Ultrasonic-guided-wave-based rapid damage imaging method for aerospace composites

Abstract

An ultrasonic-guided-wave-based rapid damage imaging method for aerospace composites. The method comprises the following steps: on the basis of material parameters of a carbon fiber composite plate, calculating a dispersion curve thereof and obtaining an optimal ultrasonic excitation angle and an ultrasonic guided wave propagation mode (S1); sequentially scanning and collecting ultrasonic guided wave signals in two orthogonal directions (S2); decomposing the guided wave signals, and extracting a target component on the basis of the magnitude of energy entropy to obtain reconstructed ultrasonic guided wave signals (S3); calculating time-frequency information and energy of the signals and, in combination with damage indexes, estimating damage probabilities in scanning paths (S4); embedding dilated kernel convolution and a dual probability distribution function into damage index curves, so as to enhance the damage indexes and sharpen edges (S5); and expanding the damage index curves in the two orthogonal directions into two-dimensional damage matrices, and performing imaging according to a composite path cross-imaging method, thereby ultimately realizing the localization and imaging of damage in the carbon fiber composite plate (S6).

Description

A rapid ultrasonic guided wave damage imaging method for aerospace composite materials Technical Field

[0001] This invention relates to the field of nondestructive testing, and in particular to a rapid ultrasonic guided wave damage imaging method for aerospace composite materials. Background Technology

[0002] Carbon fiber composites are widely used in various fields such as aircraft fuselages, high-speed trains, and subway bodies due to their advantages such as lightweight, high specific strength, high temperature resistance, corrosion resistance, and excellent design flexibility. However, due to the uncertainties in the complex manufacturing process and the external impact or fatigue loads suffered under extreme service environments, carbon fiber composites inevitably suffer from defects such as delamination, porosity, fiber breakage, fiber buckling, and fatigue cracks, which affect their mechanical properties and bring safety hazards. Therefore, it is particularly important to carry out research on health monitoring and damage detection technologies for carbon fiber composite structures.

[0003] As an elastic wave, ultrasonic guided waves have the advantages of long propagation distance, small amplitude loss and high damage sensitivity. Compared with the point-by-point scanning mode of traditional transmission and reflection methods, ultrasonic guided waves have higher detection efficiency for large-sized carbon fiber composite structural components. At the same time, the imaging of defects is often more intuitive than the ultrasonic shape parameters. Ultrasonic guided wave defect imaging has emerged, among which probabilistic damage imaging method has attracted widespread attention due to its simple calculation method and independence from prior knowledge of the structure. However, the use of existing technologies is greatly limited by the fact that the calibration of damage indicators requires manual selection, signal acquisition requires dense sensor arrays, and the imaging effect on multiple defects is poor.

[0004] Therefore, a rapid ultrasonic guided wave damage imaging method for aerospace composite materials is provided to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a rapid ultrasonic guided wave damage imaging method for aerospace composite materials, which overcomes the shortcomings of existing probabilistic damage imaging algorithms and enables accurate localization and imaging of delamination damage in carbon fiber composite laminates.

[0006] To achieve the above objectives, this invention provides a rapid ultrasonic guided wave damage imaging method for aerospace composite materials, comprising the following steps:

[0007] S1: Calculate the dispersion curve of the carbon fiber composite plate based on the material parameters of the carbon fiber composite plate, and calculate the optimal excitation angle and ultrasonic propagation mode based on the dispersion curve;

[0008] S2: Determine the scanning area, set up four scanning paths with different ultrasonic propagation directions, and scan and collect ultrasonic guided wave signals under the four scanning paths respectively;

[0009] S3: The ultrasonic guided wave signal is decomposed by fully adaptive noise set empirical mode decomposition (EMD) to obtain several ultrasonic guided wave signal components. The energy entropy of the ultrasonic guided wave signal components is calculated. Based on the energy entropy, the ultrasonic guided wave signal components are optimally reconstructed to obtain the reconstructed signal.

[0010] S4: Obtain the time-frequency information of the reconstructed signal through Hilbert transform, obtain the marginal spectrum through amplitude integration in the time-frequency information, calculate the damage index in the scanning path, and evaluate the degree of damage to the scanning path based on the damage index;

[0011] S5: The damage index in the scanning path is convolved using an expansion kernel and a probability distribution function, and the defect area of ​​the damage index curve is enhanced and the defect edge area of ​​the damage index curve is sharpened.

