JP2015161247A - Wind turbine blade damage detection method and wind turbine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a wind turbine blade damage detection method and a wind turbine capable of accurately detecting abnormality of a wind turbine blade.SOLUTION: A wind turbine blade damage detection method for a wind turbine rotor that includes at least one wind turbine blade 2, comprises: a data acquisition 18 step of acquiring vibration data indicating a vibration 20A of acoustic data of the wind turbine blade 2 or acoustic data indicating an internal sound 21 of the wind turbine blade 2; and a detection step of detecting damage to the wind turbine blade 2 on the basis of the vibration data or the acoustic data acquired in the data acquisition step, the damage to the wind turbine blade 2 being detected using only the vibration data or the acoustic data acquired when a wind velocity is within a wind velocity specified range out of the vibration data or the acoustic data acquired in the data acquisition step.

Description

  The present disclosure relates to a wind turbine blade damage detection method and a wind turbine.

As a method of detecting damage to a wind turbine blade at an initial stage, a method of performing damage detection using information acquired using a vibration sensor attached to the wind turbine blade is known.
For example, Patent Document 1 describes a windmill blade breakage detection device that includes a vibration sensor that detects vibrations on each windmill blade of a windmill, and a determination unit that compares and determines vibration patterns obtained from these vibration sensors. . In this wind turbine blade breakage detection device, a vibration signal (intensity signal) detected by a vibration sensor is converted into a frequency component by a spectrum analyzer, and further, a peak value for each section obtained by dividing a frequency region (for example, 0 to 10 MHz) is used. Pattern. The determination means compares the vibration waveform patterned in this way with a normal waveform pattern previously taken in the database. For example, if the value obtained by subtracting the normal waveform from the acquired vibration waveform varies by a predetermined ratio (for example, 20% or more), it is determined as abnormal.

JP 2001-349775 A

In the wind turbine blade breakage detection device described in Patent Document 1, since vibration is detected directly from the wind turbine blade, it is possible to detect vibration of the wind turbine blade with high sensitivity. Therefore, when an abnormality occurs in the wind turbine blade, detection is performed at an initial stage. Can do.
On the other hand, since the load applied to the wind turbine blade depends on the wind speed, the vibration of the wind turbine blade affected by the load is also affected by the wind speed. For this reason, even if the wind turbine blade breakage detection device can detect the vibration of the wind turbine blade, the wind turbine blade abnormality detection based on the detected vibration accurately detects the wind turbine blade abnormality without considering the effect of the wind speed. It ’s difficult.

  An object of at least one embodiment of the present invention is to provide a wind turbine blade damage detection method capable of accurately detecting an abnormality of a wind turbine blade.

A wind turbine blade damage detection method according to at least one embodiment of the present invention includes:
A wind turbine rotor damage detection method in a wind turbine rotor comprising at least one wind turbine blade,
A data acquisition step for acquiring vibration data indicating vibration of the wind turbine blades or acoustic data inside the wind turbine blades;
A detection step of detecting damage to the wind turbine blades based on a change over time of the vibration data or the acoustic data acquired in the data acquisition step,
In the detection step, the vibration of the wind turbine blade is damaged by using only the vibration data or the acoustic data acquired when the wind speed is within the wind speed regulation range among the vibration data or the acoustic data acquired in the data acquisition step. To detect.

Since the load applied to the wind turbine blade depends on the wind speed, the vibration data or the acoustic data acquired in the data acquisition step varies depending on the wind speed at the time of data acquisition. According to the wind turbine blade damage detection method described above, damage to the wind turbine blade is detected using only vibration data or acoustic data acquired when the wind speed is within the specified wind speed range. The influence of variation can be reduced, and wind turbine blade damage can be accurately detected.
Further, since vibration data or acoustic data necessary for the wind turbine blade damage detection method can be acquired even during operation of the wind turbine, it is not necessary to stop the wind turbine operation for the purpose of detecting wind turbine blade damage. Therefore, it is possible to detect damage to the wind turbine blades without causing a decrease in the operating rate of the wind turbine.

In some embodiments,
The windmill rotor includes a plurality of windmill blades,
In the data acquisition step, vibration data or acoustic data is acquired for each of the plurality of wind turbine blades,
Difference between vibration data or acoustic data of one detection target wind turbine blade among the plurality of wind turbine blades and a reference value reflecting vibration data or acoustic data of one or more comparison target wind turbine blades among other wind turbine blades Further comprising a difference calculating step for calculating
In the detection step, damage of the wind turbine blade to be detected is detected based on a change with time of the difference calculated in the difference calculation step.
In this case, vibration data or acoustic data of one detection target wind turbine blade among a plurality of wind turbine blades, and a reference value reflecting vibration data or acoustic data of one or more comparison target wind turbine blades among other wind turbine blades, The difference between the two is used for detecting damage to the wind turbine blade, so that it is possible to eliminate the influence on the damage detection due to the operating state such as a change in wind speed. For this reason, it becomes possible to detect the damage of a windmill blade more accurately.

In some embodiments, in the data acquisition step, the vibration data or the acoustic data is acquired using a vibration sensor or an acoustic sensor attached to an inner wall surface of the wind turbine blade.
In this case, since a vibration sensor or an acoustic sensor attached to the inner wall surface of the wind turbine blade is used for detecting damage to the wind turbine blade, cracks generated in the blade outer shell, adhesive peeling in the blade outer shell, etc. occur on the wind turbine blade. Damage can be detected.

