Monitoring system for blade of wind power generator

Abstract

To provide a monitoring system that can immediately detect an occurrence of an abnormality in a blade of a wind power generator by using AE waves, and can also detect an occurrence location of the abnormality.SOLUTION: In a monitoring system 1 for a blade of a wind power generator, an AE sensor 35 of a sensor unit 3 detects AE waves generated by a blade B, and a detection result of the AE sensor 35 is transmitted from the sensor unit 3. A central unit 4 detects an occurrence of an abnormality in the blade B based on the detection result of the AE sensor 35 received from each of the multiple sensor units 3. Furthermore, when an abnormality has occurred in the blade B, the central unit 4 detects an occurrence location of the abnormality based on the detection result of the AE sensor 35 received from each of the multiple sensor units 3.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a monitoring system that uses AE waves (acoustic emission waves) to detect abnormalities in the blades of wind turbine generators. [Background technology]

[0002] Systems for detecting abnormalities in wind turbine generators have been proposed. For example, a system has been proposed that uses a vibration sensor and an AE sensor to diagnose abnormalities in the rolling bearings of a wind turbine generator (see, for example, Patent Document 1). In this conventional system, abnormalities in the rolling bearings are diagnosed using a vibration sensor for measuring the vibration waveform of the rolling bearings and an AE sensor for detecting AE waves generated by the rolling bearings. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2011-154020 Summary of the Invention [Problem to be solved by the invention]

[0004] However, no system has been proposed to date that uses AE waves to detect abnormalities in wind turbine blades. As wind turbines have become larger in recent years, the number of lightning strikes on these devices has been increasing. In particular, because the blades reach their highest point on a wind turbine as they rotate, the number of lightning strikes on the blades is not small. When an abnormality occurs, such as damage to a blade caused by a lightning strike, it is necessary to detect it immediately. It is also necessary to detect the location of the abnormality (the location of the damaged blade).

[0005] The present invention has been made in consideration of the above-mentioned problems, and aims to provide a monitoring system that can immediately detect the occurrence of an abnormality in the blades of a wind turbine generator by using AE waves, and can also detect the location of the abnormality. [Means for solving the problem]

[0006] The monitoring system of the present invention is a monitoring system for the blades of a wind turbine generator, the monitoring system comprising: an AE waveguide provided inside the blade, which has a hollow structure; a plurality of sensor units provided on the AE waveguide inside the blade; and a central unit capable of communicating with the plurality of sensor units, wherein the AE waveguide is made of a material that has lower propagation loss of AE waves than the material constituting the blade; the sensor units comprise an AE sensor for detecting AE waves generated by the blade and a transmitter for transmitting the detection results of the AE sensor; and the central unit comprises a receiver for receiving the detection results of the AE sensor transmitted from each of the plurality of sensor units, an abnormality occurrence detection unit for detecting the occurrence of an abnormality in the blade based on the received detection results, and an abnormality location detection unit for detecting the location of the abnormality if an abnormality occurs in the blade based on the received detection results.

[0007] According to this configuration, if an abnormality occurs in a blade of a wind turbine generator, AE waves generated at the location of the abnormality are detected by the AE sensors of the multiple sensor units. Wind turbine generator blades are generally made of a material with a large propagation loss for AE waves, such as GFRP (glass fiber reinforced plastic). In the present invention, an AE waveguide made of a material with a lower propagation loss for AE waves than the material making up the blade (e.g., metals such as aluminum or iron, or ceramics) is provided inside the hollow blade. This allows AE waves that do not propagate to the AE sensor (sensor unit) to be captured by the AE waveguide and propagated to the AE sensor via the AE waveguide. Using the AE waveguide in this way makes it possible to monitor a wide area of ​​the blade with a small number of AE sensors (sensor units).

[0008] The detection results of the AE sensors in the multiple sensor units are transmitted from each sensor unit to a central unit, and the central unit determines whether or not an abnormality has occurred in the blades and where the abnormality has occurred based on the received detection results.In this way, by using AE waves, the present invention can immediately detect the occurrence of an abnormality in the blades of a wind turbine generator and can also locate the location of the abnormality.

