Abnormality determination device, abnormality determination system, abnormality determination method, and abnormality determination program

Abstract

To construct a system for determining abnormality of a wind turbine blade at low cost.SOLUTION: The abnormality determination device 4 includes a processor 43 that determines an abnormality of the wind turbine blade. The processor 43 acquires a plurality of captured images consecutive in time series generated by imaging the wind turbine blade, specifies a specific position in the wind turbine blade in the captured image, calculates a natural frequency of the wind turbine blade by frequency analysis based on the specific position specified for each of the plurality of captured images, and determines an abnormality of the wind turbine blade based on the natural frequency.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to an abnormality determination device, an abnormality determination system, an abnormality determination method, and an abnormality determination program. [Background technology]

[0002] BACKGROUND ART Conventionally, a technique for determining an abnormality in wind turbine blades in wind power generation facilities has been proposed (see, for example, Patent Document 1). In the technology described in Patent Document 1, a vibration sensor is attached to the inner wall surface of a wind turbine blade, and the vibration sensor acquires vibration data including the frequency and vibration level of the wind turbine blade. Furthermore, in this technology, a damage detection unit is provided outside the wind turbine power generation facility, and the vibration data acquired by the vibration sensor is transmitted to the damage detection unit. The damage detection unit then detects an abnormality in the wind turbine blade based on the vibration data, using the magnitude of the vibration level or changes over time in the peak vibration frequency. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6440367 Summary of the Invention [Problem to be solved by the invention]

[0004] However, when using the technology described in Patent Document 1 to determine abnormalities in wind turbine blades in existing wind power generation facilities, it is necessary to perform the cumbersome task of attaching vibration sensors to the inner wall surfaces of the existing wind turbine blades, which is very costly. Furthermore, with the technology described in Patent Document 1, it is necessary to transmit vibration data from inside the wind turbine blade, which is a rotating body, to an external damage detection unit, and building such a transmission system is very costly. Therefore, there is a demand for technology that can build a system for detecting abnormalities in wind turbine blades at low cost.

[0005] The present invention has been made in consideration of the above, and aims to provide an abnormality determination device, an abnormality determination system, an abnormality determination method, and an abnormality determination program that can be used to build a system for determining abnormalities in wind turbine blades at low cost. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the object, the abnormality determination device of the present invention comprises a processor that determines abnormalities in wind turbine blades, and the processor comprises an image acquisition unit that acquires a plurality of consecutive captured images in chronological order that are generated by capturing images of the wind turbine blades, a position identification unit that identifies a specific position on the wind turbine blade within the captured images, a frequency calculation unit that calculates the natural frequency of the wind turbine blade by frequency analysis based on the specific position identified for each of the plurality of captured images, and an abnormality determination unit that determines abnormalities in the wind turbine blades based on the natural frequency.

[0007] In addition, in the abnormality determination device according to the present invention, in the above invention, the frequency calculation unit calculates the natural frequencies of each of a plurality of modes, and the abnormality determination unit determines an abnormality in the wind turbine blade based on the natural frequencies of each of the plurality of modes.

[0008] Furthermore, the abnormality determination device according to the present invention, in the above invention, further includes a memory unit that stores relationship information indicating the relationship between the abnormal position of the wind turbine blade for each of the different modes that has been calculated in advance and the amount of change in the natural frequency, and the abnormality determination unit calculates the amount of change in the natural frequency calculated by the frequency calculation unit for each of the modes, and identifies the abnormal position of the wind turbine blade based on the amount of change in each of the natural frequencies for each of the modes and the relationship information.

[0009] Furthermore, in the abnormality determination device according to the present invention, in the above invention, the relationship information is provided for each different rate of decrease in bending stiffness of the wind turbine blade, and the abnormality determination unit identifies the rate of decrease in bending stiffness of the wind turbine blade based on the amount of change in each natural frequency for each mode and the relationship information for each rate of decrease in bending stiffness.

[0010] In the abnormality determination device according to the present invention, in the above invention, the position specifying unit specifies the position of the tip of the wind turbine blade in the captured image as the specific position.

[0011] The abnormality determination system of the present invention comprises an imaging device that images wind turbine blades and generates a plurality of successive captured images in chronological order, and an abnormality determination device having a processor that determines abnormalities in the wind turbine blades, the processor comprising an image acquisition unit that acquires the plurality of captured images, a position identification unit that identifies specific positions on the wind turbine blades within the captured images, a frequency calculation unit that calculates the natural frequency of the wind turbine blades by frequency analysis based on the specific positions identified for each of the plurality of captured images, and an abnormality determination unit that determines abnormalities in the wind turbine blades based on the natural frequency.

