Defect detection method and device for tritium-related component and medium

Abstract

The invention relates to the field of tritium-related component detection, and discloses a tritium-related component defect detection method and device and a medium, and the method comprises the following steps: filling a substitute gas into a tritium-related component; acquiring a plurality of frames of weld seam images of the tritium-related part at a certain point where the tritium-related part is detected; performing difference feature analysis on the adjacent frames of weld seam images to obtain a virtual vector matrix; and according to each frame of welding seam image and the virtual vector matrix, determining the defect of an area corresponding to the welding seam image. According to the invention, accurate, rapid and non-contact defect detection of the tritium-related part can be realized.

Description

Technical Field

[0001] This invention relates to the field of tritium-related component inspection, and more particularly to a method, apparatus, and medium for defect detection of tritium-related components. Background Technology

[0002] In fusion power plants, the reactor core, tritium plant, and critical components interact via a gaseous tritium loop. Given tritium's permeability in metallic materials, large-scale and complex tritium handling processes place extremely high demands on the sealing of tritium-related components. Therefore, rigorous weld quality inspections must be implemented during the manufacturing process of tritium-related components to minimize the risk of tritium aerosol leakage. Existing inspection methods for tritium-related components still have room for improvement in terms of accuracy and time consumption for weld defect detection, and overall suffer from insufficient accuracy and high processing time. Summary of the Invention

[0003] The present invention aims to at least solve one of the technical problems existing in the prior art. To this end, the present invention proposes a defect detection method for tritium-contact components, which can achieve accurate, rapid and non-contact defect detection for tritium-contact components.

[0004] The present invention also proposes an apparatus and a medium for detecting defects in the above-mentioned tritium-related components.

[0005] A defect detection method for a tritium-contaminated component according to a first aspect of the present invention, applied to a tritium-contaminated component, comprising: The tritium-contaminated component is filled with a substitute gas; Several frames of weld images of the tritium-contaminated component are acquired at a certain point in the component being inspected. By performing differential feature analysis on adjacent weld seam images, a virtual vector matrix is ​​obtained; Based on each frame of weld seam image and the virtual vector matrix, the defects in the region corresponding to the weld seam image are determined.

[0006] According to an embodiment of the present invention, a defect detection method for tritium-contaminated components has at least the following beneficial effects: The present invention utilizes a weld background image, i.e., a weld image, to obtain a virtual vector matrix representing the differences between adjacent frames, and then determines the defects in the corresponding area of ​​the weld image through each frame of weld image and the virtual vector matrix; The present invention realizes natural background schlieren measurement with the weld pattern as the background, and determines the defects of tritium-contaminated components by capturing the airflow differences between adjacent frames, thereby enabling accurate, fast and non-contact defect detection of tritium-contaminated components.

[0007] According to some embodiments of the present invention, determining the defects in the region corresponding to the weld image based on each frame of weld image and the virtual vector matrix includes: The set of moduli of each virtual vector matrix is ​​used to obtain the dimensionless matrix of the moduli. Determine the average weld image based on each frame of weld images; The dimensionless matrix of the modulus and the mean image of the weld are multiplied to obtain a background schlieren composite image. The presence of a through-hole defect in the area corresponding to the weld image is determined based on the background schlieren composite image.

[0008] According to some embodiments of the present invention, determining whether a through-hole defect exists in the region corresponding to the weld image based on the background schlieren composite image includes: If the modulus of the background schlieren composite image is less than a preset first threshold, then it is determined that there is no through defect in the area corresponding to the weld image. If the modulus of the background schlieren composite image is within a preset first interval, then it is determined that there is a through defect in the area corresponding to the weld image; wherein, the left endpoint of the first interval is greater than the first threshold.

[0009] According to some embodiments of the present invention, the method further includes: Obtain the original weld image; The original weld image, the background schlieren composite image determined to have no through-hole defects, and the background schlieren composite image determined to have through-hole defects are used as the training set and the validation set to train the graph convolutional neural model; wherein, the graph convolutional neural model outputs the defect type of the tritium-related component, and the defect type includes: no defect, surface defect, and through-hole defect.

[0010] According to some embodiments of the present invention, the method further includes: after obtaining the virtual vector matrix, using median filtering to remove outlier elements from each virtual vector matrix, and replacing the outlier elements with the median of the permutation of the element in each virtual vector matrix.

