A Full-Focusing Imaging Method for Pipe Axial Defects Based on Multi-Frame, Multi-Mode Wavelet Fusion

Abstract

This invention belongs to the field of nondestructive testing technology and proposes a full-focus imaging method for axial defects in pipelines based on multi-frame, multi-mode wave fusion. The method employs a detection system consisting of a full-matrix data acquisition instrument, a phased array probe, and angle wedges. Multiple frames of full-matrix data are acquired from different circumferential positions on one side of the pipeline to be inspected. For each reconstruction point within the imaging area, six wave modes are used to perform time-delay superposition and amplitude weighting processing on each frame of acquired full-matrix data. The strongest response from each frame is extracted and fused to reconstruct the contour of the prior unknown axial defect within the pipeline. This method considers the influence of pipeline curvature on ultrasonic wave propagation. By correcting the propagation time calculation, time-delay superposition and amplitude weighting processing, and multi-frame data fusion, it achieves imaging characterization and quantitative detection of prior unknown axial defects within the pipeline.

Description

Technical Field

[0001] This invention relates to the field of nondestructive testing technology, and in particular to a full-focus imaging method for axial defects in pipelines based on multi-frame, multi-mode wave fusion. Background Technology

[0002] Pipelines, serving as raw materials, product transmission media, and connecting components for pressure vessels, are widely used in fields such as petrochemicals and nuclear power. Because their production and service processes are susceptible to defects caused by processing techniques and the external environment, leading to serious production safety hazards, non-destructive testing is essential. Based on the pipe structure and defect distribution characteristics, defects are classified into circumferential defects and axial defects. Circumferential defects have their major axis perpendicular to the pipe axis, while axial defects have their major axis parallel to the pipe axis.

[0003] In current pipeline defect detection methods, ultrasonic testing and post-processing technology are widely used due to their safety and high efficiency. When detecting circumferential defects, the incident direction of the sound beam is in the same plane as the pipeline axis. In this case, the cross-section of the main sound beam propagation point in the test block is rectangular, thus often approximating a flat plate structure for detection and analysis. When detecting axial defects, the probe needs to be arranged along the circumference of the pipeline. In this case, the curvature of the inner and outer walls of the pipeline will cause the sound beam reflection and refraction path to deflect, thereby changing the propagation time of the sound. If this is not corrected, it will seriously affect the quantitative detection results of defects, causing severe distortion and discrepancies. In current ultrasonic testing research, defect contour reconstruction considering the influence of pipeline curvature has become a major focus. For example, when using the adaptive full-focusing method to detect axial defects in welds, the contour of the outer wall of the pipeline can be reconstructed first to correct the propagation time of the sound, thereby achieving contour reconstruction and quantitative detection of axial defects such as incomplete fusion and slag inclusions. L, Westerveld WJ, Haines HH, et al. Adaptive ultrasonic imaging of electric resistance welded pipeline seams [C]. 12th European Conference on Non-Destructive Testing, Gothenburg, 2018). However, this method requires the use of a water bag or water immersion coupling, and the adaptive full-focusing method has high requirements for the hardware of the detection equipment, which is often difficult to achieve in the engineering field.

[0004] The composite-mode total focusing method extracts A-scan signals from different array elements in the full matrix data and selectively performs time-delay superposition imaging processing on eight mode waves to improve the coverage of defect orientation angles. It has great potential in acquiring prior unknown defect contour features and can accurately quantify and locate defects (Jin SJ, Liu CF, Shi SQ, et al. Profile reconstruction and quantitative detection of planar defects with composite-mode total focusing method (CTFM)[J]. NDT&E International. 2021, 123: 102518). Currently, the composite-mode total focusing method is used to detect defects in planar structures; for pipeline inspection, curvature significantly affects the array signal beam path and acoustic time calculation process. Therefore, it is necessary to modify the existing total focusing method by incorporating pipeline curvature, increasing the number of acquisition frames, and optimizing the imaging mode waves to obtain more intuitive and complete defect detection images, thereby achieving identification and quantitative detection of axial defects in pipelines. Summary of the Invention

