Method for three-dimensional imaging of metal welded structures based on a matrix-addressed ultrasonic phased array

Abstract

This invention relates to the technical field of non-destructive testing of welded metal structures, and provides a three-dimensional imaging method for welded metal structures based on row-column addressed ultrasonic phased arrays. The method includes: performing ultrasonic phased array testing on the welded metal structure to acquire full matrix capture data; constructing a three-dimensional imaging Cartesian coordinate system for the welded structure; calculating the three-dimensional propagation paths of ultrasonic waves from the array's row and column sources to each imaging point in three-dimensional space according to spatial geometric relationships, and calculating the corresponding propagation times based on the propagation paths; performing denoising, delay summation, and vector coherence factor weighting on the full matrix capture data to obtain the final three-dimensional volumetric imaging result; and reconstructing and visualizing the data to obtain the three-dimensional imaging result of internal defects in the welded metal structure. Precise modeling of the ultrasonic propagation path within the three-dimensional imaging framework allows for full utilization of the multi-path information from the row and column sources, improving the imaging accuracy and spatial positioning precision of internal defects in the welded structure.

Description

Technical Field

[0001] This invention relates to the technical field of non-destructive testing of metal welded structures, particularly for the non-destructive testing of major equipment such as nuclear power and aerospace, and more specifically to a three-dimensional imaging method for metal welded structures based on row and column addressing ultrasonic phased array. Background Technology

[0002] Nuclear power equipment, as a key pressure-bearing and energy-transferring component in nuclear power generation systems, is widely used in reactor cooling systems, main steam pipelines, and pressure boundary structures. Its operational safety directly impacts the stable operation of nuclear power plants and public safety. Metal welded structures, especially pipe welds, dissimilar steel welds, and welds on thick-walled components, are subjected to complex service environments such as high temperature, high pressure, radiation, corrosion, and alternating loads. These areas are among the weakest points in terms of structural integrity and stress concentration. During manufacturing and service, defects such as porosity, slag inclusions, cracks, incomplete penetration, and lack of fusion are easily generated within welded structures. Failure to detect and accurately assess these defects in a timely manner can lead to media leakage, structural failure, or even serious nuclear safety accidents. Therefore, conducting highly reliable and precise non-destructive testing and defect characterization of critical metal welded structures has significant engineering and safety value.

[0003] Currently, methods such as radiographic testing and computed tomography (CT) are commonly used in weld inspection. While these methods offer advantages in defect resolution, they generally suffer from high radiation risks, high inspection costs, limited on-site space, and difficulty in achieving online or in-situ inspection, making it difficult to meet the practical needs of in-service inspection and rapid assessment. In contrast, ultrasonic phased array testing technology offers advantages such as strong penetration capability, no radiation to the human body, high inspection efficiency, and the ability to achieve multi-angle imaging, making it an important technical means for non-destructive testing of welded structures.

[0004] However, welded metal structures generally exhibit complex microstructures, coarse grains in the weld zone, significant anisotropy, and non-uniform microstructures, particularly prominent in multi-pass welds, welds with excessive weld reinforcement, and welds made of dissimilar materials. These microstructural characteristics lead to strong scattering, attenuation, and mode conversion phenomena during ultrasonic wave propagation in the weld region, significantly enhancing structural noise in the received ultrasonic echoes and severely impacting the imaging signal-to-noise ratio and defect contrast. Traditional ultrasonic phased array imaging methods in such coarse-grained welds often exhibit strong background noise, blurred defect boundaries, and obvious artifacts, making reliable defect identification and accurate characterization difficult.

[0005] Ultrasonic imaging techniques, such as Total Focusing Method (TFM), can theoretically achieve coherent focusing across the entire aperture and angle, and are widely used in weld inspection. However, their imaging quality is highly dependent on the coherence and signal-to-noise ratio of the echo signal. When the weld area contains coarse-grained structures or geometrically irregular structures, microstructure scattering noise accumulates significantly in the full-matrix capture data. This makes it difficult for conventional focusing algorithms, even if they can generate defect indications, to meet the high requirements of reliable defect detection and accurate assessment in welds.