In some embodiments, in the data acquisition step, the vibration data or the acoustic data is acquired using a vibration sensor or an acoustic sensor attached to a front edge portion or a rear edge portion of the wind turbine blade.
In a wind turbine blade produced by bonding the back wing skin and the abdominal wing skin with an adhesive, the adhesive may be peeled off due to deterioration over time, and the mouth may be open. In the above embodiment, since the vibration data or the acoustic data obtained by using the vibration sensor or the acoustic sensor attached to the front edge portion or the rear edge portion of the wind turbine blade is used in the damage detection of the wind turbine blade, the front edge of the wind turbine blade is used. Or it is easy to detect the mouth opening at the trailing edge.

In some embodiments, in the data acquisition step, the vibration data or the acoustic data is acquired using a vibration sensor or an acoustic sensor attached to the back side or the ventral side of the wind turbine blade.
When the shear web provided along the blade length direction of the windmill blade and the inner wall of the windmill blade are bonded together with an adhesive, the adhesive may be peeled off due to deterioration over time. In the above embodiment, since the vibration data or the acoustic data acquired using the vibration sensor or the acoustic sensor attached to the back side or the abdomen side of the wind turbine blade is used in the detection of the damage to the wind turbine blade, the back side or the abdomen of the wind turbine blade is used. It is easy to detect the adhesive peeling between the shear web and the inner wall of the wind turbine blade and the occurrence and development of cracks on the back side or the ventral side of the wind turbine blade.

In some embodiments, in the data acquisition step, the vibration data or the acoustic data is acquired using a vibration sensor or an acoustic sensor attached to a shear web of the wind turbine blade.
In this case, since the vibration sensor or the acoustic sensor attached to the shear web of the wind turbine blade is used for detecting the damage of the wind turbine blade, it is easy to detect the peeling of the adhesive material between the shear web and the inner wall of the wind turbine blade.

In some embodiments, in the data acquisition step, the vibration data or the acoustic data is acquired using a vibration sensor or an acoustic sensor attached to a root of the wind turbine blade.
If a crack or the like generated at the blade root part develops, it may lead to a serious accident such as breakage of a wind turbine blade. In the above embodiment, since the vibration data or the acoustic data acquired using the vibration sensor or the acoustic sensor attached to the blade root part of the wind turbine blade is used in the detection of the damage to the wind turbine blade, the wind turbine blade may be broken or the like. Damage to a certain blade root can be detected effectively.

In some embodiments, in the data acquisition step, the vibration data or the acoustic data is acquired using a vibration sensor or an acoustic sensor attached to a blade tip portion of the wind turbine blade.
In a windmill, there is a hub to which the windmill blade is attached and a device connected to the hub on the blade root side (rotor head side) opposite to the blade tip of the windmill blade. Vibration or sound is generated due to movement. For this reason, when a vibration sensor or an acoustic sensor is provided on the rotor head side of the wind turbine blade, vibration due to damage or noise due to vibration or sound other than sound is easily picked up by the sensor. In the above embodiment, in the wind turbine blade damage detection, vibration data or acoustic data acquired using a vibration sensor or an acoustic sensor attached to the tip of the wind turbine blade is used. Damage to the wind turbine blade can be detected in a state where the influence of noise due to vibration and sound is reduced.
Further, in the above embodiment, since the vibration data or the acoustic data acquired using the vibration sensor or the acoustic sensor attached to the tip of the wind turbine blade is used in the detection of damage to the wind turbine blade, the wind turbine generated at the tip of the blade is used. It is easy to detect mouth opening due to adhesive peeling of the wing skin.

In some embodiments, in the detection step, the failure mode is further determined based on a change with time in the peak frequency of the wind turbine blade.
In this way, more detailed damage detection can be performed by detecting damage to the wind turbine blade and determining the failure mode based on the vibration data.

A windmill according to at least one embodiment of the present invention is:
A windmill rotor comprising at least one windmill blade;
A vibration sensor or an acoustic sensor for detecting vibration of the windmill blade or sound inside the windmill blade;
A damage detection unit for detecting damage to the windmill blade, and
The damage detection unit is based on a change with time of vibration data or acoustic data acquired when the wind speed is within a prescribed range of wind speed among vibration data or acoustic data acquired from the detection result of the vibration sensor or the acoustic sensor. And configured to detect damage to the wind turbine blade.

According to the windmill described above, damage to the windmill blade is detected using only vibration data or acoustic data acquired when the wind speed is within the wind speed regulation range, so that the influence of variations in vibration data or acoustic data on the wind speed is reduced. Therefore, it is possible to accurately detect damage to the wind turbine blade.
Further, according to the wind turbine, since vibration data or acoustic data necessary for the wind turbine blade damage detection method can be acquired even during operation of the wind turbine, the wind turbine operation is stopped for the purpose of detecting wind turbine blade damage. There is no need. Therefore, it is possible to detect damage to the wind turbine blades without causing a decrease in the operating rate of the wind turbine.

  According to at least one embodiment of the present invention, it is possible to accurately detect damage to a wind turbine blade.