[0012] The abnormality determination method according to the present invention is an abnormality determination method executed by a processor of an abnormality determination device, and includes the steps of acquiring a plurality of successive captured images in chronological order that are generated by capturing images of a wind turbine blade, identifying a specific position on the wind turbine blade within the captured images, calculating the natural frequency of the wind turbine blade by frequency analysis based on the specific position identified for each of the plurality of captured images, and determining whether there is an abnormality in the wind turbine blade based on the natural frequency.

[0013] The abnormality determination program of the present invention is an abnormality determination program that causes a computer to execute the steps of acquiring a plurality of consecutive captured images in chronological order that are generated by capturing images of wind turbine blades, identifying specific positions on the wind turbine blades within the captured images, calculating the natural frequency of the wind turbine blades by frequency analysis based on the specific positions identified for each of the plurality of captured images, and determining whether there is an abnormality in the wind turbine blades based on the natural frequency. [Effects of the Invention]

[0014] According to the abnormality determination device, abnormality determination system, abnormality determination method, and abnormality determination program of the present invention, a system for determining abnormalities in wind turbine blades can be constructed at low cost. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 1 is a diagram showing an object of abnormality determination by an abnormality determination system according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating the configuration of a wind turbine blade. [Figure 3] FIG. 3 is a diagram illustrating the configuration of a wind turbine blade. [Figure 4] FIG. 4 is a block diagram showing the configuration of the abnormality determination system. [Figure 5] FIG. 5 is a flowchart showing the abnormality determination method. [Figure 6] FIG. 6 is a diagram illustrating step S3. [Figure 7] FIG. 7 is a diagram illustrating step S4. [Figure 8] FIG. 8 is a diagram illustrating step S4. [Figure 9] FIG. 9 is a diagram illustrating step S4. [Figure 10] FIG. 10 is a diagram illustrating steps S5 and S6. [Figure 11] FIG. 11 is a diagram illustrating step S7. [Figure 12] FIG. 12 is a diagram illustrating step S7. DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, a mode for carrying out the present invention (hereinafter referred to as an embodiment) will be described with reference to the drawings. Note that the present invention is not limited to the embodiment described below. Furthermore, in the description of the drawings, the same parts are given the same reference numerals.

[0017] [Target of abnormality judged by the abnormality judgement system] 1 is a diagram showing an object to be determined as abnormal by an abnormality determination system 1 according to an embodiment. In FIG. 1, the vertical axis is the Z axis, and two axes perpendicular to the Z axis are the X axis and the Y axis, respectively. First, before describing the configuration of the abnormality determination system 1, the object of abnormality determination by the abnormality determination system 1 will be described. 1 shows a wind power generation facility 100. The wind power generation facility 100 includes a main tower 101, a nacelle 102, a hub 103, and wind turbine blades 104.

[0018] The main tower 101 is a columnar structure with a substantially circular cross section that extends upward from the ground surface. The nacelle 102 is attached to the top of the main tower 101 so as to be rotatable around the Z axis (rotatable within a horizontal plane (XY plane)).

[0019] The hub 103 is journaled to the nacelle 102 via a main shaft (not shown) that extends along a horizontal plane (XY plane). A plurality of wind turbine blades 104 are provided, each extending radially from hub 103 and catching wind. The wind turbine blades 104 are targets for abnormality determination by abnormality determination system 1 according to this embodiment. The detailed configuration of the wind turbine blade 104 will be explained in the section "Configuration of Wind Turbine Blade" below.

[0020] Although not specifically shown, the nacelle 102 houses the above-mentioned main shaft, a gearbox that increases the rotation speed of the main shaft to the rotation speed required for the generator, and a generator that generates electricity using the rotational force increased by the gearbox. The electricity generated by the generator is transmitted to the power grid through conductors inside the main tower 101.

[0021] [Wind turbine blade configuration] 2 and 3 are diagrams illustrating the configuration of the wind turbine blade 104. Specifically, Fig. 2 is a diagram of the wind turbine blade 104 as seen from the side that catches wind. Note that the lower end of the wind turbine blade 104 in Fig. 2 is the part that connects to the hub 103. Fig. 3 is a cross-sectional view taken along line III-III in Fig. 2. As shown in FIG. 3, the wind turbine blade 104 is composed of a shell 105, a spar cap 106, and a share web 107.

