTION

[0021] Reference will now be made in detail to embodiments of the present invention, one or more examples of which are illustrated in the accompanying drawings. Each example is provided by way of illustration of the present invention without limiting the present invention. Indeed, it will be apparent to those skilled in the art that various modifications and variations may be made in the present invention without departing from the scope or spirit of the present invention. For example, a feature shown or described as part of one embodiment may be used in conjunction with another embodiment to produce yet another embodiment. Therefore, it is intended that the present invention covers such modifications and variations within the scope of the appended claims and their equivalents.

[0022] Now referring to the accompanying drawings, Figure 1 A perspective view of one embodiment of a wind turbine 10 according to the present disclosure is shown. As shown, wind turbine 10 generally includes a tower 12 extending from a support surface 14 , a nacelle 16 mounted on tower 12 , and a rotor 18 coupled to nacelle 16 .

[0023] like Figure 1 As shown, the rotor 18 includes a rotatable hub 20 and at least one rotor blade 22 coupled to the hub 20 and extending outwardly from the hub 20. For example, in the illustrated embodiment, the rotor 18 includes three rotor blades 22. However, in alternative embodiments, the rotor 18 may include more or fewer than three rotor blades 22. Each rotor blade 22 may be spaced about the hub 20 to facilitate rotation of the rotor 18 so that kinetic energy can be converted from the wind into usable mechanical energy and subsequently into electrical energy. For example, the hub 20 may be rotatably coupled to a generator 24 ( Figure 2 ), to allow the generation of electrical energy.

[0024] Wind turbine 10 may also include a wind turbine controller 26 centralized within nacelle 16. However, in other embodiments, controller 26 may be located within any other component of wind turbine 10 or at a location external to wind turbine 10. Furthermore, controller 26 may be communicatively coupled to any number of components of wind turbine 10 to control the components. Thus, controller 26 may include a computer or other suitable processing unit. Thus, in several embodiments, controller 26 may include suitable computer-readable instructions that, when implemented, configure controller 26 to perform various functions, such as receiving, transmitting, and / or executing wind turbine control signals.

[0025] See now Figure 2 , showing Figure 1, a simplified internal view of an embodiment of a nacelle 16 of a wind turbine 10 is shown, particularly illustrating its drivetrain components. More specifically, as shown, a generator 24 can be coupled to the rotor 18 to generate electrical energy from the rotational energy generated by the rotor 18. The rotor 18 can be coupled to a main shaft 34, which can rotate via main bearings (not shown). The main shaft 34 can, in turn, be rotatably coupled to a gearbox output shaft 36 of the generator 24 via a gearbox 30. The gearbox 30 can include a gearbox housing 38 connected to a base plate 46 via one or more torque arms 48. More specifically, in certain embodiments, the base plate 46 can be a forged component in which the main bearings (not shown) are located and through which the main shaft 34 extends. As generally understood, in response to the rotation of the rotor blades 22 and the hub 20, the main shaft 34 provides a low-speed, high-torque input to the gearbox 30. The gearbox 30 therefore converts a low speed, high torque input to a high speed, low torque output to drive the gearbox output shaft 36 and, therefore, the generator 24 .

[0026] Each rotor blade 22 may also include a pitch adjustment mechanism 32 configured to rotate each rotor blade 22 about its pitch axis 28 via a pitch bearing 40. Similarly, wind turbine 10 may include one or more yaw drive mechanisms 42 communicatively coupled to controller 26, wherein each yaw drive mechanism(s) 42 is configured to change the angle of nacelle 16 relative to the wind (e.g., by engaging a yaw bearing 44 of wind turbine 10).

[0027] Throughout the embodiments described in this disclosure, as referenced Figure 3A , 4, 5A, 5B, 5C, 5D are exemplary embodiments that can be combined with other embodiments described herein, and the rotor effective wind speed data and / or the wind speed data associated with the wind turbine can be replaced by the wind speed data associated with other wind turbines and / or combined with the wind speed data associated with other wind turbines.

[0028] Wind speed data associated with a wind turbine can be derived from physical and / or virtual sensors. Examples of physical sensors include an anemometer mounted on the nacelle. Physical sensors can be cup anemometers, propellers, ultrasonic, sonic, SODAR, and LIDAR-type sensors. Virtual sensors can be derived from model-based controllers to provide rotor effective wind speed, blade estimated wind speed, and more.

[0029] Other wind speed data associated with the wind turbine includes data derived therefrom, such as subsampled data, turbulence intensity parameters, length scale parameters, and the like.

[0030] Other wind speed data associated with the wind turbine also includes constitutive data such as rotor speed, blade pitch angle, wind turbine power coefficient, generator power, wind turbine power output, generator speed, generator torque, etc.

[0031] In an exemplary embodiment, the wind speed associated with the wind turbine is a rotor effective wind speed based on at least turbine power output, rotor speed, and blade pitch.

[0032] Throughout the embodiments described in this disclosure, as referenced Figure 3A 、 3B In the exemplary embodiments described in 5A, 5B, 5C, and 5D, which can be combined with other embodiments described herein, the first plurality of wind data may refer to rotor effective wind speed data, wind speed data associated with a wind turbine, or wind speed data associated with (one or more) existing sites. For the first plurality of wind data that refers to rotor effective wind speed data or wind speed data associated with a wind turbine, the first plurality of wind data may refer to the same wind turbine as or a different wind turbine associated with the third plurality of wind data.

[0033] In an exemplary embodiment, the first plurality of wind data is compared to an exemplary reference Figure 4A The same wind speed data is described in connection with the wind turbine.

[0034] In an exemplary embodiment, the second plurality of wind data is compared to an exemplary reference Figure 4A The reference wind speed information described is the same.

[0035] In an exemplary embodiment, the third plurality of wind data is compared with an exemplary reference Figure 3A The rotor effective wind speed data described are the same.

[0036] In embodiments where the wind turbines associated with the first and third pluralities of wind data are the same, the calibration model performs better. In embodiments where the wind turbines associated with the first and third pluralities of wind data are different, the calibration model is more robust.

[0037] Throughout the embodiments described in this disclosure, as referenced Figure 3A For the exemplary embodiment explained, the operating parameters of the wind turbine may include blade pitch angle, generator speed, converter speed, generator power, rotor speed, required torque, generator torque, aerodynamic torque, wind speed, and the like.

[0038] Throughout the embodiments described in this disclosure, as referenced Figure 3A 、 3B , 4, 5A, 5B, 5C and 5D exemplarily explained embodiments, the second plurality of wind data may refer to reference wind speed data, models or information.

[0039] Figure 3A A method for operating a wind turbine according to the present disclosure is shown. In a typical embodiment, a calibration model can be used to optimize an existing wind turbine. Rotor effective wind speed data 322 can be received. For example, rotor speed can be measured using one or more sensors. Power output can be measured using one or more sensors. Calculations can be performed by a controller, such as a model-based controller, that correlates the measured rotor speed, measured power output, and / or other operating parameters of the wind turbine to determine the rotor effective wind speed. The controller, such as a model-based controller, can store data on some data storage medium at the wind turbine or at a central controller, such as a wind farm controller.

[0040] Other operating parameters of the wind turbine may be, for example, blade pitch angle. A data connection (WiFi, cable, etc.) may be provided for data transmission. The rotor effective wind speed data and / or data used to determine the rotor effective wind speed data may be transmitted to a controller, such as a wind farm controller. The controller may include a processor and some volatile or non-volatile data storage medium.

[0041] See for example Figure 3A , a corrected turbulence intensity parameter 324 may be determined based on the received wind speed data and the correction model. The correction model may be, for example, a transfer function. The correction model may be stored on a data storage medium, which may be, for example, removable or non-removable, and may be at a central wind farm controller or some other data storage location. A data connection (WiFi, cable, etc.) may be present to transmit data, such as the correction model, to the controller.

[0042] In an example, a controller performs a calculation based on received rotor effective wind speed data to determine a corrected turbulence intensity parameter. The controller may include a processor and some volatile or non-volatile data storage medium. The corrected turbulence intensity parameter may be stored on the data storage medium, for example, at a wind turbine (e.g., a wind turbine from which the rotor effective wind speed data is received), at some central controller (e.g., a wind farm controller), or at some other data storage location.

[0043] The calibration model provides a relationship between spatially averaged wind data (e.g., turbulence intensity based on rotor effective wind speed data) and point-equivalent wind data (e.g., turbulence intensity based on meteorological wind speed data). Thus, a calibrated turbulence intensity parameter can be determined from the received rotor effective wind speed data.

[0044] The controller may calculate the corrected turbulence intensity parameter online or recursively (eg at regular time intervals, such as every 10 minutes) based on the latest received rotor effective wind speed data (particularly the rotor effective wind speed data of the last 10 minutes).

[0045] Additionally or alternatively, calculation of the corrected turbulence intensity parameter may be performed by the controller in an offline manner (e.g., once or irregularly), for example, by calculating based on historical rotor effective wind speed data, such as the previous day, week, month, year, or any time period that is not considered to be recent.

