Intelligent vacuum box detection system and method

Abstract

The application discloses a kind of intelligent vacuum box detection system and method.The present application is mainly used for the local rough detection of vacuum equipment, the intelligent vacuum box detection system includes vacuum system, image acquisition system, leak rate analysis and alarm system.The vacuum system can be pumped by adjusting device to realize the vacuum box, and the other side of the area to be detected establishes pressure difference;The image acquisition system can realize continuous image acquisition of the bubble in the vacuum box and transmit to the processing terminal.The leak rate analysis and alarm system can realize automatic interpretation of bubble diameter and bubble generation rate, and convert the leak rate value of leak point according to bubble diameter and generation rate, and if the leak rate value exceeds the acceptance range, an alarm signal is sent.The present application can realize the on-site connection of detection device, automatic determination of detection phenomenon, reduces the requirement for the number and experience of detection personnel, improves the detection efficiency and reliability.

Description

Technical Field

[0001] This application relates to the field of nondestructive testing technology, specifically to an intelligent vacuum chamber testing system and method. Background Technology

[0002] Vacuum equipment has wide applications in the power, chemical, aerospace, and shipbuilding industries. Controlling the leakage rate of vacuum equipment is a decisive factor in whether it can perform to its full potential. Especially during the construction of large-scale vacuum equipment, leak detection is necessary for quality control. Different leak detection methods have different characteristics, which determines their applicability; some are suitable for rough inspection, some for fine inspection, and some can only detect leaks in small, localized areas. In large-scale vacuum equipment, a combination of rough and fine inspection, as well as localized and overall leak rate detection, is often used. Vacuum chamber testing, as a method of localized rough inspection, is widely used in the construction and testing of large-scale vacuum equipment.

[0003] Vacuum chamber testing involves applying a special foaming agent to the area to be tested. When a local pressure differential is established in the area using a vacuum chamber, if a leak exists, bubbles will be observed continuously appearing at the leak point (see reference). Figure 1 Vacuum chamber testing, as a method of localized rough inspection, also shoulders the task of detecting major leaks as soon as possible.

[0004] Vacuum chamber inspection is a labor-intensive task, often requiring a large number of inspectors and support personnel. Furthermore, the operation of vacuum chambers demands considerable experience from operators. Inspectors use the vacuum chamber's observation window to monitor bubble size and formation rate to determine the leak rate and decide whether to rework the affected area. Developing intelligent inspection systems and methods not only reduces the workload of operators but also eliminates the reliance on human experience in vacuum chamber inspection, providing a reliable way to improve inspection efficiency and accuracy. Summary of the Invention

[0005] This application provides an intelligent vacuum chamber inspection system and method, which can solve the technical problems of high labor intensity and high dependence on personnel experience for vacuum chamber inspection operators, and avoid missed detection and false detection.

[0006] This application provides an intelligent vacuum chamber detection system, including a vacuum pumping system, an image acquisition system, and a leak rate analysis and alarm system;

[0007] The vacuum system includes a vacuum pump, a vacuum chamber, a pumping speed regulating valve, and a pumping gas pipeline. The vacuum pump is connected to the vacuum chamber through the pumping gas pipeline, and the pumping speed regulating valve is installed on the pumping gas pipeline. The vacuum pump is used to extract gas from the vacuum chamber, the vacuum chamber is used to form a local sealed space in the area being detected, and the pumping speed regulating valve is used to control the pumping speed of the vacuum pump and maintain the pressure.

[0008] The image acquisition system is used to acquire images of bubbles generated in the detection area inside the vacuum chamber and transmit the acquired images to the leak rate analysis and alarm system.

[0009] The leak rate analysis and alarm system is connected to the image acquisition system. The leak rate analysis and alarm system is used to process and analyze the acquired images. If the leak rate value is greater than the alarm threshold range, an alarm message is issued.

[0010] Furthermore, the vacuum chamber is a transparent vacuum chamber made of acrylic glass material bonded together, and a replaceable rubber ring is installed at the bottom of the vacuum chamber to form a sealed space with the test piece.