[0012] S6: Expand the damage index curves of the four scanning paths into two-dimensional image matrices, process the two-dimensional image matrices to obtain damage imaging image matrices, and fuse the damage imaging image matrices to obtain the final imaging result.

[0013] Preferably, in step S2, the ultrasonic propagation directions of the four scanning paths adopt two sets of orthogonal angles, namely 0° and 90°, and 45° and 135°.

[0014] Preferably, in step S3, the ultrasonic guided wave signal is decomposed by fully adaptive noise ensemble empirical mode decomposition (EMD) to obtain several ultrasonic guided wave signal components, specifically including the following steps:

[0015] Step 1: Incorporate Gaussian white noise into the ultrasonic guided wave signal to obtain the signal to be processed, x. i (t), x i (t) is represented as:

[0016] ;

[0017] Where x(t) represents the ultrasonic guided wave signal, μ represents the amplitude of Gaussian white noise, and c i (t) represents the Gaussian white noise of the i-th group;

[0018] Step 2: Decompose the signal x to be processed using EMD i (t), to obtain the first component IMF1 of each group, IMF1 is represented as:

[0019] ;

[0020] Where T represents the EMD decomposition period, the first component IMF1 of each group in the ultrasonic guided wave signal x(t) is removed to obtain the residual component r1(t) of the first stage, which is expressed as:

[0021] ;

[0022] Step 3: Add i sets of Gaussian white noise with amplitude μ to the residual component r1(t) of the first stage to obtain a new signal. , Represented as:

[0023] ;

[0024] For new signals EMD decomposition yields the first component IMF2 and the residual component r2(t) of the new signal in the second stage. IMF2 is expressed as:

[0025] ;

[0026] r2(t) is expressed as:

[0027] ;

[0028] Step 4: Repeat steps 1 to 3 until the residual can no longer be decomposed. At this point, the ultrasonic guided wave signal x(t) is represented as:

[0029] .

[0030] Preferably, in step S3, the energy entropy of the ultrasonic guided wave signal component is calculated, and the ultrasonic guided wave signal component is optimally reconstructed based on the energy entropy to obtain the reconstructed signal. This specifically includes the following steps:

[0031] S31: The energy entropy H(x) of the ultrasonic guided wave signal component is expressed as:

[0032] ;

[0033] Where P(x) is the probability density function, representing the proportion of the energy of a single ultrasonic guided wave signal component to the total energy;

[0034] S32: Sort the energy entropy H(x) in descending order, and select the ultrasonic guided wave signal components corresponding to the first five energy entropy H(x) for reconstruction. At this time, the ultrasonic guided wave signal x(t) is expressed as:

[0035] ;

[0036] in These are the first five ultrasonic guided wave signal components selected.

[0037] Preferably, in step S4, the calculation is performed using the Hilbert transform, specifically including the following steps:

[0038] S41: Perform a Hilbert transform on the preferred IMF component of the ultrasonic guided wave signal to obtain the principal value integral y(t), which is expressed as:

[0039] ;

[0040] Where P represents a slowly increasing distribution;

[0041] S42: Construct the analytic signal Z(t), and obtain the instantaneous frequency spectrum curve of the preferred IMF component of the ultrasonic guided wave signal based on the analytic signal Z(t). Z(t) is expressed as:

[0042] ;

[0043] Where a(t) is the instantaneous amplitude, θ(t) is the instantaneous phase, and ω(t) is the instantaneous frequency. Let be the instantaneous amplitude of the j-th order IMF component at time t and frequency j;

[0044] S43: The instantaneous amplitudes are combined in the time domain to obtain the Hilbert spectrum H(t,ω), which is expressed as:

[0045] ;

[0046] S44: Integrating the Hilbert spectrum over the time axis yields the Hilbert marginal spectrum h(t,ω), which is expressed as:

[0047] ;

[0048] S45: Obtain the damage index DI for each scan path based on the Hilbert spectrum. DI is represented as:

[0049] ;

[0050] Where E r E represents the average marginal spectral energy, and E is the marginal spectral energy value in a single scan step.

