Wind turbine and method for noise reduction for a wind turbine

Abstract

Wind turbine comprising a tower (2) supporting a nacelle (5) and a rotor (3) having a plurality of rotor blades (4) and an active noise reduction device (7), wherein the active noise reduction device (7) comprises at least one actuator (8), at least one unsteady pressure sensor (9) adapted to generate an output signal corresponding to a turbulent flow condition during operation of the rotor blades (4), at least one noise sensor (10) adapted to generate an output signal corresponding to a noise generated by the rotor blades (4) at the position of the noise sensor (10), wherein the unsteady pressure sensor (9) and the actuator (8) are arranged on at least one of the rotor blades (4) and the noise sensor (10) is arranged at the nacelle (5) and / or at the tower (2), wherein the control unit (11) is adapted to control the actuator (8) depending on the output signals of the unsteady pressure sensor (9) and the noise sensor (10) to emit an anti-noise signal at least partially reducing the noise generated by the rotor blades (4).

Description

Technical Field

[0001] This invention relates to a wind turbine comprising a tower supporting a nacelle and a rotor, and an active noise reduction device. The rotor has multiple rotor blades. The active noise reduction device includes at least one actuator and at least one unstable pressure sensor adapted to generate an output signal corresponding to turbulence conditions during rotor blade operation. Furthermore, this invention relates to a noise reduction method for a wind turbine. Background Technology

[0002] Trailing-edge noise emitted from the rotor blades during wind turbine operation is the primary noise generation mechanism in modern industrial-grade wind turbines. The noise generated by a wind turbine during operation determines the area where the wind turbine can be installed, or consequently, how it can be operated to comply with noise restrictions in the surrounding environment.

[0003] Excessive noise generated during the operation of wind turbines may necessitate operating them in a reduced mode, sacrificing power generation to limit noise emissions. Therefore, the noise emissions of wind turbines directly impact their economic viability. Consequently, noise reduction technologies, particularly for onshore wind turbines, are needed to limit the noise emitted by wind turbines during operation.

[0004] One method for limiting noise emitted by wind turbines is to use serrations at the trailing edge of the rotor blades to passively reduce noise emanating from that area. However, the maximum noise reduction achievable using passive noise reduction devices appears to be limited, making active noise cancellation systems, additionally or alternatively, suitable for actively reducing noise emitted from the rotor blades during operation. Active noise cancellation systems typically use sensors to measure surface pressure fluctuations on the rotor blade surfaces, feeding these fluctuations into an anti-noise controller that generates a counter-acoustic signal to cancel out trailing edge noise in the far field or the environment surrounding the wind turbine.

[0005] EP3249216A1 describes a rotor blade with a noise reduction device. The rotor blade includes at least one sensor for detecting the flow characteristics of fluid flowing substantially from the leading edge to the trailing edge of the rotor blade. The rotor blade also includes at least one actuator for generating an anti-noise signal to at least partially cancel edge noise caused by the flow of the rotor blade. Summary of the Invention

[0006] The object of the present invention is to provide an improved wind turbine and an improved method for using a wind turbine, which in particular allows for improved noise reduction during the operation of the wind turbine.

[0007] According to the invention, this objective is achieved by a wind turbine as originally described, wherein the active noise reduction device further includes at least one noise sensor and a control unit, the at least one noise sensor being adapted to generate an output signal corresponding to the noise generated by the rotor blades at the location of the noise sensor, wherein an unstable pressure sensor and an actuator are arranged on at least one of the rotor blades, and the noise sensor is arranged at the nacelle and / or the tower, wherein the control unit is adapted to control the actuator based on the output signals of the unstable pressure sensor and the noise sensor to generate an anti-noise signal that at least partially reduces the noise generated by the rotor blades.

[0008] The tower supports the nacelle to which the rotor is mounted, and may include one or more tower segments arranged between the ground and the nacelle. The rotor includes multiple rotor blades, such as three rotor blades, wherein at least one actuator of the active noise reduction device and at least one instability pressure sensor are mounted to at least one of the rotor blades. In particular, the active noise reduction device may include multiple actuators and multiple instability pressure sensors, which are specifically arranged and / or fixed to all rotor blades of the wind turbine to allow for noise reduction of the trailing edge noise of all rotor blades of the wind turbine.

[0009] At least one unstable pressure sensor can be mounted particularly close to the trailing edge of the rotor blades, since most of the noise emitted by the rotor blades is generated at the trailing edge. By detecting pressure fluctuations at the rotor blades, especially at the trailing edge, using at least one unstable pressure sensor, the sensor can generate an output signal corresponding to turbulence conditions during rotor blade operation. This unstable pressure condition measured by the unstable pressure sensor corresponds to near-field pressure fluctuations at the trailing edge, and particularly close to the surface of the rotor blades. These near-field pressure fluctuations are related to the far-field noise generation of the rotor blades.

