Vision-based phased array probe positioning method, system and non-destructive testing method

Abstract

The present application relates to a kind of based on vision's phased array probe positioning method, system and nondestructive testing method in the field of nondestructive testing technology.Phased array probe positioning method is by 3D scanning to the surface of detection object and obtains three-dimensional point cloud data, with the structure size in standard database is matched to extract standard parameter, constructs the three-dimensional model of detection object and generates the two-dimensional section of different positions;Phased array probe is positioned in scanning probe position in real time, while phased array ultrasonic imaging is overlaid and displayed the two-dimensional section of current position.The present application can automatically generate complex structure surface model on one hand, on the other hand can record probe position while saving atlas data, and real-time model two-dimensional section is overlaid on atlas, can effectively assist detection personnel to carry out defect determination and atlas record, so that when scanning special structure parts, defect positioning is more accurate.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of nondestructive testing, in particular to a phased array probe positioning method and system based on vision and a nondestructive testing method. BACKGROUND

[0002] When performing nondestructive testing by using a phased array ultrasonic, its recordability is a great advantage, which is convenient for post-audit and defect position judgment. With the development of pressure-bearing equipment and industrial components towards high parameters, integration and modularization, higher requirements for the applicability of nondestructive testing are put forward for their complex structures. However, when performing phased array ultrasonic testing on special structures such as pipe support type fillet welds, heat exchanger tube sheet joints and complex structure forgings, on the one hand, the existing probe (scanner) cannot be used due to its difficulty in adapting to the complex and variable surface profile. Moreover, with the change of the probe detection position, the shape and distance of the corresponding main pipe inner wall also change. Due to the characteristics of the welding groove of the welded support pipe, for example, the weld area is smallest at the 90° defect position and largest at the 270° defect position, which makes the ultrasonic detection cross section different at different positions, resulting in the continuous change of the structure amplitude signal position, which is easy to cause missed judgment and misjudgment of defects. On the other hand, the modeling function of the existing detection system cannot be adjusted in real time for the structure. The existing probe positioning method for such special structures usually only uses manual scanning and time axis recording method to save atlas data, which makes it difficult for the phased array ultrasonic scanning atlas to correspond to the actual position of the detection object, i.e. it is impossible to determine whether the detected high echo signal belongs to structure noise or defect signal. This causes the detection and evaluation problems of the buried defects of the related structure, which may lead to major accidents due to missed judgment and misjudgment. SUMMARY

[0003] In order to solve the technical problem of inaccurate defect positioning of the existing probe positioning method for special structure parts, the present application provides a phased array probe positioning method and system based on vision and a nondestructive testing method.

[0004] In the first aspect, the present application provides a phased array probe positioning method based on vision, which is used to position the actual position of the phased array probe on the detection object. The phased array probe positioning method comprises:

[0005] Scanning the detection object to obtain three-dimensional data. The structure size parameters of the detection object are calculated through the three-dimensional data, and a three-dimensional model of the detection object is generated according to the structure size parameters. The probe position at the moment is collected by a vision sensor t Three positioning points for determining a detection surface are set on the probe P RST R , S , T ​Display the model on a 3D model with one of the positioning points as the origin. t The probe position at any given time, and with P RST The plane that coincides with the 3D model is used as t A two-dimensional cross-sectional view at a given moment. Generation of objects scanned and detected by a phased array probe. t Phased array imaging at any given time. t Two-dimensional cross-sectional view at time and t The phased array images at different times are superimposed to form a spectrum.

[0006] Secondly, a vision-based phased array ultrasonic non-destructive testing method is proposed, which uses the vision-based phased array probe positioning method in the first aspect to locate and scan the phased array probe in order to perform non-destructive testing on the object to be tested.

[0007] Thirdly, a vision-based phased array probe positioning system is proposed, which uses the vision-based phased array probe positioning method described in the first aspect. This phased array probe positioning system includes: a scanning module, a model generation module, a two-dimensional cross-section generation module, and a positioning module.

