Control of wind turbine wake loss using detected downstream wake loss severity

Abstract

The invention relates to a wind turbine for controlling wake generation during operation. The wind turbine is part of a wind farm comprising a plurality of wind turbines. The invention includes receiving, from a further wind turbine of a plurality of wind turbines located downstream of the wind turbine, a severity parameter indicative of a severity of wake loss experienced at the further wind turbine. Determining one or more wake loss control actions for regulating wake generated by the wind turbine, the invention including controlling wind turbine operation according to the determined one or more wake loss control actions.

Description

Technical Field

[0001] The present invention relates to controlling wind turbines of a wind park comprising a plurality of wind turbines. In particular, the present invention relates to controlling a wake generated by a wind turbine to control the wind turbine according to one or more wake loss control actions determined based on a wake loss severity detected at another wind turbine of the plurality of wind turbines. Background Art

[0002] Wind turbines are used to capture energy from wind as it flows through them, and to generate electrical energy from the captured energy, for example to supply to an electrical grid. Often, several wind turbines are located in relatively close proximity to one another in a geographic area, where such a grouping of wind turbines may be referred to as being grouped together to form a wind farm or wind park.

[0003] The amount of wind energy that can be captured by a wind turbine depends on various environmental factors, such as wind speed and wind direction. For example, a wind turbine can often be most efficient at capturing wind energy when the turbine's rotor or nacelle faces directly in the direction of the incoming wind, i.e., when the wind turbine is "aligned" with the wind.

[0004] As the wind flows past the wind turbine, a wake is generated downstream of the wind turbine. That is, the flow of wind downstream of the wind turbine is disturbed or perturbed relative to upstream of the wind turbine. This disturbance may result in a reduction in the velocity of the wind flow and / or an increase in the turbulence of the wind flow. Each of these results may result in a reduction in the available energy that can be captured from the wind.

[0005] In a wind farm, the wake generated by a first, upstream wind turbine may impinge on a second, downstream wind turbine, causing the power generation efficiency of the downstream wind turbine to decrease (relative to if the upstream wind turbine did not exist, i.e., relative to if the wake effect caused by the upstream wind turbine did not exist). This may be referred to as the wake loss experienced by the downstream wind turbine.

[0006] It is known to perform so-called "wake steering" of wind turbines to steer the wake generated by an upstream turbine away from a downstream turbine. This may involve controlling the upstream turbine out of alignment with respect to the incoming wind, for example, by performing yaw control of the upstream turbine. While this may reduce the energy capture efficiency of the upstream turbine, the increase in the energy capture efficiency of the downstream turbine may result in an overall increase in the energy capture efficiency of the wind farm.

[0007] Known methods of performing wake steering may be disadvantageous in that wake steering is not performed when required, is performed when not required, and / or is performed incorrectly such that the wake effect experienced by downstream turbines is not properly achieved. In particular, known methods may not always achieve an overall increase in power generation (or energy capture) at the wind farm level.

[0008] An example of wake control to achieve an overall increase in power generation is found in EP 2 063 108 A2. Therein is disclosed the determination of the wake state of a downwind turbine based on data received from an upwind turbine.

[0009] It is against this background that the present invention is situated. Summary of the invention

[0010] According to one aspect of the invention, a method for controlling a wind turbine generating a wake during operation is provided. The wind turbine is part of a wind farm comprising a plurality of wind turbines. The method comprises receiving a signal from another wind turbine of the plurality of wind turbines located downstream or downwind of the wind turbine indicating a severity of a wake loss experienced at the other wind turbine. The method comprises, based on the received severity signal, determining one or more wake loss control actions for adjusting a wake generated by the wind turbine. The method comprises controlling wind turbine operation according to the determined one or more wake loss control actions.

[0011] The one or more wake loss control actions may be part of a predefined wake loss control strategy for controlling the wind turbine to adjust the wake generated by the wind turbine as a function of wind direction in the vicinity of the wind turbine.

[0012] The severity signal is a signal indicative of the strength of the wake, also denoted as a signal indicative of a wake loss experienced at the further wind turbine, for example by detecting the strength of the wake at the further wind turbine downstream of the wind turbine. The downstream wind turbine determines the strength of the wake experienced by the downstream wind turbine and converts the detected strength into a severity signal for transmission to the wind turbine generating the wake, i.e. the upstream wind turbine (or the wind turbine). The upstream wind turbine receives the severity signal as a received severity signal.

[0013] The received severity signal may be a gain. The predefined wake loss control strategy may be a gain-scheduled control strategy. The method may include applying the gain to the gain-scheduled control strategy to determine one or more wake loss control actions to be performed. Controlling the wind turbine may include controlling the wind turbine according to the gain-scheduled control strategy.

[0014] If the received gains indicate that the wake loss severity experienced at the other wind turbine is above a predefined threshold, the gain-dispatch control strategy may include executing one or more wake loss control actions to adjust the wake generated by the wind turbine.

[0015] In some examples, if the received gains indicate that the wake loss severity experienced at the other wind turbines is below a predefined threshold, no wake loss control action is performed as part of the gain-dispatching control strategy.

[0016] The method may comprise determining a severity parameter indicative of a severity of wake loss experienced by the further wind turbine; and transmitting the determined severity parameter as a severity signal to the wind turbine.

[0017] In an embodiment, the severity parameter reflects a determined lack of wind speed at the further wind turbine. The wake severity may be based on a determined lack of (rotor averaged) wind speed experienced at the further turbine. A lack of rotor averaged wind speed risks reducing power performance. A lack of rotor averaged wind speed may be determined based on a comparison of power production between an upstream turbine (the wind turbine) and a downstream turbine (the further wind turbine). Measurement differences in wind speed measurements (e.g. nacelle anemometer wind speed measurements) may also be used to indicate a lack of wind speed. However, the measurements here should serve as a basis for estimating the rotor averaged wind speed rather than simply using point measurements. Furthermore, differences in turbulence intensity between the wind turbine and the further wind turbine may be used to determine the turbulence added to the wake.

[0018] The determined severity parameter, when received at the wind turbine as a severity signal, may be compared with similar signals of the wind turbine to determine that the wind speed is insufficient.

[0019] By determining on which side of the wake the further wind turbine is located this severity parameter may be improved.

