Fan Blade Monitoring Method, Device, Computer Equipment and Storage Medium

Abstract

The present application relates to a method, device, computer device and storage medium for monitoring a wind turbine blade. The method includes the following steps: using the undamaged state signal of the wind turbine blade as the reference signal for subsequent state monitoring, and obtaining the undamaged state signal through a plurality of piezoelectric sensing neurons distributed on the inner surface of the wind turbine blade; obtaining the current health state signal of the wind turbine blade, and performing a correlation analysis on the current health state signal and the undamaged state signal to obtain the corresponding damage coefficient at the position of each piezoelectric sensing neuron; judging the existence and position of the damage of the wind turbine blade according to the change and amplitude of the corresponding damage coefficient at the position of each piezoelectric sensing neuron. The above-mentioned method, device, computer device and storage medium for monitoring a wind turbine blade can judge the damage position and degree of the wind turbine blade according to the principle that the closer to the damage in actual calculation, the greater the measured damage coefficient, and can monitor the health state of the wind turbine blade as needed at any time and anywhere.

Description

Technical Field

[0001] The present application relates to the field of structural health monitoring, and particularly to a method and device for monitoring a wind turbine blade, a computer device, and a storage medium. Background Art

[0002] As one of the most important external components of a wind turbine during operation, the wind turbine blade is in a harsh environment for a long time. It is subject to impacts and erosion caused by extreme weather such as sandstorms, hailstorms, typhoons, and lightning, and also bears the fatigue load brought by periodic rotation. These numerous external conditions and operating mechanisms inevitably cause some structural damages to the wind turbine blade, such as cracks, debonding, holes, and fiber fractures. The existence of these damages will gradually damage the wind turbine blade, the built-in motor, and the tower barrel, and even cause accidents such as the overall collapse and falling of the blade, resulting in greater economic losses and threats to personal safety.

[0003] Currently, the methods for condition monitoring and fault diagnosis of wind turbine blades at home and abroad mainly include infrared thermal imaging detection, X-ray method, acoustic emission detection, fiber optic sensor detection, and vibration detection method. These methods have certain effects on specific ranges and specific damages, but there are many limitations.

[0004] The most fundamental limitation is that the existing methods are all detections rather than monitoring. Detection requires manual operation of instruments, and corresponding operations and identifications are needed to detect the existence of damages. Most detection methods need to be tested during shutdown or even on land. Summary of the Invention

[0005] Based on this, in view of the above technical problems, it is necessary to provide a method and device for monitoring a wind turbine blade, a computer device, and a storage medium, which can monitor the health status as needed at any time and anywhere.

[0006] In a first aspect, the present application provides a method for monitoring a wind turbine blade, the method including:

[0007] Taking the undamaged state signal of the wind turbine blade as the reference signal for subsequent status monitoring, and the undamaged state signal is obtained through a plurality of piezoelectric sensor neurons distributed on the inner surface of the wind turbine blade;

[0008] Obtaining the current health status signal of the wind turbine blade, and performing a correlation analysis on the current health status signal and the undamaged state signal to obtain the corresponding damage coefficient at each position of the piezoelectric sensor neuron;

[0009] Judging the existence and position of the damage of the wind turbine blade according to the change and amplitude of the corresponding damage coefficient at each position of the piezoelectric sensor neuron.

[0010] In one embodiment, the undamaged state signal is a relative undamaged state signal.

[0011] In one embodiment, before using the undamaged state signal of the wind turbine blade as the reference signal for subsequent condition monitoring, the following steps are also included:

[0012] Obtain the reference time-domain signal collected by the piezoelectric sensor neurons, which is generated by exciting the wind turbine blade with an electromagnetic exciter.

[0013] Perform Fourier transform on the reference time-domain signal to obtain the reference frequency-domain signal, and use the frequency-domain data of the reference frequency-domain signal as the undamaged state signal of the wind turbine blade. The frequency-domain data of the reference frequency-domain signal is the reference spectrum signal.