[0008] The scanning module is used to scan the object to be detected and obtain 3D data. The model generation module is used to calculate the structural dimensional parameters of the object from the 3D data and generate a 3D model of the object based on these parameters. The 2D section generation module is used to acquire data through a vision sensor. t Probe position at any given time: Three points are set on the probe to determine one detection surface. P RST Location point R , S , T Display on the 3D model with one of the positioning points as the origin. t The probe position at any given time, and with P RST Generation of the cross plane between the plane and the 3D model t A two-dimensional cross-sectional view at a given moment. The positioning module is used to generate the image by scanning the target object with a phased array probe. t Phased array imaging at any given moment; t Two-dimensional cross-sectional view at time and t The phased array images at different times are superimposed to form a spectrum.

[0009] Fourthly, a computer-readable storage medium is also proposed, which stores a computer program that, when executed by a processor, implements the steps of the vision-based phased array probe positioning method in the first aspect.

[0010] The beneficial effects of this invention are as follows:

[0011] 1. This invention employs a visual positioning method to track and record the probe's position and orientation during phased array probe scanning. The corresponding two-dimensional cross-sectional images are then overlaid on the imaging frame of the inspection software, forming a visual atlas in conjunction with phased array imaging. This vision-based probe positioning method automatically generates three-dimensional models of complex surface structures and records the probe position while saving the atlas data. It also overlays the two-dimensional cross-section of the model onto the atlas in real time, effectively assisting inspectors in defect identification and atlas recording, thus making defect location more accurate when scanning parts with special structures.

[0012] 2. This invention can automatically generate three-dimensional models of complex surface structures, and can also perform adaptive filtering based on the structural complexity, surface roughness, and maximum size of the object being detected, thereby making the constructed three-dimensional model more accurate. Attached Figure Description

[0013] Figure 1 This is a flowchart of a vision-based phased array probe positioning method;

[0014] Figure 2 This is a schematic diagram of the probe and the object being detected;

[0015] Figure 3 It is a map obtained by scanning a fillet weld structure of a pipe using a vision-based phased array probe positioning method. Detailed Implementation

[0016] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] It should be noted that when a component is said to be "installed on" another component, it can be directly on the other component or it may be in a component that is centered on it. When a component is said to be "set on" another component, it can be directly set on the other component or it may also be in a component that is centered on it. When a component is said to be "fixed to" another component, it can be directly fixed to the other component or it may also be in a component that is centered on it.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the specification of this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "or / and" as used herein includes any and all combinations of one or more of the associated listed items.

[0019] This embodiment provides a vision-based phased array probe positioning method, which is used to locate the actual position of the phased array probe on the object being detected. Please refer to... Figure 1 The phased array probe positioning method includes the following steps:

[0020] S1. First, a laser scanner is used to perform a 3D scan of the surface to be inspected to obtain the desired surface characteristics. I Points p 1. p 2、…、 p i … p I Point cloud data P . P It can be represented in three-dimensional coordinates. Then, a filtering function is used. f ( p i )right P The three-dimensional data is obtained after filtering. This filtering function... f ( p i )for:

[0021] .

[0022] .

[0023] in, f ( p i The point where )=0 represents P The discrete points to be filtered out. p i express P The first in i One point. R L ( p i )express p i The adaptive radius. n ( p i )express R L ( p i The number of neighboring points within the range of ).N min ( p i )express p i The minimum number of neighboring points. At least ensure p i It has 3 adjacent points. express p i The k Average distance of each neighborhood. p j express P Except p i The outside j One point. d G express P The global neighborhood distance. N k ( i )express p i The k Each neighborhood. median express median function. α This represents the basic scale factor, which controls the range of outliers that will be removed as a global sensitivity factor. β This represents the density factor, which is used to control the sensitivity of the radius to local density changes, thus protecting the effective points in sparse regions. gamma This represents the density scaling factor, used to convert relative radius ratios to absolute values. It's worth noting that... α , β , gamma It can adaptively select values ​​based on the structural complexity, surface roughness, and maximum size of the object being detected, resulting in more accurate 3D data. Specifically, in this embodiment... α , β , gamma Methods for adaptive value selection include:

[0024] S11. Determine the type of object to be inspected. If the object to be inspected is a small-sized complex structure, such as a small-diameter pipe seat fillet weld, then... α Use a value between 2.0 and 3.5. If the object being tested is a large structure, such as a pipe or a forging with small curvature, then... α The range is 3.6 to 6.0. Less than 20 probe wavelengths are considered small-size, and 20 or more probe wavelengths are considered large-size (the wavelength of a 5MHz probe in steel is approximately 1mm). Objects with varying cross-sections, varying wall thicknesses, or corner joints are considered complex structures.