[0005] This invention provides a full-focus imaging method for axial defects in pipelines based on multi-frame, multi-mode wave fusion. Its purpose is to address the difficulty in reconstructing the profile of previously unknown axial defects due to beam deflection and dispersion of array signals under the curvature structure of pipelines. The method utilizes a phased array probe to acquire multiple frames of full-matrix data along the circumferential direction, and combines curvature correction with time-delay superposition processing of six wave modes. By weighting and filtering the time-delay superposition amplitudes of each reconstruction point within the detected area, a response amplitude matrix is ​​formed through fusion to reconstruct the axial defect profile and ultimately achieve quantitative detection.

[0006] The technical solution adopted in this invention is as follows:

[0007] A full-focus imaging method for axial defects in pipelines based on multi-frame, multi-mode wave fusion is proposed. The method utilizes a detection system consisting of a full-matrix data acquisition instrument 1, a phased array probe 2, and an angle wedge 3. Multiple frames of full-matrix data are acquired from different circumferential positions on one side of the pipeline to be inspected. For each reconstruction point within the imaging area, six wave modes are used to perform time-delay superposition and amplitude weighting on each frame of acquired full-matrix data. The strongest response from each frame is extracted and fused to achieve contour reconstruction and quantitative detection of previously unknown axial defects within the pipeline.

[0008] The specific steps are as follows:

[0009] Step 1. Determine the detection parameters

[0010] Based on the material, shape, and size information of the pipe to be tested, select the center frequency and number of array elements of the phased array probe 2, as well as the angle wedge 3 that matches the surface curvature of the pipe to be tested;

[0011] Step 2. Multi-frame full matrix data acquisition

[0012] Connect the full matrix data acquisition instrument 1, phased array probe 2, and angle wedge 3 in sequence. Place the phased array probe 2 on one side of the area to be detected, and move the phased array probe position along the circumference of the pipeline to acquire multiple frames of full matrix data. If the number of array elements of the selected phased array probe 2 is N, and data is acquired at M positions, then M frames containing N elements are obtained. 2 The full matrix data of each A-scan signal; during the acquisition of the s-th frame of data, the coordinates of the i-th element of the phased array probe are (x... s-i ,y s-i The coordinates of the j-th element are (x... s-j ,y s-j For the full matrix data acquired in the s-th frame, the signal transmitted by the i-th element and received by the j-th element is defined as L. s-ij (); where 1≤i≤N, 1≤j≤N, 1≤s≤M;

[0013] Step 3. Reconstruct the region grid

[0014] Using the axis of the pipe under test as the origin, the area to be tested is gridded, and each grid node is defined as an image reconstruction point, forming a reconstruction region with m×n points. The coordinates of any reconstruction point P are (a l ,b w ), where 1≤l≤m, 1≤w≤n;

[0015] Step 4. Multimode acoustic time calculation

[0016] When implementing full focusing, considering the refraction, reflection, and mode conversion of ultrasound waves at the pipe interface and defect surface, three propagation modes are generated: direct mode, half-span mode, and full-span mode, resulting in a total of 21 different mode waves. Taking into account the influence of the pipe interface and curvature, LL and TT mode waves are selected from the direct mode, TTL, TTT, and TLT mode waves from the half-span mode, and TTTT mode wave from the full-span mode as the imaging mode waves, where L represents the longitudinal wave and T represents the transverse wave. The transmit i, receive element j, and image reconstruction point P of the s-th frame of full matrix data are determined, and the total acoustic time t of the G-th mode wave is determined. s-G-ij (a l ,b w When taking the incident sound t s-G-ip (a l ,b w ) and the sound emitted at t s-G-pj (a l ,bw The sum of G, where 1 ≤ G ≤ 6;

[0017] t s-G-ij (a l ,b w )=t s-G-ip (a l ,b w )+t s-G-pj (a l ,b w )1)

