Wind power blade web detection method and related device

Abstract

The invention provides a wind power blade web detection method and a related device, and relates to the technical field of detection. The method comprises the steps that terahertz time domain data of a web to be detected are obtained to serve as first sample data, terahertz time domain data of a metal layer are obtained to serve as first reference data, and the terahertz time domain data are obtained through a probe; the distance between the probe and a first target surface, facing the probe, of the to-be-detected web is equal to the distance between the probe and a second target surface, facing the probe, of the metal layer; performing Fourier transform on the first sample data and the first reference data to obtain second sample data and second reference data; according to the second sample data and the second reference data, performing deconvolution on the second sample data to obtain third sample data; performing noise reduction processing on the third sample data to obtain target sample data; and according to the target sample data, generating a cross-section image of the to-be-detected web. In this way, a cross-sectional image capable of performing high-precision display on levels and defects can be obtained.

Description

Technical Field

[0001] The present application relates to the field of detection technologies, and more particularly, to a method and related device for detecting the web of a wind turbine blade. Background Art

[0002] In the safety inspection of wind power equipment, the health status of the blade is crucial. It is the main component for receiving wind energy and a determining factor for normal and stable operation. Moreover, the bonding quality of the web of the wind turbine blade directly affects the service life and operation stability of the blade. Due to the influence of process complexity and invisibility, web defects are difficult to detect in daily inspections. However, even minor defects, such as debonding, bubbles, delamination, or microcracks, will affect the bonding quality. If these hidden defects are not detected and repaired in a timely manner at an early stage, they may not only lead to fatigue damage and a decline in structural performance of the blade, but may also cause serious operation failures and even safety accidents in extreme cases. Therefore, how to detect the web of a wind turbine blade has become a technical problem that needs to be urgently solved by those skilled in the art. Summary of the Invention

[0003] The embodiments of the present application provide a method and related device for detecting the web of a wind turbine blade. By performing deconvolution processing on the first sample data of the web to be detected collected by a terahertz data acquisition system using the first reference data of the metal layer collected by the terahertz data acquisition system, and then performing noise reduction, a cross-sectional image of the blade web that can be clearly imaged can be obtained, and the layers and defects can be displayed with high precision.

[0004] The embodiments of the present application can be implemented as follows:

[0005] In a first aspect, the embodiments of the present application provide a method for detecting the web of a wind turbine blade, the method including:

[0006] Obtaining terahertz time-domain data of the web to be detected as the first sample data, and obtaining terahertz time-domain data of the metal layer as the first reference data, wherein the terahertz time-domain data is obtained by a probe, and the distance between the probe and the first target surface of the web to be detected facing the probe is equal to the distance between the probe and the second target surface of the metal layer facing the probe;

[0007] Performing Fourier transform on the first sample data and the first reference data to obtain second sample data and second reference data;

[0008] Performing deconvolution on the second sample data according to the second sample data and the second reference data to obtain third sample data;

[0009] Performing noise reduction processing on the third sample data to obtain target sample data;

[0010] Generate a cross-sectional image of the web to be detected according to the target sample data.

[0011] In a second aspect, an embodiment of the present application provides a wind turbine blade web detection device, which includes:

[0012] A data acquisition module, configured to acquire terahertz time-domain data of the web to be detected as first sample data, and acquire terahertz time-domain data of the metal layer as first reference data, wherein the terahertz time-domain data is obtained by a probe, and the distance between the probe and the first target surface of the web to be detected facing the probe is equal to the distance between the probe and the second target surface of the metal layer facing the probe;

[0013] A deconvolution module, configured to perform Fourier transform on the first sample data and the first reference data to obtain second sample data and second reference data;

[0014] The deconvolution module is further configured to perform deconvolution on the second sample data according to the second sample data and the second reference data to obtain third sample data;

[0015] A noise reduction module, configured to perform noise reduction processing on the third sample data to obtain target sample data;

[0016] An image generation module, configured to generate a cross-sectional image of the web to be detected according to the target sample data.

[0017] In a third aspect, an embodiment of the present application provides an electronic device, including a processor and a memory, where the memory stores machine-executable instructions that can be executed by the processor, and the processor can execute the machine-executable instructions to implement the wind turbine blade web detection method described in the foregoing embodiments.

