A synthetic aperture imaging technique for ultrasonic spiral scanning of metal round bars

Abstract

The purpose of the present invention is to provide a synthetic aperture pSAFT focusing imaging algorithm in a polar coordinate system and a synthetic aperture imaging technology for ultrasonic spiral scanning and detection of metal round bars based on the algorithm. The specific imaging technology is: firstly, after a spiral scanning of the metal round bar, a data set is obtained, and then the semi-aperture angle θ′ of the workpiece surface and the radius r of the inscribed circle that can be covered by the sound beam at all positions are calculated; for all points A (ρ, α) in the round bar, synthetic focusing calculations are performed according to the synthetic aperture focusing imaging formula in the polar coordinate system with appropriate imaging resolution, so as to obtain a data matrix Sp (i, j); finally, the polar coordinate data matrix is ​​coordinate-transformed to display a circular cross-section tomographic scan. The method can meet the online real-time detection requirements as much as possible while improving the detection resolution, and intuitively display the information such as the size and distribution of defects in the bar in the form of a tomographic scan similar to CT imaging, providing richer and more accurate information for the quality evaluation of the bar.

Description

Technical Field

[0001] The invention belongs to the field of nondestructive testing, and in particular provides a synthetic aperture imaging technology for ultrasonic spiral scanning and testing of a metal round bar. Background Art

[0002] Metal round bars and their parts made by casting, forging, powder metallurgy and other processes are widely used in various industrial fields. If the defects in the manufacturing process are not discovered and eliminated in time, it is easy to reduce the reliability of industrial equipment operation, and even cause fracture accidents, resulting in casualties and huge economic losses. Nondestructive testing and evaluation, as an important branch of material processing engineering, not only plays an important role in product material quality and production process quality control, coordination of performance and economic benefits, but also is an important technical means to ensure the safety of materials and their products. Therefore, it is of great significance to carry out research to improve the level of nondestructive testing technology for defects in metal bars and their parts and the ability to control bar quality.

[0003] For metal bars, the commonly used non-destructive testing technologies include five categories: X-ray, magnetic powder, penetration, eddy current and ultrasonic testing. Among them, ultrasonic testing technology has the advantages of high reliability, rich characteristic parameters and easy automation, and has been widely used. Most of the current ultrasonic automatic testing equipment for metal bars is based on the ultrasonic bar testing standard, which determines the detection sensitivity by artificial defects, uses ultrasonic probes to perform spiral scanning on the bars, and uses Go / No Go simple alarm method to conduct qualification assessment. For example, the national standard GB / T 37566-2019 stipulates that longitudinal wave water immersion focusing automated equipment can be used to detect transverse hole defects in round steel, that is, the size of the defect is evaluated based on the A-scan signal amplitude alarm. Because the defect morphology and distribution in the bar are complex and diverse, the defect information provided by the current bar ultrasonic automatic testing equipment is too little, and the measurement error of the defect size is also large, which is not conducive to the rational use of metal bars and the guarantee of safety.

[0004] Imaging detection of the quality status of bars and even quantitative non-destructive testing is the development trend of bar quality control technology. Ultrasonic synthetic aperture imaging technology (SAFT), which originated from synthetic aperture radar technology, has been further developed by many scholars in the fields of medical examination and industrial inspection since the 1970s due to its advantages such as high signal-to-noise ratio, azimuth resolution independent of detection depth, "near field" applicability, and combination with B-scan and C-scan imaging. Its technical content is very rich. However, most of them are based on 2D and 3D imaging in the linear scanning mode of Cartesian coordinate system. The research on synthetic aperture imaging methods for circular cross-sections of cylindrical bars began in the 1990s with medical ultrasound imaging research. O'Donnell et al. used a circular phased array sensor placed in the lumen of the coronary artery to propose a synthetic aperture method for circular cross-section catheters. In 2015, Wu Shiwei of Zhejiang University further developed this method for B-scan ultrasound imaging of cylindrical workpieces in Cartesian coordinate system for industrial inspection.