[0018] According to Fermat's theorem, the refraction point of the sound beam at the interface satisfies the minimum total sound time. When detecting axial defects in a pipe, both the incident and exit points of the sound beam are located at the interface between the curved angle wedge 3 and the pipe under test. Let the incident point of the emitted sound beam of the i-th element be A(x). s-G-ip ,y s-G-ip The incident point of the received sound beam for the j-th element is B(x). s-G-pj ,y s-G-pj If the radius of the outer surface of the pipe is R1, the coordinates of the incident point A and the exit point B satisfy:

[0019]

[0020] Let the coordinates of the reflection point of the sound beam on the inner wall of the pipe during the transmission phase be C(x). s-G-qip ,y s-G-qip The coordinates of the reflection point of the sound beam on the inner wall of the pipe during the receiving stage are D(x). s-G-qpj ,y s-G-qpj If the inner radius of the pipe is R2, then the following condition is met:

[0021]

[0022] According to Fermat's theorem and Snell's law, during direct mode imaging, the acoustic time will satisfy:

[0023]

[0024] During half-span mode imaging, the acoustic time will satisfy:

[0025]

[0026] During full-mode imaging, the acoustic timing will satisfy:

[0027]

[0028] This is used to obtain the acoustic time t of each mode wave at the reconstruction point P. s-G-ij (a l ,b w );

[0029] From Ls-ij Select the sound time t in () s-G-ij (a l ,b w The corresponding amplitude L at point ) s-ij (t s-G-ij (a l ,b w At reconstruction point P, the amplitude of the signal transmitted by array element i and received by array element j in the s-th frame and the G-th mode wave.

[0030] Step 5. Delay overlay processing

[0031] For the N in the full matrix data of the s-th frame 2 Each A-scan signal undergoes time-delay superposition processing for the aforementioned six mode waves. This process is as follows:

[0032]

[0033] In the formula, I s-G (a l ,b w The reconstructed point P(a) is obtained by applying the G-mode time-delay superposition process to the s-th frame of data. l ,b w The absolute amplitude of )

[0034] At this point, I is obtained from the full matrix data acquired in the s-th frame. s-LL (a l ,b w ), I s-LL (a l ,b w ), I s-LL (a l ,b w ), I s-LL (a l ,b w ), I s-LL (a l ,b w ) and I s-LL (a l ,b w The response amplitude matrices for a total of 6 mode waves;

[0035] Repeat steps 4-5 for all M frames of full matrix data to obtain the response amplitude matrix of each mode wave in each frame.

[0036] Step 6. Amplitude-weighted and fused imaging

[0037] Subsequently, the six response amplitude matrices of the s-th frame are weighted; the weighting considers the imaging mode wave path length and the relative central angle φ between the probe and the center of the reconstructed region at that acquisition frame. sAs for the amplitude reduction caused by the reflection of different mode waves through the inner wall of the pipe, corresponding weighting coefficients are applied to different mode waves; let the depth of the imaging region center from the outer wall of the pipe be h, then the direct mode weighting coefficient ε is defined. s-direct ε, weighting coefficient of the half-span mode s-half and the weighted coefficient ε across all modes s-full They can be represented as:

[0038]

[0039]

[0040]

[0041] In the formula, ζ is the amplitude reduction coefficient of the mode wave after passing through the inner surface of the pipe, which is a constant;

[0042] For the full matrix data acquired in the s-th frame, the maximum amplitude of different mode waves after weighting at the reconstructed point P is extracted. This process is represented as follows:

[0043]

[0044] Based on this, for the processing results of all M frames of full matrix data, the maximum amplitude of different mode waves after weighting at the reconstruction point is compared and extracted to obtain the corresponding strongest response I. max (), this process is represented as

[0045]

[0046] Repeat steps 4-6 for each reconstructed point within the test area to obtain a baseline image for defect assessment.

[0047] Step 7. Qualitative and quantitative detection of axial defects in pipelines.