[0010] Based on the output signal of the unstable pressure sensor, the control unit can determine an anti-noise signal corresponding to the turbulence conditions and thus the noise at the rotor blades. The determined anti-noise signal can be emitted from at least one actuator of the active noise reduction device to reduce noise in the far field (and therefore near or around the wind turbine).

[0011] To improve noise reduction in the far field of a wind turbine, an active noise cancellation device also includes at least one noise sensor adapted to generate an output signal corresponding to the noise generated by the rotor blades at the sensor's location. Unlike the unstable pressure sensor and actuator located at at least one rotor blade of the wind turbine, the noise sensor of the active noise cancellation device is located, for example, at the tower, and thus at a distance from the rotor blades or other components of the active noise cancellation device. This allows the use of at least one noise sensor to measure a noise signal corresponding to the noise generated by two rotor blades, which can be measured at the wind turbine tower and therefore also in the far field of the rotor blades. Knowledge of the noise detectable at the wind turbine tower location helps create an anti-noise signal that also reduces noise in the far field away from the wind turbine, such as in residential areas near the wind turbine.

[0012] Noise sensors can be positioned at the nacelle, particularly at the rear of the nacelle and / or on the wind turbine tower. The advantage of positioning the noise sensor, or at least a portion thereof, on the tower is improved blade-to-blade noise isolation, as noise measured at the sensor's location can be correlated with the noise generated by the rotor blades closest to the tower. Furthermore, the spanwise source of the noise can be located on the tower without using complex microphone array techniques.

[0013] For example, if the noise measured by at least one noise sensor still exceeds a predetermined threshold, the noise sensor allows for correction of the emitted noise reduction signal. The advantage of using at least one noise sensor is that the noise reduction function of the noise reduction device can be controlled and / or corrected during wind turbine operation. Effects such as sensor degradation and / or changes in trailing edge geometry over time, potentially stemming from damage or wear of the trailing edge, can be considered and compensated for. Furthermore, by using at least one noise sensor at the wind turbine tower to generate the noise reduction signal, changes in the flow conditions of the fluid (especially air or wind) causing turbulence at the rotor blades can be taken into account. The robustness and noise reduction capability of the active noise reduction device are improved by using at least one noise sensor. Therefore, the reliability of the active noise reduction device and the amount of noise that can be reduced using it are advantageously increased.

[0014] In a preferred embodiment of the invention, the unstable pressure sensor and actuator are arranged in at least one arrangement portion of the rotor blade housing, wherein the noise sensor is arranged in an opposing portion on the outer side of the nacelle and / or the outer side of the tower, the opposing portion being at least partially opposite to the arrangement portion of the rotor blade when the rotor blade is aligned in a downward direction parallel to the tower. As the wind turbine rotor rotates, the rotor blades are aligned parallel to the tower at specific points in time, particularly when the rotor blades are pointing vertically downward from the rotor hub.

[0015] In a position where the rotor blades are parallel to the tower, there is a relative portion on the nacelle and / or tower opposite the arrangement of the rotor blade housing, wherein at least one noise sensor of the active noise cancellation device is arranged in this relative portion. In other words, at least one noise sensor of the active noise cancellation device is arranged in the relative portion at the nacelle and / or tower, wherein this relative portion and the arrangement on the rotor blade housing are at the same distance from the ground on which the wind turbine is located.

[0016] The advantage of arranging at least one noise sensor in the relative section is that the spatial distance between at least one noise sensor and, in particular, at least one unstable pressure sensor and / or at least one actuator of the wind turbine is relatively small, especially near these time points when the rotor blades are parallel or substantially parallel to the tower alignment.

[0017] Preferably, the active noise reduction device includes multiple noise sensors arranged in a ring around the outer circumference of the tower. The tower may include, for example, multiple tower segments supporting a rotor and / or nacelle located at the top of the tower. These tower segments may be, for example, cylindrical in shape, allowing the noise reduction sensors to be arranged in a ring around the outer circumference of one of the tower segments or tower segments.

[0018] Preferably, the unstable pressure sensor and / or actuator are arranged in the tip region of the rotor blade, and / or the unstable pressure sensor is arranged on the trailing edge of the rotor blade. Noise generated during wind turbine operation is greatest in the tip region of the wind turbine blade, making it advantageous, in particular, to position the unstable pressure sensor and actuator in the tip region of the rotor blade. The tip region of the rotor blade can be, particularly, the outermost 10% to 30% of the rotor blade length, especially the outermost 20%. Since the primary source of noise generation is the trailing edge of the rotor blade, the unstable pressure sensor can be particularly arranged on the trailing edge of the rotor blade.