[0009] In addition, in the monitoring system of the present invention, the multiple AE waveguides may be arranged in parallel in a line inside the blade, and the spacing between adjacent AE waveguides may be set to be equal to or less than twice the detectable radius at which the AE sensor can detect AE waves.

[0010] With this configuration, multiple AE waveguides are arranged in parallel in a line at an appropriate interval d (an interval less than twice the detectable radius r of the AE sensor; i.e., d≦2r), so that AE waves that do not propagate directly to the AE sensor (sensor unit) can be captured by the AE waveguide and propagated to the AE sensor via the AE waveguide. By using multiple AE waveguides in this way (multiple AE waveguides arranged in parallel in a line), a wide area of ​​the blade can be monitored with a small number of AE sensors (sensor units).

[0011] In the monitoring system of the present invention, the AE waveguide may be made of a conductive material, and the sensor unit may include a power supply section that receives power supply via the AE waveguide.

[0012] According to this configuration, since the AE waveguide is made of a conductive material (for example, a metal such as aluminum or iron), power can be supplied to the sensor unit from an external power supply source via the AE waveguide.

[0013] and a second transmitting unit that transmits the correlation processing result of the correlation processing unit; and the central unit may include a second receiving unit that receives the correlation processing results transmitted from each of the plurality of sensor units, and a delay time dataset generating unit that generates a dataset of mutual delay times of the plurality of sensor units based on the received correlation processing results. The abnormality occurrence detection unit may detect that an abnormality has occurred in the blade based on the delay time dataset, and the abnormality location detection unit may detect the location of the abnormality based on the delay time dataset.

[0014] According to this configuration, a reference vibration wave (a pseudo-AE wave modulated by a pseudo-random signal) is generated from the transducer of one of the multiple synchronized sensor units, and the reference vibration wave is detected by the AE sensors of the other sensor units. The detected reference vibration wave is AD converted and then subjected to correlation processing, and the delay time from the generation of the reference vibration wave to its detection is calculated as the correlation processing result. The correlation processing results of the multiple sensor units are transmitted from each sensor unit to a central unit. The central unit detects whether an abnormality has occurred in the blade and the location of the abnormality based on a delay time data set generated based on the received correlation processing results. In this way, by using the reference vibration wave (a pseudo-AE wave modulated by a pseudo-random signal) to AD convert the detected reference vibration wave and then performing correlation processing, disturbances (noise) such as vibration waves caused by blade rotation can be eliminated. This improves the signal-to-noise ratio.

[0015] In the monitoring system of the present invention, the central unit may include a delay time dataset update unit that updates the delay time dataset by unsupervised machine learning.

[0016] With this configuration, the delay time dataset is updated by unsupervised machine learning, so it is possible to determine whether the blades of the wind turbine are normal based on the delay time dataset that reflects the current state of the blades. Furthermore, if an abnormality occurs in the blades, it is possible to appropriately detect whether an abnormality has occurred in the blades and the location of the abnormality based on the delay time dataset updated by unsupervised machine learning.

[0017] A method of the present invention is a method executed by a monitoring system for blades of a wind turbine generator, the monitoring system comprising: an AE waveguide provided inside the blade, which has a hollow structure; a plurality of sensor units provided inside the blade on the AE waveguide; and a central unit capable of communicating with the plurality of sensor units, wherein the AE waveguide is made of a material that has lower propagation loss of AE waves than a material constituting the blade. The method includes detecting AE waves generated by the blade with an AE sensor of the sensor unit; transmitting the detection result of the AE sensor from the sensor unit; receiving, by the central unit, the detection results of the AE sensor transmitted from each of the plurality of sensor units; detecting, by the central unit, that an abnormality has occurred in the blade based on the received detection results; and, when an abnormality has occurred in the blade, detecting a location of the abnormality based on the received detection results by the central unit.

[0018] With this method, as with the above-described system, if an abnormality occurs in the blades of a wind turbine generator, AE waves generated at the location of the abnormality are detected by the AE sensors of multiple sensor units. Wind turbine generator blades are generally made of a material with a large propagation loss for AE waves, such as GFRP (glass fiber reinforced plastic). In the present invention, an AE waveguide made of a material with a lower propagation loss for AE waves than the material making up the blade (e.g., metals such as aluminum or iron, or ceramics) is provided inside the hollow blade. This allows AE waves that do not propagate to the AE sensor (sensor unit) to be captured by the AE waveguide and transmitted to the AE sensor via the AE waveguide. Using the AE waveguide in this way allows a wide area of ​​the blade to be monitored with a small number of AE sensors (sensor units).