It is the schematic which shows the whole structure of the windmill which concerns on one Embodiment, FIG. 1 (A) is the figure which looked at the windmill from the side, and FIG.1 (B) is the figure which looked at the windmill from the front. It is a figure which shows the outline | summary of the whole structure of the windmill blade of the windmill shown in FIG. It is an example of AA sectional drawing of the windmill blade shown in FIG. It is an example of BB sectional drawing of the windmill blade shown in FIG. It is an example of the graph which shows the vibration data acquired at the data acquisition step which concerns on one Embodiment. It is an example of the graph which shows the vibration data acquired at the data acquisition step which concerns on one Embodiment. It is an example of the graph which shows the vibration data acquired at the data acquisition step which concerns on one Embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention, but are merely illustrative examples.

First, the structure of a windmill provided with the windmill blade used as the object of damage detection in this invention is demonstrated.
FIG. 1 is a schematic view showing an entire configuration of a windmill 1 according to an embodiment of the present invention. FIG. 1 (A) is a view of the windmill 1 as viewed from the side, and FIG. It is the figure seen from the front. As shown in FIGS. 1A and 1B, a windmill 1 includes a windmill rotor 6 including at least one windmill blade 2 and a hub 4 to which the windmill blade 2 is attached, a nacelle 8 and a nacelle 8. And a supporting tower 10. In the wind turbine 1 shown in FIG. 1, three wind turbine blades 2 are attached to the hub 4. In the wind turbine 1, when the wind hits the wind turbine blade 2, the wind turbine rotor 6 including the wind turbine blade 2 and the hub 4 rotates around the rotation axis of the wind turbine rotor 6.
The windmill 1 shown in FIG. 1 includes a damage detection unit 30 for detecting damage to the windmill blade 2.

  The windmill 1 may be a wind power generator. In this case, the nacelle 8 accommodates a power transmission mechanism for transmitting the rotation of the generator and the wind turbine rotor 6 to the generator, and the wind turbine 1 is rotated from the wind turbine rotor 6 to the generator by the power transmission mechanism. The energy may be configured to be converted to electrical energy by a generator.

The wind turbine blade 2 includes a vibration sensor 20 (20A, 20B) for detecting vibration of the wind turbine blade 2 or an acoustic sensor 21 for detecting sound inside the wind turbine blade 2. FIG. 1 shows, as an example, a wind turbine 1 in which vibration sensors 20 (20A, 20B) are attached to the blade root portions 12 of the wind turbine blades 2. Either one of the vibration sensor 20 or the acoustic sensor 21 may be attached to the wind turbine blade 2, or both the vibration sensor 20 (20A, 20B) and the acoustic sensor 21 may be attached.
In the wind turbine 1 shown in FIG. 1, the blade root portion 12 is a structural portion constituting the end portion of the wind turbine blade 2 on the hub 4 side. The blade root portion 12 is formed in a cylindrical shape and bears a bending moment transmitted from the wind turbine blade 2 to the hub 4.

The vibration sensor 20 (20A, 20B) can detect the vibration of the measurement object. From the detection result, the vibration frequency of the measurement object to which the vibration sensor 20 (20A, 20B) is attached. And vibration data such as vibration level (amplitude) can be acquired. For example, an accelerometer can be used as the vibration sensor 20.
The acoustic sensor 21 can detect the sound at the measurement point, and can acquire acoustic data such as the frequency and volume of the sound at the measurement point from the detection result. As the acoustic sensor 21, for example, a microphone can be used.

The vibration and sound signals detected by the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 attached to the wind turbine blade 2 are transferred to the data recording unit 18 accommodated in the nacelle 8 via the signal cable 16. Entered. The data recording unit 18 may be accommodated in the hub 4.
The vibration or acoustic signal input to the data recording unit 18 is input to the damage detection unit 30 via the signal cable 17 wired inside the tower 10, and the damage detection unit 30 detects damage to the wind turbine blade 2. . Here, the term “damage” refers to an event that leads to breakage of the wind turbine blades.For example, the occurrence and progress of cracks in the outer skin of the wind turbine blades, the occurrence and development of adhesive peeling, etc. Is happening.
As shown in FIG. 1, the damage detection unit 30 may be arranged outside the nacelle 8 or the tower 10, or may be arranged inside the nacelle 8 or the tower 10. Further, the damage detection unit 30 collects, monitors, and manages information related to the windmill (for example, various information such as wind speed and the number of rotations of the windmill rotor) at a remote location away from the windmill (for example, Part of SCADA).
The detection of windmill blade damage by the damage detection unit 30 will be described later.

  Next, the configuration of the wind turbine blade 2 and the mounting position of the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 on the wind turbine blade 2 will be described with reference to FIGS. FIG. 2 is a diagram showing an outline of the entire structure of the wind turbine blade 2 of the wind turbine 1 shown in FIG. 3 and FIG. 4 are examples of a cross-sectional view of the wind turbine blade shown in FIG. 2, respectively. FIG. 3 is an example of the AA cross-sectional view of FIG. 2, and FIG. It is an example of a figure.

