Ultrasonic phased array testing method for strain clamp based on two-dimensional deconvolution envelope reconstruction

Abstract

The application discloses a tension clamp phased array ultrasonic detection method based on two-dimensional deconvolution envelope reconstruction and relates to the technical field of nondestructive testing, which comprises the following steps: placing a phased array ultrasonic probe outside a tension clamp, exciting the phased array probe by adopting a fan-shaped scanning mode and storing an A-scan signal set of each phased array element; calculating the refraction path and propagation sound path of the sound beam at the wedge interface according to Snell's law, and completing ray tracing mapping from the A-scan signal time domain to two-dimensional spatial coordinates; then introducing a probe point spread function, and performing non-negative inverse iterative deconvolution on a single-polarity ultrasonic envelope signal. The method utilizes non-blind deconvolution to process the phased array ultrasonic echo of the tension clamp, can simultaneously improve the lateral and longitudinal resolution of the phased array ultrasonic imaging, suppresses background noise, realizes effective identification of the inter-strand gap in the tension clamp, has a larger applicable range and higher algorithm robustness, and has higher engineering application value.

Description

Technical Field

[0001] This invention relates to the field of nondestructive testing technology, specifically to a phased array ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction. Background Technology

[0002] Tension clamps are key hardware in high-voltage overhead transmission lines and are widely used in power systems such as high-voltage and ultra-high-voltage overhead transmission lines. Their main function is to transfer the tension borne by the conductor during long-term service to the tower through the anchoring structure, thereby ensuring the stability of the conductor in the transmission line. Because tension clamps are subjected to complex service environments such as high mechanical tension, micro-wind vibration, temperature fluctuations, and environmental corrosion, defects such as incomplete crimping, broken strands, or slippage can easily occur in their internal sleeve parts, which in turn affect the safe and stable operation of the transmission line. Currently, X-ray inspection is commonly used to detect internal defects in tension clamps. While this method can image the internal structure of the clamp, it has drawbacks such as radiation safety hazards, high inspection costs, and limitations imposed by protective conditions in field applications. Phased array ultrasonic testing, as a non-destructive testing technology, has advantages such as high sensitivity, flexible operation, and no need for complex radiation protection. It has been used to detect internal crimping defects in tension clamps. By using phased array fan-shaped scanning, ultrasonic echo images of the inside of the tension clamp can be obtained and used to identify typical crimping defects such as undervoltage and broken strands. However, tension clamps typically have an axisymmetric structure with multiple nested layers of aluminum shell, aluminum core, and steel core. There are small gaps between the internal strands. Accurate evaluation of the size and spacing of these tiny gaps is crucial for determining the crimping condition and the degree of defects. Due to the limitations of the lateral and longitudinal resolution of conventional phased array ultrasonic testing, the ultrasonic echo envelope signals generated by adjacent tiny gaps tend to overlap and are difficult to distinguish effectively, thus limiting the accuracy of identifying crimping defects inside tension clamps. Therefore, this invention proposes a phased array ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction. Summary of the Invention