[0019] In a preferred embodiment of the invention, the actuator includes a loudspeaker, and / or the unstable pressure sensor includes a pressure transducer, particularly a surface pressure transducer, and / or a microphone, and / or the noise sensor includes a pressure transducer, particularly a microphone. The actuator may, in particular, be an active loudspeaker including an amplifier, such that an anti-noise signal generated by the control unit can be amplified by the actuator before being emitted.

[0020] Preferably, the control unit is adapted to use at least one actuator transfer function to determine the noise immunity signal, the at least one actuator transfer function describing the relationship between the input signal to at least one actuator and the far-field noise.

[0021] The at least one actuator transfer function between at least one actuator and at least one far-field noise sensor can be measured directly, for example, by generating a signal through at least one actuator including, for example, a loudspeaker, and simultaneously measuring the received signal at at least one far-field noise sensor. The actuator transfer function can be expressed as an impulse response function in the discrete time domain.

[0022] An actuator transfer function describes the relationship between the input signal of at least one actuator and the far-field noise generated in a region of interest near a wind turbine, where the noise emitted by the wind turbine should be reduced. The actuator transfer function can be stored, for example, in the control unit of an active noise cancellation device. The actuator transfer function allows the determination of the noise generated by the actuator based on the input signal (e.g., an electrical signal) sent to the actuator.

[0023] Preferably, the control unit is adapted to apply at least one filter function to the output signal of the unstable pressure sensor in order to determine the input signal for the actuator to emit an anti-noise signal, wherein the control unit is adapted to adjust the filter function according to the output signal of the noise sensor, particularly in adaptive feedforward control.

[0024] Specifically, the control unit is implemented to apply a filter to the unstable pressure signal to determine the input signal of the actuator, causing the actuator to emit an anti-noise signal, particularly from at least one speaker of the actuator. The control unit is also implemented to adjust the filter based on signals from at least one far-field noise sensor, particularly in an adaptive feedforward controller. The filter function applied to the output signal from at least one unstable surface pressure sensor is specifically related to the transfer function between the unstable surface pressure and the far-field noise, although in, for example, an adaptive feedforward controller, it is not necessary to have direct knowledge of the transfer function between the unstable surface pressure and the far-field noise.

[0025] The filter function correlates the output signal from at least one unstable pressure sensor with a signal sent to at least one actuator. For example, the filter function can be implemented as a weighted linear combination of stored signal samples from at least one unstable pressure sensor. This can be mathematically represented as follows:

[0026] y m (n)=x(n)w m (n).

[0027] Here, y(n) is the signal sent to the loudspeaker in discrete time n, m is an index between 1 and M, where M is the number of loudspeakers, x is a horizontal vector containing previously recorded samples of signals from at least one unstable pressure sensor, and w is a filter function represented as a vertical vector of length x containing weighted scalars.

[0028] For example, the filter function W can be calculated using the measured relationship between the unstable pressure described by the output signal of at least one unstable pressure sensor and the far-field noise at one or more far-field noise sensors. The filter function depends in particular on the relationship between the unstable pressure signal and the far-field sound, although, due to the use of an adaptive filter function, this relationship need not be calculated in the form of a transfer function at any point.

[0029] The output signal of the noise sensor can be used to adjust the filter function w, thus allowing the filter function to be corrected based on the noise measured at the location of the noise sensor. This adjustment of the filter function is called adaptive feedforward control and can be used particularly when the intention is to reduce noise at locations further away from the wind turbine. The application of the filter function in the above form is suitable for the control of linear systems. The control system can be enhanced or modified to control the nonlinear relationship between surface pressure and far-field noise, such as frequency shifts or harmonic generation, by using, for example, neural networks or other machine learning algorithms.

[0030] Preferably, the control unit includes or is connected to a rotor orientation sensor, particularly a gravity sensor arranged on at least one rotor blade, and / or a rotary encoder coupled to the rotor, to determine the relative position between the rotor blade and the tower during wind turbine operation. The control unit is adapted to adjust a filter function for periods when at least one rotor blade is at least temporarily in a downward or nearly downward direction parallel to the tower. This allows for consideration of the rotational movement of the rotor blade relative to a stationary noise sensor or the rotational movement of an unstable pressure sensor and actuator, in order to adjust the active noise reduction device.

[0031] To reduce the impact of the rotational motion of the rotor blades, the adjustment of the filter function allows for the use of predominantly static and / or reproducible conditions during a period when at least one rotor blade is at least temporarily parallel to the tower in a nearly downward direction or parallel to the tower in a downward direction. This is achieved by minimizing and / or approximately keeping the distance between at least one actuator and at least one unstable pressure sensor at the rotor blade and the noise sensor at the tower for each adjustment of the transfer function.