[0022] The shell 105 is arranged to cover most of the surface of the wind turbine blade 104 and determines the outer shape of the wind turbine blade 104. The shell 105 is mainly made of glass fiber reinforced composite material (hereinafter referred to as GFRP), carbon fiber reinforced composite material (hereinafter referred to as CFRP), or a combination of both.

[0023] The spar cap 106 and shear web 107 are arranged over substantially the entire length of the wind turbine blade 104 in the longitudinal direction (the vertical direction in FIG. 2, the direction perpendicular to the paper surface in FIG. 3). When looking at a cross section of the wind turbine blade 104, the long side (the horizontal direction in FIG. 3) is called the edge direction, while the short side (the vertical direction in FIG. 1) is called the flap direction. The spar cap 106 serves to increase the strength and rigidity of the wind turbine blade 104 in the flap direction. On the other hand, the shear web 107 is designed to support shear loads and bending loads in the flap direction. GFRP and CFRP are mainly used for this spar cap 106, and as wind turbines become larger, CFRP, which has excellent rigidity and strength, is increasingly being used. The interior of the shell 105, other than the spar cap 106 and the share web 107, is hollow enough for a person to stand and work.

[0024] Examples of abnormalities in the wind turbine blades 104 include damage caused by lightning strikes, erosion or cracks on the surface of the shell 105, and deterioration of the components inside the shell 105.

[0025] [Configuration of abnormality determination system] FIG. 4 is a block diagram showing the configuration of the abnormality determination system 1. As shown in FIG. The abnormality determination system 1 is a system that determines an abnormality in a wind turbine blade 104. The abnormality determination system 1 includes an imaging device 2, a display device 3, and an abnormality determination device 4, as shown in FIG.

[0026] The imaging device 2 includes an imaging element such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor), and captures images of the subject, the wind turbine blade 104, under the control of the abnormality determination device 4 to generate a plurality of captured images that are consecutive in time series. For ease of explanation, the captured images generated by the imaging device 2 will be referred to as blade images below. That is, the imaging device 2 captures images of the wind turbine blade 104 to generate moving images. Note that the imaging device 2 can be configured as a general-purpose camera rather than a special high-speed camera, because it is expected that the natural frequency of the wind turbine blade 104 will be low, at a few Hz, and that the amount of deflection will also be large.

[0027] 1, the imaging device 2 is placed on the ground near the wind power generation facility 100 so as to be able to capture an image of the wind turbine blades 104 as a subject. Note that the location of the imaging device 2 is not limited to the ground, and it may be placed in another location, such as the outer surface of the main tower 101, as long as it is able to capture an image of the wind turbine blades 104 as a subject.

[0028] The imaging device 2 is connected to the abnormality determination device 4 so as to be able to communicate with it, and outputs the generated blade images to the abnormality determination device 4. Note that the communication between the imaging device 2 and the abnormality determination device 4 may be wireless communication or wired communication.

[0029] The display device 3 is configured with a display using a liquid crystal or organic EL (Electro Luminescence) display, etc., and displays a predetermined image under the control of the abnormality determination device 4.

[0030] The abnormality determination device 4 determines an abnormality in the wind turbine blade 104 based on a plurality of blade images (moving images) generated by the imaging device 2. As shown in FIG. 4, the abnormality determination device 4 includes an input unit 41, a storage unit 42, and a processor 43.

[0031] The input unit 41 is configured using operation devices such as a mouse, a keyboard, and a touch panel, and receives user operations. The input unit 41 then outputs an operation signal to the processor 43 in response to the user operations.

[0032] The storage unit 42 stores various programs (including the abnormality determination program according to the present invention) executed by the processor 43, a plurality of blade images (moving images) output from the imaging device 2, and data required when the processor 43 performs processing, etc. An example of the data required when the processor 43 performs processing is related information. Details of the related information will be explained in the "Abnormality Determination Method" section below.

[0033] The processor 43 is realized by a controller such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) executing various programs stored in the storage unit 42, and controls the operation of the entire abnormality determination device 4. The processor 43 is not limited to a CPU or an MPU, and may be configured by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array).

