Metal sheet detection method, system and device, computer equipment and storage medium

Abstract

The invention relates to a metal sheet detection method, system and device, computer equipment and a storage medium. The method comprises the following steps: controlling an ultrasonic guided-wave probe to generate an ultrasonic guided-wave signal under the condition that a metal sheet is transmitted to a preset detection area, so that the ultrasonic guided-wave signal is propagated in the metal sheet; acquiring a reflection echo signal of the metal sheet; performing feature analysis on the reflection echo signal to obtain a feature analysis result; and detecting whether the metal sheet has defects or not according to a feature analysis result to obtain a defect detection result of the metal sheet. The method is beneficial to improving the detection efficiency of the metal sheet.

Description

Technical Field

[0001] This application relates to the technical field of non-destructive testing, and particularly to a method, system, device, computer equipment, computer-readable storage medium, and computer program product for detecting metal sheets. Background Art

[0002] Since thin sheets are produced by rolling raw materials, during the production process, small defects in the raw materials can form defects extending along the rolling direction after rolling. These defects will seriously affect the performance and service life of the thin sheets and may even cause safety accidents. Therefore, it is necessary to comprehensively and accurately evaluate the internal quality of the thin sheets.

[0003] At present, there is a problem of low efficiency in the methods for detecting surface and internal defects of metal sheets. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a method, system, device, computer equipment, computer-readable storage medium, and computer program product for detecting metal sheets that can improve the detection efficiency of metal sheets.

[0005] In a first aspect, this application provides a method for detecting a metal sheet, including:

[0006] When the metal sheet is conveyed to a preset detection area, controlling an ultrasonic guided wave probe to generate an ultrasonic guided wave signal so that the ultrasonic guided wave signal propagates in the metal sheet;

[0007] Obtaining the reflected echo signal of the metal sheet;

[0008] Performing feature analysis on the reflected echo signal to obtain a feature analysis result;

[0009] According to the feature analysis result, detecting whether there are defects in the metal sheet to obtain a defect detection result of the metal sheet.

[0010] In one embodiment, the method further includes:

[0011] Performing wavelet transform on the reflected echo signal to obtain a wavelet transform result;

[0012] When the defect detection result indicates that there are defects in the metal sheet, determining the defect position and defect size according to the wavelet transform result;

[0013] Generating and pushing an alarm message according to the defect detection result, defect position, and defect size.

[0014] In an exemplary embodiment, determining the defect position and defect size according to the wavelet transform result includes:

[0015] According to the wavelet transform result, determine the energy of the low-frequency component pulse signal and the wave packet centroid time in the reflected echo signal;

[0016] According to the energy and the wave packet centroid time, determine the defect position and the defect size.

[0017] In an exemplary embodiment, the method further includes:

[0018] Compare the energy with a preset energy threshold. If the energy is greater than or equal to the preset energy threshold, it is determined that there is a defect in the metal sheet.

[0019] In an exemplary embodiment, the method further includes:

[0020] In the case of determining that there is a defect in the metal sheet, generate a grayscale image based on the energy of the low-frequency component pulse signal in the reflected echo signal;

[0021] Analyze the shape characteristics of the target area in the grayscale image to obtain a shape characteristic analysis result, where the target area is the potential defect area;

[0022] Based on the shape characteristic analysis result, determine the defect type.

[0023] In one of the embodiments, perform feature analysis on the reflected echo signal to obtain a feature analysis result, including:

[0024] Perform feature analysis on the reflected echo signal to determine the amplitude of the reflected echo signal. The feature analysis result includes the amplitude of the reflected echo signal;

[0025] According to the feature analysis result, detect whether there is a defect in the metal sheet to obtain a defect detection result of the metal sheet, including:

[0026] According to the amplitude of the reflected echo signal and a preset amplitude threshold, detect whether there is a defect in the metal sheet to obtain a defect detection result of the metal sheet.

[0027] In one of the embodiments, according to the feature analysis result, detect whether there is a defect in the metal sheet to obtain a defect detection result of the metal sheet, and further includes:

[0028] According to a preset defect discrimination criterion and the feature analysis result, detect whether there is a defect in the metal sheet to obtain a defect detection result of the metal sheet. The preset defect discrimination criterion is determined based on feature extraction and quantitative analysis of the ultrasonic guided wave signal reflected by the metal sheet containing known defects.

[0029] In a second aspect, the present application further provides a metal sheet detection system. The system includes a processor, an ultrasonic guided wave probe, a conveying component, and a metal sheet. The processor is connected to the ultrasonic guided wave probe;

[0030] A transfer component for transferring a metal sheet to a preset detection area;

[0031] An ultrasonic guided wave probe for transmitting an ultrasonic guided wave signal into the metal sheet when the metal sheet is transferred to the preset detection area, receiving the reflected echo signal of the metal sheet, and sending the reflected echo signal to the processor;

[0032] A processor for performing the steps in any of the above metal sheet detection method embodiments to detect defects in the metal sheet.

[0033] In one embodiment, the ultrasonic guided wave probe further includes a couplant spraying device, an ultrasonic guided wave excitation and reception device, and a couplant recovery device;

[0034] The couplant spraying device for spraying the couplant onto the metal sheet;

[0035] The ultrasonic guided wave excitation and reception device for generating an electrical signal with a preset frequency and waveform based on the material characteristics of the metal sheet, amplifying the electrical signal, converting the amplified electrical signal into an ultrasonic guided wave signal, transmitting the ultrasonic guided wave signal into the metal sheet, receiving the reflected echo signal of the metal sheet, and sending the reflected echo signal to the processor;

[0036] The couplant recovery device for recovering the couplant on the metal sheet.

[0037] In a third aspect, the present application further provides a metal sheet detection device, including:

[0038] A signal acquisition module for controlling the ultrasonic guided wave probe to generate an ultrasonic guided wave signal when the metal sheet is transferred to the preset detection area, so that the ultrasonic guided wave signal propagates in the metal sheet; acquiring the reflected echo signal of the metal sheet;

[0039] A feature analysis module for performing feature analysis on the reflected echo signal to obtain a feature analysis result;

[0040] A defect detection module for detecting whether there are defects in the metal sheet according to the feature analysis result to obtain a defect detection result of the metal sheet.

[0041] In a fourth aspect, the present application further provides a computer device, including a memory and a processor, the memory stores a computer program, and the processor implements the steps in any of the above metal sheet detection method embodiments when executing the computer program.

[0042] In a fifth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, and the computer program implements the steps in any of the above metal sheet detection method embodiments when executed by the processor.

[0043] In a sixth aspect, the present application further provides a computer program product, including a computer program which, when executed by a processor, implements the steps in any of the above metal sheet detection method embodiments.

