Method and assembly for locating an anomaly in a rotor blade of a wind turbine

Abstract

The invention relates to a method for locating an anomaly in a rotor blade (6) of a wind turbine (1), said method having the following steps: - detecting airborne noise using at least one airborne noise sensor (11, 21, 31) arranged within the rotor blade (6); - detecting structure-borne noise using at least one structure-borne noise sensor (12, 22, 32) arranged in or on the rotor blade (6); - ascertaining times (tK, tL) at which an event during which airborne noise and structure-borne noise are emitted at an unknown source position in or on the rotor blade (6) is reflected in the detected airborne noise and in the detected structure-borne noise; - determining a time difference (Δt) between the times (tK, tL); and - ascertaining the source position on the basis of the time difference (Δt) and the positions of the airborne noise sensor (11, 21, 31) and the structure-borne noise sensor (12, 22, 32). The invention further relates to an assembly suitable for carrying out the method for locating an anomaly in a rotor blade (6) of a wind turbine (1).

Description

[0001] Method and arrangement for locating an anomaly in a rotor blade of a wind turbine

[0002] The invention relates to a method and an arrangement for locating an anomaly in a rotor blade of a wind turbine. The wind turbine has at least one airborne sound sensor arranged within the rotor blade and at least one structure-borne sound sensor arranged in or on the rotor blade.

[0003] An anomaly in a rotor blade is defined as any deviation of the rotor blade's structure and / or behavior compared to a normal state. Anomalies can be material damage to the rotor blade or events such as a lightning strike. Single events may cause no further damage or lead to additional anomalies, such as damage.

[0004] It is known that anomalies can be inferred from the vibrations of a rotor blade during the operation of a wind turbine. These anomalies can indicate, for example, the condition of the rotor blade itself or the occurrence of unique events, such as the accumulation or shedding of ice, or the aforementioned lightning strike. A method for analyzing vibrations in the rotor blade of a wind turbine is described, for example, in publication WO 2006 / 012 827 A1. In vibration analysis, measured vibration spectra are typically compared with reference spectra to identify anomalies based on typical defect patterns and, if necessary, to derive the causes of these defects.

[0005] In addition to simply detecting an anomaly, determining its location on or within the rotor blade can also be crucial for its assessment. For example, only knowing the position of a detected defect allows for a reliable assessment of whether the damage prevents the continued operation of the wind turbine and what measures are necessary to repair it.

[0006] Typically, the location of detected damage or general anomalies is determined through a time-consuming on-site inspection or a visual inspection using a drone. This is particularly complex and expensive for offshore wind turbines. A method for locating a lightning strike on a wind turbine rotor blade is known from US 2017 / 336532 A1. In this method, several microphones are mounted on the nacelle of the wind turbine to detect airborne sound. When lightning strikes a rotor blade located above the microphones, the travel time of the resulting sonic boom to the microphones is measured. The position of the lightning strike on the blade is then calculated from this travel time and the differences in travel time between the boom and the various microphones.However, due to the measurement of airborne sound using microphones on the gondola, this type of localization is only suitable for events associated with a very loud, explosive sound event such as a lightning strike.

[0007] It is an object of the present invention to reliably determine the position of an event associated with the emission of sound in or on a rotor blade. The detection should be sensitive enough to reliably detect even sound events that are less explosive than a lightning strike. Furthermore, detection should also be possible in positions of the rotor blade where the blade is not located above the nacelle.

[0008] This task is solved by a method or arrangement with the features of the respective independent claim. Advantageous embodiments and further developments are the subject of the dependent claims.

[0009] A method according to the invention for locating an anomaly in a rotor blade of a wind turbine comprises the following steps: Airborne sound is detected by means of at least one airborne sound sensor arranged within the rotor blade, and structure-borne sound is detected by means of at least one structure-borne sound sensor arranged in or on the rotor blade. Furthermore, the times at which an event, in which airborne sound and structure-borne sound are emitted at an unknown origin position in or on the rotor blade, is reflected in the detected airborne sound and in the detected structure-borne sound are determined, and a time difference between these times is calculated. The origin position is then determined based on the time difference and the positions of the airborne sound sensor and the structure-borne sound sensor.