[0034] The processor 43 has the functions of an image acquisition unit, a position identification unit, a vibration frequency calculation unit, and an abnormality determination unit according to the present invention. The details of the functions of the processor 43 as the captured image acquisition unit, position identification unit, vibration frequency calculation unit, and abnormality determination unit will be explained in the "Abnormality Determination Method" section below.

[0035] [Abnormality determination method] Next, an abnormality determination method executed by the processor 43 will be described. FIG. 5 is a flowchart showing the abnormality determination method. First, the processor 43 controls the operation of the imaging device 2, causes the imaging device 2 to capture images, and generates a plurality of time-series consecutive blade images (moving images) of the wind turbine blades 104 (step S1). Note that step S1 captures images for a predetermined period of time, and is performed, for example, only once a day. In the following description, it is assumed that step S1 (the cycle of steps S1 to S6 that are repeatedly executed) is performed only once a day.

[0036] After step S1, the processor 43 acquires a plurality of blade images (moving images) that are consecutive in time series and that are generated by the imaging device 2 capturing images of the wind turbine blades 104 in step S1 (step S2). That is, the processor 43 has a function as a captured image acquisition unit according to the present invention.

[0037] It is also possible to provide a storage unit in the imaging device 2, store all of the plurality of blade images generated in the imaging of step S1 in the storage unit, and have the processor 43 acquire all of the plurality of blade images collectively in step S2. It is also possible for the processor 43 to acquire the blade images for each frame generated in the imaging of step S1 sequentially in step S2.

[0038] After step S2, the processor 43 identifies a specific position of the wind turbine blade 104 in each of the blade images acquired in step S2 (step S3). That is, the processor 43 has a function as a position identification unit according to the present invention.

[0039] In this embodiment, three wind turbine blades 104 are provided, as shown in Fig. 1. Therefore, the processor 43 identifies specific positions on each of the three wind turbine blades 104 in step S3.

[0040] Fig. 6 is a diagram for explaining step S3. Specifically, Fig. 6 shows one frame of a blade image F1. In this embodiment, as shown in Fig. 6, the tip position P0 of the wind turbine blade 104 is adopted as the specific position according to the present invention. There are three wind turbine blades 104. Therefore, the tip positions P0 of the wind turbine blades 104 are distinguished as tip positions P1 to P3 (Fig. 6). In Fig. 6, for ease of explanation, the tip position P1 is represented by a hollow circle symbol, the tip position P2 is represented by a hollow square symbol, and the tip position P3 is represented by a solid diamond symbol.

[0041] Specifically, in step S3, the processor 43 makes the contrast between the wind turbine blade 104 in the blade image F1 and the background of the wind turbine blade 104 more pronounced by image processing such as binarization, and identifies the positions P1 to P3 of the tip of each wind turbine blade 104 by image processing such as edge detection.

[0042] Here, the tip positions P1 to P3 identified in step S3 are pixel positions (X coordinate value, Y coordinate value) within the blade image F1. The X coordinate value of the pixel positions is the value on the X axis, which is the horizontal axis shown in Fig. 6. The Y coordinate value of the pixel positions is the value on the Y axis, which is the vertical axis shown in Fig. 6.

[0043] The method for identifying the positions P1 to P3 of the tips of the wind turbine blades 104 is not limited to the above-mentioned method, and the following method may also be used, for example. The processor 43 identifies the positions P1 to P3 of the tips of each wind turbine blade 104 in the blade image F1 by image recognition using a trained model (image recognition using AI (Artificial Intelligence)). The trained model is a model obtained by using a plurality of captured images of each wind turbine blade 104 taken at various angles as training images, and by performing machine learning (for example, deep learning) on ​​the positions of the tips of each wind turbine blade 104 based on the plurality of training images.

[0044] The processing of steps S4 to S7 described below is actually performed for each wind turbine blade 104, in other words, for each of the tip positions P1 to P3 identified for all of the multiple blade images acquired in step S2. However, for ease of explanation, the processing of steps S4 to S7 described below will be explained focusing on one wind turbine blade 104, in other words, focusing on tip position P0 of any of tip positions P1 to P3 identified for all of the multiple blade images acquired in step S2.

[0045] After step S3, the processor 43 calculates the natural frequency of the wind turbine blade 104 by frequency analysis based on the tip positions P0 identified for all of the multiple blade images acquired in step S2 (step S4). That is, the processor 43 has a function as a frequency calculation unit according to the present invention.

