Airtightness experimental device for detecting micro-leakage under multiple factors

Abstract

The application discloses a kind of detection under the gas tightness experimental device of multiple factors micro leakage;The device includes gas storage module, experimental body module, soap solution dip coating module, data acquisition module, dual-camera three-dimensional reconstruction module and leak point sealing module.Gas storage module includes gas compressor and pressure stabilizing vessel.Soap solution dip coating module includes hose, guide rail and soap solution container.Data acquisition module includes pressure sensor for detecting the pressure inside pressure stabilizing vessel.Camera three-dimensional reconstruction module includes first industrial camera and second industrial camera.The application uses stainless steel sheet welding to form a cubic experimental body, and different diameter, different shape leak holes are opened on the surface and weld of the experimental body, so as to simulate different gas leakage scenarios of metal shell structure, so that the growth of soap bubble of different leak holes can be studied;Further, the relationship between the shape and size of the hollow leak hole and the growth of soap bubble can be obtained, and the accuracy of detecting leakage according to the soap bubble method is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of non-destructive testing of industrial equipment, and in particular relates to an airtightness experimental device for detecting small leaks under multiple factors; Background Art

[0002] Currently, welding is the most common method for joining metals. Poor welding techniques can easily lead to cracking in the welds during use, resulting in tiny leaks on the container surface. This compromises the airtightness of the sealed container, and gas leaks can lead to serious consequences such as fires and equipment explosions. Therefore, gas storage containers and pneumatic equipment undergo a series of airtightness tests before leaving the factory.

[0003] As detection requirements increase, methods such as electron probes, spectroscopy, ultrasound, tracer gas detection, infrared thermal imaging, and fast algorithms using machine learning have emerged. Machine learning algorithms include neural networks and support vector machines. While these methods offer high sensitivity, the computational complexity of machine learning and deep learning algorithms makes them difficult, and they also require complex modeling, making their practicality limited. Other methods are costly and present limitations for large-scale deployment.

[0004] Currently, the most commonly used traditional air tightness testing methods can be divided into two categories: water testing and soap bubble testing. These two methods are commonly used for manual testing in enterprises. The soap bubble method involves applying a soap solution of a certain concentration to the surface of the object to be tested. As the leak continues to leak gas, a gradually enlarging soap bubble forms at the leak point, which can be used to determine the leak location. The soap bubble method is highly safe, economical, and simple, making it a commonly used air tightness testing method in enterprises. Due to the existence of tiny leaks in reality, it is important to study tiny leaks of different pore sizes, shapes, and surface styles. Furthermore, since different pressures, soap solution thicknesses, and the inclination of the plane where the leak is located all have a significant impact on the formation and growth of soap bubbles, these mechanism studies have important academic and application value for air tightness testing. Summary of the Invention

[0005] The purpose of the present invention is to provide an airtightness experimental device for detecting micro-leakage under multiple factors.

[0006] In its first aspect, the present invention provides an airtightness test device for detecting micro-leakages under multiple factors. The device comprises a gas storage module, a test body module, a soap solution dipping module, a data acquisition module, a dual-camera 3D reconstruction module, and a leak point sealing module. The gas storage module includes a gas compressor and a pressure-stabilizing container. The gas compressor is used to deliver gas to the pressure-stabilizing container, causing the pressure in the pressure-stabilizing container to reach a preset pressure.

[0007] The experimental body module includes an experimental body. The experimental body adopts a hollow metal shell structure and is provided with an air supply pipe. The experimental body is provided with a plurality of leakage holes of different sizes, shapes, and positions.

[0008] The soap solution dipping module includes a hose, a guide rail, and a soap solution container. The guide rail guides the experimental body, allowing it to switch between being immersed in and out of the soap solution container. The air supply pipe of the experimental body is connected to the pressure-stabilizing container via a hose and a third shut-off valve. The leak point sealing module seals any leak hole on the experimental body, ensuring that only the observed leak hole can leak.