[0010] This method utilizes the time-of-flight differences of various sound types to determine the sound's origin. The recorded sound measurements reflect the event, allowing the event to be detected and the time of its arrival at each sensor to be determined. The determination of the origin can then be made taking into account the different speeds of airborne and structure-borne sound.

[0011] Localization, in this context, means determining the origin position of the sound in at least one spatial direction, for example in the direction of a longitudinal axis of a rotor blade.

[0012] In an advantageous embodiment of the method, the at least one airborne sound sensor is arranged in essentially the same position as the at least one structure-borne sound sensor. The original position can be determined in this sensor geometry based on the different speeds of airborne and structure-borne sound, which would not be possible when measuring only one type of sound. Advantageously, both required sensors can therefore be arranged in a single measurement module in this embodiment, which simplifies signal processing and minimizes assembly effort. Alternatively or additionally, the at least one airborne sound sensor and the at least one structure-borne sound sensor can also be arranged in different positions.In the event that the airborne sound sensor and the structure-borne sound sensor are located in the same position, a further development may provide for the use of a single sensor for both airborne sound and structure-borne sound. Sound sensors are generally sensitive to both airborne and structure-borne sound. This is exploited here to detect the successively arriving signals of structure-borne and airborne sound with a single sensor, which simplifies the design of the arrangement for carrying out the method.

[0013] In a further advantageous embodiment of the method, at least one amplitude value of the airborne sound and / or structure-borne sound is analyzed to determine the times of arrival. In a technically simple embodiment that is also advantageous with regard to the energy requirements of the evaluation, the analysis of the at least one amplitude value of the airborne sound and / or structure-borne sound can include a comparison with a limit value. This evaluation is particularly suitable for one-off events, e.g., lightning strikes, which generate a large sound amplitude. The sound released by such events generally propagates in the form of wavefronts that arrive at the two sensors, i.e., the structure-borne sound sensor and the airborne sound sensor, at different times. Preferably, phase information of these wavefronts of airborne sound and / or structure-borne sound can be evaluated to determine the times of arrival with particular precision.Due to dispersion, the shape of the wavefront, which is reflected in the phase information, changes depending on the distance from the origin. This allows for an additional distance measurement that can be taken into account during localization. To reliably detect events with high selectivity, the analysis of at least one amplitude value of the airborne and / or structure-borne sound can be performed in a frequency-dependent manner. This is based on the understanding that certain events are more pronounced in specific frequency ranges of sound.

[0014] In a further advantageous embodiment of the method, an amplitude spectrum and / or an amplitude profile is analyzed using pattern recognition, and the time points are determined based on the occurrence of a recognized pattern. This type of evaluation enables the detection of the anomaly in the recorded sound signal even in the case of recurring events not associated with high sound amplitudes. Additionally, a recognized pattern can provide information about the type of anomaly, which can then be characterized accordingly using pattern recognition. For this purpose, comparisons with reference spectra of known anomalies can be made, for example.

[0015] In a further advantageous embodiment of the method, information from the at least one airborne sound sensor and the at least one structure-borne sound sensor is acquired and evaluated synchronously. This time-synchronous acquisition of the measured quantities enables accurate determination of the transit times. Preferably, the sensor readings are acquired with a time resolution and correlation to each other in the range of less than 10 milliseconds (ms), and particularly preferably less than 250 microseconds (ps), and most preferably in the range of 1 ps.

[0016] In the context of this application, "time-synchronous" means that the measurements either actually take place simultaneously and are transmitted to an evaluation unit, or that each measurement is timestamped so that measurements in the evaluation unit can be interpolated to the same point in time. Preferably, the timestamp is provided by an internal or external clock that sensors or the units reading them can access to establish a common time reference.

[0017] An arrangement according to the invention for locating an anomaly in a rotor blade of a wind turbine comprises at least one airborne sound sensor arranged within the rotor blade and at least one structure-borne sound sensor arranged in or on the rotor blade. The arrangement is characterized by an evaluation unit for carrying out such a method. The advantages described in connection with the method result.

[0018] The evaluation device can, for example, be arranged in a measuring module, preferably a measuring module near the leaf root, or within the distributor of the hub or downstream of this distributor.