As shown in FIGS. 2 and 3, the wind turbine blade 2 includes a hollow blade body 3 and two shear webs 36 extending in the blade length direction of the wind turbine blade 2 inside the blade body 3. Is provided. The wind turbine blade 2 is attached to the hub 4 using a bolt or the like at a blade root 62 which is one end of the wind turbine blade 2 (see FIG. 1). The end of the wind turbine blade 2 opposite to the blade root 62 is a blade tip 64.
As shown in FIG. 3, the wing body 3 includes a dorsal outer skin 32 constituting the outer surface of the dorsal side 22 and an abdominal side skin 34 constituting the outer surface of the abdominal side 24. The outer skin 34 is bonded to the wind turbine blade 2 at the front edge 26 and the rear edge 28 via an adhesive 38. As shown in FIGS. 3 and 4, the shear web 36 and the inner wall surface 5 of the wind turbine blade 2 are bonded to each other on the back side 22 and the ventral side 24 of the wind turbine blade 2 via an adhesive 39. In the present specification, the abdominal skin and the back skin may be collectively referred to as “skin” or “wing skin”.

A vibration sensor 20 (20A, 20B) or an acoustic sensor 21 is attached to the windmill blade 2. The vibration sensor 20 (20A, 20B) or the acoustic sensor 21 is attached to the inside of the wind turbine blade 2 in order to detect the vibration of the wind turbine blade 2 or the sound inside the wind turbine blade 2 without being influenced by outside such as wind pressure.
As shown in FIGS. 3 and 4, the vibration sensor 20 (20A, 20B) is fixed to the inside of the wind turbine blade 2 by using an adhesive 42, and for example, the inner wall surface 5 of the wind turbine blade 2 (that is, the inside of the blade body 3). Wall surface) and the surface of the shear web 36.
Further, as shown in FIG. 4, the acoustic sensor 21 uses a tape or the like on the inner wall surface 5 of the wind turbine blade 2 via the cushion material 44 so that solid sound or the like from the wind turbine blade 2 is not transmitted to the acoustic sensor 21. Fixed. For example, on the inner wall surface 5 of the wind turbine blade 2, a cushion material 44 having a sufficient thickness for absorbing solid sound or the like from the wind turbine blade 2 is laid, and the acoustic sensor 21 is placed on the cushion material 44. Along with the cushion material 44, the tape is attached to the inner wall surface 5 of the wind turbine blade 2 with tape at several places and fixed. The cushion material 44 absorbs solid vibration and solid sound propagating from the wind turbine blade 2, and is formed of, for example, a sponge.

As shown in FIG. 4, the vibration sensor 20 (20 </ b> A, 20 </ b> B) or the acoustic sensor 21 is connected to the data logger 54 included in the data recording unit 18 by the signal cable 16. The vibration or sound signal detected by the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 is input to the data logger 54 through the signal cable 16 and recorded. From the vibration or sound signal recorded in the data logger 54, vibration data such as frequency and vibration level or acoustic data such as frequency and volume can be obtained. Used to detect damage. The data logger 54 may be connected to the damage detection unit 30 by the signal cable 17 (see FIG. 1).
As shown in FIG. 4, a partition plate 46 is provided in the blade root portion 12 of the wind turbine blade 2, and a manhole 48 that can be opened and closed is provided in the partition plate 46. The signal cable 17 may connect the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 and the data logger 54 through an opening 49 provided in the manhole 48.

  The vibration sensor 20 (20A, 20B) or the acoustic sensor 21 may be connected to the data logger 54 via the amplifier 52. In this case, the vibration or sound signal detected by the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 is amplified by the amplifier 52, and the amplified signal is input to the data logger 54. FIG. 4 shows an example in which the vibration sensor 20 (20A, 20B) is connected to the data logger 54 via the amplifier 52.

In one embodiment, as shown in FIGS. 3 and 4, vibration data is acquired using a vibration sensor 20 </ b> A attached to the inner wall surface 5 of the wind turbine blade 2.
In this case, by using the vibration sensor 20 </ b> A attached to the inner wall surface 5 of the wind turbine blade 2, cracks occurring in the back side skin 32 or the ventral side skin 34, and adhesion between the back side skin 32 and the ventral side skin 34. Damage occurring in the wind turbine blade 2 such as peeling of the agent 38, peeling of the adhesive 39 between the back side skin 32 or the ventral side skin 34 and the shear web 36 can be detected.

In the example shown in FIG. 3, the vibration sensor 20 </ b> A is attached to the front edge portion 27 and the rear edge portion 29 of the wind turbine blade 2. The front edge portion 27 or the rear edge portion 29 of the wind turbine blade 2 is about 25% from the front edge 26 or the rear edge 28 in the cord direction connecting the front edge 26 and the rear edge 28 of the wind turbine blade 2, respectively. It is the part of the range.
In this case, it is easy to detect a mouth opening at the front edge 26 or the rear edge 28 of the wind turbine blade 2 (that is, peeling of the adhesive 38 between the back skin 32 and the ventral skin 34).

In the example shown in FIG. 4, the vibration sensor 20 </ b> A is attached to the back side 22 and the ventral side 24 of the wind turbine blade 2.
In this case, peeling of the adhesive 39 between the shear web 36 and the inner wall surface 5 of the windmill blade, which occurs on the back side 22 or the abdomen side 24 of the windmill blade 2, or cracks on the backside 22 or the abdomen side 24 of the windmill blade 2 are generated. Easy to detect occurrence and progress.
During operation of the windmill 1, the windmill blade 2 receives the wind with the abdomen side 24 facing the front side (windward side), and static pressure due to wind is constantly applied to the abdomen side 24 of the windmill blade 2. The wind turbine blade 2 is easily subjected to a bending load in the direction connecting the back side 22 and the ventral side 24 (flap direction). For this reason, in the windmill blade 2, the back side 22 or the abdomen side 24 tends to crack. In the example shown in FIG. 4, since it is easy to detect such a crack on the back side 22 or the ventral side 24, damage detection of the wind turbine blade 2 can be performed effectively.