[0032] The adjustment of the unstable pressure filter function occurs specifically during a time period in which at least one rotor blade is arranged at least temporarily in an almost downward direction parallel to the tower. Therefore, the transfer function is adjusted over a time period including the points in time when the rotor blades are arranged parallel to the tower or when the rotor blades move toward the tower through parallel positions. The rotor position, particularly the position of each rotor blade, can be determined by a rotor orientation sensor of the control unit or a rotor orientation sensor connected to the control unit.

[0033] The rotor orientation sensor can be a gravity sensor arranged on at least one rotor blade and / or a rotary encoder coupled to the rotor. The gravity sensor arranged on at least one rotor blade can determine the position of the rotor blade relative to the direction of gravity, and thus determine the point in time when at least one rotor blade is in a downward or nearly downward direction parallel to the tower. A rotary encoder coupled to the rotor can also be used to determine the rotor orientation. Such a rotary encoder can, for example, be part of the motor used as a generator and / or the drivetrain of a wind turbine, wherein the rotary encoder determines the position of the drivetrain shaft and / or the rotor of the motor, and thus also determines the position of the wind turbine rotor, which is coupled to the drivetrain and the generator.

[0034] Preferably, at least one actuator and at least one unstable pressure sensor are arranged on each of the plurality of rotor blades, wherein the control unit is adapted to use a separate filter function for each rotor blade, and is adapted to adjust each individual filter function during the time period during which the corresponding rotor blade is at least temporarily in a downward direction parallel to or nearly downward of the tower. This allows for individual adjustment of the emissivity of the noise immunity signal for each rotor blade, making it possible to effectively compensate for effects that may only occur on a single blade, such as a damaged trailing edge.

[0035] In a preferred embodiment of the invention, the control unit is adapted to use an adaptive filter of the filtered x-Least Mean Square (FxLMS) algorithm as the filter function. The filtered x-LMS algorithm is an adaptive feedforward control that includes the adaptive filter. This adaptive filter can be provided as a filter function describing the correlation between the output signal of the unstable pressure sensor and the noise-resistant signal to be sent to the actuator.

[0036] An adaptive filter can be adjusted based on the output signal from a noise sensor. The noise sensor's output signal can be compared specifically to a expected noise level at the sensor's location, where the expected noise level is determined using a model of the wind turbine, particularly a model of the rotor blades and the active noise system. From this model, the expected noise level under the current operating conditions of the wind turbine or rotor blades can be obtained. For example, the filter can be adjusted based on the difference between the expected noise level and the measured noise level described by the noise sensor's output signal. Alternatively, the adaptive filter can be adjusted to minimize the noise at the location of at least one far-field noise sensor, for example, in the sense of mean square pressure.

[0037] In either case, the control unit may include a control algorithm intended to minimize a prescribed cost function, where the cost function may be the difference between the desired noise level and the measured noise level at the far-field noise sensor, or the cost function may be the noise level at the far-field sensor. The former embodiment would require modeling the relationship between near-field unstable pressure and far-field noise to generate the expected noise level based on the unstable surface pressure. Minimization of the noise level can itself be achieved, for example, using a gradient-based minimization algorithm to adjust the filter function without requiring a model of the expected noise level.

[0038] Preferably, at least one rotor blade includes a trailing edge having a passive noise reduction device, particularly a serrated trailing edge profile. Adding a passive noise reduction device, such as a serrated trailing edge profile, can further reduce noise emitted by the wind turbine. The passive noise reduction device (particularly the serrated trailing edge profile) can be arranged across the entire trailing edge of the rotor blade, not just near components of active noise reduction devices mounted on the rotor blade.

[0039] The noise reduction method for a wind turbine according to the present invention uses a wind turbine including a tower supporting a nacelle and a rotor, and an active noise reduction device. The rotor has a plurality of rotor blades. The active noise reduction device includes at least one actuator, at least one unstable pressure sensor, at least one noise sensor, and a control unit. The at least one unstable pressure sensor is adapted to generate an output signal corresponding to turbulence conditions during operation of the wind turbine blades. The at least one noise sensor is adapted to generate an output signal corresponding to noise generated by the rotor blades at the location of the noise sensor. The unstable pressure sensor and the actuator are arranged on at least one of the rotor blades, and the noise sensor is arranged at the nacelle and / or the tower. The control unit controls the actuator to emit an anti-noise signal that at least partially reduces the noise generated by the rotor blades based on the output signals of the unstable pressure sensor and the noise sensor.

[0040] Preferably, the control unit applies at least one filter function to the output signal of at least one unstable pressure sensor in order to determine the input signal for the actuator to emit an anti-noise signal, wherein the control unit adjusts the filter function according to the output signal of the noise sensor, particularly in adaptive feedforward control.