[0025] S12, Judgment PUniformity: If the average surface roughness of the object being tested is... Ra If the roughness threshold is less than 1.6 μm in this embodiment, then it is determined that... P Uniform, and β Take a value of 0.1 to 0.3. If... Ra If the roughness threshold is ≥, then it is determined that... P Uneven, and β Take a value of 0.5 to 1.0.

[0026] S13, Calculation P global average density rho Simultaneously, a joint judgment is made based on the dimensions of the object being inspected. If the maximum dimension is <50mm, rho >100 points / cubic centimeter, then gamma Take 0.3 to 0.5. If 50mm ≤ maximum size < 500mm, 100 points / cubic centimeter ≥ rho >10 points / cubic centimeter, then gamma Take a value of 0.1 to 0.3. If the maximum dimension is ≥500mm, rho ≤10 points / cubic centimeter, then gamma Take a value of 0.05 to 0.15.

[0027] at this time α , β , gamma The characteristics of the object to be detected are determined, and the adaptive value is substituted into the filtering function. f ( p i )right P Filtering is performed to obtain the final required 3D data.

[0028] S2. Calculate the structural dimensional parameters of the object to be inspected using 3D data. Match the structural dimensional parameters to a standard database. Upon successful matching, a 3D model of the object is obtained. This standard database is pre-built and includes: manufacturing standards corresponding to the type of object being inspected (e.g., pipe fittings, butt welds, fillet welds, pipe seat fillet welds, forgings, etc.). These parameters are represented as feature vectors. V p,1 , V p,2 … V p,m … V p,M The structural dimensions calculated from the 3D data are stored in a standard database. During the matching process, the structural dimension parameters are converted into feature vectors. V q Euclidean distance is used for calculation. V q Feature vectors of all standard objects in the standard databaseV p,1 , V p,2 … V p,m … V p,M similarity D 1. D 2、…、 D m … D M The calculation formula is:

[0029] .

[0030] like D m <Matching threshold D max Then the corresponding V p,m Successfully matched the detected object. Extract the corresponding [object]. V p,m The complete parameter set is used to call a predefined parametric modeling script with a CAD kernel, based on its type, to generate a 3D model of the object being inspected. D 1. D 2、…、 D m … D M All greater than or equal to D max If the match fails, the matching will fail. For matching failures, and for parameters not found in the standard database such as surface curvature, branch pipe bevel length, bevel angle, weld height, and weld width, a 3D model of the object to be inspected can be directly generated based on the average value of the structural dimension parameters in the 3D data.

[0031] S3. The phased array probe scanning process includes: a computer equipped with detection software sends commands to the controller for signal transmission and reception. The probe, using a coupling agent, tightly adheres to the surface of the object being detected. It receives electrical signals from the controller, converts them into ultrasonic waves, and transmits them into the object. Echo signals caused by discontinuities in the object's internal structure are received by the probe, converted into electrical signals, and transmitted back to the controller. The controller converts the electrical signals into digital signals of a specific format and transmits them to the computer. The computer processes the acquired data and displays images, such as those obtained through sector scanning or full-focus imaging, on the screen. One innovation of this invention is the additional use of a visual sensor for data acquisition. t The probe's position at any given time. A visual sensor, such as a camera, determines the current position of the probe. Specifically, such as... Figure 2As shown, in this embodiment, the probe is configured as a pentagonal prism, with one side serving as the scanning surface. Using the scanning surface as the bottom surface, the center point of the top of the probe's front end is set as the positioning point. R Set the center point of the bottom of the probe tip as the positioning point. S Set the center point of the bottom rear end of the probe as the positioning point. T These three positioning points R , S , T Forming a uniquely defined detection surface P RST Using the positioning point S The origin is used to match the coordinate system of the 3D model and is displayed on the 3D model. t The probe's position within the object being inspected can be determined by its position at a given time. P RST The plane that coincides with the 3D model (i.e.) P RST (The cut surface of the 3D model) as t A two-dimensional cross-sectional view at a given time.

