FIELD OF THE INVENTION

The present invention relates to devices, systems and methods for controlling wind turbines, and more particularly, to devices, systems and methods for reducing extreme loads on wind turbines. Further, the invention may be suitable for reducing fatigue.

BACKGROUND OF THE INVENTION

Wind turbines are an alternative energy technology that can provide clean power from wind. Wind turbine control systems play an important role in obtaining maximum power production and durability of wind turbines. The goal of wind turbine control systems is, inter alia, to maximize energy production, while protecting wind turbine components from extreme or fatigue loads that can cause component failure. Such loads can for example be caused by gusts. To avoid extreme loads caused by gusts, the wind speed at a pre-determined distance from wind turbines may be used in a feed-forward fashion to de-rate wind turbines, before or at the time the gust arrives at the wind turbine. The wind turbine controller may for example decrease the operational set-point of rotor speed or generator speed when an incoming gust is detected, which will cause the wind turbine to de-rate before the gust impacts the wind turbine. However, at present, it is not optimal to implement such a procedure in existing wind turbines, as the wind turbine controllers would have to be re-programmed to perform such pro- active control. Such re-programming may be hard and expensive to implement, due to physical access to the wind turbine controller, as well as complicated re- programming. Further, re-programming is in many instances be inaccessible from the wind turbine manufacturer. Additionally, new type certificates are needed when new controller implementations are performed.

Hence, improved devices or methods to reduce extreme loads on wind turbines would be advantageous, and in particular a more efficient and simple way to reduce extreme loads on wind turbines without having to re-programme existing wind turbine controllers would be advantageous. OBJECT OF THE INVENTION

An object of the present invention is to provide an alternative to the prior art.

In particular, it may be seen as a further object of the present invention to provide devices, system and methods that solve the above mentioned problems of the prior art.

SUMMARY OF THE INVENTION

Thus, the above described object and several other objects are intended to be obtained in a first aspect of the present invention by providing a signal interfering device for interfering with a speed signal provided by a speed sensor arranged to sense the speed of a drivetrain of a wind turbine comprising a wind turbine controller controlling the operational mode of the wind turbine based inter alia on a received speed signal, the signal interfering device being configured to:

identify from a wind signal of a wind sensor measuring upwind an approaching wind change, such as rapid wind change which would result in a critical load response of the wind turbine, then interfere with the speed signal by, preferably gradually, increasing the speed signal to a second speed and provide this, preferably gradually, increasing speed signal to the wind turbine controller, or

- provide to the wind turbine controller the un-interfered speed signal, if the wind signal is not indicating or comprising, such as inherently comprising information on a rapid wind change.

By "the signal interfering device being configured" is typically, meant that the implementation of the device (similarly for the method) is made by use of software running on a computer hardware platform and comprising suitable data exchange capabilities to allow data, such as data represented the wind signal and speed signal (and other signals) used by the various aspects and embodiments of the invention. Accordingly, the various aspects and embodiments of the invention may be implemented on a computer platform by use of software.

"Rapid wind change" is typically used to reference a wind change that occurs within a predetermined time period, typically evaluated in combination with a predetermined magnitude. This could be evaluated as ^an ™ ^ah > threshold where Vmean is a mean (averaged) wind speed and threshold is a pre-determined magnitude. The values of ^dv ™ ^an and threshold is typically based on empiric observations, since they depends on the actual dynamic of the wind turbine.

"Rapid wind change" may be defined as a change in wind speed which the wind turbine controller cannot counteract due to its fast/rapid time variation and might cause higher loading than the observed during normal operation. The time-scales differ from the wind turbine size, rotational speed (due to rotor speed inertia), controller type and others.

"A rapid wind change" may in accordance with embodiments disclosed herein preferably refer to "a gust", "a peak" and a "cut-off wind speed".

"A gust" or a "wind gust" preferably refers to a brief increase in the speed of wind, such as less than 20 seconds, although this is not limiting to the scope of the invention.. In some implementations of the invention, a gust cannot normally be counteracted by the wind turbine controller (normal operation) of the wind turbine.

"A peak" may preferably refer to a spike on time series.

By "wind signal is not indicating or comprising, such as inherently comprising information on a rapid wind change" is preferably meant that the approaching does not represent a wind change which would result in a critical load response of the wind turbine.

"A rated wind speed" preferably refers to the wind speed at which the wind turbine is able to generate electricity at its maximum, or rated, capacity.

"A cut-off wind speed" preferably refers to a wind speed exceeding a rated wind speed for the wind turbine, typically the wind speed above which the wind turbine should be shut-down. The cut-off wind speed is typically higher than the rated wind speed. "Critical load response" is typically used to reference an operation condition of the wind turbine such as an operation condition being outside prescribed limits, e.g. rotor speed.