[0018] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the wind turbine blade web detection method described in the foregoing embodiments.

[0019] The wind turbine blade web detection method and related device provided by the embodiments of the present application first obtain the terahertz time-domain data of the web to be detected as the first sample data, and obtain the terahertz time-domain data of the metal layer as the first reference data. The terahertz time-domain data is obtained by a probe, and the distance between the probe and the first target surface of the web to be detected facing the probe is equal to the distance between the probe and the second target surface of the metal layer facing the probe. Then, Fourier transform is performed on the first sample data and the first reference data to obtain the second sample data and the second reference data. After that, deconvolution is performed on the second sample data according to the second sample data and the second reference data to obtain the third sample data. Next, noise reduction processing is performed on the third sample data to obtain the target sample data. Finally, a cross-sectional image of the web to be detected is generated according to the target sample data. In this way, deconvolution processing is performed on the first sample data of the web to be detected collected by the terahertz data acquisition system in combination with the first reference data of the metal layer collected by the terahertz data acquisition system, and then noise reduction is performed, so that a cross-sectional image that can clearly image the blade web can be obtained, and the layers and defects can be displayed with high precision. Description of the Drawings

[0020] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 Schematic diagram of the communication between the terahertz data acquisition system and the electronic device provided by the embodiments of the present application;

[0022] Figure 2 Schematic diagram of data acquisition using the terahertz data acquisition system provided by the embodiments of the present application;

[0023] Figure 3 Schematic diagram of the first target surface and the second target surface provided by the embodiments of the present application;

[0024] Figure 4 Block diagram of the electronic device provided by the embodiments of the present application;

[0025] Figure 5 Flowchart of the wind turbine blade web detection method provided by the embodiments of the present application;

[0026] Figure 6 Schematic diagram of the time-domain waveform corresponding to the first reference data;

[0027] Figure 7 The time-domain waveform schematic diagram and tomography image corresponding to the first sample data;

[0028] Figure 8 is Figure 5 The flow schematic diagram of the sub-steps included in step S130 in;

[0029] Figure 9 is Figure 5 The flow schematic diagram of the sub-steps included in step S140 in;

[0030] Figure 10 is Figure 9 The flow schematic diagram of the sub-steps included in sub-step S142 in;

[0031] Figure 11 is Figure 6 The time-domain waveform schematic diagram corresponding to the sixth sample data corresponding to the first reference data shown in, and the tomography image corresponding to the sixth sample data;

[0032] Figure 12 The block schematic diagram of the wind turbine blade web detection device provided by the embodiment of the present application.

[0033] Icon: 100 - Terahertz data acquisition system; 110 - Platform control system; 120 - Power module; 130 - Power control and drive board; 140 - Depth detection control device; 150 - Probe; 160 - Time-domain signal acquisition card; 170 - Wifi module; 200 - Electronic device; 210 - Memory; 220 - Processor; 230 - Communication unit; 300 - Wind turbine blade web detection device; 310 - Data acquisition module; 320 - Deconvolution module; 330 - Noise reduction module; 340 - Image generation module. Detailed implementation manners

[0034] 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. The components of the embodiments of the present application usually described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations.

[0035] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application claimed, but merely represents the selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0036] It should be noted that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0037] Currently, the following non-destructive testing techniques are used to detect the webs of wind turbine blades.

[0038] Ultrasonic testing: By transmitting ultrasonic beams into the interior of the blade web, detecting the reflection of the sound waves, and then identifying defects. This method has relatively high operating requirements, and the probe needs to be in contact with the surface of the web, which has requirements for the detection location.

[0039] Radiographic testing: By scanning the blade with X-rays to capture internal images. Although this method has high precision, it has high requirements for the site and the scanning process is slow.

[0040] Infrared thermal imaging testing: By using infrared imaging technology to detect abnormal conditions caused by defects. This method can quickly scan a large area, but it is easily affected by environmental temperature and light factors, and it is difficult to detect deep defects.