[0011] According to some embodiments of the present invention, the method further includes: after acquiring several frames of weld seam images of the tritium-containing component, performing digital image correction on each frame of weld seam images; and registering each frame of weld seam images by translation after correction.

[0012] According to some embodiments of the present invention, the step of performing difference feature analysis on adjacent frame weld seam images to obtain a virtual vector matrix includes: For each frame of weld seam image, perform cross-correlation calculation or optical flow algorithm on the weld seam image of that frame and the weld seam images of adjacent frames to obtain the virtual vector matrix.

[0013] A defect detection apparatus for a tritium-contaminated component according to a second aspect of the present invention is applied to a tritium-contaminated component and performs a defect detection method for a tritium-contaminated component according to any one of the first aspects, comprising: Replacement for gas cylinders, image acquisition devices, and controllers; The image acquisition device is set at a certain point on the tritium-contaminated component and is used to acquire several frames of weld images of the tritium-contaminated component. The alternative gas cylinder is used to fill the tritium-contaminated component with alternative gas; The controller is connected to the valve of the substitute gas cylinder and the image acquisition device, respectively, and is used to receive several frames of weld seam images sent by the image acquisition device; to perform difference feature analysis on adjacent frames of weld seam images to obtain a virtual vector matrix; and to determine the defects in the region corresponding to the weld seam image based on each frame of weld seam images and the virtual vector matrix.

[0014] According to some embodiments of the present invention, the apparatus further includes: A light source, connected to the controller, is used to receive switching commands from the controller and provide light for the weld image; A magnifying lens assembly is disposed between the image acquisition device and the tritium-contaminated component to magnify the weld image.

[0015] According to a third aspect of the present invention, a storage medium stores computer-executable instructions for performing the method as described in any one of the first aspects.

[0016] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0017] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the technical solutions of the present invention, and do not constitute a limitation on the technical solutions of the present invention.

[0018] Figure 1 This is a flowchart of a defect detection method for tritium-related components provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a background schlieren composite image in a defect detection method for tritium-related components provided in an embodiment of the present invention; Figure 3 This is a diagram showing the effect of median filtering before and after using a defect detection method for tritium-related components according to an embodiment of the present invention; Figure 4 This is a structural block diagram of a defect detection device for tritium-related components provided in another embodiment of the present invention; Figure 5 This is a schematic diagram of a defect detection device for tritium-related components provided in another embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0020] It should be understood that in the description of the embodiments of the present invention, "multiple" (or "amounts") means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. If "first," "second," etc., are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0021] like Figure 1 As shown, this embodiment of the invention provides a defect detection method for tritium-contaminated components, applied to tritium-contaminated components, including: Step S100: Inject the replacement gas into the tritium-contacting component; Step S200: Obtain several frames of weld images of the tritium-involved component at a certain point of defect detection in the tritium-involved component; Step S300: Perform differential feature analysis on adjacent frame weld seam images to obtain a virtual vector matrix; Step S400: Determine the defects in the area corresponding to the weld image based on each frame of weld image and the virtual vector matrix.

[0022] This invention utilizes weld background images, i.e. weld images, to obtain virtual vector matrices representing the differences between adjacent frames. Then, it uses each frame of weld images and the virtual vector matrix to determine the defects in the corresponding area of ​​the weld image. This invention achieves natural background schlieren measurement with weld patterns as the background, and determines the defects of tritium-related components by capturing the airflow differences between adjacent frames. It can achieve accurate, fast, and non-contact defect detection for tritium-related components.

[0023] It should be noted that the inspection of tritium-related components presents the following challenges: First, tritium has a small atomic radius and extremely high penetration and diffusion capabilities in metals, imposing stringent requirements on the near-zero defect sealing integrity of tritium-related components; second, there are numerous welds on tritium-related components and pipes within fusion devices, and existing X-ray-based weld inspection methods pose radioactive risks, and inspecting each weld with X-rays results in lengthy overall maintenance times; finally, most tritium-related components in fusion reactors are installed within vacuum chambers and blanket modules, making subsequent maintenance difficult once the reactor is built and put into operation; therefore, this invention achieves an accurate, efficient, safe, and non-contact inspection method.