[0003] To address the aforementioned technical issues, a phased array ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction is provided. This technical solution solves the problem that conventional phased array ultrasonic testing has limited lateral and longitudinal resolution, making it difficult to distinguish the echoes of adjacent tiny gaps inside the tension clamp, thus affecting the accurate identification of crimping defects.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A phased array ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction includes: S1. Determine the phased array ultrasonic testing parameters based on the material, geometric dimensions, and testing range of the tension clamp under test, and place the phased array ultrasonic probe in the testing area outside the tension clamp. S2. Based on the phased array ultrasonic detection parameters, a phased array sector scan is performed on the tension wire strand crimping area. Deflection excitation and echo reception are performed along a set angle range to obtain the original A-scan signals at multiple sector scan angles and form the original phased array ultrasonic A-scan signal matrix. S3. Perform Hilbert transform on each column of the A-scan signal in the original phased array ultrasound A-scan signal matrix and take the absolute value to obtain a fully positive envelope signal matrix. S4. Calculate the refraction path of the ultrasonic wave at the interface between the wedge and the tension clamp under test according to Snell's law, and combine the sound velocity of the wedge, the sound velocity of the tension clamp under test, the height of the wedge and the sampling time to map the time sampling points in the envelope signal matrix to the spatial coordinates in the detection section. S5. Construct the time-domain point spread function corresponding to the axial impulse response of the phased array ultrasonic probe. Perform non-blind deconvolution envelope reconstruction on the envelope signal matrix along the time axis under non-negative constraints to obtain the polar coordinate envelope matrix after axial iterative reconstruction. S6. Extract the arc sequence that is at the same sampling point in the polar coordinate envelope matrix and spans all fan-scan angles, construct the lateral point spread function corresponding to the response in the fan-scan angle direction, and perform non-blind deconvolution envelope reconstruction along the fan-scan angle direction to obtain the final two-dimensional envelope matrix. S7. Based on the spatial coordinates, normalize the two-dimensional final envelope matrix and reconstruct the image to generate a phased array ultrasonic sector scan image after two-dimensional deconvolution envelope reconstruction, so as to identify the gap between the strands inside the tension clamp.

[0005] Preferably, in step S1, determining the phased array ultrasonic testing parameters includes: Based on the material, geometric dimensions, and testing range of the tension clamp under test, determine the center frequency, number of array elements, system sampling frequency, and sector scanning angle range of the phased array ultrasonic probe; The phased array ultrasonic probe is controlled according to the fan-shaped scanning angle range to deflect and excite the tension wire strand crimping area and receive the echo.

[0006] Preferably, in step S2, obtaining the original phased array ultrasound A-scan signal matrix includes: Select N sector sweep angles within the set angle range; At each sector scan angle, the phased array ultrasonic probe is excited and the echo signal is received to obtain a series of raw A-scan signals. Each column of raw A-scan signals contains M time sampling points. The raw A-scan signals corresponding to the M time sampling points and N sector scan angles constitute the raw phased array ultrasound A-scan signal matrix. .

[0007] Preferably, in step S3, the Hilbert envelope extraction includes: The original phased array ultrasound A-scan signal matrix Perform a Hilbert transform on each column of A-scan signals p(t); The envelope signal e(t) at the corresponding sampling time is obtained by taking the absolute value of the complex analytic signal after Hilbert transform; The envelope signal matrix is ​​composed of the envelope signals at each sector scan angle. .

[0008] Preferably, in step S4, the spatial coordinate mapping includes: Let the angle of refraction of the current sector scan beam within the tension clamp of the tested wire be . The angle of incidence of the ultrasonic wave within the wedge was calculated based on Snell's law. ; Based on wedge height and incident angle Calculate the two-way propagation time of sound waves penetrating the wedge. and the lateral offset of the probe's exit point ; For greater than At sampling time t, calculate the one-way propagation distance inside the specimen. The formula is as follows: according to and angle of refraction Calculate the spatial depth coordinate Z and lateral coordinate X in the test section, where c is the sound velocity of the tension clamp under test.

[0009] Preferably, in step S5, constructing the time-domain point spread function includes: The envelope of ultrasonic unipolar pulses received by a phased array ultrasonic probe is simulated using a normalized discrete feature sequence. The time-domain point spread function is determined by the discrete sampling point index, the length of the discrete sequence of the point spread function, the system time sampling step size, and the damping attenuation coefficient. The damping attenuation coefficient is related to the actual bandwidth of the phased array ultrasonic probe.

[0010] Preferably, in step S5, performing non-blind deconvolution envelope reconstruction along the time axis includes: Under non-negativity constraints, a Poisson probability distribution model corresponding to a single sampling point of a phased array ultrasonic probe is established. The log-likelihood gradient of the Poisson probability distribution model is introduced into the error backpropagation process; Using the time-domain point spread function as a known point spread function, perform a one-dimensional non-blind iterative deconvolution along the time axis on each column of the envelope signal matrix; After completing the set number of axial iterations, the polar coordinate envelope matrix after axial iteration reconstruction is obtained.