[0020] The method may include, at the further wind turbine: determining a severity parameter indicative of a severity of wake losses experienced at the further wind turbine; and transmitting the determined severity parameter as a severity signal to the wind turbine.

[0021] It may be advantageous to determine the severity parameter for the further wind turbine without taking into account the signal from the (upwind) wind turbine. In this way, wake loss control actions may be based on the actual wake state at the further wind turbine.

[0022] In an embodiment, the severity signal and / or the severity parameter is determined based on sensor measurements at said wind turbine experiencing the wake loss.

[0023] The method may include receiving sensor signals from one or more sensors of the further wind turbine and determining an imbalance parameter indicative of a rotor load imbalance of the further wind turbine based on the received sensor signals. A severity parameter may be determined based on the determined imbalance parameter and be a magnitude indicative of a load imbalance.

[0024] The sensor signals from the one or more sensors may be blade load signals from one or more blade load sensors of rotor blades of the further wind turbine.The imbalance parameter may be a yaw moment of the rotor of the further wind turbine determined based on the received blade load signals.

[0025] The severity parameter may be determined based on the magnitude of the imbalance parameter and the wind direction relative to the full wake condition experienced by the wind turbine.

[0026] The method may include determining a curve describing the imbalance parameter as a function of wind direction or nacelle yaw position based on the magnitude of the imbalance parameter and relative to a defined wind direction. The method may include comparing the determined shape with a plurality of defined shapes, each defined shape being associated with a respective severity parameter. The severity parameter may be determined by the comparison.

[0027] The imbalance parameter may be normalized based on one or more operating variables. The severity parameter may be determined based on the normalized imbalance parameter. Optionally, the one or more operating variables may include one or more of: a defined peak value of the imbalance parameter; wind speed; absolute output power; output power normalized to rated power; and, an estimated boost level of the further wind turbine.

[0028] The signal indicative of wake loss severity may in embodiments be received from two or more turbines, wherein the one or more wake loss control actions to adjust the wake are determined based on the severity of the severity signals received from the two or more turbines.

[0029] The one or more wake loss control actions may include at least one of the following actions: performing yaw control to rotate the nacelle and the rotor of the wind turbine relative to the tower of the wind turbine about a yaw angle to adjust the direction of the wake generated by the wind turbine; performing pitch control to generate a pitch moment about a pitch axis to adjust the direction of the wake generated by the wind turbine; collective pitch control of the rotor blades of the wind turbine; and individual pitch control of the rotor blades of the wind turbine.

[0030] According to another aspect of the present invention, there is provided a non-transitory, computer-readable storage medium having stored therein instructions which, when executed by one or more computer processors, cause the one or more computer processors to perform the method defined above.

[0031] According to another aspect of the invention, a controller for controlling a wind turbine that generates a wake during operation is provided. The wind turbine is part of a wind farm that includes a plurality of wind turbines. The controller is configured to receive a signal from another wind turbine of the plurality of wind turbines that is located downstream or downwind of the wind turbine indicating a wake loss severity experienced by the other wind turbine. The controller is configured to determine one or more wake loss control actions based on the received severity signal to adjust the wake generated by the wind turbine. The controller is configured to control the wind turbine operation based on the determined one or more wake loss control actions.

[0032] According to another aspect of the invention, a control system for a wind park is provided, the wind park comprising the wind turbine as described above and the further wind turbine. The control system comprises a controller as defined above. The control system comprises a further controller for controlling the further wind turbine, the further controller being configured to: determine a severity parameter indicating a severity of wake loss experienced by the further wind turbine; and transmit the determined severity parameter as a severity signal to the controller of the wind turbine.

[0033] According to an aspect of the invention, there is provided a wind turbine comprising a controller as defined above.

[0034] According to another aspect of the present invention, there is provided a wind farm comprising the control system defined above. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Examples of the present invention will now be described with reference to the accompanying drawings, in which:

[0036] Figure 1 Schematically illustrates a wind farm comprising a plurality of wind turbines according to an aspect of the invention;

[0037] Figure 2 (a) Schematic diagram showing the Figure 1 The wake generated downstream of a wind turbine when it is aligned with the incoming wind direction, Figure 2 (b) The schematic diagram shows the Figure 2 The wake generated when the wind turbine in (a) is not aligned with respect to the incoming wind direction;

[0038] Figure 3 Show Figure 1illustrative graph of estimated yaw moment versus absolute wind direction downstream of one of the wind turbines generating a wake downstream of another wind turbine;

[0039] Figure 4 Schematically shown Figure 1 One of the downstream wind turbines is subjected to another upstream wind turbine Figure 1 The effects of wind turbine-generated wakes, where: Figure 4 (a) shows the downstream wind turbine in full wake state; Figure 4 (b) shows the downstream wind turbine in the left half-plane wake state; Figure (c) shows the downstream wind turbine in the right half-plane wake state;

[0040] Figure 5 The present invention is shown in one aspect Figure 1 A schematic diagram of a controller of one of the wind turbines generating a wake downstream thereof; and

[0041] Figure 6 The present invention is shown in one aspect Figure 5 The controller performs the steps of the method. DETAILED DESCRIPTION

[0042] The present invention provides a method and system for monitoring wake losses of one or more downstream wind turbines in a wind farm relative to upstream wind turbines, and controlling the upstream wind turbines based on the wake losses monitored at these downstream turbines, for example by performing wake steering of the upstream turbines. In particular, the effects of the wake generated by the upstream wind turbine on the downstream wind turbines, for example the severity of the loads experienced by one or more components in the downstream turbines (particularly, for specific wind conditions), are monitored, and appropriate control of the upstream turbines to mitigate these effects based on the identified wake loss severity is performed, for example, in a manner that increases or maximizes the power output of the entire wind farm. This is in contrast to some known wake steering methods, in which only the wind conditions of the (upstream) wind turbine to be controlled, or only the wind conditions of one or more downstream turbines, are considered when determining how to control the wind turbines.

[0043] Figure 1A schematic illustration of a wind farm or wind plant 10 including a plurality of wind turbines 12 is shown. Each wind turbine 12 includes a tower 121, a nacelle disposed at the top or upper end of the tower, and a rotor operably coupled to a generator housed within the nacelle. In addition to the generator, the nacelle also houses other components required to convert wind energy into electrical energy and various components required to operate, control and optimize the performance of the wind turbine 12. The rotor of the wind turbine 12 includes a central hub and three rotor blades 122 protruding outwardly from the central hub.