[0048] Based on the multi-frame, multi-mode wave fusion full-focus image obtained in step 6, defect assessment is performed, and the contour information of unknown axial defects in the pipeline is extracted for further qualitative identification. Finally, the coordinates of the defect endpoints are read, and the decibel descent method is used to calculate quantitative information such as the size, depth, and orientation angle of the defect.

[0049] The beneficial effects of this invention are as follows: This multi-frame, multi-mode wave fusion-based full-focus imaging method for axial defects in pipelines considers the influence of pipeline curvature on ultrasonic wave propagation. By optimizing the imaging mode waves, implementing time-delay superposition processing, and performing composite weighting and multi-frame image fusion on the mode wave amplitudes, it achieves contour reconstruction of previously unknown axial defects in pipelines while maintaining high quantitative accuracy. Furthermore, this method can be integrated into full-matrix data acquisition and processing instruments and implemented in conjunction with scanners, demonstrating significant application prospects and promotional value. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of the detection system used in this invention.

[0051] Figure 2 These are schematic diagrams of cross-sections of carbon steel pipe test blocks with different inclination angles for processing axial artificial grooves; (a) is an axial artificial groove with an orientation angle of -45°; (b) is an axial artificial groove with an orientation angle of -30°; (c) is an axial artificial groove with an inclination angle of 0°; (d) is an axial artificial groove with an inclination angle of 30°; (e) is an axial artificial groove with an inclination angle of 45°; where the vertical direction is 0° and the clockwise direction is positive.

[0052] Figure 3 The results are full-focus imaging of axial defects in pipelines based on multi-frame, multi-mode wave fusion; (a) is an axial artificial groove with an orientation angle of -45°; (b) is an axial artificial groove with an orientation angle of -30°; (c) is an axial artificial groove with an tilt angle of 0°; (d) is an axial artificial groove with an tilt angle of 30°; (e) is an axial artificial groove with an tilt angle of 45°.

[0053] In the diagram: 1-Full matrix data acquisition instrument; 2-Phase array probe; 3-Angle wedge. Detailed Implementation

[0054] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings and technical solutions.

[0055] A pipe axial defect full-focusing imaging method based on multi-frame, multi-mode wave fusion, employing a detection system such as... Figure 1 As shown, the system includes a full-matrix data acquisition instrument 1, a phased array probe 2, and an angle wedge made of tilted plexiglass 3. The dashed box inside the pipe represents the area to be imaged. The specific detection and processing steps are as follows:

[0056] Step (a): The test object is a carbon steel pipe block with a wall thickness of 20mm and an outer diameter of 200mm. An axial groove with a length of 5mm and a center depth of 10mm is machined inside the pipe wall, with orientation angles of -45°, -30°, 0°, 30°, and 45°. The vertical direction is 0°, and the clockwise direction is positive. Figure 2 As shown.

[0057] Step (b): Using a full matrix data acquisition instrument 1, a phased array probe 2 with a center frequency of 5MHz and an array element number of N=32 is used in conjunction with a curved surface wedge block 3 with a 45° longitudinal wave to detect defects. The sampling frequency is 125MHz, the longitudinal wave velocity of the wedge block 3 is 2337m / s, the transverse wave velocity of the carbon steel test block is 3230m / s, and the longitudinal wave velocity is 5900m / s.

[0058] Step (c): Position the phased array probe 2 on one side of the imaging area, and change the circumferential position of the phased array probe 2 to the relative central angle φ. s At positions of 13°, 18° and 23°, array signals were acquired using the full matrix data acquisition function of the full matrix data acquisition instrument 1, resulting in 3 frames of full matrix data containing 32×32 A-scan signals.