[0051] Preferably, in step S5, the probability distribution function P(d) is expressed as:

[0052] ;

[0053] Where d is the distance from the scanning path to the center line of the scanning area, σ is a parameter controlling the dispersion of the damage probability, μ is the center position of the normal distribution, D is the diameter of the ultrasonic transducer, and when σ is 2.4 and μ is +2, the probability distribution function is P. L (d) When σ is 2.4 and μ is -2, the probability distribution function is P R (d)

[0054] Preferably, step S5 specifically includes the following steps:

[0055] S51: Perform convolution operation on the damage exponential curve using a one-dimensional expansion kernel with size 1, length 11, and expansion rate of 4;

[0056] S52: When entering the damaged area, the high-weighted region is biased towards the leading edge of the scanning area, through the probability distribution function P. L (d) Perform convolution operation on the damage index curve, with the operation direction set from left to right;

[0057] S53: When leaving the damaged area, bias the high-weighted region towards the trailing edge of the scanned area, using the probability distribution function P. R (d) Perform convolution operation on the damage index curve, with the operation direction set from left to right.

[0058] Preferably, in step S6, the two-dimensional image is filled with several zero values ​​around its edges, and the two-dimensional image matrix is ​​a two-dimensional image matrix with equal rows and columns. The damage imaging image matrix M is represented as:

[0059] ;

[0060] Among them, M0, M 45 M 90 and M 135 These represent the damage index distribution matrices under scanning directions of 0°, 45°, 90°, and 135°, respectively.

[0061] Therefore, the present invention employs the above-mentioned rapid imaging method for ultrasonic guided wave damage in aerospace composite materials, which has the following beneficial effects:

[0062] (1) The damage index calculation method adopted does not require the selection of a threshold manually, which reduces the uncertainty when testing different samples and can effectively improve the robustness of damage detection.

[0063] (2) Introducing the energy entropy of the signal as the basis for selecting the optimal signal component when calculating the Hilbert spectrum can effectively reduce the interference of noise signals and improve the signal-to-noise ratio of the image;

[0064] (3) Preprocessing the damage index curve by expanding the verification can effectively increase the damage area index and improve the contrast between the damaged and undamaged areas.

[0065] (4) Using a dual probability function to sharpen the damage edge can reduce the problem of damage edge enlargement caused by factors such as probe width and improve the accuracy of damage imaging;

[0066] (5) The present invention uses orthogonal image matrix fusion, which can further increase the difference between damaged and undamaged areas and improve imaging quality;

[0067] (6) A composite path cross imaging method is proposed, which can identify multiple damages in flat plate samples, avoid the generation of artifacts, and improve detection accuracy and efficiency.

[0068] The method of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0069] Figure 1 is a flowchart of a rapid ultrasonic guided wave damage imaging method for aerospace composite materials according to the present invention;

[0070] Figure 2 is a diagram of the phase velocity curves in all directions in an embodiment of the present invention;

[0071] Figure 3 is a schematic diagram of the biorthogonal scanning of the sample plate in an embodiment of the present invention;

[0072] Figure 4 shows the effect of the expanded core on enhancing the damage index of the damaged area in an embodiment of the present invention.

[0073] Figure 5 shows the optimization effect of the dual probability distribution function on the damage index curve in an embodiment of the present invention;

[0074] Figure 6 shows the imaging effect of a single lesion in an embodiment of the present invention;

[0075] Figure 7 shows the imaging effect of multiple damages in an embodiment of the present invention. Detailed Implementation

[0076] The method of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0077] Unless otherwise defined, the methodological or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0078] The terms "comprising" or "including" as used in this invention mean that the element preceding the term encompasses the element listed after the term, and do not exclude the possibility of encompassing other elements. Terms such as "inner," "outer," "upper," and "lower" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. When the absolute position of the described object changes, the relative positional relationship may also change accordingly. In this invention, unless otherwise explicitly specified and limited, the term "attached" and similar terms should be interpreted broadly. For example, it can refer to a fixed connection, a detachable connection, or an integral part; it can refer to a direct connection or an indirect connection through an intermediate medium; it can refer to the internal communication of two elements or the interaction relationship between two elements. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances. Example

[0079] In this embodiment, two carbon fiber composite material plates with different sizes and damage degrees were selected as test plates. The first test plate was set as a single-damage test plate with a size of 600mm×300mm, a thickness of 2.4mm, and a delamination diameter of 10mm. The second test plate was set as a multi-damage test plate with a size of 400mm×400mm, a thickness of 2.4mm, and delamination diameters of 10mm and 5mm, respectively. During the experiment, the two delamination damages were scanned simultaneously.