[0019] In an advantageous embodiment of the arrangement, at least one measuring module is arranged in the rotor blade, comprising at least one airborne sound sensor and / or at least one structure-borne sound sensor. One or, preferably, several sensors can be arranged in a measuring module. For this purpose, the at least one measuring module is connected, for example, via an electrical cable for power supply and / or data exchange to a distribution box located in a hub of the wind turbine. This is particularly suitable for measuring modules located in the area of ​​a blade root. A power supply and data connection via an electrical cable is especially advantageous for an arrangement in the area of ​​the blade root. It is robust, cost-effective to implement, and, moreover, safe in the area of ​​the blade root with regard to interference radiation and the influence of strong fields, for example, caused by lightning near the wind turbine.

[0020] The measurement module can supply power to the sensors and / or preprocess the sensor signals, e.g., amplification, filtering, and / or digitization. Using one measurement module for multiple sensors simplifies signal processing and minimizes assembly effort. If at least two measurement modules are spaced apart within the at least one rotor blade, it is preferred that these measurement modules be coupled to each other for the transmission of power and / or signals or data. The at least two spaced-apart measurement modules can be coupled to each other via at least one optical fiber for the transmission of power and / or signals or data.The system may include a radiation source, in particular a light-emitting diode (LED) or a semiconductor laser, in one of the measurement modules connected via cable, and a photovoltaic cell in a connected measurement module that converts light received via the optical fiber into electricity used to operate the measurement module. Furthermore, an electrical energy storage unit may be provided in the measurement module to collect transmitted energy and to operate energy-intensive components of the measurement module in pulsed mode. Data can be transmitted using modulated light in the same or a separate optical fiber. The power and / or data connection via the at least one optical fiber is particularly resistant to interference.

[0021] In further advantageous embodiments of the arrangement, the at least one airborne sound sensor is mounted in a structure-borne sound damping manner and / or the at least one structure-borne sound sensor is mounted in a sound-insulated manner.

[0022] Airborne sound sensors are generally also sensitive to structure-borne sound vibrations, i.e., vibrations transmitted via the sensor's mounting point. Conversely, structure-borne sound sensors also typically detect airborne sound. The measures mentioned above advantageously reduce the proportion of each unwanted type of sound at the respective sensor.

[0023] The invention is explained in more detail below with reference to an exemplary embodiment and the accompanying figures. The figures show:

[0024] Fig. 1 shows a spatial representation of part of a wind turbine;

[0025] Fig. 2 shows a schematic representation of an arrangement for locating anomalies in rotor blades of the wind turbine of Fig. 1;

[0026] Fig. 3 four partial images a) to d) each with a schematic representation of a rotor blade in longitudinal section with at least one measuring module; and Fig. 4 five partial images a) to e) each with a schematic representation of a rotor blade in cross-section with at least one measuring module.

[0027] In the figures, identical reference symbols denote identical or equivalent elements. For clarity, not all elements in every figure are labeled with reference symbols.

[0028] Fig. 1 shows, by way of example, an upper part of a wind turbine 1, in which a nacelle 3 is rotatably mounted on a tower 2. A rotor 4 with hub 5 and, in this case, three rotor blades 6, is rotatably mounted on the nacelle 3 about a substantially horizontal axis. A drive train with gearbox and generator is located downstream of the rotor 4 in the nacelle 3.

[0029] According to the invention, at least one airborne sound sensor and at least one structure-borne sound sensor are used in at least one of the rotor blades 6 to locate anomalies. However, it is also advantageous to use two or more airborne sound sensors or two or more structure-borne sound sensors.

[0030] In the example shown in Figure 1, each of the rotor blades 6 has a measuring module 10 near the blade root and a measuring module 20 further away from the blade root, in which the aforementioned airborne sound sensors and / or structure-borne sound sensors are arranged. The measuring modules 10 and 20 serve to house the sensors, supply them with power, and at least preprocess the signals or data from the sensors. A distributor 40 is arranged in the hub 5 of the rotor 4. This distributor provides power to the measuring modules 10 and 20 in the rotor blades 6 and collects and forwards the measurement results from the measuring modules 10 and 20.

