Defect detection method and device based on multi-modal Lamb waves

Abstract

The invention relates to the technical field of sensors and nondestructive testing, in particular to a defect detection method and device based on multi-mode Lamb waves. In order to solve the problem that effective modal signals corresponding to defects are difficult to accurately identify due to the multi-modal characteristics of the Lamb waves, the invention provides a novel defect detection method and device based on the multi-modal Lamb waves, and the method comprises the following steps: 1) simultaneously exciting two kinds of Lamb waves with different frequencies in a sample; 2) receiving detection echoes of the sample and forming a time domain signal diagram; 3) generating a time-frequency graph by using a wavelet transform efficient time-frequency analysis means; 4) two frequency values are selected, frequency straight lines parallel to the time axis are constructed respectively, the two frequency straight lines are used as references, slicing operation in the vertical direction is carried out on the generated time-frequency graph, and finally a slice graph is obtained; and 5) analyzing the slice map to obtain the position of the defect of the sample. The method provided by the invention can realize accurate identification and positioning of the defect.

Description

Technical Field

[0001] The present invention relates to the field of sensors and non-destructive testing technology, and in particular to a defect detection method and device based on multimodal Lamb waves. Background Art

[0002] Nondestructive testing (NDT) technology accurately identifies internal defects in materials and structures through non-destructive techniques, providing key support for high-end manufacturing quality control, the safe operation of industrial equipment, and infrastructure lifespan assessment. As a new technology in NDT, electromagnetic ultrasonic guided waves (EUSG) utilize the interaction of electromagnetic induction and ultrasonic effects to generate ultrasonic guided waves that propagate along the surface or internal structure of a material to detect potential defects. This technology offers high detection efficiency, the ability to rapidly scan a large area using a single-point excitation, ease of operation, the absence of coupling agents, and compatibility with high temperatures.

[0003] Lamb waves, a type of ultrasonic guided wave, are sensitive to defects such as delamination, cracks, and corrosion, and are widely used in nondestructive testing of various materials or structures, such as wing skin testing in the aerospace field, thin body structure testing in the automotive and rail transportation industries, and thin plate material testing in the electronics industry. However, the multimodal nature of Lamb waves has always been a key obstacle to their application and development. The multimodal nature of Lamb waves is mainly manifested in the simultaneous excitation of multiple modes at the same excitation frequency, such as the symmetric modes S0 and S1 and the antisymmetric modes A0 and A1, as shown in [1]. Figure 1 As shown in , due to the differences in velocity and dispersion characteristics of different modes and the non-uniformity of energy distribution, the modal aliasing is serious, making it difficult to accurately identify the defect information corresponding to the effective modal signal, such as Figure 2 As shown ( Figure 2 The traditional detection method described in the preceding text uses Lamb waves of a single frequency to excite the defect, which in turn affects the reliability of defect location, quantification, and type judgment.

[0004] Existing technologies suppress non-target modes from the wave source by optimizing the structural parameters of electromagnetic ultrasonic transducers (such as coil layout and permanent magnet configuration) or use signal processing algorithms to perform modal separation on existing multi-modal signals. However, optimizing transducer parameters can only suppress a limited number of modes. For example, at low frequencies, Lamb waves exist in both A0 and S0 modes, and non-destructive testing can be performed by suppressing one of them. However, when Lamb waves exist in multiple modes (more than three) at the same time, it is impossible to suppress all non-target modes. The use of algorithms to process multi-modal signals relies on an accurate dispersion model and has high computational complexity, making it difficult to adapt to the needs of rapid on-site detection. Summary of the Invention

[0005] In order to solve the problem that the multimodal characteristics of Lamb waves make it difficult to accurately identify the effective modal signals corresponding to defects, the present invention provides a new defect detection method and device based on multimodal Lamb waves.

[0006] The present invention is achieved by adopting the following technical solutions: A defect detection method based on multimodal Lamb waves comprises the following steps: 1) Simultaneously excite two Lamb waves of different frequencies in the sample; 2) Receive the detection echo of the sample and form a time domain signal graph; 3) Using wavelet transform, an efficient time-frequency analysis method, the signal is mapped from the time domain to the time-frequency domain, generating a time-frequency graph showing how the signal frequency changes over time; 4) Select two frequency values that are consistent with the two excitation frequencies in step 1) and construct frequency lines parallel to the time axis. Using the two frequency lines as references, slice the generated time-frequency graph vertically (the vertical direction is parallel to the amplitude direction) to obtain a slice diagram of the modal signal amplitude over time. 5) Analyze the slice diagram based on the similar characteristics of A0 modal velocity detected at different excitation frequencies to obtain the location of the sample defect.