[0046] Therefore, online correction has the advantages of optimizing power generation and load control, such as more precise control strategies.

[0047] An advantage of offline correction is that life estimates, such as damage equivalent load estimates or remaining turbine life, are improved or corrected, thereby adjusting load management and power management based on the corrected life estimate information.

[0048] See for example Figure 3A , operating parameters of the wind turbine, such as blade pitch angle or generator torque, may be determined based on the corrected turbulence intensity parameters 326. For example, the corrected turbulence intensity parameters may be stored on a data storage medium. A data connection (an internal data connection (e.g., within the machine or on a processor) or an external connection (e.g., a cable, Wi-Fi, etc.)) may be present to transmit the data to a controller. A controller, such as a model-based controller, may retrieve the stored corrected turbulence intensity parameters via a data connection, such as Wi-Fi, a cable, etc. The controller may perform calculations to determine operating parameters based on the corrected turbulence intensity. The calculated operating parameters may be stored on the data storage medium.

[0049] The controller may transmit appropriate wind turbine control signals corresponding to the determined operating parameters via data connections to control components, such as pitch adjustment mechanisms, generator-side converters, nacelle yaw drives, generator contactors, etc. Thus, operating parameters may be determined or the wind turbine may be operated based on the corrected turbulence intensity parameters.

[0050] The corrected turbulence intensity parameter more accurately reflects the wind conditions experienced by the wind turbine. Thus, by determining operating parameters of the wind turbine based on the corrected turbulence intensity parameter, improved control of the wind turbine is achieved. For example, improved control of power output, such as higher power generation, is achieved. In another example, better control of structural loads (e.g., fatigue loads or extreme loads) on the wind turbine due to turbulence or changing wind speeds is achieved.

[0051] Improved estimates of turbulence intensity, such as the corrected turbulence intensity parameter, offer the advantage of maximizing the operational life of a wind turbine, for example by better maintaining the turbine operating within its designed load limits. Improved estimates of turbulence intensity also have the advantage of maximizing power generation, for example by strategically exploiting specific turbulence conditions.

[0052] In this way, the operating parameters of the wind turbine may be improved, such as the operation of existing wind turbines. Thus, efficient wind turbine operation is achieved.

[0053] The correction based on the correction model can be understood as correcting the spatially averaged measured value of the turbulence intensity, such as the turbulence intensity based on the rotor effective wind speed, to obtain a point equivalent estimate of the turbulence intensity, for example, as shown in the exemplary embodiment with reference to Figure 3A As stated.

[0054] Therefore, the corrected turbulence intensity parameter may be better than a turbulence intensity parameter calculated based on rotor effective (or associated with an existing wind turbine) wind speed data and without using a correction model.

[0055] For example, using the described correction model to correct a turbulence intensity parameter (which is calculated using a controller based on wind speed data associated with a wind turbine (such as received rotor effective wind speed data)) can correct for effects such as vertical extrapolation, for example, as caused by a difference between the altitude at which the reference wind speed information is measured and the altitude at which the wind speed data associated with the wind turbine (such as the rotor effective wind speed data) is determined, and / or correct for effects such as horizontal extrapolation.

[0056] The correction may correct for effects such as seasonal and long-term trends, changes, and / or variations. The correction may correct for effects such as operating wind turbine wake effects. Thus, the correction model may be used to correct wind speed data from wind speed data associated with a wind turbine, such as rotor effective wind speed data.

[0057] Figure 3B A method for siting a wind vortex according to the present disclosure is shown. A method for developing a wind farm is provided, which may include a method for siting a wind turbine. In an embodiment, a site correction model is used to optimize (one or more) new wind turbines and / or (one or more) new wind farms. Wind speed data 342 associated with a location such as a new location or a new wind farm may be received. For example, wind speed is measured using one or more sensors and stored on some volatile or non-volatile data storage medium. A mast-mounted anemometer, such as a sonic anemometer mounted on a meteorological mast, may be such a sensor. The sensor may be located at the new location or new wind farm. A data connection (WiFi or cable, etc.) may exist to transfer data from the one or more sensors to the data storage medium. The data storage medium may be located at the new location or new wind farm, or at a remote data center.

[0058] See for example Figure 3B, the adjusted turbulence intensity parameter may be determined based on the received wind speed data associated with the new location and the location correction model 344. The location correction model may be, for example, a transfer function. In an exemplary embodiment, the exemplary reference here is Figure 3B Described location correction model and exemplary reference Figure 4B The location correction model described is the same.

[0059] The location correction model can be stored on a data storage medium, which can be, for example, removable or non-removable, and can be located at a central wind farm controller or some other data storage location. A data connection (WiFi, cable, etc.) can be provided to transmit data such as the location correction model and / or wind speed data associated with the new location to the controller. In one example, the controller performs a calculation based on the received wind speed data associated with the new location to determine an adjusted turbulence intensity parameter. The controller can include a processor and some volatile or non-volatile data storage medium. The adjusted turbulence intensity parameter can be stored on the data storage medium, for example, at a data storage center.

[0060] Therefore, the adjusted turbulence intensity parameter may be better than a turbulence intensity parameter calculated based on the new site-associated wind speed data and without using the site-corrected model.

[0061] For example, adjustments to turbulence intensity parameters calculated using a controller based on wind speed data associated with a new location using the described location correction model may correct for effects such as vertical extrapolation (e.g., as caused by a difference between the altitude at which the wind speed data associated with the new location was measured and the altitude at which the new wind turbine operates) and / or horizontal extrapolation.

[0062] Adjustments can correct for effects such as seasonal and long-term trends, changes, and / or variations. Adjustments can correct for effects such as the effects of wakes from operating wind turbines. Thus, a site-corrected model can be used to correct wind speed data from future wind farms or new locations.

[0063] In this way, siting parameters of new wind turbines, for example at new locations, can be improved. Thus, the siting of new wind turbines is optimized. Adjustment based on the site correction model can be understood as adjusting point measurements of turbulence intensity, for example from anemometers mounted on meteorological masts, to determine an operationally relevant parameterization of turbulence intensity, for example as determined by the exemplary reference Figure 3B The embodiments described are as follows.

[0064] Thus, based on the corrected model, turbulence intensity parameters determined from wind speed data associated with the new location (e.g., wind speed data based on anemometers mounted on meteorological masts) may be adjusted to determine equivalent spatially averaged turbulence intensity parameters, such as turbulence intensity parameters associated with the new wind turbine.

[0065] In some embodiments, the adjusted turbulence intensity parameter may be understood to be the same as the corrected turbulence intensity parameter. Such an adjusted turbulence intensity parameter may be applicable to situations where one or more wind turbines already exist and the new wind turbine or new wind farm may be a wind turbine installed at the same location as or adjacent to the existing wind turbine(s).

[0066] See for example Figure 3B , a siting parameter for a new wind turbine can be determined based on the adjusted turbulence intensity parameter 346. For example, the adjusted turbulence intensity parameter can be stored on a data storage medium. A data connection (internal data connection (e.g., within the machine or on the processor) or external connection (e.g., cable, Wi-Fi, etc.)) can be present to transmit the data to the controller.

[0067] A simulator, such as an aeroelastic simulator, can retrieve the stored corrected turbulence intensity parameters via a data connection, such as Wi-Fi or a cable. The simulator can then perform calculations to determine siting parameters based on the adjusted turbulence intensity. The calculated siting parameters can be stored on a data storage medium. A new wind turbine can be installed based on the calculated siting parameters. Thus, siting parameters can be determined or a new wind turbine can be sited based on the adjusted turbulence intensity parameters.

[0068] The adjusted turbulence intensity parameter can be used to determine the optimal characteristics of a new wind turbine. Examples of siting parameters include power rating, location, hub height, structural load-bearing characteristics, etc. The advantage is that the new wind turbine is optimized for the local wind conditions. This improves turbulence intensity estimation and enables optimal wind turbine siting.

[0069] Exemplary reference site calibration model and Figure 3B The described embodiments may be combined with or independent of the exemplary references Figure 3A and 4A For example, a controller may apply a site correction model to determine an adjusted turbulence intensity parameter to determine siting parameters for a new wind turbine, without or independently of a controller receiving rotor effective wind speed data, a controller determining operating parameters of the wind turbine, or a controller calibrating a correction model.

[0070] Figure 4A A method of calibrating a calibration model according to the present disclosure is shown. In a typical embodiment, the calibration model may be calibrated as follows. Wind speed data 422 associated with a wind turbine may be received. In an embodiment, as described herein with reference to Figure 4A The received wind speed data associated with the wind turbine described in the exemplary illustrated embodiment may be compared with the wind speed data in the example illustrated in FIG. Figure 3AThe received rotor effective wind speeds described in the exemplary explained embodiments are the same.

[0071] More than one source of wind speed data associated with the wind turbine may be received, such as rotor effective wind speed, wind speed data from a wind sensor mounted on the nacelle, etc. Increasing the amount of data used for calibration, such as calibrating the calibration model using more than one source of wind speed data associated with the wind turbine, may improve performance of the calibration model.