[0009] The data acquisition module includes a pressure sensor for detecting the internal pressure of the pressure-stabilizing container. The dual-camera 3D reconstruction module includes a first industrial camera and a second industrial camera. The first and second industrial cameras are positioned directly above and to the side of the experimental body, respectively, with their lenses facing the experimental body. The angle between the first and second industrial cameras is between 10° and 75°. During operation, the first and second industrial cameras capture images of the experimental body and the soap bubbles generated by the air leak. The resulting images are reconstructed in 3D to produce 3D models of the soap bubbles at different times.

[0010] Preferably, the pressure stabilizing container comprises a gas storage tank, a micro pressure gauge connected to the gas storage tank via a first shut-off valve, a fine-adjustable shut-off valve and an exhaust valve connected to the gas storage tank. A second shut-off valve is provided between the gas storage tank and the pressure sensor.

[0011] Preferably, the experiment body module further includes a digital level, which is used to detect the posture of the experiment body.

[0012] Preferably, a plurality of first leakage holes of different diameters are arranged in sequence on one of the flat surfaces of the experimental body; a plurality of second leakage holes of the same cross-sectional area but different shapes are arranged in sequence on one of the flat surfaces of the experimental body; a plurality of third leakage holes of different diameters are arranged in sequence on one of the weld seams of the experimental body;

[0013] Preferably, the test body is formed by welding six rectangular metal plates and an air supply pipe, and each third leak hole is provided on a weld where two adjacent rectangular metal plates are connected.

[0014] Preferably, the first leakage hole and the third leakage hole are both circular.

[0015] Preferably, a pressure reducing valve and a pressure gauge are provided between the gas compressor and the pressure stabilizing container.

[0016] Preferably, the pressure sensor is connected to a computer system via a data collector, and the first industrial camera and the second industrial camera transmit the captured videos to the computer system for processing.

[0017] Preferably, the leak point sealing module adopts a sealing block. The sealing block is provided with a sealing ring. The sealing block and the outer surface of the experimental body are connected by a detachable fastening structure; the sealing ring on the sealing block surrounds the outer side of the corresponding leak point.

[0018] In a second aspect, the present invention provides a method for simulating detection of small leaks, which uses the aforementioned airtightness experimental device; the method for simulating detection of small leaks comprises the following steps:

[0019] The method for using the airtightness experimental device for detecting small leaks under multiple factors is as follows:

[0020] Step 1: Based on the simulated leak scenario, select one or more leak holes on the experimental body as target leak holes. Seal all leak holes except the target leak holes. Adjust the experimental body's posture so that the position of the target leak hole relative to the experimental body matches the leak point in the simulated leak scenario.

[0021] Step 2: The gas compressor charges the pressure stabilizing container so that the pressure in the pressure stabilizing container reaches the working pressure of the simulated leakage scenario.

[0022] Step 3: Add soap liquid to the soap liquid container; immerse the test body in the soap liquid container so that a soap liquid layer is formed on the surface where the target leak hole is located.

[0023] Step 4: Open the third shutoff valve between the pressure-stabilizing container and the experimental body. A data acquisition device monitors pressure changes in the pressure-stabilizing container. The first and second industrial cameras capture images of the experimental body and the soap bubble formed on the target leak. Using these two images from different angles, a 3D reconstruction is performed to obtain a 3D model of the soap bubble at different times.

[0024] Step 5: After the soap bubbles break, turn off the first industrial camera and the second industrial camera, and release the pressure in the pressure-stabilizing container.

[0025] Preferably, the process of performing three-dimensional reconstruction using images captured by the first industrial camera and the second industrial camera is as follows:

[0026] Step (1): Calibrate the first industrial camera and the second industrial camera to determine the internal and external parameters of the cameras;

[0027] Step (2): using the first industrial camera and the second industrial camera to capture images;

[0028] Step (3): Perform distortion correction on the captured image using camera internal parameters;

[0029] Step (4): extract features from the image corrected in step (3);

[0030] Step (5): Use image matching algorithm to perform image matching;

[0031] Step (6): Generate workpiece sparse point cloud based on image features;

[0032] Step (7): Use deep learning stereo matching algorithm for dense matching, obtain depth map, generate Mesh grid, and complete 3D reconstruction.