In the example shown in FIG. 3, the vibration sensor 20 </ b> B is attached to the shear web 36 of the wind turbine blade 2.
In this case, it is easy to detect damage to the wind turbine blade 2 related to the shear web 36, and for example, it is easy to detect peeling of the adhesive 39 on the inner wall surface 5 of the shear web 36 and the wind turbine blade 2.

Similarly to the vibration sensor 20 (20A, 20B), the acoustic sensor 21 can be attached to the inner wall surface 5 or the shear web 36 of the wind turbine blade 2. Further, when the acoustic sensor 21 is attached to the inner wall surface 5 of the wind turbine blade 2, the acoustic sensor 21 may be attached to the front edge portion 27 or the rear edge portion 29 of the wind turbine blade 2, or the back side 22 or the ventral side 24.
In the example shown in FIG. 4, the acoustic sensor 21 is attached to the back side 22 of the inner wall surface 5 of the wind turbine blade 2.

In some embodiments, the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 may be attached to the blade root portion 12 (see FIG. 2) of the wind turbine blade 2. At this time, the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 may be attached to the front edge portion 27 or the rear edge portion, or the back side 22 or the abdomen side 24 in the blade root portion 12 of the wind turbine blade 2.
If the crack etc. which generate | occur | produced in the blade root part 12 will advance, it may lead to serious accidents, such as breakage of the windmill blade 2. FIG. In the above-described embodiment in which the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 is attached to the blade root portion 12, damage to the blade root portion 12 leading to breakage of the wind turbine blade 2 or the like can be detected effectively.

In some embodiments, the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 may be attached to the blade tip 14 (see FIG. 2) of the wind turbine blade 2. At this time, the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 may be attached to the front edge portion 27 or the rear edge portion 29, or the back side 22 or the ventral side 24 in the blade tip portion 14 of the wind turbine blade 2. . The blade tip 14 of the wind turbine blade 2 is a portion of the wind turbine blade 2 in a range of about 25% from the blade tip 64 in the blade length direction connecting the blade root 62 and the blade tip 64.
In the wind turbine 1, the hub 4 to which the wind turbine blade 2 is attached and the equipment connected to the hub 4 exist on the blade root 12 side (rotor head side) opposite to the blade tip 14 of the wind turbine blade 2. During the operation of the windmill 1, vibrations and sounds are generated due to the operation of the hub 4 and equipment. For this reason, when the vibration sensor 20 (20A, 20B) and the acoustic sensor 21 are provided on the rotor head side of the wind turbine blade 2, noise due to vibration and sound that is not caused by damage is easily picked up by the sensor. In the above-described embodiment, the vibration data or the acoustic data acquired by using the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 attached to the blade tip 14 of the wind turbine blade 2 is used in the damage detection of the wind turbine blade 2. It is possible to detect damage to the wind turbine blade 2 in a state where the influence of noise due to vibration or sound other than vibration due to damage or sound is reduced.
Moreover, in the said embodiment, in the damage detection of the windmill blade 2, the vibration acquired using the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 attached to the front edge part 27 or the rear edge part 29 of the windmill blade 2 is demonstrated. Since data or acoustic data is used, it is easy to detect mouth opening due to adhesive peeling between the dorsal outer skin 32 and the ventral outer skin 34 generated at the wing tip 14.

  The mounting position of the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 is not limited to the position described above. In the wind turbine blade 2, damage (cracking, adhesive peeling, etc.) to be detected by vibration or sound is detected. You may attach in the position which is easy to generate | occur | produce. For example, peeling is likely to occur at the bonded portion at the end of the shear web 36. Further, a lightning or erosion is likely to occur at the front edge near the blade tip 14, and a lightning due to lightning is likely to occur at the rear edge near the blade tip 14. In addition, lightning strikes and cracks due to excessive loads may occur in the entire wing skin. You may attach the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 in the position where the damage of such a detection target is easy to generate | occur | produce.

In addition, about the vibration sensor 20 (20A, 20B) or the acoustic sensor 21, it is not necessary to attach all the sensors shown in FIG. 3 and FIG. The vibration sensor 20 (20A, 20B) or the acoustic sensor 21 may be attached.
Moreover, either the vibration sensor 20 (20A, 20B) or the acoustic sensor 21 may be attached to the wind turbine blade 2, or both the vibration sensor 20 (20A, 20B) and the acoustic sensor 21 may be attached. .

Next, a wind turbine blade damage detection method according to an embodiment will be described.
5 to 7 are examples of graphs showing vibration data acquired in the data acquisition step according to the embodiment, respectively.

  In the wind turbine blade damage detection method according to an embodiment, the above-described wind turbine blade 2 of the wind turbine 1 is subjected to damage detection, and the data acquisition step for acquiring vibration data indicating the vibration of the wind turbine blade 2 and the data acquisition step are used. And a detecting step for detecting damage to the wind turbine blade 2 based on a change with time of the vibration data. In the detection step, damage to the wind turbine blade 2 is detected using only the vibration data acquired when the wind speed is within the specified wind speed range among the vibration data acquired in the data acquisition step.