[0072] For example, more data for calibration provides a wider range of parameters such as turbulence intensity and length scale on which to calibrate the reference wind speed information and wind speed data associated with the wind turbine.

[0073] Exemplary References Figure 4A In one embodiment, rotor effective wind speed data 422 is received. For example, rotor speed may be measured using one or more sensors. Power output may be measured using one or more sensors. Calculations may be performed by a controller, such as a model-based controller, that correlates the measured rotor speed, measured power output, and / or other operating parameters of the wind turbine to determine the rotor effective wind speed. This controller, such as a model-based controller, may store data on some data storage medium at the wind turbine or at a central controller, such as a wind farm controller.

[0074] Other operating parameters of the wind turbine used to determine the effective wind speed data for the rotor may include blade pitch angle. A data connection (WiFi, cable, etc.) may be provided to transmit data, such as received wind speed data associated with the wind turbine. The wind speed data associated with the wind turbine and / or data used to determine the wind speed data associated with the wind turbine may be sent to a controller. The controller may include a processor and some volatile or non-volatile data storage medium to process and store the wind speed data associated with the wind turbine.

[0075] Exemplary References Figure 4A , a wind characteristic quantity associated with the wind turbine may be determined based on the received wind speed data associated with the wind turbine 424. For example, a data connection (WiFi, cable, etc.) may be provided to transmit data such as wind speed data associated with the wind turbine from a data storage medium to the controller.

[0076] The controller may perform calculations based on the received wind speed data associated with the wind turbine to determine wind characteristics associated with the wind turbine. The wind characteristics associated with the wind turbine may be parameters such as turbulence intensity, length scale, vertical shear, wind spectrum, etc. The controller may store the wind characteristics associated with the wind turbine on a storage medium.

[0077] Exemplary References Figure 4AIn one embodiment, the wind spectrum, turbulence intensity, and length-scale wind characteristics are calculated by the controller based on the received rotor effective wind speed data associated with the wind turbine and stored by the controller on a volatile or non-volatile data storage medium.

[0078] Exemplary References Figure 4A , reference wind speed information 442 may be received. The reference wind speed information may be received via a data connection such as wifi, cable, etc., and stored on a volatile or non-volatile data storage medium. In an exemplary embodiment, the reference wind speed information is obtained by measuring from one or more wind speed sensors (e.g., a 2D sonic anemometer) mounted on a meteorological mast.

[0079] In an exemplary embodiment, the wind speed sensor used for the reference wind speed information may be located at the same wind farm as the wind turbine that is located at the reference wind farm. Figure 4A The exemplary embodiments described herein are associated with wind speed data associated with the wind turbines.

[0080] In an exemplary embodiment, the reference wind speed information is obtained from a set of aeroelastic simulations.The reference wind speed information may illustratively extend over the same time period(s) as the wind speed data associated with the wind turbine.

[0081] Exemplary References Figure 4A , a reference wind characteristic quantity 444 may be determined based on the received reference wind speed information. For example, a data connection (WiFi, cable, etc.) may be provided to enable the controller to retrieve the reference wind speed information. The controller may use a processor to calculate at least one reference wind characteristic quantity based on the received reference wind speed information, and store the at least one calculated reference wind characteristic quantity on a volatile or non-volatile data storage medium. In an exemplary embodiment, the controller calculates the same type of wind characteristic quantity for the reference wind speed data 444 as for the wind speed data 424 associated with the wind turbine. In an exemplary embodiment, the controller uses the reference wind speed information to calculate wind spectrum, turbulence intensity, and length scale characteristics.

[0082] Exemplary References Figure 4A Based on the wind characteristic quantities associated with the wind turbine and the reference wind characteristic quantities, the calibration model may be calibrated 446. For example, the controller may use a processor to calculate the calibration model based on the wind characteristic quantities associated with the wind turbine and the reference wind characteristic quantities, and store the calibration model on a volatile or non-volatile data storage medium.

[0083] In an exemplary embodiment, the correction model is a spectral correction model. The spectral correction model is illustratively a transfer function between a wind spectrum of wind speed data associated with the wind turbine and a wind spectrum of reference wind speed information. In an exemplary embodiment, the correction model is a transfer function in the frequency domain. In an exemplary embodiment, the correction model is corrected by the controller using nonlinear regression.

[0084] In an exemplary embodiment, the controller uses a processor to calibrate the correction model based on at least one of: wind speed, turbulence intensity, and length scale.

[0085] By providing a robust mapping of reference wind spectra to wind spectra associated with wind turbines (and vice versa), wind speed data associated with wind turbines may be used to obtain corrected turbulence intensity, and / or reference wind speed information may be used to obtain adjusted turbulence intensity.

[0086] Thus, turbulence intensity estimation is improved and efficient wind turbine operation is achieved.

[0087] Figure 4B A method for calibrating a site correction model according to the present disclosure is shown. In an exemplary embodiment, the site correction model may be calibrated as follows. Wind speed information associated with (one or more) existing sites may be received, such as wind speed information 462 from an anemometer mounted on a meteorological mast in (one or more) existing wind farms.

[0088] Exemplary References Figure 4B In an embodiment, wind speed data associated with the existing location(s), such as collected by sensors mounted on a meteorological mast at the potential new wind farm or potential new location, is retrieved 462. For example, wind speed may be measured using one or more sensors. The sensors may be, for example, 2D anemometers. In some embodiments, the existing location(s) associated with the wind speed information are the same as the reference wind speed information, such as, for example, a reference wind speed information. Figure 4A 、 5A , 5B, 5C and 5D.

[0089] A data connection (WiFi, cable, etc.) may be provided to transmit data, such as received wind speed information associated with the existing location(s). The wind speed information associated with the existing location(s) and / or data used to determine the wind speed information associated with the existing location(s) may be transmitted to the controller. The controller may have a processor and some volatile or non-volatile data storage medium to process and store the wind speed information associated with the existing location(s).

[0090] Exemplary References Figure 4B, one or more wind characteristics 464 associated with the (one or more) existing locations may be determined based on the received wind speed information associated with the (one or more) existing locations. For example, a data connection (WiFi, cable, etc.) may be present to transmit information such as wind speed information associated with the (one or more) existing locations from a data storage medium to the controller. The controller may perform calculations based on the received wind speed information associated with the (one or more) existing locations to determine one or more wind characteristics associated with the (one or more) existing locations. The wind speed information associated with the (one or more) existing locations may be parameters such as turbulence intensity, length scale, vertical shear, wind spectrum, etc. The controller may store the one or more wind characteristics associated with the (one or more) existing locations on a storage medium.

[0091] Exemplary References Figure 4B In one embodiment, wind characteristics of wind spectrum, turbulence intensity and length scale are calculated by the controller based on the received wind speed information associated with (one or more) existing locations, and stored by the controller on a volatile or non-volatile data storage medium.

[0092] Exemplary References Figure 4B , (one or more) corrected turbulence intensity parameters or information for determining (one or more) corrected turbulence intensity parameters may be received 482. (One or more) corrected turbulence intensity parameters or information for determining (one or more) corrected turbulence intensity parameters may be received via a data connection such as WiFi, cable, etc. and stored on a volatile or non-volatile data storage medium. In an exemplary embodiment, as shown in the exemplary reference Figure 3A The described embodiments determine the corrected turbulence intensity parameter(s) as described.

[0093] The corrected turbulence intensity parameter(s) or information used to determine the corrected turbulence intensity parameter(s) may exemplarily extend over the same time period(s) as the wind speed information associated with the existing location(s).

[0094] In the case of receiving information for determining (one or more) corrected turbulence intensity parameters, the controller may use the processor to calculate (one or more) corrected turbulence intensity parameters based on the received information for determining (one or more) corrected turbulence intensity parameters, and store the calculated (one or more) corrected turbulence intensity parameters on a volatile or non-volatile data storage medium. In some embodiments, determining or receiving information corresponding to the exemplary reference in step 482 Figure 4BThe same wind characteristic quantities described in step 464 may be used, for example, one or more wind characteristic quantities associated with the existing location(s), such as turbulence intensity, length scale, and average wind speed.

[0095] Exemplary References Figure 4B Based on one or more wind characteristic quantities (such as turbulence intensity) associated with the existing location(s) and the corrected turbulence intensity parameters(s), the location correction model 486 may be corrected. For example, the controller may use a processor to calculate the location correction model based on one or more wind characteristic quantities associated with the existing location(s) and the corrected turbulence intensity parameters(s), and store the location correction model on a volatile or non-volatile data storage medium.

[0096] In an exemplary embodiment, in addition to one or more wind characteristic quantities associated with an existing site and (one or more) corrected turbulence intensity parameters, site correction model calibration is performed based on other quantities, such as length scale, average wind speed, turbulence intensity or turbulence intensity bins, vertical wind shear, etc., such as described with respect to the correction model described throughout the specification.