[0033] The present invention has the following beneficial effects.

[0034] 1. The present invention uses stainless steel metal plates to weld into a cubic experimental body, and provides leak holes of different diameters and shapes on the surface and welds of the experimental body to simulate different leakage scenarios of the metal shell structure, thereby being able to study the soap bubble growth of different leak holes; and furthermore, the relationship between the shape and size of the hollow leak hole and the soap bubble growth can be obtained, thereby improving the accuracy of leak detection based on the soap bubble method.

[0035] 2. The present invention uses dual cameras to form a dual-camera system, which can more accurately detect the three-dimensional data of soap bubbles, perform three-dimensional modeling of soap bubbles, and more accurately describe the generated soap bubble conditions, thus preparing for actual operations during on-site work.

[0036] 3. The present invention can simply and conveniently simulate factors that affect bubble generation and growth, such as different pore sizes, shapes, surface styles, inclinations, and pressures, thereby being able to study the situation of tiny leaks under multiple factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0038] Figure 2 It is a three-dimensional schematic diagram of the experimental body in the present invention.

[0039] Figure 3 This is a schematic diagram corresponding to the first metal plate of the experimental body in the present invention.

[0040] Figure 4 This is a schematic diagram corresponding to the third metal plate of the experimental body in the present invention.

[0041] Figure 5 This is a schematic diagram corresponding to the fifth metal plate of the experimental body in the present invention.

[0042] Figure 6 This is a flowchart of dual-camera 3D reconstruction in the present invention. DETAILED DESCRIPTION

[0043] The following content will provide a complete and clear description of the technical solution of the present invention in conjunction with the accompanying drawings. The present invention includes but is not limited to this type of technical solution.

[0044] In the following description, all terms related to orientation and position are based on the orientation or positional relationships in the accompanying drawings. Their purpose is simply to facilitate the simplified description of this utility patent, rather than to explicitly indicate the corresponding device-related positions and specific orientations, such as "upper," "lower," and "inner." Furthermore, the stainless steel material used in the experimental body is merely representative of one material, and other materials such as plastic, metal, and ceramic can be used instead.

[0045] like Figure 1 As shown, an airtightness experimental device for detecting small leaks under multiple factors includes an air storage module, an experimental body module, a soap solution dipping module, a data acquisition module, a dual-camera 3D reconstruction module, a leak point sealing module, and a pipeline connection module.

[0046] The gas storage module consists of a gas compressor 1 and a pressure-stabilizing vessel. Gas compressor 1 provides a high-pressure gas source, producing the required pressure for the experiment. This creates the internal pressure environment for the experiment itself, ensuring the basic testing requirements for leak detection. The pressure-stabilizing vessel includes a gas storage tank V1, a fine-tuning shut-off valve 3, a first shut-off valve 4, a micro-pressure gauge P2, a pressure sensor P3, a second shut-off valve 5, and an exhaust valve 6. These valves are mounted on top of gas storage tank V1. The fine-tuning shut-off valve 3 precisely adjusts the internal pressure of the system to meet the experimental requirements during the experiment. The exhaust valve 6 allows for quick and safe exhaust after the experiment. The micro-pressure gauge P2 is mounted above the first shut-off valve 4, and the pressure sensor P3 is mounted above the second shut-off valve 5. The first shut-off valve 4 and the second shut-off valve 5, respectively, are located between gas storage tank V1, the micro-pressure gauge P2, and the pressure sensor P3, to throttle and cut off the gas flow while ensuring airtightness. The micro-manometer P2 is used to accurately read the air pressure in the system in real time, and the pressure sensor P3 is used to monitor the air pressure in the system and cooperate with the data acquisition module.