[0011] Furthermore, the edge of the vacuum chamber is engraved with scales to calibrate the size of bubbles in the acquired images.

[0012] Furthermore, the image acquisition system includes an LED lighting source, a CCD camera, and an image acquisition card; the LED lighting source is located on the side wall of the vacuum chamber, the CCD camera is located on the top of the vacuum chamber, and the image acquisition card is electrically connected to the CCD camera;

[0013] The LED lighting source is used to provide supplemental lighting for the CCD camera, which is used to photograph and record bubbles generated in the detection area inside the vacuum chamber. The image acquisition card is used to collect and store the photographed image information as an acquired image and transmit it to the leak rate analysis and alarm system.

[0014] Furthermore, the leakage rate analysis and alarm system includes a computing module and an alarm flashing light;

[0015] The computing module is connected to the image acquisition card. The computing module is used to process and analyze the acquired images. If the leakage rate value is greater than the alarm threshold range, the alarm flashing light is controlled to flash.

[0016] This application also provides a method for detecting a smart vacuum chamber, including the following steps:

[0017] Step 1: Deploy the intelligent vacuum chamber detection system described above on the area to be tested, and ensure its airtightness;

[0018] Step 2: Perform a vacuuming operation on the vacuum chamber corresponding to the area to be tested. After the test pressure is reached, close the pumping speed regulating valve and disconnect the vacuum pump from the vacuum chamber to maintain the pressure inside the vacuum chamber.

[0019] Step 3: After the intelligent vacuum chamber detection system has maintained pressure for the first time, a leak detection is performed. The image acquisition system is controlled to acquire images of air bubbles generated in the detection area inside the vacuum chamber, and the acquired images are transmitted to the leak rate analysis and alarm system. The leak rate analysis and alarm system is used to process and analyze the acquired images. If the leak rate value is greater than the alarm threshold range, it is determined that a leak has been found in the detection area and an alarm message is issued. Otherwise, the detection area is determined to be qualified.

[0020] Furthermore, the step of evacuating the vacuum chamber for the corresponding area to be detected includes:

[0021] After setting up the intelligent vacuum chamber detection system, apply a special foaming agent to the area to be tested, cover the area to be tested with the vacuum chamber, open the pumping speed regulating valve connected to the vacuum chamber, and create negative pressure in the vacuum chamber to adsorb the part to be tested in the area to be tested.

[0022] Furthermore, the leak rate analysis and alarm system is used for processing and analyzing the acquired images, including:

[0023] The acquired image is sharpened to extract bubble edges, and the image is smoothed using a Gaussian function. The Laplacian-Gaussian operator is combined with the zero-crossing point of the second derivative to detect the image edges.

[0024] By measuring the projected area of ​​the bubble and converting the area into the bubble diameter; whereby... The projected area of ​​the bubble can be obtained by measuring the pixels of the acquired image; where S represents the area of ​​the image, x and y are the pixel values, C is the connected region within the extracted bubble edge, and f is a function related to the calibration scale inside the vacuum chamber; the measured area is obtained through... Converted to the equivalent bubble diameter D;

[0025] pass The evolution time t of the bubble is calculated, where F represents the number of frames and Fr represents the frame rate of the CCD camera, according to... The evolution rate v of the bubble is obtained by taking the reciprocal of time;

[0026] In vacuum chamber leak detection, through The leakage rate at the leak point is calculated; where v avg D represents the average rate of bubble generation. avg P is the average diameter of the bubble. b This represents the gas pressure inside the bubble.

[0027] This application also provides a method for detecting a smart vacuum chamber, including the following steps:

[0028] Step 1: Deploy the intelligent vacuum chamber detection system described above on the area to be tested, and ensure its airtightness;

[0029] Step 2: Perform a vacuuming operation on the vacuum chamber corresponding to the area to be tested. After the test pressure is reached, close the pumping speed regulating valve and disconnect the vacuum pump from the vacuum chamber to maintain the pressure inside the vacuum chamber.