"Critical load response" or "abnormally high load" is typically used to reference loading on any wind turbine component which is not comprehended within the normal operation value which is caused, for example, by a slow response on the wind turbine control when rapid/peak/gust variations impact the wind turbine rotor.

A wind sensor may be (non-limiting examples) :

a wind based system, such as a Lidar based system,

a wind sensor located on a mast, preferably placed in a proximity to the wind turbine allowing detection of wind changes sufficient early to interfere with the speed signal;

a loading sensor of an upstream turbine, e.g. if a peak on the thrust or blade flapwise moment is observed in an upstream wind turbine, that signal could be used as a trigger for the gust alleviation of the invention; in such as well as in other situations, the position of the sensor relative to the wind turbine controlled should preferably be known; However, in Lidar based systems, the wind turbine typically yaws with the system, thus the relative measuring position is typically the same;

a power sensor, such as similar to similar to a loading sensor, as one may relate power dips (above rated) to higher loading,

a cup anemometer,

a hot wire anemometer.

A wind sensor used in the present invention may preferably determine the wind speed, but as disclosed above, indirect measurements of wind speed such as loading may be used in connection with the present invention.

One main idea of the present invention may be seen as to use a signal interfering device to detect an abrupt change in wind conditions that can lead to high loads on a wind turbine, and send a modified speed signal, as the drivetrain of a wind turbine, to the wind turbine controller, when an abrupt change in wind conditions that can lead to high loads is detected. Accordingly, the sensed actual speed signal is modified and this modified speed signal is forwarded to the controller of the wind turbine thereby making the controller to believe that the actual speed is higher than it is in reality. In this way, the wind turbine will de-rate to avoid extreme loads on its components. In some embodiments, this may be seen as the wind turbine will change its operational point, by changing e.g. the rotor speed and pitch angle to a suboptimal point, thus avoiding extreme loads on its components. The signal interfering device of the present invention can be used to improve control over wind turbines without direct feedforward control and without re-programming the wind turbine controllers themselves. This is advantageous, as the device of the present invention can be used as a plug-in solution on existing wind turbines or wind turbines initially designed without feedforward control. And, advantageously, the controller is not modified, only the speed signal received by the controller is.

"Drivetrain" as used herein is typically used in a manner being ordinary to a skilled person and is preferably used to mean one or more of a rotor, rotor hub, generator, gearbox, shaft or the like.

In preferred embodiments, the wind sensor is a Light Detecting and Ranging (LIDAR) sensor. However, the wind sensor may be any type of sensor that can be used to indicate an abrupt change in wind conditions that can lead to high loads on a wind turbine.

The term "signal" herein refers to a numerical value or information that can be understood or processed to be a numerical value by the signal interfering device and the wind turbine controller. As disclosed herein, one of the signals may be a wind signal provided by a wind sensor, such as a sensor providing a wind speed. Accordingly, the wind signal may be a wind speed signal.

The term "speed sensor" as used herein may be a mechanical, electrical or optical sensor, such as an rpm sensor, configured to monitor drivetrain speed (e.g. at least one of generator speed and rotor speed) of the wind turbine. The speed sensor may be configured to sense the speed of the drivetrain directly, such as by mechanical means or optical pulses, such as laser pulses, or indirectly, such as through measured voltage and/or frequency of the generator output. The speed sensor may, for example, be located on or in a nacelle of the wind turbine.

The one or more wind conditions monitored by the wind sensor is preferably wind speed, but may also include wind acceleration and/or wind direction and/or wind shear, or any other signal that can be used as an approximation for wind speed variations, such as the pitch angle and/or flapwise blade root bending moment or the like of an upstream placed wind turbine.

The wind sensor preferably measures the one or more wind conditions at a pre- determined distance of at least 50 meters from the wind turbine, such as at least 100 meters, such as at least 150 meters, such as at least 200 meters.

In some embodiments, there is more than one wind sensor, such as two or three or four or five sensors. In such cases, the signal interfering device must be configured to process the different wind signals from the more than one sensors and be able to determine whether or not the more than one wind signals are indicative of an abrupt change in wind conditions that can lead to high loads, such as an incoming gust. Each sensor may be located on or in a nacelle of the wind turbine, elsewhere on the wind turbine, or even at a location detached from the wind turbine that is preferable for detecting wind conditions.

In some embodiments, one wind sensor may monitor one or more wind conditions at a first pre-determined distance, whereas another wind sensor may monitor one or more wind conditions at a second pre-determined distance closer to the wind turbine than the first pre-determined distance. In this way, the signal interfering device may use the data from the more than one wind sensors, to determine if an abrupt change in wind conditions that can lead to high loads on a wind turbine is approaching.