[0019] The detection results of the AE sensors in the multiple sensor units are transmitted from each sensor unit to a central unit, and the central unit determines whether or not an abnormality has occurred in the blades and where the abnormality has occurred based on the received detection results.In this way, by using AE waves, the present invention can immediately detect the occurrence of an abnormality in the blades of a wind turbine generator and can also locate the location of the abnormality. [Effects of the Invention]

[0020] According to the present invention, by using AE waves, it is possible to immediately detect the occurrence of an abnormality in the blades of a wind turbine generator, and also to locate the location of the abnormality. [Brief explanation of the drawings]

[0021] [Figure 1] 1 is an explanatory diagram showing the configuration of a monitoring system for blades of a wind turbine generator according to an embodiment of the present invention; [Figure 2] 1A is a diagram illustrating an example of an AE waveguide according to an embodiment of the present invention, and FIG. 1B is a diagram illustrating an example (comparative example) in which an AE waveguide is not provided. [Figure 3] FIG. 2 is a block diagram showing a configuration of a sensor unit according to the embodiment of the present invention. [Figure 4] FIG. 2 is a block diagram showing the configuration of a central unit according to the embodiment of the present invention. [Figure 5] FIG. 10 is a diagram illustrating a delay time data set according to an embodiment of the present invention. [Figure 6] FIG. 3 is a sequence diagram for explaining the operation (calibration and normality determination) of the monitoring system according to the embodiment of the present invention. [Figure 7] FIG. 3 is a sequence diagram for explaining the operation (detection of an abnormality in a blade and detection of an abnormal location) of the monitoring system according to the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0022] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A monitoring system according to an embodiment of the present invention will be described below with reference to the accompanying drawings. In this embodiment, a monitoring system used for monitoring the blades of a wind turbine generator will be described as an example.

[0023] The configuration of a monitoring system according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is an explanatory diagram showing the configuration of the monitoring system according to this embodiment. As shown in FIG. 1, a monitoring system 1 for a blade B of a wind turbine generator includes an AE waveguide 2, multiple sensor units 3, and a central unit 4. The AE waveguide 2 is provided inside the blade B of the wind turbine generator, which has a hollow structure, and the multiple sensor units 3 are provided above the AE waveguide 2 inside the blade B. The multiple sensor units 3 are synchronized with each other. The central unit 4 is provided inside a nacelle N of the wind turbine generator, and is capable of communicating with the multiple sensor units 3.

[0024] Blades B of the wind turbine generator are made of a material such as GFRP (glass fiber reinforced plastic). On the other hand, AE waveguide 2 is made of a material (for example, metal such as aluminum or iron, or ceramic) that has less propagation loss of AE waves than the material that makes up blade B. AE waveguide 2 in this embodiment is made of a conductive material, specifically, metal such as aluminum or iron.

[0025] Fig. 2(a) is a diagram illustrating an example of the AE waveguide 2 of this embodiment. As shown in Fig. 2(a), in this embodiment, multiple AE waveguides 2 are provided in parallel in a line inside the blade B, and the distance d between adjacent AE waveguides 2 is set to be equal to or less than twice the detectable radius r at which the sensor unit 3 (the AE sensor thereof) can detect AE waves.

[0026] In this way, multiple AE waveguides 2 are arranged in parallel in a line at appropriate intervals d (intervals less than twice the detectable radius r of the sensor unit 3; i.e., d≦2r), so that AE waves that do not propagate directly to the sensor unit 3 (its AE sensor) can be captured by the AE waveguide 2. For example, the AE wave generated at position P in Figure 2(a) does not propagate directly to the sensor unit 3 (sensor No.: S11). This is because the distance from position P to the sensor unit 3 (sensor No.: S11) is greater than the detectable radius r of the sensor unit 3 (its AE sensor). However, in this case, the distance from position P to the AE waveguide 2 (the AE waveguide 2 on the left in Figure 2) is shorter than the detectable radius r of the sensor unit 3 (its AE sensor). Therefore, the AE wave generated at position P can be captured by the AE waveguide 2 and propagated to the sensor unit 3 (its AE sensor) via the AE waveguide 2.