[0007] Furthermore, the two Lamb waves of different frequencies excited in step 1) are composed of a high-frequency Lamb wave and a low-frequency Lamb wave, and the high-frequency Lamb wave is convenient for detecting smaller defects.

[0008] Furthermore, the frequency of the high-frequency Lamb wave is 1.13 MHz, and the frequency of the low-frequency Lamb wave is 510 KHz.

[0009] A multi-modal Lamb wave-based defect detection device includes a dual-frequency excitation sensor and a dual-frequency receiving sensor, a dual-channel ultrasonic flaw detector, and a host computer; The dual-frequency excitation sensor is used to excite Lamb waves of two different frequencies, and the dual-frequency receiving sensor is used to receive detection echoes. The dual-frequency excitation sensor is connected to the two excitation ports of the dual-channel ultrasonic flaw detector, the dual-frequency receiving sensor is connected to the receiving port of the dual-channel ultrasonic flaw detector, and the communication port of the dual-channel ultrasonic flaw detector is connected to the host computer; The host computer is used to send command signals to the dual-channel ultrasonic flaw detector, receive and store the digital signals sent by the dual-channel flaw detector, and convert the digital signals into time-domain signal graphs. Then, the wavelet transform is used to efficiently analyze the time and frequency, mapping the signal from the time domain to the time-frequency domain to generate a time-frequency graph that can clearly show the time-varying characteristics of the signal frequency. Finally, two excitation frequency values are selected and frequency lines parallel to the time axis are constructed respectively. The generated time-frequency graph is sliced in the vertical direction (the vertical direction is parallel to the amplitude) based on the two frequency lines, and finally a slice graph showing the time-varying characteristics of the modal signal amplitude is obtained.

[0010] During use, an aluminum plate is used as a sample, and a dual-frequency excitation sensor and a dual-frequency receiving sensor are placed on the aluminum plate in a straight line along the length direction of the aluminum plate. The dual-channel ultrasonic flaw detector inputs two excitation electrical signals of different frequencies to the dual-frequency excitation sensor. The dual-frequency excitation sensor excites two Lamb waves of different frequencies in the aluminum plate. The dual-frequency receiving sensor then receives the detection echo, which is amplified and filtered by the dual-channel ultrasonic flaw detector and discretized into a digital signal. The digital signal is then transmitted to the host computer, which converts the digital signal into a time domain signal graph and generates a slice graph reflecting the details of the change of the marker modal signal over time based on the time domain signal graph. Signal features are extracted by analyzing the slice graph. The signal features include the flight time of the marker modal direct wave and the flight time of the defect echo. Finally, the defect information within the straight line range of the two sensors on the aluminum plate is obtained by signal feature analysis (signal feature analysis is common knowledge among those skilled in the art, that is, by the formula d=v×t q Obtain the defect location, where v=H / t z , v is the propagation velocity of the marker mode, H is the distance between the dual-frequency excitation sensor and the dual-frequency receiving sensor, t z is the direct wave flight time of the marker mode, t q is the defect echo flight time, d is the distance between the dual-frequency excitation sensor and the defect). After the defect detection corresponding to the linear position is completed, the dual-frequency excitation sensor and the dual-frequency receiving sensor are simultaneously moved along the width direction of the aluminum plate to another linear position to detect the defects corresponding to the linear position until all defects of the aluminum plate are detected.

[0011] Furthermore, the dual-frequency excitation sensor includes a double-pitch laminated cooperative coil and an excitation permanent magnet pressed thereon. The double-pitch laminated cooperative coil includes two excitation folded coils with different folding pitches (as is known to those skilled in the art: the folded coils described in this application are coils arranged in a rectangular wave folding pattern), stacked one on top of the other. The double-pitch laminated cooperative coil is designed to input excitation electrical signals of two different frequencies.

[0012] Furthermore, the two excitation folded coils are stacked up and down through a PCB process, thereby improving spatial coupling efficiency.