[0006] Furthermore, current ultrasonic testing of welds in engineering practice still primarily relies on two-dimensional imaging, which typically only obtains defect information from a single cross-section or limited perspective of the weld, making it difficult to fully reflect the three-dimensional spatial distribution, geometric morphology, and extension characteristics of defects. For strip-shaped defects distributed along the weld direction or defects with complex spatial morphology, two-dimensional imaging often requires multiple scans and manual judgment based on experience, resulting in low detection efficiency and the risk of missed detections. This makes it difficult to meet the equipment's requirements for high reliability and automated testing.

[0007] In recent years, row-column addressed ultrasonic phased arrays have shown great potential for engineering applications in the field of three-dimensional ultrasonic imaging due to their ability to significantly reduce the number of system channels while maintaining equivalent full-aperture imaging capabilities. However, existing research has mostly focused on fields such as medical ultrasound, and their application in the inspection of welded metal structures remains relatively limited. In particular, there is a lack of mature and effective technical solutions for achieving high-quality three-dimensional imaging under conditions of complex weld geometry and coarse-grained microstructure.

[0008] Therefore, how to fully utilize the three-dimensional imaging advantages of row-column addressable ultrasonic phased arrays under the complex microstructure and geometry of welded structures, construct a reasonable three-dimensional acoustic propagation model, and significantly suppress structural noise caused by coarse-grained microstructures through effective signal processing and imaging strategies to achieve high signal-to-noise ratio three-dimensional imaging and visualization characterization of welding defects has become a key technical problem that urgently needs to be solved in the field of weld nondestructive testing. Summary of the Invention

[0009] To address the aforementioned technical problems, this invention provides a three-dimensional imaging method for metal welded structures based on row-column addressing ultrasonic phased arrays, comprising: S1, ultrasonic phased array testing of metal welded structures is carried out using an oblique incidence method with organic glass wedges, and full matrix capture data is obtained using a row and column addressing ultrasonic phased array. S2, based on the single row excitation and all column reception modes of row- and column excitation and all row reception modes of row- and column-addressed ultrasonic phased array, constructs a three-dimensional imaging rectangular coordinate system for welded structures; and confirms the geometric positions and spatial mapping relationships between array row sources, column sources and spatial three-dimensional imaging points. S3. Based on the spatial geometric relationship, calculate the three-dimensional propagation path of the ultrasonic wave from the array row source and column source to each imaging point in three-dimensional space, and calculate the corresponding propagation time based on the propagation path; S4 performs non-convex overlap group sparse variational denoising on the full matrix capture data to suppress structural noise caused by the coarse grain structure of the welded structure. S5, based on the propagation time, performs time-delay summation on the denoised full matrix capture data to achieve three-dimensional full-focus imaging of the welded structure and obtain the initial three-dimensional volume imaging result; S6. The initial three-dimensional volumetric imaging results are weighted by vector coherence factor to enhance the coherence of the defect echo and further suppress background noise, thus obtaining the final three-dimensional volumetric imaging results. S7 performs 3D reconstruction and visualization of the final 3D volumetric imaging results to obtain 3D imaging results of internal defects in the metal welded structure.

[0010] Furthermore, in step S1, the full matrix capture data includes ultrasonic echo data obtained under two acquisition methods: single row excitation and full column reception, and single column excitation and full row reception.

[0011] Furthermore, in step S2, the upper surface of the metal welded structure is set as... The reference plane and the geometric axisymmetric center of the row and column addressing ultrasonic phased array are defined as the origin O(0,0,0), and a three-dimensional imaging rectangular coordinate system including the thickness direction of the welded structure, the weld direction and the transverse direction is established. Among them, the row sources of the array are parallel to Plane, column source parallel to Plane and perpendicular to Plane, the imaging point is located flat.