The one or more wind sensors is preferably configured to continuously and real time send information on the one or more wind conditions to the signal interfering device. For example, the one or more wind sensors may send a second signal to the signal interfering device every second, such as every miliseconds. In preferred embodiments, the signal interfering device identifies the approach of a rapid wind change which would result in a critical load response of the wind turbine, by the difference between a wind signal at a first time point and a wind signal at a second time point. If the difference between a wind signal at a first time point and a wind signal at a second time point is above a pre-determined threshold value, a rapid wind change which would result in a critical load response of the wind turbine is identified by the wind sensor.

In some preferred embodiments, the signal interfering device will preferably not react on a slow increase in wind speed, but will only react to sudden (abrupt) changes in wind conditions. For example, the difference between a first wind signal at a first time point and a second wind signal at a second time point may be above a pre-determined threshold value when the wind speed difference between the two points in time is more than 5 mph (2.24 m/s), such as more than 10 mph (4.47 m/s), such as more than 15 mph (6.71 m/s). The wind signal at the first time point and the wind signal at the second time point is received by the signal interference device within a period of less than 20 seconds, such as less than 10 seconds, preferably less than 5 seconds, such as less than 2 seconds, such as 1 second. The pre-determined threshold depends on the environment in which the wind turbine is located, the design of the wind turbine, as well as the chosen time interval in which the first and second time points are received. The skilled person will be able to determine such a threshold.

In some embodiments, the speed signal is gradually increased and subsequently gradually decreased over a pre-determined period of time, when the wind signal is indicative of an abrupt change in wind conditions that can lead to high loads. In this way, the speed signal is gradually offset when a gust is detected at a pre- determined distance from the wind turbine. The increased speed signal is never less than the original speed signal.

The pre-determined period of time in which the speed signal is gradually increased and subsequently gradually decreased may be at least 5 seconds, such as 10 seconds, such as at least 20 seconds, such as at least 30 seconds, such as at least 1 minute, such as at least 2 minutes, such as at least 5 minutes. The signal interfering device may be configured to estimate the expected arrival time of a gust at the wind turbine, based on the one or more wind signals. In this way, the speed signal is offset in a way, so at the expected arrival time of the gust, the wind turbine is de-rated to an extend such that the incoming gust will not cause extreme loads on the wind turbine. This may in some embodiment result in that the wind turbine changes its operational point, by changing the rotor speed and pitch angle to a suboptimal point, thus avoiding extreme loads on its components.

The signal interfering device may be configured to estimate the expected load on the wind turbine components from an incoming gusts and modify the speed signal accordingly, such that the processed speed signal is increased more or less depending on the expected load of the incoming gusts. In this way, in some embodiments, the extend of offset of speed signal will depend on the expected load of the incoming gusts on the wind turbine components. Thus, for higher wind speeds, the wind turbine could e.g. off-set (de-rate) more than for lower wind speeds

The modified speed signal must preferably always return to it's real value after some time after the gust has passed. In some embodiments, the speed signal will not decrease before the difference between the wind signal at the first time point and a second wind signal at a third time point is below a pre-determined value. In this way, the wind turbine will not return to normal operating conditions, before the gust has past.

In some embodiments, the signal interfering device is a micro-controller, microcomputer or a programmable logic controller (PLC).

By changing the input of a drivetrain of the wind turbine to the wind turbine controller, for example rotor speed, the turbine works in a different operational conditions which makes a de-rate of the wind turbine.

Thus, the de-rate is a consequence of changing the operational point. Preferably, the speed signal may return to the actual value of the speed sensor if the wind signal is indicative of wind speed level that corresponds to the normal wind turbine operation which will not cause extreme loading, if the gust has passed the wind turbine rotor or, in general, if the wind signal is not indicating or comprising, such as inherently comprising information of a rapid wind change.

In some embodiments, the wind turbine de-rate by altering one or more operating conditions of the wind turbine, such as generator speed or rotor speed and/or pitch angle, to decrease the wind turbine power output. However, this will depend on the turbine class.

In some preferred embodiments of the signal interfering device, the device may be configured to increase the speed signal for a period of time, such as a pre- defined period of time, in case the wind change is a gust.

In some preferred embodiments of the signal interfering device, the device may be configured to increase the speed signal for a period of time, such as a predefined period of time, in case the wind change is a peak.

In some preferred embodiments, of the signal interfering device, the device may be configured to increase the speed signal to a value larger than or equal to an overspeed of the wind turbine, in case the wind change is result in a cut-off wind- speed.

In some preferred embodiments of the signal interfering device, the device may be configured for increasing and/or decreasing the speed signal by a gradually increase and/or decrease.

In some preferred embodiments of the signal interfering device, the device may be configured for increasing and/or decreasing the speed signal by a step-wise increase and/or decrease.