[0046] 7 to 9 are diagrams illustrating step S4. Specifically, FIG. 7 is a three-axis graph, with the X and Y axes of the blade image and the frame number of the blade image, in which the tip positions P1 to P3 are arranged in chronological order (in the order of the frames of the blade image). FIG. 8(a) is a two-axis graph, with the X axis of the blade image and the frame number of the blade image, in which the X coordinate values ​​of the tip positions P1 to P3 are arranged in chronological order (in the order of the frames of the blade image). FIG. 8(b) is a two-axis graph, with the Y axis of the blade image and the frame number of the blade image, in which the Y coordinate values ​​of the tip positions P1 to P3 are arranged in chronological order (in the order of the frames of the blade image). Note that in FIGS. 7 and 8, as in FIG. 6, the tip position P1 is represented by a hollow circle, the tip position P2 is represented by a hollow square, and the tip position P3 is represented by a solid diamond. FIG. 9 is a graph showing the results of the frequency analysis performed in step S4.

[0047] 7 and 8, the tip position P0 vibrates sequentially in a time series. The waveform of the vibration at this tip position P0 includes not only vibration caused by the rotation of the wind turbine blade 104 but also vibration caused by an abnormality (damage) of the wind turbine blade 104. Therefore, in this embodiment, in step S4, the processor 43 calculates the natural frequency of the wind turbine blade 104 that is susceptible to the influence of an abnormality (damage) of the wind turbine blade 104 by performing frequency analysis on the vibration waveform at the tip position P0 identified for each of the multiple blade images acquired in step S2.

[0048] Specifically, in step S4, the processor 43 performs a fast Fourier transform on the vibration waveform at the tip position P0 identified for each of the multiple blade images acquired in step S2 to obtain a power spectrum, and calculates the natural frequency of the wind turbine blade 104 from the power spectrum. In this embodiment, the processor 43 calculates the natural frequencies of multiple modes. Figure 9 illustrates an example in which the natural frequency of the first mode is calculated to be "0.673 Hz", the natural frequency of the second mode is calculated to be "2.280 Hz", the natural frequency of the third mode is calculated to be "5.261 Hz", the natural frequency of the fourth mode is calculated to be "9.641 Hz", and the natural frequency of the fifth mode is calculated to be "15.44 Hz".

[0049] After step S4, the processor 43 determines whether there is an abnormality in the wind turbine blade 104 based on the natural frequencies of the multiple modes calculated in step S4 (steps S5 to S7). That is, the processor 43 has a function as an abnormality determination unit according to the present invention.

[0050] Fig. 10 is a diagram illustrating steps S5 and S6. Specifically, Fig. 10 is a graph in which the natural frequency values ​​for each mode calculated in step S4 in the repeatedly executed cycle of steps S1 to S6 are arranged in the order of the cycle (in the order of the number of days elapsed since the start of measurement). In Fig. 10, the natural frequency value of the first mode is represented by a triangle symbol, the natural frequency value of the second mode is represented by a white square symbol, and the natural frequency value of the third mode is represented by a white circle symbol.

[0051] First, the processor 43 calculates the amount of change in the natural frequency for each mode (step S5). Specifically, in step S5, the processor 43 calculates, for each mode, the amount of change by which the natural frequency calculated in step S4 falls from a reference value that serves as a reference for the natural frequency. Examples of the reference value include the natural frequency calculated in step S4 in the first cycle (first day) of the repeatedly executed cycles of steps S1 to S6, or a natural frequency that is set in advance by simulation or the like. In other words, the reference value is the natural frequency when there is no damage to the wind turbine blades 104.

[0052] After step S5, the processor 43 determines whether or not any of the changes in the natural frequencies of the modes calculated in step S5 is equal to or greater than a specific threshold (step S6). Note that the specific threshold may be a different value for each mode, or may be a common value for all modes. If it is determined that all of the changes in the natural frequency for each mode are less than the specific threshold value (step S6: No), the processor 43 determines that there is no abnormality in the wind turbine blade 104, and returns to step S1. On the other hand, if it is determined that any of the changes in the natural frequency for each mode is equal to or greater than a specific threshold (step S6: Yes), the processor 43 determines that the wind turbine blade 104 is abnormal, and proceeds to step S7.