[0013] Furthermore, the dual-frequency excitation sensor includes a first upper shell and a first lower shell. The first upper shell is a rectangular cover with an opening downward, and first pressure strips for pressing the excitation permanent magnet between the first upper shell and the first lower shell are vertically fixed on all four sides of the inner side of the cover top. The first lower shell is a rectangular box structure with an opening upward. First connecting holes are provided at the four corners of the first upper shell and the first lower shell. The first fastening bolt passes through the first connecting holes of the first upper shell and the first lower shell and is screwed to the first fastening nut to achieve fixed connection between the first upper shell and the first lower shell, thereby facilitating the integration and standardization of the sensor structure.

[0014] Furthermore, the folding distance between the excitation return coils located in the upper layer is 2.5 mm, and the folding distance between the excitation return coils located in the lower layer is 1.25 mm.

[0015] Furthermore, the dual-frequency receiving sensor includes a receiving loop coil and a magnet array pressed against the receiving loop coil. The magnet array consists of multiple receiving permanent magnets arranged side by side along the loop direction of the receiving loop coil. The pitch of the receiving loop coil is the least common multiple of the A0 mode wavelengths at the two excitation frequencies. The pitch of the receiving loop coil is designed to simultaneously receive guided wave signals at two frequencies.

[0016] Furthermore, the dual-frequency receiving sensor includes a second upper shell and a second lower shell. The second upper shell is a rectangular cover with an opening downward, and second pressure strips for pressing the array magnet between the second upper shell and the second lower shell are vertically fixed on all four sides of the inner side of the cover top. The second lower shell is a rectangular box structure with an opening upward. Second connecting holes are provided at the four corners of the second upper shell and the second lower shell. The second fixing bolt passes through the second connecting holes of the second upper shell and the second lower shell and is screwed to the second fixing nut to achieve fixed connection between the second upper shell and the second lower shell, thereby facilitating the integration and standardization of the sensor structure.

[0017] The beneficial effects of the present invention are as follows: the detection method and device of the present invention cleverly utilize the similar velocity characteristics of the A0 mode to perform defect detection, overcoming the difficult problems of accurate identification of effective modal signals corresponding to the multimodal characteristics of Lamb waves and difficulty in locating defects, so that defects can be accurately located; at the same time, this detection method can still accurately identify and locate defects even when there are many Lamb wave modes (more than 3 types) excited, and it consumes less computing resources and can be adapted to rapid on-site detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1 Schematic diagram of the dispersion curve of Lamb wave in a 3mm aluminum plate; Figure 2 This is the time domain signal diagram obtained by using the traditional detection method; Figure 3Flow chart of the detection method of the present invention; Figure 4 Schematic diagram of the structure of the dual-frequency excitation sensor of the present invention; Figure 5 This is a schematic structural diagram of the dual-frequency receiving sensor of the present invention; Figure 6 This is a schematic diagram of the arrangement of the two sensors of the present invention when in use; Figure 7 A time domain signal diagram detected by the detection method of the present invention; Figure 8 This is the time-frequency diagram obtained by transforming the detection signal according to the present invention: Figure 9 This is the sliced graph obtained by slicing the time-frequency graph as described in the present invention.

[0021] In the figure: 1-aluminum plate, 2-dual-frequency excitation sensor, 3-dual-frequency receiving sensor, 4-defect, 5-first upper shell, 6-first pressure strip, 7-excitation permanent magnet, 8-double-pitch laminated cooperative coil, 9-first lower shell, 10-second upper shell, 11-second pressure strip, 12-array magnet, 13-receiving return coil, 14-second lower shell, 15-marker modal direct wave, 16-marker modal defect echo. DETAILED DESCRIPTION

[0022] In order to more clearly understand the above-mentioned objectives, features and advantages of the present invention, the scheme of the present invention will be further described below. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features therein can be combined with each other.

[0023] In the description, it should be noted that the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance. It should be noted that, unless otherwise expressly specified or limited, the terms "installed," "connected," and "connected" should be understood broadly. For example, they can refer to fixed connections, removable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms based on specific circumstances.

[0024] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present invention, rather than all the embodiments.