In a second aspect, the invention relates to a method for controlling a wind turbine by means of a signal interfering device, preferably according to the first aspect, wherein the method preferably comprises the steps of: the speed signal and the wind signal are received by a signal interfering device from a speed sensor and a wind sensor, respectively;

the signal interfering device determines whether or not the wind signal (4) is indicative of a an approaching rapid wind change which would result in a critical load response of the wind turbine (10); wherein

if the wind signal (4) is indicative of an approaching rapid wind change which would result in a critical load response of the wind turbine , the signal interfering device (1) interfere with the speed signal (2) by, preferably gradually, increasing the speed signal (2) to a second speed and provide this, preferably gradually, increasing speed signal (2) to the wind turbine controller (8), or provide to the wind turbine controller (8) the un-interfered speed signal (2), if wind signal (4) is not indicating or comprising, such as inherently comprising information on a rapid wind change.

In a third aspect, the present invention relates to a system comprising a wind turbine, a wind turbine controller and a signal interference device according to the first aspect of the present invention.

In some embodiments, the system according to the third aspect of the present invention, further comprise a speed sensor for monitoring e.g. generator speed or rotor speed and a sensor for monitoring one or more wind conditions, wherein the signal interfering device is communicatively coupled to the speed sensor, the wind sensor and the wind turbine controller.

The first, second and third aspect of the present invention may each be combined with any of the other aspects. These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. In the same way, the different embodiments may be each be combined with any other embodiments. BRIEF DESCRIPTION OF THE FIGURES

The signal interfering device 1, the system and the method according to the present invention will now be described in more detail with regard to the accompanying figures. The figures show one way of implementing the present invention and is not to be construed as being limiting to other possible

embodiments falling within the scope of the attached claim set.

Figure 1 schematically illustrates a system according to an embodiment of the present invention,

Figure 2 schematically illustrates a method of controlling a wind turbine using a signal interfering device, according to an embodiment of the present invention,

Figures 3 -5 each schematically illustrates three graphs showing when a gust is detected (3A, 4A), how the speed signal is modified via an offset in response to detecting of a gust (3B) and when the gust arrives at the wind turbine (3C, 4C), according to different embodiments of the present invention,

Figure 6 schematically illustrates two graphs showing when a wind peak (a wind velocity exceeding a rated wind speed for a period of time) is detected, and how the speed signal is modified in response to the detected wind peak, such situation is herein referred to as peak shaving;

Figure 7 schematically illustrates two graphs showing when a cut-off wind speed (a wind speed, typically prevailing, exceeding a cut-off wind speed) is detected, and how the speed signal is modified in response to the detected cut-off wind speed resulting in shut-down of the wind turbine. In the figure, a Ati is shown. The size of this time difference should preferably be made in accordance with the normal shut-down system of the wind turbine, which shut-down procedure is in many embodiments invoked by the wind turbine controller received the interfered speed signal.

DETAILED DESCRIPTION OF AN EMBODIMENT

Reference is made to figure, illustrating a system according to an embodiment of the present invention. The system of the present invention comprise a signal interference device 1 which is connected to a wind turbine 10 and its wind turbine controller 8, both located in the nacelle 11. The system further comprises, a speed sensor 3 for sensing the speed of a drivetrain of the wind turbine 10 and a wind sensor 5 for monitoring one or more wind conditions. Although the speed sensor 3 and wind sensor 5 are associated with the nacelle 11 in this drawing, in

alternative embodiments of the present invention, one or more of these sensors may be located elsewhere in the system. The wind turbine controller 8 is configured to control the wind turbine 10.

The signal interfering device 1 is configured for interfering with a speed signal 2 being send to the wind turbine controller 8 from a speed sensor 3 and for modifying the speed signal 2 when a wind signal 4 from a wind sensor 5 measure an upwind approaching rapid wind change which would result in a critical load response of the wind turbine 10. As will become apparent from the following, the invention may find use a different modes of operation including gust alleviation, peak shaving and shut-down.

The speed sensor 3 may for example be configured to monitor rotor speed of the wind turbine 10. The wind sensor 5 is configured to monitor one or more wind conditions, such as wind speed and/or wind direction, at a predetermined distance from the wind turbine 10. The wind sensor 5 may be a Light Detecting and Ranging (LIDAR) sensor or any other radar capable of measuring one or more wind conditions at a predetermined distance from a wind turbine 10. Thus, the speed signal 2 comprise information on the speed of a drivetrain of the wind turbine 10; and the wind signal 4 comprise information on the one or more wind conditions at a predetermined distance from the wind turbine 10, such that an upwind approaching rapid wind change which would result in a critical load response of the wind turbine 10, can be detected by the wind sensor 5.

The signal interfering device 1 may be a programmable logic controller (PLC).

The signal interfering device 1 comprise an input module 6, a processor 9 and an output module 7. The input module 6 is configured for receiving the speed signal 2 from the speed sensor 3 and for receiving the wind signal 4 from the wind sensor 5. The processor 9 is communicatively coupled to the input module and configured for determining whether or not the wind signal 4 is indicative of an upwind approaching rapid wind change which would result in a critical load response of the wind turbine 10. When the wind signal 4 is indicative of an upwind

approaching rapid wind change which would result in a critical load response of the wind turbine 10, the speed signal 2 is modified, such that a the speed signal 2 is gradually increased over a period of time. Further, the speed signal reverts to its normal value (un-interfered) after the gust has passed the wind turbine. When the wind signal 4 does not indicate an upwind approaching rapid wind change which would result in a critical load response of the wind turbine 10, the speed signal 2 is not modified.