[0041] In current wind power safety inspections, the daily inspection requirements for wind turbine blades are increasing, which is crucial for the safe operation of the entire wind power equipment. The inventors of this application have found through research that the above-mentioned detection methods have the following limitations and defects.

[0042] Ultrasonic testing requires the use of a coupling agent to contact the probe with the surface of the web to ensure the effective transmission of signals. However, in a complex high-altitude environment (such as strong wind, temperature difference, and humidity changes), using a coupling agent is not only difficult to operate but also has an adverse impact on the accuracy of detection. Using air coupling can avoid the coupling agent, but since the propagation speed of air is much lower than that of the coupling agent, the signal attenuation is severe, which will significantly reduce the detection depth and accuracy, and it is difficult to detect deep defects.

[0043] Although radiographic testing has high penetration power, due to its dependence on ionizing radiation, its radiation characteristics pose safety hazards. The protection measures for ground and high-altitude environments are complex, and limited by cost, complexity, and portability, it is difficult to use in daily inspections.

[0044] Infrared detection mainly relies on temperature differences for defect identification. However, the fan blades are affected by various meteorological factors such as solar radiation, wind speed, and humidity in the high-altitude environment, resulting in a large change in the temperature field and making it difficult to obtain accurate detection results under complex meteorological conditions.

[0045] To solve a series of problems existing in the existing high-altitude and ground detection technologies for fan blades, such as the need for a couplant in ultrasonic detection, the radiation hazards and difficulty in portability of ray detection, and the susceptibility of infrared detection to environmental factors, the embodiments of the present application provide a method and related device for detecting the web of a wind turbine blade, which can perform a health assessment on the web of the wind turbine blade, penetrate the blade surface and the composite material layer, identify the bonding levels and defects of the blade web, and eliminate hardware and environmental noise through algorithm optimization, improving the level, defect resolution, and robustness while having flexibility, and solving the problems of difficulty and insufficient accuracy in online detection of blades in the air and on the ground by the existing detection methods.

[0046] The following will describe in detail some embodiments of the present application with reference to the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0047] When it is necessary to detect the web to be detected, the terahertz time-domain data of the web to be detected can be obtained as the first sample data, and the terahertz time-domain data of the metal layer can be obtained as the first reference data. Then, the first sample data can be deconvolved with the first reference data and then denoised to generate a blade web cross-sectional image that can display the levels and defects with high precision.

[0048] Among them, the first sample data and the first reference data can be obtained by a terahertz data acquisition system, and the deconvolution processing, denoising processing, and image generation can be processed by an electronic device. Optionally, the terahertz data acquisition system can be used to collect data in any way, and the data can be sent to the electronic device.

[0049] As a possible implementation, as Figure 1 shown, the terahertz data acquisition system 100 is communicatively connected to the electronic device 200, and the terahertz data acquisition system 100 can directly send the collected terahertz time-domain data to the electronic device 200.

[0050] The terahertz data acquisition system 100 includes a probe 150 for obtaining terahertz time-domain data. When detecting the web to be detected, as Figure 2 shown, the terahertz data acquisition system 100 can be located in the inner cavity of the blade, that is, the probe 150 is located in the inner cavity; the probe 150 can be located on the web to be detected (i.e., Figure 2above the blade web in it), and the distance between the two can be described as the height of the probe 150. As Figure 3 shown, the web to be detected includes a first target surface facing the probe, and the metal layer includes a second target surface facing the probe. The distances from the first target surface and the second target surface to the probe are equal, which facilitates deconvolution processing of the first sample data corresponding to the web to be detected based on the first reference data corresponding to the metal layer.

[0051] As Figure 1 shown, the terahertz data acquisition system 100 may further include a platform control system 110, a power module 120, a power control and drive board 130, a depth control device 140, a time-domain signal acquisition card 160, and a wifi module 170 that are communicatively connected. The platform control system 110 can control the power control and drive board 130 through the power module 120, so as to control the working states of the depth control device 140, the probe 150, etc. The probe 150 can be a transceiver integrated probe, which can be specifically determined according to actual requirements. The depth control device 140 is used to control the detection depth of the probe 150. As Figure 2 shown, the detection distance (i.e., depth) into the web is the detection depth. During actual detection, the specific detection depth can be determined according to actual requirements. The wifi module 170 can be used to send the terahertz time-domain data collected by the time-domain signal acquisition card 160 to the electronic device 200.