[0044] Each wind turbine 12 includes a control system or controller (not shown). Figure 1 ). The controllers may be placed in a nacelle, in a tower, or distributed in multiple locations inside (or outside) the turbine 12 and communicatively connected to each other. In addition, the wind farm 10 may include a (central) controller communicatively connected to the wind turbine controllers.

[0045] The rotor blades 122 are pitch adjustable. The rotor blades 122 can be adjusted according to a collective pitch setting, wherein each of the blades is set to the same pitch value. In addition, the rotor blades 122 can be adjusted according to individual pitch settings, wherein each blade 122 can be provided with an individual pitch setting value. The control system / controller of the corresponding wind turbine 12 can determine the collective and / or individual pitch settings, and output / transmit control signals to appropriate actuators of the wind turbine 12 to actuate the pitch bearings of the wind turbine 12 to control the pitch angle of the rotor blades 122 according to the determined pitch setting.

[0046] Each wind turbine 12 may be configured to adjust yaw, for example, relative to the wind in the vicinity of the respective wind turbine 12. In particular, each turbine 12 may include a yaw bearing between the tower 121 and the nacelle, which allows rotational movement of the nacelle (and accompanying components, including the rotor and rotor blades 122) relative to the tower in order to adjust the yaw angle of the wind turbine 12 relative to the wind, i.e., rotation (lateral or horizontal adjustment) about the tower axis of the turbine 12. The control system / controller of the respective wind turbine 12 may determine a desired yaw angle for the wind turbine 12 and output a control signal to control a yaw actuation mechanism of the wind turbine 12 to rotate the nacelle relative to the tower 121 via the yaw bearing according to the desired yaw angle.

[0047] Wake steering can also be achieved through pitch moment control, whereby a single pitch mechanism generates a pitch moment on the rotor which can steer the wake in the vertical direction.

[0048] Each of the wind turbines 12 in the wind farm 10 is configured to capture energy from the wind flow and convert the captured wind energy into electrical energy, such as to be provided to a power grid. It is often desirable to maximize the amount of wind energy captured by the wind turbines in order to maximize the amount of power generated by the turbines. Each wind turbine 12 monitors wind conditions in its vicinity, and appropriately controls / adjusts one or more components of the wind turbine 12 based on the monitored wind conditions to maximize the capture of wind energy. Each wind turbine 12 may include one or more sensors to measure one or more aspects of wind conditions in the vicinity of the turbine 12, such as wind speed, wind direction, etc. For example, each turbine 12 may include one or more accelerometers, such as located in a nacelle, for this purpose.

[0049] Each wind turbine 12 may be controlled to balance maximizing the captured energy / power production of the turbine against (minimizing) the loads experienced by one or more components of the turbine 12. If the loads experienced by a wind turbine component (e.g., extreme or fatigue loads) are too high, this may lead to a shortened component life or even failure. Each turbine 12 may include sensors for monitoring the loads on different wind turbine components. For example, each turbine 12 may include a blade load sensor placed at or near the root end of each blade 122 in such a way that the sensor detects the load in the blade 122. Depending on the placement and type of the sensor, the load may be detected in the flap (flap-wise) direction (in-plane / out-of-plane) or the edge (edge-wise) direction (in-plane). For example, such a sensor may be a strain gauge sensor or an optical Bragg sensor.

[0050] Generally, in order to maximize the energy that a wind turbine captures from the wind, the wind turbine can be controlled to align with the incoming wind direction. That is, the wind turbine can be controlled so that the rotor or nacelle points directly into the incoming wind or into the wind. The difference between the wind direction and the nacelle / rotor direction - that is, where the wind turbine is out of alignment with the wind direction - can be called yaw error.

[0051] Figure 1 The direction 14 of the wind flow of the wind farm 10 is schematically illustrated. As the wind flows past the first turbine 12a in the wind farm 10, a wake is generated downstream of the wind turbine 12a. This means that the wind flow downstream of the wind turbine 12a is disturbed or disrupted relative to the wind flow upstream of the wind turbine 12a, resulting in a reduction in the wind flow speed and / or an increase in the turbulence of the wind flow.

[0052] Depending on the position of the other wind turbines 12b relative to the (first) wind turbine 12a in the wind park area 10, the wind flow through one or more other wind turbines 12b may include a wake effect caused by the wind flow through the first wind turbine 12a. The wind turbine that generates / causes the wake may be referred to as an upstream or upwind wind turbine 12a, while the one or more wind turbines that are affected by the generated wake may be referred to as a downstream or downwind wind turbine 12b.

[0053] Due to the influence of the wake of the upstream wind turbine on the downstream wind turbine, the upstream wind turbine tends to generate more energy than the downstream wind turbine. In particular, the wake effect from the upstream wind turbine results in a reduction in wind speed and an increase in turbulence near the downstream wind turbine relative to the upstream wind turbine. It is known to control the upstream wind turbine to adjust the generated wake in a manner intended to reduce the influence of the wake on one or more wind turbines downstream of the upstream wind turbine. In particular, for example, so-called wake steering can be performed to change the direction of the generated wake. This can be achieved by misaligning the upstream wind turbine with respect to the incoming wind direction.

[0054] Figure 2 Schematic showing how to use wake steering to adjust the generated wake. In particular, Figure 2 (a) shows a situation where the upstream wind turbine 12a is aligned with the incoming wind direction 14. In this case, it can be seen that the wake 20 generated downstream of the upstream wind turbine 12a is directed toward another wind turbine 12b downstream of the upstream wind turbine 12a. Since the downstream wind turbine 12b is affected by the generated wake 20, this then reduces the amount of wind energy that can be captured by the downstream wind turbine 12b. Figure 2 (b) shows a situation where the upstream wind turbine 12a is not aligned with respect to the incoming wind direction 14, e.g. Figure 2 (a) The yaw angle of the upstream wind turbine 12a is adjusted. It can be seen that this changes the direction of the generated wake 20 so that the downstream wind turbine 12b is not affected by the generated wake 20, or at least has its impact reduced.