[0011] Preferably, in step S6, performing non-blind deconvolution envelope reconstruction along the sector scan angle direction includes: In the polar coordinate envelope matrix after axial iterative reconstruction, extract the arc sequence that is at the same sampling point in time and spans all sector angles; The sector sweep angle index is used as the lateral index, and a lateral point spread function is constructed. Based on the lateral point spread function, perform a lateral one-dimensional non-blind iterative deconvolution on the arc-shaped sequence; After completing the set number of horizontal iterations, the reconstructed sequence from the horizontal iterations is refilled into the matrix to obtain the final two-dimensional envelope matrix.

[0012] Preferably, in step S7, generating the phased array ultrasonic sector scan image after two-dimensional deconvolution envelope reconstruction includes: The two-dimensional final envelope matrix is ​​normalized. Based on the spatial depth coordinates Z and lateral coordinates X obtained in step S4, the envelope amplitude in the final two-dimensional envelope matrix is ​​mapped to the spatial position of the detection section using the difference function. The reconstructed phased array ultrasonic sector scan image is obtained after two-dimensional deconvolution envelope reconstruction.

[0013] Preferably, after step S7, the method further includes: Echo features corresponding to the gaps between strands inside the tension clamp are extracted from the phased array ultrasonic sector scan image reconstructed by two-dimensional deconvolution envelope. The gap echo features reconstructed from the two-dimensional deconvolution envelope are compared with the gap echo features in the original phased array ultrasonic sector scan image. The identification effect of the internal strand gap of the tension clamp is evaluated based on the full width at half maximum (FWHM), signal-to-noise ratio (SNR), or degree of gap echo separation.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention proposes to extract the Hilbert envelope from the phased array sector scan signal and perform non-blind deconvolution envelope reconstruction along the time axis. This effectively compresses the ultrasonic pulse width, solves the problem of difficulty in distinguishing the echo envelope signals of adjacent small gaps in conventional phased array ultrasonic testing, and improves the axial resolution of the gaps between strands inside the tension clamp. By extracting the arc sequence spanning all sector scan angles at the same time sampling point from the polar coordinate envelope matrix and performing lateral non-blind deconvolution envelope reconstruction along the sector scan angle direction, the lateral resolution of the phased array ultrasonic sector scan image can be improved, and the increased computational load and spatial distortion caused by traditional Cartesian grid interpolation transformation can be avoided. Through two-dimensional deconvolution envelope reconstruction along the time axis and sector scan angle direction, the background noise level can be reduced, making the echo characteristics of the gaps inside the tension clamp clearer, thereby improving the accuracy of crimping defect identification and engineering applicability. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the system structure used in this invention; Figure 3 This is a schematic diagram of the tension clamp of the present invention; Figure 4 These are the original phased array ultrasonic detection sector scan images of this invention; Figure 5 This is a phased array ultrasonic detection sector scan image after two-dimensional deconvolution envelope reconstruction according to the present invention; Figure 6 This is the original envelope signal at the gap of the tension clamp in this invention; Figure 7 It is the envelope signal after two-dimensional deconvolution envelope reconstruction at the gap of the tension clamp of the present invention. Detailed Implementation