[0080] As shown in Figure 1, this embodiment provides a rapid ultrasonic guided wave damage imaging method for aerospace composite materials, including the following steps:

[0081] S1: As shown in Figure 2, the dispersion curve of the carbon fiber composite plate is calculated based on the material parameters of the carbon fiber composite plate. The optimal excitation angle and ultrasonic propagation mode are calculated based on the dispersion curve. In this embodiment, the A0 mode with the largest out-of-plane displacement is taken as the excitation target. A suitable probe frequency is selected and the optimal incident angle of the probe is calculated. The probes are placed symmetrically according to the optimal incident angle. In this embodiment, the selected probe frequency is 200kHz, the optimal incident angle is 14.5°, and the probe spacing is 10cm.

[0082] S2: As shown in Figure 3, determine the scanning area and set up four scanning paths with different ultrasonic propagation directions. Scan and collect ultrasonic guided wave signals under the four scanning paths respectively. When scanning different scanning paths, the scanning direction of the probe is always along the Y-axis. After scanning in a single direction, rotate the carbon fiber composite plate at a fixed angle to realize the acquisition of guided wave signals under different directions.

[0083] In step S2, the ultrasonic propagation directions of the four scanning paths adopt two sets of orthogonal angles, namely 0° and 90°, and 45° and 135°. By scanning with orthogonal angles, the location of the damage can be accurately located, while eliminating misjudgments caused by multiple damages.

[0084] S3: The ultrasonic guided wave signal is decomposed by fully adaptive noise set empirical mode decomposition (EMD) to obtain several ultrasonic guided wave signal components. The energy entropy of the ultrasonic guided wave signal components is calculated. Based on the energy entropy, the ultrasonic guided wave signal components are optimally reconstructed to obtain the reconstructed signal.

[0085] In step S3, the ultrasonic guided wave signal is decomposed by fully adaptive noise ensemble empirical mode decomposition (EMD) to obtain several ultrasonic guided wave signal components. This specifically includes the following steps:

[0086] Step 1: Incorporate Gaussian white noise into the ultrasonic guided wave signal to obtain the signal to be processed, x. i (t), x i (t) is represented as:

[0087] ;

[0088] Where x(t) represents the ultrasonic guided wave signal, μ represents the amplitude of Gaussian white noise, and c i (t) represents the Gaussian white noise of the i-th group;

[0089] Step 2: Decompose the signal x to be processed using EMD i (t), to obtain the first component IMF1 of each group, IMF1 is represented as:

[0090] ;

[0091] Where T represents the EMD decomposition period, the first component IMF1 of each group in the ultrasonic guided wave signal x(t) is removed to obtain the residual component r1(t) of the first stage, which is expressed as:

[0092] ;

[0093] Step 3: Add i sets of Gaussian white noise with amplitude μ to the residual component r1(t) of the first stage to obtain a new signal. , Represented as:

[0094] ;

[0095] For new signals EMD decomposition yields the first component IMF2 and the residual component r2(t) of the new signal in the second stage. IMF2 is expressed as:

[0096] ;

[0097] r2(t) is expressed as:

[0098] ;

[0099] Step 4: Repeat steps 1 to 3 until the residual can no longer be decomposed. At this point, the ultrasonic guided wave signal x(t) is represented as:

[0100] ;

[0101] In step S3, the energy entropy of the ultrasonic guided wave signal component is calculated, and the ultrasonic guided wave signal component is reconstructed based on the energy entropy to obtain the reconstructed signal. This specifically includes the following steps:

[0102] S31: The energy entropy H(x) of the ultrasonic guided wave signal component is expressed as:

[0103] ;

[0104] Where P(x) is the probability density function, representing the proportion of the energy of a single ultrasonic guided wave signal component to the total energy, thereby determining the magnitude of the component energy and its proportion in the overall signal;

[0105] S32: Sort the energy entropy H(x) in descending order, and select the ultrasonic guided wave signal components corresponding to the first five energy entropy H(x) for reconstruction. At this time, the ultrasonic guided wave signal x(t) is expressed as:

[0106] ;

[0107] in For the first five ultrasonic guided wave signal components, the energy entropy of each component is introduced to optimize the signal components, which can reduce the influence of noise in the guided wave signal.