[0097] In an example, the controller uses the processor to calibrate the site correction model based on at least one of: wind speed, turbulence intensity or turbulence intensity bins, and length scale.

[0098] In an exemplary embodiment, the corrected turbulence intensity parameter(s) are associated with the same location(s) as the wind speed information associated with the existing location(s).

[0099] By providing a robust mapping of wind speed information associated with an existing site(s) to an adjusted turbulence intensity, and using the wind speed information associated with the existing site(s) to obtain the adjusted turbulence intensity, turbulence intensity estimation is improved and optimal wind turbine siting is achieved.

[0100] Figure 5A A method of calibrating a correction model according to the present disclosure is shown. In embodiments that may be combined with other embodiments described herein, such as with reference to Figure 1 、 2 , 3A, 3B, 4, 5C, and 5D, the wind spectrum model may be used to calibrate the correction model as follows. Wind speed data associated with the wind turbine may be received, such as rotor effective wind speed 512 obtained using a controller based on the wind turbine model. For example, the wind speed data associated with the wind turbine may be received via a data connection such as Wi-Fi, a cable, or the like, and stored on a volatile or non-volatile data storage medium.

[0101] Exemplary References Figure 5A , a wind spectrum model associated with the wind turbine, such as Kaimal wind spectrum model 514, may be determined based on each received wind speed data. For example, the received wind speed data associated with the wind turbine stored on the data storage medium may be retrieved by the controller using a processor via a data connection such as Wi-Fi, a cable, etc. The controller may use the processor to calculate a wind spectrum model based on the received wind speed data associated with the wind turbine, and store the calculated wind spectrum model on a volatile or non-volatile data storage medium.

[0102] Exemplary References Figure 5A , reference wind speed information may be received, such as wind speed data 522 from a weather mast. For example, the reference wind speed information may be sent to the controller via a data connection such as WiFi, a cable, etc. The reference wind speed information may be stored on a volatile or non-volatile data storage medium.

[0103] Exemplary References Figure 5A A reference wind spectrum model, such as Kaimal wind spectrum model 524, may be determined based on the received reference wind speed information. For example, a reference wind spectrum model 524 may be determined for each received reference wind speed data. For example, the received reference wind speed information stored on a data storage medium may be retrieved by the controller using a processor. The controller may use a processor to calculate a wind spectrum model based on the received reference wind speed information and store the calculated wind spectrum model on a volatile or non-volatile data storage medium.

[0104] Exemplary References Figure 5A A correction model 526, such as a spectral correction model, may be calibrated based on a wind spectrum model associated with the wind turbine and a reference wind spectrum model, for example, using nonlinear regression. For example, the controller may use a processor to calculate the correction model based on the wind spectrum model associated with the wind turbine and the reference wind spectrum model. The controller may store the correction model on a volatile or non-volatile data storage medium. In an exemplary embodiment, the correction model is a spectral correction model.

[0105] In reference Figure 5A In the embodiments exemplarily illustrated in other figures such as Figures 4, 5C, and 5D, a wind spectrum may replace a wind spectrum model. For example, a controller may calculate a correction model based on a wind spectrum associated with a wind turbine and a reference wind spectrum, rather than a wind spectrum model associated with the wind turbine and a reference wind spectrum model. The correction model calculated by the controller has the effect of improving turbulence intensity estimation and can achieve efficient wind turbine operation or optimal wind turbine siting.

[0106] Figure 5B A method of calibrating a correction model according to the present disclosure is shown. In embodiments that may be combined with other embodiments described herein, such as with reference to Figure 1 、 2 , 3A, 3B, 4, 5C, and 5D, the turbulence intensity parameter may be used to calibrate the correction model as follows. Wind speed data associated with the wind turbine may be received, such as nacelle anemometer wind speed data 532. For example, the wind speed data associated with the wind turbine may be received via a data connection such as Wi-Fi, a cable, or the like, and stored on a volatile or non-volatile data storage medium.

[0107] Exemplary References Figure 5B The turbulence intensity parameter may be determined based on the received wind speed data 534 associated with the wind turbine. For example, the received wind speed data associated with the wind turbine stored on a data storage medium may be retrieved by the controller using a processor. The controller may use the processor to calculate the turbulence intensity parameter based on the received wind speed data associated with the wind turbine and store the calculated turbulence intensity parameter on a volatile or non-volatile data storage medium.

[0108] Exemplary References Figure 5B , reference wind speed information may be received, such as wind speed data 542 from a set of aeroelastic simulations. For example, the reference wind speed information may be sent to the controller via a data connection such as WiFi, a cable, etc. The reference wind speed information may be stored on a volatile or non-volatile data storage medium.

[0109] Exemplary References Figure 5B , a turbulence intensity parameter 544 may be determined based on the received reference wind speed information. For example, the received reference wind speed data stored on a data storage medium may be retrieved by the controller using a processor via a data connection such as Wi-Fi, a cable, or the like. The controller may use the processor to calculate a turbulence intensity parameter based on the received reference wind speed data. The controller may store the calculated turbulence intensity parameter on a volatile or non-volatile data storage medium.

[0110] Exemplary References Figure 5B The correction model, such as transfer function 546, may be calibrated based on a turbulence intensity parameter associated with the wind turbine and a reference turbulence intensity parameter, for example using a correction scaling. For example, the controller may use a processor to calculate the correction model based on the turbulence intensity parameter associated with the wind turbine and the reference turbulence intensity parameter. The controller may store the correction model on a volatile or non-volatile data storage medium. In an exemplary embodiment, the correction model is a ratio of the reference turbulence intensity parameter to the turbulence intensity parameter associated with the wind turbine, which is a function of at least one of: wind speed, turbulence intensity, and length scale. In an exemplary embodiment, the turbulence intensity parameter is a ratio of the variance of the wind speed to the mean of the wind speed.

[0111] In reference Figure 5BIn the embodiments exemplarily explained in FIG4 , FIG5C and FIG5D , correction scaling or linear regression is used to obtain a correction model. The correction model calculated by the controller has the effect of improving the turbulence intensity estimation and can achieve efficient wind turbine operation or optimal wind turbine site selection. In addition, as shown in FIG5C , FIG5D Figure 5B The correction model described in the exemplary embodiment has a speed advantage. The online or real-time correction of turbulence intensity can be improved accordingly.

[0112] Figure 5C A method of calibrating a correction model according to the present disclosure is shown. In embodiments that may be combined with other embodiments described herein, such as with reference to Figure 1 、 2 , 3A, 3B, 4, 5A, and 5B, the calibration model may be calibrated against reference data as follows. Wind speed data associated with a wind turbine, such as nacelle anemometer wind speed data 552, may be received. For example, the wind speed data associated with the wind turbine may be received via a data connection such as Wi-Fi, a cable, or the like, and stored on a volatile or non-volatile data storage medium.

[0113] Exemplary References Figure 5C , a wind characteristic quantity associated with the wind turbine, such as the length scale parameter 554, may be determined based on each received wind speed data. For example, the received wind speed data associated with the wind turbine stored on a data storage medium may be retrieved by the controller using a processor. The controller may calculate the wind characteristic quantity based on the received wind speed data associated with the wind turbine, and store the calculated wind characteristic quantity on a volatile or non-volatile data storage medium.

[0114] Exemplary References Figure 5C , reference wind speed data may be received, such as non-wake wind speed measurements 562 from (one or more) LIDAR sensors at the wind farm. For example, the received reference wind speed data may be sent to the controller via a data connection such as WiFi, a cable, etc. The reference wind speed data may be stored on a volatile or non-volatile data storage medium.

[0115] Exemplary References Figure 5C , a reference wind characteristic quantity, such as Kaimal wind spectrum model 564, may be determined based on the received wind speed data. For example, the received reference wind speed data stored on a data storage medium may be retrieved by the controller using a processor. The controller may use the processor to calculate a wind characteristic quantity based on the received reference wind speed data and store the calculated wind characteristic quantity on a volatile or non-volatile data storage medium.

[0116] Exemplary References Figure 5CA correction model 566, such as a scaling function, may be calibrated based on wind characteristics associated with the wind turbine and reference wind characteristics, for example, using linear regression. For example, the controller may use a processor to calculate the correction model based on wind characteristics associated with the wind turbine and reference wind characteristics. The controller may store the correction model on a volatile or non-volatile data storage medium. In an exemplary embodiment, the correction model is a spectral correction model.

[0117] Figure 5D A method of calibrating a correction model according to the present disclosure is shown. In embodiments that may be combined with other embodiments described herein, such as with reference to Figure 1 、 2 , 3A, 3B, 4, 5A, and 5B, the calibration model may be calibrated against the reference model as follows. Wind speed data associated with the wind turbine may be received, such as estimated blade wind speed 572 obtained from a controller based on a wind turbine model. For example, the wind speed data associated with the wind turbine may be received via a data connection such as Wi-Fi, a cable, or the like, and stored on a volatile or non-volatile data storage medium.