[0014] In one embodiment, obtaining the current health state signal of the wind turbine blade includes:

[0015] Truncate the current time-domain signal according to the position where the voltage value of the obtained current time-domain signal first reaches 0.01.

[0016] Obtain the truncated current time-domain signal after window function and filtering processing.

[0017] Obtain the effective time-domain signal segment of the current time-domain signal. The effective time-domain signal segment is obtained by identifying the tail end of the pulse signal emitted by the electromagnetic exciter through the signal attenuation rate, then intercepting 20% of the length of the pulse signal backward, and finally performing noise reduction.

[0018] Perform Fourier transform on the effective time-domain signal segment to obtain the current frequency-domain signal, and use the frequency-domain data of the current frequency-domain signal as the current health state signal of the wind turbine blade. The frequency-domain data of the current frequency-domain signal is the current spectrum signal.

[0019] In one embodiment, after obtaining the corresponding damage coefficient at each position of the piezoelectric sensor neurons, the following steps are also included:

[0020] Calibrate the judgment threshold of the damage coefficient according to the corresponding damage coefficient values at different positions of the piezoelectric sensor neurons.

[0021] In one embodiment, judging the existence and location of damage to the wind turbine blade according to the change and amplitude of the corresponding damage coefficients at each position of the piezoelectric sensor neurons includes:

[0022] Judge whether there is damage to the wind turbine blade according to the difference between the damage coefficient at each position of the piezoelectric sensor neuron and the damage coefficient judgment threshold.

[0023] Judge the position of the damage to the fan blade according to the change of the difference value.

[0024] In a second aspect, the present application provides a fan blade monitoring device, which includes:

[0025] A first acquisition module, configured to use the undamaged state signal of the fan blade as a reference signal for subsequent state monitoring, and the undamaged state signal is acquired by a plurality of piezoelectric sensing neurons distributed on the inner surface of the fan blade;

[0026] A second acquisition module, configured to acquire the current health state signal of the fan blade, and perform a correlation analysis on the current health state signal and the undamaged state signal to obtain the corresponding damage coefficient at the position of each piezoelectric sensing neuron;

[0027] A comparison module, configured to judge the existence and position of the damage to the fan blade according to the change and amplitude of the corresponding damage coefficient at the position of each piezoelectric sensing neuron.

[0028] In a third aspect, the present application provides a computer device, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0029] In a fourth aspect, the present application provides a computer-readable storage medium, storing a computer program, and when the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0030] In a fifth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0031] The above-mentioned fan blade monitoring method, device, computer device and storage medium pre-enter the undamaged state signal of the fan blade as the subsequent state monitoring signal of the fan blade, acquire the current health state signal of the fan blade when needed, and compare the current health state signal with the undamaged state signal. Finally, according to the comparison result, the damage coefficients at the positions of different piezoelectric sensing neurons are obtained, and then according to the principle that the closer to the damage in the actual calculation, the larger the measured damage coefficient, the damage position and degree of the fan blade are judged. Using this method, the health state of the fan blade can be monitored on demand at any time and anywhere. Description of the Drawings

[0032] Figure 1 It is a flowchart of the fan blade monitoring method for the first embodiment;

[0033] Figure 2 It is a flowchart of the fan blade monitoring method for the second embodiment;

[0034] Figure 3 Flow chart of the fan blade monitoring method for the third embodiment;

[0035] Figure 4 Flow chart of the fan blade monitoring method for the fourth embodiment;

[0036] Figure 5 Module diagram of the fan blade monitoring device for one embodiment;

[0037] Figure 6 Schematic diagram of the structure of a piezoelectric sensing neuron for one embodiment;

[0038] Figure 7 Schematic diagram of the arrangement of cross-sectional sensing neurons of a fan blade for one embodiment;

[0039] Figure 8 Schematic diagram of the time-domain and frequency-domain signals of a sensing neuron far from damage for one embodiment;

[0040] Figure 9 Schematic diagram of the time-domain and frequency-domain signals of a sensing neuron close to damage for one embodiment;

[0041] Figure 10 Three-dimensional columnar schematic diagram of the damage coefficient for one embodiment;

[0042] Figure 11 Internal structure diagram of a computer device for one embodiment. Detailed implementation manners

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the scope of protection of the present application.