[0044] Different from the traditional method of first hoisting and then detecting a metal sheet in the metal sheet detection method, by integrating a defect detection function on the production line, when the metal sheet is conveyed to a preset detection area, an ultrasonic guided wave probe is controlled to generate an ultrasonic guided wave signal and transmit it into the metal sheet, the reflected echo signal of the metal sheet is obtained, the reflected echo signal is subjected to feature analysis, and whether there are defects in the metal sheet is detected according to the obtained feature analysis result. On the one hand, by performing on-line detection during the conveyance of the metal sheet, it is beneficial to improve the detection efficiency. Moreover, there is no need to hoist the metal sheet, reducing the possibility of secondary damage to the metal sheet caused by defect detection. At the same time, through automated defect detection, manual intervention is reduced, further improving the detection efficiency. On the other hand, based on the method of ultrasonic guided wave signal feature analysis for detecting defects in the metal sheet, it is beneficial to improve the accuracy and reliability of defect detection. Further, it is beneficial to predict potential quality problems according to the defect detection results and optimize the production process, thereby helping to reduce the probability of producing unqualified products and improving product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the following described drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0046] Figure 1 It is an application environment diagram of the metal sheet detection method in an embodiment;

[0047] Figure 2 It is a flow schematic diagram of the metal sheet detection method in an embodiment;

[0048] Figure 3 It is a flow schematic diagram of the metal sheet detection method in another embodiment;

[0049] Figure 4 It is a flow schematic diagram of the metal sheet detection method in yet another embodiment;

[0050] Figure 5 It is a structural block diagram of the metal sheet detection device in an embodiment;

[0051] Figure 6 is the structural block diagram of the metal sheet detection system in an embodiment;

[0052] Figure 7 is the structural schematic diagram of the metal sheet detection system in another embodiment;

[0053] Figure 8 is the schematic diagram of the aluminum plate defect detection of the uncoiler in an embodiment;

[0054] Figure 9 is the internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0055] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] The metal sheet detection method provided by the embodiment of the present application can be applied to an application environment as Figure 1 shown. Among them, the conveying component is the conveying component on the metal sheet processing production line. The ultrasonic guided wave probe 102 can be coupled with the metal sheet through a coupling agent. The ultrasonic guided wave probe 102 is communicatively connected to the processor 104, so as to detect defects in the metal sheet during the conveying process of the metal sheet. Specifically, when the metal sheet is conveyed to a preset detection area, the processor 104 sends a control signal to the ultrasonic guided wave probe 102. The ultrasonic guided wave probe 102 generates an ultrasonic guided wave signal in response to the control signal of the processor, so that the ultrasonic guided wave signal propagates in the metal sheet. Secondly, the processor 104 acquires the reflected echo signal of the metal sheet, performs feature analysis on the reflected echo signal to obtain a feature analysis result, and detects whether there are defects in the metal sheet according to the feature analysis result to obtain the defect detection result of the metal sheet.

[0057] In an exemplary embodiment, as Figure 2 shown, a metal sheet detection method is provided. Taking the method applied to the Figure 1 processor 104 in it as an example, the following steps (hereinafter simply referred to as S) S100 to S400 are included. Among them:

[0058] S100, when the metal sheet is conveyed to a preset detection area, control the ultrasonic guided wave probe to generate an ultrasonic guided wave signal, so that the ultrasonic guided wave signal propagates in the metal sheet.

[0059] Among them, the ultrasonic guided wave signal is an acoustic wave signal that propagates using the guided wave properties of the medium. The ultrasonic guided wave signal may include SH waves (Horizontal Shear Wave) in the plate, SV waves (Vertical Shear Wave), Lamb waves, and leaky Lamb waves, etc. The ultrasonic guided wave signal is the ultrasonic guided wave signal transmitted into the metal thin plate. The metal thin plate may be a thin plate made of materials such as stainless steel, aluminum alloy, and copper alloy, which is not limited herein. The thickness of the metal thin plate is generally between 0.2 mm and 2 mm. It can be understood that it can also be other thicknesses other than 0.2 mm to 2 mm, which is not limited herein.

[0060] In practical applications, to improve the detection efficiency, the defect detection function can be integrated on the metal thin plate processing production line. During the process of automatically transporting the metal thin plate through the conveying component on the metal thin plate processing production line, on-line defect detection is carried out. Subsequently, a specific space is preset on the production line as the detection area, and a position sensor is set in the detection area to detect whether the metal thin plate reaches the detection area. Then, the conveying speed of the conveying component can also be set according to the production speed of the metal thin plate and the detection performance of the processor, so that the conveying component conveys the metal thin plate to the preset detection area. It can be understood that the conveying speed of the conveying component can be set based on production processing requirements and detection accuracy.

[0061] In this embodiment, considering the plate structure and detection requirements of the metal thin plate, defect detection of the metal thin plate is carried out based on Lamb waves. Specifically, during implementation, the metal thin plate is placed on the conveying component of the metal thin plate processing production line, and the conveying component conveys the metal thin plate at the set conveying speed. When the metal thin plate is conveyed to the preset detection area, it can be that the position sensor sends a detection signal to the processor, so that the processor controls the ultrasonic guided wave probe to generate an electrical signal with a preset frequency and a preset waveform, amplify the electrical signal to meet the power required for effective excitation of the ultrasonic guided wave in the metal thin plate, convert the amplified electrical signal into an ultrasonic guided wave signal and transmit it into the metal thin plate. Among them, the frequency range of the preset frequency is adjusted between 1 MHz and 10 MHz (megahertz) according to the material characteristics of the metal thin plate (such as thickness, acoustic impedance, and Poisson's ratio) and detection requirements. The preset waveform can be a customized pulse sequence, sine wave, square wave, etc. Exemplarily, according to the material characteristics of the metal thin plate, the excitation frequency is determined to be 2 MHz and the waveform is a pulse wave to excite a single mode of Lamb wave. It can be understood that it can also be that the position sensor sends a detection signal to the ultrasonic guided wave probe, so that the ultrasonic guided wave probe generates an ultrasonic guided wave signal and transmits the ultrasonic guided wave signal into the metal thin plate.

[0062] S200, obtain the reflected echo signal of the metal thin plate.

[0063] Among them, the reflected echo signal is the signal reflected back to the ultrasonic guided wave probe from the metal thin plate.

[0064] In practical applications, after the ultrasonic guided wave probe transmits an ultrasonic guided wave signal into the metal thin plate, the ultrasonic guided wave signal propagates in the metal thin plate. After propagation, it reflects back the signal carrying the internal structure information of the metal thin plate. It can be that after the ultrasonic guided wave probe receives the reflected echo signal from the metal thin plate, the reflected echo signal is sent to the processor.

[0065] S300, perform feature analysis on the reflected echo signal to obtain a feature analysis result.

[0066] Among them, the feature analysis result can include signal frequency components, amplitude, and phase, etc., and can be obtained through feature analysis such as time-domain and frequency-domain analysis of the reflected echo signal.

[0067] In practical applications, in the case where there are defects inside the metal thin plate, during the propagation of the ultrasonic guided wave signal inside the metal thin plate, phenomena such as reflection, refraction, or scattering will occur, thus causing the reflected echo signal to change. Therefore, it is possible to determine whether there are defects in the metal thin plate by performing feature analysis on the ultrasonic guided wave signal to detect whether the reflected ultrasonic guided wave signal from the metal thin plate has changed.

[0068] Specifically, before performing feature analysis on the reflected echo signal, it can be that the processor performs preprocessing such as denoising (such as filtering) and normalization on the obtained reflected echo signal to improve the signal-to-noise ratio. Through feature analysis such as time-domain analysis and frequency-domain analysis of the reflected echo signal, a feature analysis result is obtained.

[0069] Exemplarily, the processor detects features such as the amplitude change and arrival time difference of the reflected echo signal through time-domain analysis to obtain a feature analysis result; the processor converts the time-domain signal into a frequency-domain signal by performing a fast Fourier transform on the reflected echo signal and analyzes the change in the signal frequency components to obtain a feature analysis result; the processor performs multi-scale wavelet transform on the reflected echo signal, decomposes the signal at different scales, and identifies local features to obtain a feature analysis result.

[0070] S400, according to the feature analysis result, detect whether there are defects in the metal thin plate to obtain a defect detection result of the metal thin plate.