[0118] Exemplary References Figure 5D Based on each received wind speed data, a wind characteristic quantity associated with the wind turbine, such as the average horizontal wind speed 574, may be determined. For example, the received wind speed data associated with the wind turbine stored on a data storage medium may be retrieved by the controller using a processor. The controller may use the processor to calculate a wind characteristic quantity based on the received wind speed data. The controller may store the calculated wind characteristic quantity on a volatile or non-volatile data storage medium.

[0119] Exemplary References Figure 5D , a wind speed model, such as a Mann spectrum model 582, may be received. For example, the reference wind speed model may be sent to the controller via a data connection such as WiFi, a cable, etc. The reference wind speed model may be stored on a volatile or non-volatile data storage medium.

[0120] Exemplary References Figure 5D , a reference wind characteristic quantity, such as a power spectrum density, may be determined based on each received wind speed model 584. For example, the received reference wind speed data model stored on a data storage medium may be retrieved by the controller using a processor via a data connection, such as Wi-Fi or a cable. The controller may use the processor to calculate a wind characteristic quantity based on the received reference wind speed data. The controller may store the calculated wind characteristic quantity on a volatile or non-volatile data storage medium.

[0121] Exemplary References Figure 5DA correction model 586, such as a spectral correction model, may be calibrated based on wind characteristics associated with the wind turbine and reference wind characteristics, for example, using non-parametric regression. For example, the controller may use a processor to calculate the correction model based on wind characteristics associated with the wind turbine and reference wind characteristics. The controller may store the correction model on a volatile or non-volatile data storage medium.

[0122] In an exemplary embodiment, the correction model is a spectral correction model that is a function of at least one of: wind speed, turbulence intensity, and length scale. In an exemplary embodiment, the correction model is calculated using a nonlinear regression model.

[0123] Figure 6 An example according to the present disclosure is shown. In an embodiment, a wind spectrum model is used to calibrate a correction model against reference data to operate an existing wind turbine. Figure 6 A controller can be used to determine a reference wind spectrum model based on observed reference wind speed data. A controller can be used to determine a wind spectrum model associated with the wind turbine based on wind speed data associated with the wind turbine. Thus, a controller can be used to calibrate a correction model based on the wind spectrum model associated with the wind turbine and the reference wind spectrum model. In an embodiment, a controller can be used to determine a corrected turbulence intensity parameter based on the calibrated correction model and, for example, received rotor effective speed data obtained from an existing wind turbine. In an embodiment, the controller can determine operating parameters of the wind turbine, such as a target rotational speed, based on the corrected turbulence intensity parameter. As a result, the wind turbine can be operated at an optimal power generation level or with an optimal lifespan.

[0124] The following describes embodiments related to wind characteristic quantities. Wind characteristic quantities are generally parameters or quantities derived from wind information, specifically wind speed data or a wind speed model. Wind characteristic quantities can be understood as derived quantities that characterize wind information.

[0125] The characteristic quantity of wind may be any of the following: wind spectrum, wind spectrum model, power spectrum density, turbulence intensity, turbulence intensity squared, integration time scale, integration length scale, vertical wind shear, average wind speed or any combination thereof.

[0126] In an exemplary embodiment, at least one wind characteristic is used. In an exemplary embodiment, the wind characteristic comprises a wind spectrum model, a turbulence intensity, and an integral length scale.

[0127] The wind characteristic quantity may be derived based on a specific time length of wind data (eg, 10 minutes, 5 minutes, 15 minutes, 30 minutes, etc.).

[0128] The following describes embodiments related to operating parameters of a wind turbine. These operating parameters may be related to load management. For example, the target rotor speed may be inversely proportional to the turbulence intensity to maximize the lifespan of the wind turbine and improve the levelized cost of energy. In another example, the accuracy of a wind turbine lifetime odometer may be improved by retrospectively correcting load estimates based on the corrected turbulence intensity.

[0129] The loads or component loads described here are, for example, loads on components of a wind turbine. Components of a wind turbine are, for example, Figure 1 and Figure 2 The components described are, for example, rotor blade(s), tower, shaft, gearbox and generator. The loads or component loads may be extreme loads and / or fatigue loads.

[0130] Operating parameters can also be related to power performance optimization. For example, using a corrected model to provide an estimate of turbulence intensity with lower uncertainty can enable the wind turbine to operate closer to its rated power, thereby improving power production.

[0131] In other examples, the power output may be maximized, or the power output may be fixed at a constant power level or at the rated power of the wind turbine. Maximizing the power output may be advantageous in variable speed operation to maximize power generation. In other examples, fixing the power output may be advantageous when the grid requires a stable output.

[0132] Document US9587628B2 provides examples of efficiency improvements in annual energy production during operation below the rated power of a wind turbine. For example, under low turbulence, the wind turbine is operated at a rotational speed (or tip speed ratio) that includes a small speed deviation (or tip speed ratio deviation) from the optimal rotational speed, for example, a maximum deviation of 8% or even a maximum deviation of 5% from the maximum rotational speed (or optimal tip speed ratio), and typically a deviation of at least 1% or 2%.

[0133] As understood herein, a speed deviation or tip speed ratio deviation from an optimal rotational speed or optimal tip speed ratio should be understood as an expected deviation as disclosed herein. Deviations from the optimal rotational speed or optimal tip speed ratio should not be mistaken for deviations due to changing wind conditions, wherein, however, the wind turbine controller may be configured to avoid any such deviations.

[0134] At higher turbulence, the wind turbine may be operated at high deviations from the optimal rotational speed or tip speed ratio, for example at least 5%, 10% or even 15% deviation from the optimal rotational speed or tip speed ratio.

[0135] Therefore, operating parameters of the wind turbine, such as the rotational speed, may be determined based on the turbulence intensity.

[0136] Other embodiments related to operating parameters of a wind turbine that can be combined with other embodiments described herein are described below. In certain examples, if the turbulence intensity estimated according to the embodiments described herein is lower or higher than assumed, or lower than a conventionally estimated turbulence intensity, the wind turbine can be operated based on the turbulence intensity estimated according to the embodiments described herein, for example, based on a corrected turbulence intensity.

[0137] In certain examples, the turbulence intensity estimated according to the embodiments described herein may be lower than assumed or lower than conventionally estimated turbulence intensity. In certain examples, the wind turbine operates in a region (knee region) between a variable speed or variable power output region of the power curve and a rated power region or maximum power output of the power curve.

[0138] In certain examples, based on the estimated lower corrected turbulence intensity, the wind turbine can be operated more aggressively. Control parameters such as the pitch and speed of the rotor blade(s) can be manipulated. The transition between the variable speed region and the rated power speed region can be smoothed. Operating parameters can be determined to capture power more aggressively while still controlling the load within the operating range.

[0139] In the variable speed region, the pitch may not change. In the variable speed region, the rotor speed may be variable and track the optimal speed to maximize power output. In the rated power region, the pitch may vary. In the rated power region, the rotor speed may be fixed to maintain constant power output.

[0140] In the transition region between the variable speed region and the rated power region, the start of changing the pitch can be delayed, or the pitch can be changed in a controlled or slow manner rather than suddenly. In the transition region, the acceleration of the rotor speed can be slowed down slowly or in a controlled manner rather than suddenly stopped.

[0141] According to embodiments described herein, operating parameters, and hence the operation of the wind turbine, may be determined based on the corrected turbulence intensity parameter.The operating parameters may include any other known operating parameters, such as yaw direction and generator torque demand.

[0142] The following describes embodiments related to siting parameters for new wind turbines. Offline wind resource assessment or wind characterization is related to the ability to build a database of environmental conditions to draw conclusions about current assets or future projects. Accurate estimates of the spatial and temporal wind conditions experienced by the blades provide better load estimates or better wind characterization functions. Consequently, improved turbulence intensity estimates can improve siting for new turbines.

[0143] For example, measurements of wind characteristics, such as turbulence intensity, at a new wind farm site can be used to determine load and performance characteristics of a new wind turbine located at the new wind farm site. Using a site-corrected model, adjusted wind characteristics, such as an adjusted turbulence intensity parameter, can be used to more accurately determine the load and performance characteristics of the new wind turbine.

[0144] Thus, physical properties of a new wind turbine, such as site rated power or load capacity, may be determined more accurately / optimally (eg, with less uncertainty / design margin) based on adjusted wind characteristics, such as adjusted turbulence intensity.

[0145] The following describes embodiments related to wind speed data associated with a wind turbine. Wind speed data can be obtained via virtual or physical sensors. Virtual sensors can be based on operational and state parameters of the wind turbine, such as generated power, rotor speed, and blade pitch dynamics. Virtual sensors, based on a model-based controller, can provide information such as rotor effective wind speed and blade estimated wind speed.

[0146] Alternatively or additionally, an estimated wind speed at the blades may be provided. The physical sensor may be an anemometer, such as a cup anemometer or a sonic anemometer. The anemometer may be placed on the nacelle of the wind turbine or on a mast at the height of the wind turbine.