[0044] As Figure 1 shown, in one embodiment, a fan blade monitoring method includes the following steps:

[0045] Step S110: Use the undamaged state signal of the fan blade as the reference signal for subsequent state monitoring, and the undamaged state signal is obtained by multiple piezoelectric sensing neurons distributed on the inner surface of the fan blade.

[0046] Specifically, the un-damaged state signal is a relative un-damaged state signal rather than an absolute one. Before obtaining the un-damaged state signal of the wind turbine blade, it is necessary to determine the neural network cross-section unit according to the size, overall structure, position of the stiffeners and the position of the existing damage of the wind turbine blade. For example, Figure 7 As shown, six groups of piezoelectric sensing neurons are respectively arranged around the wind turbine blade. After determining the key cross-section, the positions for pasting the piezoelectric sensing neurons are designed according to the structural strength inside the cross-section. For example, Figure 6 As shown, the outer surface of the piezoelectric sensing neuron is a flexible printed circuit, and the internal sensor is an active piezoelectric sheet sensor. The two are coupled through bonding and welding. The active piezoelectric sheet sensor can receive wide-band vibration and wave signals from sound waves to ultrasonic waves, and obtain rich structural information. The flexible printed circuit can effectively protect and seal the sensor, and the cost of this piezoelectric sensing neuron is extremely low.

[0047] Step S120: Obtain the current health state signal of the wind turbine blade, and perform a correlation analysis on the current health state signal and the un-damaged state signal to obtain the corresponding damage coefficients at the positions of each piezoelectric sensing neuron.

[0048] Specifically, after using the relative un-damaged signal of the wind turbine blade as the reference signal for subsequent condition monitoring, the signals obtained by subsequent monitoring of the wind turbine blade are all current health state signals, which are used to compare and analyze with the un-damaged state signal, and the damage coefficients at the monitoring positions of each piezoelectric sensing neuron are judged according to the difference and the magnitude of the difference between the two.

[0049] Step S130: Judge the existence and position of the damage of the wind turbine blade according to the change and amplitude of the corresponding damage coefficients at the positions of each piezoelectric sensing neuron.

[0050] Specifically, as Figure 10 shown, after actual calculation, it is known that the damage coefficient monitored at the position of the piezoelectric sensing neuron closer to the damage is larger. If the damage of the wind turbine blade is between multiple piezoelectric sensing neurons, the position and degree of the damage of the wind turbine blade can be judged according to the magnitude of the damage coefficients monitored at the positions of the piezoelectric sensing neurons.

[0051] For the above-mentioned wind turbine blade monitoring method, by pre-recording the un-damaged state signal of the wind turbine blade as the subsequent condition monitoring signal of the wind turbine blade, obtaining the current health state signal of the wind turbine blade when needed, comparing the current health state signal with the un-damaged state signal, finally obtaining the damage coefficients at the positions of different piezoelectric sensing neurons according to the comparison result, and then judging the damage position and degree of the wind turbine blade according to the principle that the closer to the damage, the larger the measured damage coefficient obtained in the actual calculation. Using this method, the health state of the wind turbine blade can be monitored at any time as needed.

[0052] As Figure 2 shown, in this embodiment, the undamaged state signal of the fan blade is used as the reference signal for subsequent condition monitoring, and the following steps are also included before:

[0053] Step S210, obtaining the reference time-domain signal collected by the piezoelectric sensor neuron, which is generated by exciting the fan blade with an electromagnetic exciter.

[0054] Specifically, the electromagnetic exciter will excite the fan blade to generate vibration, and the piezoelectric sensor neuron will be affected by the vibration and collect the reference time-domain signal of the vibration. Among them, the electromagnetic exciter can actively excite the fan blade according to the set amplitude pulse and the set time interval. The piezoelectric sensor neuron can collect the pulse vibration and the feedback signal when the whole fan blade is plucked according to the set sampling rate, sampling time and sampling interval.