[0031] Electrical lines 41 for power and data are routed from the distributor 40 to the measuring modules 10 located near the blade root. In the illustrated embodiment, optical fibers 42 run from the measuring modules 10 located near the blade root to the measuring modules 20 located further away from the blade root. Energy in the form of light is transmitted via these fibers from the measuring modules 10 to the measuring modules 20 for their power supply, and data is transmitted bidirectionally. In this way, it is possible to connect the measuring modules 20 located far from the blade root deep within the rotor blade 6 without the risk of high voltages being induced in the electrical supply lines, e.g., by lightning, which could destroy the measuring modules 20 themselves or connected components. "Deep within the rotor blade" means, for example, at 1 / 3 of the length of the rotor blade 6 or even at the blade tip.

[0032] Fig. 2 shows the arrangement of the measuring modules 10, 20 and the distributor 40 via the electrical lines 41 and the optical fibers 42 again schematically.

[0033] In the embodiment shown in Figure 2, each measuring module 10, 20 is provided with an airborne sound sensor 11 or 21 and a structure-borne sound sensor 12 or 22. Examples of other possible arrangements are shown in Figures 3 and 4.

[0034] In the example shown, the measuring modules 10 near the leaf root are each positioned directly behind a leaf sheath 7. Energy and data are transferred from each measuring module 10 to the associated measuring module 20 via at least one optical fiber 42. For this purpose, an emitter 13 is arranged in each measuring module 10 near the leaf root, which feeds light into the at least one optical fiber 42. The emitter can be a high-power LED or a semiconductor laser. Each measuring module 20 further from the leaf root includes a receiver for the light with a photovoltaic cell 23, which converts the light back into electrical energy. A DC-DC converter is connected downstream of the photovoltaic cell 23 to provide operating voltage for the measuring module 20.

[0035] For data transmission, additional transceivers, each comprising an emitter and a receiver, are provided in each measurement module 10, 20. Data transmission can take place via the same optical fiber 42 through which the energy is also transmitted, or alternatively via a separate optical fiber.

[0036] Data transmitted from the measuring modules 20 furthest from the blade root to the measuring modules 10 closest to the blade root are forwarded from there to the distributor 40. All data can then be output via a wired connection 43 from the distributor 40, provided that wired data transmission between the nacelle 3 and the hub 5 of the wind turbine 1 is supported. Alternatively, a wireless transmission from the distributor 40 to a receiver in the nacelle 3 can be configured. Particularly in the case of wireless transmission, a timestamp, possibly provided by an internal or external clock, can be advantageous to prevent latency from affecting the time-synchronous evaluation of the measurements. The port 43 can be configured for data and power transmission and, for example, be designed as a PoE (Power over Ethernet) connection.If data transmission is wireless, port 43 can also be used solely for power supply.

[0037] According to the invention, airborne sound is detected by the at least one airborne sound sensor 11, 21 arranged within the rotor blade 6. For the sake of simplicity, the airborne sound sensor 11, 21 is hereinafter also referred to as a microphone 11, 21. Furthermore, structure-borne sound is detected by the at least one structure-borne sound sensor 12, 22 arranged in or on the rotor blade.

[0038] The respective microphone 11, 21 can be, for example, a condenser microphone, an electromagnetic microphone, or a piezoelectric microphone. Microphones are generally also sensitive to structure-borne vibrations, i.e., vibrations transmitted via a mounting point of the microphone. It is therefore advantageous to mount the microphone in such a way that the proportion of structure-borne vibrations transmitted and detected by the microphone 11, 21 is as low as possible. For this purpose, the microphone 11, 21 can be mounted or suspended in a damped manner, for example, by attaching it to a soft foam that transmits as little structure-borne vibration as possible.

[0039] A piezoelectric sensor and / or an accelerometer based on MEMS (Micro Electro-Mechanical System) or a fiber optic system can be used as the structure-borne sound sensor 12, 22. In contrast to the airborne sound sensor 11, 21 described above, the structure-borne sound sensor 12, 22 is advantageously mounted in such a way that it is as insensitive as possible to airborne sound, i.e., it does not detect airborne sound or detects it only with very low signal amplitudes. For example, the structure-borne sound sensor 12, 22 can be enclosed in a sound-insulating housing. Conversely, its detection of structure-borne sound is advantageously increased by mounting it in the most direct and secure connection possible to the substrate, i.e., a part of the rotor blade 6.