[0030] Step 4: Perform dynamic pressure change leak detection in the intelligent vacuum chamber detection system, control the image acquisition system to acquire images of bubbles generated in the detection area inside the vacuum chamber, and transmit the acquired images to the leak rate analysis and alarm system; the leak rate analysis and alarm system is based on a BP neural network to realize dynamic judgment of leak point and leak rate.

[0031] Furthermore, the BP neural network is trained before use, and the steps for training the BP neural network include:

[0032] The initial parameters of the neural network are set, with the average bubble diameter and average evolution rate as the input vectors and the leak rate of the standard sample as the expected output. Under a specified pumping speed, the relationship between the average bubble diameter, average bubble evolution rate and leak rate in standard samples with different leak rates is obtained.

[0033] The actual output is calculated, and a recursive method is used to return from the output node to the intermediate hidden layer to adjust the weights for optimization until the calculation error is less than a preset threshold, so that the neural network realizes the input-output mapping relationship.

[0034] The intelligent vacuum chamber testing system and method provided in this application can efficiently and accurately perform localized rough inspections of vacuum equipment through intelligent interpretation methods under single-person operation. This reduces the labor intensity of testing operators and is not highly dependent on personnel experience, avoiding missed or incorrect inspections and improving the accuracy of product testing.

[0035] The detection method of this invention enables on-site connection of the detection device and automatic determination of the detected phenomena. This detection system and method can perform localized coarse inspections of LNG membrane enclosure systems under single-person operation, based on machine interpretation, reducing the requirements for the number and experience of inspection personnel and improving detection efficiency and reliability. Attached Figure Description

[0036] The technical solution and other beneficial effects of this application will become apparent from the following detailed description of specific embodiments in conjunction with the accompanying drawings.

[0037] Figure 1 This is a schematic diagram illustrating the implementation principle of the intelligent vacuum chamber detection system provided in this application embodiment.

[0038] Figure 2 for Figure 1 The flowchart for the testing implementation of the provided intelligent vacuum chamber testing system.

[0039] Figure 3 This is a schematic diagram of the intelligent vacuum chamber detection system provided in an embodiment of this application.

[0040] Figure 4 This is a graph showing the relationship between bubble diameter and the number of frames in the acquired image, provided in an embodiment of this application.

[0041] Figure 5 A flowchart illustrating the steps for training a BP neural network as provided in an embodiment of this application. Detailed Implementation

[0042] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0043] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication between two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0044] The following disclosure provides many different implementations or examples for carrying out different structures of this application. To simplify the disclosure of this application, the components and arrangements of specific examples are described below. Of course, these are merely examples and are not intended to limit this application.

[0045] This invention discloses an intelligent vacuum chamber detection system for leak detection, mainly used for localized rough inspection of vacuum equipment. The intelligent vacuum chamber detection system includes a vacuum pumping system, an image acquisition system, and a leak rate analysis and alarm system. The vacuum pumping system can evacuate air from the vacuum chamber through an adjustment device, establishing a pressure difference with the other side of the area to be inspected. The image acquisition system can continuously acquire images of bubbles within the vacuum chamber and transmit them to a processing terminal. The leak rate analysis and alarm system can automatically determine the bubble diameter and bubble generation rate, converting these parameters into a leak rate value for the leak point. If the leak rate value exceeds the acceptable range, an alarm signal is issued.

[0046] Specifically, in combination Figure 1 , Figure 2 The principle shown is as follows: Figure 3 As shown, the intelligent vacuum chamber detection system includes a vacuum pumping system, an image acquisition system, and a leak rate analysis and alarm system. The vacuum pumping system includes a vacuum pump 11, a vacuum chamber 12, a pumping speed regulating valve 13, and a pumping gas pipeline 14. The vacuum pump 11 is connected to the vacuum chamber 12 via the pumping gas pipeline 14, and the pumping speed regulating valve 13 is installed on the pumping gas pipeline 14. The vacuum pump 11 is used to extract gas from the vacuum chamber 12, and the vacuum chamber 12 is used to form a locally sealed space in the detected area. The pumping speed regulating valve 13 is used to control the pumping speed of the vacuum pump 11 and maintain the pressure. The image acquisition system is used to acquire images of bubbles generated in the detected area within the vacuum chamber 12 and transmit the acquired images to the leak rate analysis and alarm system. The leak rate analysis and alarm system is connected to the image acquisition system and is used to process and analyze the acquired images. If the leak rate value exceeds the alarm threshold range, an alarm message is issued.