[0055] Step S220, performing Fourier transform on the reference time-domain signal to obtain the reference frequency-domain signal, and using the frequency-domain data of the reference frequency-domain signal as the undamaged state signal of the fan blade. The frequency-domain data of the reference frequency-domain signal is the reference spectrum signal.

[0056] Specifically, the undamaged state signal of the fan blade is a spectrum signal. Therefore, the time-domain signal obtained by the piezoelectric sensor neuron needs to be converted. First, the time-domain signal is converted into a frequency-domain signal by Fourier transform, and the frequency-domain data of the frequency-domain signal is the spectrum signal.

[0057] As Figure 3 shown, in this embodiment, obtaining the current health state signal of the fan blade specifically includes the following steps:

[0058] Step S121, truncating the current time-domain signal according to the position where the voltage value of the obtained current time-domain signal first reaches 0.01.

[0059] Specifically, for the time-domain signal matrix V(t, i) collected by the piezoelectric sensor neuron, where i is the number of the piezoelectric sensor neuron, because there is a bias voltage in it, the voltage signal in the starting segment will have an invalid signal segment that gradually returns from the negative full scale to the zero scale line. Therefore, it is necessary to truncate the signal by identifying the position where the signal first reaches the voltage value of 0.01 to avoid the influence of the bias voltage, so as to obtain V1(t, i).

[0060] Step S122, obtaining the truncated current time-domain signal after window function and filtering processing.

[0061] Specifically, in order to reduce the noise impact of signal mutations on time-domain and frequency-domain analysis, a Tukey Window function needs to be added to the truncated signal. The description of the window function is shown in Equation (1). The length of the window function is the number of rows of V1(t,i), that is, the length of the signal after time-domain truncation. V2(t,i) is obtained by multiplying V1(t,i) by Tukeywin

[0062]

[0063] In the formula, W(x) represents the window function, and r represents the windowing range.

[0064] In addition, in order to reduce the influence of surrounding electrical equipment on the collected signal and the influence brought by the low-frequency operation of the fan blades. It should be noted that the operating frequency of most electrical equipment is 50Hz, and the operating frequency of large equipment is lower than 50Hz. A high-pass filter needs to be set to filter out the signal components below 50Hz, and the filtered time-domain signal V3(t,i) is obtained.

[0065] Step S123: Obtain the effective time-domain signal segment of the current time-domain signal. The effective time-domain signal segment is obtained by identifying the tail end of the pulse signal emitted by the electromagnetic exciter through the signal attenuation rate, then intercepting 20% of the pulse signal length backward, and finally performing noise reduction.

[0066] Specifically, in order to improve the calculation efficiency, the tail end of the pulse signal emitted by the electromagnetic exciter is automatically identified according to the signal attenuation rate, and then 20% of the pulse signal length is intercepted backward. After intercepting the effective signal, a filter is also needed for noise reduction. Here, a high-pass filter is used to filter out environmental wind noise, motor operating frequency interference, and other low-frequency noises. Finally, the effective time-domain signal segment V4(t,i) is obtained, which not only ensures the effectiveness and continuity of the signal but also reduces the subsequent calculation amount.

[0067] Step S124: Perform Fourier transform on the effective time-domain signal segment to obtain the current frequency-domain signal, and use the frequency-domain data of the current frequency-domain signal as the current health status signal of the fan blade. The frequency-domain data of the current frequency-domain signal is the current spectrum signal.

[0068] Specifically, for the obtained effective time-domain signal segment, a fast Fourier transform needs to be performed. Denote the lowest passing frequency of the high-pass filter as f start , and a spectrum signal F1(f,i) with a frequency range of f start -f s / 2 is obtained, where f s is the sampling frequency of the acquisition card.