[0053] The example of Fig. 10 shows a case where, when the number of days elapsed since the start of measurement is "60 days," any of the amounts of change in the natural frequencies of the first mode, second mode, and third mode is equal to or greater than a specific threshold, and the wind turbine blade 104 is determined to be abnormal. In other words, the example of Fig. 10 shows a case where, when the number of days elapsed since the start of measurement is "0 day" to "59 days," all of the amounts of change in the natural frequencies of the first mode, second mode, and third mode are less than a specific threshold, and the wind turbine blade 104 is determined to be normal.

[0054] In step S7, the processor 43 determines that the wind turbine blade 104 is abnormal, and identifies the abnormal position and the rate of decrease in bending rigidity of the wind turbine blade 104 based on the amount of change in the natural frequency for each mode calculated in step S5.

[0055] 11 and 12 are diagrams illustrating step S7. Specifically, Fig. 11 is a graph showing the relationship between the amount of change [Hz] in the natural frequency for each mode and the abnormal position [m] of the wind turbine blade 104 when the rate of change [%] in the bending rigidity of the wind turbine blade 104 is 50 [%]. The abnormal position [m] shown on the horizontal axis of Fig. 11 indicates the position in the longitudinal direction of the wind turbine blade 104 where an abnormality has occurred, with the side connected to the hub 103 being set as 0 [m]. Fig. 12 is a graph showing the relationship between the amount of change [Hz] in the natural frequency for each mode and the rate of decrease [%] in the bending rigidity of the wind turbine blade 104 when the abnormal position of the wind turbine blade 104 is 50 [m].

[0056] First, the amount of change in the natural frequency and the rate of decrease in the bending rigidity of the wind turbine blade 104 have the relationship shown in FIG. Specifically, as shown in Fig. 12, the amount of change in the natural frequency increases as the rate of decrease in the bending stiffness of the wind turbine blade 104 increases. The amount of change in the natural frequency also differs depending on the mode. In particular, the amount of change in the natural frequency increases as the mode becomes higher. That is, the amount of change in the natural frequency in a higher mode is more sensitive to the rate of decrease in the bending stiffness of the wind turbine blade 104.

[0057] The amount of change in the natural frequency and the abnormal position of the wind turbine blade 104 have the relationship shown in FIG. Specifically, as shown in Fig. 11, the amount of change in the natural frequency varies depending on the abnormal position of the wind turbine blade 104. The amount of change in the natural frequency also varies depending on the mode. In particular, the amount of change in the natural frequency varies more significantly with changes in the abnormal position of the wind turbine blade 104 as the mode becomes higher.

[0058] 11 corresponds to the relationship information according to the present invention. The relationship information is stored in the storage unit 42. The relationship information is provided for each different bending stiffness reduction rate [%] of the wind turbine blade 104.

[0059] Then, in step S7, the processor 43 identifies the abnormal position and the rate of decrease in bending rigidity of the wind turbine blade 104 based on the relationship information for each different rate of decrease in bending rigidity [%] of the wind turbine blade 104 stored in the memory unit 42 and the amount of change in the natural frequency for each mode calculated in step S5.

[0060] Specifically, step S7 will be described using an example in which the amount of change in the natural frequency of the first to third modes is calculated in step S5. In this case, the amount of change in the natural frequency of the first mode is set to "0.028 Hz," the amount of change in the natural frequency of the second mode is set to "0.103 Hz," and the amount of change in the natural frequency of the third mode is set to "0.152 Hz" (FIG. 11).

[0061] In step S7, the processor 43 extracts an abnormal position of the wind turbine blade 104 corresponding to "0.028 Hz," which is the amount of change in the natural frequency of the first mode, based on the relationship information corresponding to a predetermined decrease rate of bending stiffness stored in the storage unit 42. In the example of FIG. 11, the abnormal position is 50 [m]. Similarly, the processor 43 extracts an abnormal position of the wind turbine blade 104 corresponding to "0.103 Hz," which is the amount of change in the natural frequency of the second mode, based on the relationship information corresponding to a predetermined decrease rate of bending stiffness stored in the storage unit 42. In the example of FIG. 11, the abnormal positions are three positions: 12 [m], 50 [m], and 78 [m]. Similarly, the processor 43 extracts an abnormal position of the wind turbine blade 104 corresponding to "0.152 Hz," which is the amount of change in the natural frequency of the third mode, based on the relationship information corresponding to a predetermined decrease rate of bending stiffness stored in the storage unit 42. In the example of Fig. 11, the abnormal positions are five positions: 10 [m], 30 [m], 50 [m], 64 [m], and 88 [m]. Then, the processor 43 identifies 50 [m], which is a position that matches in all modes, as the abnormal position from among the extracted abnormal positions. Note that if there is no position that matches in all modes from among the extracted abnormal positions, the above-mentioned process is repeated using relationship information with different bending stiffness reduction rates until a matching position is found. Note that if there is a position that matches in all modes from among the extracted abnormal positions, the bending stiffness reduction rate of the used relationship information is identified as the bending stiffness reduction rate of the wind turbine blade 104.