[0108] S4: Obtain the time-frequency information of the reconstructed signal through Hilbert transform, obtain the marginal spectrum through amplitude integration in the time-frequency information, calculate the damage index in the scanning path, and evaluate the degree of damage to the scanning path based on the damage index;

[0109] In step S4, the calculation is performed using the Hilbert transform, specifically including the following steps:

[0110] S41: Perform a Hilbert transform on the preferred IMF component of the ultrasonic guided wave signal to obtain the principal value integral y(t), which is expressed as:

[0111] ;

[0112] Where P represents a slowly increasing distribution;

[0113] S42: Construct the analytic signal Z(t), and obtain the instantaneous frequency spectrum curve of the preferred IMF component of the ultrasonic guided wave signal based on the analytic signal Z(t). Z(t) is expressed as:

[0114] ;

[0115] Where a(t) is the instantaneous amplitude, θ(t) is the instantaneous phase, and ω(t) is the instantaneous frequency. Let be the instantaneous amplitude of the j-th order IMF component at time t and frequency j;

[0116] S43: The instantaneous amplitudes are combined in the time domain to obtain the Hilbert spectrum H(t,ω), which is expressed as:

[0117] ;

[0118] S44: Integrating the Hilbert spectrum over the time axis yields the Hilbert marginal spectrum h(t,ω), which is expressed as:

[0119] ;

[0120] S45: Obtain the damage index DI for each scan path based on the Hilbert spectrum. DI is represented as:

[0121] ;

[0122] Where E r E is the average value of the marginal spectral energy, and E is the value of the marginal spectral energy in a single scan step. When calculating the damage index DI, there is no need to input a manually selected threshold. This calculation method can be applied to different scenarios.

[0123] In step S5, the probability distribution function P(d) is expressed as:

[0124] ;

[0125] Where d is the distance from the scanning path to the center line of the scanning area, σ is a parameter controlling the dispersion of the damage probability, μ is the center position of the normal distribution, D is the diameter of the ultrasonic transducer, and when σ is 2.4 and μ is +2, the probability distribution function is P. L (d) When σ is 2.4 and μ is -2, the probability distribution function is P R(d) The introduction of the dual probability function effectively solves the problem of damage edge expansion caused by probe size during the scanning process.

[0126] Combining fully adaptive noise set empirical mode decomposition (EMD) with Hilbert transform can solve the problem of mode aliasing in the decomposed signal.

[0127] S5: The damage index in the scanning path is convolved using an expansion kernel and a probability distribution function, and the defect area of ​​the damage index curve is enhanced and the defect edge area of ​​the damage index curve is sharpened.

[0128] Step S5 specifically includes the following steps:

[0129] S51: A one-dimensional expansion kernel with a size of 1, a length of 11, and an expansion rate of 4 is used to perform convolution operation on the damage index curve. The introduction of the expansion kernel effectively increases the index of the damaged region in the damage index curve and decreases the index of the undamaged region. The effect of the expansion kernel on enhancing the damage index of the damaged region is shown in Figure 4.

[0130] S52: When entering the damaged area, the high-weighted region is biased towards the leading edge of the scanning area, through the probability distribution function P. L (d) Perform convolution operation on the damage index curve, with the operation direction set from left to right;

[0131] S53: When leaving the damaged area, bias the high-weighted region towards the trailing edge of the scanned area, using the probability distribution function P. R (d) Perform convolution operation on the damage index curve, with the operation direction set from left to right;

[0132] probability distribution function P L (d) and P R (d) The optimization effect on the damage index curve is shown in Figure 5.