[0047] Furthermore, the vacuum chamber 12 is a transparent vacuum chamber made of bonded acrylic glass material. A replaceable rubber ring is installed at the bottom of the vacuum chamber 12 to form a sealed space with the inspected object. The transparent vacuum chamber, made of bonded acrylic glass material, allows for visual inspection of the test results, enabling manual verification in addition to machine interpretation. A vacuum pump unit is connected to the vacuum chamber, and the pumping speed is adjusted via a regulating valve to achieve the specified test pressure within the process documentation.

[0048] Furthermore, the edge of the vacuum chamber 12 is engraved with scales for calibrating the size of bubbles in the acquired images.

[0049] like Figure 3As shown, the image acquisition system includes an LED lighting source 21, a CCD camera 22, and an image acquisition card 23. The LED lighting source 21 is located on the side wall of the vacuum chamber 12, the CCD camera 22 is located on the top of the vacuum chamber 12, and the image acquisition card 23 is electrically connected to the CCD camera 22. The LED lighting source 21 is used to provide supplementary lighting for the CCD camera 22, the CCD camera 22 is used to photograph and record bubbles generated in the detection area inside the vacuum chamber, and the image acquisition card 23 is used to collect and store the photographed image information as an acquired image and transmit it to the leak rate analysis and alarm system.

[0050] The image acquisition system, with its CCD camera integrated within the vacuum chamber, can acquire detected image information under LED illumination and transmit it to the processing module of the leak rate analysis and alarm system via a data acquisition card.

[0051] like Figure 3 As shown, the leak rate analysis and alarm system includes a calculation module 31 and an alarm flashing light 32; wherein the calculation module 31 is connected to the image acquisition card 23, and the calculation module 31 is used to process and analyze the acquired image. If the leak rate value is greater than the alarm threshold range, the alarm flashing light 32 is controlled to flash.

[0052] The leak rate analysis and alarm system processes CCD-acquired images to obtain the diameter and generation rate of bubbles. Combined with the working pressure of the vacuum chamber, the leak rate can be calculated. If the leak rate value exceeds the acceptable range, the alarm light will flash.

[0053] This application provides a smart vacuum box detection method in the embodiments, including the following steps:

[0054] Step 1: Deploy the intelligent vacuum chamber detection system described above on the area to be tested, and ensure its airtightness;

[0055] Step 2: Perform vacuuming on the vacuum chamber 12 corresponding to the area to be tested. After the test pressure is reached, close the pumping speed regulating valve and disconnect the vacuum pump from the vacuum chamber to maintain the pressure inside the vacuum chamber 12.

[0056] Step 3: After the intelligent vacuum chamber detection system has maintained pressure for the first time, a pressure-holding leak detection is performed. The image acquisition system is controlled to acquire images of the air bubbles generated in the detection area inside the vacuum chamber 12, and the acquired images are transmitted to the leak rate analysis and alarm system. The leak rate analysis and alarm system is used to process and analyze the acquired images. If the leak rate value is greater than the alarm threshold range, it is determined that a leak has been found in the detection area and an alarm message is issued. Otherwise, the detection area is determined to be qualified.

[0057] Furthermore, the step of performing a vacuuming operation on the vacuum chamber 12 corresponding to the area to be detected includes:

[0058] After setting up the intelligent vacuum chamber detection system, apply a special foaming agent to the area to be tested, cover the area to be tested with the vacuum chamber, open the pumping speed regulating valve connected to the vacuum chamber, and create negative pressure in the vacuum chamber to adsorb the part to be tested in the area to be tested.