[0032] S4. Phased array probe scan detection object generation t Phased array imaging at any given time. t Two-dimensional cross-sectional view at time and t The phased array images are superimposed at specific moments to form the ultrasonic spectrum after probe positioning. Specifically, the phased array probe scan uses a time-axis recording method. The probe moves at a constant speed, and the camera synchronously records the probe position. The probe position, the video recording from the vision sensor, and the phased array imaging information are saved in real time in the computer, achieving accurate recording of the spectrum. When a report needs to be generated or reviewed, the phased array imaging and probe position information can be directly accessed. At this time, the screen will display the phased array imaging with a superimposed two-dimensional cross-sectional view and the camera video recording during the scan, which facilitates accurate determination of the location, length, and other information of defects in the inspected object.

[0033] The vision-based phased array probe positioning method of this invention is used to detect and image a fillet weld structure of a pipe fitting. A three-dimensional model is generated based on the scanning results and parameters such as structural dimensions in the pipe fitting forging standard (19326-2012). Then, a two-dimensional cross-section of the structure is generated according to the probe position. Figure 3As shown, the two-dimensional cross-section is represented by a solid-line frame, where the red dashed frame represents the weld area, overlaid with a phased array full-focus image. The reddish areas in the full-focus image indicate strong acoustic signatures. Two high-signature areas, A and B, exist in the image. A is located near coordinates (27, 14), and B is located near coordinates (34, 12). According to the cross-sectional diagram, area A is located inside the weld area and can be considered a weld defect. Area B is outside the weld and follows the contour of the inner wall of the main pipe; therefore, it is considered a structural echo rather than a defect. When the structure is unclear and the weld location is uncertain, the two high-signature areas are easily mistaken for defects, leading to misjudgment. By adding the probe's detection of the cross-sectional shape at the corresponding location to the imaging results, the structural amplitude signal and defect signal can be distinguished. This allows the vision-based phased array probe positioning method used in this invention to accurately locate defect signatures during imaging and distinguish between defect signatures and structural echo signatures.

[0034] In another embodiment, a vision-based phased array ultrasonic nondestructive testing method is proposed, which uses the vision-based phased array probe positioning method as described in the above embodiment to locate and scan the phased array probe in order to perform nondestructive testing on the object to be tested.

[0035] In another embodiment, a vision-based phased array probe positioning system is proposed, which uses the vision-based phased array probe positioning method described in the above embodiment. This phased array probe positioning system includes: a scanning module, a model generation module, a two-dimensional cross-section generation module, and a positioning module.

[0036] The scanning module is used to scan the object to be detected and obtain 3D data. The model generation module is used to calculate the structural dimension parameters of the object to be detected using the 3D data; it then matches these parameters against a standard database, and upon successful matching, obtains a 3D model of the object. The 2D section generation module is used to acquire data via a vision sensor. t Probe position at any given time: Three points are set on the probe to determine one detection surface. P RST Location point R , S , T Display on the 3D model with one of the positioning points as the origin. t The probe position at any given time, and with P RST Generation of the cross plane between the plane and the 3D model t A two-dimensional cross-sectional view at a given moment. The positioning module is used to generate this view during phased array probe scanning. t Phased array imaging at any given moment; t Two-dimensional cross-sectional view at time and t The phased array images at different times are superimposed to form a spectrum.

[0037] In another embodiment, a computer-readable storage medium is also proposed. This computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the vision-based phased array probe positioning method as described in the above embodiment.