[0133] S6: Expand the damage index curves of the four scanning paths into two-dimensional image matrices, process the two-dimensional image matrices to obtain damage imaging image matrices, and fuse the damage imaging image matrices to obtain the final imaging result.

[0134] In step S6, the two-dimensional image is filled with several zero values ​​around its perimeter to ensure that no data is lost when the two-dimensional image is reverse-rotated and restored according to the corresponding rotation angle.

[0135] The two-dimensional image matrix is ​​a two-dimensional image matrix with equal rows and columns;

[0136] The damage imaging matrix M is represented as:

[0137] ;

[0138] Among them, M0, M45 M 90 and M 135 The damage index distribution matrices represent the damage indices at scanning directions of 0°, 45°, 90°, and 135°, respectively. The two-dimensional image matrix is ​​restored by inversely rotating it according to the ultrasonic propagation angle during scanning. The two-dimensional image matrices with mutually perpendicular rotation angles are added pairwise to increase the weight of the damage location. The two added two-dimensional image matrices are then multiplied to effectively expand the numerical value between the damaged and undamaged areas, eliminate artifacts, and obtain the final damage imaging image matrix, realizing damage localization imaging in carbon fiber composite plates and micro-curved surface structures.

[0139] As shown in Figures 6 and 7, the ultrasonic guided wave damage rapid imaging method for aerospace composite materials provided in this embodiment can accurately locate both single-damage and multi-damage regions.

[0140] Therefore, the present invention employs the above-mentioned rapid ultrasonic guided wave damage imaging method for aerospace composite materials, which reduces the uncertainty in the detection of different samples, effectively improves the robustness of damage detection, image signal-to-noise ratio and contrast between damaged and undamaged areas, reduces the interference of noise signals and the problem of damage edge expansion caused by factors such as probe width, identifies multiple damages in flat plate samples, avoids the generation of artifacts, and further improves the detection accuracy and efficiency of damage imaging.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the method of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the method of the present invention, and these modifications or equivalent substitutions should not cause the modified method to deviate from the spirit and scope of the method of the present invention.

Claims

1. An ultrasonic guided wave damage rapid imaging method for aerospace composites, characterized in that: Includes the following steps: S1: Calculate the dispersion curve of the carbon fiber composite plate based on the material parameters of the carbon fiber composite plate, and calculate the optimal excitation angle and ultrasonic propagation mode based on the dispersion curve; S2: Determine the scanning area, set up four scanning paths with different ultrasonic propagation directions, and scan and collect ultrasonic guided wave signals under the four scanning paths respectively; S3: The ultrasonic guided wave signal is decomposed by fully adaptive noise set empirical mode decomposition (EMD) to obtain several ultrasonic guided wave signal components. The energy entropy of the ultrasonic guided wave signal components is calculated. Based on the energy entropy, the ultrasonic guided wave signal components are optimally reconstructed to obtain the reconstructed signal. S4: Obtain the time-frequency information of the reconstructed signal through Hilbert transform, obtain the marginal spectrum through amplitude integration in the time-frequency information, calculate the damage index in the scanning path, and evaluate the degree of damage to the scanning path based on the damage index; S5: The damage index in the scanning path is convolved using an expansion kernel and a probability distribution function, and the defect area of ​​the damage index curve is enhanced and the defect edge area of ​​the damage index curve is sharpened. S6: Expand the damage index curves of the four scanning paths into two-dimensional image matrices, process the two-dimensional image matrices to obtain damage imaging image matrices, and fuse the damage imaging image matrices to obtain the final imaging result.

2. The method of claim 1, wherein: In step S2, the ultrasonic propagation directions of the four scanning paths adopt two sets of orthogonal angles, namely 0° and 90°, and 45° and 135°.