[0147] Alternatively or additionally, the SODAR and / or LIDAR measurement system may provide wind speed data. Each individual wind speed data received may improve the calibration of the correction model.

[0148] Thus, any combination of wind speed data associated with the wind turbines may also be obtained and / or used.

[0149] The rotor effective wind speed can be understood as follows. It can represent the average of the wind field over the entire rotor area. The rotor effective wind speed can also be understood as the average longitudinal wind speed component or the spatial average of the wind field over the entire rotor plane. The rotor effective wind speed can be based on using the entire rotor as an anemometer.

[0150] The rotor effective wind speed can be determined by considering turbine model characteristics and several measured signals.Determining the rotor effective wind speed can be understood as determining the size of the undisturbed wind field that produces a (unique) combination of power generation, rotational speed and pitch angle at the turbine.

[0151] A power balance estimator, a Kalman filter-based estimator, and / or an extended Kalman filter-based estimator may be used to determine the rotor effective wind speed. In some embodiments, wind speed data associated with the wind turbine, such as the rotor effective wind speed data, may be pre-processed. For example, specific filters may be applied to the input signal, such as filtering high-frequency and / or low-frequency content, and filtering specific frequencies (e.g., blade pass frequency).

[0152] The following describes embodiments related to reference wind speed data. The reference wind speed data can be obtained from any of the following: aeroelastic simulations, large eddy wind models, Mann spectrum models, real-time test measurements in the field, sensors on a meteorological tower (or Met MAST or meteorological mast), data from nearby weather station(s), experimental analogs (such as models in wind chambers), weather forecast models, satellite-based climate data, or any combination thereof.

[0153] Turbulence intensity is described as follows. Turbulence is the random fluctuation of wind vectors in the atmosphere. Therefore, turbulence intensity is a measure that describes the typical amplitude of fluctuations over a specified time period. Turbulence intensity TI can be measured as the standard deviation of wind speed magnitudes over a specific time period (e.g., 10 minutes). Divide by the average wind speed (e.g. over 10 minutes) The ratio is as shown in the formula: .

[0154] In this example, the mean and standard deviation of wind speed are calculated over a 10-minute period. Turbulence intensity can be expressed as a percentage. Turbulence intensity ranges from 0% to 50%. The average turbulence intensity value in the atmosphere at the height of the wind turbine can be between 10% and 20%.

[0155] A 10 minute time scale is typically used (as described elsewhere in this specification) to assess turbulence intensity, but shorter or longer time scales may also be used, for example time periods between 5 and 30 minutes, such as 15 minutes, 20 minutes, etc.

[0156] A meteorological mast with a cup anemometer or sonic anemometer located at the height of the wind turbine is a possible way to measure the turbulence intensity.

[0157] The following describes an embodiment related to wind spectrum model estimation, especially for reference wind speed data. The initial database can be formed as follows. A 10-minute long longitudinal wind speed component u can be obtained. x and the lateral wind speed component u y Large database of time series.

[0158] 2D acoustic anemometers provide longitudinal and transverse wind speed data. The wind speed data may be sampled at a rate of 1 Hz or higher. Data quality checks, such as verifying proper sensor operation and / or, in some cases, such as for reference wind speed measurements, verifying that the sensor or sensor data is non-wake. Flow representing freestream conditions is considered non-wake. A non-wake sensor is understood to be upstream of a wake-generating structure.

[0159] Thus, an initial database can be formed.

[0160] For 2D wind speed data, the longitudinal wind speed can be determined as follows. For each time series, the projection in the longitudinal direction can be determined. The 10-minute average wind direction vector in two dimensions is, It can be given as x- and y-wind speed components according to the following formulas: Therefore, according to the equation , by transforming the two-dimensional 10-minute average wind vector Normalize by its magnitude to get the longitudinal vector .

[0161] Therefore, the x- and y-wind speed components can be projected onto the component vector and The time series of longitudinal wind speed is obtained from the equation: Therefore, the longitudinal wind speed u(t) can be determined.

[0162] Detrending can be performed as follows. Linear detrending can be performed on each time series as follows. Determine the least squares regression of the longitudinal vector u(t) , where a and u0 are the fitted regression parameters. Therefore, the trend-eliminated time series It can be given by the following formula: .

[0163] The power spectrum can be determined using a Fourier transform. The Fourier transform can be performed on each time series as follows. According to the formula , which can be achieved by eliminating the trend of the time series Apply fast Fourier transform to obtain the one-sided power spectrum .

[0164] Binning can be performed as follows. One or more bin sizes can be used. The most typical first bin size is average wind speed. Other optional bin sizes can be used. The second bin size can be illustratively turbulence intensity. The third bin size can be illustratively atmospheric stability. Other bin parameters can be used. Different numbers of bins and / or combinations of bin parameters can also be used.

[0165] Therefore, the power spectra belonging to a given bin can be averaged. Spectral averaging should be performed under similar conditions, such as wind speed and turbulence intensity. For example, all one-sided power spectra belonging to a given bin can be averaged. The average value for each bin can be used for spectral fitting.

[0166] Spectrum fitting can be performed as follows. For each bin or bin combination, a Kaimal model fit can be used to obtain an integral length scale. Spectrum fitting can be performed by defining a Kaimal fit function. The minimum and maximum frequencies for fitting can be defined, for example, from 1 / 600 to 1 Hz.

[0167] The fitting function and the spectrum to be fitted can be normalized by using the variance of the spectrum range within the fitting range, that is, the variance of the spectrum between the minimum and maximum frequencies.

[0168] A fitting cost function can be defined so as to give appropriate weights to the frequencies of interest. The integral length scale Lint can be obtained by minimizing the cost function. Thus, the integral length scale Lint can be determined for each bin.

[0169] A check on the data quality can be performed. This check can be based on the bin count and fit accuracy. For example, bins containing fewer than 10 spectra may be marked as having reduced quality or may be ignored. In another example, minimum fit accuracy can also be used to remove poor quality data, such as poor quality Kaimal power spectrum fits.

[0170] Embodiments related to length scaling and time scaling are described below.

[0171] The integrated time scale or length scale may be understood to represent the time or spatial period, respectively, that represents the typical size or persistence of the most energetic eddies.

[0172] In the atmosphere, vortex size is generally proportional to the altitude at which it is measured. Furthermore, vortex size also depends on atmospheric conditions. At an altitude of 100 meters, a typical vortex size might be around 100 meters. During the night, when atmospheric conditions are relatively stable, vortex sizes can drop to 10 meters. During the day, when atmospheric conditions are unstable, vortex sizes can reach several kilometers.

[0173] In an embodiment, the length scale may be characterized as follows: Note that the terms "integrated time scale" and "time scale" are used interchangeably. The terms "integral length scale" and "length scale" are also used interchangeably.

[0174] Wind speed data can be obtained in the form of a time series of wind speed u(t). The time series of wind speed can be understood as wind speed measurements taken at fixed spatial points. The autocorrelation function , given by the formula: , describes the signal persistence of a statistically stationary process, where s is the lag.

[0175] According to the equation , the time scale can be Determined as the autocorrelation function The autocorrelation function is therefore a time scale that can be used to characterize the time series of wind speed. Time Scale It is a measure of the time-dependent distance of a statistically static process.

[0176] Therefore, when the wind speed is known, the length scale can be determined from the time scale That is, the length scale can be determined as the horizontal wind speed HWS and the time scale according to the following formula The product of: .

[0177] In an embodiment, the time scale It can also be determined from the power spectrum. The power spectrum describes the time scale of the turbulent energy. In other words, the power spectrum It describes whether the turbulent energy is slow or fast, etc. Therefore, it can be understood that the wind spectrum is a function of atmospheric conditions. The energy spectrum is directly related to the autocorrelation.

[0178] Since the autocorrelation function Based on power spectrum, time scale , so the autocorrelation function The integral of is given by the formula: ,in , .

[0179] In an embodiment, the length scale can also be estimated directly using an estimator , for example, by defining model equations that model or equivalently model the dynamics of the wind flow, in particular the meaningful or dominant dynamics.

[0180] The model equation can be expressed as a low-pass filter: , where the cutoff frequency can be determined from the turbulent dynamics of the wind flow.

[0181] According to the formula , the length scale estimator can take the form of a first-order low-pass filter, where It is a turbulent process. is the average wind speed (e.g., the 10-minute average wind speed over a moving window), is available with the cutoff frequency The associated integer length scale, and is the zero-mean Gaussian white noise turbulence process noise. The symbol refers to the time derivative of the turbulent process .

[0182] According to the following , the noise of the turbulent process can be approximated as a Gaussian distribution, where . Turbulence intensity and average wind speed Can be used to approximate turbulent process noise .

[0183] Cutoff frequency , which can be calculated according to the following formula: ,in is maximized frequency.

[0184] Turbulence can be modeled using the Kaimal power spectrum. The Kaimal power spectrum can be approximated as a first-order transfer function. For example, the formula Give the power spectrum .

[0185] Length scale (meters) , turbulence intensity and average wind speed May be as described previously or elsewhere in this specification.