[0047] like Figure 2 、 Figure 3 、 Figure 4 and Figure 5 As shown, the experimental body module includes a digital level S1 and an experimental body 10. The digital level S1 is mounted on the top surface of the experimental body 10 and is used to accurately measure the current tilt of the experimental body. The experimental body 10 is a stainless steel cavity workpiece formed by welding six rectangular metal plates and an air supply pipe 10-10.

[0048] The six rectangular metal plates are the first metal plate 10-1, the second metal plate 10-2, the third metal plate 10-3, the fourth metal plate 10-4, the fifth metal plate 10-5, and the sixth metal plate 10-6. The first metal plate 10-1 forms the top surface of the experimental body 10; the sixth metal plate 10-6 forms the bottom surface of the experimental body 10. The second metal plate 10-2, the third metal plate 10-3, the fourth metal plate 10-4, and the fifth metal plate 10-5 form the four side surfaces of the experimental body 10. The air supply pipe 10-10 is connected to the through hole on the third metal plate 10-3 by welding to ensure its airtightness. A sealing rubber ring is installed at the input end of the air supply pipe 10-10, which is connected to the output end of the third air supply pipeline.

[0049] The top surface of the experimental body 10 is laser-cut with multiple first leak holes 10-7 of varying diameters, arranged at equal intervals. The first leak holes 10-7 are circular. The bottom surface of the experimental body 10 is laser-cut with multiple second leak holes 10-8 of the same cross-sectional area but varying shapes, arranged at equal intervals. The shapes of the different second leak holes 10-8 include ellipse, triangle, regular hexagon, square, and long waist. The long waist shape consists of a rectangle with two semicircles at either end. Each first leak hole 10-7 and each second leak hole 10-8 are used to simulate the effects of different hole diameters and shapes on soap bubble formation during a small leak.

[0050] A plurality of third leak holes 10-9 of different diameters are arranged in sequence and at equal intervals by laser at one of the steel plate welds of the experimental body 10; the third leak holes 10-9 are circular; the first leak hole 10-7 and the third leak hole 10-9 are used together to simulate the influence of different surface patterns on soap bubble generation during small leaks.

[0051] The soap liquid dipping module includes a hose 9, a guide rail 11 and a soap liquid container 14. The soap liquid container 14 contains the soap liquid required for the experiment; the vertical guide rail 11 is fixed to the frame; the experimental body 10 is slidably connected to the guide rail 11 through a slider to facilitate the lifting and lowering of the experimental body 10 so that the experimental body can be immersed in the soap liquid container 14, thereby obtaining a soap liquid layer of uniform thickness on the surface of the experimental body 10. The posture of the experimental body 10 can be adjusted. The output end of the third air supply pipeline is connected to the output port of the gas storage tank V1 through a flexible hose with elasticity, a flexible joint 8, and a third shut-off valve 7.

[0052] The leak point sealing module is used to seal irrelevant leak points during the experiment to ensure the airtightness of the experimental workpiece, so that soap bubbles are only generated at the observed leak points.

[0053] In some embodiments, the leak sealing module uses waterproof, highly viscous tape. This tape is applied to the leak to be sealed and can test the pressure generated by the internal pressure of the main body 10. By applying tape to different leaks, different experimental requirements can be met.

[0054] In some other embodiments, the leakage point sealing module adopts a sealing block. A sealing ring is provided on the sealing block. A detachable fastening structure that cooperates with the sealing block is provided on the outer surface of the experimental body. When the sealing block is against the outer side surface of the experimental body, the sealing ring surrounds the outside of the corresponding leakage point to ensure the sealing effect. The detachable fastening structure can adopt a bolt and nut connection structure or a snap connection structure. When using the bolt and nut connection structure, a fixing nut is welded on the experimental body, and a bolt is used to pass through the through hole on the sealing block and connect with the nut.