[0041] In a preferred embodiment of the invention, the control unit includes or is connected to a rotor orientation sensor, particularly a gravity sensor arranged on at least one rotor blade and / or a rotary encoder coupled to the rotor, to determine the relative position between the rotor blade and the tower during operation of the wind turbine, wherein the control unit adjusts a filter function for at least a short period of time when at least one rotor blade is at least temporarily in a downward or nearly downward direction parallel to the tower.

[0042] Preferably, at least one actuator and at least one unstable pressure sensor are arranged on each of the plurality of rotor blades, wherein the control unit uses a separate filter function for each blade and adjusts each separate filter function for at least a temporary period of time during which the corresponding rotor blade is in a downward direction parallel to or nearly downward of the tower.

[0043] Preferably, the control unit uses an adaptive filter based on the least mean square (FxLMS) algorithm of the filter as the filter function.

[0044] All the details and advantages described above with respect to the wind turbine according to the invention are correspondingly applicable to the noise reduction method for the wind turbine according to the invention. Attached Figure Description

[0045] Other objects and features of the invention will become apparent from the following detailed description taken in conjunction with the accompanying drawings. However, the drawings are merely schematic diagrams, designed for illustrative purposes only, and are not intended to limit the invention. The drawings show:

[0046] Figure 1 This is an embodiment of the wind turbine according to the present invention.

[0047] Figure 2 This is a detail of the rotor blades of a wind turbine according to the present invention.

[0048] Figure 3 It is a control unit for a wind turbine according to the present invention, which is suitable for use in embodiments of the method according to the present invention, and

[0049] Figure 4 This is a block diagram of the control system used in the wind turbine according to the invention and in embodiments of the noise reduction method for the wind turbine according to the invention. Detailed Implementation

[0050] exist Figure 1 The image shows a wind turbine 1. The wind turbine 1 includes a tower 2 that supports a rotor 3 having a plurality of rotor blades 4. The rotor 3 is mounted to a nacelle 5 on the tower 2. The rotor blades 4 of the rotor 3 are connected to a hub 6 of the rotor 3, wherein the rotor 3 and thus the rotor blades 4 rotate during operation of the wind turbine 1.

[0051] The wind turbine 1 includes an active noise reduction device 7, wherein the active noise reduction device 7 includes at least one actuator 8, a plurality of unstable pressure sensors 9, a plurality of noise sensors 10, and a control unit 11. The actuator 8 and the unstable pressure sensors 9 are arranged and / or fixed to at least one rotor blade 4 of the wind turbine 1. In particular, at least one actuator 8 and at least one unstable pressure sensor 9 are arranged on each rotor blade 4.

[0052] Noise sensor 10 is arranged on tower 2, particularly on a tower segment of tower 2 that supports the nacelle 5 of tower 2. The noise sensor 10 is arranged in a ring around tower 2. Alternatively, the noise sensor 10, or some of the noise sensors 10, may be arranged in the nacelle 5, particularly at the rear. Control unit 11 is connected to actuator 8, unstable pressure sensor 9, and noise sensor 10 via at least one wireless connection and / or via at least one cable connection (not shown).

[0053] Actuator 8 and a plurality of unstable pressure sensors 9 are arranged in arrangement portion 12 of housing 13 of rotor blade 4, wherein the unstable pressure sensors 9 are arranged near trailing edge 14 of rotor blade 4. The arrangement portion 12, in which actuator 8 and unstable pressure sensors 9 are arranged, is located in the tip region of rotor blade 4, wherein the tip region, and therefore the arrangement portion 12, covers a portion of the length of rotor blade 4 from the tip, particularly 10% to 30%, preferably 20%.

[0054] The noise sensor 10 at the tower 2 is arranged in the opposing portion 15, wherein, when the rotor blades 4 are arranged parallel to the tower 2 of the wind turbine 1 as depicted, the opposing portion 15 is positioned relative to the arrangement portion 12 at the rotor blades 4. In other words, when the rotor blades are aligned parallel to the tower 2 in a downward direction toward the ground 16, both the arrangement portion 14 and the opposing portion 15 have the same distance from the ground 16 on which the wind turbine 1 is mounted.

[0055] The wind turbine 1 also includes at least one orientation sensor 17 for determining the relative position between the rotor blades 4 and the tower 2 during operation of the wind turbine 1. This at least one orientation sensor 17 is connected to a control unit 11. Alternatively, the control unit 11 may also be arranged, for example, in the rotor blades 4 of the wind turbine 1, wherein the control unit 11 may include the rotor orientation sensor 17. The rotor orientation sensor 17 arranged in the rotor blades 4 of the wind turbine 1 may, in particular, be a gravity sensor 18. Additionally, or alternatively, a rotary encoder 19 provided as part of the rotor 20 of the generator 21 of the wind turbine 1 arranged within the nacelle 5 may be used as the orientation sensor 17.