[0186] Can be calculated to maximize Frequency , we get the formula: This relationship enables the determination of the length scale from the characteristic frequencies of the spectrum.

[0187] Therefore, using Kaimal power spectrum, we can use the formula Give the cutoff frequency Average wind speed and the length scale L may be as described previously or elsewhere in this specification.

[0188] Therefore, using the cutoff frequency derived from the Kaimal spectrum model , the state model of the estimator can be given as: Length scale The first-order derivative of can be modeled as zero-mean Gaussian white noise process noise , which can also be called a random walk process.

[0189] In the state model, It is a turbulent process. is the length scale, is the zero-mean Gaussian white noise turbulent process noise, as described above. The cutoff frequency It can be given by the following formula: ,in, is the frequency that maximizes f * S(f). f is the frequency, and S(f) is the Kaimal spectrum.

[0190] Tentatively, the wind process can be modeled as two components: a slow motion process (e.g., the average wind speed ) and fast-moving processes (e.g., turbulent winds ). The measured or estimated wind speed can be written as the average wind speed and turbulent wind speed The sum of the turbulent wind speed The wind speed can be measured or estimated by comparing it with the average wind speed. The measurement model of the estimator can be given by the following formula: ,in The distribution is usually as follows: ,and is the measured or estimated wind speed.

[0191] With the defined state model and measurement model, an estimator is designed to estimate the integral length scale using real-time measured or estimated wind speed data from simulation or wind farm. The estimator can be a least squares estimator or an extended Kalman filter or an unscented Kalman filter or a Monte Carlo filter that can operate on an embedded device.

[0192] Next, embodiments related to the calibration model will be described.

[0193] Examples of calibration models include a simple calibration scaling model, a linear regression model, a nonlinear regression model, or a nonparametric regression model.

[0194] In an exemplary embodiment, the correction model is a spectral correction model, for example, derived from the rotor effective wind speed data PSD REWS The power spectral density and wind speed data PSD from the reference sonic anemometer Sonic The transfer function between the power spectral density of

[0195] For example, the calibration model can be based on the PSD REWS = A(f)* PSD Sonic This takes the form of a correction factor A(f). In this case, the corrected turbulence intensity is The correction factor A(f) may be in any of the following exemplary forms:

[0196] ,

[0197] ,

[0198] ,or

[0199] A(f) =

[0200] , where α, β represent fitting constants, L represents the length scale, V represents the average wind speed, for example, the 10-minute average wind speed, and D is a physical constant.

[0201] Wind speed characteristics such as turbulence intensity TI and length scale L may be estimated in a recursive or online manner. For example, updated turbulence intensity parameter TI and length scale parameter L are recalculated at regular time intervals (eg, 10 minute intervals).

[0202] The updated parameters may be based on the latest wind speed data. In an exemplary embodiment, the updated turbulence intensity parameter and the updated length scale parameter L are used to update the calibration model, e.g. .

[0203] Alternatively, any one factor, other factors, or any combination of factors may be updated. Other factors may include, for example, vertical wind shear. Recursive or online estimation improves the dynamic performance of the calibration model and enables efficient operation of the wind turbine.

[0204] In some embodiments, the correction model may take the form of a ratio. For example, based on the turbulence intensity parameters from the rotor effective wind speed and the reference wind speed data, the ratio of the two turbulence intensity parameters may be empirically derived.

[0205] The ratio is exemplarily a function of at least one of the following parameters: average wind speed, turbulence intensity and length scale. The turbulence intensity parameter is exemplarily the change in wind speed over a time period (eg a 10 minute period) and the average wind speed over the same time period.

[0206] Hence, a fast correction is obtained and an efficient operation of the wind turbine is achieved.

[0207] Some advantages associated with various aspects and / or embodiments are described below: A turbulence intensity estimator for estimating turbulence intensity may be based on a ratio of a standard deviation to a mean value of the rotor effective wind speed.

[0208] This turbulence estimator leads to under-predictions of the turbulence intensity because the rotor effective wind speed is an average of the wind field over a large spatial range. Therefore, this turbulence intensity wind characteristic does not accurately describe the loading conditions experienced by the blade.

[0209] Therefore, turbulence intensity estimation based on the ratio of the standard deviation to the mean value of the rotor effective wind speed leads to poor load predictions.

[0210] In order to more accurately estimate the turbulence intensity of the wind incident on the turbine, it is necessary to first obtain the real-time wind conditions before the wind hits the blades. The real-time wind conditions near the turbine can be provided by transforming or processing the rotor effective wind speed signal.

[0211] The transformation or model may be obtained by learning mast wind speed and rotor effective wind speed estimates offline or online.The transformation or model may use an accurate length scale metric obtainable by an online recursive estimation technique.

[0212] This transformation of the rotor effective wind speed signal uses the estimated length scale to provide a corrected wind speed characteristic that can be used for online load and performance management.

[0213] Alternative approaches include using different correction factors and correction methods. Additional sensors can also be used in the predictor and corrector pathways to improve the accuracy of the turbulence intensity estimate.

[0214] The approach of using a correction model is unique in that the correction model can be applied in an online manner so that the estimated length scale and turbulence intensity can be used for load and performance management. The inventors have demonstrated the concept through simulations using both high fidelity modeling data and data from the field.

[0215] Good online estimation of turbulence intensity and length scale is important for good load management and performance estimation. Performance estimation directly affects annual energy production.

[0216] Typically, a 1% error in turbulence intensity estimates translates to a 1% error in load estimates and a 0.2% error in average cost of energy estimates. Current turbulence estimates typically have an error of about 30%.

[0217] Therefore, improving the accuracy of the turbulence intensity estimate based on the rotor effective wind speed estimate in the model-based controller is important for good load management and performance estimation.

[0218] Improving length-scale estimates is crucial. A 25% uncertainty in length-scale estimates can result in a 2.5% change in damage-equivalent load estimates and a 0.5% change in annual energy production estimates.

[0219] Thus, improved estimates of real-time turbulence intensity for individual wind turbines improve load estimation for lower-level energy costs. Power production control is also improved to achieve higher annual energy production.

[0220] Corrected wind speed characteristics increased annual energy production by more than 0.25% and the damage equivalent load by 1%.

[0221] 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. If such other embodiments 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, then it is intended that such other examples be within the scope of the claims.

[0222] In an example, a method of operating a wind turbine includes determining a correction model associated with the wind turbine; determining a correction turbulence intensity parameter associated with the wind turbine based on the correction model; and / or operating the wind turbine based on the correction turbulence intensity parameter.

[0223] In an example, a method of siting a wind turbine at a new site includes determining a site correction model, determining an adjusted turbulence intensity parameter associated with the new wind turbine at the new site, and / or siting the wind turbine at the new site based on the adjusted turbulence intensity parameter.

[0224] In an example, a method of siting a new wind turbine includes determining a site-corrected model, determining an adjusted turbulence intensity parameter associated with the new wind turbine based on the site-corrected model, and / or siting the new wind turbine based on the adjusted turbulence intensity parameter.

[0225] In an example, an apparatus includes: a wind turbine including at least two rotor blades; and / or a model-based controller configured to determine an average wind speed of the rotor. The wind turbine can be operated or sited according to embodiments described herein.

[0226] In an example, a wind farm comprises at least one wind turbine and / or a wind farm controller.The wind farm controller may comprise a processor configured to perform a method according to embodiments described herein.

[0227] In an example, a system for operating a wind turbine includes a processor configured to perform a method according to embodiments described herein.

[0228] In an example, a method of operating a wind turbine includes determining a correction model associated with the wind turbine; determining a correction turbulence intensity parameter associated with the wind turbine based on the correction model; and / or operating the wind turbine based on the correction turbulence intensity parameter.

[0229] In an example, a method for determining a calibration model includes: receiving a first plurality of wind data and a second plurality of wind data; and processing the first plurality of wind data and / or the second plurality of wind data to determine a first set of estimates and / or a second set of estimates, respectively. Each set of estimates may include at least one from the group consisting of a normalized plurality of wind data, a power spectral density of the plurality of wind data, a turbulence intensity parameter, and a turbulence length scale parameter; and / or calibrating the calibration model based on at least one of the first set of estimates and at least one of the second set of estimates.

[0230] In an example, the first plurality of wind data is wind speed associated with the wind turbine and / or includes wind speed from at least one of the group consisting of rotor average wind speed, blade estimated wind speed, nacelle anemometer wind speed, SODAR based wind speed, and LIDAR based wind speed.

[0231] In an example, the second plurality of wind data includes wind speeds obtained from at least one of the group consisting of a meteorological mast, a set of aeroelastic simulations, and / or a set of measurements associated with a test wind turbine.

[0232] In an example, determining the corrected turbulence intensity includes receiving a third plurality of wind data and / or processing the third plurality of wind data to determine a third set of estimates. The third set of estimates may include an uncorrected turbulence intensity parameter and / or optionally include at least one of the group consisting of a normalized plurality of wind data, a power spectrum, and a turbulence length scale parameter.