[0059] Furthermore, the leak rate analysis and alarm system is used for processing and analyzing the acquired images, including:

[0060] The acquired image is sharpened to extract bubble edges, and the image is smoothed using a Gaussian function. The Laplacian-Gaussian operator is combined with the zero-crossing point of the second derivative to detect the image edges.

[0061] By measuring the projected area of ​​the bubble and converting the area into the bubble diameter; whereby... The projected area of ​​the bubble can be obtained by measuring the pixels of the acquired image; where S represents the area of ​​the image, x and y are the pixel values, C is the connected region within the extracted bubble edge, and f is a function related to the calibration scale inside the vacuum chamber; the measured area is obtained through... Converted to the equivalent bubble diameter D;

[0062] pass The evolution time t of the bubble is calculated, where F represents the number of frames and Fr represents the frame rate of the CCD camera, according to... The evolution rate v of the bubble is obtained by taking the reciprocal of time;

[0063] In vacuum chamber leak detection, through The leakage rate at the leak point is calculated; where v avg D represents the average rate of bubble generation. avg P is the average diameter of the bubble. b This represents the gas pressure inside the bubble.

[0064] Another embodiment of this application provides a smart vacuum chamber detection method, including the following steps:

[0065] Step 1: Deploy the intelligent vacuum chamber detection system described above on the area to be tested, and ensure its airtightness;

[0066] Step 2: Perform vacuuming on the vacuum chamber 12 corresponding to the area to be tested. After the test pressure is reached, close the pumping speed regulating valve and disconnect the vacuum pump from the vacuum chamber to maintain the pressure inside the vacuum chamber 12.

[0067] Step 4: Perform dynamic pressure change leak detection in the intelligent vacuum chamber detection system, control the image acquisition system to acquire images of bubbles generated in the detection area inside the vacuum chamber 12, and transmit the acquired images to the leak rate analysis and alarm system; the leak rate analysis and alarm system is based on a BP neural network to realize dynamic judgment of leak points and leak rates.

[0068] The BP neural network is trained before use, such as Figure 5 As shown, the steps for training a BP neural network include:

[0069] The initial parameters of the neural network are set, with the average bubble diameter and average evolution rate as the input vectors and the leak rate of the standard sample as the expected output. Under a specified pumping speed, the relationship between the average bubble diameter, average bubble evolution rate and leak rate in standard samples with different leak rates is obtained.

[0070] The actual output is calculated, and a recursive method is used to return from the output node to the intermediate hidden layer to adjust the weights for optimization until the calculation error is less than a preset threshold, so that the neural network realizes the input-output mapping relationship.

[0071] The intelligent vacuum chamber testing system and method provided in this application can efficiently and accurately perform localized rough inspections of vacuum equipment through intelligent interpretation methods under single-person operation. This reduces the labor intensity of testing operators and is not highly dependent on personnel experience, avoiding missed or incorrect inspections and improving the accuracy of product testing.

[0072] The detection method of this invention enables on-site connection of the detection device and automatic determination of the detected phenomena. This detection system and method can perform localized coarse inspections of LNG membrane enclosure systems under single-person operation, based on machine interpretation, reducing the requirements for the number and experience of inspection personnel and improving detection efficiency and reliability.

[0073] In summary, this invention is an intelligent vacuum chamber detection system and method for leak detection, and the implementation process is as follows: Figure 2 This invention enables efficient and accurate localized coarse inspection of vacuum systems through intelligent interpretation methods, allowing for single-person operation. The technical solution employed in this invention combines... Figures 3 to 4 A detailed explanation of the principles and procedures to be followed when using the above system to test intelligent vacuum chambers in vacuum equipment:

[0074] Step 1: Reference Figure 3 Complete the setup for the intelligent vacuum chamber testing system before testing, following the connection method shown in the diagram.

[0075] Step 2: After completing the system layout, apply a special foaming agent to the area to be detected, cover the area to be inspected with a vacuum box, open the pumping speed regulating valve connected to the vacuum box, press the vacuum box against the part to be detected, and wait until the pressure reaches the detection pressure P b After that, pressure-holding leak detection or rapid leak detection based on the pressure change during the air extraction process can be carried out.