[0056] exist Figure 2 The image shows, in more detail, an unstable pressure sensor 9 arranged on a rotor blade 4, depicting a section or arrangement portion 12 of the rotor blade 4 in the tip region of the rotor blade 4. The unstable pressure sensor 9 is arranged near the trailing edge 14 of the rotor blade 4. The trailing edge 14 includes a passive noise reduction device 22, which is provided as a serrated trailing edge profile 23, the serrated trailing edge profile 23 including a plurality of teeth arranged along the trailing edge 14 of the rotor blade 4. An actuator 8 is also arranged on the housing 13, wherein the actuator 8 is provided as a loudspeaker for emitting an acoustic noise reduction signal. Each actuator 8 may include an amplifier for amplifying the noise reduction signal generated by the control unit 11.

[0057] During operation of the wind turbine 1, noise is generated, particularly at the trailing edge 14 of the rotor blade 4, by turbulence caused by air or wind flowing along the profile of the rotor blade 4 as indicated by arrow 24. To reduce the noise emitted by the rotor blade 4 in the far field or near the wind turbine 1, the actuator 8 is used to generate an anti-noise signal that at least partially reduces the noise generated by the rotor blade 4 during operation of the wind turbine 1.

[0058] like Figure 3 As shown, the control unit 11 receives the output signal of an unstable pressure sensor 9 disposed at the rotor blade 4. Furthermore, the control unit 11 receives the output signal of a noise sensor 10 located at the tower 2 and / or nacelle 5 of the wind turbine 1. The control unit 11 is adapted to control the actuator 8 based on the output signals of the unstable pressure sensor 9 and the noise sensor 10 to emit an anti-noise signal that at least partially reduces the noise generated by the rotor blade 4. Therefore, the control unit 11 is adapted to determine the anti-noise signal emitted by the actuator 8 using at least one transfer function describing the relationship between the output signal of the unstable pressure sensor 9 and the far-field noise. Furthermore, the control unit 11 is adapted to apply a filter function to the output signal of the unstable pressure sensor 9 to obtain an input signal for the actuator 8. The actuator 8 emits the anti-noise signal based on the input signal from the control unit 11.

[0059] The control unit 11 is adapted to adjust the filter function according to the output signal of the noise sensor 10, so that, in particular, when noise generated by the rotor 4 is measured at the location of the noise sensor 10 at the tower 2 of the wind turbine 1, the filter function can be corrected.

[0060] To account for the rotating components of the active noise reduction device 7, particularly the actuator 8 and the unstable pressure sensor 9, the control unit 11 is adapted to adjust the filter function during the time period during which at least one rotor blade 4 is at least temporarily in a downward or nearly downward direction parallel to the tower 2. Therefore, the control unit 11 is connected to the orientation sensor 17. The control unit 11 can use the filter function in adaptive feedforward control. Specifically, the control unit 11 uses a separate filter function for each rotor blade 4, allowing for individual reduction of noise generated by each rotor blade 4. The control unit 11 adjusts the filter function for each rotor blade 4 at or near the point in time when the corresponding rotor blade is aligned parallel to the tower 2.

[0061] exist Figure 4Block diagram 25 shows the feedforward control algorithm used in control unit 11. In block diagram 25, vector x represents the surface pressure measured by multiple unstable pressure sensors 9. Block 26 describes the adaptive control filter as vector w, which is used as a filter function to determine the vector of the control signal y sent to actuator 8.

[0062] Box 27 describes the physical relationship G between the actuator 8 and the noise sensor 10 located on the tower 2 of the wind turbine 1 in the form of an actuator transfer function. The actuator transfer function describes, for example, the relationship between the input signal of at least one actuator 8 and the far-field noise generated in a region of interest near the wind turbine 1, in which the noise emitted by the wind turbine should be reduced.

[0063] The filter function correlates the output signal x from at least one unstable pressure sensor with a signal sent to at least one actuator. For example, the filter function is implemented as a weighted linear combination of stored signal samples from at least one unstable pressure sensor. This can be mathematically represented as follows:

[0064] y m (n)=x(n)w m (n).

[0065] Here, y(n) is a vector of signals sent to actuator 8 in discrete time n, m is an index between 1 and M, where M is the number of actuators 8, x is a horizontal vector containing previously recorded samples of signals from unstable pressure sensor 9, and w is a filter function, which is represented as a vertical vector of length x containing weighted scalars.

[0066] The filter function w in block 27 can be calculated using the measured relationship between the unstable pressure described by the output signal of the unstable pressure sensor 9 and the far-field noise measured by the far-field noise sensor 10. The filter function depends on the relationship between the unstable pressure signal from the unstable noise sensor 9 and the far-field sound, although this relationship does not need to be calculated in the form of a transfer function at any point due to the use of an adaptive filter function.