First, the data acquisition step will be described.
In the data acquisition step, for example, in the wind turbine blade 2 in which the vibration sensor 20 is attached to the blade root portion 12, vibration data is acquired from the vibration signal of the wind turbine blade 2 obtained by the vibration sensor 20.
Examples of vibration data of the wind turbine blade 2 that can be acquired by the vibration sensor 20 include a vibration level (vibration intensity) and a peak frequency (natural frequency). The peak frequency is obtained by performing a frequency analysis on the vibration level measured for a certain time as shown in FIG. 5 and converting the vibration level into a frequency component. In FIG. 5, the peak frequency f _P is the natural frequency of the wind turbine blade 2.
Moreover, the wind speed at the time of measuring the vibration is also measured at the same time, and the relationship between the wind speed and the vibration level or peak frequency of the wind turbine blade 2 as shown in FIGS. 6 and 7 can be acquired as vibration data. .
The graph of FIG. 6 shows the average value of the vibration level measured for a fixed time or the vibration level of the peak frequency of vibration measured for a fixed time.
Moreover, the graph of FIG. 7 shows the peak frequency of vibration measured for a certain time.

  Next, the detection step will be described.

In some embodiments, the detecting step determines that the wind turbine blade 2 is damaged when the magnitude of the vibration level of the wind turbine blade 2 exceeds a predetermined threshold. The threshold value is determined in consideration of safety from the vibration level measured in a state where the wind turbine blade 2 is healthy.
At this time, since the load applied to the wind turbine blade depends on the wind speed, the magnitude of the vibration level (vibration data) acquired in the data acquisition step varies depending on the wind speed when the vibration signal is acquired by the vibration sensor.
For example, when a crack or adhesive peeling that causes damage occurs, the vibration level tends to increase with time. At this time, when the wind speed is high, the variation in the vibration signal detected by the vibration sensor 20 increases, and the magnitude of the vibration level (vibration data) obtained from the vibration signal also varies as shown in FIG. Becomes larger.
Therefore, only the vibration data when the wind speed is within the specified wind speed range (for example, the wind speed of 10 to 13 m / s) among the acquired vibration data is used, and the magnitude of the vibration level measured within the specified wind speed range is calculated. When a measured value outside the predetermined range (Th ₁ to Th ₂ shown in FIG. 6) is detected, it is determined that the wind turbine blade 2 is damaged.
Here, when the vibration data outside the threshold range is detected not only once but a plurality of times over time, it may be determined that the wind turbine blade 2 is damaged. In this case, it is possible to detect damage to the wind turbine blade 2 with higher accuracy.

In some embodiments, in the detection step, it is determined that the wind turbine blade 2 is damaged when the rate of change of the magnitude of the vibration level of the wind turbine blade 2 with respect to time exceeds a predetermined threshold value. It is considered that the magnitude of the vibration level increases with the progress of damage to the wind turbine blade 2. For this reason, it is considered possible to detect damage to the wind turbine blade 2 based on the rate of change in the vibration level of the wind turbine blade. The threshold value is determined in consideration of safety from the vibration level measured in a state where the wind turbine blade 2 is healthy.
At this time, since the load applied to the wind turbine blade depends on the wind speed, the rate of change (vibration data) of the magnitude of the vibration level acquired in the data acquisition step has a value corresponding to the wind speed when the vibration signal is acquired by the vibration sensor. It varies.
Therefore, only the vibration data when the wind speed is within the specified wind speed range (for example, the wind speed of 10 to 13 m / s) among the acquired vibration data is used, and the magnitude of the vibration level measured within the specified wind speed range is used. When a measurement point whose rate of change exceeds a threshold value occurs, it is determined that the wind turbine blade 2 is damaged.
Here, when the vibration data outside the threshold range is detected not only once but a plurality of times over time, it may be determined that the wind turbine blade 2 is damaged. By doing in this way, it becomes possible to detect damage to the wind turbine blade 2 with higher accuracy.

In some embodiments, the detection step determines that there is damage to the wind turbine blade 2 when the peak frequency of the wind turbine blade 2 exceeds a predetermined threshold range. When damage such as cracking or adhesive peeling that causes breakage or the like occurs in the wind turbine blade 2, the peak frequency (natural frequency) of the wind turbine blade 2 changes. Therefore, damage can be detected based on the change in the peak frequency of the wind turbine blade 2. The threshold is determined in consideration of safety from the peak frequency measured in a state where the wind turbine blade 2 is healthy.
At this time, since the load applied to the wind turbine blade depends on the wind speed, the peak frequency (vibration data) acquired in the data acquisition step may vary depending on the wind speed when the vibration signal is acquired by the vibration sensor.
Therefore, among the acquired vibration data, only the vibration data when the wind speed is within the specified wind speed range (for example, the wind speed of 10 to 13 m / s) is used, and the vibration frequency measured within the specified wind speed range is within the threshold range. When a measurement value that is outside (Th ₃ to Th ₄ shown in FIG. 7) appears, it is determined that the wind turbine blade 2 is damaged.
Here, when the vibration data outside the threshold range is detected not only once but a plurality of times over time, it may be determined that the wind turbine blade 2 is damaged. In this case, it is possible to detect damage to the wind turbine blade 2 with higher accuracy.
As described above, the peak frequency (natural frequency) of the wind turbine blade 2 can be obtained from the vibration signal of the wind turbine blade 2 obtained by the vibration sensor 20, and is measured for a certain time as shown in FIG. This is obtained by performing frequency analysis on the vibration level and converting the vibration level into a frequency component. In the detection step, the change with time of the peak frequency obtained in this way is monitored.