[0024] It should also be noted that in step S100, the substitute gas is a "simulated leakage medium" used to replace tritium, which is radioactive and highly permeable, and is filled into the tritium-related component and maintained at a certain pressure. A gas with strong chemical inertness and a density difference with air, such as helium, is used to ensure that the detection process is safe, harmless, and non-radioactive.

[0025] In one embodiment, in step S300, difference feature analysis is performed on adjacent frame weld seam images to obtain a virtual vector matrix including: For each frame of weld seam image, perform cross-correlation calculations or optical flow algorithms on the weld seam image of that frame and the weld seam images of its adjacent frames to obtain a virtual vector matrix.

[0026] In one embodiment, the cross-correlation calculation uses Fast Fourier Transform; in step S200, 500 frames of weld seam images are continuously acquired at a certain point, with a fixed acquisition frame rate, and cross-correlation calculation or optical flow algorithm is performed on adjacent two frames of weld seam images until the difference feature analysis is performed on the 500 frames of weld seam images and their adjacent frames respectively, and the resulting virtual vector matrix is ​​250.

[0027] It should be noted that the virtual vector matrix represents a two-dimensional distribution of micron-level virtual vectors, capturing minute displacements caused by differences in the airflow density of the substitute gas.

[0028] In one embodiment, in step S400, determining the defects in the region corresponding to the weld image based on each frame of weld image and the virtual vector matrix includes: The set of moduli of each virtual vector matrix is ​​used to obtain the dimensionless matrix of the moduli. Determine the average weld image based on each frame of weld images; The dimensionless matrix of the model and the mean image of the weld are multiplied to obtain the background schlieren composite image. Determine whether there is a through-defect in the area corresponding to the weld image based on the background schlieren composite image.

[0029] It's easy to understand that an image is equivalent to a matrix. Determining the weld mean image means averaging the pixel values ​​of each pixel in each frame of the weld image to obtain a weld mean image.

[0030] In one embodiment, to ensure that the dimensionless matrix of the model and the weld mean image can be multiplied, the dimensionless matrix of the model and the weld mean image are subjected to matrix size assimilation processing before the multiplication operation is performed.

[0031] In one embodiment, determining whether a through-hole defect exists in the region corresponding to the weld image based on the background schlieren composite image includes: If the modulus of the background schlieren composite image is less than the preset first threshold, it is determined that there is no through defect in the area corresponding to the weld image. If the modulus of the background schlieren composite image is within a preset first interval, then it is determined that there is a through defect in the area corresponding to the weld image; wherein, the left endpoint of the first interval is greater than a first threshold.

[0032] like Figure 2 As shown, in one embodiment, if the modulus of the background schlieren composite image is within [0, first threshold], then the modulus is approximately 0, and the area corresponding to the weld image may only have local surface defects, and there are no through defects. If the modulus of the background schlieren composite image is approximately 1, meaning the first interval includes 1, it is determined that the area corresponding to the weld image has air leakage and a through-hole defect. The product result will highlight the through-hole defect area. Figure 2 The rightmost image shows the result of the enhancement of the natural background texture of the weld, i.e., the product result. It indicates that the area circled in the right image has a drastic change in the thermal gas density compared to the original weld image in the left image. By analogy to a leaking weld, this indicates that there is a high probability of local gas leakage in the weld.

[0033] In one embodiment, the method further includes: Obtain the original weld image; The original weld image, the background schlieren composite image (determined to have no through-hole defects), and the background schlieren composite image (determined to have through-hole defects) are used as the training set and the validation set to train the graph convolutional neural model. The graph convolutional neural model outputs the defect type of the tritium-involved component, which includes: no defect, surface defect, and through-hole defect.

[0034] It is easy to understand that the original weld image is used as data for no defects, the background schlieren composite image that determines there are no through defects is used as data for surface defects, and the background schlieren composite image that determines there are through defects is used as data for through defects.

[0035] In one embodiment, a CNN model (Convolutional Neural Network model) is used as the graph convolutional neural network model; the training method for the graph convolutional neural network model includes: The training set is input into the graph convolutional neural model to obtain the training output. Based on the training output and the defect type corresponding to the training set, adjust the parameters of the graph convolutional neural model and repeat the above steps until the loss function of the graph convolutional neural model meets the conditions for training completion. The validation set is input into the graph convolutional neural model to complete the model validation.