[0005] Domestic and foreign scholars have studied SAFT imaging of circular cross-sections or spiral scanning. In addition to improving imaging resolution and imaging inversion speed, they also need to study special synthetic aperture algorithms for cylindrical surface structures. There are three main ways to detect cylindrical surface structures: one is to use a one-dimensional linear phased array probe to detect cylindrical workpieces; the other is to use a water-immersion focused probe to perform spiral scanning relative to the workpiece, which is the mainstream of round bar industrial detection; and another is to use a phased array probe for spiral scanning, such as detecting large-diameter rollers. There are three main types of special synthetic aperture methods for curved surface structures: one is to use the fast marching method (FMM) for flight time correction; the second is to use the explicit approximate wave solution (CCWS) of the acoustic field relative to the curved surface structure or the curved surface vector method (UFVA) to transform the SAFT; the third is to perform phase delay in the frequency domain based on the wave equation derived in the cylindrical coordinate system.

[0006] However, there is no practical SAFT method that can promote the technological advancement of existing ultrasonic automated testing equipment for bars. Summary of the invention

[0007] The purpose of the present invention is to provide a synthetic aperture pSAFT focusing imaging algorithm in a polar coordinate system and a synthetic aperture imaging technology for ultrasonic spiral scanning and detection of metal round bars based on the algorithm, which can meet the online real-time detection requirements as much as possible while improving the detection resolution, and intuitively display information such as the defect size and distribution in the bar in the form of a tomographic scan similar to CT imaging, thereby providing richer and more accurate information for bar quality evaluation.

[0008] The technical solution of the present invention is as follows:

[0009] The SAFT imaging process of linear scanning of the probe in the classic Cartesian coordinate system is as follows: Figure 1As shown, for any point A(x, z) in the workpiece, it will be in the probe sound field of each detection position 1…i…n with an effective synthetic aperture length l (l=0.84λz / d, where λ is the wavelength, d is the probe wafer diameter, and z is the distance from the target position to the probe), and the signal at each position is S k (t), then the synthesized signal strength at this point is S SAFT (x, z), the delayed superposition formula of the SAFT algorithm is:

[0010]

[0011] Where i represents the probe position opposite to point A, τ ki represents the signal delay time of the kth probe position relative to the ith position ( Δx is the probe stepping distance, C is the sound velocity in the workpiece). Figure 1 It can be seen that the signal interval M of the defect at point A can be transformed into point M′ after SAFT time-delay superposition processing, thereby improving the imaging resolution.

[0012] However, when performing spiral scanning of a metal rod, the classic SAFT method needs to be modified. Based on this, the present invention provides a synthetic aperture pSAFT focusing imaging algorithm in a polar coordinate system, which is characterized by:

[0013] Assume that a water-immersion focusing probe with a diameter of D and a focal length of F is used to detect a round rod with a radius of R, so that the focus is located on the surface of the rod. The round bar is scanned spirally with the step angle.

[0014] like Figure 2 As shown in the figure, the sound beam in the workpiece can be regarded as the reverse expansion of the incident sound beam that narrows before the focus, so the focus of the probe can be set as the virtual aperture of the SAFT reconstruction process. Since the focus is located at the heterogeneous interface between the coupling agent and the sample, the refraction effect must be considered when calculating the aperture angle θ′ of the virtual source. The incident angle and semi-aperture angle of the sound beam are:

[0015]

[0016]

[0017] In the formula, C1 represents the sound velocity in water, and C2 represents the sound velocity in the sample;

[0018] When the target point is located at the center, it will be covered by the probe sound field at all scanning positions (such as Figure 2 a), therefore, the signals at all scanning angles will participate in SFAT reconstruction, that is, the effective synthetic aperture (SAR) is 2π; when the target point is located in the near-surface area below the rod surface, it will be Figure 2 The sound field coverage at each scanning angle on the near field side shown in b will also be Figure 2 c shows the acoustic field coverage at each scanning angle on the back side. The signal sets of these two cases are as follows: Figure 2 d and 2e. The dividing line between these two cases can be Figure 2 The dotted circle with radius r in a represents:

[0019]

[0020] Establish as Figure 3 In the polar coordinate system shown in FIG. 1 , when the target point A (ρ, α) is located on the near field side (r<ρ<R), its SAR is γ. Figure 3 The geometric relationship of a can be obtained:

[0021]

[0022] Target point A can also be Figure 2 c shows the back side being detected, according to Figure 3 The geometric relationship of b shows that the SAR for dorsal detection is γ′:

[0023]

[0024] Each scanning position The signal delay time τ of SAFT relative to the target point A(ρ, α) ki for:

[0025]

[0026] Therefore, the synthetic aperture (pSAFT) focusing imaging formula in the polar coordinate system is:

[0027]

[0028]

[0029] In the formula,

[0030] Based on this, the present invention also provides a synthetic aperture imaging technology (ST-SAFT) for ultrasonic spiral scanning detection of metal round bars, which is characterized by the following specific steps:

[0031] 1) After a spiral scan of the metal round bar, a data set is obtained The semi-aperture angle θ′ of the workpiece surface is calculated according to the probe parameters and formula (3); and the radius r of the inscribed circle that can be covered by the sound beam at all positions is calculated according to formula (4);

[0032] 2) For all points A(ρ, α) in the circular rod (where ρ∈0~R, α∈0~2π), use appropriate imaging resolution Δx (e.g., each pixel represents 0.1 mm) to The synthetic focusing calculation is performed according to the synthetic aperture (pSAFT) focusing imaging formula in the polar coordinate system. When ρ≤r, it is calculated according to formula (8-1). When ρ>r, it is calculated according to formulas (8-2) and (8-3). Thus, the data matrix Sp(i, j) is obtained, where i represents the radial count from 0 to R in Δx, and j represents the circumferential count from 0 to 2π in Count of;

[0033] 3) Transform the polar coordinate data matrix to display the circular cross-section tomography image. The coordinate transformation formula is:

[0034]

[0035] Note that when ρ is small, we can choose to multiply Calculations are performed to reduce the amount of calculations and increase the imaging speed. Based on the above ideas, the algorithm can be further improved to improve the imaging quality and speed.

[0036] The ultrasonic detection data is processed by ST-SAFT and imaged into a circular cross-section tomographic image; the image edge recognition method is then used to quantitatively evaluate the quantitative positioning resolution of this method for defect detection. The results show that the ST-SAFT aperture measurement value is equivalent to the actual value; the horizontal hole positioning is accurate; the imaging resolution is significantly better than the B-scan result; the imaging speed of each circular section can reach milliseconds, which can match the mechanical scanning speed of a bar inspection cycle. This technology can be used to improve the technical level of ultrasonic testing equipment for metal round bars and improve the safety guarantee capability of bar use.

[0037] The beneficial effects of the present invention are:

[0038] 1. The pSAFT imaging algorithm of the present invention transforms the classic SAFT imaging technology in a rectangular coordinate system into a time-delay superposition calculation in a polar coordinate system. The ST-SAFT imaging technology is a process of processing the collected data by using pSAFT and imaging it into a circular cross section. The method of the present invention is particularly suitable for ultrasonic testing of metal round bars.

[0039] 2. The pSAFT imaging algorithm of the present invention has the same advantages as SAFT in Cartesian coordinate system: high signal-to-noise ratio, azimuth resolution is independent of detection depth, etc. The angular resolution of the experiment can reach its theoretical value.

[0040] 3. The tomographic scanning imaging method of the ST-SAFT of the present invention is superior to the B-scan imaging method and can more intuitively and accurately reflect the size, location and distribution of defects, so as to carry out quantitative, localized and qualitative evaluation of defects, which can be used to accurately evaluate defects and help improve the production process of materials.

[0041] 4. The imaging speed of a circular cross section of the ST-SAFT of the present invention can match the spiral mechanical scanning speed of the rod, which can meet the needs of fast real-time detection.

[0042] 5. The scanning method of the probe in the prior art is: the probe moves axially - the rod rotates a certain angle - the probe moves axially again. In industry, automated testing equipment is usually used to perform spiral scanning on the rod, and online real-time imaging is required as much as possible. The ST-SAFT technology provided by the present invention can meet the requirements of spiral scanning for imaging of cross sections, and can also perform axial and three-dimensional imaging of the data processed by pSAFT.