[0027] By using multiple AE waveguides 2 in this way, it becomes possible to monitor the presence or absence of AE waves over a wide area of ​​blade B (the area shown by the dot pattern in FIG. 2(a)) using a small number of sensor units 3 (AE sensors). On the other hand, FIG. 2(b) is a diagram illustrating an example (comparative example) in which no AE waveguide 2 is provided. As shown in FIG. 2(b), in this case, in order to monitor the wide area of ​​blade B, it is necessary to use a large number of sensor units 3 (AE sensors), and the area that can be monitored in the example of FIG. 2(b) is smaller than the area that can be monitored in the example of FIG. 2(a). For example, in the example of FIG. 2(b), position Q is not covered by the monitorable area.

[0028] Fig. 3 is a block diagram showing the configuration of the sensor unit 3. As shown in Fig. 3, the sensor unit 3 includes a radio unit 30, a power supply unit 31, a control unit 32, a drive circuit 33, a vibrator 34, an AE sensor 35, and a signal processing unit 36. The control unit 32 is configured with an MCU or the like, and includes a modulation processing unit 320 and a correlation processing unit 321 as functional blocks.

[0029] The wireless section 30 of the sensor unit 3 has the function of performing wireless communication with the central unit 4. This wireless section 30 has the function of transmitting the detection results of the AE sensor 35, and corresponds to the transmitter of the present invention. This wireless section 30 also has the function of transmitting the correlation processing results of the correlation processing section 321, and corresponds to the second transmitter of the present invention. The power supply section 31 has a power generation section 310 that generates electricity using kinetic energy generated by the rotational motion of the blade B, and a power supply section 311 that receives power from an external power supply source (not shown) via the AE waveguide 2. Note that known technologies can be used for these power generation and power supply.

[0030] The modulation processing unit 320 performs processing to modulate the vibration source of the AE wave with a pseudo-random signal. The vibration source of the pseudo-AE wave modulated with the pseudo-random signal is supplied to the vibrator 34 via the drive circuit 33. In this way, the vibrator 34 is configured to generate a pseudo-AE wave (reference vibration wave) modulated with the pseudo-random signal. This pseudo-AE wave (reference vibration wave) modulated with the pseudo-random signal is used when performing calibration.

[0031] The AE sensor 35 is a vibration sensor that has sensitivity in the frequency band of AE waves (several tens of kHz to several MHz). The AE sensor 35 can detect AE waves generated when an abnormality occurs in the blade B (AE waves generated due to an abnormality in the blade B). As described above, the detection results of the AE sensor 35 are transmitted from the wireless section 30 of the sensor unit 3. The AE sensor 35 can also detect pseudo AE waves (reference vibration waves) generated by other sensor units 3 during calibration. The AE sensor 35 may also have sensitivity in a frequency band wider than the frequency band of AE waves (for example, a frequency band of several Hz to several kHz).

[0032] The signal processing unit 36 ​​has a function of AD converting a reference vibration wave generated by another sensor unit 3 during calibration when the reference vibration wave is detected by the AE sensor 35. The correlation processing unit 321 has a function of performing correlation processing on the AD converted reference vibration wave to determine the delay time from the generation to the detection of the reference vibration wave. The delay time determined in this manner is output from the correlation processing unit 321 as the correlation processing result. The correlation processing result output from the correlation processing unit 321 is transmitted from the wireless unit 30 of the sensor unit 3, as described above.

[0033] Fig. 4 is a block diagram showing the configuration of the central unit 4. As shown in Fig. 4, the central unit 4 includes a radio unit 40, a control unit 41, and a storage unit 42. The control unit 41 is configured with an MCU or the like, and includes, as functional blocks, a delayed data set generation unit 410, an abnormality occurrence detection unit 411, an abnormality location detection unit 412, and a machine learning unit 413.