[0076] Step 3: If pressure-holding leak detection is performed on the system, after reaching the detection pressure P b close the pumping speed regulating valve, the CCD starts to collect data. After waiting for n seconds (the acquisition time can be appropriately extended to obtain more representative data), the CCD camera completes the photo acquisition. After the terminal finishes data processing, if the leak rate exceeds the preset Q A , the alarm light changes from being constantly on to flashing, indicating that a leak point is found in the detection area; otherwise, the detection is qualified.

[0077] Specifically, the CCD converts the optical signal of the bubble into an electrical signal, then converts the electrical signal into a digital image through an image acquisition card, and performs image processing and analysis through an operation module for judgment. Since the bubbles form on the surface of the inspected part, in the grayscale image, the gray value at the edge of the bubble is lower than that of the background. Image processing technology is combined to identify the generation of bubbles.

[0078] First, through image sharpening, the edge of the bubble is extracted. The Gaussian function is used to smooth the image, combined with the Laplacian of Gaussian operator, and the edge of the image is detected according to the zero-crossing of the second derivative.

[0079] Secondly, calculate the diameter of the bubble. Bubbles often form in a circular or elliptical shape and deform under the action of force during the movement process. There are often deviations in the measurement of the diameters along different radial directions of the bubble. In this invention, the projection area of the bubble is measured, and the area is converted into the bubble diameter. The projection area of the image pixel points can be obtained by measuring according to the following formula.

[0080]

[0081] where S represents the area of the figure, x and y are pixel values, C is the connected domain inside the extracted bubble edge, and f is a function related to the calibration scale inside the vacuum box. The measured area is converted into the equivalent diameter D of the bubble through the following formula. As the bubble generates, during the process of continuously increasing in volume under the pressure difference until the bubble finally bursts and disappears, the change of the bubble diameter follows Figure 4 the change rule, and the diameter of a single bubble takes the maximum diameter during its evolution process.

[0082]

[0083] Finally, the evolution time t of the bubble can be calculated using the following formula, where F represents the number of frames and Fr represents the frame rate of the CCD camera. The evolution rate v of the bubble can be obtained by taking the reciprocal of the time according to formula (4).

[0084]

[0085]

[0086] To obtain more general data, the bubble radius and bubble evolution rate are calculated by averaging the maximum diameter data of the first m bubbles captured by the camera. Here, m is an integer, typically around 5.

[0087] In vacuum chamber leak detection, the leak rate at the leak point is calculated using the following formula (4).

[0088]

[0089] Where v avg D represents the average rate of bubble generation. avg P is the average diameter of the bubble. b P is the gas pressure inside the bubble. b The magnitude of P is equal to the sum of the atmospheric pressure at the bubble surface, the surface tension of the bubble, and the pressure of the liquid column indicating leakage. Since the surface tension and the pressure of the liquid column indicating leakage are usually negligible compared to the atmospheric pressure at the bubble surface, P is generally considered to be... b It is approximately equal to the atmospheric pressure at the surface of the bubble, i.e., the pressure P inside the vacuum chamber. V .

[0090] Step 4: If the system is performing rapid leak detection due to pressure changes during evacuation, the CCD will simultaneously begin image acquisition as soon as the vacuum pump starts evacuation. The image acquisition will continue until the detection pressure P is reached. b After CCD image acquisition, the calculation of bubble diameter and generation speed is performed as per step 3. The detection of this area ends, and the image information is transmitted to the computing module for leak rate determination based on a BP neural network. During the evacuation process, the pressure inside the vacuum chamber should change approximately linearly.

[0091] refer to Figure 5 Specifically, the first step is to provide a training set for the system. This training set is obtained by conducting vacuum chamber experiments on standard samples with known leak rates at a specified pumping rate. This yields the relationship between the average bubble diameter, average bubble evolution rate, and leak rate for standard samples with different leak rates. The average bubble diameter and average evolution rate will serve as input vectors, while the leak rate value of the standard samples will be the expected output.