[0071] Among them, the defect detection result can include whether there are defects in the metal thin plate.

[0072] In practical applications, an ultrasonic guided wave probe can transmit an ultrasonic guided wave signal into a metal thin plate in a known defect-free state in advance. The processor acquires the ultrasonic guided wave signal reflected from the metal thin plate in the known defect-free state, performs feature analysis on the ultrasonic guided wave signal to obtain a feature analysis result, compares the feature analysis result corresponding to the metal thin plate to be detected with the feature analysis result corresponding to the metal thin plate in the known defect-free state, determines whether there are difference points, thereby detecting whether there are defects in the metal thin plate and obtaining a metal thin plate detection result. In other embodiments, it can also be to train a machine learning model in advance according to the feature analysis result corresponding to the metal thin plate in the known defect-free state, automatically perform defect detection through the processor, and adjust the model parameters according to the actual situation to obtain a trained machine learning model, thereby detecting whether there are defects in the metal thin plate based on the feature analysis result and the machine learning model.

[0073] In the above metal thin plate detection method, different from the traditional method of first hoisting and then detecting the metal thin plate, by integrating the defect detection function on the production line, when the metal thin plate is conveyed to a preset detection area, the ultrasonic guided wave probe is controlled to generate an ultrasonic guided wave signal and transmit it into the metal thin plate, the reflected echo signal of the metal thin plate is acquired, the reflected echo signal is subjected to feature analysis, and whether there are defects in the metal thin plate is detected according to the obtained feature analysis result to obtain a defect detection result. On the one hand, by performing online detection during the conveying process, it is beneficial to improve the detection efficiency, and there is no need to hoist the metal thin plate, reducing the possibility of secondary damage to the metal thin plate caused by defect detection. At the same time, through automated defect detection, manual intervention is reduced, further improving the detection efficiency. On the other hand, based on the method of ultrasonic guided wave signal feature analysis for the defects of the metal thin plate, it is beneficial to improve the accuracy and reliability of defect detection. Further, it is beneficial to predict potential quality problems according to the defect detection result, optimize the production process, thereby helping to reduce the probability of producing unqualified products and improve product quality.

[0074] To further analyze the detailed information of the defect, in an exemplary embodiment, as Figure 3 shown, the metal thin plate detection method further includes S500 to S700. Among them:

[0075] S500, perform wavelet transform on the reflected echo signal to obtain a wavelet transform result.

[0076] Among them, the wavelet transform result may include the decomposed signals after decomposition.

[0077] In practical applications, the server performs three-layer wavelet transform on the reflected echo signal, and decomposes the reflected echo signal into four decomposed signals. Among them, the four decomposed signals include one layer of approximation component and three layers of detail components. The approximation component contains the low-frequency information of the signal, and the detail components contain the high-frequency information at different scales.

[0078] S600. When the defect detection result indicates that there are defects in the metal thin plate, determine the defect position and defect size according to the wavelet transform result.

[0079] Among them, the defect types can include but are not limited to cracks, holes, delamination, slag inclusions, pits, etc. The void in the metal thin plate refers to the void formed when gas fails to be discharged in time during the smelting or casting process of the metal material, and the hole will affect the durability of the product. The slag inclusion in the metal thin plate refers to the foreign substances existing inside the metal material, such as oxides, sulfides, etc. The signal characteristics of the ultrasonic guided wave signals reflected by the metal thin plate for different defect types are different, and the defect types can be discriminated by analyzing the signal characteristics of the reflected echo signal. The defect position is the position of the defect in the metal thin plate.

[0080] In practical applications, to determine the defect position, it can be based on the wavelet transform result, extract the energy distribution characteristics of the reflected echo signal at different scales, compare the extracted energy distribution characteristics with the preset reference energy distribution characteristics, and identify the energy distribution abnormal area. Among them, the reference energy distribution characteristics can be obtained by performing wavelet transform on the ultrasonic guided wave signal reflected by a known defect-free metal thin plate with the same material properties. The energy distribution abnormal area can be the range where the defect is located. According to the time information of the energy distribution abnormal area, calculate the time delay of the signal reflection caused by the defect, and combine the known group velocity of the Lamb wave to determine the defect position. It can be understood that it can also be to convert the reflected echo signal into a pulse signal through the oscilloscope screen of the processor, determine the defect position according to the pulse signal, or determine it according to the position of the ultrasonic guided wave probe when the defect is detected, which is not limited here.

[0081] In one embodiment, to determine the defect size according to the wavelet transform result, it can be to determine the energy amplitude of the reflected echo signal according to the wavelet transform result. If the defect size is larger, the reflected wave energy is higher. The corresponding relationship between the defect size and the reflected energy can be set according to experience, or the corresponding relationship between the defect size and the reflected energy can be obtained through actual measurement, and a database containing the corresponding relationship between the defect size and the reflected energy is pre-constructed. Determine the defect size through the energy amplitude of the reflected echo signal and the pre-constructed database.

[0082] In other embodiments, the severity of the defect can also be determined based on the reflected echo signal. Specifically, an evaluation criterion for assessing the severity of the defect in the metal sheet can be set in advance according to the material and application standards of the metal sheet, and the severity of the defect can be determined based on the evaluation criterion and the reflected echo signal. Exemplarily, the severity evaluation criterion can include the area range and depth range of the defect corresponding to different severity levels. Among them, the severity levels can include minor defects, general defects, and severe defects. The defect area range corresponding to minor defects can be a relatively small range that does not affect the overall structural performance, and the corresponding defect depth range can be no more than 10% of the thickness of the metal sheet. The defect area range corresponding to severe defects can be a relatively large range that affects the function and structure of the metal sheet, and the corresponding defect depth range can be more than 25% of the material thickness, or close to or reaching the degree of complete penetration.

[0083] In other embodiments, when the defect detection result indicates that there is a defect in the metal sheet, the defect type can also be determined based on the reflected echo signal. Specifically, the operator can preset the defect type discrimination rules according to the characteristics of the ultrasonic guided wave signals reflected by the metal sheet for different defect types. For example, cracks usually cause additional reflection peaks or changes in spectral components; holes may cause significant energy loss and time delay of the reflected wave; delamination generates strong reflection signals, etc., so as to set corresponding discrimination rules for different defect types.

[0084] In other embodiments, when the defect detection result indicates that there is a defect in the metal sheet, determining the defect type based on the reflected echo signal can also include: determining the defect type according to the result of the characteristic analysis of the reflected echo signal and the preset defect type discrimination rules. It can also be that the processor obtains the reference reflected echo signals reflected from the incoming ultrasonic guided wave signals of the metal sheet corresponding to known different defect types, compares the reflected echo signal with the reference reflected echo signals of different defect types, and identifies the defect type according to the signal difference.

[0085] In one embodiment, the processor determines the boundary of the defect based on the amplitude and position of the reflected echo signal, obtains the defect image through scanning imaging techniques (such as A-scan, B-scan, phased array imaging, etc.) to determine the defect area; measures the time for the ultrasonic guided wave signal to be reflected back from the surface to the defect, and determines the defect depth according to the sound velocity. It can also be to perform signal frequency analysis on the reflected echo signal and determine the defect depth through the frequency difference between the reflected echo signal and the ultrasonic guided wave signal. After the processor determines the area and depth of the defect based on the reflected echo signal, it uses the preset severity evaluation criterion to determine the severity level of the defect.

[0086] S700, generate and push an alarm message according to the defect detection result, defect position, and defect size.