[0055] Known methods for performing wake steering may be based on monitored wind conditions in the vicinity of an (upstream) wind turbine to be controlled, and on retrievable information about the layout of the wind park, i.e. the positioning of the wind turbines relative to each other in the wind park. For example, for a specific measured - or otherwise determined (e.g. estimated) - wind direction in the vicinity of an upstream wind turbine to be controlled, it may be predicted that a wake with a specific direction and / or a specific intensity / severity will be generated downstream of the wind turbine, e.g. when the wind turbine is aligned with the wind direction. If the predicted wake direction and / or severity is such that its impact is expected to be experienced by further wind turbines downstream of the upstream wind turbine (based on the wind park layout information), one or more wake control actions may be performed, e.g. a wake steering of the upstream wind turbine may be performed to adjust the direction and / or severity of the wake generated by said wind turbine.

[0056] However, this known method may not always be able to accurately predict when the downstream wind turbine will experience the generated wake effect, and the amount of wind energy that the downstream turbine may capture is not conducive. There may be several reasons for this. For example, the wind farm layout available for the upstream wind turbine may be inaccurate, that is, the relative positioning of the wind turbines in the wind farm may be inaccurate, so that when the generated wind flow is directed to one or more downstream turbines, it is incorrectly predicted. In addition, other aspects of the prevailing wind conditions, such as wind speed, turbulence level, wind shear / steering, atmospheric stability, can affect the wake generated downstream of the wind turbine and its development. In addition, different aspects of the wind farm, such as terrain and / or vegetation between different turbines, can affect the development and path of the wake. The wind direction measurement and / or positioning (e.g., yaw angle) of the rotor or nacelle of the upstream turbine used to determine and adjust the wake may be inaccurate (e.g., if the sensor used to measure these quantities is faulty or calibrated incorrectly), which can also lead to differences between the actual and predicted wake effects.

[0057] The present invention is advantageous in that it provides a method and system for reducing the wake losses (i.e., the reduction in wind energy capture efficiency or capability) incurred or experienced by wind turbines in a wind farm in a manner that can increase or maximize the overall wind energy capture of a wind farm comprising a plurality of wind turbines. In particular, this is achieved by monitoring the (actual) impact of an upstream wind turbine on the wake generated by one or more downstream wind turbines, and using these monitored effects from the downstream turbines to determine how to control the upstream turbine to reduce the wake losses experienced on the downstream turbines, thereby increasing the overall wind power generation level.

[0058] The present invention particularly uses the monitored downstream wake effect to determine a wake loss severity experienced by the downstream wind turbine, and uses the determined wake loss severity to determine how to control the upstream wind turbine with respect to the generated wake adjustment. For example, if the wake loss severity experienced by the downstream wind turbine is above a defined threshold severity, it may be determined to control the upstream wind turbine to perform one or more wake loss control actions, such as by actuating a predefined wake loss control strategy of the upstream wind turbine. For example, if the downstream wake loss severity is below a certain threshold, controlling the upstream turbine in a manner that reduces the downstream wake effect may not achieve the desired effect of increasing the overall power generation of the wind farm.

[0059] An estimate of when to initiate a wake loss control strategy at an upstream wind turbine to maximize overall wind farm power generation may already be included in the upstream wind turbine control strategy. However, the estimated or predicted downstream wake loss severity may be based on wind conditions monitored at the upstream turbine or at the wind farm. Due to one or more of the causes / sources of error listed above, the wake losses actually experienced downstream may differ from the expected level. The present invention therefore beneficially uses the wake loss severity actually monitored at one or more downstream wind turbines to determine whether executing one or more wake control actions on the upstream turbine actually has the expected result of increasing overall wind farm power generation.

[0060] Reference Figure 1 and Figure 2 , the controller of the upstream wind turbine 12a can be configured to implement a predefined wake loss control strategy as a function of wind direction and other monitored wind conditions in the vicinity of the wind turbine 12a. The predefined wake loss control strategy may involve the controller performing one or more control actions to reduce or mitigate wake losses experienced by one or more of the downstream wind turbines 12b under specific monitored wind directions predicted or expected to cause downstream wake losses. For example, the control action may include yaw angle control of the upstream turbine 12a to redirect the downstream wake away from the downstream turbine 12b. The predefined control strategy may be activated to perform control actions, such as wake steering, to be monitored within a predefined range or wind direction interval ("wake sector") that is believed to cause wake losses. On the other hand, if the monitored wind direction is outside the predefined range, the predefined control strategy may be deactivated so that the control action to mitigate the wake loss is not performed. When the predefined control strategy is not activated (deactivated), the upstream wind turbine 12a may be controlled according to a standard control strategy, for example, to maximize power generation by aligning the wind turbine 12a with the incoming wind direction 14. The predefined wake loss control strategy, i.e., which control action is performed for which wind condition, may be determined offline based on historical, simulated or experimental data, or in any other suitable manner (e.g., including machine learning methods) such that it is known a priori.

[0061] As well as wind direction, the predefined strategy may also take into account (i.e., take it as a function of) other wind conditions, such as wind speed. For example, even if the wind direction is such that the downstream turbine is predicted to be in the generated wake, if the wind speed is relatively low, it may not be worthwhile for the upstream turbine to perform a wake turn. This may be because the reduction in wake losses at one or more downstream turbines is sufficient to offset the reduction in energy capture efficiency of the upstream turbine caused by the wake turn. Thus, in one example, for example, the predefined control strategy may only be activated when the wind speed is sufficiently high.

[0062] In order to determine the actual severity of the wake generated by the upstream wind turbine 12a and experienced by one or more downstream wind turbines 12b, one or more aspects of the operation of the downstream wind turbine(s) 12b are monitored. For example, the loads experienced by the rotor blades 122 of the downstream wind turbine 12b may be monitored as a means to detect the wake effect or wake loss experienced by the downstream wind turbine 12b, e.g., the blade loads may increase as the severity of the wake flow in which the wind turbine is operating increases.

[0063] In some examples, a parameter indicative of a downstream turbine rotor load imbalance may be determined and used as an indication of wake loss. In one such example, an estimated or measured rotor tilt or yaw moment, for example based on blade load sensor signals, may be used to detect wake loss. If there are other factors affecting the rotor load imbalance, these factors may be removed or compensated for before performing subsequent analysis. For example, if the wind turbine has separate pitch control (IPC) activity, the (measured) rotor yaw moment may be compensated to account for the imbalance correction performed by the IPC.