[0043] 6. ST-SAFT can be used on the basis of existing bar automatic inspection equipment without adding additional hardware expenses except for dedicated imaging computers. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 Schematic diagram of SAFT imaging in Cartesian coordinate system, including (a) plane linear scan, (b) data set, and (c) SAFT.

[0045] Figure 2 Schematic diagram of pSAFT imaging, including (a) circumferential scan, (b) near-field side scan, (c) back-side scan, (d) defect B data set, (e) defect A data set, and (f) circular cross section.

[0046] Figure 3 Schematic diagram of SAR calculation, where (a) is the near-field side and (b) is the back side.

[0047] Figure 4 Schematic diagram of artificial defect specimen, including (a) physical image and (b) coordinate diagram.

[0048] Figure 5 Imaging of the scanning results of a certain week, including (a) B-scan image and (b) original data cross-sectional image.

[0049] Figure 6 Imaging images after ST-SAFT processing, including (a) circumferential extension of the radial depth position of the defect, and (b) cross-sectional tomogram after ST-SAFT processing.

[0050] Figure 7 Resolution curves after ST-SAFT processing, where (a) before ST-SAFT, (b) after ST-SAFT.

[0051] Figure 8 Magnified image after ST-SAFT processing.

[0052] Fig. 9Initial serial imaging after ST-SAFT processing, including (a) 3-D image and (b) axial spatial slices.

[0053] Fig.10 The final continuous imaging after ST-SAFT processing, including (a) columnar image and (b) axial spatial cross-section. DETAILED DESCRIPTION

[0054] The ultrasonic imaging experimental equipment consists of mechanical equipment and ultrasonic detection electronic equipment. In the three-axis sliding module used in the experiment, the Y-axis motor is used to adjust the probe to align with the generatrix of the round rod, the Z-axis motor is used to adjust the focal position of the probe relative to the round rod, the X-axis motor moves the probe along the axis of the round rod, and the U-axis motor drives the round rod to rotate and move, thereby realizing the surrounding spiral ultrasonic scanning and detection of the round rod.

[0055] The diameter of the experimental round steel rod is 65mm, the height is 35mm, and the sound velocity is 5900m / s. The experiment uses a 5MHz water immersion focusing probe with a chip diameter of 13mm and a focal length of 55mm. The probe sound beam is incident perpendicular to the surface of the sample to be tested, and the water distance is 55mm. The sampling frequency is 50MHz. The axial scan is 20mm, the pitch is 1mm, and the circumferential step angle is 1°.

[0056] The round steel artificial defect specimen is designed with 11 transverse holes with a hole depth of 25 mm. The actual picture of the specimen and the position distribution of each transverse hole on its end surface are shown in the figure below. Figure 4 a and 4b and Table 1. Among them, 1~2# and 10~11#φ2.0mm transverse holes are all located within the r zone, 1~2# are used to examine the circumferential resolution within the r zone; 3~5# are used to examine the resolution of different circumferential spacings; 6~8# are used to examine the resolution of defects of different sizes; 9~11# are used to examine the radial resolution across the r zone.

[0057] ST-SAFT imaging:

[0058] 1) After a spiral scan of the metal round bar, a data set is obtained The semi-aperture angle θ′ of the workpiece surface is calculated according to the probe parameters and formula (3); and the radius r of the inscribed circle that can be covered by the sound beam at all positions is calculated according to formula (4);

[0059] 2) For all points A(ρ, α) in the circular rod (where ρ∈0~R, α∈0~2π), use appropriate imaging resolution Δx (e.g., each pixel represents 0.1 mm) to The synthetic focusing calculation is performed according to the synthetic aperture (pSAFT) focusing imaging formula in the polar coordinate system. When ρ≤r, it is calculated according to formula (8-1). When ρ>r, it is calculated according to formulas (8-2) and (8-3). Thus, the data matrix Sp(i, j) is obtained, where i represents the radial count from 0 to R in Δx, and j represents the circumferential count from 0 to 2π in Count of;

[0060] 3) Transform the polar coordinate data matrix to display the circular cross-section tomography image. The coordinate transformation formula is:

[0061] Test results and analysis

[0062] (1) Comparison of images before and after ST-SAFT processing

[0063] The ultrasonic echo signals at each scanning angle position obtained in one round of detection are sampled and recorded, and then the raw data is subjected to B-scan imaging (such as Figure 5 a), by converting the sampling time into the depth inside the sample and performing a simple coordinate transformation, the scanning image of this section can be obtained (the result is shown in Figure 5 b). Figure 5 It can be seen from a that defects 1 to 2#, 3 to 5# and 6 to 8# are difficult to distinguish. Figure 5 b can distinguish 1~2#, which shows that different imaging methods have different abilities to distinguish defects. The cross-sectional imaging method is better, but it still cannot reflect the true distribution of defects without SAFT processing.

[0064] After the original data is processed according to the aforementioned ST-SAFT method, the circumferential development diagram and cross-sectional imaging diagram of the radial depth position of the defect are shown as follows: Figure 6 a and 6b.

[0065] contrast Figure 5 and Figure 6 b, it can be seen that the cross-sectional tomography after ST-SAFT processing can well and intuitively reflect the defect condition.

[0066] (2) Defect resolution capability of ST-SAFT imaging

[0067] Will Figure 5 The data of a is gated at a sampling depth of 150 to 600 pts, and the maximum amplitude value within the gate of each circumferential scanning position signal is read to obtain the circumferential distribution curve of the defect (such as Figure 7 As shown in a), since the 6dB resolution rule is not met, only 4 defects out of 11 can be identified, which will exaggerate the size of the defects in actual detection. Figure 6 The circumferential distribution curve of defects is obtained by data processing of a. Figure 7 As shown in b, 11 defects can be completely resolved. Figure 7 b shows that the image signal-to-noise ratio is as high as 26dB. Figure 6 Zoom in to identify defect edges Figure 8The comparison between the defect size and the true value is shown in Table 1, which shows that when measured with a pixel length of 0.2mm, the defect size measurement error is within two pixel length units (≈0.4mm); the polar diameter ρ measurement error is 0.8~2.1mm, and then corrected with the artificial defect radius, the radial positioning error of the defect is about 1mm; the polar angle α measurement error is 1°. Therefore, the defect quantitative positioning capability of ST-SAFT can meet the needs of engineering inspection.

[0068] Table 1 Artificial defect location and size measurement

[0069]

[0070] Note: φ-horizontal hole diameter (mm), ρ-polar radius (mm), α-polar angle (°)

[0071] Depend on Figure 7 b According to the peak drop 6dB method, the circumferential resolution of 1~2# in r zone is 2.43°; the resolution of 3~5# with different circumferential spacing is 2.09°, which is basically consistent with the theoretical value of azimuth resolution δφ≈D / 2R≈0.02rad (where D is the focal spot diameter and R is the sample radius); the resolution of 6~8# defects of different sizes is 3.26°. Figure 8 The signal distribution curve is drawn along the radial direction according to the -6dB method, and the radial resolution of the 9-11# span r zone is 0.58mm, which is smaller than the size of the artificial defect. It can be seen that its azimuth resolution is independent of depth and other characteristics are consistent with the advantages of the classic SAFT.

[0072] (3) ST-SAFT imaging speed

[0073] The CPU is 2.6 GHz. Core TM On an i7 laptop, using MATLAB programming, it takes about 250ms to perform ST-SAFT imaging of a cross section after a full scan, which means that the ST-SAFT imaging speed can match the mechanical scanning speed of 4 scans per second. If machine language (such as VC) is used, the imaging speed can be doubled. If special instruments such as FPGA and DSP are used, the imaging speed is even faster, which can adapt to the fast real-time detection speed of mechanical scanning of 10 scans per second.

[0074] The imaging mode of ST-SAFT during continuous detection can be as follows Fig. 9 360° unfolded three-dimensional imaging and slice imaging can also be Fig.10 The three-dimensional columnar imaging and the cross-sectional imaging of the continuous video method can be used to comprehensively evaluate the three-dimensional distribution information of the internal defects of the sample.