[0087] In practical applications, the processor may obtain the defect detection time, integrate the defect detection results, defect detection time, defect location, and defect size, generate an alarm message, and push it. The ways to push the alarm message may include pushing the alarm message in the notification bar of the processor's display, or displaying the alarm message in the form of a full-screen or half-screen pop-up window. Additionally, based on this, an audible alarm prompt can be made by emitting a specific sound or vibration pattern, and a visual alarm prompt can be made by the flashing of an indicator light, etc. It can be understood that the method of pushing the alarm message can be a combination of the aforementioned method of pushing the alarm message and any one or more of the aforementioned alarm prompt methods, which is not limited herein. In other embodiments, the alarm signal may further include the defect type and severity.

[0088] In this embodiment, wavelet transform is performed on the reflected echo signal. When a defect in the metal thin plate is detected, further analysis is carried out based on the wavelet transform result to determine the detailed defect information, such as the defect location, defect type, and severity, and an alarm signal is generated. This is beneficial for processing the defective metal thin plate in a timely manner according to the alarm message to improve product quality. At the same time, it is beneficial to analyze the cause of the defect based on the alarm message, thereby optimizing the production parameters of the metal thin plate.

[0089] In an exemplary embodiment, determining the defect location and defect size based on the wavelet transform result includes:

[0090] Based on the wavelet transform result, determine the energy and wave packet centroid time of the low-frequency component pulse signal in the reflected echo signal.

[0091] Based on the energy and wave packet centroid time, determine the defect location and defect size.

[0092] Among them, the wave packet centroid time is the energy distribution center of the signal on its time axis.

[0093] In practical applications, for the decomposed low-frequency (frequency range 0 - 0.65 MHz) signal (i.e., the low-frequency component pulse signal), the energy E of the low-frequency component pulse signal in the reflected echo signal is determined by the following formula:

[0094]

[0095] The wave packet centroid time is determined by the following formula:

[0096]

[0097] Among them, \(t1 / t2\) is the starting time of the wave packet, that is, the time range of the wave packet when the peak of the wave packet drops by 6 dB, and \(s(t)\) is the detected signal. During the detection process, if it is found that the centroid of the reflected signal wave packet in a certain area is significantly different from the centroid of the wave packet in the surrounding defect-free area, it may be due to the existence of a defect.

[0098] After obtaining the energy \(E\) and the centroid time of the wave packet, according to the energy and the centroid time of the wave packet, the defect position can be determined. It can be to determine the distance between the defect and the probe according to the centroid time of the wave packet and the sound velocity. Specifically, the centroid time of the wave packet * sound velocity = the distance between the defect and the probe. According to the energy and the centroid time of the wave packet, the defect size can be determined. It can be to establish a relationship model between the energy, the centroid time of the wave packet and the actual defect size in advance according to experiments. Obtain this model, test it with standard defect samples, record its energy and centroid time characteristics of the wave packet, and then train a regression model based on these data for evaluating the defect size. Among them, the standard defect samples contain known defects, and the defect sizes of the defects are measured in advance. In specific implementation, taking the energy and the centroid time of the wave packet as inputs, call the trained defect size evaluation model to obtain the defect size.

[0099] In this embodiment, by using the energy and the centroid time of the wave packet to determine the defect position and the defect size, it is beneficial to improve the accuracy of evaluating the defect position and the defect size.

[0100] In an exemplary embodiment, the method further includes: comparing the energy with a preset energy threshold. If the energy is greater than or equal to the preset energy threshold, it is determined that there is a defect in the metal thin plate.

[0101] Among them, the preset energy threshold is determined according to a metal thin plate sample containing standard reference defects. Exemplarily, the preset energy threshold is the energy of the reflected echo signal of a \(\varphi1\) mm through-hole (standard reference defect) perpendicular to the plate surface of the metal thin plate sample.

[0102] In practical applications, detect whether the energy \(E\) is greater than or equal to the preset energy threshold \(E\) h , if the energy \(E\) is greater than or equal to the preset energy threshold \(E\) h , it is determined that there is a defect in the metal thin plate.

[0103] In this embodiment, by using the preset energy threshold to judge whether there is a defect in the metal thin plate according to the energy of the low-frequency component pulse signal of the reflected echo signal, the efficiency and accuracy of defect judgment are improved.

[0104] To improve the simplicity of detection, in an exemplary embodiment, feature analysis is performed on the reflected echo signal to obtain a feature analysis result, including S320, where:

[0105] S320. Analyze the characteristics of the reflected echo signal to determine the amplitude of the reflected echo signal. The result of the characteristic analysis includes the amplitude of the reflected echo signal.

[0106] In practical applications, it can be the processor that performs time-domain and frequency-domain analyses on the reflected echo signal, extracts the signal amplitude of the reflected wave or scattered wave in the reflected echo signal, and obtains the result of the characteristic analysis.

[0107] According to the result of the characteristic analysis, detect whether there are defects in the thin metal plate, and obtain the defect detection result of the thin metal plate, including S420, where:

[0108] S420. According to the amplitude of the reflected echo signal and a preset amplitude threshold, detect whether there are defects in the thin metal plate, and obtain the defect detection result of the thin metal plate.

[0109] In practical applications, defect detection can be performed on a pre-prepared comparison specimen to determine the amplitude threshold as a reference. Specifically, an ultrasonic guided wave signal can be transmitted into the comparison specimen through an ultrasonic guided wave probe, and the processor obtains the ultrasonic guided wave signal reflected by the comparison specimen, and determines the highest amplitude of the reflected wave for the defect as the amplitude threshold.

[0110] In specific implementation, the processor can detect whether there are defects in the thin metal plate by comparing the amplitude of the reflected echo signal with a preset amplitude threshold. Specifically, when the amplitude of the reflected echo signal exceeds the preset amplitude threshold, it is determined that there are defects in the thin metal plate; otherwise, it is determined that there are no defects in the thin metal plate. Among them, the comparison specimen is close to the material, thickness, and geometric shape of the actually detected thin metal plate and contains known defects (such as cracks, holes, delaminations, etc.).

[0111] In this embodiment, defect detection is performed on the thin metal plate through the amplitude of the reflected echo signal and a preset amplitude threshold, which improves the detection efficiency and simplicity.

[0112] In an exemplary embodiment, according to the reflected echo signal, determine the defect type, including S520 to S540, where:

[0113] S520. Perform wavelet transform on the reflected echo signal to obtain the wavelet transform result.

[0114] S540. Using the wavelet transform result as input, call a trained defect discrimination model to obtain the defect type of the thin metal plate. The defect discrimination model is trained based on the wavelet transform results of historical ultrasonic guided wave signals carrying defect category labels.

[0115] Among them, the wavelet transform result can be a time-frequency diagram, which characterizes the changes of the signal at different times and frequencies.