[0059] Step (d): Establish a Cartesian coordinate system and divide the imaging area into 100×100 rectangular grids. Using the multi-frame, multi-mode wave fusion-based full-focus imaging method for axial defects in pipelines proposed in this invention, for each grid point in the imaging area, the three frames of full matrix data acquired are time-delayed, superimposed, and weighted using six wave modes. Subsequently, for each grid point in the imaging area, the imaging amplitudes of the multi-frame, multi-mode waves are compared, and the strongest amplitude is extracted and fused to obtain the defect imaging characterization result, such as... Figure 3 As shown in the image, the major axis contours of each artificial groove are effectively represented, with quantitative errors of 0.59 mm, 1.14 mm, 0.30 mm, 0.83 mm, and 0.61 mm for length, 0.73°, 0.32°, 1.08°, 0.26°, and 1.85° for angle, and 0.55 mm, 0.15 mm, 0.05 mm, 0.20 mm, and 0.20 mm for depth. The positioning and quantitative errors are small, meeting engineering requirements.

[0060] The descriptions of the exemplary embodiments presented above are merely illustrative of the technical solutions of the present invention and are not intended to be exhaustive or to limit the invention to the precise forms described. Obviously, those skilled in the art can make many changes and variations based on the above teachings. The exemplary embodiments were chosen and described to explain the specific principles of the invention and its practical applications, thereby enabling other those skilled in the art to understand, implement, and utilize the various exemplary embodiments of the invention and their various alternatives and modifications. The scope of protection of the present invention is intended to be defined by the appended claims and their equivalents.