[0055] The pipeline connection module includes a pressure reducing valve 2, a pressure gauge P1, a third stop valve 7, a flexible joint 8, a first air supply pipeline, a second air supply pipeline, and a third air supply pipeline. The first air supply pipeline, the second air supply pipeline, and the third air supply pipeline are respectively used to connect the gas compressor 1 to the input port of the gas storage tank V1, the air outlet of the gas storage tank V1 to the flexible joint 8, and the hose 9 to the experimental body 10. The pressure reducing valve 2 and the pressure gauge P1 are arranged on the first air supply pipeline, and are respectively used to adjust the internal air pressure of the system and read the air pressure in the entire device in real time. The third stop valve 7 is arranged on the second air supply pipeline, and is used to throttle and cut off the gas to ensure air tightness. The flexible joint 8 is arranged on the second air supply pipeline. When it is necessary to change experimental factors such as the inclination of the device and the shooting angle, the connection between the flexible joint and the stop valve can be loosened, and the experimental body in the middle can be rotated.

[0056] The data acquisition module includes a pressure sensor P3, a data collector 12, and a computer system 13. The pressure sensor P3 monitors the air pressure in the system, and the data collector 12 collects data and transmits it to the computer system 13 for analysis of experimental data.

[0057] The dual-camera 3D reconstruction module includes a first industrial camera 15, a second industrial camera 16, and a computer system 13. The first and second industrial cameras 15, 16 are positioned directly above and to the side of the experimental unit, respectively, for recording. The angle between the first and second industrial cameras 15, 16 ranges from 10° to 75°, and can be adjusted as needed to simultaneously capture views of the soap bubble from two different directions. Because the soap bubble itself is transparent and has few features, the first and second industrial cameras 15, 16 capture the entire workpiece for 3D reconstruction. The captured videos are then transmitted to the computer system 13 for processing.

[0058] like Figure 6As shown, the specific steps of dual-camera 3D reconstruction in this embodiment are:

[0059] Step 1: Calibrate the first industrial camera 15 and the second industrial camera 16 to determine the internal and external parameters of the cameras;

[0060] Step 2: Use the first industrial camera 15 and the second industrial camera 16 to capture the workpiece image;

[0061] Step 3: Perform distortion correction on the captured image using the camera’s intrinsic parameters;

[0062] Step 4: Use ORB feature detection algorithm for feature extraction;

[0063] Step 5: Use the image matching algorithm to match the left and right images, and use the RANSAC algorithm to eliminate false matches;

[0064] Step 6: Based on the image features, use the sfM algorithm (Structure from Motion) to perform "structure recovery from motion" processing to generate a sparse point cloud of the workpiece;

[0065] Step 7: Use the deep learning stereo matching algorithm to perform dense matching and obtain the workpiece depth map;

[0066] Step 8: Generate a Mesh based on the dense point cloud to complete the dual-camera 3D reconstruction.

[0067] The method for using the airtightness experimental device for detecting small leaks under multiple factors is as follows:

[0068] Step 1: Select the required leak hole from the first leak hole 10-7, the second leak hole 10-8, and the third leak hole 10-9 based on the research needs. Determine the required workpiece angle, adjust the experimental body 10's posture, and observe the digital level S1 to ensure that the experimental body aligns with the studied situation. Adjust the first industrial camera 15 and the second industrial camera 16 to the appropriate position based on the experimental needs. Use the leak sealing module to seal the entire experimental body 10.

[0069] Step 2: Open pressure reducing valve 2 and third shut-off valve 7, close all other valves, turn on gas compressor 1 to produce high-pressure gas and pass it into the device for 2 seconds. Turn off gas compressor 1 and pressure reducing valve 2, and observe whether the micro-pressure gauge P2 reading is stable to determine the airtightness of the device. If the micro-pressure gauge P2 reading fluctuates too much and cannot be stopped, it indicates poor airtightness and needs to be rechecked and steps 1 and 2 repeated. If the micro-pressure gauge P2 reading is stable, it indicates good airtightness. Open exhaust valve 6 to release the gas and then start the experiment.