[0116] In practical applications, it can be that the processor pre-gets the historical signal of the reflected echo signal obtained by defect detection of the metal sheet in a historical time period, performs multi-scale wavelet transform on the historical signal of the reflected echo signal, extracts the time-frequency domain features of the historical signal of the reflected echo signal, generates a time-frequency diagram, labels the corresponding defect type labels (such as crack, hole, delamination, etc.) for the time-frequency diagram, and constructs a defect feature library according to the time-frequency diagrams of different defect types and the defect type labels corresponding to the time-frequency diagrams. The operator constructs an initial defect discrimination model based on a deep convolutional neural network (Deep Convolutional Neural Networks, DCNN), and controls the processor to iteratively train the initial defect discrimination model through the defect feature library to learn the classification pattern of defects until a preset training end condition is reached, and a trained defect discrimination model is obtained. Among them, the deep convolutional neural network can be models such as ResNet (residual network) and Inception-v3. The preset training end condition can be that the value of the loss function is less than the preset loss threshold for a continuous preset number of times. Exemplarily, it can be that the processor respectively obtains 1000 Lamb wave signals (reflected echo signals) uploaded by the operator for defect types such as cracks, holes, and delaminations, performs multi-scale wavelet transform on each signal, extracts the time-frequency domain features, generates a time-frequency diagram, labels the corresponding defect type labels for the time-frequency domain and stores them in the defect feature library, and trains the DCNN model through the defect feature library to optimize the network parameters to obtain a defect discrimination model for discriminating the defect types of metal sheets.

[0117] After obtaining the trained defect discrimination model, apply this model to determine the defect type of the detected defect. In specific implementation, perform multi-scale wavelet transform on the reflected echo signal, extract the time-frequency domain features of the reflected echo signal, generate a time-frequency diagram, use the time-frequency diagram as the input, call the trained defect discrimination model, and obtain the defect type of the metal sheet.

[0118] In this embodiment, by extracting the time-frequency features of the reflected echo signal and combining with a deep learning model for defect detection, the accuracy and efficiency of defect detection are improved.

[0119] To improve the accuracy of defect type discrimination, in an exemplary embodiment, as Figure 4 shown, it further includes S820 to S860, where:

[0120] S820, in the case of determining that there is a defect in the metal sheet, generate a grayscale image based on the energy of the low-frequency component pulse signal in the reflected echo signal.

[0121] S840, analyze the shape features of the target region in the grayscale image to obtain a shape feature analysis result, where the target region is a defect potential region.

[0122] S860, determine the defect type based on the shape feature analysis result.

[0123] Among them, the shape feature analysis result may include the shape type of the target area. The target area is the potential defect area, and can be identified from the grayscale image through the edge detection algorithm. The defect type may include, but is not limited to, pores, slag inclusions, cracks, etc.

[0124] In practical applications, generating a grayscale image based on the energy of the low-frequency component pulse signal in the reflected echo signal may be to segment the low-frequency component pulse signal of the reflected echo signal into multiple time windows of a fixed length. The processor calculates the energy of the low-frequency component for each window to obtain a set of energy values. Normalize the energy values to the range of 0% to 100%, map the normalized energy values to different grayscale values, and then fill each grayscale value into the corresponding position of the graph to generate a grayscale image. Then, perform smoothing processing, denoising, and artifact removal on the grayscale image to obtain the grayscale image to be detected.

[0125] After the processor obtains the grayscale image to be detected, it may be based on an edge detection algorithm (such as Canny edge detection) to identify the edges in the grayscale image to detect the boundaries of the defects and determine the target area. Extract the shape feature parameters of the target area, such as area, perimeter, circularity, aspect ratio, and moment, etc. Match the shape feature parameters of the target area with the shape feature parameters corresponding to the preset shape type to determine the shape type of the target area and obtain the shape feature analysis result of the target area. The shape feature analysis result includes the shape type of the target area. Among them, the types of the target area include, but are not limited to, dot-like, spherical, elliptical, and strip-shaped.

[0126] After the processor obtains the shape feature analysis result, it discriminates the defect type according to the shape type of the target area and the preset judgment rules. Specifically, the preset judgment rules may include: if there is a dot-like or spherical target area in the image, it is determined that the defect type is a pore; if there is an elliptical target area in the image, it is determined that the defect type is a slag inclusion; if there is a narrow strip-shaped target area in the image, it is determined that the defect type is a crack.

[0127] In this embodiment, signal processing and image analysis are combined to discriminate the defect type, improving the accuracy and efficiency of defect detection.

[0128] In an exemplary embodiment, according to the feature analysis result, detect whether there are defects in the metal sheet to obtain the defect detection result of the metal sheet, and further include:

[0129] Based on a preset defect discrimination criterion and the result of feature analysis, detect whether there are defects in the metal sheet to obtain the defect detection result of the metal sheet. The preset defect discrimination criterion is determined based on feature extraction and quantitative analysis of the ultrasonic guided wave signals reflected by the metal sheets containing known defects.

[0130] Among them, the preset defect discrimination criterion may include the discrimination criteria for the ultrasonic guided wave signal features (such as the amplitude of the reflected wave, time delay, and frequency components, etc.) reflected by the metal sheet for different defect types, and can be determined by feature extraction and quantitative analysis of the ultrasonic guided wave signals reflected by the metal sheets containing known defects. The ultrasonic guided wave signals reflected by the metal sheets containing known defects can be obtained from the local database.

[0131] Exemplarily, the processor may pre - perform wavelet transform and Fourier transform, etc. on the ultrasonic guided wave signals reflected by the metal sheets containing known defects in the database, extract the time - frequency features and frequency - domain features of each ultrasonic guided wave signal, such as the amplitude of the reflected wave, time delay, pulse width, spectral peak, frequency components, etc. Analyze the signal features of the same defect type, and set quantitative criteria for different defects. For example, for cracks, set the defect discrimination criterion to include: if the amplitude of the reflected wave of the ultrasonic guided wave signal > 80%, and the time delay is between 10 (microseconds) and 20 , then it is determined that there are cracks in the metal sheet.

[0132] For holes, set the defect discrimination criterion to include: if the amplitude of the scattered wave of the ultrasonic guided wave signal > 50%, and the frequency components are concentrated between 1 MHz and 2 MHz, then it is determined that there are holes in the metal sheet.

[0133] For delamination, set the defect discrimination criterion to include: if the ultrasonic guided wave signal includes multiple reflected waves, and the time interval is between 5 and 10 , it is determined that there is delamination in the metal sheet. In a specific implementation, according to the preset defect discrimination criterion and the result of feature analysis, determine whether the signal features meet the defect discrimination criterion. If so, it is determined that there are defects in the metal sheet, and further determine the defect type. It can be understood that the set discrimination criterion is specifically determined according to the material and specifications of the metal sheet to be detected, as well as the defect type, and is not limited to the content included in the above - mentioned example of the defect discrimination criterion.

[0134] In this embodiment, defect detection is performed through the preset defect discrimination criterion and the result of feature analysis, improving the defect detection efficiency and accuracy.

[0135] To make a clearer description of the metal sheet detection method provided in this application, the following describes a specific embodiment, which includes the following steps:

[0136] S1. When the metal sheet is conveyed to a preset detection area, control the ultrasonic guided wave probe to generate an ultrasonic guided wave signal so that the ultrasonic guided wave signal propagates in the metal sheet.

[0137] S2. Obtain the reflected echo signal of the metal sheet.

[0138] S3. Conduct feature analysis on the reflected echo signal to determine the amplitude of the reflected echo signal. The feature analysis result includes the amplitude of the reflected echo signal.

[0139] S4. According to the amplitude of the reflected echo signal and a preset amplitude threshold, detect whether there are defects in the metal sheet to obtain the defect detection result of the metal sheet.

[0140] S5. Perform wavelet transform on the reflected echo signal to obtain the wavelet transform result. When the defect detection result indicates that there are defects in the metal sheet, based on the wavelet transform result, determine the energy of the low-frequency component pulse signal and the wave packet centroid time in the reflected echo signal, and determine the defect location and defect size according to the energy and the wave packet centroid time.