[0064] In addition, the imbalance parameter is normalized based on one or more operating variables of the downstream turbine 12 b, and the normalized imbalance parameter determines the severity parameter. Such operating variables may include a defined peak value of the imbalance parameter, wind speed, absolute output power, output power normalized based on rated power, and / or an estimated boost level of the downstream wind turbine 12 b.

[0065] Furthermore, the severity parameter may be normalized based on one or more factors. For example, it may be normalized based on measurements of other turbines in the wind farm. If the distance between a first turbine and a turbine of another turbine is small, the severity parameter of the first turbine may be greater than the severity parameter of the other turbine. The severity parameters of each turbine may be sent to a central unit, for example in the wind farm, and they may be compared and normalized with respect to the largest severity parameter and then sent back to the respective turbine.

[0066] When the maximum yaw peak is identified, it can be normalized based on the yaw moment peaks from other turbines (possibly also in other wind farms) with the same rotor size and distance to the upwind turbine. The starting point for normalization can be simulation curves for various scenarios (for a given rotor type, power rating, turbulence conditions, wind shear conditions, etc.).

[0067] The severity parameter may also be normalized based on values ​​of a severity measurement database of other turbines operating at different wind farms, a predefined most severe wake considered possible / feasible (e.g. yaw moment trajectory as a function of wake), and / or other wind directions from the same turbine. Such normalization may be to avoid a situation where the maximum severity that a (downstream) turbine may experience always results in a maximum wake loss control of an upstream wind turbine, even if the maximum severity is not globally large compared to the other turbines.

[0068] Figure 3 Illustrative graphs / curves 30, 32, 34 are shown of the rotor yaw moment (along the y-axis, for example in kNm) versus the absolute wind direction (along the x-axis, in degrees) for the downstream wind turbine 12b, which allow for the detection of wake conditions when operating the turbine 12b, and the severity of said wake conditions. Figure 4 , the rotor yaw moment allows determining whether the downstream wind turbine 12b is in a so-called "full wake state", "left half plane wake state", "right half plane wake state", somewhere between these defined wake states, or outside these wake states.

[0069] Figure 4 (a) schematically shows a schematic diagram of the downstream wind turbine 12b in a full wake state. In particular, in the full wake state, the downstream wind turbine 12b is fully in the wake 20 generated by the upstream wind turbine 12a. When in this case, the blade loading effect caused by the wake at the downstream turbine 12b may be substantially balanced (equal) between the left and right sides / halves of the rotor plane, so that the yaw moment experienced by the rotor of the downstream turbine 12b is substantially zero, corresponding to Figure 3 The complete wake state points are 301, 321, 341 in Figures 30, 32, and 34 respectively.

[0070] Figure 4(b) schematically shows the downstream wind turbine 12b in a left half plane wake condition. In particular, in the left half plane wake condition, (only) the left half of the rotor plane of the downstream turbine 12b - e.g. defined as the swept area of ​​the rotor blades 122 of the turbine 12b - is in (or is affected by) the wake 20 generated by the upstream wind turbine 12a. In this case, this may result in an imbalance of the maximum level / amount of blade loading between the left and right halves of the rotor plane, such that the magnitude of the yaw moment experienced by the rotor of the downstream turbine 12b is maximized. This corresponds to Figure 3 The left half plane trail state points 302, 322, 342 are shown in Figures 30, 32, and 34 respectively.

[0071] Figure 4 (c) schematically shows the downstream wind turbine 12b in a right half plane wake condition. This corresponds to the left half plane wake condition, except that (only) the right half rotor plane is in the wake 20 generated by the upstream wind turbine 12a. This corresponds to Figure 3 The right half plane trail state points 303, 323, 343 in Figures 30, 32, and 34 respectively.

[0072] refer to Figure 3 Combination Figure 4 , it will be appreciated that the points 304, 305, 324, 325, 344, 345 of the estimated yaw moment diagrams 30, 32, 34 correspond to wind directions where the downstream turbine 12b is just outside the wake 20 generated by the upstream turbine 12a, i.e., points where the generated wake does not affect the estimated or measured yaw moment. It may be desirable for the wake loss control strategy for the upstream turbine 12a to be activated / deactivated at wind directions 304, 305, 324, 325, 344, 345; however, the activation / deactivation wind direction may be set at any suitable wind direction based on the estimated or measured yaw moment. For example, it may be desirable for the downstream turbine to be sufficiently within the generated wake before activating the upstream control strategy to mitigate the effects of the downstream wake loss. It may also be desirable to activate and deactivate the wake loss control strategy at different wind directions, i.e., to introduce hysteresis into the control strategy. This may prevent repeated activation and deactivation cycles of the control strategy and may also prevent deactivation of the control strategy from causing an increase in the yaw moment of the downstream turbine (since the wake may move back in the direction of the downstream turbine).

[0073] It can be seen that, therefore, Figure 332 (i.e., point 322) and is minimum for FIG. 34 (i.e., point 342). The variation in the rotor yaw moment for a given wind direction at which the downstream wind turbine 12b experiences wake losses, such as wind directions between points 304, 324, 344 and points 305, 325, 345, may vary. For example, when in a left half plane wake condition, the magnitude of the rotor yaw moment is maximum for FIG. 32 (i.e., point 322) and minimum for FIG. 34 (i.e., point 342). The variation in the rotor yaw moment for a given wind direction may be the result of different factors. For example, a greater wind speed at the wind farm 10 may result in a greater magnitude of rotor yaw moment being experienced. A greater level of turbulence in the wind may also result in a greater magnitude of rotor yaw moment being experienced; however, in the semi-wake condition, a greater level of turbulence may result in a lower yaw moment peak value because at a higher turbulence level, wake mixing may be higher (wake losses may be lower).

[0074] For a given wind direction, it can be considered that the greater the magnitude of the rotor yaw moment of the downstream turbine 12b, the greater the wake loss experienced by the downstream turbine 12b, i.e., the more severe the impact on the energy capture capability of the downstream turbine. Therefore, a severity parameter based at least in part on the rotor yaw moment can be determined as an indication of the wake loss experienced at the downstream wind turbine 12b.