[0052] In this embodiment, the platform control system 110 can perform mobile control on the terahertz data acquisition system 100 (a mobile acquisition system). As Figure 2 and Figure 3 shown, the control system performs data acquisition in the X-axis direction, controls the detection depth of the probe 150 through the depth control device 140, and uses the probe 150 to collect terahertz time-domain spectral signals in the Y-axis direction. Among them, when generating an image, when collecting the first sample data in the above manner, the z value and the y value corresponding to the first sample data are the same.

[0053] The above terahertz data acquisition system 100 can perform mobile operations inside the wind turbine blade cavity, so as to adjust the position of the terahertz signal probe, and can collect terahertz time-domain data in real time, which is convenient for sampling in the narrow blade cavity.

[0054] Please refer to Figure 4 , Figure 4The block diagram of the electronic device 200 provided by the embodiment of the present application is shown. The electronic device 200 may be, but is not limited to, a computer, a server, etc. The electronic device 200 includes a memory 210, a processor 220, and a communication unit 230. Each of the memory 210, the processor 220, and the communication unit 230 is directly or indirectly electrically connected to each other to achieve data transmission or interaction. For example, these components may be electrically connected to each other through one or more communication buses or signal lines.

[0055] Among them, the memory 210 is used to store programs or data. The memory 210 may be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc.

[0056] The processor 220 is used to read / write the data or programs stored in the memory 210 and perform corresponding functions. For example, a wind turbine blade web detection device 300 is stored in the memory 210. The wind turbine blade web detection device 300 includes at least one software function module that can be stored in the memory 210 in the form of software or firmware. The processor 220 executes various functional applications and data processing by running the software programs and modules stored in the memory 210, such as the wind turbine blade web detection device 300 in the embodiment of the present application, thereby implementing the wind turbine blade web detection method in the embodiment of the present application.

[0057] The communication unit 230 is used to establish a communication connection between the electronic device 200 and other communication terminals through a network and is used to send and receive data through the network.

[0058] It should be understood that Figure 4 The structure shown is only the structural schematic diagram of the electronic device 200. The electronic device 200 may also include more or fewer components than those shown Figure 4 herein, or have a different configuration from that shown Figure 4 herein. Figure 4 Each of the components shown herein may be implemented by hardware, software, or a combination thereof.

[0059] Please refer to Figure 5 ,Figure 5 It is a schematic flowchart of the method for detecting the web of a wind turbine blade provided by the embodiment of the present application. The method can be applied to the above-mentioned electronic device. The following elaborates in detail on the specific process of the method for detecting the web of a wind turbine blade. In this embodiment, the method may include steps S110 to S150.

[0060] Step S110: Obtain the terahertz time-domain data of the web to be detected as the first sample data, and obtain the terahertz time-domain data of the metal layer as the first reference data.

[0061] In this embodiment, the web to be detected is the web that needs to be health-assessed. The metal layer can be specifically determined according to actual needs. The terahertz time-domain data of the web to be detected can be obtained as the first sample data, and the terahertz time-domain data of the metal layer can be obtained as the first reference data by any means. Among them, the first reference data can be obtained in advance and obtained from the corresponding data storage when the first sample data needs to be processed.

[0062] When collecting the first reference data, the metal layer can be adjusted to an appropriate time-domain peak for reference sampling according to the height position of the probe in the terahertz data acquisition system (i.e., the distance between the probe and the second target surface of the metal layer), so as to obtain the first reference data. The schematic diagram of the time-domain waveform corresponding to the first reference data can be as Figure 6 shown. Among them, when collecting the first reference data, the probing depth used can be determined according to actual needs.