[0025] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0026] A defect detection method based on multimodal Lamb waves, such as Figure 3 The following steps are shown: 1) Simultaneously excite two Lamb waves of different frequencies in the sample; 2) Receive the detection echo of the sample and form a time domain signal graph; 3) Using wavelet transform, an efficient time-frequency analysis method, the signal is mapped from the time domain to the time-frequency domain, generating a time-frequency graph showing how the signal frequency changes over time; 4) Select two frequency values that are consistent with the two excitation frequencies in step 1) and construct frequency lines parallel to the time axis. Using the two frequency lines as references, slice the generated time-frequency graph vertically (the vertical direction is the direction parallel to the amplitude of the time-frequency graph) to obtain a slice graph showing the change of the modal signal amplitude over time. 5) The slice diagram is analyzed based on the similar characteristics of the A0 modal velocity detected at different excitation frequencies to obtain the location of the sample defect 4.

[0027] In a specific implementation, the two Lamb waves of different frequencies excited in step 1) are composed of a high-frequency Lamb wave and a low-frequency Lamb wave. The high-frequency Lamb wave is convenient for detecting smaller defects 4.

[0028] In a specific implementation, the frequency of the high-frequency Lamb wave is 1.13 MHz, and the frequency of the low-frequency Lamb wave is 510 KHz.

[0029] A multi-modal Lamb wave-based defect detection device includes a dual-frequency excitation sensor 2 and a dual-frequency receiving sensor 3, a dual-channel ultrasonic flaw detector, and a host computer; The dual-frequency excitation sensor 2 is used to excite Lamb waves of two different frequencies, and the dual-frequency receiving sensor 3 is used to receive the detection echo. The dual-frequency excitation sensor 2 is connected to the two excitation ports of the dual-channel ultrasonic flaw detector, and the dual-frequency receiving sensor 3 is connected to the receiving port of the dual-channel ultrasonic flaw detector. The communication port of the dual-channel ultrasonic flaw detector is connected to the host computer; The host computer is used to send command signals to the dual-channel ultrasonic flaw detector, receive and store the digital signals sent by the dual-channel flaw detector, and convert the digital signals into time-domain signal graphs. Then, the wavelet transform is used to efficiently analyze the time and frequency, mapping the signal from the time domain to the time-frequency domain to generate a time-frequency graph that can clearly show the time-varying characteristics of the signal frequency. Finally, two excitation frequency values are selected and frequency lines parallel to the time axis are constructed respectively. The generated time-frequency graph is sliced in the vertical direction (the vertical direction is parallel to the amplitude) based on the two frequency lines, and finally a slice graph showing the time-varying characteristics of the modal signal amplitude is obtained.

[0030] When implementing it specifically, Figure 4As shown, the dual-frequency excitation sensor 2 includes a double-pitch laminated cooperative coil 8 and an excitation permanent magnet 7 pressed thereon. The double-pitch laminated cooperative coil 8 comprises two excitation folded coils with different folding pitches (as is known to those skilled in the art: the folded coils described herein are coils arranged in a rectangular wave pattern), stacked one on top of the other. The double-pitch laminated cooperative coil 8 is designed to input excitation electrical signals of two different frequencies.

[0031] In a specific implementation, the two excitation folded coils are stacked up and down through a PCB process, thereby improving the spatial coupling efficiency.

[0032] In specific implementation, the dual-frequency excitation sensor 2 includes a first upper shell 5 and a first lower shell 9. The first upper shell 5 is a rectangular cover with an opening downward, and the inner sides of the top of the cover are vertically fixed with first pressure strips 6 for pressing the excitation permanent magnet 7 between the first upper shell 5 and the first lower shell 9. The first lower shell 9 is a rectangular box structure with an opening upward. The four corners of the first upper shell 5 and the first lower shell 9 are provided with first connecting holes. The first fastening bolt passes through the first connecting holes of the first upper shell 5 and the first lower shell 9 and is screwed to the first fastening nut to achieve fixed connection between the first upper shell 5 and the first lower shell 9, which facilitates the integration and standardization of the sensor structure.

[0033] In a specific implementation, the folding distance between the excitation return coils located in the upper layer is 2.5 mm, and the folding distance between the excitation return coils located in the lower layer is 1.25 mm.

[0034] When implementing it specifically, Figure 5 As shown, the dual-frequency receiving sensor 3 includes a receiving folded coil 13 and an array magnet 12 pressed against the receiving folded coil 13. The array magnet 12 is composed of multiple receiving permanent magnets arranged side by side along the folded direction of the receiving folded coil 13. The fold spacing of the receiving folded coil 13 is the least common multiple of the A0 mode wavelengths at the two excitation frequencies. The fold spacing of the receiving folded coil 13 is designed to enable simultaneous reception of guided wave signals at two frequencies.

