Composite material defect detection method and system based on double-wave interference laser ultrasound

Abstract

The invention discloses a composite material defect detection method and system based on double-wave interference laser ultrasound, and relates to the technical field of laser ultrasound nondestructive testing, the composite material defect detection method and system based on double-wave interference laser ultrasound realize high-precision detection of composite material defects through double-wavelength laser interference and multi-mode signal processing; the dual-wavelength laser excitation module enables interference light spots to directionally excite sound waves of different depths through dynamic modulation of phase difference, the problem that traditional single-wavelength excitation depth resolution is insufficient is solved, and the depth detection precision reaches 0.1 mm; multi-mode sound wave collection is combined with a laser Doppler vibration meter and self-adaptive Kalman filtering, the signal-to-noise ratio is increased to 20 dB or above, 15-25 MHz high-frequency transverse waves and 1-5 MHz low-frequency longitudinal waves are effectively separated, and the bottleneck that traditional ultrasound is insufficient in micron-order defect sensitivity is broken through; and the signal processing module realizes type identification of defects such as interlayer debonding and fiber fracture through wavelet packet decomposition and DCNN classification, so that the accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of laser ultrasonic nondestructive testing, and in particular to a composite material defect detection method and system based on dual-wave interference laser ultrasound. Background Art

[0002] Existing composite material defect detection methods and systems based on dual-wave interference laser ultrasound still have the following drawbacks in actual use:

[0003] In the aviation and new energy sectors, carbon fiber composites (CFRP) and ceramic matrix composites (CMC) are widely used due to their excellent performance. However, they are prone to hidden defects such as interlaminar debonding and fiber breakage, which seriously affect the safety and reliability of components. Current detection technologies have significant limitations: contact ultrasonic testing requires the use of coupling agents, which not only damages the material surface but also has extremely low detection efficiency in scenarios such as aircraft skins. Laser ultrasonic single-point excitation can only excite single-frequency sound waves, which is insufficiently sensitive for detecting micron-level defects, with a signal-to-noise ratio often below 6dB. Although laser interferometry can achieve non-contact measurement, traditional single-wavelength interferometry is easily affected by material surface roughness and cannot effectively distinguish between near-surface and deep-seated defects. For example, it is difficult to distinguish between near-surface pores with a depth of less than 0.5mm and interlaminar debonding greater than 2mm. These technical bottlenecks make it difficult to accurately identify internal defects in composite materials. More efficient detection methods are urgently needed to meet the stringent requirements of the aviation and new energy sectors for composite component quality control. Summary of the Invention

[0004] The purpose of the present invention is to provide a composite material defect detection method and system based on dual-wave interference laser ultrasound to solve the above-mentioned problems.

[0005] In order to achieve the above object, the present invention provides the following technical solution: a composite material defect detection method based on dual-wave interference laser ultrasound, comprising the following steps:

[0006] S1. Dual-wavelength laser interference excitation:

[0007] Receive 532nm pulsed laser beam and 1064nm pulsed laser beam;

[0008] The phase difference between the 532nm pulsed laser beam and the 1064nm pulsed laser beam is modulated by a spatial light modulator to generate a dynamically adjustable interference spot. The fringe spacing of the interference spot is satisfy ,in is the wavelength, is the angle between the two beams;

[0009] Focusing the interference light spot on the surface of the composite material to excite a deep-directional acoustic wave signal;

[0010] S2. Multimodal sound wave acquisition:

[0011] receiving a vibration signal generated by the propagation of sound waves on the surface of the composite material;

[0012] converting the vibration signal into an electrical signal using a laser Doppler vibrometer;

[0013] Output the original sound wave signal containing high-frequency shear wave components and low-frequency longitudinal wave components;

[0014] S3. Defect feature decoupling:

[0015] receiving the original sound wave signal;

[0016] The 15-25MHz high-frequency band signal and the 1-5MHz low-frequency band signal are separated by wavelet packet decomposition algorithm;

[0017] Generate high-frequency feature vectors and low-frequency eigenvectors ;

[0018] S4. Defect depth inversion:

[0019] Receive the high frequency feature vector and low-frequency eigenvectors ;

[0020] Based on the coherence factor model Calculate defect depth ,in is the material constant, is the frequency ratio, is the attenuation coefficient;

[0021] Output defect depth coordinates and type classification results.

[0022] Furthermore, the generation of the dynamically adjustable interference spot in step S1 includes:

[0023] Receive preset target depth parameters ;

[0024] According to the formula Dynamically adjust the phase difference of the spatial light modulator, wherein is the speed of sound, is the laser repetition frequency);

[0025] Generate and Matched interference fringe density.