[0070] Step 3: Determine the required pressure for the experiment based on research needs. Open pressure reducing valve 2, close all other valves, and open gas compressor 1 to produce high-pressure gas. Use fine-tuning stop valve 3 to finely adjust the pressure within the device. Observe the micro-pressure gauge P2 until the desired pressure is reached. Then, close gas compressor 1, pressure reducing valve 2, and fine-tuning stop valve 3. Prepare a soap solution of a certain concentration based on research needs and place it in soap solution container 14. Remove the leak sealing module from the desired leak point. Adjust the position of the third air supply line on guide rail 11 so that the experimental body 10 is lowered and immersed in soap solution container 14. Then, lift it up to obtain a uniform soap solution layer on the surface of the target leak point.

[0071] Step 4: Start computer system 13 to prepare for receiving experimental data. Start first and second industrial cameras 15, 16 to begin capturing experimental images. Start data collector 12 to collect air pressure data. Open third shutoff valve 7. Gas within the device will leak from the target leak point, forming bubbles. First and second industrial cameras 15, 16 capture images and transmit them to computer system 13. After modeling, computer system 13 saves and records the experimental images and pressure data.

[0072] Step 5: After the soap bubbles are broken, turn off the first industrial camera 15, the second industrial camera 16, and the data collector 12, open the exhaust valve 6, and quickly depressurize the container, so that the experiment ends safely and quickly.

[0073] The above content is a further explanation of the present invention by combining specific examples, and the specific implementation of the present invention should not be considered to be limited to these examples. For those skilled in the art to which the present invention belongs, simple deductions or substitutions to a certain extent can be made without departing from the inventive concept, and all of these should be considered to fall within the scope of protection of the present invention.

Claims

1. An airtightness test device for detecting micro-leakage under multiple factors, comprising an air storage module, a test body module, and a soap solution dipping module; characterized in that: It also includes a data acquisition module, a dual-camera three-dimensional reconstruction module, and a leak point sealing module; the gas storage module includes a gas compressor (1) and a pressure-stabilizing container; the gas compressor (1) is used to deliver gas to the pressure-stabilizing container so that the pressure-stabilizing container reaches a preset pressure; The experimental body module includes an experimental body (10); the experimental body (10) adopts a hollow metal shell structure and is provided with an air supply pipe (10-10); the experimental body (10) is provided with a plurality of leakage holes of different sizes, and / or different shapes, and / or different positions; The soap solution dipping module comprises a hose (9), a guide rail (11) and a soap solution container (14); the guide rail (11) is used to provide guidance for the experimental body (10), so that the experimental body (10) can switch between a state of being immersed in the soap solution container (14) and a state of being separated from the soap solution container (14); the air supply pipe (10-10) of the experimental body (10) is connected to the pressure-stabilizing container through a hose (9) and a third stop valve (7); the leakage point sealing module is used to seal the leakage hole at any position, so that only the observed leakage hole on the experimental body (10) can leak; The data acquisition module includes a pressure sensor (P3) for detecting the internal pressure of the pressure-stabilizing container; the dual-camera three-dimensional reconstruction module includes a first industrial camera (15) and a second industrial camera (16); the first industrial camera (15) and the second industrial camera (16) are respectively arranged directly above and above the experimental body (10), and the lenses are facing the experimental body (10); the angle between the first industrial camera (15) and the second industrial camera (16) is 10° to 75°; during operation, the first industrial camera (15) and the second industrial camera (16) collect images of the experimental body (10) and soap bubbles generated by air leakage; by three-dimensionally reconstructing the obtained images, a three-dimensional model of the soap bubbles at different times is obtained; A plurality of first leakage holes (10-7) with different diameters arranged in sequence are provided on one of the flat surfaces of the experimental body (10); a plurality of second leakage holes (10-8) with the same cross-sectional area but different shapes are arranged in sequence are provided on one of the flat surfaces of the experimental body (10); a plurality of third leakage holes (10-9) with different diameters arranged in sequence are provided on one of the weld seams of the experimental body (10); The experimental body (10) is formed by welding six rectangular metal plates and an air supply pipe (10-10); each third leak hole (10-9) is arranged on a weld at a connection point between two adjacent rectangular metal plates.