[0062] Then, the processor 43 causes the display device 3 to display an image (e.g., a message) showing the wind turbine blade 104 that has been determined to have an abnormality, as well as the abnormal position and bending rigidity reduction rate of the wind turbine blade 104 identified in step S7, as a result of the determination of an abnormality in the wind turbine blade 104.

[0063] According to the present embodiment described above, the following effects are achieved. In the abnormality determination device 4 according to this embodiment, the processor 43 acquires a plurality of blade images that are consecutive in time series and are generated by capturing images of the wind turbine blades 104, and sequentially identifies the tip positions P0 of the wind turbine blades 104 in the blade images. The processor 43 also calculates the natural frequency of the wind turbine blade 104 by frequency analysis, based on the tip positions P0 identified for each of the plurality of blade images. The processor 43 then determines whether there is an abnormality in the wind turbine blade 104, based on the natural frequency. Therefore, even when determining an abnormality in the wind turbine blades 104 in an existing wind power generation facility 100, the complicated work of attaching a vibration sensor to the inner wall surface of the wind turbine blades 104 as in the technology described in Patent Document 1 is not necessary, and the simple work of installing the imaging device 2 outside the wind turbine power generation facility 100 is sufficient. Furthermore, since the imaging device 2 that generates input data (blade images) for the abnormality determination device 4 to determine an abnormality in the wind turbine blades 104 can be installed outside the wind turbine power generation facility 100, there is no need to transmit the input data (vibration data) from inside the wind turbine blades 104, which are rotating bodies, to the outside as in the technology described in Patent Document 1, and the input data can be transmitted outside the wind turbine power generation facility 100. Therefore, according to the abnormality determination device 4 of this embodiment, the abnormality determination system 1 that determines abnormalities in the wind turbine blades 104 can be constructed at low cost.

[0064] Incidentally, the technology described in Patent Document 1 uses the magnitude of the vibration level to determine whether the wind turbine blade 104 is abnormal. However, the magnitude of the vibration level is a parameter that fluctuates greatly due to various factors such as wind speed, blade angle, wind direction, and rotation speed. For this reason, it is difficult to accurately determine whether the wind turbine blade 104 is abnormal when the magnitude of the vibration level is used. Furthermore, the technology described in Patent Document 1 uses the peak frequency, which is the peak value of the natural frequency of the wind turbine blade 104, to determine whether the wind turbine blade 104 is abnormal. However, the natural frequency has higher modes, and the mode in which the change in frequency appears varies depending on the location and degree of the abnormality (damage). For this reason, it is difficult to accurately determine whether the wind turbine blade is abnormal by simply monitoring the peak value of the natural frequency.

[0065] The inventors of the present application have noticed that there is a relationship shown in Fig. 11 between the amount of change in the natural frequency for each mode and the rate of change in the bending stiffness of the wind turbine blade 104, and that there is a relationship shown in Fig. 12 between the amount of change and the abnormal position of the wind turbine blade 104. Then, the processor 43 identifies the abnormal position and the rate of decrease in bending stiffness of the wind turbine blade 104 based on the amount of change in each natural frequency for each mode and the relationship information stored in the storage unit 42. That is, the abnormality determination device 4 according to this embodiment determines an abnormality in the wind turbine blade 104 based on the natural frequencies of each of the multiple modes, and therefore can accurately determine an abnormality in the wind turbine blade 104. Furthermore, it is possible to identify not only the presence or absence of an abnormality in the wind turbine blade 104, but also the position of the abnormality and the rate of decrease in bending rigidity of the wind turbine blade 104, and subsequently take effective measures.

[0066] Furthermore, in the abnormality determination device 4 according to this embodiment, the position P0 of the tip of the wind turbine blade 104 is adopted as the specific position according to the present invention. Therefore, there is no need to apply markings or the like to the outer surface of the wind turbine blade 104 in order to identify a specific position of the wind turbine blade 104, and the tip position P0, which is the specific position, can be easily identified.