[0026] Furthermore, after step S2, the method further includes:

[0027] Signal enhancement steps:

[0028] receiving the original sound wave signal;

[0029] Eliminate environmental vibration noise through adaptive Kalman filter;

[0030] Output enhanced sound wave signal with signal-to-noise ratio ≥ 20dB.

[0031] Furthermore, the wavelet packet decomposition algorithm of step S3 includes:

[0032] Receiving the original sound wave signal ;

[0033] The 6-layer decomposition is performed using the db8 wavelet basis, extracting the node 3,0 corresponding to the 15-25MHz frequency band and the node 1,1 corresponding to the 1-5MHz frequency band;

[0034] Output the reconstructed high-band signal and low-band signals .

[0035] Furthermore, the defect type classification in step S4 includes:

[0036] Receive the high frequency feature vector and low-frequency eigenvectors ;

[0037] Will Input a pre-trained DCNN classifier, which contains 3 convolutional layers and 2 fully connected layers;

[0038] Output defect type label.

[0039] Furthermore, the energy density of the 532nm pulse laser is 1.5-2.0J / cm², and the energy density of the 1064nm pulse laser is 3.0-4.0J / cm².

[0040] Furthermore, in the defect depth inversion of step S4, the defect with a depth of ≤0.5 mm is For defects with a depth of ≥1mm, .

[0041] A composite material defect detection system using dual-wave interference laser ultrasound based on the method includes:

[0042] Dual-wavelength laser excitation module: including a 532nm pulsed laser source, a 1064nm pulsed laser source, and a spatial light modulator coupled to the 532nm pulsed laser source and the 1064nm pulsed laser source;

[0043] Acoustic wave acquisition module: comprising a laser Doppler vibrometer aimed at the surface of the composite material, wherein the output end of the laser Doppler vibrometer is connected to the signal conditioning circuit;

[0044] Signal processing module: including the following connected in sequence:

[0045] a) a wavelet packet decomposition unit configured to perform frequency band separation;

[0046] b) a coherence factor calculation unit configured to perform a depth inversion model;

[0047] c) a DCNN classification unit, configured to perform defect classification;

[0048] Three-dimensional imaging module: receives the output of the signal processing module and generates a defect spatial distribution map.

[0049] Furthermore, the dual-wavelength laser excitation module further includes:

[0050] an acousto-optic modulator, disposed in the optical path of the 532 nm pulsed laser source, for controlling the pulse width to be 5-10 ns;

[0051] A polarization beam splitter is used to combine the 532nm pulsed laser beam and the 1064nm pulsed laser beam into a coaxial beam.

[0052] Furthermore, the signal processing module further includes an adaptive filtering unit, and the adaptive filtering unit adopts a Kalman filter, whose state equation satisfies:

[0053]

[0054] in is the state transition matrix, is the observation matrix, 、 are process noise and observation noise, respectively.

[0055] Compared with the existing technology, the composite material defect detection method and system based on dual-wave interference laser ultrasound provided by the present invention have the following beneficial effects:

[0056] This composite material defect detection method and system based on dual-wave interference laser ultrasound achieves high-precision detection of composite material defects through dual-wavelength laser interference and multimodal signal processing. The dual-wavelength laser excitation module dynamically modulates the phase difference to make the interference spot directionally excite sound waves at different depths, alleviating the problem of insufficient depth resolution of traditional single-wavelength excitation and achieving a depth detection accuracy of 0.1mm. Multimodal sound wave acquisition combined with a laser Doppler vibrometer and adaptive Kalman filtering improves the signal-to-noise ratio to above 20dB, effectively separating 15-25MHz high-frequency shear waves and 1-5MHz low-frequency longitudinal waves, breaking through the bottleneck of traditional ultrasound's insufficient sensitivity to micron-level defects. The signal processing module uses wavelet packet decomposition and DCNN classification to realize the type identification of defects such as interlayer debonding and fiber breakage, improving the accuracy. The three-dimensional imaging module generates a defect distribution map with a spatial resolution of 0.2mm. The overall detection efficiency is improved compared with traditional contact ultrasound, meeting the needs of non-destructive testing of aviation composites. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] 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. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0058] Figure 1 It is a schematic diagram of the overall process of the present invention. DETAILED DESCRIPTION

[0059] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0060] See also Figure 1 A composite material defect detection method based on dual-wave interference laser ultrasound includes the following steps:

[0061] S1. Dual-wavelength laser interference excitation:

[0062] Receive 532nm pulsed laser beam and 1064nm pulsed laser beam;

[0063] The spatial light modulator modulates the phase difference between the 532nm pulsed laser beam and the 1064nm pulsed laser beam to generate a dynamically adjustable interference spot. The fringe spacing of the interference spot is satisfy ,in is the wavelength, is the angle between the two beams;

[0064] Focus the interference spot on the surface of the composite material to stimulate the acoustic wave signal in a directional depth;