3. The method of claim 1, wherein: In step S3, the ultrasonic guided wave signal is decomposed by fully adaptive noise ensemble empirical mode decomposition (EMD) to obtain several ultrasonic guided wave signal components. This specifically includes the following steps: Step one: combine the Gaussian white noise into the ultrasonic guided wave signal to obtain the signal to be processed x i (t), x i (t) is expressed as: ； in x(t) represents the ultrasonic guided wave signal, μ represents the amplitude of the Gaussian white noise, c i (t) represents the Gaussian white noise of the i-th group; Step two: decompose the signal x to be processed by EMD i (t), the first component of each group IMF1 is obtained, and IMF1 is expressed as: ； Where T represents the EMD decomposition period, the first component IMF1 of each group in the ultrasonic guided wave signal x(t) is removed to obtain the residual component r1(t) of the first stage, which is expressed as: ； Step three: Add i set of Gaussian white noise with amplitude μ to the residual component r1(t) of the first stage, to obtain a new signal ， Represented as: ； New signal EMD decomposition yields the first component IMF2 and the residual component r2(t) of the new signal in the second stage. IMF2 is expressed as: ； r2(t) is expressed as: ； Step 4: Repeat steps 1 to 3 until the residual can no longer be decomposed. At this point, the ultrasonic guided wave signal x(t) is represented as: ； 4. The method of claim 1, wherein: In step S3, the energy entropy of the ultrasonic guided wave signal component is calculated, and the ultrasonic guided wave signal component is reconstructed based on the energy entropy to obtain the reconstructed signal. This specifically includes the following steps: S31: The energy entropy H(x) of the ultrasonic guided wave signal component is expressed as: ； 4. Where P(x) is the probability density function, representing the proportion of the energy of a single ultrasonic guided wave signal component to the total energy; S32: Sort the energy entropy H(x) in descending order, and select the ultrasonic guided wave signal components corresponding to the first five energy entropy H(x) for reconstruction. At this time, the ultrasonic guided wave signal x(t) is expressed as: ； wherein These are the first five ultrasonic guided wave signal components selected.

5. The method of claim 1, wherein: In step S4, the calculation is performed using the Hilbert transform, specifically including the following steps: S41: Perform a Hilbert transform on the preferred IMF component of the ultrasonic guided wave signal to obtain the principal value integral y(t), which is expressed as: ； Where P represents a slowly increasing distribution; S42: Construct the analytic signal Z(t), and obtain the instantaneous frequency spectrum curve of the preferred IMF component of the ultrasonic guided wave signal based on the analytic signal Z(t). Z(t) is expressed as: ； where a(t) is the instantaneous amplitude, θ(t) is the instantaneous phase, and ω(t) is the instantaneous frequency, Let be the instantaneous amplitude of the j-th order IMF component at time t and frequency j; S43: The instantaneous amplitudes are combined in the time domain to obtain the Hilbert spectrum H(t,ω), which is expressed as: ； S44: Integrating the Hilbert spectrum over the time axis yields the Hilbert marginal spectrum h(t,ω), which is expressed as: ； S45: Obtain the damage index DI for each scan path based on the Hilbert spectrum. DI is represented as: ； where E r is the average of the marginal spectrum energy, E is the value of the marginal spectrum energy in a single scan step.

6. The method of claim 1, wherein: In step S5, the probability distribution function P(d) is expressed as: ； where d is the distance of the scan path to the centerline of the scan region, σ is a parameter that controls the spread of the damage probability distribution, μ is the center position of the normal distribution, D is the diameter of the ultrasound transducer, and the probability distribution function is P L (d) when σ is 2.4 and μ is -2. R (d).

7. The method of claim 6, wherein: Step S5 specifically includes the following steps: S51: Perform convolution operation on the damage exponential curve using a one-dimensional expansion kernel with size 1, length 11, and expansion rate of 4; S52: When entering the damage area, the high weight area is deviated to the scanning area front, and the probability distribution function P is used L (d) Convolution operation is performed on the damage index curve, and the operation direction is set from left to right; S53: When leaving the damage area, the high weight area is biased to the trailing edge of the scan area by the probability distribution function P R (d) Convolution operation is performed on the damage index curve, and the operation direction is set from left to right.

8. The method of claim 1, wherein: In step S6, the two-dimensional image is filled with several zero values ​​around its edges, and the two-dimensional image matrix is ​​a two-dimensional image matrix with equal rows and columns. The damage imaging image matrix M is represented as: ； in M0, M 45 , M 90 and M 135 represent the damage index distribution matrix at 0°, 45°, 90° and 135° scanning direction, respectively.

Images