[0012] Furthermore, in step S3, The equivalent line source positions of row and column sources in three-dimensional space are determined. The geometric axisymmetry center of the row-column addressed ultrasonic phased array is defined as the origin. Taking the position Z of the short side of the wedge (bottom of the row source and the first linear array element of the column source) as the reference, and combining the line source spacing and the wedge angle, the spatial positions of each equivalent row and column source are solved to obtain the linear equation of each line source. ; Propagation time from each imaging point to the column source for:

[0013] In the formula, , For the source in The x and y coordinates of the projection points on the plane are obtained by solving the equation of the line source. , For imaging points exist The x and y coordinates of the projection point on the plane. and These represent the sound velocity in the plexiglass wedge and the metal welding material, respectively. The x-coordinate of the point of refraction of the sound wave emitted by the source after refraction at the interface between the wedge and the workpiece is obtained by combining formulas (2) and (3);

[0014] In the formula, It is the angle between the incident sound wave and the normal to the refracting plane in the plexiglass wedge; It is the angle between the refracted sound wave and the normal to the refracting plane in the metal welding material; Since the row source does not have any special characteristics in the three-dimensional coordinate system, the problem of finding the optimal distance from a pixel to the row source is transformed into finding the optimal point on the row source. And calculate the optimal point. Considering refraction in three-dimensional space for the imaging point The sound propagation path; the refraction point where the sound wave emitted from the source is refracted at the interface between the wedge and the workpiece. Introducing a key scaling parameter To describe, the point of refraction Expressed as:

[0015] Therefore, the propagation time from each imaging point to the line source... for:

[0016] In the formula, For the best With imaging point The horizontal distance between the projection points on the Z=0 plane; ; Formula (5) uses the one-dimensional optimization method in MATLAB to obtain the shortest path. That is, the optimal point for ultrasound transmission. propagation to imaging point The propagation time (ToF).

[0017] Furthermore, in step S4, the non-convex overlapping group sparse variational denoising process utilizes the group sparsity characteristics of defect echoes in time or space to suppress structured scattering noise in the welded structure.

[0018] Furthermore, in step S5, the initial three-dimensional volumetric imaging results The expression is:

[0019] In the formula, M and N are the number of row sources and column sources of the phased array, respectively. This is the Hilbert transform of the ultrasonic signal.

[0020] Furthermore, in step S6, the final expression for the three-dimensional volumetric imaging result is as follows:

[0021] Among them, vector coherence factor for:

[0022] In the formula, For the index of the summation loop, for The real part; for The imaginary part.

[0023] Furthermore, the internal defects in step S7 include porosity defects, slag inclusion defects, crack defects, and incomplete penetration defects.

[0024] Furthermore, the welded structure is a stainless steel weld, a low alloy steel weld, or a dissimilar metal weld in nuclear power, aerospace, or other large-scale high-end equipment.

[0025] The present invention has the following beneficial effects: (1) By introducing a row-column addressable ultrasonic phased array, the present invention significantly reduces the number of system channels and hardware complexity, while realizing three-dimensional volume imaging of welded structures. This breaks through the limitations of traditional two-dimensional ultrasonic imaging in terms of spatial information acquisition and can fully reflect the three-dimensional spatial morphology and distribution characteristics of welding defects.

[0026] (2) The present invention performs precise modeling of the ultrasonic propagation path under the three-dimensional imaging framework, so that the multi-path information of row source and column source can be fully utilized, thereby improving the imaging accuracy and spatial positioning accuracy of internal defects in welded structures.

[0027] (3) In the process of three-dimensional imaging, the row source and column source in the row and column addressing ultrasonic phased array are equivalent to line sources. Based on their spatial position in the three-dimensional coordinate system, the ultrasonic propagation path between the line source and any imaging point is calculated. In the process of calculating the propagation path, the refraction propagation of ultrasonic waves at the interface between the wedge and the metal welded structure, as well as the straight propagation characteristics inside the welded structure, are comprehensively considered. The effective propagation path of the corresponding imaging point is determined according to the principle of the shortest propagation time, so as to obtain the propagation time information for three-dimensional imaging.

[0028] (4) This invention introduces non-convex overlap group sparse variational denoising processing at the data level and introduces vector coherence factor enhancement mechanism at the imaging level, thereby effectively suppressing noise in coarse-grained weld structures, significantly improving the imaging signal-to-noise ratio and defect contrast, and enabling smaller defects and complex morphological defects to be presented more clearly and reliably, thereby improving the reliability and engineering applicability of non-destructive testing of welded structures.

[0029] (5) The present invention can achieve high-quality three-dimensional defect imaging under complex weld structure conditions, which has good engineering applicability and promotion value, and is of great significance for improving the non-destructive testing level of key welded structures of high-end equipment. Attached Figure Description

[0030] Figure 1 This is a flowchart of the three-dimensional imaging method in this invention.