Claims

1. A method for full-focus imaging of axial defects in pipelines based on multi-frame, multi-mode wave fusion, characterized in that, This method is based on a detection system consisting of a full matrix data acquisition instrument (1), a phased array probe (2), and an angle wedge (3). Multiple frames of full matrix data are acquired from different circumferential positions on one side of the pipeline to be inspected. For each reconstruction point in the imaging area, six mode waves are used to perform time-delay superposition and amplitude weighting processing on each frame of acquired full matrix data. The strongest response in each frame of imaging results is extracted for data fusion, thereby realizing the contour reconstruction and quantitative detection of prior unknown axial defects in the pipeline. The specific steps are as follows: Step 1. Determine the detection parameters Based on the material, shape and size information of the pipe to be tested, select the center frequency and number of array elements of the phased array probe (2), as well as the angle wedge (3) that matches the surface curvature of the pipe to be tested. Step 2. Multi-frame full matrix data acquisition Connect the full matrix data acquisition instrument (1), the phased array probe (2), and the angle wedge (3) in sequence. Place the phased array probe (2) on one side of the pipe to be tested, and move the position of the phased array probe (2) around the circumference of the pipe to be tested to acquire multiple frames of full matrix data. If the number of array elements of the selected phased array probe (2) is N, and data is acquired at M positions, then M frames containing N elements are obtained. 2 The full matrix data of each A-scan signal; during the acquisition of the s-th frame of data, the coordinates of the i-th element of the phased array probe are (x... s-i ,y s-i The coordinates of the j-th element are (x... s-j ,y s-j For the full matrix data acquired in the s-th frame, the signal transmitted by the i-th element and received by the j-th element is defined as L. s-ij (); where 1≤i≤N, 1≤j≤N, 1≤s≤M; Step 3. Reconstruct the region grid Using the axis of the pipe under test as the origin, the area to be tested is gridded, and each grid node is defined as an image reconstruction point, forming a reconstruction region with m×n points. The coordinates of any reconstruction point P are (a l , b w ), where 1≤l≤m, 1≤w≤n; Step 4. Multimode acoustic time calculation When implementing full focusing, considering the refraction, reflection, and mode conversion of ultrasound waves at the pipe interface and defect surface, three propagation modes are generated: direct mode, half-span mode, and full-span mode, resulting in a total of 21 different mode waves. Taking into account the influence of the pipe interface and curvature, LL and TT mode waves are selected from the direct mode, TTL, TTT, and TLT mode waves from the half-span mode, and TTTT mode wave from the full-span mode as the imaging mode waves, where L represents the longitudinal wave and T represents the transverse wave. The transmit i, receive element j, and image reconstruction point P of the s-th frame of full matrix data are determined, and the total acoustic time t of the G-th mode wave is determined. s-G-ij (a l , b w When taking the incident sound, t s-G-ip (a l , b w ) and the sound output t s-G-pj (a l , b w The sum of G, where 1 ≤ G ≤ 6; 1）； According to Fermat's theorem, the refraction point of the sound beam at the interface satisfies the minimum total sound time. When detecting axial defects in a pipe, the incident and exit points of the sound beam are both located at the interface between the curved angle wedge (3) and the pipe under test. Let the incident point of the emitted sound beam of the i-th element be A(x). s-G-ip , y s-G-ip The incident point of the received sound beam for the j-th element is B(x). s-G-pj , y s-G-pj If the radius of the outer surface of the pipe to be measured is R1, and the coordinates of the incident point A and the exit point B satisfy: 2）； Let the coordinates of the reflection point of the sound beam on the inner wall of the pipe under test during the emission phase be C(x). s-G-qip , y s-G-qip The coordinates of the reflection point of the sound beam on the inner wall of the pipe under test during the receiving stage are D(x). s-G-qpj , y s-G-qpj If the inner radius of the pipe to be tested is R2, then the following conditions are met: 3）； According to Fermat's theorem and Snell's law, during direct mode imaging, the acoustic time will satisfy: 4）； During half-span mode imaging, the acoustic time will satisfy: 5）； During full-mode imaging, the acoustic time will satisfy: 6）； This is used to obtain the acoustic time t of each mode wave at the reconstruction point P. s-G-ij (a l , b w ); From L s-ij Select the sound time t in () s-G-ij (a l , b w The corresponding amplitude L at point ) s-ij (t s-G-ij (a l , b w At reconstruction point P, the amplitude of the signal transmitted by array element i and received by array element j in the s-th frame and the G-th mode wave. Step 5. Delay overlay processing For the N in the full matrix data of the s-th frame 2 Each A-scan signal undergoes time-delay superposition processing for the aforementioned six mode waves. This process is as follows: 7）； In the formula, I s-G (a l , b w The reconstructed point P(a) is obtained by applying the G-mode time-delay superposition process to the s-th frame of data. l , b w The absolute amplitude of ) At this point, I is obtained from the full matrix data acquired in the s-th frame. s-LL (a l , b w ), I s-LL (a l , b w ), I s-LL (a l , b w ), I s-LL (a l , b w ), I s-LL (a l , b w ) and I s-LL (a l , b w The response amplitude matrices for a total of 6 mode waves; Repeat steps 4-5 for all M frames of full matrix data to obtain the response amplitude matrix of each mode wave in each frame. Step 6. Amplitude-weighted and fused imaging Subsequently, the six response amplitude matrices of the s-th frame are weighted; the weighting considers the imaging mode wave path length and the relative central angle φ between the probe and the center of the reconstructed region at that acquisition frame. s As for the amplitude reduction caused by the reflection of different mode waves through the inner wall of the pipe, corresponding weighting coefficients are applied to different mode waves; let the depth of the imaging region center from the outer wall of the pipe be h, then the direct mode weighting coefficient ε is defined. s-direct ε, weighting coefficient of the half-span mode s-half and the weighted coefficient ε across all modes s-full They can be represented as: 8）； 9）； 10）； In the formula, ζ is the amplitude reduction coefficient of the mode wave after passing through the inner surface of the pipe, which is a constant; For the full matrix data acquired in the s-th frame, the maximum amplitude of different mode waves after weighting at the reconstructed point P is extracted. This process is represented as follows: 11）； Based on this, for the processing results of all M frames of full matrix data, the maximum amplitude of different mode waves after weighting at the reconstruction point is compared and extracted to obtain the corresponding strongest response I. max (), this process is represented as 12）； Repeat steps 4-6 for each reconstructed point within the test area to obtain a baseline image for defect assessment. Step 7. Qualitative and quantitative detection of axial defects in pipelines. Based on the multi-frame, multi-mode wave fusion full-focus image obtained in step 6, defect assessment is performed, and the contour information of unknown axial defects in the pipeline is extracted for further qualitative identification. Finally, the coordinates of the defect endpoints are read, and the decibel descent method is used to calculate quantitative information such as the size, depth, and orientation angle of the defect.

Images