The signal interfering device 1 further comprise an output module 7

communicatively coupled to the processor 9, configured for forwarding the modified speed signal 2 to the wind turbine controller 8.

Reference is made to figure 2, illustrating a method for controlling a wind turbine 10 by means of the signal interfering device 1 according to an embodiment of the present invention. It is noted that the reference in figure 2 in the processer 9 to a "gust" is an example and that the method is applicable the other wind phenome, e.g. as disclosed herein as peak and cut-off wind speed. In a first step, the speed signal 2 and a wind signal 4 are send to the signal interfering device 1 from the speed sensor 3 and the wind sensor 5, respectively.

Next, the signal interfering device 1 determines whether or not the received wind signal 4 is indicative of an upwind approaching rapid wind change which would result in a critical load response of the wind turbine 10, e.g. by the difference between a wind signal at a first time point 4a and a wind signal at a second time point 4b is above a pre-determined threshold value. The wind signal at the first time point 4a and the wind signal at the second time point 4b is preferably received by the signal interference device 1 within a period of less than 20 seconds, such as less than 10 seconds, preferably less than 5 seconds, such as less than 2 seconds. If the signal interfering device 1 determines that the wind signal 4 is indicative of an upwind approaching rapid wind change which would result in a critical load response of the wind turbine 10, the signal interfering device 1 modify the speed signal 2, such that the speed signal 2 is increased over a period of time. Further, the speed signal reverts to its normal value (un-interfered) after the gust has passed the wind turbine. If the wind signal 4 is not indicative of an upwind approaching rapid wind change which would result in a critical load response of the wind turbine 10, the first signal 2 is not modified.

The signal interfering device 1 forwards the speed signal 2 to the wind turbine controller 8 and as a result the wind turbine 10 de-rates if the speed signal 2 increase above a predetermined set point. Typically, this results in that the wind turbine de-rates if the speed signal is larger than the measured signal. The wind turbine controller 8 preferably "de-rate" by altering one or more operating conditions of the wind turbine 10, such as generator speed or rotor speed and/or pitch angle, to decrease the wind turbine power output. Since the input to the wind turbine controller is an interfered signal, the speed signal received by the controller is an artificial speed signal, and the wind turbines "does not know it is" de-rating. By changing the rotor speed the wind turbine will, in some

embodiments, pitch out and apply torque which is similar to "de-rate" but it is not.

It is noted that although figure 2 illustrates the speed sensor 3 and the wind sensor 5 as located in the nacelle 11, this is considered a non-limiting example. In other embodiments, instead of the nacelle 11, the box in fig. 2 labelled 11, may be the wind turbine 10 as such

Reference is made to figure 3, illustrating an embodiment where the speed signal 2 increases when the wind signal 4 is indicative of an upwind approaching rapid wind change which would result in a critical load response of the wind turbine 10.

Figure 3A shows the wind signal 4, which in this case is information on wind speed, at a pre-determined distance from the wind turbine 10, over a period of time. Figure 3B illustrates the value of the speed signal (the off-set) over a period of time. As illustrated the speed signal is un-interfered until point 4a (see fig. 3A) where after the speed signal is off-set by a ramp-up and ramp-down. . Figure 3C illustrates wind speed at the wind turbine site, typically the position of the wind turbine in question.

In figure 3, the speed signal 2 is gradually increased and subsequently gradually decreased over a period of 30 seconds, when the wind signal 4 indicates an upwind approaching rapid wind change which would result in a critical load response of the wind turbine 10. Preferably, the modified speed signal 2 is never less than the original speed signal 2. As can be seen from the figure, the rapid wind which would result in a critical load response of the wind turbine 10 arrives at the wind turbine site, at the time the speed signal 2 have increased to a maximum. When the speed signal 2 increases, the wind turbine will de-rate and thus extreme loading on the wind turbine components are avoided. Again, since the input to the wind turbine controller is an interfered signal, the speed signal received by the controller is an artificial speed signal, and the wind turbines controller "does not know it is" is de-rating. By changing the rotor speed the wind turbine will, in some embodiments, pitch out and apply torque which is similar to "de-rate" but it is not. Figure 3B may actually be viewed in two different ways, as showing the offset applied to the original speed signal (OSS), or the interfered speed signal (ISS), as:

Where constant (t) is an increasing (or decreasing) constant with respect to time.