[0016] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0017] Reference Figure 1 As shown, the phased array ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction includes: S1. Determine the phased array ultrasonic testing parameters based on the material, geometric dimensions, and testing range of the tension clamp under test, and place the phased array ultrasonic probe in the testing area outside the tension clamp. S2. Based on the phased array ultrasonic detection parameters, a phased array sector scan is performed on the tension wire strand crimping area. Deflection excitation and echo reception are performed along a set angle range to obtain the original A-scan signals at multiple sector scan angles and form the original phased array ultrasonic A-scan signal matrix. S3. Perform Hilbert transform on each column of the A-scan signal in the original phased array ultrasound A-scan signal matrix and take the absolute value to obtain a fully positive envelope signal matrix. S4. Calculate the refraction path of the ultrasonic wave at the interface between the wedge and the tension clamp under test according to Snell's law, and combine the sound velocity of the wedge, the sound velocity of the tension clamp under test, the height of the wedge and the sampling time to map the time sampling points in the envelope signal matrix to the spatial coordinates in the detection section. S5. Construct the time-domain point spread function corresponding to the axial impulse response of the phased array ultrasonic probe. Perform non-blind deconvolution envelope reconstruction on the envelope signal matrix along the time axis under non-negative constraints to obtain the polar coordinate envelope matrix after axial iterative reconstruction. S6. Extract the arc sequence that is at the same sampling point in the polar coordinate envelope matrix and spans all fan-scan angles, construct the lateral point spread function corresponding to the response in the fan-scan angle direction, and perform non-blind deconvolution envelope reconstruction along the fan-scan angle direction to obtain the final two-dimensional envelope matrix. S7. Based on the spatial coordinates, normalize the two-dimensional final envelope matrix and reconstruct the image to generate a phased array ultrasonic sector scan image after two-dimensional deconvolution envelope reconstruction, so as to identify the gap between the strands inside the tension clamp.

[0018] like Figure 2 As shown, the ultrasonic testing system used in this embodiment of the invention includes a phased array ultrasonic testing instrument, a phased array ultrasonic probe, and a wedge. The phased array ultrasonic probe is positioned on the outside of the tension clamp in the area to be inspected via the wedge, and is used to perform a phased array fan-shaped scan of the tension clamp's strand crimping area. Figure 3 The tension clamp structure and the location of its internal gaps in an embodiment of the present invention are shown.

[0019] Furthermore, in step S1, determining the phased array ultrasonic testing parameters includes: Based on the material, geometric dimensions, and testing range of the tension clamp under test, determine the center frequency, number of array elements, system sampling frequency, and sector scanning angle range of the phased array ultrasonic probe; The phased array ultrasonic probe is controlled according to the fan-shaped scanning angle range to deflect and excite the tension wire strand crimping area and receive the echo.

[0020] In one specific embodiment, a tension clamp used in actual power transmission lines was selected as the workpiece to be inspected. Its outer shell thickness was approximately 10 mm. The sound velocity of the test piece material was measured. There are minute gaps at the internal pressing interface. The center frequency is selected for detection. A phased array probe and wedge with 32 array elements. Set the system sampling frequency of the phased array flaw detector. The fan-shaped scanning angle range is set to -40° to 40°.

[0021] In step S2, the acquisition of the original phased array ultrasound A-scan signal matrix includes: Select N sector sweep angles within the set angle range; At each sector scan angle, the phased array ultrasonic probe is excited and the echo signal is received to obtain a series of raw A-scan signals; Each column of raw A-scan signals contains M time sampling points, and the raw A-scan signals corresponding to the M time sampling points and N sector scan angles constitute the raw phased array ultrasound A-scan signal matrix. .

[0022] In step S3, Hilbert envelope extraction includes: Let the sampling time be The original phased array ultrasound A-scan signal matrix Perform a Hilbert transform on each column of A-scan signals p(t); Taking the absolute value of the complex analytic signal after Hilbert transform yields the envelope signal e(t) at the corresponding sampling time. The envelope signal matrix is ​​composed of the envelope signals at each sector scan angle. .

[0023] The envelope signal e(t) is expressed as: Where e(t) represents the amplitude of the envelope signal at sampling time t, p(t) represents the amplitude of the original A-scan radio frequency signal at sampling time t, and i is the imaginary unit. Pi For integration variables, Representing the integral variable The corresponding original A-scan radio frequency signal amplitude, This represents the time difference between the current sampling time and the integral variable. This represents the modulus value of the complex analytic signal.