[0040] Optionally, one or more of the measuring modules 10, 20 can be equipped with an accelerometer and / or a gyroscope, which can be used to determine the operating states of the rotor blade 6 and thus indirectly also of the wind turbine 1. Advantageously, the power supply and signal processing infrastructure available for the microphones 11, 21 and the structure-borne sound sensors 12, 22 can also be used for accelerometers and / or gyroscopes.

[0041] Figure 3 shows, by way of example and not exhaustively, four different combinations in the partial images a - d) of how at least one microphone 11 , 21 and at least one structure sound sensor 12, 22 can be arranged in a rotor blade 6 to implement the method according to the invention.

[0042] In each of the partial images a - d) a rotor blade 6 is shown in a longitudinal section, with a blade root with a blade bulkhead 7 located on the left side of the figure 3 and the blade tip on the right side.

[0043] In the embodiment shown in partial figure a), a microphone 11 is arranged in the area of ​​the leaf sheath 7, i.e., in the area of ​​the leaf root, and a structure-borne sound sensor 22 is located in the area of ​​the leaf center. In this embodiment, the microphone 11 and the structure-borne sound sensor 22 are advantageously arranged in different measuring modules, for example, the microphone 11 in the measuring module 10 near the leaf root according to Figure 1 and the structure-borne sound sensor 22 in the measuring module 20 furthest from the leaf root. When evaluating the sound recorded by the microphone 11 or the structure-borne sound sensor 22 and determining the position of the event that emitted this sound, the radial position of the measuring modules 10, 20, and thus of the microphone 11 and the structure-borne sound sensor 22, must be taken into account accordingly.

[0044] In the embodiment shown in partial image b), only one measuring module 10 is present, which is mounted here for illustrative purposes at approximately one-third of the length of the rotor blade 6, away from the blade root and the blade shroud 7. Within this single measuring module 10, both the microphone 11 and the airborne sound sensor 21 are located in essentially the same position.

[0045] For the case of Figure 3 in sub-figure b), where one microphone 11 is located at the same location as one structure-borne sound sensor 12, the following explains how an origin position from which airborne and structure-borne sound are emitted can be determined. The origin position indicates the location of an anomaly, for example, an event such as a lightning strike. The following steps are performed: 1. The structure-borne sound sensor 12 measures a signal that can be attributed to the event at the origin position at a time t«

[0046] 2. It holds true that: x / VK = tK- to, where vx is the structure-borne sound velocity, to is the time of emission of the sound at the origin position, and x is the distance between the origin position and the position of the measuring module 10, which has the structure-borne sound sensor 12 and the microphone 11.

[0047] 3. Microphone 11 receives a signal that can be attributed to the (same) event at the original position at a time tu

[0048] 4. It holds true that: x / VL = ti_ — to, where VL is the speed of sound in air.

[0049] 5. From 2 and 4 it follows: x (1 / VL — 1 / VK) = | - ti_| = At ​​or x = At ​​ / (1 / VL - 1 / VK).

[0050] From the distance x between the measuring module 10 and the original position, and the known position of the measuring module 10, the original position can be determined as the location of the anomaly, for example, a defect. If both sensors, the microphone 11 and the structure-borne sound sensor 12, are not located in the same position, but there is a distance Ax between them, the previously given formulas can be adapted by replacing x with (x + Ax) in either step 2 or step 4.

[0051] In the case shown, the result is not unambiguous because, although the distance between the origin position and the measuring module 10 can be determined, the direction—i.e., whether the origin position lies radially outside or radially inside the measuring module 10—cannot be determined. If at least one additional sensor, either a second microphone 21 and / or a second structure-borne sound sensor 22, is present, the direction in which the origin position of the anomaly lies can also be unambiguously determined.

[0052] In the embodiment shown in partial diagram c), two microphones 11 and 21 and a structure-borne sound sensor 22 are present. One of the microphones, here microphone 11, is arranged separately in the measuring module 10 in the area of ​​the leaf bulkhead 7, whereas the other two sensors, microphone 21 and the structure-borne sound sensor 22, are arranged together in the measuring module 20 at a distance from the leaf bulkhead 7. The position of the sound source can be determined using the time-of-flight differences of the structure-borne sound to the position of the structure-borne sound sensor 22 and the airborne sound to either microphone 11 or microphone 21. In particular, the speed of sound propagation in air is highly dependent on the air parameters, for example, the temperature and, to a lesser extent, the humidity.In the embodiments shown in sub-images a) and b) of Figure 3, it is possible and useful to measure environmental parameters such as temperature and, if applicable, humidity, and to obtain the sound velocity in air values ​​used to calculate the original position from tables or characteristic curves, taking the measured parameters into account. The dependence of the sound velocity on environmental parameters can also be considered in the embodiment shown in sub-image c) of Figure 3. Alternatively or additionally, the sound velocity in air in the specific situation can be measured from the time difference with which a sound event is detected by the two microphones 11 and 21, and this measurement can be used to determine the position of the anomaly.