  As described above, in the detection step, an example of a procedure for detecting damage to the wind turbine blade 2 based on any time-dependent change of the magnitude of the vibration level of the wind turbine blade 2, the rate of change of the magnitude of the vibration level, or the peak frequency. However, in one embodiment, in the detection step, the vibration level of the wind turbine blade 2, the rate of change of the magnitude of the vibration level, and the peak frequency are selected based on changes over time of two or more. Damage to the wind turbine blade 2 may be detected. For example, the damage to the wind turbine blade 2 may be detected based on the temporal change in the magnitude of the vibration level of the wind turbine blade 2 and the temporal change in the peak frequency of the wind turbine blade 2. Thus, it can detect more accurately by using in combination a plurality of indicators for damage detection.

In one embodiment, in the detection step, the failure mode is further determined based on a change over time in the peak frequency of the wind turbine blade 2.
First, the tendency of the change in peak frequency (natural frequency) when a failure (damage) that causes breakage of the wind turbine blade 2 occurs and the relationship between the failure modes are grasped in advance. Here, the failure mode refers to the type of failure (crack, adhesive peeling, etc.), the magnitude and position of the failure, and there is a natural frequency corresponding to each failure mode.
Then, when the peak frequency (vibration data) acquired by the vibration sensor 20 becomes approximately the same as the natural frequency in the failure mode grasped in advance, the failure mode occurs in the wind turbine blade 2. Is determined. For example, when the difference between the acquired peak frequency (vibration data) and the natural frequency in the failure mode grasped in advance is about 10% or less (that is, the acquired peak frequency (vibration data) is expressed as f1. When the natural frequency in the failure mode ascertained in advance is f0, when | f1-f0 | /f0≦0.1), it is determined that the failure mode is occurring in the wind turbine blade 2. May be.
In this way, more detailed damage detection can be performed by detecting damage to the wind turbine blade and determining the failure mode based on the vibration data.
Also, in the above-described failure mode determination, only the vibration data acquired when the wind speed is within the specified wind speed range (for example, the wind speed of 10 to 13 m / s) among the vibration data acquired in the data acquisition step is used. Thus, it is possible to detect damage to the wind turbine blade 2 with higher accuracy.

  The procedure for acquiring vibration data (vibration level and frequency) using the vibration sensor 20 in the data acquisition step and detecting the possibility of damage to the wind turbine blade 2 based on the change over time of the vibration data in the detection step is described above. did. As in the case where the possibility of damage to the wind turbine blade 2 is detected using the vibration data in this way, the acoustic data (volume or frequency) is acquired using the acoustic sensor 21 in the data acquisition step, and the acoustic data is detected in the detection step. It is possible to detect damage to the wind turbine blade 2 based on the change over time.

In the wind turbine blade damage detection method according to the embodiment, the wind turbine blade 2 of the wind turbine 1 including the plurality of wind turbine blades 2 described above is subjected to damage detection, and vibration data indicating vibration of each wind turbine blade 2 is acquired. A data acquisition step, a difference calculation step, and a detection step.
In the following, a case where damage to the wind turbine blade is detected using vibration data will be described. Similarly, damage can also be detected using acoustic data.

  In the data acquisition step, as in the data acquisition step described above, for example, in the wind turbine blade 2 in which the vibration sensor 20 is attached to the blade root portion 12, the vibration frequency and vibration level are determined from the vibration signal of the wind turbine blade 2 obtained by the vibration sensor 20. Acquire vibration data including.

In the difference calculation step, the difference between the vibration data of one detection target wind turbine blade among the plurality of wind turbine blades 2 and the reference value reflecting the vibration data of one or more comparison target wind turbine blades of the other wind turbine blades 2. Is calculated.
As the vibration data of the wind turbine blade 2, for example, the magnitude of the vibration level of the wind turbine blade 2 can be used. In this case, as the reference value, for example, the magnitude of the vibration level of one comparison target wind turbine blade among the other wind turbine blades 2 can be used. Further, as the reference value, an average value of vibration levels of two or more comparison target wind turbine blades among the other wind turbine blades 2, or two or more comparison target wind turbine blades among the other wind turbine blades 2, The average value of the vibration level of the one wind turbine blade to be detected can also be used.

In the detection step, damage of the wind turbine blade to be detected is detected based on the change over time of the difference calculated in the difference calculation step.
In this manner, the difference between the vibration data of one detection target wind turbine blade among the plurality of wind turbine blades 2 and the reference value reflecting the vibration data of one or more comparison target wind turbine blades among the other wind turbine blades is determined. By using it for the detection of damage to the blade 2, it is possible to eliminate the influence on the damage detection due to the operating state such as a change in wind speed. For this reason, it is possible to more accurately detect an abnormality caused by damage to the wind turbine blade 2.