[0031] Figure 2 This is a schematic diagram of the row-column addressing ultrasonic phased array excitation and reception method in this invention.

[0032] Figure 3 This is a schematic diagram of the ultrasonic propagation path calculation of a point-value source in a three-dimensional coordinate system according to the present invention.

[0033] Figure 4 This is a schematic diagram of the ultrasonic propagation path calculation of a point-value row source in a three-dimensional coordinate system according to the present invention.

[0034] Figure 5 This is a three-dimensional imaging comparison result of the porosity defects in the metal welded joint in this invention.

[0035] Figure 6 This is a three-dimensional imaging comparison result of slag inclusion defects in metal welded joints in this invention.

[0036] Figure 7 It is a three-dimensional scan view of the entire weld area of ​​a metal pipe welded joint.

[0037] Figure 8 Radiographic inspection image of welded joints in metal pipes. Detailed Implementation

[0038] The technical solution of the present invention will be further described in detail below with reference to specific embodiments. However, these embodiments are not intended to limit the present invention. Any similar structures and similar variations of the present invention should be included in the protection scope of the present invention. The commas in the present invention all indicate the relationship between and. The English letters in the present invention are case-sensitive.

[0039] This embodiment provides a three-dimensional imaging method for metal welded structures based on row and column addressing ultrasonic phased array, which is suitable for three-dimensional detection and visualization of defects in metal pipe welded joints under complex conditions such as weld reinforcement and coarse grain structure.

[0040] like Figure 1 As shown, the three-dimensional imaging method includes: S1, ultrasonic phased array testing of the metal welded structure was conducted using an oblique incidence method with plexiglass wedges. A row-column addressed ultrasonic phased array was used to acquire full matrix capture data. The full matrix capture data includes ultrasonic echo data obtained under two acquisition methods: single row excitation with all columns received, and single column excitation with all rows received. Figure 2 As shown in the figure and This indicates the sequential index of column and row sources, thereby obtaining equivalent full-aperture ultrasonic echo information under the condition of significantly reduced channel number.

[0041] In actual testing, the phased array probe is coupled to the surface of the weld joint via a wedge, scanning the weld area in an oblique incidence manner to adapt to the weld reinforcement structure and improve the acoustic beam coverage of the weld root and sidewall fusion area. Through the above-mentioned row and column addressing excitation and reception method, full matrix capture data containing multi-angle and multi-propagation path information is acquired, providing basic data support for subsequent 3D imaging.

[0042] S2, based on the single row excitation and all column reception modes of row- and column excitation and all row reception modes of row- and column-addressed ultrasonic phased array, constructs a three-dimensional imaging rectangular coordinate system for welded structures; and confirms the geometric positions and spatial mapping relationships between array row sources, column sources and spatial three-dimensional imaging points. Specifically, the upper surface of the metal welded structure is set as... The reference plane and the geometric axisymmetric center of the row and column addressing ultrasonic phased array are defined as the origin O(0,0,0). A three-dimensional imaging rectangular coordinate system is established, which includes the thickness direction of the welded structure, the weld direction, and the transverse direction, for spatial positioning and three-dimensional reconstruction of internal defects in the weld. Among them, the row sources of the array are parallel to Plane, column source parallel to Plane and perpendicular to Plane, the imaging point is located flat.

[0043] S3. Based on the spatial geometric relationship, calculate the three-dimensional propagation path of the ultrasonic wave from the array row source and column source to each imaging point in three-dimensional space, and calculate the corresponding propagation time based on the propagation path; like Figures 3-4As shown, specifically: the equivalent line source positions of row and column sources in three-dimensional space are determined. The geometric axisymmetric center of the row-column addressed ultrasonic phased array is defined as the origin of the coordinate system. Taking the position of the short side of the wedge (bottom of the row source and the first linear array element of the column source) Z=5mm as the reference, and combining the line source spacing and the wedge angle, the spatial positions of each equivalent row and column source are solved to obtain the linear equation of each line source. The ultrasonic propagation path between the line source and any imaging point is calculated. During the calculation of the propagation path, the refraction propagation of ultrasonic waves at the interface between the wedge and the metal welded structure, as well as the straight-line propagation characteristics inside the welded structure, are comprehensively considered. Based on the principle of the shortest propagation time, the effective propagation path of the corresponding imaging point is determined, thereby obtaining the propagation time information for three-dimensional imaging.