[0075] Although the above description refers to monitoring parameters indicative of load imbalance - and in particular rotor yaw moment - to determine the severity or level of wake loss experienced by a (downstream) wind turbine, it will be appreciated that different parameters of the wind turbine may be considered for this purpose. For example, parameters indicative of turbulence may be used, specific frequency content in the wind turbine fore-aft acceleration (e.g. 3P content), pitch / yaw controller pitch actuation (at 1P) to correct a possible single pitch control amplitude to determine yaw moment to reduce asymmetric rotor planar moment without influence from pitch application alone, side-to-side acceleration of the wind turbine tower or nacelle, and / or blade edge or flapping moment acceleration / variation. In particular, parameters that may be used to determine rotor mean wind speed deficiency may be used.

[0076] As part of the predefined wake loss control strategy for the upstream wind turbine 12a, it may be known which wind directions are associated with different wake states for the downstream wind turbine 12b. For example, it may be known that the wind directions corresponding to points 301, 321, 341 correspond to a full wake state at the downstream turbine 12b, the wind directions corresponding to points 302, 322, 342 correspond to a left half-plane wake state at the downstream turbine 12b, the wind directions corresponding to points 303, 323, 343 correspond to a right half-plane wake state at the downstream turbine 12b, and so on.

[0077] An indication of the severity of the wake loss - for example, via the magnitude of the downstream rotor yaw moment - combined with the wind direction, allows appropriate wake control actions to be performed at the upstream turbine 12a. In one example, this allows a determination to be made that the downstream wind turbine 12b is operating at Figure 3 which of the graphs / curves 30, 32, 34 is on, and this determination may be used to select the appropriate wake loss control action to be performed on the upstream turbine 12a.

[0078] Figure 5 Elements of a controller 50 of an upstream wind turbine 12a are schematically shown. For example, the controller 50 may be located in a nacelle of the turbine 12a. The controller 50 includes one or more computer processors 501, and may include a data storage or memory 502. The controller 50 is configured to receive an input signal 504, for example via an input of the controller 50. The input signal 504 may include a signal from one or more downstream wind turbines 12b, the signal indicating the severity of wake loss experienced by the downstream turbine 12b. The controller 50 is configured to output / transmit a control signal 505 via an output of the controller 50. The output signal 505 may include one or more control signals for controlling the operation of the wind turbine 12a, for example, to control a pitch actuator to adjust the rotor blade pitch according to a determined pitch reference value, and / or to control the rotor or generator speed according to a determined speed reference.

[0079] The controller 50 described may be in the form of any suitable computing device, such as implementing one or more functional units or modules on one or more computer processors. Such functional units may be provided by suitable software running on any suitable computing substrate using conventional or custom processors and memory. The one or more functional units may use a common computing substrate (e.g., they may run on the same server) or a separate substrate, or one or both of them may themselves be distributed among multiple computing devices. A computer memory may store instructions for executing the method performed by the controller, and the processor(s) may execute the stored instructions to perform the method.

[0080] As mentioned above, an indication of the severity of the wake loss experienced by the downstream turbine 12b may be transmitted from (the controller of) the downstream turbine 12b to the upstream turbine controller 50. The indication of the severity may take different forms in different examples. Specifically, different processing steps of the overall process may be performed at different locations, including one or more of the downstream turbine controller(s), the upstream turbine controller 50, and the controller 10 of the wind farm.

[0081] In some examples, the indication of severity may be (raw) sensor data from one or more sensors of the downstream controller 12b. In an example where rotor yaw moment is used to indicate wake loss severity, blade load sensor data from one or more blade load sensors of the downstream turbine 12a may be transmitted to the upstream turbine controller 50. Based on the received sensor data the controller 50 may then determine the rotor yaw moment.

[0082] In further examples, a wake loss severity parameter, such as rotor yaw moment, may be determined at a downstream turbine controller, and the determined parameter then communicated from the downstream turbine controller to the upstream turbine controller 50 .

[0083] In still further examples, an indication of how to implement or adjust the wake loss control strategy based on the determined severity parameter may be determined at the downstream turbine controller and then communicated to the upstream turbine controller 50. In one such example, this may be applied in the form of a gain to the wake loss control strategy, or one or more wake control actions thereof, as described in more detail below.

[0084] Figure 6 The steps of a method 60 for controlling a wind turbine 12a according to an example of the present invention performed by the controller 50 are summarized. It should be understood, however, that Figure 6 One or more of the method steps shown, and / or some or all of the other steps that may in some examples form part of an overall method, may be performed remotely from the upstream wind turbine 12a, for example, by a controller of one or more of the downstream wind turbines 12b, and / or by a controller of the wind farm 10.

[0085] At step 601 of the method 60, the controller 50 receives an indication of the severity of wake loss experienced by one or more wind turbines 12b downstream of the upstream wind turbine 12a. As mentioned above, the indication may be in any suitable form. For example, the indication may be in the form of raw sensor data from one or more sensors of the downstream turbine 12b, a determined parameter (e.g., rotor yaw moment) indicating the severity of the wake loss, and / or an indication of how to implement the wake loss control strategy of the upstream turbine 12a, such as a gain.

[0086] At step 602 of method 60, the controller 50, based on the received wake loss severity indication, determines one or more wake loss control actions to control or adjust the wake generated by the upstream wind turbine 12a. In one example, the controller 50 is configured to implement a predefined wake loss control strategy at the upstream turbine 12a. This may include performing one or more wake control actions, such as wake steering, as a function of wind direction. For example, the predefined wake loss control strategy may be activated at a particular wind direction, such as a wind direction in which the wake generated by the upstream turbine 12a generates a wake loss at the downstream turbine 12b, such as Figure 3 Some or all of the wind directions corresponding between midpoints 304, 324, 344 and points 305, 325, 345. The specific control actions taken as part of the predefined strategy may be different for different wind directions in which the control strategy is active. For example, for wind directions corresponding to a full wake state (e.g., wind directions at points 301, 321, 341) relative to wind directions closer to points 304, 324, 344 or points 305, 325, 345, a greater wake steering may be implemented, such as by actuation of a greater yaw angle (i.e., a greater misalignment relative to the wind direction). However, for a given wind direction, it may also be the case that the misalignment angle of the upstream turbine varies with the severity of the wake loss (yaw moment) at the downstream turbine.