[0141] S6. When it is determined that there are defects in the metal sheet, generate a grayscale image based on the energy of the low-frequency component pulse signal in the reflected echo signal, analyze the shape features of the target area in the grayscale image to obtain the shape feature analysis result, and determine the defect type based on the shape feature analysis result. The target area is the defect potential area.

[0142] S7. Generate and push an alarm message according to the defect detection result, defect type, defect location, and defect size.

[0143] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise clearly stated in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments, and the execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0144] In an exemplary embodiment, as Figure 5 shown, a metal sheet detection device 600 is provided, including: a signal acquisition module 610, a feature analysis module 620, and a defect detection module 630, where:

[0145] A signal acquisition module 610, configured to, when the metal sheet is conveyed to a preset detection area, control an ultrasonic guided wave probe to generate an ultrasonic guided wave signal, so that the ultrasonic guided wave signal propagates in the metal sheet; and acquire the reflected echo signal of the metal sheet.

[0146] A feature analysis module 620, configured to perform feature analysis on the reflected echo signal to obtain a feature analysis result.

[0147] A defect detection module 630, configured to detect whether there is a defect in the metal sheet according to the feature analysis result to obtain a defect detection result of the metal sheet.

[0148] In an exemplary embodiment, the metal sheet detection device 600 further includes a wavelet transform module 640, a defect analysis module 650, and a defect alarm module 660, where:

[0149] The wavelet transform module 640 is configured to perform wavelet transform on the reflected echo signal to obtain a wavelet transform result.

[0150] The defect analysis module 650 is configured to, when the defect detection result indicates that there is a defect in the metal sheet, determine the defect position and defect size according to the wavelet transform result.

[0151] The defect alarm module 660 generates and pushes an alarm message according to the defect detection result, the defect position, and the defect size.

[0152] In an exemplary embodiment, the defect analysis module 650 is further configured to determine the energy and the wave packet centroid time of the low-frequency component pulse signal in the reflected echo signal according to the wavelet transform result; and determine the defect position and defect size according to the energy and the wave packet centroid time.

[0153] In an exemplary embodiment, the metal sheet detection device 600 further includes a defect judgment module 670, configured to compare the energy with a preset energy threshold, and if the energy is greater than or equal to the preset energy threshold, determine that the metal sheet has a defect.

[0154] In an exemplary embodiment, the feature analysis module 620 is further configured to perform feature analysis on the reflected echo signal to determine the amplitude of the reflected echo signal, and the feature analysis result includes the amplitude of the reflected echo signal.

[0155] The defect detection module 630 is further configured to detect whether there is a defect in the metal sheet according to the amplitude of the reflected echo signal and a preset amplitude threshold to obtain a defect detection result of the metal sheet.

[0156] In an exemplary embodiment, the defect analysis module 640 is further configured to perform wavelet transform on the reflected echo signal to obtain a wavelet transform result, and call a trained defect discrimination model with the wavelet transform result as the input to obtain the defect type of the metal sheet. The defect discrimination model is trained based on the wavelet transform results of historical ultrasonic guided wave signals carrying defect category labels.

[0157] In an exemplary embodiment, the metal sheet detection device 600 further includes a defect type discrimination module 680, configured to generate a grayscale image based on the energy of the low-frequency component pulse signal in the reflected echo signal when it is determined that the metal sheet has a defect, analyze the shape characteristics of the target region in the grayscale image to obtain a shape feature analysis result, and determine the defect type based on the shape feature analysis result. The target region is the potential defect region.

[0158] In an exemplary embodiment, the defect detection module 630 is further configured to detect whether the metal sheet has a defect according to a preset defect discrimination criterion and the feature analysis result, and obtain the defect detection result of the metal sheet.

[0159] Each module in the above metal sheet detection device 600 can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor in the computer device in hardware form or independent of the processor, or stored in the memory in the computer device in software form, so that the processor can call and execute the operations corresponding to the above respective modules.

[0160] In an exemplary embodiment, as Figure 6 shown, a metal sheet detection system is provided, including an ultrasonic guided wave probe 120, a processor 140, a conveying component 160, and a metal sheet 180. The processor 140 is connected to the ultrasonic guided wave probe 120. Among them:

[0161] The conveying component 160 is configured to convey the metal sheet 180 to a preset detection area.

[0162] The ultrasonic guided wave probe 120 is configured to emit an ultrasonic guided wave signal when the metal sheet 180 is conveyed to the preset detection area, transmit the ultrasonic guided wave signal into the metal sheet 180, receive the reflected echo signal of the metal sheet 180, and send the reflected echo signal to the processor 140.

[0163] The processor 140 is configured to execute the steps in any of the above metal sheet detection method embodiments to perform defect detection on the metal sheet 180.

[0164] Among them, the conveying component 160 is a conveying component on a metal sheet processing production line. The ultrasonic guided wave probe 120 can be coupled with the metal sheet 180 through a coupling agent. The ultrasonic guided wave probe and the processor detect defects in the metal sheet during the conveying process of the metal sheet.

[0165] In practical applications, a specific space is preset on the production line as a detection area, and a position sensor is set in the detection area to detect whether the metal sheet reaches the detection area. Then, the conveying speed of the conveying component can also be set according to the production speed of the metal sheet and the detection performance of the processor, so that the conveying component conveys the metal sheet to the preset detection area.

[0166] Before detection, an operator can select corresponding ultrasonic guided wave excitation frequencies, probe types, and arrangement methods according to parameters such as the material, thickness, and size of the metal sheet. By adjusting the parameters of the signal generator and the position of the ultrasonic guided wave probe, the ultrasonic guided wave probe 120 can be made to excite and receive effective ultrasonic guided wave signals and be in an optimal working state.

[0167] In specific implementation, the metal sheet is placed on the conveying component of the metal sheet processing production line. The conveying component conveys the metal sheet at the set conveying speed. When the metal sheet is conveyed to the preset detection area, the position sensor can send a detection signal to the processor, so that the processor controls the ultrasonic guided wave probe to generate an electrical signal with a preset frequency and a preset waveform, amplify the electrical signal to meet the power required for effective excitation of the ultrasonic guided wave in the metal sheet, convert the amplified electrical signal into an ultrasonic guided wave signal and transmit it into the metal sheet. The processor acquires the reflected echo signal of the metal sheet, performs defect detection based on the reflected echo signal, refers to the above-mentioned processor to perform feature analysis on the reflected echo signal to obtain a feature analysis result, and based on the feature analysis result, detects whether there are defects in the metal sheet. The steps in the method for obtaining the defect detection result of the metal sheet are not elaborated here.

[0168] In this embodiment, an automated defect detection system for metal sheets is designed. Different from the traditional method of first hoisting and then detecting metal sheets, by integrating the defect detection function on the production line, online detection is performed during the conveying process of the metal sheet, without hoisting the metal sheet, thereby reducing the possibility of secondary damage to the metal sheet and being beneficial to improving the efficiency of defect detection for large-sized metal sheets. At the same time, based on ultrasonic guided wave detection, it is beneficial to improve the accuracy and reliability of defect detection, thereby being beneficial to timely discovering quality problems in the production process of metal sheets and improving the use safety of products.

[0169] In an exemplary embodiment, such as Figure 7As shown, the ultrasonic guided wave probe 120 further includes a couplant spraying device 121, an ultrasonic guided wave excitation and reception device 122, and a couplant recovery device 123, where:

[0170] The couplant spraying device 121 is used to spray the couplant onto the metal sheet 180.