[0024] In step S4, the spatial coordinate mapping includes: Let the angle of refraction of the current sector scan beam within the tension clamp of the tested wire be . The angle of incidence of the ultrasonic wave within the wedge was calculated based on Snell's law. The formula is as follows: in, denoted as , where c is the sound velocity of the wedge block and c is the sound velocity of the specimen.

[0025] Based on wedge height and incident angle Calculate the two-way propagation time of sound waves penetrating the wedge. and the lateral offset of the probe's exit point ; For greater than At sampling time t, calculate the one-way propagation distance inside the specimen. for: according to and angle of refraction Calculate the spatial depth coordinate Z and lateral coordinate X in the test section, where c is the sound velocity of the tension clamp under test.

[0026] The true spatial depth coordinates Z and lateral coordinates X mapped by this sampling point in the detection cross section are: Using envelope signal matrix By combining the spatial coordinates of each time sampling point, the unprocessed original phased array ultrasound sector scan image can be reconstructed, such as... Figure 4 As shown.

[0027] In step S5, constructing the time-domain point spread function includes: The envelope of ultrasonic unipolar pulses received by a phased array ultrasonic probe is simulated using a normalized discrete feature sequence. The time-domain point spread function is determined by the discrete sampling point index, the length of the discrete sequence of the point spread function, the system time sampling step size, and the damping attenuation coefficient. The damping attenuation coefficient is related to the actual bandwidth of the phased array ultrasonic probe.

[0028] The time-domain point spread function h(n) is expressed as: Where n and m are the index numbers of the discrete sampling points, h(n) is the amplitude of the point spread characteristic function at the nth sampling point, and L is the length of the discrete sequence of the point spread function. The system time sampling step size, The damping attenuation coefficient is related to the actual bandwidth of the ultrasonic phased array probe. This represents summing over all L discrete sampling points.

[0029] In step S5, performing non-blind deconvolution envelope reconstruction along the time axis includes: Under non-negativity constraints, a Poisson probability distribution model corresponding to a single sampling point of a phased array ultrasonic probe is established. The probability distribution model for a single sampled signal is established as follows: in, This indicates that the conditional expected parameter is at the k-th iteration. The probability of the envelope signal e(t) occurring at sampling point t is given by the following condition: e(t) represents the amplitude of the envelope signal at sampling point t. This represents the conditional expectation parameter of the Poisson probability distribution corresponding to sampling point t at the k-th iteration; k represents the iteration number. Represents a base of the natural constant and an exponent of 1. The exponential term; e(t)! represents the factorial term corresponding to the envelope signal e(t).

[0030] The log-likelihood gradient of the Poisson probability distribution model is introduced into the error backpropagation process; Using the time-domain point spread function h(n) as a known point spread function, the envelope signal matrix is... Each column of envelope signal in the process undergoes a time-axis one-dimensional non-blind iterative deconvolution. Let the estimated signal of the k-th iteration be... The discrete iterative mathematical model is as follows: in, This represents the estimated signal obtained in the k-th iteration; denoted as the estimated signal obtained in the (k+1)th iteration; M represents the total number of time sampling points in each envelope signal; n represents the index number of the discrete sampling point in the time axis direction; h(tn) represents the time domain point spread function response between sampling time t and sampling point index n; Represent the natural logarithm function; The log-likelihood function of the Poisson probability distribution model is expressed relative to the conditional expectation parameter. The partial derivatives; +1 is the iterative correction term. After completing the set number of axial iterations, the polar coordinate envelope matrix after axial iterative reconstruction is obtained. ,in, This represents the polar coordinate envelope matrix after reconstruction by non-blind deconvolution envelope along the time axis.

[0031] After completing the set number of axial iterations, the polar coordinate envelope matrix after axial iteration reconstruction is obtained.

[0032] In one specific embodiment, for the axial pulse of a phased array ultrasonic probe, a normalized point spread characteristic function with a length of 300 sampling points is constructed. Combined with a Poisson probability log-likelihood gradient model, the envelope matrix is... Perform 15 one-dimensional non-negative iterative deconvolutions on each column to obtain the polar coordinate envelope matrix after axial iterative reconstruction. .