[0038] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0039] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A vision-based phased array probe positioning method, used to locate the actual position of the phased array probe on the object being detected, characterized in that, It includes: Point cloud data is obtained by 3D scanning of the inspection surface using a laser scanner. P ; Through the filtering function f ( p i )right P The three-dimensional data is obtained after filtering; f ( p i The expression is: ， ， In the formula, f ( p i The point where )=0 represents P The discrete points to be filtered out; p i express P The first in i One point; R L ( p i )express p i The adaptive radius; n ( p i )express R L ( p i The number of neighboring points within the range; N min ( p i )express p i The minimum number of neighborhood points; representing p i The k Average distance of each neighborhood; p j express P Except p i The outside j One point; d G express P The global neighborhood distance; N k ( i )express p i The k Each neighborhood; median express median function; α Indicates the basic scale factor. β Density factor γ Represents the density scaling factor; where, α , β , γ The methods for obtaining values ​​include: Determine the type of the object being detected; if the object is a small-sized, complex structure, then α Use a value between 2.0 and 3.5; if the object being tested is a large structure, then... α The range is 3.6 to 6.0; among them, less than 20 probe wavelengths are considered small size, and greater than or equal to 20 probe wavelengths are considered large size; the object being tested has a variable cross-section, variable wall thickness, or corner joints, which are considered complex structures; judge P Uniformity: If the average surface roughness of the object being tested is... Ra If the roughness threshold is less than 1, then it is determined that... P Uniform, and β Take 0.1 to 0.3; if Ra If the roughness threshold is ≥, then it is determined that... P Uneven, and β Take a value of 0.5 to 1.0; calculate P global average density ρ If the maximum size is <50mm, ρ >100 points / cubic centimeter, then γ Take 0.3 to 0.5; if 50mm ≤ maximum size < 500mm, 100 points / cubic centimeter ≥ ρ >10 points / cubic centimeter, then γ Take a value of 0.1 to 0.3; if the maximum size is ≥500mm, ρ ≤10 points / cubic centimeter, then γ Take a value of 0.05 to 0.15; The structural dimension parameters of the object being inspected are calculated using 3D data, and a 3D model of the object is generated based on these parameters. Collected by a vision sensor t Probe position at any given time: Three points are set on the probe to determine one detection surface. P RST Location point R , S , T ; Display on the 3D model with one of the positioning points as the origin. t The probe position at any given time, and with P RST The plane that coincides with the 3D model is used as t Two-dimensional cross-sectional view at time; Phased array probe scanning and detection object generation t Phased array imaging at any given moment; Will t Two-dimensional cross-sectional view at time and t The phased array images at different times are superimposed to form a spectrum.

2. The vision-based phased array probe positioning method according to claim 1, characterized in that, Methods for generating 3D models of the objects being inspected include: Matching based on structural size parameters in a standard database: converting the structural size parameters of the detected object into a feature vector. V q ; Euclidean distance calculation V q Feature vectors of all standard objects in the standard database V p,1 , V p,2 … V p,m … V p,M similarity D 1. D 2、…、 D m … D M ; like D m <Matching threshold D max Then the corresponding V p,m Matched successfully with the detected object; extract the corresponding... V p,m The complete set of parameters is used to call a predefined parametric modeling script with a CAD kernel based on its type, and generate a 3D model of the object to be detected. like D 1. D 2、…、 D m … D M All greater than or equal to D max If the matching fails, a 3D model of the detected object is directly generated based on the mean value of the structural dimension parameters in the 3D data.

3. The vision-based phased array probe positioning method according to claim 1, characterized in that, The probe is designed in the shape of a pentagonal prism, with one side serving as the scanning surface. Using the scanning surface as the bottom surface, the center point of the top of the probe tip is set as the positioning point. R Set the center point of the bottom of the probe tip as the positioning point. S Set the center point of the bottom rear end of the probe as the positioning point. T Location point R , S , T Forming the detection surface P RST .

4. The vision-based phased array probe positioning method according to claim 1, characterized in that, Vision-based phased array probe localization methods also include: During phased array probe scanning, the probe position, video recording from the vision sensor, and phased array imaging are recorded simultaneously using a time-axis recording method.

5. A vision-based phased array ultrasonic nondestructive testing method, characterized in that, It uses the vision-based phased array probe positioning method as described in any one of claims 1 to 4 to locate and scan the phased array probe in order to perform non-destructive testing on the object to be tested.

6. A vision-based phased array probe positioning system, characterized in that, It uses the vision-based phased array probe positioning method as described in any one of claims 1 to 4; It includes: The scanning module is used to scan and detect objects to obtain three-dimensional data. The model generation module is used to calculate the structural dimension parameters of the object being inspected using 3D data, and to generate a 3D model of the object based on the structural dimension parameters. A two-dimensional section generation module is used to acquire data via a vision sensor. t Probe position at any given time: Three points are set on the probe to determine one detection surface. P RST Location point R , S , T Display on the 3D model with one of the positioning points as the origin. t The probe position at any given time, and with P RST Generation of the cross plane between the plane and the 3D model t Two-dimensional cross-sectional view at time; The positioning module is used to generate a detection object by scanning the phased array probe. t Phased array imaging at any given time; t Two-dimensional cross-sectional view at time and t The phased array images at different times are superimposed to form a spectrum.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the vision-based phased array probe positioning method as described in any one of claims 1 to 4.

Images