[0065] S2. Multimodal sound wave acquisition:

[0066] Receive vibration signals generated by sound wave propagation on the surface of the composite material;

[0067] The vibration signal is converted into an electrical signal by a laser Doppler vibrometer;

[0068] Output the original sound wave signal containing high-frequency shear wave components and low-frequency longitudinal wave components;

[0069] S3. Defect feature decoupling:

[0070] Receive original sound wave signal;

[0071] The 15-25MHz high-frequency band signal and the 1-5MHz low-frequency band signal are separated by wavelet packet decomposition algorithm;

[0072] Generate high-frequency feature vectors and low-frequency eigenvectors ;

[0073] S4. Defect depth inversion:

[0074] Receive high-frequency feature vectors and low-frequency eigenvectors ;

[0075] Based on the coherence factor model Calculate defect depth ,in is the material constant, is the frequency ratio, is the attenuation coefficient;

[0076] Output defect depth coordinates and type classification results.

[0077] The generation of the dynamically adjustable interference spot in step S1 includes:

[0078] Receive preset target depth parameters ;

[0079] According to the formula Dynamically adjust the phase difference of the spatial light modulator, where is the speed of sound, is the laser repetition frequency);

[0080] Generate and Matched interference fringe density.

[0081] After step S2, the method further includes:

[0082] Signal enhancement steps:

[0083] Receive original sound wave signal;

[0084] Eliminate environmental vibration noise through adaptive Kalman filter;

[0085] Output enhanced sound wave signal with signal-to-noise ratio ≥ 20dB.

[0086] The wavelet packet decomposition algorithm of step S3 includes:

[0087] Receive original sound wave signal ;

[0088] The 6-layer decomposition is performed using the db8 wavelet basis, extracting the node 3,0 corresponding to the 15-25MHz frequency band and the node 1,1 corresponding to the 1-5MHz frequency band;

[0089] Output the reconstructed high-band signal and low-band signals .

[0090] The defect type classification in step S4 includes:

[0091] Receive high-frequency feature vectors and low-frequency eigenvectors ;

[0092] Will Input the pre-trained DCNN classifier, which contains 3 convolutional layers and 2 fully connected layers;

[0093] Output defect type label.

[0094] DCNN classifier implementation

[0095] Input layer: 128-dimensional feature vector (64-dimensional high-frequency + 64-dimensional low-frequency)

[0096] Convolutional layer 1: 16 3×3 convolution kernels, ReLU activation

[0097] Convolutional layer 2: 32 3×3 convolution kernels, ReLU activation

[0098] Convolutional layer 3: 64 3×3 convolution kernels, ReLU activation

[0099] Fully connected layer 1: 128 neurons, Dropout (0.5)

[0100] Fully connected layer 2: 3 neurons (corresponding to three defect types: debonding, fiber breakage, and pores).

[0101] The energy density of 532nm pulsed laser is 1.5-2.0J / cm², and the energy density of 1064nm pulsed laser is 3.0-4.0J / cm².

[0102] In the defect depth inversion of step S4, the defect with a depth of ≤0.5 mm is For defects with a depth of ≥1mm, .

[0103] A composite material defect detection system based on a dual-wave interference laser ultrasound method, characterized by comprising:

[0104] Dual-wavelength laser excitation module: includes a 532nm pulsed laser source, a 1064nm pulsed laser source, and a spatial light modulator coupled to the 532nm pulsed laser source and the 1064nm pulsed laser source;

[0105] Acoustic wave acquisition module: including a laser Doppler vibrometer aimed at the surface of the composite material, the output end of the laser Doppler vibrometer is connected to the signal conditioning circuit;

[0106] Signal processing module: including the following connected in sequence:

[0107] a) a wavelet packet decomposition unit configured to perform frequency band separation;

[0108] b) a coherence factor calculation unit configured to perform a depth inversion model;

[0109] c) a DCNN classification unit, configured to perform defect classification;

[0110] 3D imaging module: receives the output of the signal processing module and generates a defect spatial distribution map.

[0111] The dual-wavelength laser excitation module also includes:

[0112] An acousto-optic modulator (AOM) is placed in the optical path of a 532nm pulsed laser source to control the pulse width to 5-10ns.

[0113] A polarization beam splitter is used to combine a 532nm pulsed laser beam and a 1064nm pulsed laser beam into a coaxial beam.

[0114] The signal processing module further includes an adaptive filtering unit, which uses a Kalman filter, and its state equation satisfies:

[0115]

[0116] in is the state transition matrix, is the observation matrix, 、 are process noise and observation noise, respectively.