[0053] In partial image d) of Figure 3, both the measuring module 10 and the further measuring module 20 are equipped with a microphone 11 or 21 and a structure-borne sound sensor 12 or 22, respectively. In this embodiment, both the air velocity can be determined by comparing the time-of-flight differences of the two microphones 11 and 21, and the structure-borne sound velocity can be determined based on the time-of-flight differences of an event reaching the structure-borne sound sensors 12 and 22 from the known positions of the measuring modules 10 and 20. The position of an anomaly can also be calculated based on various combinations of the values ​​of the microphones 11 and 21 and the structure-borne sound sensors 12 and 22, whereby this redundant calculation contributes to more reliable and, if necessary, more accurate position determinations.

[0054] The method can be applied to a non-periodic signal from sensors 11, 12, 21, 22, for example, one caused by a single event. For this purpose, a threshold value for a signal amplitude can be defined. If a signal from a microphone 11, 21 or a structure-borne sound sensor 12, 22 exceeds a respective defined amplitude value, this is defined as time ti_, tK, at which an event is reflected in the detected airborne sound or structure-borne sound, respectively. However, for periodic signals, such as those synchronized with the rotational speed of the wind turbine rotor or synchronous with the natural modes of vibration of the rotor blade 6, time ti_ can also be extracted at which the recurring event is reflected in the detected airborne sound or structure-borne sound, respectively.For this purpose, pattern recognition is preferably performed on the recorded signal profile in order to recognize a recurring signal pattern and to determine the times ti and k as the times of occurrence of the pattern.

[0055] The evaluation device, which evaluates the signals or data of the sensors 11, 12, 21, 22, can, for example, be arranged in one of the measuring modules 10, 20, preferably a measuring module 10 near the leaf root, or be arranged within the distributor 40 in the hub 5 or be located downstream of the distributor 40.

[0056] Figure 4 shows a schematic cross-sectional view of a rotor blade 6 in five partial images (a-e). Figure 4 illustrates various lateral arrangement possibilities of microphones 11, 21, 31 and structure-borne sound sensors 12, 22, 32 in one measuring module 10 (partial image b)), in two measuring modules 10, 20 (partial images a, c, d)), and in three measuring modules 10, 20, 30.

[0057] The measuring modules 10, 20, 30 with microphones 11, 21, 31 and structure-borne sound sensors 12, 22, 32, respectively, are arranged at different positions in the blade cross-section. The depicted rotor blade 6 is divided into three segments by two blade webs 8 extending along the blade's longitudinal axis. Measuring modules 10, 20, 30 can be mounted, for example, on the blade webs 8 that define the segments. Mounting is also possible on the inner surface of the blade's outer skin, for example, in the area of ​​a blade leading edge 9, or on one of the blade sides or between the blade sides. The arrangement of the measuring modules 10, 20, 30 in the different blade segments makes it possible, particularly based on the amplitudes of the microphones 11, 21, 31, to narrow down the origin of the anomaly to one of the segments. Reference numeral

[0058] 1 wind turbine

[0059] 2 Tower

[0060] 3 gondolas

[0061] 4 Rotor

[0062] 5 hub

[0063] 6 rotor blades

[0064] 7 leaf sheath

[0065] 8 leaf web

[0066] 9 Leaf leading edge

[0067] 10 measuring module

[0068] 11 Airborne sound sensor (microphone)

[0069] 12 Structure-borne sound sensor

[0070] 13 emitters

[0071] 20 measuring module

[0072] 21 Airborne sound sensor (microphone)

[0073] 22 Structure-borne sound sensor

[0074] 23 photovoltaic cells

[0075] 30 measuring module

[0076] 31 Airborne sound sensor (microphone)