[0034] The wireless unit 40 of the central unit 4 has the function of performing wireless communication with the multiple sensor units 3. This wireless unit 40 has the function of receiving the detection results of the AE sensors 35 transmitted from each of the multiple sensor units 3 when an abnormality occurs in the blade B, and corresponds to the receiving unit of the present invention. Furthermore, this wireless unit 40 has the function of receiving the correlation processing results transmitted from each of the multiple sensor units 3 during calibration, and corresponds to the second receiving unit of the present invention.

[0035] The delay data set generation unit 410 has a function of generating a data set of mutual delay times between the multiple sensor units 3 based on the received correlation processing result. FIG. 5 is a diagram showing an example of a delay time data set. The delay data set generation unit 410 generates a mesh-like delay time data set as shown in FIG. 5. The generated delay time data set is stored in the storage unit 42.

[0036] The abnormality occurrence detection unit 411 has a function of detecting, when an abnormality occurs in blade B, that an abnormality has occurred in blade B based on the detection result of the AE sensor 35 received from the sensor unit 3. That is, if the detection result of the AE sensor 35 is that an AE wave generated due to an abnormality in blade B has been detected, it is determined that an abnormality has occurred in blade B (the occurrence of an abnormality in blade B is detected). On the other hand, if the detection result of the AE sensor 35 is not that an AE wave generated due to an abnormality in blade B has been detected, it is determined that blade B is normal.

[0037] Furthermore, the abnormality occurrence detection unit 411 has a function of detecting, during calibration, that an abnormality has occurred in blade B based on the delay time dataset generated by the delay dataset generation unit 410. For example, if the value of the delay time dataset generated by the delay dataset generation unit 410 is outside a predetermined allowable range, it is determined that an abnormality has occurred in blade B (the occurrence of an abnormality in blade B is detected). On the other hand, if the value of the delay time dataset generated by the delay dataset generation unit 410 is within a predetermined allowable range, it is determined that blade B is normal.

[0038] The abnormality location detection unit 412 has the function of detecting the location of an abnormality when an abnormality occurs in blade B based on the detection results received from multiple sensor units 3 and the delay time data set stored in the memory unit 42.

[0039] For example, the coordinates of the sensor unit 3 are

number

number

number

number

number

[0040] The machine learning unit 413 has a function of updating the delay time dataset stored in the storage unit 42 by unsupervised machine learning. Known techniques can be used for unsupervised machine learning. For example, the machine learning unit 413 may use a CNN (convolutional neural network) for the covariance matrix and an RNN (recurrent neural network) for the spatial distribution as a temporal sequence model to learn a normal pattern (pattern without anomalies) by unsupervised machine learning. This machine learning unit 413 corresponds to the delay time dataset update unit of the present invention. The delay time dataset updated by unsupervised machine learning is stored in the storage unit 42.

[0041] The operation of the monitoring system 1 configured as above will be described with reference to the drawings.

[0042] In the monitoring system 1 for blade B of a wind turbine generator according to this embodiment, when calibration and normality determination are performed, one of the multiple sensor units 3 (sensor unit A) modulates the vibration source with a pseudo-random signal (S10), and generates a reference vibration wave from the oscillator 34 (S11), as shown in Fig. 6. When this reference vibration wave is detected by another of the multiple sensor units 3 (sensor unit B) (S12), the reference vibration wave is converted to analog to digital (S13), and then correlation processing is performed to calculate a delay time (S14). The calculated delay time is transmitted to the central unit 4 (S15).

[0043] When the central unit 4 receives the delay times from the sensor unit 3 (S16), a delay time data set as shown in FIG. 5 is generated (S17), and the delay time data set stored in the storage unit 42 is updated by unsupervised machine learning (S18). Then, based on the updated delay time data set, it is determined whether or not blade B is normal (S19).

[0044] Next, in the monitoring system 1 for blade B of the wind turbine generator of this embodiment, when an abnormality in blade B occurs and abnormality detection and abnormal location detection for that blade B are performed, for example, one sensor unit 3 (sensor unit A) among the multiple sensor units 3 detects an AE wave caused by the abnormality in blade B (S20), and the detection result is transmitted to the central unit 4 (S21). The transmitted detection result is received by the central unit 4 (S22).