[0044] Propagation time from each imaging point to the column source for:

[0045] In the formula, , For the source K in The x and y coordinates of the projection point on the plane. , For imaging points exist The x and y coordinates of the projection point on the plane. and These represent the sound velocity in the plexiglass wedge and the metal welding material, respectively. Let Q be the abscissa of the refraction point Q after the sound wave emitted by the source is refracted through the interface between the wedge and the workpiece. It is obtained by combining formulas (2) and (3).

[0046] In the formula, It is the angle between the incident sound wave and the normal to the refracting plane in the plexiglass wedge; It is the angle between the refracted sound wave and the normal to the refracting plane in the metal welding material.

[0047] Since the row source does not have any special characteristics in the three-dimensional coordinate system, the problem of finding the optimal distance from a pixel to the row source is transformed into finding the optimal point on the row source. And calculate the optimal point. Considering refraction in three-dimensional space for the imaging point The sound propagation path; the refraction point where the sound wave emitted from the source is refracted at the interface between the wedge and the workpiece. Introducing a key scaling parameter To describe, the point of refraction Expressed as:

[0048] Therefore, the propagation time from each imaging point to the line source... for:

[0049] In the formula, For the best With imaging point The horizontal distance between the projection points on the Z=0 plane; ; Formula (5) uses the one-dimensional optimization method in MATLAB to obtain the shortest path. That is, the optimal point for ultrasound transmission. propagation to imaging point The propagation time (ToF).

[0050] S4 addresses the strong scattering noise problem caused by coarse-grained structures commonly found in welded structures by performing non-convex overlapping group sparse variational denoising on the full matrix capture data. This processing method utilizes the group continuity characteristics of defect echoes in time and space to effectively suppress structured scattering noise, thereby improving the signal-to-noise ratio of the effective defect echo signal and providing a cleaner data foundation for subsequent imaging.

[0051] The specific steps for non-convex overlapping group sparse variational denoising processing of the full matrix captured data are as follows: The following mathematical model is set for the ultrasonic echo received by the ultrasonic probe:

[0052] in, To capture data across the entire matrix, For a valid FMC signal, This is structural scattering noise.

[0053] In order to restore a valid signal Solve the following energy minimization problem:

[0054] in, For data fidelity items, For regularization terms, For regularization parameters, The regularization term represents the optimal FMC signal recovered after noise reduction. Defined as a sparse prior of a non-convex overlapping group:

[0055] In the formula, It is a sparse transformation matrix. For a predefined set of groups that allows overlap, Indicates that the coefficients belong to the group Part; For non-convex sparse promoting functions, such as Norms, in their role, impose more radical sparsity constraints on the energy of the entire group, and can more accurately approximate the real sparse structure than convex canonicals.

[0056] Because non-convex terms are coupled with transformations and are difficult to solve directly, the Alternating Direction Multiplier Method (ADMM) framework is adopted to decompose them through variable splitting. First, auxiliary variables are introduced. The problem is transformed into an equivalent constrained optimization form, and its augmented Lagrangian function is constructed:

[0057] in, For Lagrange multipliers, For penalty parameters, The sparse coefficient subvector corresponding to the g-th overlapping group; ADMM uses an alternating direction multiplier framework to alternately update three variables, fixing the g-th group. Sparse domain auxiliary variable in the next iteration and the Lagrange multipliers in the next iteration ,about Minimize the Lagrange function:

[0058] This is about The quadratic convex optimization problem is obtained by solving the following normal equation:

[0059] In the formula, For the first The effective FMC signal after the next iteration update For FMC data dimension identity matrix; For the inverse / transpose of the sparse transformation; For the sparse domain auxiliary variable z, let Updates are performed in parallel by group:

[0060] In the formula, For the first The median value of the sparse domain corresponding to each group; For intermediate input variables in the sparse domain; For the first The second iteration The sparsity coefficient of the group; To handle nonconvexity, an iterative reweighting strategy is adopted. Approximated as the first Group No. Weight of the next iteration This transforms the non-convex problem into a series of weighted group soft thresholding operations, for the ... Groups:

[0061] in, Dynamically calculated based on the results of the previous iteration:

[0062] In the formula, For soft thresholding operators, weak energy groups are set to zero, and strong energy groups are collected. Non-convex function The first derivative; This operator sets all group coefficients below a threshold to zero, otherwise shrinks them, effectively promoting group sparsity. Finally, the Lagrange multipliers are updated:

[0063] In the formula, For the first The Lagrange multipliers updated in the next iteration; Repeat the update steps for the three variables until convergence (relative). The change in the variable is less than Or reach the maximum number of iterations The final output target signal .

[0064] S5, based on the propagation time, performs time-delay summation on the denoised full-matrix capture data to achieve three-dimensional full-focus imaging of the welded structure and obtain the initial three-dimensional volumetric imaging result. The expression is:

[0065] In the formula, M and N are the number of row sources and column sources of the phased array, respectively. This is the Hilbert transform of the ultrasonic signal.

[0066] S6. The initial 3D volumetric imaging results are weighted by vector coherence factor to obtain the final 3D volumetric imaging results, expressed as follows:

[0067] Among them, vector coherence factor for:

[0068] In the formula, For the index of the summation loop, for The real part; for The imaginary part.

[0069] By analyzing the phase consistency of the complex signal corresponding to the imaging point, the coherent defect echo is enhanced, while incoherent noise and sidelobe interference are suppressed, so that the defect area presents higher contrast and clearer boundary in the three-dimensional imaging result.

[0070] S7 performs three-dimensional reconstruction and visualization of the final three-dimensional volume imaging results to obtain three-dimensional imaging results of internal defects in the metal welded structure; internal defects include porosity defects, slag inclusion defects, crack defects and incomplete penetration defects.

[0071] To verify the effectiveness of the method of the present invention, typical welding defects such as porosity and slag inclusions were set in the welded joint of metal pipes, and the method of the present invention was used to perform three-dimensional volume imaging to perform non-destructive testing on the internal defects of the welded joint. Figure 5 The results show the three-dimensional imaging comparison of porosity defects in metal welded joints. Figure 6 The results are three-dimensional imaging comparisons of slag inclusion defects in welded joints, among which... Figure 5 , Figure 6 Figures (a)-(d) are three-dimensional images of internal porosity defects and slag inclusion defects of the welded joint obtained without non-convex overlap group sparse variational denoising processing, while other processing is consistent with the method of the present invention. Figure 5 , Figure 6 Figures (e)-(h) are three-dimensional images of internal porosity and slag inclusion defects in welded joints obtained using the method of this invention; wherein, Figure 5 , Figure 6 Figures (a) and (e) are three-dimensional views. Figure 5 , Figure 6 Figures (b) and (f) show the location of the defect. Sectional view, Figure 5 , Figure 6 Figures (c) and (g) are Sectional view, Figure 5 , Figure 6 Figures (d) and (h) are Cross-sectional view. From Figure 5 and Figure 6 Figures (a)-(d) show that the un-denoised images have high background noise, low overall signal-to-noise ratio, and overlapping defect signals with background noise. Defect edges are blurred, contrast is insufficient, and artifacts exist in some sections, affecting defect identification and localization. Figure 5As can be seen from Figures (e)-(h), the imaging results generated by the method of this invention have significantly suppressed background noise, produced clear and sharp defect signals, and accurately restored the spatial positions and morphologies of the two pore defects in the three-dimensional stereoscopic view. The defect outlines are clear and accurately positioned in each two-dimensional cross-sectional view, with no obvious artifacts or trailing phenomena. Figure 6 As can be seen from Figures (e)-(h), the background noise of the imaging results generated by the method of the present invention is significantly suppressed, the contrast between the defect signal and the background is greatly improved, the spatial position and extension morphology of the slag inclusion defect in the three-dimensional stereoscopic view are presented more clearly, the defect contours in each two-dimensional cross-sectional view are sharper and the positioning is more accurate, effectively eliminating background noise interference, significantly improving the imaging signal-to-noise ratio and defect recognition accuracy, and realizing high-precision three-dimensional visualization non-destructive testing of slag inclusion defects in welded joints.