[0067] In summation node 28, the noise suppression signal generated by actuator 8 and the noise n generated at rotor blade 4 are... a The residual error noise signal e, measured at noise sensor 10, is the output signal of noise sensor 10 and describes the residual noise generated from rotor blade 4, and is fed into summation node 29. There, the error noise signal e is added to or subtracted from the expected noise level, which is determined using the noise system in block 30. The model and the measured surface pressure x are determined. The output of node 29 is used to adjust the adaptive filter W in block 26 to reduce the error noise signal e measured at noise sensor 10.

[0068] System in box 30 The model describes, for example, the propagation path between the actuator 8, which acts as a second source, and the noise sensor 10. The filter function or adaptive filter w is adjusted based on the expected noise level at the location of the noise sensor 10 under the current operating state and the actual error noise signal e measured at the location of the noise sensor 10. Specifically, when the measured error noise signal e deviates from the model in box 30... When the expected noise level is determined, the adaptive filter w in block 26 is adjusted. Block diagram 25 shown is an example of the minimum mean square (FxLMS) algorithm for filtering.

[0069] Both the unstable pressure sensor 9 and the noise sensor 10 can be provided as unstable pressure transducers, particularly microphones. As previously mentioned, the adjustment of the filter w in block 26 occurs during the time period when the rotor blade 4 is arranged parallel to the tower 2, and therefore occurs when the unstable pressure sensor 9 is positioned close to the noise sensor 10. The adjustment of the filter w in block 26 allows for consideration of effects such as the current flow conditions of the fluid flowing along the rotor blade 4, as well as slowly occurring degradation effects during the operation of the wind turbine 1, such as the degradation of the trailing edge 14 of the rotor blade 4 and / or the degradation of the unstable pressure sensor 9. The adaptive filter w in block 26 can alternatively be adjusted to minimize the noise at the location of the noise sensor 10, for example, in the sense of mean square pressure. The minimization of the noise level itself can be achieved, for example, by a gradient-based minimization algorithm, in models where the expected noise level is not required. Adjust the filter function as needed.

[0070] Although the invention has been described in detail with reference to preferred embodiments, the invention is not limited to the disclosed examples, and those skilled in the art can derive other variations from the disclosed examples without departing from the scope of the invention.

Claims

1. A wind turbine, comprising a tower (2) supporting a nacelle (5) and a rotor (3) and an active noise reduction device (7), wherein the rotor (3) has multiple rotor blades (4), wherein, The active noise reduction device (7) includes at least one actuator (8), at least one unstable pressure sensor (9), at least one noise sensor (10), and a control unit (11). The at least one unstable pressure sensor (9) is adapted to generate an output signal corresponding to the turbulence condition during the operation of the rotor blade (4), and the at least one noise sensor (10) is adapted to generate an output signal corresponding to the noise generated by the rotor blade (4) at the location of the noise sensor (10). The unstable pressure sensor (9) and the actuator (8) are arranged on the rotor blade (4). At least one of them, and the noise sensor (10) is arranged at the nacelle (5) and / or the tower (2), wherein the control unit (11) is adapted to control the actuator (8) according to the output signals of the unstable pressure sensor (9) and the noise sensor (10) to emit an anti-noise signal that at least partially reduces the noise generated by the rotor blades (4), wherein the control unit (11) is adapted to apply at least one filter function to the output signal of the unstable pressure sensor (9) in order to determine the input signal for the actuator (8) to emit the anti-noise signal.

2. The wind turbine according to claim 1, characterized in that, The unstable pressure sensor (9) and the actuator (8) are arranged in at least one arrangement portion (12) of the housing of the rotor blade (4), wherein the noise sensor (10) is arranged in an opposing portion (15) on the outside of the nacelle (5) and / or on the outside of the tower (2), the opposing portion (15) being at least partially opposite to the arrangement portion (12) of the rotor blade (4) when the rotor blade (4) is aligned in a downward direction parallel to the tower (2).

3. The wind turbine according to claim 1 or 2, characterized in that, The active noise reduction device (7) includes multiple noise sensors (10), wherein the noise sensors (10) are arranged in a ring around the outer circumference of the tower (2).

4. The wind turbine according to claim 1 or 2, characterized in that, The unstable pressure sensor (9) and / or the actuator (8) are arranged in the tip region (14) of the rotor blade (4) and / or the unstable pressure sensor (9) is arranged on the trailing edge (14) of the rotor blade.

5. The wind turbine according to claim 1 or 2, characterized in that, The actuator (8) includes a loudspeaker, and / or the unstable pressure sensor (9) includes a pressure transducer, and / or the noise sensor (10) includes a pressure transducer.