[0033] In step S6, performing non-blind deconvolution envelope reconstruction along the sector scan angle direction includes: polar coordinate envelope matrix after axial iterative reconstruction In this process, an arc sequence is extracted that is at the same sampling point t and spans all sector scanning angles; the arc sequence represents the set of envelope signals in different sector scanning angle directions at the same sampling point.

[0034] The sector sweep angle index is used as the lateral index, and a lateral point spread function is constructed. Let the estimated signal of the k-th horizontal iteration be... ,in, Let be the lateral angle index of the signal to be estimated, and the conditional expectation parameter of the corresponding Poisson probability distribution is: The one-dimensional transverse non-blind deconvolution envelope reconstruction, similar to the time-axis non-blind deconvolution, is performed again on the arc signal y. Its discrete iterative model is as follows: in, This represents the estimated signal obtained in the k-th horizontal iteration; This represents the estimated signal obtained in the (k+1)th lateral iteration; N represents the total number of sector sweep angles; and i represents the lateral angle index number in the sector sweep angle direction. This indicates the lateral angle index of the signal to be estimated; This represents the angle index i in the direction of the sector sweep angle and the lateral angle index to be estimated. The transverse point spread function response between; y(i) represents the transverse envelope signal amplitude corresponding to the i-th sector sweep angle in the arc sequence; This represents the conditional expectation parameter of the Poisson probability distribution corresponding to the horizontal angle index i at the kth horizontal iteration. The conditional expected parameter is The probability of obtaining the transverse envelope signal y(i) at that time; The log-likelihood function of the lateral probability distribution model is expressed relative to the conditional expectation parameter. The partial derivatives of .

[0035] After completing the set number of horizontal iterations, the reconstructed sequence from the horizontal iterations is re-filled into the matrix to obtain the final two-dimensional envelope matrix. .in, This represents the final envelope matrix after reconstruction by two-dimensional deconvolution of the envelope along the time axis and the fan sweep angle.

[0036] In one specific embodiment, in the polar coordinate envelope matrix processed in step S5, an arc sequence that is at the same sampling point t and spans all scanning angles is extracted, a point diffusion feature function with a length of 20 sampling points is constructed, and 10 horizontal one-dimensional iterative deconvolutions are performed on it to obtain a two-dimensional reconstructed envelope matrix.

[0037] In step S7, generating the phased array ultrasonic sector scan image after two-dimensional deconvolution envelope reconstruction includes: For the two-dimensional final envelope matrix Perform normalization processing; Based on the spatial depth coordinates Z and lateral coordinates X obtained in step S4, the envelope amplitude in the final two-dimensional envelope matrix is ​​mapped to the spatial position of the detection section using the difference function. The reconstructed phased array ultrasonic sector scan image is obtained after two-dimensional deconvolution envelope reconstruction.

[0038] in, The two-dimensional final envelope matrix is ​​represented by X; X represents the lateral coordinate in the detection section; Z represents the spatial depth coordinate in the detection section; (X,Z) represents the real spatial coordinates of each sampling point in the envelope signal matrix after mapping; normalization is used to adjust the envelope amplitude in the two-dimensional final envelope matrix to a uniform amplitude range; the difference function is used to map the envelope amplitude in the two-dimensional final envelope matrix to the spatial position of the detection section according to the spatial coordinates (X,Z), thereby reconstructing the phased array ultrasonic sector scan image after two-dimensional deconvolution envelope reconstruction.

[0039] Figure 4 and Figure 5 The pre-processed phased array ultrasonic sector scan image and the reconstructed image are presented respectively. After two-dimensional deconvolution envelope reconstruction, the gap echo resolution in the sector scan image is improved, the signal-to-noise ratio is increased, and the gap characteristics of the internal strands of the tension clamp are clearer.