[0171] The ultrasonic guided wave excitation and reception device 122 is used to generate an electrical signal with a preset frequency and waveform based on the material properties of the metal sheet 180, amplify the electrical signal, convert the amplified electrical signal into an ultrasonic guided wave signal, and transmit the ultrasonic guided wave signal into the metal sheet 180, receive the reflected echo signal of the metal sheet 180, and send the reflected echo signal to the processor 140.

[0172] The couplant recovery device 123 is used to recover the couplant on the metal sheet 180.

[0173] In practical applications, the ultrasonic guided wave excitation and reception device dynamically matches the excitation frequency (1 - 10 MHz) and waveform (such as a customized pulse sequence) based on the material properties of the metal sheet (such as acoustic impedance, Poisson's ratio), thereby exciting a single-mode Lamb wave, reducing clutter interference, and greatly improving the signal-to-noise ratio.

[0174] The ultrasonic guided wave excitation and reception device 122 is divided into two parts ( Figure 7 not shown in the figure), one is the ultrasonic guided wave excitation device, and the other is the ultrasonic guided wave reception device. Among them, the ultrasonic guided wave excitation device includes a signal generator, a power amplifier, and an excitation probe. Specifically:

[0175] The signal generator is used to generate an electrical signal with a specific frequency and waveform, and its frequency range can be adjusted between 1 MHz and 10 MHz according to the material properties of the metal sheet and the detection requirements. The waveform can be a sine wave, a square wave, a pulse wave, etc.

[0176] The power amplifier is connected to the signal generator and is used to amplify the electrical signal to meet the power required for effective excitation of ultrasonic guided waves in the metal sheet, ensuring that an ultrasonic guided wave signal with sufficient intensity can propagate in the metal sheet.

[0177] The excitation probe is made of piezoelectric ceramic material and is tightly coupled to the surface of the metal sheet through epoxy resin to convert the amplified electrical signal into an ultrasonic guided wave signal and transmit it into the metal sheet. The size and shape of the probe are optimized according to the thickness and material of the metal sheet to ensure that the excited guided wave mode is single and has good propagation characteristics.

[0178] The ultrasonic guided wave reception device includes a reception probe, a preamplifier, and a data acquisition card. Specifically:

[0179] The receiving probe is made of piezoelectric ceramic material and coupled to the surface of the metal sheet. It is used to receive the ultrasonic guided wave signal that carries the internal information of the metal sheet after propagating in the metal sheet and convert it into an electrical signal. The position of the receiving probe maintains a specific distance and angle relationship with the excitation probe to obtain the best signal reception effect.

[0180] The preamplifier is connected to the receiving probe to preliminarily amplify the weak electrical signal output by the receiving probe.

[0181] The data acquisition card converts the analog electrical signal output by the preamplifier into a digital signal and transmits it to the processor 140 .

[0182] The coupling agent spraying device 121 may include a coupling agent storage tank, a gradient pressure control device, and a micro nozzle array. The coupling agent storage tank is used to store coupling agent (such as water, oil, or special ultrasound gel). The gradient pressure control device is used to dynamically adjust the coupling agent spraying pressure and flow rate in different areas according to the shape and size of the metal sheet so that the liquid film is evenly covered. The micro nozzle array is used to spray the coupling agent.

[0183] The coupling agent recovery device 123 may include a negative pressure generator, a recovery pipeline, a filter device and a recovery storage tank. The negative pressure generator is used to generate negative pressure through a vacuum pump or a negative pressure fan. The recovery pipeline is used to collect excess coupling agent on the metal sheet. The filter device is used to filter the recovered coupling agent to remove impurities and contaminants therefrom. The recovery storage tank is used to store the recovered coupling agent for recycling.

[0184] In this embodiment, the ultrasonic guided wave excitation and receiving device of the ultrasonic guided wave probe is optimized to excite and receive ultrasonic guided wave signals with a single mode and good propagation characteristics, which is conducive to improving the sensitivity of detecting small defects inside the metal sheet, thereby improving the accuracy and reliability of detection. At the same time, the coupling agent spraying device and the coupling agent recovery device are used for automatic spraying and recovery, which improves the degree of automation, improves the recycling rate of the coupling agent, and reduces the possibility of residual pollution.

[0185] In a detailed embodiment, Figure 8 As shown, the integrated defect detection on the aluminum alloy sheet unwinding line is used to perform online detection of metal sheets (aluminum alloy sheets) as an example. Specifically:

[0186] This embodiment uses an aluminum alloy sheet with a material grade of 5A06, a thickness of 1 mm, a width of 2200 mm, and an aluminum sheet uncoiling line with a finished product length of 2200 mm.

[0187] (1)To remove the interference caused by surface unevenness to the flaw detection process, a flattening and straightening process is added before the punching process; after the straightening process is completed, the burrs in the width direction are removed, and then it enters the edge trimming and cleaning process to trim the width of the aluminum plate to 2200 mm. Then, the whole plate is cleaned with a cleaning liquid and enters the drying process to dry the whole plate.

[0188] (2)The structure of the ultrasonic guided wave probe is divided into three parts. The front part is a couplant spraying device used to evenly spray the couplant (glycerol: water = 1:1) on the material surface. The middle part is an ultrasonic guided wave excitation and reception device, which is also equipped with a marking device. The end part is a couplant recovery device, which can realize the recovery of the couplant on the plate surface.

[0189] (3)For the ultrasonic guided wave probe, select an ultrasonic guided wave probe that emits a 2 MHz (megahertz) ultrasonic guided wave signal. Place the ultrasonic guided wave probe on a reference specimen made of the same material and specification as the metal sheet to be detected and having defects (such as a φ1 mm through hole on the plate surface). Adjust the angle of the ultrasonic guided wave probe, select the Lamb wave S0 excitation mode, and transmit the Lamb wave into the reference specimen for defect detection.

[0190] (4)The ultrasonic guided wave probe emits Lamb waves to scan the reference specimen. The processor detects the defects on the reference specimen by comparing the Lamb waves reflected back in the reference specimen and determines the energy of the Lamb waves reflected back from the defects, and sets the energy as the alarm threshold (energy threshold).

[0191] (5)Before the on-line detection of the metal sheet, install the ultrasonic guided wave probe on the aluminum plate uncoiler line and ensure good contact between the ultrasonic guided wave probe and the aluminum plate to be detected (aluminum alloy sheet). Set the uncoiling speed at 20 m / min (meters per minute) and start the uncoiler line.

[0192] (6)During the detection process of the metal sheet based on ultrasonic guided waves, the couplant spraying device at the front end of the ultrasonic guided wave probe first sprays the couplant to ensure good contact between the probe and the aluminum plate surface, so that the ultrasonic guided wave excitation and reception device can excite and receive the ultrasonic guided wave signals reflected back from the aluminum plate. After the excitation and reception of the ultrasonic guided wave signals are completed, the couplant is recovered and filtered through the couplant recovery device at the end of the ultrasonic guided wave probe device.

[0193] (7) The ultrasonic guided wave probe sends the reflected echo signal to the processor. The processor performs wavelet transform on the reflected echo signal to obtain the wavelet transform result, determines the energy of the low-frequency component pulse signal of the reflected echo signal according to the wavelet transform result, compares the energy with a preset energy threshold. If the energy is greater than or equal to the preset energy threshold, the processor generates and pushes an alarm message, causing the processor to emit a beeping sound for alarm. At the same time, an alarm signal is sent to the ultrasonic guided wave probe, and the marking device of the ultrasonic guided wave probe marks the position where the defect appears in the metal sheet, and the processor starts to record the reflected echo signal of the defect.