[0063] The terahertz time-domain data of the web to be detected can be collected at the same probe height position, so as to obtain the first sample data. That is, the distance between the probe and the first target surface of the web to be detected facing the probe is equal to the distance between the probe and the second target surface of the metal layer facing the probe. Among them, when collecting the first sample data, the probing depth used can be greater than the total depth of the web to be detected, can be less than the total depth of the web to be detected, or data can be collected both when the probing depth is greater than the total depth of the web to be detected to generate the corresponding image at this time, and when the probing depth is less than the total depth of the web to be detected to generate the corresponding image at this time, which can be specifically determined according to requirements. The time-domain waveform of the obtained first sample data is as shown in a of Figure 7 7. Due to the influence of the superposition of external noise and hardware noise, the signal effect is poor, the imaging of defects is blurred and the hierarchical information cannot be distinguished. The tomographic image corresponding to the time-domain signal with noise (i.e., the first sample data) is as shown in b of 7. It is necessary to eliminate the noise of the signal through a noise reduction algorithm to improve the imaging quality.

[0064] Step S120: Perform Fourier transform on the first sample data and the first reference data to obtain second sample data and second reference data.

[0065] Step S130: Deconvolve the second sample data according to the second sample data and the second reference data to obtain third sample data.

[0066] Deconvolution is performed on the frequency-domain signal. In signal processing, deconvolution is a commonly used technique for recovering the original signal from a signal containing noise or other interference. Therefore, after obtaining the first sample data and the first reference data, the first sample data can be subjected to Fourier transform to convert the first sample data from time-domain data to frequency-domain data, thereby obtaining the second sample data; similarly, the first reference data is subjected to Fourier transform to convert the first reference data from time-domain data to frequency-domain data, thereby obtaining the second parameter data. Then, in combination with the second reference data, the second sample data can be deconvolved, and the second sample data after deconvolution is used as the third sample data.

[0067] Step S140: Denoise the third sample data to obtain target sample data.

[0068] Step S150: Generate a cross-sectional image of the web to be detected according to the target sample data.

[0069] In the case of obtaining the third sample data, the corresponding denoising processing method can be determined according to actual needs for denoising, and the processed third sample data is used as the target sample data, and then a cross-sectional image of the web to be detected is generated based on the target sample data. Among them, the above method can be applied to high-altitude detection or ground detection.

[0070] The above-mentioned wind turbine blade web detection method provided in this embodiment is a non-destructive detection method for wind turbine blade webs based on terahertz spectroscopy, which solves the technical problems such as difficult detection and insufficient accuracy in existing non-destructive detection technologies at high altitudes or on the ground, and aims to accurately identify the bonding levels and defects of wind turbine blade webs. By performing deconvolution and noise reduction filtering on the data collected by terahertz, clear bonding levels and defects are finally obtained.

[0071] Optionally, deconvolution can be performed in the Figure 8 shown manner. Please refer to Figure 8 , Figure 8 which is Figure 5 a schematic flow diagram of the sub-steps included in step S130 in

[0072] Sub-step S131: Determine a target delay value according to the second sample data, and obtain a filtering function based on the target delay value.

[0073] Sub-step S132: Perform deconvolution on the second sample data according to the filtering function, the second sample data, and the second reference data to obtain the third sample data.

[0074] In this embodiment, deconvolution parameters can be set according to the delay data of the sampling frequency to generate a filtering function: where LF and HF represent filtering parameters; t 0 represents a delay value corresponding to the peak position of the second sample data in the delay array, that is, t 0 represents the target delay value, and the target delay value corresponds to the peak in the second sample data; t represents the value in the delay array; represents the filtering function, and h(t) represents the value after the value t is processed by the filtering function.

[0075] Both the first sample data and the first reference data are terahertz time-domain data. After performing Fourier transforms on them respectively, the second sample data sam(f) and the second reference data ref(f) that are frequency-domain data can be obtained. Then, based on the second sample data sam(f) and the second reference data ref(f) that are frequency-domain data, deconvolution is performed on the second sample data sam(f) to obtain the third sample data.

[0076] Optionally, in this embodiment, during the deconvolution process, the second sample data can also be deconvolved according to a preset compensation function, the filtering function, the second sample data, and the second reference data to obtain the third sample data. Among them, the preset compensation function is used for signal enhancement. In this way, the formula for deconvolution is as follows: where Com(f) represents the preset compensation function, and the formula for the preset compensation function is as follows: Com(f) = e -αf , α represents the attenuation factor, which controls the compensation intensity and is used to attenuate high-frequency signals; f represents the value of the second sample data.