[0087] In such an example, the indication of wake loss severity may be used to determine whether to actuate a predefined wake loss control strategy, and / or whether / how to adjust the wake control actions of the predefined control strategy.

[0088] A parameter indicative of wake loss severity, such as a rotor yaw moment, may be determined, either at the controller 50, based on sensor data received from the downstream turbine 12b, or at the downstream turbine 12b, and then transmitted to the upstream turbine controller 50. In an example, a gain determination is based on a wake loss severity parameter, which is to be applied to a predefined wake loss control strategy. Again, the gain may be determined by the upstream turbine controller 50, or may be determined at the downstream turbine controller (or wind farm controller) and then communicated to the upstream turbine controller 50.

[0089] The gain may be applied in any suitable manner. In an example where the gain is used to activate / deactivate a predefined wake loss control strategy, it may be applied as follows. The wind direction may be monitored, for example, by a sensor of the upstream turbine 12a. If the wind direction is in a wind direction range that is expected / predicted to cause a wake loss of the downstream turbine 12a such that the predefined wake loss control strategy is to be activated, then in this case, a further determination is made based on the determined gain. In particular, in the illustrative example, if the determined wake loss severity parameter value is above a defined threshold, the gain may be determined to be one, and if the determined wake loss severity parameter value is below a defined threshold, the gain may be determined to be zero.

[0090] It is noted that when the severity parameter value is known, and in the case where it is the rotor yaw moment, then when the wind direction is also detected, it can be determined in which of a plurality of defined graphs / curves 30, 32, 34 the downstream turbine 12b is operating. The gain may be determined based on a predefined graph / curve describing the current downstream turbine 12b operation.

[0091] The gain may be applied to a predefined control strategy and may also be applied to a wake loss control action to be performed as part of the predefined control strategy. In this way, the predefined control strategy may be a gain-scheduled control strategy. Thus, the result of this example may be that if the gain is one, a wake control action is performed at the upstream turbine 12a according to the predefined control strategy. On the other hand, if the gain is zero, no wake loss control action is performed (i.e., the wake loss control action that would otherwise be performed is suppressed). For example, this may mean that for a given wind direction, for a severity level above a threshold, the wind turbine is yawed by a defined yaw angle (e.g., 20 degrees), while for a severity level below the threshold, no control action is taken. However, it will be appreciated that any suitable strategy may be implemented. For example, different defined angles of yaw may be achieved for different severity levels, e.g., 0 degrees for a severity level below a first threshold severity, 10 degrees for a severity between a first threshold severity and a second threshold severity (greater than the first threshold severity), and 20 degrees for a severity above the second threshold severity.

[0092] The defined threshold may correspond to a downstream wake loss severity above which it is worthwhile to adjust the upstream generated wake because it may lead to an overall increase in the power generation of the wind farm 10. On the other hand, if the wake loss severity is relatively low, i.e. below the defined threshold, the wake loss effect at the downstream turbine 12b is not sufficient to justify compromising the power generation capacity of the upstream turbine 12a, e.g. by misalignment with respect to the incoming wind direction. The defined threshold may be determined in any suitable manner, e.g. via experiments, simulations, historical data, etc.

[0093] It will be appreciated that the gain to be applied to the predefined wake loss control strategy may be determined in any suitable form. For example, the gain may include a value between zero and one, such that in certain circumstances, a reduced level of wake loss intervention may be commanded, such as a "mid-level" wake loss severity. This may establish an adjustment of the predefined wake loss control actions, rather than activating / deactivating them. For example, a lower amount of wake steering may be implemented for a relatively lower wake loss severity value, and a higher amount of wake steering may be implemented for a relatively higher wake loss severity value.

[0094] In a different example, instead of incorporating the detected downstream wake loss severity into the wake loss control strategy implemented by the controller 50 using a gain, the severity may be incorporated into part of a control loop to minimize or eliminate downstream wake losses. For example, a parameter indicative of severity may be determined, such as a rotor yaw moment, as described above, and based on the determined severity parameter, one or more wake control actions may be determined to reduce the level of wake loss experienced at the downstream turbine 12b. The determined wake loss control action is executed, and then the wake loss severity parameter value is re-determined to assess whether the control action has achieved the desired effect. An updated control action may then be determined based on the updated severity parameter value. The goal of the control loop may be to reduce the severity parameter value to zero, or to below a particular threshold level. For example, the goal of the control loop may be, in one example, to reduce the rotor yaw moment to zero. The control loop may be a proportional-integral (PI) control loop or a proportional-integral-derivative (PID) control loop.

[0095] In this way, feedback is provided to the controller 50 indicating the effect of the control action performed. If, for example, the wake loss severity at the downstream turbine is relatively low, the control action (i.e., wake deviation from wake steering) may also be relatively low, or even deactivated / disabled (not performed). Sources of varying wake loss severity may be varying atmospheric stability, varying turbulence, wind shear, temperature, terrain heating, etc.

[0096] An embodiment is therefore provided in which the wind turbine is controlled in a control loop to operate according to one or more wake loss control actions determined to reduce the severity signal to a predefined level. The predefined level may be set to zero, or to a certain threshold level.

[0097] At step 603 of method 60, the controller 50 controls the upstream wind turbine 12a based on the determined one or more wake loss control actions. Note that this may be to take no action, for example, if zero gain is applied. The control action may include controlling any suitable operating mode of the upstream wind turbine 12a to control / adjust the wake generated downstream thereof. For example, the control action may include performing yaw control to rotate the nacelle and rotor of the upstream wind turbine 12a relative to the wind turbine tower about a yaw angle to adjust the (lateral) direction of the wake generated by the upstream wind turbine 12a. The control action may also include performing tilt moment control to direct the generated wake toward the ground. The control action may also include performing collective and / or individual pitch control of the rotor blades 122 of the wind turbine in a manner that changes the generated wake as desired.