In the wind turbine 1 according to the embodiment, the damage detection unit 30 may detect damage to the wind turbine blades.
Of the vibration data or acoustic data of the wind turbine blade 2 acquired from the detection result of the vibration sensor 20 or the acoustic sensor 21, the damage detection unit 30 acquires the vibration data or the acoustic data acquired when the wind speed is within the specified wind speed range. Is configured to detect damage to the wind turbine blade 2 based on the change over time.

&shy; The vibration data may be a vibration level or frequency of the wind turbine blade 2. The acoustic data may be a sound volume or frequency measured inside the wind turbine blade 2.

The damage detection unit 30 stores a predetermined wind speed regulation range and a predetermined threshold value of vibration data or acoustic data. The damage detection unit 30 stores the vibration data or acoustic data of the wind turbine blade 2 acquired from the detection result of the vibration sensor 20 or the acoustic sensor 21 when the measured value of the wind speed is in the stored wind speed regulation range. It may be configured to detect damage of the relevant wind turbine blade when the vibration data or the acoustic data exceeds the threshold value by comparing with the detected threshold value.
As for the wind speed, an anemometer may be provided in the windmill 1, and data measured by the anemometer may be input to the damage detection unit 30.

  According to the wind turbine 1 including the damage detection unit configured as described above, damage to the wind turbine blade 2 is detected using only vibration data acquired when the wind speed is within the wind speed regulation range. Can be reduced, and the wind turbine blade 2 can be accurately detected for damage.

DESCRIPTION OF SYMBOLS 1 Windmill 2 Windmill blade 3 Blade body 4 Hub 5 Inner wall surface 6 Windmill rotor 8 Nacelle 10 Tower 12 Blade root part 14 Blade tip part 16 Signal cable 18 Data recording part 20 Vibration sensor 21 Acoustic sensor 22 Back side 24 Abdominal side 26 Front edge 27 Front edge portion 28 Rear edge 29 Rear edge portion 30 Damage detection portion 32 Back side skin 34 Ventral side skin 36 Shear web 38 Adhesive 39 Adhesive 42 Adhesive 44 Cushion material 46 Partition plate 48 Manhole 49 Opening 52 Amplifier 54 Data logger 62 Wing root 64 Wing tip

Claims (10)

A wind turbine rotor damage detection method in a wind turbine rotor comprising at least one wind turbine blade,
A data acquisition step of acquiring vibration data indicating vibrations of the wind turbine blades or acoustic data indicating sound inside the wind turbine blades;
A detection step of detecting damage to the wind turbine blades based on a change over time of the vibration data or the acoustic data acquired in the data acquisition step,
In the detection step, the wind turbine blade is damaged using only the vibration data or the acoustic data acquired when the wind speed is within a predetermined range, among the vibration data or the acoustic data acquired in the data acquisition step. A wind turbine blade damage detection method characterized by detecting. The windmill rotor includes a plurality of windmill blades,
In the data acquisition step, vibration data or acoustic data is acquired for each of the plurality of wind turbine blades,
Difference between vibration data or acoustic data of one detection target wind turbine blade among the plurality of wind turbine blades and a reference value reflecting vibration data or acoustic data of one or more comparison target wind turbine blades among other wind turbine blades Further comprising a difference calculating step for calculating
The wind turbine blade damage detection method according to claim 1, wherein in the detection step, damage of the detection target wind turbine blade is detected based on a change with time of the difference calculated in the difference calculation step.   3. The wind turbine blade according to claim 1, wherein, in the data acquisition step, the vibration data or the acoustic data is acquired using a vibration sensor or an acoustic sensor attached to an inner wall surface of the wind turbine blade. Damage detection method.   The vibration data or the acoustic data is acquired in the data acquisition step using a vibration sensor or an acoustic sensor attached to a front edge portion or a rear edge portion of the wind turbine blade. Wind turbine blade damage detection method.   4. The wind turbine blade according to claim 3, wherein in the data acquisition step, the vibration data or the acoustic data is acquired using a vibration sensor or an acoustic sensor attached to a back side or an abdomen side of the wind turbine blade. Damage detection method.   In the data acquisition step, the vibration data or the acoustic data is acquired using a vibration sensor or an acoustic sensor attached to a shear web provided along the blade length direction of the wind turbine blade inside the wind turbine blade. The wind turbine blade damage detection method according to claim 1 or 2.   7. The data acquisition step according to claim 1, wherein the vibration data or the acoustic data is acquired using a vibration sensor or an acoustic sensor attached to a root portion of the wind turbine blade. The wind turbine blade damage detection method described.   7. The vibration data or the acoustic data is acquired in the data acquisition step using a vibration sensor or an acoustic sensor attached to a blade tip portion of the wind turbine blade. Wind turbine blade damage detection method according to claim 1.   The wind turbine blade damage detection method according to any one of claims 1 to 8, wherein in the detection step, a failure mode is further determined based on a change with time of a peak frequency of the wind turbine blade. A windmill rotor comprising at least one windmill blade;
A vibration sensor or an acoustic sensor for detecting vibration of the windmill blade or sound inside the windmill blade;
A damage detection unit for detecting damage to the windmill blade, and
The damage detection unit is based on a change with time of vibration data or acoustic data acquired when the wind speed is within a prescribed range of wind speed among vibration data or acoustic data acquired from the detection result of the vibration sensor or the acoustic sensor. A wind turbine configured to detect damage to the wind turbine blades.

Images