[0045] Similarly, another sensor unit 3 (sensor unit B) among the multiple sensor units 3 also detects AE waves caused by an abnormality in blade B (S23), and transmits the detection result to the central unit 4 (S24).The transmitted detection result is then received by the central unit 4 (S25).

[0046] The central unit 4 detects the occurrence of an abnormality in blade B based on the detection result received from the sensor unit 3 as described above (S26). The central unit 4 also detects the abnormal location in blade B based on the detection result received from the sensor unit 3 as described above and the data set of delay times stored in the memory unit 42 (S27).

[0047] According to the monitoring system 1 of this embodiment, when an abnormality occurs in a blade B of a wind turbine generator, AE waves generated at the location of the abnormality are detected by the AE sensors 35 of the multiple sensor units 3. The blade B of a wind turbine generator is generally made of a material that has a large propagation loss for AE waves, such as glass fiber reinforced plastic (GFRP). In this embodiment, the AE waveguide 2 made of a material that has a smaller propagation loss for AE waves than the material that makes up the blade B is provided inside the hollow blade B. Therefore, AE waves that do not propagate to the AE sensor 35 (sensor unit 3) can be captured by the AE waveguide 2 and propagated to the AE sensor 35 via the AE waveguide 2. By using the AE waveguide 2 in this way, a wide area of ​​the blade B can be monitored with a small number of AE sensors 35 (sensor unit 3).

[0048] The detection results of the AE sensors 35 of the multiple sensor units 3 are transmitted from each sensor unit 3 to the central unit 4, and the central unit 4 determines whether or not an abnormality has occurred in blade B and where the abnormality has occurred based on the received detection results. In this embodiment, by using AE waves in this way, it is possible to immediately detect that an abnormality has occurred in blade B of the wind turbine generator and to locate the location of the abnormality.

[0049] Furthermore, in this embodiment, multiple AE waveguides 2 are arranged in parallel in a line at appropriate intervals d (intervals that are equal to or less than twice the detectable radius r of the AE sensor 35; that is, d≦2r), so that AE waves that do not propagate directly to the AE sensor 35 (sensor unit 3) can be captured by the AE waveguides 2 and propagated to the AE sensor 35 via the AE waveguides 2. By using multiple AE waveguides 2 in this way (multiple AE waveguides 2 arranged in parallel in a line), a wide area of ​​the blade B can be monitored with a small number of AE sensors 35 (sensor units 3).

[0050] Furthermore, in this embodiment, since the AE waveguide 2 is made of a conductive material (for example, a metal such as aluminum or iron), power can be supplied to the sensor unit 3 from an external power supply source via the AE waveguide 2.

[0051] In this embodiment, a reference vibration wave (a pseudo AE wave modulated by a pseudo random signal) is generated from the transducer 34 of one of the sensor units 3 among the multiple sensor units 3 synchronized with each other, and the reference vibration wave is detected by the AE sensor 35 of the other sensor units 3. The detected reference vibration wave is AD converted and then subjected to correlation processing, and the delay time from the generation of the reference vibration wave to the detection thereof is obtained as the correlation processing result. The correlation processing results of the multiple sensor units 3 are transmitted from each sensor unit 3 to the central unit 4. The central unit 4 detects whether or not an abnormality has occurred in the blade B and the location of the abnormality based on a data set of delay times generated based on the received correlation processing results. In this way, by using the reference vibration wave (a pseudo AE wave modulated by a pseudo random signal) to AD convert the detected reference vibration wave and then performing correlation processing, disturbances (noise) such as vibration waves accompanying the rotation of the blade B can be eliminated. This improves the signal-to-noise ratio.

[0052] Furthermore, in this embodiment, the delay time dataset is updated by unsupervised machine learning, so that it is possible to determine whether blade B of the wind turbine generator is normal or not during calibration based on the delay time dataset that reflects the current state of blade B. Furthermore, if an abnormality occurs in blade B, it is possible to appropriately detect whether an abnormality has occurred in blade B and the location of the abnormality at the time of the abnormality, based on the delay time dataset updated by unsupervised machine learning.