6. The wind turbine according to claim 1 or 2, characterized in that, The actuator (8) includes a speaker, and / or the unstable pressure sensor (9) includes a surface pressure transducer and / or a microphone, and / or the noise sensor (10) includes a microphone.

7. The wind turbine according to claim 1 or 2, characterized in that, The control unit (11) is adapted to adjust the filter function according to the output signal of the noise sensor (10).

8. The wind turbine according to claim 1 or 2, characterized in that, The control unit (11) is adapted to adjust the filter function according to the output signal of the noise sensor (10) in adaptive feedforward control.

9. The wind turbine according to claim 7, characterized in that, The control unit includes a rotor orientation sensor (17) or is connected to a rotor orientation sensor (17) to determine the relative position between the rotor blades (4) and the tower (2) during operation of the wind turbine (1), wherein the control unit (11) is adapted to adjust the filter function for at least one rotor blade (4) being at least temporarily in a downward direction parallel to the tower for a period of time.

10. The wind turbine according to claim 9, characterized in that, The rotor orientation sensor (17) is a gravity sensor (18) arranged on at least one rotor blade (4) and / or a rotary encoder (19) connected to the rotor (3).

11. The wind turbine according to claim 7, characterized in that, At least one actuator and at least one unstable pressure sensor are arranged on each of the plurality of rotor blades (4), wherein the control unit (11) is adapted to use a separate filter function for each of the rotor blades (4) and is adapted to adjust each of the separate filter functions for a period of time during which the corresponding rotor blade (4) is at least temporarily in a downward direction or nearly downward direction parallel to the tower (2).

12. The wind turbine according to claim 7, characterized in that, The control unit is adapted to use an adaptive filter based on the least mean square (FxLMS) algorithm of filtering as the filter function.

13. The wind turbine according to any one of claims 1-2 and 9-12, characterized in that, At least one rotor blade (4) includes a trailing edge (14) having a passive noise reduction device (22).

14. The wind turbine according to any one of claims 1-2 and 9-12, characterized in that, At least one rotor blade (4) includes a trailing edge (14) having a serrated trailing edge profile (23).

15. A noise reduction method for a wind turbine (1), the wind turbine comprising a tower (2) supporting a nacelle (5) and a rotor (3) and an active noise reduction device (7), the rotor (3) having a plurality of rotor blades (4), wherein, The active noise reduction device (7) includes at least one actuator (8), at least one unstable pressure sensor (9), at least one noise sensor (10), and a control unit (11). The at least one unstable pressure sensor (9) is adapted to generate an output signal corresponding to the turbulence condition during the operation of the rotor blade (4), and the at least one noise sensor (10) is adapted to generate an output signal corresponding to the noise generated by the rotor blade (4) at the location of the noise sensor (10). The unstable pressure sensor (9) and the actuator (8) are arranged on the rotor blade (4). At least one of them, and the noise sensor (10) is arranged at the nacelle (5) and / or the tower (2), wherein the control unit (11) controls the actuator (8) based on the output signals of the unstable pressure sensor (9) and the noise sensor (10) to emit an anti-noise signal that at least partially reduces the noise generated by the rotor blades (4), wherein the control unit (11) applies at least one filter function to the output signal of the at least one unstable pressure sensor (9) to determine the input signal for the actuator (8) to emit the anti-noise signal.

16. The method according to claim 15, characterized in that, The control unit (11) adjusts the filter function based on the output signal of the noise sensor (10).

17. The method according to claim 15, characterized in that, The control unit (11) adjusts the filter function in adaptive feedforward control based on the output signal of the noise sensor (10).

18. The method according to claim 16, characterized in that, The control unit includes a rotor orientation sensor (17) or is connected to a rotor orientation sensor (17) to determine the relative position between the rotor blades (4) and the tower (2) during operation of the wind turbine (1), wherein the control unit (11) adjusts the filter function for at least one rotor blade (4) at least temporarily in a downward or nearly downward direction parallel to the tower.

19. The method according to claim 18, characterized in that, The rotor orientation sensor (17) is a gravity sensor (18) arranged on at least one rotor blade (4) and / or a rotary encoder (19) connected to the rotor (3).

20. The method according to any one of claims 16-19, characterized in that, At least one actuator (8) and at least one unstable pressure sensor (9) are arranged on each of the plurality of rotor blades (4), wherein the control unit (11) uses a separate filter function for each of the rotor blades (4) and adjusts each of the separate filter functions for at least a temporary period of time during which the corresponding rotor blade (4) is in a downward or nearly downward direction parallel to the tower (2).

21. The method according to any one of claims 16-19, characterized in that, The control unit uses an adaptive filter based on the least mean square (FxLMS) algorithm of filtering as the filter function.

Images