[0117] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

Claims

1. A composite material defect detection method based on dual-wave interference laser ultrasound, characterized in that: The following steps are involved: S1. Dual-wavelength laser interference excitation: Receive 532nm pulsed laser beam and 1064nm pulsed laser beam; The phase difference between the 532nm pulsed laser beam and the 1064nm pulsed laser beam is modulated by a spatial light modulator to generate a dynamically adjustable interference spot. The fringe spacing of the interference spot is satisfy ,in is the wavelength, is the angle between the two beams; Focusing the interference light spot on the surface of the composite material to excite a deep-directional acoustic wave signal; S2. Multimodal sound wave acquisition: receiving a vibration signal generated by the propagation of sound waves on the surface of the composite material; converting the vibration signal into an electrical signal using a laser Doppler vibrometer; Output the original sound wave signal containing high-frequency shear wave components and low-frequency longitudinal wave components; S3. Decoupling of defect features: receiving the original sound wave signal; The 15-25MHz high-frequency band signal and the 1-5MHz low-frequency band signal are separated by wavelet packet decomposition algorithm; Generate high-frequency feature vectors and low-frequency eigenvectors ; S4. Defect depth inversion: Receive the high frequency feature vector and low-frequency eigenvectors ; Based on the coherence factor model Calculate defect depth ,in is the material constant, is the frequency ratio, is the attenuation coefficient; Output defect depth coordinates and type classification results.

2. The composite material defect detection method based on dual-wave interference laser ultrasound according to claim 1, characterized in that: The generation of the dynamically adjustable interference spot in step S1 includes: Receive preset target depth parameters ; According to the formula Dynamically adjust the phase difference of the spatial light modulator, wherein is the speed of sound, is the laser repetition frequency); Generate and Matched interference fringe density.

3. The composite material defect detection method based on dual-wave interference laser ultrasound according to claim 1, characterized in that: After step S2, the method further includes: Signal enhancement steps: receiving the original sound wave signal; Eliminate environmental vibration noise through adaptive Kalman filter; Output enhanced sound wave signal with signal-to-noise ratio ≥ 20dB.

4. The composite material defect detection method based on dual-wave interference laser ultrasound according to claim 1, characterized in that: The wavelet packet decomposition algorithm of step S3 includes: Receiving the original sound wave signal ; The 6-layer decomposition is performed using the db8 wavelet basis, extracting the node 3,0 corresponding to the 15-25MHz frequency band and the node 1,1 corresponding to the 1-5MHz frequency band; Output the reconstructed high-band signal and low-band signals .

5. The composite material defect detection method based on dual-wave interference laser ultrasound according to claim 1, characterized in that: The defect type classification in step S4 includes: Receive the high frequency feature vector and low-frequency eigenvectors ; Will Input a pre-trained DCNN classifier, which contains 3 convolutional layers and 2 fully connected layers; Output defect type label.

6. The composite material defect detection method based on dual-wave interference laser ultrasound according to claim 1, characterized in that: The energy density of the 532nm pulse laser is 1.5-2.0J / cm², and the energy density of the 1064nm pulse laser is 3.0-4.0J / cm².

7. The composite material defect detection method based on dual-wave interference laser ultrasound according to claim 1, characterized in that: In the defect depth inversion of step S4, the defect with a depth of ≤ 0.5 mm is inverted. For defects with a depth of ≥1mm, .

8. A composite material defect detection system based on the method according to any one of claims 1 to 7 using dual-wave interference laser ultrasound, characterized in that: include: Dual-wavelength laser excitation module: including a 532nm pulsed laser source, a 1064nm pulsed laser source, and a spatial light modulator coupled to the 532nm pulsed laser source and the 1064nm pulsed laser source; Acoustic wave acquisition module: comprising a laser Doppler vibrometer aimed at the surface of the composite material, wherein the output end of the laser Doppler vibrometer is connected to the signal conditioning circuit; Signal processing module: including the following connected in sequence: a) a wavelet packet decomposition unit configured to perform frequency band separation; b) a coherence factor calculation unit configured to perform a depth inversion model; c) a DCNN classification unit, configured to perform defect classification; Three-dimensional imaging module: receives the output of the signal processing module and generates a defect spatial distribution map.

9. The composite material defect detection system based on dual-wave interference laser ultrasound according to claim 8, characterized in that: The dual-wavelength laser excitation module also includes: an acousto-optic modulator, disposed in the optical path of the 532 nm pulsed laser source, for controlling the pulse width to be 5-10 ns; A polarization beam splitter is used to combine the 532nm pulsed laser beam and the 1064nm pulsed laser beam into a coaxial beam.

10. The composite material defect detection system based on dual-wave interference laser ultrasound according to claim 8, characterized in that: The signal processing module further includes an adaptive filtering unit, which uses a Kalman filter, and its state equation satisfies: in is the state transition matrix, is the observation matrix, 、 are process noise and observation noise, respectively.

Images