[0077] Optionally, noise reduction can be performed in the Figure 9 shown manner. Please refer to Figure 9 Figure 9 is Figure 5 a schematic flowchart of the sub-steps included in step S140 in

[0078] Sub-step S141: Perform band-pass filtering on the third sample data to obtain the fourth sample data.

[0079] Sub-step S142: Obtain the target sample data according to the fourth sample data.

[0080] In this embodiment, after obtaining the third sample data through deconvolution processing, band-pass filtering may be performed on the third sample data to reduce noise and obtain the fourth sample data. The formula for band-pass filtering is as follows: where f H represents the high-frequency end of the filter and is the upper limit frequency at which the signal can pass; f L f H represents the low-frequency end of the filter and is the lower limit frequency at which the signal can pass; n represents the steepness of the filter; H(f) represents the value after the value f is processed by the filtering function, and f represents the value of the third sample data.

[0081] Optionally, in obtaining the fourth sample data, the fourth sample data may be converted from frequency-domain data to time-domain data, and then the converted time-domain data may be directly used as the target sample data, and the cross-sectional image may be generated. In this way, the cross-sectional image can be obtained quickly.

[0082] Optionally, as another possible implementation, the target sample data may be obtained in the manner Figure 10 shown. Please refer to Figure 10 , Figure 10 which is Figure 9 a schematic flow diagram of the sub-steps included in sub-step S142 in . In this embodiment, sub-step S142 may include sub-steps S1421 to S1422.

[0083] Sub-step S1421: Convert the fourth sample data from a frequency-domain signal to a time-domain signal to obtain a fifth sample data.

[0084] Sub-step S1422: Filter the fifth sample data to obtain the target sample data.

[0085] In this embodiment, after obtaining the fourth sample data through band-pass filtering, perform an inverse Fourier transform on the fourth sample data to convert the fourth sample data from a frequency-domain signal to a time-domain signal, obtaining a fifth sample data Rec(t). The fifth sample data Rec(t) is a relatively stable signal data with a stable peak. The fifth sample data may be filtered in combination with the set filtering method, and then the target sample data may be obtained. Among them, the filtering method may be specifically determined according to actual requirements.

[0086] As a possible implementation, the fifth sample data may be filtered in the following manner. First, generate an asymmetric Gaussian kernel according to the fifth sample data: σ xrepresents the standard deviation in the x direction, σ y represents the standard deviation in the y direction, G(x, y) represents an asymmetric Gaussian kernel function, and after substituting values, G(x, y) represents the value generated based on the value at the (x, y) position of the fifth sample data; then, Gaussian filtering is performed on the fifth sample data according to the asymmetric Gaussian kernel to reduce high-frequency noise and smooth the signal. The formula for the Gaussian filtering process is as follows: f out (x, y) represents the value of the Gaussian-filtered data at the (x, y) position, G(i, j) represents the value of the asymmetric Gaussian filtering kernel at the (i, j) position, f(x + i, y + j) represents the value of the fifth sample data at the (x + i, y + j) position; then, Kalman filtering is performed on the fifth sample data after Gaussian filtering to further remove low-frequency noise and errors and provide a high-quality signal, obtaining the sixth sample data. Among them, in Kalman filtering, the following state equation is established: where, x k represents the state vector at time k (i.e., amplitude data), x k-1 represents the state vector at time k - 1, u k represents the position of the terahertz time-domain signal, f(x k-1 , u k ) represents the state transition function, u represents the previous state value, σ 2 represents the peak signal width, w k represents the process noise.

[0087] Optionally, after obtaining the sixth sample data, the maximum and minimum values in the amplitude data of the sixth sample data can be determined, and then, according to a preset target interval, the maximum and minimum values, the amplitude data of the sixth sample data is normalized to obtain the target sample data. The above processing process can be expressed by the following formula:

[0088]

[0089] where, f n (x, y) represents the value of the normalized data at the (x, y) position, f(x, y) is the value of the time-domain signal at the (x, y) position, X max represents the maximum value in the amplitude data of the sixth sample data, X min represents the minimum value in the amplitude data of the sixth sample data, a represents the minimum value of the preset target interval, and b represents the maximum value of the preset target interval.