[0035] During specific implementation, the dual-frequency receiving sensor 3 includes a second upper shell 10 and a second lower shell 14. The second upper shell 10 is a rectangular cover with an opening downward, and second pressure strips 11 for pressing the array magnet 12 between the second upper shell 10 and the second lower shell 14 are vertically fixed on all four sides of the inner side of the cover top. The second lower shell 14 is a rectangular box structure with an opening upward. Second connecting holes are provided at the four corners of the second upper shell 10 and the second lower shell 14. The second fixing bolt passes through the second connecting holes of the second upper shell 10 and the second lower shell 14 and is screwed to the second fixing nut to achieve fixed connection between the second upper shell 10 and the second lower shell 14, which facilitates the integration and standardization of the sensor structure.

[0036] When in use, take a 3mm aluminum plate 1 as a sample, place the dual-frequency excitation sensor 2 and the dual-frequency receiving sensor 3 in a straight line along the length direction of the aluminum plate 1. Figure 6 As shown, the dual-channel ultrasonic flaw detector inputs two excitation electrical signals of different frequencies (1.13MHz and 510KHz respectively) to the dual-frequency excitation sensor 2. The dual-frequency excitation sensor 2 excites two Lamb waves of different frequencies in the aluminum plate 1. Then the dual-frequency receiving sensor 3 receives the detection echo, which is amplified and filtered by the dual-channel ultrasonic flaw detector and discretized into a digital signal. The digital signal is then transmitted to the host computer, which converts the digital signal into a time domain signal as shown in the figure. Figure 7 As shown, and according to the time domain signal diagram, it is converted into a time-frequency diagram as shown Figure 8 Finally, a slice diagram reflecting the time-varying details of the marker modal signal is generated. Figure 9 As shown. By analyzing the slice diagram, signal features are extracted. The signal features include the flight time of the direct wave 15 of the marker mode and the flight time of the defect echo 16 of the marker mode. Finally, the information of the defect 4 within the straight line range of the two sensors on the aluminum plate 1 is obtained through signal feature analysis (signal feature analysis is common knowledge among those skilled in the art, that is, by the formula d=v×t q Obtain the position of defect 4, where v=H / t z , v is the propagation velocity of the marker mode, H is the distance between the dual-frequency excitation sensor 2 and the dual-frequency receiving sensor 3, t z is the flight time of the direct wave 15 of the marked mode, t q is the flight time of the mark modal defect echo 16, d is the distance between the dual-frequency excitation sensor 2 and the defect 4). After the detection of the defect 4 corresponding to the linear position is completed, the dual-frequency excitation sensor 2 and the dual-frequency receiving sensor 3 are simultaneously moved along the width direction of the aluminum plate 1 to another linear position to detect the defect 4 corresponding to the linear position until all defects 4 of the aluminum plate 1 are detected.

[0037] compared to Figure 2 The various modal wave packets in the mixed, can not identify the defect echo wave packet to obtain defect 4 information, and by Figure 9 The defect 4 information can be accurately detected and analyzed.

[0038] The above description is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Although detailed descriptions have been made with reference to the aforementioned embodiments, those skilled in the art should understand that they may still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents; and such modifications or replacements do not deviate from the essence of the corresponding technical solutions within the scope of the technical solutions of the embodiments, and they should all be included in the scope of protection of the claims.

Claims

1. A defect detection method based on multimodal Lamb waves, characterized in that: The steps include: 1) Simultaneously excite two Lamb waves of different frequencies in the sample; 2) Receive the detection echo of the sample and form a time domain signal graph; 3) Using wavelet transform, an efficient time-frequency analysis method, the signal is mapped from the time domain to the time-frequency domain, generating a time-frequency graph showing how the signal frequency changes over time; 4) Select two frequency values that are consistent with the two excitation frequencies in step 1) and construct frequency lines parallel to the time axis. Use the two frequency lines as references to slice the generated time-frequency diagram vertically, and finally obtain a slice diagram of the modal signal amplitude changing with time. 5) Analyze the slice diagram based on the similar characteristics of A0 modal velocity detected at different excitation frequencies to obtain the location of the sample defect (4).

2. The defect detection method based on multimodal Lamb waves according to claim 1, characterized in that: The two Lamb waves of different frequencies excited in step 1) are composed of a high-frequency Lamb wave and a low-frequency Lamb wave.