Reference is made to figure 4. Figure 4A shows the wind speed "Gust impact time" is shown as impact at 150 seconds on the wind turbine at the wind turbine. Figure 4B "LIDAR detection" shows, that the LIDAR system detects the gust at 130 seconds lasting until approximately 170 seconds. The system detects the gust around 130 seconds. The "Derate Flag" signal provides the information on how long the rotor speed signal will be changed. The duration of that (in this case around 40 seconds) is chosen based on wind turbine, site, gust in this example. Thus, 4B represents the time the rotor speed signal is going to be modified, thus altering the original rotor speed signal.

Figure 4C: "b" shows that the speed signal is ramped up by up till 1.1 and ramped down to 1 during the period [130sec; 170sec]. As shown in figure 4Bs y-axis, the curve shows the offset value. The Offset value of 1.1 is also an example. The value could be more or less (but, in principle, preferably higher than 1). It is also possible to see in Figure 4C that while the De-rate Flag is not active, the offset value is 1 meaning that the original signal is not modified.

Reference is made to figure 5, which is similar to figure 3, except that the speed signal 2 in figure 4, is not decreased (at least for the time period shown). In some embodiments, the speed signal 2 increases over a pre-determined period of time, and does not decrease again, before the difference between the wind signal at the first time point 4a and a wind signal at a third time point 4c is below a pre- determined value (not shown in fig. 4 since it is further in time and since the offset plot (4B) has not started to decrease to its original value). In this way, the wind turbine only return to normal operating conditions, when the gust have past.

Embodiments pertaining to peak shaving

Reference is made to figure 6 which schematically illustrates two graphs showing when a wind peak is detected and how the speed signal 2 is modified in response to the detected wind peak in an embodiments which may be termed "peak shaving". A wind peak is as illustrated typically considered to be a wind velocity exceeding a rated wind speed for a period of time, where the period of time during which the speed signal is offset being determined e.g. empirically, that is in relation to figure 6 between [t2;t3]. The rated wind speed is typically a wind speed below cut-off wind speed at which the wind turbine is be shut-down. It is noted, that the figure showing a situation in which the wind speed after the peak is above the rated wind speed. The purpose of offsetting (interfering) of speed signal is to smooth out the transition region between partial and full load of the wind turbine. As illustrated, peak shaving is preferably only supposed to act between a partial and full load region (around rated wind speed) where the mean thrust level is higher. By offsetting (interfering) the rotor speed the wind turbine may start pitching before the highest (mean) thrust peak occurs. The offset returns to 1, that is no interference with the speed signal, once the controller is working on the full load region and then the normal controller kicks-in using as an input the measured/real rotor speed signal, that is not interfered speed signal. Please note in fig. 6 that the value of the offset= l is illustrated although the scale of the y-axis "Offset of Speed signal" is illustrative only, so that the offset= l refers to un-interfered speed signal and offset 1 refers to interfered speed signal.

It is noted that as the wind speed is determined upstream of the wind turbine (in the Lidar measuring plane) the duration of the peak can in some instances not be determined a priori and the approaching wind may be categorised as a gust or peak as long as the wind signal is lower than the cut-off wind speed and above the rated wind speed. If the wind signal exceed the cut-off wind signal, the procedure disclosed below as "shut-down" is invoked.

Further, if wind speed does not decrease relatively fast after peaking, the wind speed may be categorised as a gust and the procedure disclosed in relation to figures 5 may be invoked.

The procedure referred to as peak shaving in figure 6 may be disclosed as follows. The wind sensor 5, such as a Lidar system measures upstream (in the Lidar measuring plane) of the wind turbine the speed of the approaching wind. If the signal interfering device 1 detects that the wind speed exceeds the rated wind speed, the speed signal 2 is increased by the interfering device 1 and this changed speed signal 2 goes into the wind turbine controller 8. In figure 6, this occurs at t=t2. The Lidar system continues to measure the speed of the

approaching wind and feeds this into the interfering device 1. After some time, the wind turbine controller has set the wind turbine settings e.g. pitch to operate at the higher wind and the interference with the speed signal is terminated (offset returns to 1) and the actual measured speed signal is fed un-interfered to the wind turbine controller. This occurs at t=t3. Since the actual values of tå and depends on parameters such as inertia of the wind turbine and/or response speed of the wind turbine controller, the actual values of tå and may advantageously be determined empirically.

It is noted, that the changes made to the speed signal in figure 6 are illustrated as abrupt changes (a step function). However, a gradually increase and/or decrease in speed signal as disclosed in connection with figures 3-5 may be applied as alternative to the abrupt change(s). Further, while the above implementation is based on a comparison on whether a measured wind speed is above or below the rated wind speed, the procedure disclosed in relation to figures 3-5 may be applied (or vice versa). That is, considering the difference between a wind signal at a first time point (4a) and a wind signal at a second time point (4b) and invoking the interference of the speed signal if this difference is above the a certain threshold and/or stop interfering if below the rated wind speed or the wind turbine controller has adopted the wind turbine to the new wind regime. In other words a peak shaving procedure is typically applied in a situation were: wsp( wsp(2)

> threshold.