[0069] It should be noted that since the pulse frequencies emitted by different electromagnetic exciters are different, it is necessary to intercept the spectrum so as to make subsequent monitoring faster and more accurate. For example, in this embodiment, the pulse frequency band emitted by the electromagnetic exciter mainly concentrates in the frequency range of f start -f s / 8. Therefore, an effective pulse spectrum F2(f,i) is obtained by intercepting the spectrum of the pulse frequency band.

[0070] As Figure 4 shown, in this embodiment, the presence and location of the damage to the fan blade are judged according to the change and amplitude of the corresponding damage coefficient at the position of each piezoelectric sensing neuron, which specifically includes the following steps:

[0071] Step S131, judge whether there is damage to the fan blade according to the difference between the damage coefficient at the position of each piezoelectric sensing neuron and the damage coefficient judgment threshold.

[0072] Specifically, the calculation of the correlation damage coefficient array (Damage Index) is shown in formula (2)

[0073]

[0074] In the formula, D1(f,i) and D2(f,i) are the frequency spectral densities of the undamaged and damaged fan blades respectively, which are transformed from F2(f,i). The integral upper and lower limits f1 and f2 respectively correspond to the frequency upper and lower limits of the effective pulse spectrum. i represents the number of the sensing neuron.

[0075] Step S132, judge the location of the damage to the fan blade according to the change of the difference.

[0076] Specifically, as Figure 8 and Figure 9 shown, by combining the time-domain and frequency-domain signals of the sensing neurons far from the damage and the time-domain and frequency-domain signals of the sensing neurons close to the damage, a three-dimensional columnar diagram of the damage coefficient as shown in Figure 10 can be obtained. This three-dimensional diagram marks the damage coefficient according to the position of the piezoelectric sensing neuron, and it can be seen from the figure that the closer to the damage position, the greater the damage coefficient monitored by the sensing neuron. Therefore, the damage position on the fan blade can be judged according to the change law and magnitude of the damage coefficient.

[0077] In this embodiment, after obtaining the corresponding damage coefficient at the position of each piezoelectric sensing neuron, it further includes: calibrating the damage coefficient judgment threshold according to the size of the corresponding damage coefficient at the positions of different piezoelectric sensing neurons.

[0078] Specifically, the above-mentioned correlation coefficient DI is a damage index that fluctuates numerically between 0 and 1. The higher its value, the worse the correlation, that is, the greater the difference between the two measurements. This spectral shift is caused by damage existing on the fan blade. The determination of damage requires a calibration process. Specifically, artificial damage needs to be created to determine the lowest threshold for the change in the correlation coefficient. Here, △DI is used to determine the existence of damage, and the calibration of damage is obtained by measuring the fan blade sample pasted on the outside.

[0079] The specific calibration process is as follows: Paste a fan blade sample of a specified size between two piezoelectric sensor neurons. The signal measured before pasting is recognized as "Baseline", and the signal measured after pasting is "Monitor". The correlation coefficient DI of the two sensor neurons near the artificially pasted damage obtained through the above calculation process s (n) and DI s (n + 1), and the smaller value min{DI s (n), DI s (n + 1)} is used as the correlation coefficient threshold for determining the existence of damage.

[0080] The process of damage determination is as follows: First, measure the artificially recognized standard signal "Baseline", then measure the "Monitor" signal periodically, continuously, or as needed at any time, and obtain the damage coefficients monitored at the positions of each piezoelectric sensor neuron. When the damage coefficients of one or more sensor neurons exceed the threshold DI s , it is determined that damage exists.

[0081] If only one sensor neuron exceeds the damage threshold DI s , the damage location is near the sensor neuron. If multiple sensor neurons in adjacent sensing sections exceed the damage coefficient threshold, assuming the difference is ΔDI n , a circle with a radius of ΔDI n *r and a linearly increasing shape is formed centered on the sensor neuron according to the ratio of the damage coefficient difference. The common area obtained is the location where damage exists. If the sensing units of multiple sections exceed the threshold, then for every two sections, the above process is executed to determine the locations where multiple damages exist.