[0072] from Figures 5-6 As can be seen, the three-dimensional imaging results generated by the method of the present invention effectively suppress background noise and artifacts, while significantly improving the imaging contrast and contour clarity of weld joint defects, and realizing accurate three-dimensional positioning and morphological restoration of defects.

[0073] In addition, the overall 3D view obtained after performing a 3D scan of the entire weld area of ​​the metal pipe weld joint is as follows: Figure 7 As shown, the spatial distribution of two types of defects within the weld—porosity and slag inclusions—is fully presented. The central 3D view displays the overall imaging result of the weld area, the upper left magnified view clearly shows the imaging characteristics of slag inclusions, and the lower right magnified view clearly shows the imaging characteristics of porosity. From... Figure 7 As can be seen, the non-destructive testing method of this invention can achieve three-dimensional visualization scanning of the entire weld seam. It can not only accurately locate the three-dimensional spatial coordinates of different types of defects within the weld seam, but also clearly distinguish the morphological characteristics of slag inclusions and porosity defects: slag inclusion defects are distributed in clusters, with concentrated echo signals over a large area; porosity defects are distributed as isolated points, with concentrated echo signals and clear boundaries. This overall three-dimensional view intuitively reflects the location, shape, and distribution of defects within the weld seam, providing a comprehensive and intuitive three-dimensional visualization basis for weld joint quality assessment, defect tracing, and safety evaluation.

[0074] The test results show that the present invention can clearly and stably detect a variety of defects inside the welded joint and accurately reflect its three-dimensional spatial morphology.

[0075] Meanwhile, radiographic testing was used to compare and verify the same welded joints, and the test results are as follows: Figure 8 As shown, Figure 8The lower center shows the overall radiographic image of the weld, while the upper center shows magnified views of two defect areas. The magnified image on the left indicates slag inclusion, and the magnified image on the right indicates porosity, clearly demonstrating the image characteristics of these two types of defects in radiographic inspection. The comparative results show that the three-dimensional ultrasonic imaging results obtained by the method of this invention are highly consistent with the radiographic inspection results in terms of defect location, morphology, and distribution characteristics: the discontinuous, strip-like distribution characteristics of slag inclusions and the circular, isolated distribution characteristics of porosity defects are consistently presented in both detection methods, and the spatial location of the defects completely corresponds to the actual location of the weld. This comparison verifies the accuracy and reliability of this invention in detecting porosity and slag inclusion defects in welded joints, proving that its imaging results can truly reflect the actual state of internal defects in the weld and can provide a reliable basis for the quality assessment of welded components.

[0076] In summary, the non-destructive testing method for welded joints of metal pipes based on row and column addressing ultrasonic phased array three-dimensional imaging described in this embodiment can achieve high signal-to-noise ratio three-dimensional defect imaging under complex weld structure and geometry, significantly improving defect detection capability and spatial characterization capability, and has good engineering application prospects.