Where threshold typically is defined empirically. Embodiments pertaining to shut-down

Reference is made to figure 7 schematically illustrating two graphs showing when a cut-off wind speed (a wind speed, typically prevailing, exceeding a cut-off wind speed) is detected, and how the speed signal is modified in response to the detected cut-off wind speed resulting in shut-down of the wind turbine.

Please note in fig. 7 that the value of the offset= l is illustrated although the scale of the y-axis "Offset of Speed signal" is illustrative only, so that the offset= l refers to un-interfered speed signal and offset 1 refers to interfered speed signal . However, the offset is preferably chosen so as to produce an interfered speed signal being higher than the rated rotor speed as illustrated by "Overspeed".

As illustrated, the speed of the approaching wind is measured, e.g. by a Lidar system, in the Lidar measuring plane. The interfering device detects that at t=ti the wind speed exceed the cut-off wind speed (being predetermined). As shut- down normally is considered a drastic step to be carried out preferably only when needed and typically to be in accordance with the wind turbine controller definition of "shut-down", the interference of the speed signal is as illustrated typically delayed an amount of Ati. This delay is advantageously used to make it more certain that the approaching wind exceeding the cut-off speed is not a momentary spike not requiring cut-off. However, the invention is not limited to such delay as the interfering may be invoked as soon as approaching wind speed exceed the cut-off speed (at t=ti).

If a delay is used, the duration of the delay may be calculated or estimated in the following manner. Estimated time of arrival, ETA, for the wind speed at which the turbine is to shut-down is approximated by:

AS

^ETA =— ^v wf

Where AS is the distance upwind the wind turbine to the position where the wind speed is measured and v _wf is the velocity of the wind front. Preferably, the ETA should be greater than a response time for the wind turbine, allowing the wind turbine to shut-down before the wind front hits the wind turbine. Thus, the time At _± at which interference of the speed signal (off-set) is introduced may be determined as difference between ETA and response time t _r e.g. as:

At _± = ETA t _r

The velocity of the wind front may as a first approximation be set to the measured velocity of the wind, and the response time may be determined or estimated by empirical research.

In the embodiment of figure 7, the off-set (interference) is show as a step function, but the off-set could be left constant including a larger off-set (sufficient to trigger shut-down) after its onset. The proposed step function wherein the speed signal is raised to the magnitude of overspeed has shown to be highly effective as it signals to the controller that the rotor is running at an overspeed rpm (although in reality, it does not). Once the wind turbine has acted upon the interfered speed signal, by the wind turbine controlling receiving the interfered (offset) speed signal, the interference with the speed signal is ended, as illustrated in figure 7 as the offset of speed signal returns to 1. At that point in time, the wind turbine controller kicks-in using as an input the measured/real rotor speed signal, that is no interference of the speed signal.

That the interfered speed signal being offset to a value higher than the rated rotor speed may by many wind turbine controllers be evaluated as a faulty operation condition and the offset produced by the present invention in such situtations may be perceived as generating an artificial fault signal forcing the wind turbine controller to shut-down.

As disclosed also in relation to figure 6, the off-set of the speed signal is shown in figure 7 as an abrupt change (step-function). However, a gradually increase or decrease of the interference may be applied (e.g. as disclosed in relation to figures 3-5).

Depending on how the wind turbine is shut-down, the non-interfered speed signal after the shut-down has ended may be zero (fully braked rotor during shut-down) or an value different from zero (if the rotor is not fully braked during shut-down).

Further, while the above implementation is based on a comparison on whether a measured wind speed is above or below the cut-off wind speed, the procedure disclosed in relation to figures 3-5 may be applied (or vice versa). That is, considering the difference between a wind signal at a first time point (4a) and a wind signal at a second time point (4b) and invoking the interference of the speed signal if this difference is above the cut-off wind speed (or stop interfering if below the cut-off wind speed).

Instantaneous and averaged wind speed

Wind speed is often fluctuating with short-living (e.g. duration less than 1-2 seconds) high wind occurrences. Further, fluctuations may appear on further time scales, such as the time scale of turbulence, time scales of eddies (e.g. from upwind wind turbines) and time scales of duration of gusts, peaks, high wind etc.

While the invention disclosed may operate based on measured instantaneous wind speed, it may be beneficial to smoothen wind speed measurements e.g. by applying a moving average:

Wherein n is the latest n wind measurements. This may be implemented in the above by instead of using the instantaneous wind speed measured in the determination as to whether or not to interfere the speed signal, the mean velocity Vmean is used. The value of n and the time tn-ti are typically determined empirically. It is noted that e.g. a Lidar measurement system inherently performs some averaging, whereby the instantaneous wind speed provided by the Lidar system is inherently averaged by the Lidar system, and the above averaging may not needed.