[0092] The actual output is then calculated, and a recursive method is used to return from the output node to the intermediate hidden layer to adjust the weights for optimization until the error meets the requirements, so that the neural network can realize the given input-output mapping relationship.

[0093] The data from the final actual detection application will still serve as the basis for optimizing the neural network.

[0094] Case 1: Using the above system to perform the pressure-holding leak test described in step 3 on a test piece with welding defects, first, a special foaming agent is applied to the area to be tested. A vacuum chamber is then used to cover the area. The pumping speed regulating valve connected to the vacuum chamber is opened, causing the pressure to drop to 500 mbar for pressure-holding leak testing. The pumping speed regulating valve is then closed, and after 10 seconds, the alarm device flashes. In the calculation module, the leak rate preset value Q... A 1×10 -5 Pa.m 3 The detected bubble generation rate was 2 bubbles / s, with a bubble diameter of 1.2 mm. The leak rate calculated by the calculation module was 9 × 10⁻⁶. -5 Pa.m 3 / s, leakage rate value exceeds Q A Therefore, the system is running well.

[0095] Case 2: The above system was used for rapid leak detection of pressure changes during the evacuation process described in step 4. The detection pressure was 500 mbar. A set of 100 test specimens with known leak rates at the detection pressure was used as the training set, covering a leak rate range of 3 × 10⁻⁶ mbar. -4 Pa.m 3 / s to 6×10 -5 Pa.m 3 / s. Using the above system, when the pressure inside the vacuum chamber decreases from atmospheric pressure to the detection pressure (500 mbar) at a rate of 100 mar / s, the average evolution and average diameter of bubbles in each test piece are measured. These values ​​are used as input vectors to a neural network, with the leak rate value as the expected output. The trained neural network is then applied to rapid leak detection under pressure changes, and the leak rate detected by the above system is 5 × 10⁻⁶. -5 Pa.m 3 For a sample size of / s, the calculated output value is 4.98 × 10⁻⁶. - 5 Pa.m 3 / s.

[0096] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0097] The above provides a detailed description of an intelligent vacuum chamber detection system provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the technical solutions and core ideas of this application. Those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for detecting a smart vacuum chamber, characterized in that, Including the following steps: Step 1: Deploy an intelligent vacuum chamber detection system on the area to be tested and ensure its airtightness; the intelligent vacuum chamber detection system includes a vacuum pumping system, an image acquisition system, and a leak rate analysis and alarm system; wherein, the vacuum pumping system includes a vacuum pump (11), a vacuum chamber (12), a pumping speed regulating valve (13), and a pumping air pipeline (14); the vacuum pump (11) is connected to the vacuum chamber (12) through the pumping air pipeline (14), and the pumping speed regulating valve (13) is installed on the pumping air pipeline (14); the vacuum pump (11) is used to extract the vacuum chamber ( 12) The vacuum chamber (12) is used to form a local sealed space in the area to be detected. The pumping speed regulating valve (13) is used to control the pumping speed and maintain the pressure of the vacuum pump (11). The image acquisition system is used to acquire images of the bubbles generated in the area to be detected in the vacuum chamber (12) and transmit the acquired images to the leak rate analysis and alarm system. The leak rate analysis and alarm system is connected to the image acquisition system. The leak rate analysis and alarm system is used to process and analyze the acquired images. If the leak rate value is greater than the alarm threshold range, an alarm message is issued. Step 2: Perform vacuuming operation on the vacuum chamber (12) corresponding to the area to be tested. After the test pressure is reached, close the pumping speed regulating valve and disconnect the vacuum pump from the vacuum chamber to maintain the pressure inside the vacuum chamber (12). Step 3: After the intelligent vacuum chamber detection system has maintained pressure for the first time, a pressure test is performed. The image acquisition system is controlled to acquire images of the air bubbles generated in the detection area inside the vacuum chamber (12) and transmits the acquired images to the leak rate analysis and alarm system. The leak rate analysis and alarm system is used to process and analyze the acquired images. If the leak rate value is greater than the alarm threshold range, it is determined that a leak has been found in the detection area and an alarm message is issued. Otherwise, the detection area is determined to be qualified. The leak rate analysis and alarm system is used for processing and analyzing the acquired images, including: The acquired image is sharpened to extract bubble edges, and the image is smoothed using a Gaussian function. The Laplacian-Gaussian operator is combined with the zero-crossing point of the second derivative to detect the image edges. By measuring the projected area of ​​the bubble and converting the area into the bubble diameter; whereby... The projected area of ​​the bubble can be obtained by measuring the pixels of the acquired image; whereby... S Represents the area of ​​the figure. x,y For pixel values, C The connected components within the extracted bubble edges. f It is a function related to the calibration scale inside the vacuum chamber; the measured area is obtained through... Converted to the equivalent bubble diameter D; pass The evolution time t of the bubble was calculated, where F Represents the number of frames. Fr Represents the frame rate of a CCD camera, according to The rate of bubble evolution is obtained by taking the reciprocal of time. v ; In vacuum chamber leak detection, through The leakage rate at each leakage point is calculated; where v avg The average rate of bubble generation, D avg The average diameter of the bubble. P b The gas pressure inside the bubble 。 2. The intelligent vacuum chamber detection method as described in claim 1, characterized in that, The step of performing a vacuuming operation on the vacuum chamber (12) for the corresponding area to be tested includes: After setting up the intelligent vacuum chamber detection system, apply a special foaming agent to the area to be tested, cover the area to be tested with the vacuum chamber, open the pumping speed regulating valve connected to the vacuum chamber, and create negative pressure in the vacuum chamber to adsorb the part to be tested in the area to be tested.