[0067] (Other embodiments) Although the embodiments for carrying out the present invention have been described above, the present invention should not be limited to only the above-described embodiments. In the above-described embodiment, the position P0 of the tip of the wind turbine blade 104 is used as the specific position according to the present invention, but this is not limitative and other positions of the wind turbine blade 104 may be used. Furthermore, the specific position according to the present invention is not limited to one, and may be multiple.

[0068] The flow of the abnormality determination method (FIG. 5) explained in the above embodiment may be changed within a consistent range. For example, in step S7, the abnormal position and the rate of decrease in bending rigidity of the wind turbine blade 104 are identified as the process for determining whether there is an abnormality in the wind turbine blade 104, but this is not limiting, and it is also possible to identify only whether there is an abnormality. Also, for example, step S6 may be omitted, and step S7 may be executed after step S5. [Explanation of symbols]

[0069] 1. Abnormality detection system 2. Imaging device 3 Display device 4 Abnormality determination device 41 Input section 42 Storage section 43 processors 100 Wind power generation facilities 101 Main tower 102 Nacelle 103 Hub 104 Windmill Blades 105 Shell 106 Super Cap 107 Share Web F1 Blade Images P0~P3 Tip position

Claims

1. A processor is provided to determine abnormalities in the wind turbine blades, The processor: an image acquisition unit that acquires a plurality of time-series consecutive captured images generated by capturing images of the wind turbine blades; a position specifying unit that specifies a specific position on the wind turbine blade within the captured image; a frequency calculation unit that calculates a natural frequency of the wind turbine blade by frequency analysis based on the specific positions identified for each of the plurality of captured images; an abnormality determination unit that determines an abnormality in the wind turbine blade based on the natural frequency.

2. The frequency calculation unit Calculating the natural frequencies of each of the plurality of modes; The abnormality determination unit The abnormality determination device according to claim 1 , wherein an abnormality in the wind turbine blade is determined based on the natural frequencies of the plurality of modes.

3. a storage unit configured to store relationship information indicating a relationship between an abnormal position of the wind turbine blade for each of the modes different from each other and a change amount of the natural frequency, the relationship being calculated in advance; The abnormality determination unit 3. The abnormality determination device according to claim 2, wherein an amount of change in the natural frequency calculated by the frequency calculation unit is calculated for each of the modes, and an abnormal position of the wind turbine blade is identified based on the amount of change in each of the natural frequencies for each of the modes and the relationship information.

4. The related information is a bending stiffness reduction rate for each of the wind turbine blades, The abnormality determination unit The abnormality determination device according to claim 3 , wherein the rate of decrease in bending rigidity of the wind turbine blade is identified based on the amount of change in each of the natural frequencies for each of the modes and the relationship information for each rate of decrease in bending rigidity.

5. The position identification unit The abnormality determination device according to claim 1 , wherein the specific position is determined to be a position of a tip of the wind turbine blade in the captured image.

6. an imaging device that captures images of the wind turbine blades and generates a plurality of time-series consecutive captured images; an abnormality determination device having a processor for determining an abnormality in the wind turbine blades, The processor: a captured image acquisition unit that acquires the plurality of captured images; a position specifying unit that specifies a specific position on the wind turbine blade within the captured image; a frequency calculation unit that calculates a natural frequency of the wind turbine blade by frequency analysis based on the specific positions identified for each of the plurality of captured images; an abnormality determination unit that determines an abnormality in the wind turbine blade based on the natural frequency.

7. An abnormality determination method executed by a processor of an abnormality determination device, acquiring a plurality of time-series consecutive captured images generated by capturing images of the wind turbine blades; identifying a specific position on the wind turbine blade within the captured image; calculating a natural frequency of the wind turbine blade by frequency analysis based on the specific positions identified for each of the plurality of captured images; and determining an abnormality in the wind turbine blade based on the natural frequency.

8. acquiring a plurality of time-series consecutive captured images generated by capturing images of the wind turbine blades; identifying a specific position on the wind turbine blade within the captured image; calculating a natural frequency of the wind turbine blade by frequency analysis based on the specific positions identified for each of the plurality of captured images; and determining an abnormality in the wind turbine blade based on the natural frequency.

Images