[0036] like Figure 3 As shown, the method for detecting tritium-related components further includes: after step S300, after obtaining the virtual vector matrix, median filtering is used to remove outlier elements from each virtual vector matrix, and the outlier elements are replaced with the median of the arrangement of the element in each virtual vector matrix, thereby reducing the deviation caused by image post-processing. Figure 3 The left and right images are comparisons of the image before and after processing in this embodiment. By filtering out outliers from the micrometer-level virtual vector distribution and reducing biases caused by image post-processing, the vector arrows of the background schlieren generated by the hot airflow are obtained (e.g., ...). Figure 3 (The green arrow circled in the right image).

[0037] The method for detecting tritium-contaminated components further includes: after step S200, acquiring several frames of weld seam images of the tritium-contaminated component, performing digital image correction on each frame of weld seam images; and registering each frame of weld seam images by translation after correction.

[0038] It should be noted that digital image correction mainly includes camera parameter space correction methods, which are used to reduce optical errors such as spherical errors generated during camera shooting.

[0039] like Figure 4 , Figure 5 As shown, this embodiment of the invention also provides a defect detection device for tritium-contaminated components, applied to tritium-contaminated component 1, for performing the aforementioned defect detection method for tritium-contaminated components, including: Replacement gas cylinder 10, image acquisition device 20, controller 30; Image acquisition device 20 is set at a certain point on the tritium-contaminated component 1 and is used to acquire several frames of weld images of the tritium-contaminated component 1. Alternate gas cylinder 10, used to fill tritium-contacting component 1 with alternative gas; The controller 30 is connected to the valve of the substitute gas cylinder 10 and the image acquisition device 20, respectively, and is used to receive several frames of weld seam images sent by the image acquisition device 20; to perform difference feature analysis on adjacent frames of weld seam images to obtain a virtual vector matrix; and to determine the defects in the area corresponding to each frame of weld seam image based on each frame of weld seam image and the virtual vector matrix.

[0040] It should be noted that the image acquisition device 20 is located on one side of the straight line where the weld is located, and the weld can be observed.

[0041] like Figure 5 As shown, in one embodiment, the location of the image acquisition device 20 is situated on one side of the centerline of the weld area; The device also includes: The light source 40, connected to the controller 30, is located on the other side of the center line of the weld area. It is used to receive the switching command of the controller 30 and provide light to the image acquisition device so that the acquired weld image is clearer.

[0042] This embodiment adds a light source, making the acquired weld image clearer and achieving localized lighting to illuminate the weld in a simple and low-cost manner.

[0043] like Figure 4 , Figure 5 As shown, in one embodiment, the apparatus further includes: The magnifying lens assembly 50 is positioned between the image acquisition device 20 and the tritium-contaminated component 1 to magnify the weld image.

[0044] In this embodiment, the detection energy is focused on a local micro-area, namely the weld seam, by using a "light source 40 + magnifying lens assembly 50" to form a uniform local light spot on the weld seam, thereby achieving micron-level defect detection.

[0045] This invention also provides an electronic device, which includes, but is not limited to: Memory, used to store programs; The processor is used to execute the program stored in the memory. When the processor executes the program stored in the memory, the processor is used to execute the aforementioned defect detection method for tritium-related components.

[0046] The processor and memory can be connected via a bus or other means.

[0047] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs, such as the method described in the embodiments of the present invention. The processor implements the above method by running the non-transitory software program and instructions stored in the memory.

[0048] The memory may include a program storage area and a data storage area, wherein the program storage area may store the operating system and application programs required for at least one function; the data storage area may store data for executing the methods described above. Furthermore, the memory may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0049] The non-transitory software program and instructions required to implement the above terminal selection method are stored in memory and are executed by one or more processors.

[0050] This invention also provides a storage medium storing computer-executable instructions for performing the above-described methods.

[0051] In one embodiment, the storage medium stores computer-executable instructions that are executed by one or more control processors.

[0052] The embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0053] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0054] This document describes embodiments of the invention, including preferred embodiments known to the inventors for carrying out the invention. Variations of these embodiments will become apparent to those skilled in the art upon reading the foregoing description. The inventors encourage those skilled in the art to adopt such variations as appropriate, and the inventors intend to practice embodiments of the invention in ways other than those specifically described herein. Therefore, the scope of the invention includes all modifications and equivalents of the subject matter set forth in the appended claims, as permitted by applicable law. Furthermore, the scope of the invention covers any combination of the foregoing elements in all possible variations thereof, unless otherwise indicated herein or otherwise clearly contradicted by the context.