[0233] In an example, the third plurality of wind data is wind speed associated with the wind turbine and / or includes at least one from the group consisting of rotor average wind speed, nacelle anemometer wind speed, SODAR based wind speed, and LIDAR based wind.

[0234] In an example, operating the wind turbine based on the corrected turbulence intensity includes determining an operating parameter of the wind turbine based on the corrected turbulence parameter.

[0235] In an example, the correction model includes at least one from the group consisting of a scaling model, a linear regression model, a nonlinear regression model, and a nonparametric regression model.

[0236] In an example, the determination of the corrected turbulence intensity parameter is performed recursively at regular time intervals, which may be between 5 minutes and 30 minutes.

[0237] In an example, the location correction model is calibrated based on at least the second plurality of wind data and / or the corrected turbulence intensity parameter.

[0238] In an example, an adjusted turbulence intensity parameter is determined based on a fourth plurality of wind data and a site-corrected model.

[0239] In an example, new wind turbines are sited based on the adjusted turbulence intensity parameter.

[0240] In an example, the fourth plurality of wind data includes wind speeds obtained from a meteorological mast that is at a location different from the location(s) associated with the second plurality of wind data.

[0241] In an example, a wind turbine includes at least two rotor blades and / or a model-based controller configured to determine a rotor average wind speed.The wind turbine may be operated according to or sited according to embodiments described herein.

[0242] In an example, a wind farm comprises at least one wind turbine and / or a wind farm controller.The wind farm controller may comprise a processor configured to perform a method according to embodiments described herein.

[0243] In an example, a system for operating a wind turbine includes a processor configured to perform a method according to embodiments described herein.

[0244] The present disclosure is intended to include the following listed embodiments, which can be combined with any other embodiments disclosed herein:

[0245] 1. A method for operating a wind turbine, the method comprising:

[0246] determining a calibration model associated with the wind turbine;

[0247] determining a corrected turbulence intensity parameter associated with the wind turbine based on the corrected model; and

[0248] The wind turbine is operated based on the corrected turbulence intensity parameter.

[0249] 2. The method of embodiment 1, wherein determining the calibration model comprises:

[0250] receiving a first plurality of wind data and a second plurality of wind data;

[0251] processing the first plurality of wind data and the second plurality of wind data to determine a first set of estimates and a second set of estimates, respectively, each set of estimates comprising at least one from the group consisting of a normalized plurality of wind data, a power spectral density of the plurality of wind data, a turbulence intensity parameter, and a turbulence length scale parameter; and

[0252] The calibration model is calibrated based on at least one of the first set of estimates and at least one of the second set of estimates.

[0253] 3. The method of embodiment 2, wherein the first plurality of wind data is wind speeds associated with the wind turbine and comprises wind speeds from at least one of the group consisting of rotor average wind speed, blade estimated wind speed, nacelle anemometer wind speed, SODAR-based wind speed, and LIDAR-based wind speed.

[0254] 4. The method of embodiment 2 or 3, wherein the second plurality of wind data comprises wind speeds obtained from at least one of the group consisting of a meteorological mast, a set of aeroelastic simulations, and a set of measurements associated with a test wind turbine.

[0255] 5. The method of any preceding embodiment, wherein determining the corrected turbulence intensity comprises:

[0256] receiving a third plurality of wind data; and

[0257] A third plurality of wind data is processed to determine a third set of estimates, the third set of estimates comprising an uncorrected turbulence intensity parameter and, optionally, at least one from the group consisting of a normalized plurality of wind data, a power spectrum, and a turbulence length scale parameter.

[0258] 6. The method of embodiment 5, wherein the third plurality of wind data is wind speed associated with the wind turbine and comprises at least one from the group consisting of rotor average wind speed, nacelle anemometer wind speed, SODAR-based wind speed, and LIDAR-based wind speed.

[0259] 7. The method of any preceding embodiment, wherein operating the wind turbine based on the corrected turbulence intensity comprises determining an operating parameter of the wind turbine based on the corrected turbulence parameter.

[0260] 8. The method according to any of the preceding embodiments,

[0261] wherein the correction model comprises at least one from the group consisting of a scaling model, a linear regression model, a nonlinear regression model, and a nonparametric regression model; or

[0262] The determination of the corrected turbulence intensity parameter is performed recursively at regular time intervals, wherein the time intervals are between 5 minutes and 30 minutes.

[0263] 9. The method of any preceding embodiment, further comprising calibrating the site correction model based on at least a second plurality of wind data and the corrected turbulence intensity parameter.

[0264] 10. The method of any preceding embodiment, further comprising determining an adjusted turbulence intensity parameter based on a fourth plurality of wind data and a site-corrected model.

[0265] 11. The method of any preceding embodiment, further comprising siting a new wind turbine based on the adjusted turbulence intensity parameter.

[0266] 12. The method of any preceding embodiment, wherein the fourth plurality of wind data comprises wind speeds obtained from a meteorological mast located at a different location than the location(s) associated with the second plurality of wind data.

[0267] 13. A wind turbine comprising at least two rotor blades and a model-based controller configured to determine a rotor average wind speed, wherein the wind turbine is operated according to embodiment 1 or is sited according to embodiment 9.

[0268] 14. A wind farm comprising at least one wind turbine and a wind farm controller, the wind farm controller comprising a processor configured to execute the method according to embodiment 1 or embodiment 9.

[0269] 15. A system for operating a wind turbine, the system comprising a processor configured to perform the method according to embodiment 1 or embodiment 9.

Claims

1. A method for operating a wind turbine, the method comprising: determining a calibration model associated with the wind turbine; determining a calibrated turbulence intensity parameter associated with the wind turbine based on the calibrated model and the received rotor effective wind speed data; and operating the wind turbine based on the corrected turbulence intensity parameter, The rotor effective wind speed is based on the average value of the wind field over the rotor area using the entire rotor as an anemometer.

2. The method according to claim 1, characterized in that Determining the correction model includes: receiving a first plurality of wind data and a second plurality of wind data; processing the first plurality of wind data and the second plurality of wind data to determine a first set of estimates and a second set of estimates, respectively, each set of estimates comprising at least one from the group consisting of a normalized plurality of wind data, a power spectral density of the plurality of wind data, a turbulence intensity parameter, and a turbulence length scale parameter; and The calibration model is calibrated based on at least one of the first set of estimates and at least one of the second set of estimates.

3. The method according to claim 2, characterized in that The first plurality of wind data is wind speeds associated with the wind turbine and includes wind speeds from at least one of the group consisting of rotor average wind speed, blade estimated wind speed, nacelle anemometer wind speed, SODAR based wind speed, and LIDAR based wind speed.

4. The method according to claim 2 or claim 3, characterized in that The second plurality of wind data includes wind speeds obtained from at least one of the group consisting of a meteorological mast, a set of aeroelastic simulations, and a set of measurements associated with a test wind turbine.

5. The method according to claim 1, wherein Determining the corrected turbulence intensity includes: receiving a third plurality of wind data; and The third plurality of wind data is processed to determine a third set of estimates comprising an uncorrected turbulence intensity parameter and, optionally, at least one from the group consisting of a normalized plurality of wind data, a power spectrum, and a turbulence length scale parameter.

6. The method according to claim 5, characterized in that The third plurality of wind data is wind speed associated with the wind turbine and includes at least one from the group consisting of rotor average wind speed, nacelle anemometer wind speed, SODAR based wind speed, and LIDAR based wind speed.

7. The method according to claim 1, characterized in that Operating the wind turbine based on the corrected turbulence intensity includes determining an operating parameter of the wind turbine based on the corrected turbulence parameter.

8. The method according to claim 1, It is characterized in that The correction model comprises at least one from the group consisting of a scaling model, a linear regression model, a nonlinear regression model, and a nonparametric regression model; or The determining of the corrected turbulence intensity parameter is performed recursively at regular time intervals, wherein the time intervals are between 5 minutes and 30 minutes.

9. The method according to claim 2, characterized in that The method further includes calibrating a site correction model based on at least the second plurality of wind data and the corrected turbulence intensity parameter.

10. The method according to claim 9, characterized in that The method further includes determining an adjusted turbulence intensity parameter based on a fourth plurality of wind data and the site-corrected model.

11. The method according to claim 10, characterized in that The method further includes siting a new wind turbine based on the adjusted turbulence intensity parameter.

12. The method according to claim 10 or claim 11, characterized in that The fourth plurality of wind data includes wind speeds obtained from a meteorological mast located at a location different from one or more locations associated with the second plurality of wind data.

13. A wind turbine comprising at least two rotor blades and a model-based controller configured to determine a rotor average wind speed, wherein the wind turbine is operated according to claim 1 or is sited according to claim 9.

14. A wind farm comprising at least one wind turbine and a wind farm controller, the wind farm controller comprising a processor configured to perform the method according to claim 1 or claim 9.

15. A system for operating a wind turbine, the system comprising a processor configured to perform the method of claim 1 or claim 9.

Images