[0082] As Figure 5 shown, in one embodiment, a fan blade monitoring device includes a first acquisition module 110, a second acquisition module 120, and a comparison module 130.

[0083] The first acquisition module 110 is configured to use the undamaged state signal of the fan blade as the reference signal for subsequent state monitoring. The undamaged state signal is acquired through multiple piezoelectric sensor neurons distributed on the inner surface of the fan blade.

[0084] A second acquisition module 120, configured to acquire a current health status signal of the fan blade, and perform a correlation analysis on the current health status signal and the undamaged status signal to obtain a corresponding damage coefficient at each piezoelectric sensor neuron position.

[0085] A comparison module 130, configured to determine the presence and location of damage to the fan blade according to the change and amplitude of the corresponding damage coefficients at each piezoelectric sensor neuron position.

[0086] In this embodiment, the first acquisition module 110 is specifically configured to acquire a reference time-domain signal collected by the piezoelectric sensor neurons, where the reference time-domain signal is generated by exciting the fan blade with an electromagnetic exciter; perform a Fourier transform on the reference time-domain signal to obtain a reference frequency-domain signal, and use the frequency-domain data of the reference frequency-domain signal as the undamaged status signal of the fan blade, and the frequency-domain data of the reference frequency-domain signal is the reference spectrum signal.

[0087] In this embodiment, the second acquisition module 120 is specifically configured to truncate the current time-domain signal according to the position where the voltage value of the acquired current time-domain signal first reaches 0.01; acquire the truncated current time-domain signal after window function and filtering processing; acquire the effective time-domain signal segment of the current time-domain signal, where the effective time-domain signal segment is obtained by identifying the tail end of the pulse signal emitted by the electromagnetic exciter through the signal attenuation rate and then intercepting 20% of the length of the pulse signal backward; perform a Fourier transform on the effective time-domain signal segment to obtain a current frequency-domain signal, and use the frequency-domain data of the current frequency-domain signal as the current health status signal of the fan blade, and the frequency-domain data of the current frequency-domain signal is the current spectrum signal.

[0088] In this embodiment, the comparison module 130 is specifically configured to determine whether there is damage to the fan blade according to the difference between the damage coefficient at each piezoelectric sensor neuron position and the damage coefficient judgment threshold; determine the location of the damage to the fan blade according to the change in the difference.

[0089] In this embodiment, the fan blade monitoring device further includes a calibration module, configured to calibrate the damage coefficient judgment threshold according to the corresponding damage coefficient magnitudes at different piezoelectric sensor neuron positions.

[0090] In one embodiment, a computer device is provided. The computer device may be an intelligent terminal, and its internal structure diagram may be as Figure 11As shown. The computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a method for monitoring fan blades.

[0091] Those skilled in the art can understand that Figure 11 the structure shown in is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0092] In one embodiment, a computer device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, it implements the steps in the above method embodiments.

[0093] In one embodiment, a computer storage medium stores a computer program. When the computer program is executed by the processor, it implements the steps in the above method embodiments.

[0094] In one embodiment, a computer program product or a computer program is provided. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps in the above method embodiments.

[0095] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0096] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0097] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. A method for monitoring a wind turbine blade, characterized in that, The method includes: Taking the undamaged state signal of the wind turbine blade as the reference signal for subsequent condition monitoring, and obtaining the undamaged state signal through a plurality of piezoelectric sensing neurons distributed on the inner surface of the wind turbine blade; Obtaining the current health state signal of the wind turbine blade, including: truncating the current time-domain signal according to the position where the voltage value of the obtained current time-domain signal first reaches 0.01; obtaining the truncated current time-domain signal after window function and filtering processing; obtaining the effective time-domain signal segment of the current time-domain signal, and the effective time-domain signal segment is obtained by identifying the tail end of the pulse signal emitted by the electromagnetic exciter through the signal attenuation rate and then intercepting 20% of the length of the pulse signal backward; performing Fourier transform on the effective time-domain signal segment to obtain the current frequency-domain signal, and taking the frequency-domain data of the current frequency-domain signal as the current health state signal of the wind turbine blade, and the frequency-domain data of the current frequency-domain signal is the current spectrum signal; and performing correlation analysis on the current health state signal and the undamaged state signal to obtain the corresponding damage coefficient at each position of the piezoelectric sensing neuron; Judging the existence and position of the damage of the wind turbine blade according to the change and amplitude of the corresponding damage coefficient at each position of the piezoelectric sensing neuron, wherein the closer the piezoelectric sensing neuron is to the damage, the greater the damage coefficient monitored at its position. If the damage of the wind turbine blade is between multiple piezoelectric sensing neurons, the position and degree of the damage of the wind turbine blade can be judged according to the size of the damage coefficient monitored at the position of the piezoelectric sensing neuron.