[0098] The method described above takes into account the severity of the wake loss experienced by the downstream wind turbine. In some examples, the method may take into account further information about the effect of the wake loss at the downstream turbine in order to control the upstream turbine to adjust the wake it generates. For example, in an example in which the upstream turbine controller 50 is configured to control the upstream turbine 12a according to a predefined wake loss control strategy that performs wake control actions as a function of wind direction, the operation of the downstream turbine 12b may be monitored to ensure that the assumption under which the predefined control strategy operates, i.e., which wind directions cause downstream wake losses, is actually correct. The predefined control strategy may be set under the assumption that a particular wind direction causes the downstream turbine 12b to be in a full wake state. If the monitoring sensor data from the downstream turbine 12b indicates that a full wake state is actually experienced at a different wind direction, the predefined control strategy may be adjusted to offset the wind direction in which the wake loss control action is performed by the difference between the assumed wind direction and the monitored wind direction.

[0099] The upstream turbine may receive a signal indicating a wake loss severity from each of the plurality of downstream turbines. A wake loss control action performed by the upstream controller may therefore be determined based on a combination of the received severity indications (e.g., an accumulated severity signal). The indications of severity may be combined (and optionally normalized) in any suitable manner. For example, a greater weight may be placed on a severity signal received from a downstream turbine that is closer to the upstream turbine (relative to a downstream turbine that is located farther from the upstream turbine).

[0100] Many modifications may be made to the examples described without departing from the scope of the appended claims.

[0101] In the described example, the (upstream) wind turbine to be controlled receives data indicating the wake state for a particular wind direction from a single downstream wind turbine. It will be appreciated, however, that the upstream wind turbine may receive wake state data from multiple downstream wind turbines in the wind farm. This data may be combined to determine the appropriate wake steering control of the upstream wind turbine to result in the maximum increase in energy capture efficiency of the wind farm as a whole.

Claims

1. A method for controlling a wind turbine generating a wake during operation, the wind turbine being part of a wind park comprising a plurality of wind turbines, the method comprising: receiving, from an additional wind turbine of the plurality of wind turbines downstream of the wind turbine, a signal indicative of a severity of wake losses experienced at the additional wind turbine; determining one or more wake loss control actions to adjust a wake generated by the wind turbine based on the received severity signal; and, The wind turbine operation is controlled based on the determined one or more wake loss control actions.

2. The method according to claim 1, wherein: The one or more wake loss control actions are part of a predefined wake loss control strategy for controlling the wind turbine to adjust a wake generated by the wind turbine as a function of a wind direction in the vicinity of the wind turbine.

3. The method according to claim 2, wherein: The received severity signal is a gain, wherein the predefined wake loss control strategy is a gain-scheduled control strategy, the method comprises: applying the gain to the gain-scheduled control strategy to determine the one or more wake loss control actions to be performed, and wherein controlling the wind turbine comprises controlling the wind turbine according to the gain-scheduled control strategy.

4. The method according to claim 3, wherein: The gain-scheduled control strategy comprises: if the received gain indicates that the severity of the wake loss experienced at the further wind turbine is above a predefined threshold, performing one or more wake loss control actions to adjust the wake generated by the wind turbine; and wherein if the received gain indicates that the severity of the wake loss experienced at the further wind turbine is below the predefined threshold, no wake loss control action is performed as part of the gain-scheduled control strategy.

5. A method according to any one of the preceding claims, comprising: determining, at the further wind turbine, a severity parameter indicative of a severity of wake losses experienced at the further wind turbine; And, transmitting the determined severity parameter as a severity signal to the wind turbine.

6. The method according to claim 5, wherein: The severity parameter reflects a determined insufficient wind speed at the further wind turbine.

7. The method according to claim 5, comprising: receiving sensor signals from one or more sensors of the additional wind turbine; and, based on the received sensor signals, determining an imbalance parameter indicative of a load imbalance on the rotor of the further wind turbine, wherein the severity parameter is a magnitude determined based on the determined imbalance parameter and indicative of the load imbalance.

8. The method according to claim 7, wherein: The sensor signals from the one or more sensors are blade load signals from one or more blade load sensors of rotor blades of the further wind turbine, and wherein the imbalance parameter is a yaw moment of the rotor of the further wind turbine determined based on the received blade load signals.

9. The method according to claim 7 or claim 8, wherein: The severity parameter is determined based on the magnitude of the imbalance parameter and based on the wind direction relative to a defined wind direction at which the further wind turbine experiences a full wake condition.

10. The method according to claim 9, comprising: determining a curve describing the imbalance parameter as a function of the wind direction or the nacelle yaw position based on the magnitude of the imbalance parameter and the wind direction relative to the defined wind direction; And, comparing the determined shape against a plurality of defined shapes, each defined shape being associated with a respective severity parameter, wherein the severity parameter is determined based on the comparison.

11. The method according to any one of claims 7 to 10, wherein: The imbalance parameter is normalized based on one or more operating variables; and wherein the severity parameter is determined based on the normalized imbalance parameter; optionally, wherein the one or more operating variables include one or more of the following: a defined peak amplitude of the imbalance parameter; wind speed; absolute output power; output power normalized based on rated power, a value from a database of measurements of severity of other turbines operating at different wind farms; and an estimated boost level of the further wind turbine.

12. A method according to any one of the preceding claims, wherein: Signals indicative of wake loss severity are received from two or more turbines, and wherein the one or more wake loss control actions to adjust the wake are determined based on the severity of the severity signals received from the two or more turbines.

13. A method according to any one of the preceding claims, wherein: The method comprises controlling, in a control loop, wind turbine operation according to the determined one or more wake loss control actions to reduce the severity signal to a predefined level.

14. A controller for controlling a wind turbine generating a wake during operation, the wind turbine being part of a wind park comprising a plurality of wind turbines, the controller being configured to: receiving, from an additional wind turbine of the plurality of wind turbines downstream from the wind turbine, a signal indicative of a severity of wake losses experienced at the additional wind turbine; determining one or more wake loss control actions to adjust a wake generated by the wind turbine based on the received severity signal; and Wind turbine operation is controlled based on the determined one or more wake loss control actions.

15. A control system for a wind farm according to claim 14, the control system comprising: The controller according to claim 14; and a further controller for controlling the further wind turbine, the further controller being configured to: determining a severity parameter indicative of a severity of wake losses experienced at the further wind turbine; and The determined severity parameter is transmitted as a severity signal to the controller of the wind turbine.

16. A wind turbine comprising a controller according to claim 14, or a wind farm comprising a control system according to claim 15.

Images