[0040] The process following step S7 also includes: Echo features corresponding to the gaps between strands inside the tension clamp are extracted from the phased array ultrasonic sector scan image reconstructed by two-dimensional deconvolution envelope. The gap echo features reconstructed from the two-dimensional deconvolution envelope are compared with the gap echo features in the original phased array ultrasonic sector scan image; the comparison indicators include the full width at half maximum (FWHM) of the gap echo, the signal-to-noise ratio (SNR), or the degree of gap echo separation.

[0041] The identification effect of the internal strand gap of the tension clamp is evaluated based on the full width at half maximum (FWHM), signal-to-noise ratio (SNR), or degree of gap echo separation.

[0042] Figure 6 and Figure 7 The original gap echo signal and the reconstructed signal are presented separately, and the half-width at half maximum (FWHM) of the gap echo are used to evaluate the imaging resolution. The results show that the FWHM of the gap echo in the original signal is 0.98 μs; after reconstruction using the 2D deconvolution envelope, the FWHM of the gap echo is reduced to 0.37 μs, only 37.7% of the original signal's FWHM. This demonstrates that after reconstruction using the 2D deconvolution envelope, the gap signal is sharper and has higher resolution in the A-scan envelope signal.

[0043] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A phased array ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction, characterized in that, include: S1. Determine the phased array ultrasonic testing parameters based on the material, geometric dimensions, and testing range of the tension clamp under test, and place the phased array ultrasonic probe in the testing area outside the tension clamp. S2. Based on the phased array ultrasonic detection parameters, a phased array sector scan is performed on the tension wire strand crimping area. Deflection excitation and echo reception are performed along a set angle range to obtain the original A-scan signals at multiple sector scan angles and form the original phased array ultrasonic A-scan signal matrix. S3. Perform Hilbert transform on each column of the A-scan signal in the original phased array ultrasound A-scan signal matrix and take the absolute value to obtain a fully positive envelope signal matrix. S4. Calculate the refraction path of the ultrasonic wave at the interface between the wedge and the tension clamp under test according to Snell's law, and combine the sound velocity of the wedge, the sound velocity of the tension clamp under test, the height of the wedge and the sampling time to map the time sampling points in the envelope signal matrix to the spatial coordinates in the detection section. S5. Construct the time-domain point spread function corresponding to the axial impulse response of the phased array ultrasonic probe. Perform non-blind deconvolution envelope reconstruction on the envelope signal matrix along the time axis under non-negative constraints to obtain the polar coordinate envelope matrix after axial iterative reconstruction. S6. Extract the arc sequence that is at the same sampling point in the polar coordinate envelope matrix and spans all fan-scan angles, construct the lateral point spread function corresponding to the response in the fan-scan angle direction, and perform non-blind deconvolution envelope reconstruction along the fan-scan angle direction to obtain the final two-dimensional envelope matrix. S7. Based on the spatial coordinates, normalize the two-dimensional final envelope matrix and reconstruct the image to generate a phased array ultrasonic sector scan image after two-dimensional deconvolution envelope reconstruction, so as to identify the gap between the strands inside the tension clamp.

2. The ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction according to claim 1, characterized in that, In step S1, determining the phased array ultrasonic testing parameters includes: Based on the material, geometric dimensions, and testing range of the tension clamp under test, determine the center frequency, number of array elements, system sampling frequency, and sector scanning angle range of the phased array ultrasonic probe; The phased array ultrasonic probe is controlled according to the fan-shaped scanning angle range to deflect and excite the tension wire strand crimping area and receive the echo.

3. The ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction according to claim 1, characterized in that, In step S2, the acquisition of the original phased array ultrasound A-scan signal matrix includes: Select N sector sweep angles within the set angle range; At each sector scan angle, the phased array ultrasonic probe is excited and the echo signal is received to obtain a series of raw A-scan signals. Each column of raw A-scan signals contains M time sampling points. The raw A-scan signals corresponding to the M time sampling points and N sector scan angles constitute the raw phased array ultrasound A-scan signal matrix. .