2. The method for monitoring a wind turbine blade according to claim 1, characterized in that, The undamaged state signal is a relative undamaged state signal.

3. The method for monitoring a wind turbine blade according to claim 1, characterized in that, Before taking the undamaged state signal of the wind turbine blade as the reference signal for subsequent condition monitoring, it further includes: Obtaining the reference time-domain signal collected by the piezoelectric sensing neuron, and the reference time-domain signal is generated by exciting the wind turbine blade through the electromagnetic exciter; Performing Fourier transform on the reference time-domain signal to obtain the reference frequency-domain signal, and taking the frequency-domain data of the reference frequency-domain signal as the undamaged state signal of the wind turbine blade, and the frequency-domain data of the reference frequency-domain signal is the reference spectrum signal.

4. The method for monitoring a wind turbine blade according to claim 3, characterized in that, After obtaining the corresponding damage coefficient at each position of the piezoelectric sensing neuron, it further includes: Calibrating the judgment threshold of the damage coefficient according to the size of the corresponding damage coefficient at different positions of the piezoelectric sensing neuron.

5. The method for monitoring a wind turbine blade according to claim 4, characterized in that, Judging the existence and position of the damage of the wind turbine blade according to the change and amplitude of the corresponding damage coefficient at each position of the piezoelectric sensing neuron, including: Judging whether there is damage to the wind turbine blade according to the difference between the damage coefficient at each position of the piezoelectric sensing neuron and the damage coefficient judgment threshold; Judging the position of the damage of the wind turbine blade according to the change of the difference.

6. A device for monitoring a wind turbine blade, characterized in that, The device includes: The first acquisition module is used to take the undamaged state signal of the wind turbine blade as the reference signal for subsequent condition monitoring, and obtain the undamaged state signal through a plurality of piezoelectric sensing neurons distributed on the inner surface of the wind turbine blade; A second acquisition module, configured to acquire the current health status signal of the wind turbine blade, including: truncating the current time-domain signal according to the position where the voltage value of the acquired current time-domain signal first reaches 0.01; acquiring the truncated current time-domain signal after window function and filtering processing; acquiring the effective time-domain signal segment of the current time-domain signal, where the effective time-domain signal segment is obtained by identifying the end of the pulse signal emitted by the electromagnetic exciter through the signal attenuation rate and then intercepting 20% of the pulse signal length backward; performing Fourier transform on the effective time-domain signal segment to obtain the current frequency-domain signal, and using the frequency-domain data of the current frequency-domain signal as the current health status signal of the wind turbine blade, where the frequency-domain data of the current frequency-domain signal is the current spectrum signal; and performing correlation analysis on the current health status signal and the undamaged status signal to obtain the corresponding damage coefficients at each piezoelectric sensor neuron position. A comparison module, configured to judge the existence and position of the damage of the wind turbine blade according to the change and amplitude of the corresponding damage coefficients at each piezoelectric sensor neuron position. Among them, the damage coefficient monitored at the piezoelectric sensor neuron position closer to the damage is larger. If the damage of the wind turbine blade is between multiple piezoelectric sensor neurons, the position and degree of the damage of the wind turbine blade can be judged according to the size of the damage coefficients monitored at the piezoelectric sensor neuron positions.

7. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 5.

Images