3. The intelligent vacuum chamber detection method as described in claim 1, characterized in that, After transmitting the acquired images to the leak rate analysis and alarm system, the system also includes: The leak rate analysis and alarm system is based on a BP neural network to achieve dynamic judgment of leak points and leak rates.

4. The intelligent vacuum chamber detection method as described in claim 3, characterized in that, The BP neural network is trained before use. The steps for training the BP neural network include: The initial parameters of the neural network are set, with the average bubble diameter and average evolution rate as the input vectors and the leak rate of the standard sample as the expected output. Under a specified pumping speed, the relationship between the average bubble diameter, average bubble evolution rate and leak rate in standard samples with different leak rates is obtained. The actual output is calculated, and a recursive method is used to return from the output node to the intermediate hidden layer to adjust the weights for optimization until the calculation error is less than a preset threshold, so that the neural network realizes the input-output mapping relationship.

5. The intelligent vacuum chamber detection method as described in claim 1, characterized in that, The vacuum chamber (12) is a transparent vacuum chamber made of acrylic glass material. The bottom of the vacuum chamber (12) is equipped with a replaceable rubber ring so as to form a sealed space with the test piece.

6. The intelligent vacuum chamber detection method as described in claim 1, characterized in that, The edge of the vacuum chamber (12) is engraved with scales for calibrating the size of bubbles in the acquired images.

7. The intelligent vacuum chamber detection method as described in claim 1, characterized in that, The image acquisition system includes an LED lighting source (21), a CCD camera (22), and an image acquisition card (23); the LED lighting source (21) is located on the side wall of the vacuum chamber (12), the CCD camera (22) is located on the top of the vacuum chamber (12), and the image acquisition card (23) is electrically connected to the CCD camera (22). The LED lighting source (21) is used to provide supplementary lighting for the CCD camera (22), the CCD camera (22) is used to take pictures and record the bubbles generated in the detection area inside the vacuum chamber, and the image acquisition card (23) is used to collect and store the image information recorded by the photograph as an acquired image and transmit it to the leak rate analysis and alarm system.

8. The intelligent vacuum chamber detection method as described in claim 1, characterized in that, The leakage rate analysis and alarm system includes a computing module (31) and an alarm flashing light (32). The computing module (31) is connected to the image acquisition card (23). The computing module (31) is used to process and analyze the acquired image. If the leakage rate value is greater than the alarm threshold range, the alarm flashing light (32) is controlled to flash.

Images