[0194] (8) According to the wavelet transform result, determine the energy and wave packet centroid time of the low-frequency component pulse signal in the reflected echo signal. According to the energy and wave packet centroid time, determine the defect position and defect size.

[0195] (9) The processor performs defect grayscale imaging based on the energy of the low-frequency component pulse signal in the reflected echo signal, and performs smoothing processing on the generated grayscale image to remove noise and artifacts. Specifically, the energy value of the reflected echo signal is normalized to the range of 0% to 100%, and the normalized energy value is mapped to a grayscale value. Among them, the grayscale value corresponding to the normalized energy value of 0% is 0 (pure black), and the grayscale value corresponding to the normalized energy value of 100% is 255 (pure white, in an 8-bit system). The grayscale value of each scan point is filled into the corresponding position of the graph to generate a grayscale image.

[0196] (10) Analyze the shape characteristics of the white area in the grayscale image to obtain the shape type of the white area. According to the shape type of the white area and a preset judgment rule, judge the defect type. Specifically, the preset judgment rule may include: if there are dot-like or spherical white areas in the image, it is determined that the defect type is a pore; if there are elliptical white areas in the image, it is determined that the defect type is a slag inclusion; if there are narrow strip-shaped white areas in the image, it is determined that the defect type is a crack.

[0197] (11) After the aluminum plate detection is completed, it enters the backend punching process. For qualified products (such as those without defects or with defect degrees lower than the preset defect degree threshold), they are stacked and transferred normally and enter the next process. Conversely, for unqualified products, they are removed before stacking.

[0198] After each defect detection, the processor records the defect detection results, including the thin plate number, detection time, detection location, whether there are defects and detailed information about the defects (such as defect type, defect location, and defect size), stores them in the constructed thin plate quality detection file, and pushes them to the display for subsequent query and traceability. In the case where the defect detection results indicate the existence of defects, according to the defect detection results, the production process parameters of the metal thin plate are adjusted and optimized, such as processing temperature, pressure, and speed, etc., to reduce the possibility of internal defects in the metal thin plate and improve the production quality and yield. At the same time, the defect detection results are fed back to the production management system to realize quality monitoring and closed-loop control of the production process.

[0199] In an exemplary embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 9 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. 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, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. The computer program, when executed by the processor, implements a metal thin plate detection method.

[0200] Those skilled in the art can understand that Figure 9 the structure shown in

[0201] is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0202] In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in any one of the above-mentioned metal thin plate detection method embodiments are implemented.

[0203] In one embodiment, a computer program product is provided, including a computer program which, when executed by a processor, implements the steps in any of the above-described embodiments of the metal sheet detection method.

[0204] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0205] 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, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., without limitation.

[0206] 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 recorded in the present application.

[0207] 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 fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A method for detecting a thin metal plate, applied to a thin metal plate detection system, characterized in that The system includes a conveying component, an ultrasonic guided wave probe, and a metal thin plate; the method includes: When the metal thin plate is conveyed to a preset detection area, controlling the ultrasonic guided wave probe to generate an ultrasonic guided wave signal so that the ultrasonic guided wave signal propagates in the metal thin plate; Obtaining the reflected echo signal of the metal thin plate; Performing feature analysis on the reflected echo signal to obtain a feature analysis result; According to the feature analysis result, detecting whether there is a defect in the metal thin plate to obtain a defect detection result of the metal thin plate.

2. The method according to claim 1, characterized in that The method further includes: Performing wavelet transform on the reflected echo signal to obtain a wavelet transform result; When the defect detection result indicates that there is a defect in the metal thin plate, determining the defect position and defect size according to the wavelet transform result; Generating and pushing an alarm message according to the defect detection result, the defect position, and the defect size.

3. The method according to claim 2, characterized in that, The determining the defect position and defect size according to the wavelet transform result includes: According to the wavelet transform result, determining the energy and the wave packet centroid time of the low-frequency component pulse signal in the reflected echo signal; Determining the defect position and defect size according to the energy and the wave packet centroid time.

4. The method according to claim 3, wherein The method further includes: Comparing the energy with a preset energy threshold, and if the energy is greater than or equal to the preset energy threshold, determining that there is a defect in the metal thin plate.

5. The method according to claim 3, wherein The method further includes: When it is determined that there is a defect in the metal thin plate, generating a grayscale image based on the energy of the low-frequency component pulse signal in the reflected echo signal; Analyzing the shape characteristics of the target area in the grayscale image to obtain a shape characteristic analysis result, where the target area is a potential defect area; Determining the defect type based on the shape characteristic analysis result.

6. The method according to claim 1, wherein The performing feature analysis on the reflected echo signal to obtain a feature analysis result includes: Performing feature analysis on the reflected echo signal to determine the amplitude of the reflected echo signal, and the feature analysis result includes the amplitude of the reflected echo signal; According to the feature analysis result, detecting whether there is a defect in the metal thin plate to obtain a defect detection result of the metal thin plate, includes: Detecting whether there is a defect in the metal thin plate according to the amplitude of the reflected echo signal and a preset amplitude threshold to obtain a defect detection result of the metal thin plate.

7. The method according to any one of claims 1 to 6, characterized in that, The detecting whether there is a defect in the metal thin plate according to the feature analysis result to obtain a defect detection result of the metal thin plate further includes: Detecting whether there is a defect in the metal thin plate according to a preset defect discrimination criterion and the feature analysis result to obtain a defect detection result of the metal thin plate, where the preset defect discrimination criterion is determined based on feature extraction and quantitative analysis of the ultrasonic guided wave signal reflected by a metal thin plate containing known defects.

8. A metal sheet detection system, characterized in that, The system includes a processor, an ultrasonic guided wave probe, a conveying component, and a metal thin plate, and the processor is connected to the ultrasonic guided wave probe; The conveying component is used to convey the metal thin plate to a preset detection area; The ultrasonic guided wave probe is used to emit ultrasonic guided wave signals, transmit the ultrasonic guided wave signals into the metal sheet, receive the reflected echo signals of the metal sheet, and send the reflected echo signals to the processor when the metal sheet is conveyed to the preset detection area; The processor is used to execute the metal sheet detection method according to any one of claims 1 to 6, and perform defect detection on the metal sheet.

9. The system according to claim 8, wherein The ultrasonic guided wave probe further includes a coupling agent spraying device, an ultrasonic guided wave excitation and reception device, and a coupling agent recovery device; The coupling agent spraying device is used to spray the coupling agent onto the metal sheet; The ultrasonic guided wave excitation and reception device is used to generate an electrical signal with a preset frequency and waveform based on the material characteristics of the metal sheet, amplify the electrical signal, convert the amplified electrical signal into an ultrasonic guided wave signal, transmit the ultrasonic guided wave signal into the metal sheet, receive the reflected echo signals of the metal sheet, and send the reflected echo signals to the processor; The coupling agent recovery device is used to recover the coupling agent on the metal sheet.

10. A metal sheet detection device, characterized in that, The device includes: A signal acquisition module, which is used to control the ultrasonic guided wave probe to generate ultrasonic guided wave signals when the metal sheet is conveyed to the preset detection area, so that the ultrasonic guided wave signals propagate in the metal sheet; and acquire the reflected echo signals of the metal sheet; A feature analysis module, which is used to perform feature analysis on the reflected echo signals to obtain a feature analysis result; A defect detection module, which is used to detect whether there are defects in the metal sheet according to the feature analysis result and obtain the defect detection result of the metal sheet.

Images