[0077] 32 Structure-borne sound sensor

[0078] 40 distributors

[0079] 41 electrical line

[0080] 42 optical fibers

Claims

Claims 1. Method for locating an anomaly in a rotor blade (6) of a wind turbine (1 ), comprising the following steps: - Detection of airborne sound by means of at least one airborne sound sensor (11 , 21 , 31) arranged within the rotor blade (6 ); - Detection of structure-borne sound by means of at least one structure-borne sound sensor (12, 22, 32) arranged in or on the rotor blade (6); - Determining time points ( , ti_) at which an event in which airborne sound and structure-borne sound are emitted at an unknown origin position in or on the rotor blade (6) is reflected in the detected airborne sound and in the detected structure-borne sound; - Determining a time difference (At) between the time points (tK, ti_); and - Determining the origin position based on the time difference (At) and the positions of the airborne sound sensor (11 , 21 , 31 ) and the structure-borne sound sensor (12, 22, 32).

2. Method according to claim 1, wherein the at least one airborne sound sensor (11, 21, 31) is arranged in the same position as the at least one structure-borne sound sensor (12, 22, 32).

3. Method according to claim 2, wherein a single sensor functions as both the at least one airborne sound sensor (11 , 21 , 31 ) and the at least one structure-borne sound sensor (12, 22, 32).

4. Method according to one of claims 1 to 3, wherein the at least one airborne sound sensor (11 , 21 , 31 ) and the at least one structure-borne sound sensor (12, 22, 32) are arranged at different positions.

5. Method according to one of claims 1 to 4, wherein at least one amplitude value of the airborne sound and / or the structure-borne sound is analyzed to determine the time points (tK, ti).

6. Method according to claim 5, wherein phase information of wavefronts of airborne sound and / or structure-borne sound is evaluated to determine the time points (tK, L).

7. Method according to claim 5 or 6, wherein the analysis of the at least one amplitude value of the airborne sound and / or the structure-borne sound comprises a comparison with a limit value.

8. Method according to one of claims 5 to 7, wherein the analysis of the at least one amplitude value of the airborne sound and / or the structure-borne sound is frequency-dependent.

9. Method according to claim 8, wherein an amplitude spectrum and / or an amplitude profile is analyzed using pattern recognition and the time points (tK, ti) are determined based on the occurrence of a recognized pattern.

10. Method according to claim 9, wherein the anomaly is characterized by means of pattern recognition.

11. Arrangement for localizing an anomaly in a rotor blade (6) of a wind turbine (1), comprising at least one airborne sound sensor (11, 21, 31) arranged within the rotor blade (6) and at least one structure-borne sound sensor (12, 22, 32) arranged in or on the rotor blade (6), characterized in that an evaluation device for carrying out a method according to one of claims 1 to 10 for analyzing detected airborne sound and detected structure-borne sound is provided.

12. Arrangement according to claim 11, wherein at least one measuring module (10, 20, 30) is arranged in the rotor blade (6), comprising the at least one airborne sound sensor (11, 21, 31) and / or the at least one structure-borne sound sensor (12, 22, 32).

13. Arrangement according to claim 12, wherein the at least one measuring module (10, 20, 30) is connected via an electrical line (41) for power supply and / or data exchange to a distributor (30) which is arranged in a hub (5) of a rotor (4).

14. Arrangement according to claim 12 or 13, wherein the at least one measuring module (10, 20, 30) is arranged in the area of ​​a leaf bulkhead (7).

15. Arrangement according to one of claims 12 to 14, wherein at least two of the measuring modules (10, 20, 30) are spaced apart from each other within the at least one rotor blade (6) and are coupled to each other for the transmission of energy and / or signals or data.

16. Arrangement according to claim 15, wherein the at least two spaced-apart measuring modules (10, 20, 30) are coupled to each other via at least one optical fiber (42) for the transmission of energy and / or signals or data.

17. Arrangement according to one of claims 11 to 16, wherein the at least one airborne sound sensor (11 , 21 , 31 ) is mounted in a structure-borne sound damping manner.

18. Arrangement according to one of claims 11 to 17, wherein the at least one structure-borne sound sensor (12, 22, 32) is mounted in a sound-insulated manner.

Images