4. The ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction according to claim 1, characterized in that, In step S3, Hilbert envelope extraction includes: The original phased array ultrasound A-scan signal matrix Perform a Hilbert transform on each column of A-scan signals p(t); The envelope signal e(t) at the corresponding sampling time is obtained by taking the absolute value of the complex analytic signal after Hilbert transform; The envelope signal matrix is ​​composed of the envelope signals at each sector scan angle. .

5. The ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction according to claim 1, characterized in that, In step S4, the spatial coordinate mapping includes: Let the angle of refraction of the current sector scan beam within the tension clamp of the tested wire be . The angle of incidence of the ultrasonic wave within the wedge was calculated based on Snell's law. ; Based on wedge height and incident angle Calculate the two-way propagation time of sound waves penetrating the wedge. and the lateral offset of the probe's exit point ; For greater than At sampling time t, calculate the one-way propagation distance inside the specimen. The formula is as follows: according to and angle of refraction Calculate the spatial depth coordinate Z and lateral coordinate X in the test section, where c is the sound velocity of the tension clamp under test.

6. The ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction according to claim 1, characterized in that, In step S5, constructing the time-domain point spread function includes: The envelope of ultrasonic unipolar pulses received by a phased array ultrasonic probe is simulated using a normalized discrete feature sequence. The time-domain point spread function is determined by the discrete sampling point index, the length of the discrete sequence of the point spread function, the system time sampling step size, and the damping attenuation coefficient. The damping attenuation coefficient is related to the actual bandwidth of the phased array ultrasonic probe.

7. The ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction according to claim 6, characterized in that, In step S5, performing non-blind deconvolution envelope reconstruction along the time axis includes: Under non-negativity constraints, a Poisson probability distribution model corresponding to a single sampling point of a phased array ultrasonic probe is established. The log-likelihood gradient of the Poisson probability distribution model is introduced into the error backpropagation process; Using the time-domain point spread function as a known point spread function, perform a one-dimensional non-blind iterative deconvolution along the time axis on each column of the envelope signal matrix; After completing the set number of axial iterations, the polar coordinate envelope matrix after axial iteration reconstruction is obtained.

8. The ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction according to claim 1, characterized in that, In step S6, performing non-blind deconvolution envelope reconstruction along the sector scan angle direction includes: In the polar coordinate envelope matrix after axial iterative reconstruction, extract the arc sequence that is at the same sampling point in time and spans all sector angles; The sector sweep angle index is used as the lateral index, and a lateral point spread function is constructed. Based on the lateral point spread function, perform a lateral one-dimensional non-blind iterative deconvolution on the arc-shaped sequence; After completing the set number of horizontal iterations, the reconstructed sequence from the horizontal iterations is refilled into the matrix to obtain the final two-dimensional envelope matrix.

9. The ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction according to claim 1, characterized in that, In step S7, generating the phased array ultrasonic sector scan image after two-dimensional deconvolution envelope reconstruction includes: The two-dimensional final envelope matrix is ​​normalized. Based on the spatial depth coordinates Z and lateral coordinates X obtained in step S4, the envelope amplitude in the final two-dimensional envelope matrix is ​​mapped to the spatial position of the detection section using the difference function. The reconstructed phased array ultrasonic sector scan image is obtained after two-dimensional deconvolution envelope reconstruction.

10. The ultrasonic testing method for tension clamps based on two-dimensional deconvolution envelope reconstruction according to claim 1, characterized in that, Following step S7, the following is also included: Echo features corresponding to the gaps between strands inside the tension clamp are extracted from the phased array ultrasonic sector scan image reconstructed by two-dimensional deconvolution envelope. The gap echo features reconstructed from the two-dimensional deconvolution envelope are compared with the gap echo features in the original phased array ultrasonic sector scan image. The identification effect of the internal strand gap of the tension clamp is evaluated based on the full width at half maximum (FWHM), signal-to-noise ratio (SNR), or degree of gap echo separation.

Images