[0077] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

Claims

1. A three-dimensional imaging method for metal welded structures based on row-column addressed ultrasonic phased array, characterized in that, include: S1, ultrasonic phased array testing of metal welded structures is carried out using an oblique incidence method with organic glass wedges, and full matrix capture data is obtained using a row and column addressing ultrasonic phased array. S2, based on the single row excitation and all column reception modes of row- and column excitation and all row reception modes of row- and column-addressed ultrasonic phased array, constructs a three-dimensional imaging rectangular coordinate system for welded structures; and confirms the geometric positions and spatial mapping relationships between array row sources, column sources and spatial three-dimensional imaging points. S3. Based on the spatial geometric relationship, calculate the three-dimensional propagation path of the ultrasonic wave from the array row source and column source to each imaging point in three-dimensional space, and calculate the corresponding propagation time based on the propagation path; Specifically, the equivalent line source positions of row and column sources in three-dimensional space are determined. The geometric axisymmetry center of the row-column addressed ultrasonic phased array is defined as the origin of the coordinate system. Taking the short side position Z of the wedge as the reference, and combining the line source spacing and the wedge angle, the spatial positions of each equivalent row and column source are solved to obtain the linear equation of each line source. ; Propagation time from each imaging point to the column source for: In the formula, , For the source in The x and y coordinates of the projection point on the plane. , For imaging points exist The x and y coordinates of the projection point on the plane. and These represent the sound velocity in the plexiglass wedge and the metal welding material, respectively. The x-coordinate of the point of refraction of the sound wave emitted by the source after refraction at the interface between the wedge and the workpiece is obtained by combining formulas (2) and (3); In the formula, It is the angle between the incident sound wave and the normal to the refracting plane in the plexiglass wedge; It is the angle between the refracted sound wave and the normal to the refracting plane in the metal welding material; Since the row source does not have any special characteristics in the three-dimensional coordinate system, the problem of finding the optimal distance from a pixel to the row source is transformed into finding the optimal point on the row source. And calculate the optimal point. Considering refraction in three-dimensional space for the imaging point The sound propagation path; the refraction point where the sound wave emitted from the source is refracted at the interface between the wedge and the workpiece. Introducing a key scaling parameter To describe, the point of refraction Expressed as: Therefore, the propagation time from each imaging point to the line source... for: In the formula, For the best With imaging point The horizontal distance between the projection points on the Z=0 plane; ; Formula (5) uses the one-dimensional optimization method in MATLAB to obtain the shortest path. That is, the optimal point for ultrasound transmission. propagation to imaging point The duration of transmission; S4 performs non-convex overlapping group sparse variational denoising on the full matrix captured data. S5, based on the propagation time, performs time-delay summation on the denoised full matrix capture data to achieve three-dimensional full-focus imaging of the welded structure and obtain the initial three-dimensional volume imaging result; S6, the initial three-dimensional volume imaging results are weighted by vector coherence factor to obtain the final three-dimensional volume imaging results; S7 performs 3D reconstruction and visualization of the final 3D volumetric imaging results to obtain 3D imaging results of internal defects in the metal welded structure.

2. The three-dimensional imaging method for metal welded structures based on row-column addressing ultrasonic phased array according to claim 1, characterized in that, In step S1, the full matrix capture data includes ultrasonic echo data obtained under two acquisition methods: single row excitation and full column reception, and single column excitation and full row reception.

3. The three-dimensional imaging method for metal welded structures based on row-column addressing ultrasonic phased array according to claim 1, characterized in that, In step S2, the upper surface of the metal welded structure is set as... The reference plane and the geometric axisymmetric center of the row and column addressing ultrasonic phased array are defined as the origin O(0,0,0), and a three-dimensional imaging rectangular coordinate system including the thickness direction of the welded structure, the weld direction and the transverse direction is established. Among them, the row sources of the array are parallel to Plane, column source parallel to Plane and perpendicular to Plane, the imaging point is located flat.

4. The three-dimensional imaging method for metal welded structures based on row-column addressing ultrasonic phased array according to claim 1, characterized in that, In step S4, the non-convex overlapping group sparse variational denoising process utilizes the group sparsity characteristics of defect echoes in time or space to suppress structured scattering noise in the welded structure.

5. The three-dimensional imaging method for metal welded structures based on row-column addressing ultrasonic phased array according to claim 1, characterized in that, In step S5, the initial three-dimensional volumetric imaging results The expression is: In the formula, M and N are the number of row sources and column sources of the phased array, respectively. This is the Hilbert transform of the ultrasonic signal.

6. The three-dimensional imaging method for metal welded structures based on row-column addressing ultrasonic phased array according to claim 5, characterized in that, In step S6, the final expression of the three-dimensional volumetric imaging result is as follows: Among them, vector coherence factor for: In the formula, For the index of the summation loop; for The real part; for The imaginary part.

7. The three-dimensional imaging method for metal welded structures based on row-column addressing ultrasonic phased array according to claim 1, characterized in that, The internal defects in step S7 include porosity defects, slag inclusion defects, crack defects, and incomplete penetration defects.

8. The method for three-dimensional imaging of metal welded structures based on row-column addressing ultrasonic phased array according to claim 1, characterized in that, The welded structure is a stainless steel weld, a low alloy steel weld, or a dissimilar metal weld in large-scale high-end equipment used in nuclear power and aerospace.

Images