Claims

1. A defect detection method for tritium-contacting components, applied to tritium-contacting components, characterized in that, include: The replacement gas is then introduced into the tritium-contaminated component; Several frames of weld images of the tritium-contaminated component are acquired at a certain point in the component being inspected. By performing differential feature analysis on adjacent weld seam images, a virtual vector matrix is ​​obtained; Based on each frame of weld seam image and the virtual vector matrix, the defects in the region corresponding to the weld seam image are determined.

2. The defect detection method for a tritium-containing component according to claim 1, characterized in that, The step of determining the defects in the region corresponding to the weld image based on each frame of weld image and the virtual vector matrix includes: The set of moduli of each virtual vector matrix is ​​used to obtain the dimensionless matrix of the moduli. Determine the average weld image based on each frame of weld images; The dimensionless matrix of the modulus and the mean image of the weld are multiplied to obtain a background schlieren composite image. The presence of a through-hole defect in the area corresponding to the weld image is determined based on the background schlieren composite image.

3. The defect detection method for a tritium-containing component according to claim 2, characterized in that, The step of determining whether there is a through defect in the region corresponding to the weld image based on the background schlieren composite image includes: If the modulus of the background schlieren composite image is less than a preset first threshold, then it is determined that there is no through defect in the area corresponding to the weld image. If the modulus of the background schlieren composite image is within a preset first interval, then it is determined that there is a through defect in the area corresponding to the weld image; wherein, the left endpoint of the first interval is greater than the first threshold.

4. The defect detection method for a tritium-containing component according to claim 2, characterized in that, The method also includes: Obtain the original weld image; The original weld image, the background schlieren composite image determined to have no through-hole defects, and the background schlieren composite image determined to have through-hole defects are used as the training set and the validation set to train the graph convolutional neural model; wherein, the graph convolutional neural model outputs the defect type of the tritium-related component, and the defect type includes: no defect, surface defect, and through-hole defect.

5. The defect detection method for a tritium-containing component according to claim 1, characterized in that, The method also includes: after obtaining the virtual vector matrix, using median filtering to remove outlier elements from each virtual vector matrix, and replacing the outlier elements with the median of the permutation of the element in each virtual vector matrix.

6. The defect detection method for a tritium-containing component according to claim 1, characterized in that, The method further includes: after acquiring several frames of weld seam images of the tritium-containing component, performing digital image correction on each frame of weld seam images; and registering each frame of weld seam images by translation after correction.

7. The defect detection method for a tritium-containing component according to claim 1, characterized in that, The step of performing difference feature analysis on adjacent weld seam images to obtain the virtual vector matrix includes: For each frame of weld seam image, perform cross-correlation calculation or optical flow algorithm on the weld seam image of that frame and the weld seam images of adjacent frames to obtain the virtual vector matrix.

8. A defect detection device for tritium-contacting components, applied to tritium-contacting components, characterized in that, For performing the method as described in any one of claims 1 to 7, comprising: Replacement for gas cylinders, image acquisition devices, and controllers; The image acquisition device is set at a certain point on the tritium-contaminated component and is used to acquire several frames of weld images of the tritium-contaminated component. The alternative gas cylinder is used to fill the tritium-contaminated component with alternative gas; The controller is connected to the valve of the substitute gas cylinder and the image acquisition device, respectively, and is used to receive several frames of weld seam images sent by the image acquisition device; to perform difference feature analysis on adjacent frames of weld seam images to obtain a virtual vector matrix; and to determine the defects in the region corresponding to the weld seam image based on each frame of weld seam images and the virtual vector matrix.

9. A defect detection device for tritium-related components according to claim 8, characterized in that, The device also includes: A light source, connected to the controller, is used to receive switching commands from the controller and provide light for the weld image; A magnifying lens assembly is disposed between the image acquisition device and the tritium-contaminated component to magnify the weld image.

10. A storage medium, characterized in that, The device stores computer-executable instructions for performing the method as described in any one of claims 1 to 7.

Images