2. The airtightness test device for detecting micro-leakage under multiple factors according to claim 1, characterized in that: The pressure-stabilizing container comprises a gas storage tank (V1), a micro-pressure gauge (P2) connected to the gas storage tank (V1) via a first stop valve (4), a fine-adjustment stop valve (3) and an exhaust valve (6) connected to the gas storage tank (V1); and a second stop valve (5) is provided between the gas storage tank (V1) and the pressure sensor (P3).

3. The airtightness test device for detecting micro-leakage under multiple factors according to claim 1 is characterized in that: The experimental body module further includes a digital level (S1); the digital level (S1) is used to detect the posture of the experimental body (10).

4. The airtightness test device for detecting micro-leakage under multiple factors according to claim 1, characterized in that: The first leakage hole (10-7) and the third leakage hole (10-9) are both circular.

5. The airtightness test device for detecting micro-leakage under multiple factors according to claim 1 is characterized in that: A pressure reducing valve (2) and a pressure gauge (P1) are provided between the gas compressor (1) and the pressure stabilizing container.

6. The airtightness test device for detecting micro-leakage under multiple factors according to claim 1, characterized in that: The pressure sensor (P3) is connected to the computer system (13) via the data acquisition device (12); the first industrial camera (15) and the second industrial camera (16) transmit the captured video to the computer system (13) for processing.

7. A method for simulating detection of small leaks, characterized by: Using the airtightness experimental device for detecting micro-leakage under multiple factors as described in claim 1; the micro-leakage simulation detection method includes the following steps: The method for using the airtightness experimental device for detecting small leaks under multiple factors is as follows: Step 1: Select one or more leak holes as target leak holes on the experimental body (10) according to the simulated leakage scenario; seal all leak holes other than the target leak holes; adjust the posture of the experimental body (10) so that the position of the target leak hole relative to the experimental body (10) matches the leak point in the simulated leakage scenario; Step 2: The gas compressor (1) charges the pressure stabilizing container so that the pressure in the pressure stabilizing container reaches the working pressure of the simulated leakage scenario; Step 3: Add soap liquid to the soap liquid container (14); immerse the experimental body (10) in the soap liquid container (14) so ​​that a soap liquid layer is formed on the surface where the target leak hole is located; Step 4: Open the third stop valve (7) between the pressure-stabilizing container and the experimental body (10); the data acquisition device (12) detects the pressure change of the pressure-stabilizing container; the first industrial camera (15) and the second industrial camera (16) collect images of the experimental body (10) and the soap bubble formed on the target leak hole; use the two images at different angles to perform three-dimensional reconstruction to obtain a three-dimensional model of the soap bubble at different times; Step 5: After the soap bubbles are broken, the first industrial camera (15) and the second industrial camera (16) are turned off, and the pressure-stabilizing container is depressurized.

8. The method for simulating detection of a small leak according to claim 7, characterized in that: The process of performing three-dimensional reconstruction using the images captured by the first industrial camera (15) and the second industrial camera (16) is as follows: Step (1): calibrate the first industrial camera (15) and the second industrial camera (16) to determine the internal and external parameters of the cameras; Step (2): using the first industrial camera (15) and the second industrial camera (16) to capture images; Step (3): Perform distortion correction on the captured image using camera internal parameters; Step (4): extract features from the image corrected in step (3); Step (5): Use image matching algorithm to perform image matching; Step (6): Generate workpiece sparse point cloud based on image features; Step (7): Use the deep learning stereo matching algorithm to perform dense matching, obtain the depth map, and generate the Mesh mesh to complete the 3D reconstruction.

Images