[0075] In the experiment, φ1.0mm transverse holes (roughly corresponding to the highest quality grade of AAA) can be reliably detected before and after SAFT. Although the benefits of ST-SAFT in improving sensitivity are not shown, this is because the round steel material used in this embodiment has low background noise, low interfering random noise and strong artificial defect signals. However, SAFT synthesizes multiple signals near each target point during imaging (see formula (8)). It has the function of average filtering to suppress noise, which can improve the signal-to-noise ratio and detection sensitivity. Therefore, compared with conventional bar ultrasonic flaw detection technology, ST-SAFT can not only provide more abundant information such as defect quantification, location and distribution, but also detect small defects with high signal-to-noise ratio. Some bars that were originally difficult to detect (such as coarse-grained materials and powder metallurgy materials) can be solved by trying to use SAFT.

[0076] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable people familiar with the technology to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made according to the spirit of the present invention should be included in the protection scope of the present invention.

[0077] Additionally, descriptions of well-known structures and techniques are omitted herein to avoid unnecessarily obscuring the concepts of the present invention.

Claims

1. A synthetic aperture pSAFT focusing imaging method in a polar coordinate system, characterized in that: Use a water-immersion focusing probe with a diameter of D and a focal length of F to detect a metal round bar with a radius of R, so that the focus is located on the surface of the round bar. The rod is spirally scanned with a step angle, and the focus of the probe is set as the virtual aperture of the SAFT reconstruction process. The acoustic beam incident angle and the semi-aperture angle of the virtual source are: Where, C1 represents the sound velocity in water, C2 represents the sound velocity in the sample, θ represents the aperture angle of the focused sound beam, and θ' represents the aperture angle of the virtual source; When the target point is located at the center, it will be covered by the acoustic field of the probes at all scanning positions, and the signals at all scanning angles will participate in SFAT reconstruction, that is, the effective synthetic aperture radian SAR is 2π; When the target point is located in the near-surface area below the rod surface, it will be covered by the sound field under each scanning angle on the near-field side, and will also be covered by the sound field under each scanning angle on the back side. The dividing line between these two situations is a dotted circle with a radius of r, that is: Establish a polar coordinate system. When the target point A(ρ,α) is located on the near field side, its SAR is γ: When the target point A(ρ,α) is on the back side, its SAR is γ′: Each scanning position The signal delay time τ of SAFT relative to the target point A(ρ,α) ki for: The synthetic aperture pSAFT focusing imaging formula in the polar coordinate system is: Where: in Indicates the step angle of spiral scanning of the round bar; S k (t) represents each position signal.

2. The synthetic aperture pSAFT focusing imaging method in the polar coordinate system according to claim 1, characterized in that: The specific steps are as follows: 1) After a spiral scan of the metal round bar, a data set is obtained The semi-aperture angle θ′ of the workpiece surface is calculated according to the probe parameters and formula (I); and the radius r of the inscribed circle that can be covered by the sound beam at all positions is calculated according to formula (II); In the formula, C1 represents the sound velocity in water, C2 represents the sound velocity in the sample, θ represents the aperture angle of the focused sound beam, θ′ represents the aperture angle of the virtual source, and R represents the detection radius of the water immersion focused probe; 2) For all points A(ρ,α) in the circular rod, use imaging resolution Δx to The synthetic focusing calculation is performed according to formula (III), where ρ∈0~R, α∈0~2π, thereby obtaining the data matrix Sp(i,j), where i represents the count from 0 to R in radial direction with Δx, and j represents the count from 0 to 2π in circumferential direction with Count of; In the formula, in Indicates the step angle of spiral scanning of the round bar; S k (t) represents each position signal; signal delay time 3) Transform the polar coordinate data matrix to display the circular cross-section tomography image. The coordinate transformation formula is:

3. The synthetic aperture pSAFT focusing imaging method in the polar coordinate system according to claim 2, characterized in that: In step 2), the imaging resolution Δx is such that each pixel represents 0.1 mm.

4. The synthetic aperture pSAFT focusing imaging method in the polar coordinate system according to claim 2, characterized in that: The probe used for detection is a point-focusing or line-focusing probe.

Images