[0090] Finally, the cross-sectional image of the web to be detected is generated according to the above target sample data. Among them, in the time-domain waveform corresponding to the first reference data, asFigure 6 As shown, the time-domain waveform corresponding to the first sample reference data is as Figure 7 In the case shown in a of the figure, after performing deconvolution, band-pass filtering, and Gaussian Kalman filtering on the first sample reference data, the waveform of the terahertz time-domain signal with noise removed (i.e., the sixth sample data) is as Figure 11 shown in a of the figure. According to threshold normalization imaging, a tomographic image (i.e., the cross-sectional image of the web to be detected) as shown in Figure 11 b of the figure is obtained. From the comparison of the tomographic images before and after noise reduction, it can be seen that the above-mentioned wind turbine blade web detection method can significantly improve the clarity of defects, the structural hierarchy can be clearly highlighted, and the longitudinal accuracy can reach the micron level.

[0091] The above-mentioned wind turbine blade web detection method is a noise reduction algorithm based on terahertz spectroscopy. By performing deconvolution on the first sample data containing noise in combination with the first reference data, and then performing noise reduction processing, the cross-sectional image of the blade web can be clearly imaged, and the hierarchy and defects can be displayed with high precision.

[0092] To execute the corresponding steps in the above-mentioned embodiments and each possible manner, the following provides an implementation manner of a wind turbine blade web detection device 300. Optionally, the wind turbine blade web detection device 300 may adopt the device structure of the above-mentioned Figure 4 shown electronic device 200. Further, please refer to Figure 12 , Figure 12 which is a block diagram of the wind turbine blade web detection device 300 provided by an embodiment of the present application. It should be noted that the basic principle and the technical effects generated by the wind turbine blade web detection device 300 provided in this embodiment are the same as those of the above-mentioned embodiments. For the sake of brief description, for the parts not mentioned in this embodiment, reference may be made to the corresponding content in the above-mentioned embodiments. In this embodiment, the wind turbine blade web detection device 300 may include: a data acquisition module 310, a deconvolution module 320, a noise reduction module 330, and an image generation module 340.

[0093] The data acquisition module 310 is configured to acquire the terahertz time-domain data of the web to be detected as the first sample data, and acquire the terahertz time-domain data of the metal layer as the first reference data. Among them, the terahertz time-domain data is obtained through a probe, and the distance between the probe and the first target surface of the web to be detected facing the probe is equal to the distance between the probe and the second target surface of the metal layer facing the probe.

[0094] The deconvolution module 320 is configured to perform Fourier transform on the first sample data and the first reference data to obtain the second sample data and the second reference data.

[0095] The deconvolution module 320 is further configured to deconvolute the second sample data according to the second sample data and the second reference data to obtain third sample data.

[0096] The noise reduction module 330 is used to perform noise reduction processing on the third sample data to obtain target sample data.

[0097] The image generation module 340 is used to generate a cross-sectional image of the web to be inspected according to the target sample data.

[0098] Optionally, the above modules can be stored in the form of software or firmware. Figure 4 The memory 210 shown in the figure may be fixed in the operating system (OS) of the electronic device 200 and may be Figure 4 Meanwhile, the data and program codes required for executing the above modules may be stored in the memory 210.

[0099] An embodiment of the present application further provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the wind turbine blade web detection method is implemented.

[0100] In summary, the embodiments of the present application provide a method and a related device for detecting the web of a wind turbine blade. First, terahertz time domain data of the web to be detected is obtained as the first sample data, and terahertz time domain data of the metal layer is obtained as the first reference data. The terahertz time domain data is obtained through a probe, and the distance between the probe and a first target surface of the web to be detected facing the probe is equal to the distance between the probe and a second target surface of the metal layer facing the probe; then, the first sample data and the first reference data are Fourier transformed to obtain second sample data and second reference data; thereafter, the second sample data is deconvolved according to the second sample data and the second reference data to obtain third sample data; then, the third sample data is denoised to obtain target sample data; finally, a cross-sectional image of the web to be detected is generated according to the target sample data. In this way, the first sample data of the web to be inspected collected by the terahertz data acquisition system is deconvolved in combination with the first reference data of the metal layer collected by the terahertz data acquisition system, and then noise reduction is performed, so that a cross-sectional image of the blade web can be obtained that can clearly image the image, and the layers and defects can be displayed with high precision.