On magnitude and duration of the off-set

By applying an offset magnitude higher than 1 (note that in accordance with the presented preferred form, the offset is typically applied in the form of speed signal multiplied by offset), it may be ensured that the torque and pitch setting by the wind turbine controller by the modified (higher) signals may be in accordance with the objective of the invention, preferably meaning a higher torque which will slow down the wind turbine and a higher pitch (depend on reference system) which will make the wind turbine blade more transparent to the wind.

The duration, offset and ramp-up/down sequence of interference of the rotor speed signal should be in accordance with the wind turbine dynamics such as, the alteration of the measured signal does not inflict higher loading than the

phenomena it is trying to avoid (i.e. gust). The time start (to) may depend on the preview time given by the wind sensor (or equivalent). The (peak/higher) magnitude of the offset will depend on the source of the detected phenomena as the available preview time, as the measured rotor speed signal at to.

The magnitude of the rotor speed signal, can vary from 1, where the signal is unaltered to a pulse value (or infinite) which will produce the wind turbine to shut- down.

ITEMIZED LIST OF PREFERRED EMBODIMENTS

1. A signal interfering device (1) for interfering with a speed signal (2) provided by a speed sensor (3) arranged to sense the speed of a drivetrain of a wind turbine (10) comprising a wind turbine controller (8) controlling the operational mode of the wind turbine (10) based inter alia on a received speed signal (2), the signal interfering device (1) being configured to:

identify from a signal (4) of a sensor (5) measuring upwind an approaching rapid wind change which would result in a critical load response of the wind turbine (10), then interfere with the speed signal (2) by gradually increasing speed signal (2) to a second speed and provide this gradually increasing speed signal (2) to the wind turbine controller (8), or

provide to the wind turbine controller (8) the un-interfered speed signal (2), if signal (4) is not indicating or inherently comprising information on a rapid wind change.

2. A signal interfering device (1) according to item 1, wherein the sensor (5) is a Light Detecting and Ranging (LIDAR) sensor. 3. A signal interfering device (1) according to any of the preceding items, wherein the speed signal (2) is gradually increased to a second speed when the difference between a signal at a first time point (4a) and a signal at a second time point (4b) is above a pre-determined threshold value. 4. A signal interfering device (1) according to any of the preceding items, wherein the speed signal (2) is gradually increased and subsequently gradually decreased over a pre-determined period of time, when the signal (4) of a sensor (5) identify an approaching rapid wind change which would result in a critical load response of the wind turbine .

5. A signal interfering device (1) according to any of items 3-4, wherein the speed signal (2) is not decreased before the difference between the signal at the first time point (4a) and a second signal at a third time point (4c) is below a pre- determined value. 6. A signal interfering device (1) according to any of the preceding items, wherein the signal interfering device (1) is a programmable logic controller (PLC).

7. A method for controlling a wind turbine (10) by means of a signal interfering device (1) according to any of items 1-6, wherein the method comprise the steps of:

the speed signal (2) and the signal (4) are received by the signal interfering device (1) from the speed sensor (3) and the sensor (5), respectively;

the signal interfering device (1) determines whether or not the signal (4) is indicative of a an approaching rapid wind change which would result in a critical load response of the wind turbine (10); wherein

if the signal (4) is indicative of an approaching rapid wind change which would result in a critical load response of the wind turbine , the signal interfering device (1) interfere with the speed signal (2) by gradually increasing speed signal (2) to a second speed and provide this gradually increasing speed signal (2) to the wind turbine controller (8), or provide to the wind turbine controller (8) the un- interfered speed signal (2), if signal (4) is not indicating or inherently comprising information on a rapid wind change

8. A system comprising a wind turbine (10), a wind turbine controller (8) and a signal interference device (1) according to any of items 1-7.

9. A system according to item 10, wherein the system further comprise the speed sensor (2)and the sensor (5)s, wherein the signal interfering device (1) is communicatively coupled to the speed sensor (2), the sensor (5) and the wind turbine controller (8).

Although the present invention has been described in connection with the specified embodiments, it should not be construed as being in any way limited to the presented examples. The scope of the present invention is set out by the accompanying claim set. In the context of the claims, the terms "comprising" or "comprises" do not exclude other possible elements or steps. Also, the mentioning of references such as "a" or "an" etc. should not be construed as excluding a plurality. The use of reference signs in the claims with respect to elements indicated in the figures shall also not be construed as limiting the scope of the invention. Furthermore, individual features mentioned in different claims, may possibly be advantageously combined, and the mentioning of these features in different claims does not exclude that a combination of features is not possible and advantageous.

LIST OF REFERENCE SYMBOLS USED

1 signal interfering device

2 speed signal

3 speed sensor

4 wind signal

5 wind sensor

6 input module

7 output module

8 wind turbine controller

9 processor

10 wind turbine

11 Nacelle