[0053] Although the embodiments of the present invention have been described above by way of example, the scope of the present invention is not limited to these, and can be modified and changed according to the purpose within the scope of the claims. [Industrial Applicability]

[0054] As described above, the monitoring system of the present invention has the effect of being able to immediately detect an abnormality in the blades of a wind turbine generator by using AE waves, and also being able to detect the location of the abnormality, and is therefore useful for monitoring the blades of wind turbine generators, etc. [Explanation of symbols]

[0055] 1. Surveillance System 2 AE waveguide 3 Sensor Unit 4 Central Unit 30 Radio Section 31 Power supply section 310 Power Generation Department 311 Power Supply Unit 32 Control section 320 Modulation processing section 321 Correlation Processing Unit 33 Drive circuit 34 oscillator 35 AE sensor 36 Signal Processing Section 40 Radio Section 41 Control Unit 410 Delayed Dataset Generation Unit 411 Abnormality detection unit 412 Abnormality detection unit 413 Machine Learning Department 42 Storage section B-Blade N nacelle

Claims

1. A monitoring system for a blade of a wind turbine, comprising: The monitoring system includes: an AE waveguide provided inside the blade having a hollow structure; a plurality of sensor units provided on the AE waveguide inside the blade; a central unit capable of communicating with the plurality of sensor units; Equipped with The AE waveguide is The antenna is made of a material that has a lower propagation loss of AE waves than the material that makes up the blade, The sensor unit includes: an AE sensor for detecting AE waves generated by the blade; a transmitter that transmits the detection result of the AE sensor; Equipped with The central unit comprises: a receiving unit that receives the detection results of the AE sensors transmitted from each of the plurality of sensor units; an abnormality occurrence detection unit that detects an abnormality occurring in the blade based on the received detection result; an abnormality location detection unit that, when an abnormality occurs in the blade, detects a location of the abnormality based on the received detection result; A monitoring system comprising:

2. the plurality of AE waveguides are provided in parallel in a line inside the blade, 2. The monitoring system according to claim 1, wherein the interval between adjacent AE waveguides is set to be equal to or less than twice the detectable radius at which the AE sensor can detect AE waves.

3. the AE waveguide is made of a conductive material, The monitoring system according to claim 1 , wherein the sensor unit includes a power supply section that receives power via the AE waveguide.

4. the plurality of sensor units are synchronized with one another; The sensor unit includes: A vibrator that generates a reference vibration wave that is a pseudo AE wave modulated by a pseudo random signal; a correlation processing unit that, when the AE sensor detects the reference vibration wave generated by another sensor unit, performs correlation processing on the reference vibration wave after AD conversion, and outputs a delay time from the generation of the reference vibration wave to the detection thereof as a correlation processing result; a second transmission unit that transmits a correlation processing result of the correlation processing unit; Equipped with The central unit comprises: a second receiving unit that receives the correlation processing results transmitted from each of the plurality of sensor units; a delay time data set generation unit that generates a data set of mutual delay times between the plurality of sensor units based on the received correlation processing result; Equipped with the abnormality occurrence detection unit detects that an abnormality has occurred in the blade based on the data set of the delay time; The monitoring system according to claim 1 , wherein the abnormality location detection unit detects the location where the abnormality has occurred based on a data set of the delay time.

5. The central unit comprises: The monitoring system according to claim 4 , further comprising a delay time dataset update unit that updates the delay time dataset by unsupervised machine learning.

6. 1. A method performed in a monitoring system for a blade of a wind power plant, comprising: The monitoring system includes: an AE waveguide provided inside the blade having a hollow structure; a plurality of sensor units provided on the AE waveguide inside the blade; a central unit capable of communicating with the plurality of sensor units; Equipped with The AE waveguide is The antenna is made of a material that has a lower propagation loss of AE waves than the material that makes up the blade, The method comprises: Detecting AE waves generated by the blade with an AE sensor of the sensor unit; transmitting a detection result of the AE sensor from the sensor unit; receiving, at the central unit, the detection results of the AE sensors transmitted from each of the plurality of sensor units; detecting, in the central unit, that an abnormality has occurred in the blade based on the received detection result; When an abnormality occurs in the blade, the central unit detects a location of the abnormality based on the received detection result; A method comprising:

Images