[0101] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to multiple embodiments of the present application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0102] In addition, each functional module in various embodiments of the present application can be integrated together to form an independent part, or each module can exist alone, or two or more modules can be integrated to form an independent part.

[0103] If the described functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0104] The above are only optional embodiments of the present application and are not used to limit the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting the web of a wind turbine blade, characterized in that: The method comprises: Obtaining terahertz time domain data of the web to be detected as first sample data, and obtaining terahertz time domain data of the metal layer as first reference data, wherein the terahertz time domain data is obtained by a probe, and a distance between the probe and a first target surface of the web to be detected facing the probe is equal to a distance between the probe and a second target surface of the metal layer facing the probe; Performing Fourier transform on the first sample data and the first reference data to obtain second sample data and second reference data; Deconvolute the second sample data according to the second sample data and the second reference data to obtain third sample data; Performing noise reduction processing on the third sample data to obtain target sample data; A cross-sectional image of the web to be inspected is generated according to the target sample data.

2. The method according to claim 1, characterized in that The step of deconvolving the second sample data according to the second sample data and the second reference data to obtain third sample data includes: Determine a target delay value according to the second sample data, and obtain a filter function according to the target delay value, wherein the target delay value corresponds to a peak value in the second sample data; According to the filter function, the second sample data and the second reference data, the second sample data is deconvolved to obtain the third sample data.

3. The method according to claim 2, characterized in that The step of deconvolving the second sample data according to the filter function, the second sample data and the second reference data to obtain the third sample data comprises: According to a preset compensation function, the filter function, the second sample data and the second reference data, the second sample data is deconvolved to obtain the third sample data, wherein the preset compensation function is used for signal enhancement.

4. The method according to claim 2, characterized in that: The performing noise reduction processing on the third sample data to obtain target sample data includes: Performing bandpass filtering on the third sample data to obtain fourth sample data; The target sample data is obtained according to the fourth sample data.

5. The method according to claim 4, characterized in that The step of obtaining the target sample data according to the fourth sample data includes: Convert the fourth sample data from a frequency domain signal to a time domain signal to obtain fifth sample data; The fifth sample data is filtered to obtain the target sample data.

6. The method according to claim 5, characterized in that The filtering the fifth sample data to obtain the target sample data includes: generating an asymmetric Gaussian kernel according to the fifth sample data; Performing Gaussian filtering and Kalman filtering on the fifth sample data according to the asymmetric Gaussian kernel to obtain sixth sample data; The target sample data is obtained according to the sixth sample data.

7. The method according to claim 6, characterized in that The step of obtaining the target sample data according to the sixth sample data includes: Determining the maximum value and the minimum value in the amplitude data of the sixth sample data; According to the preset target interval, the maximum value and the minimum value, the amplitude data of the sixth sample data is normalized to obtain the target sample data.

8. A wind turbine blade web detection device, characterized in that: The device comprises: a data acquisition module, used to obtain terahertz time domain data of the web to be detected as first sample data, and to obtain terahertz time domain data of the metal layer as first reference data, wherein the terahertz time domain data is obtained by a probe, and a distance between the probe and a first target surface of the web to be detected facing the probe is equal to a distance between the probe and a second target surface of the metal layer facing the probe; a deconvolution module, configured to perform Fourier transform on the first sample data and the first reference data to obtain second sample data and second reference data; The deconvolution module is further used to deconvolute the second sample data according to the second sample data and the second reference data to obtain third sample data; A noise reduction module, used for performing noise reduction processing on the third sample data to obtain target sample data; An image generation module is used to generate a cross-sectional image of the web to be detected according to the target sample data.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor can execute the machine executable instructions to implement the wind turbine blade web detection method according to any one of claims 1 to 7.

10. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the wind turbine blade web detection method according to any one of claims 1 to 7 is implemented.

Images