3. The defect detection method based on multimodal Lamb waves according to claim 2, characterized in that: The frequency of the high-frequency Lamb wave is 1.13 MHz, and the frequency of the low-frequency Lamb wave is 510 KHz.

4. A defect detection device based on multimodal Lamb waves, characterized in that: It includes a dual-frequency excitation sensor (2) and a dual-frequency receiving sensor (3), a dual-channel ultrasonic flaw detector, and a host computer; The dual-frequency excitation sensor (2) is used to excite Lamb waves of two different frequencies, and the dual-frequency receiving sensor (3) is used to receive the detection echo. The dual-frequency excitation sensor (2) is connected to the two excitation ports of the dual-channel ultrasonic flaw detector, the dual-frequency receiving sensor (3) is connected to the receiving port of the dual-channel ultrasonic flaw detector, and the communication port of the dual-channel ultrasonic flaw detector is connected to the host computer; The host computer is used to send command signals to the dual-channel ultrasonic flaw detector, receive and store the digital signals sent by the dual-channel flaw detector, and convert the digital signals into time-domain signal graphs. Then, the wavelet transform is used to efficiently analyze the time and frequency, and the signal is mapped from the time domain to the time-frequency domain to generate a time-frequency graph that can clearly show the characteristics of the signal frequency changing with time. Finally, two excitation frequency values are selected, and frequency lines parallel to the time axis are constructed respectively. The two frequency lines are used as references to perform vertical slicing operations on the generated time-frequency graphs, and finally a slice graph showing the change of the modal signal amplitude with time is obtained.

5. The defect detection device based on multimodal Lamb waves according to claim 4, characterized in that: The dual-frequency excitation sensor (2) includes a double-pitch laminated cooperative coil (8) and an excitation permanent magnet (7) pressed thereon. The double-pitch laminated cooperative coil (8) includes two excitation folded coils with different folding intervals, and the two excitation folded coils are stacked up and down.

6. The defect detection device based on multimodal Lamb waves according to claim 5, characterized in that: The two excitation folded coils are stacked up and down using a PCB process.

7. The defect detection device based on multimodal Lamb waves according to claim 6, characterized in that: The dual-frequency excitation sensor (2) comprises a first upper shell (5) and a first lower shell (9). The first upper shell (5) is a rectangular cover with an opening facing downward, and first pressure strips (6) for pressing the excitation permanent magnet (7) between the first upper shell (5) and the first lower shell (9) are vertically fixed on all four sides of the inner side of the cover top. The first lower shell (9) is a rectangular box structure with an opening facing upward. The first upper shell (5) and the first lower shell (9) are provided with first connecting holes at their four corners. A first fixing bolt passes through the first connecting holes of the first upper shell (5) and the first lower shell (9) and is screwed onto a first fixing nut to achieve fixed connection between the first upper shell (5) and the first lower shell (9).

8. The defect detection device based on multimodal Lamb waves according to claim 7, characterized in that: The folding distance between the excitation return coils located in the upper layer is 2.5 mm, and the folding distance between the excitation return coils located in the lower layer is 1.25 mm.

9. The defect detection device based on multimodal Lamb waves according to claim 8, characterized in that: The dual-frequency receiving sensor (3) includes a receiving folded coil (13) and an array magnet (12) pressed on the receiving folded coil (13), wherein the array magnet (12) is composed of a plurality of receiving permanent magnets arranged side by side along the folding direction of the receiving folded coil (13), and the folding distance of the receiving folded coil (13) is the least common multiple of the A0 mode wavelength at two excitation frequencies.

10. The defect detection device based on multi-modal Lamb waves according to claim 9, characterized in that: The dual-frequency receiving sensor (3) comprises a second upper shell (10) and a second lower shell (14). The second upper shell (10) is a rectangular cover with an opening facing downward, and second pressure strips (11) for pressing the array magnet (12) between the second upper shell (10) and the second lower shell (14) are vertically fixed on all four sides of the inner side of the cover top. The second lower shell (14) is a rectangular box structure with an opening facing upward. Second connection holes are provided at the four corners of the second upper shell (10) and the second lower shell (14). A second fixing bolt passes through the second connection holes of the second upper shell (10) and the second lower shell (14) and is screwed onto a second fixing nut to achieve fixed connection between the second upper shell (10) and the second lower shell (14).

Images