Gas meter leakage detection device and detection method thereof

Abstract

The invention relates to a gas meter leakage detection device and a detection method thereof. The detection device comprises a vacuum box, a vacuumizing device, a gas charging device, a helium detection device, a gas infrared thermal imaging device, a control valve, a pressure sensor and a pipeline for connection. The detection method comprises omnibearing detection from large-scale leakage to tiny leakage, and the accuracy of the gas meter under different leakage degrees is ensured. Large-scale leakage is recognized through pre-vacuumizing and pressure maintaining testing, then low-pressure helium and high-pressure helium are used for filling, the actual working environment is simulated to detect medium-scale and small leakage, and therefore the detection result is more comprehensive and reliable. By means of the high permeability characteristic of helium, high-precision detection equipment and an automatic control system are combined, and it is ensured that each step in the detection process reaches extremely high precision. Meanwhile, the gas infrared thermal imaging technology and the image enhancement technology are adopted, the leakage process can be completely recorded, and accurate positioning of leakage points is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of gas meter leakage detection, in particular to a gas meter leakage detection device and a detection method thereof. Background Art

[0002] In the production process of gas meters, ensuring their air tightness is a key link in maintaining product quality and safety. However, traditional air tightness detection technologies, such as water detection methods, have several shortcomings. The water detection method usually requires that the gas meter be filled with high-pressure gas and immersed in water, and the leakage is judged by observing the generation of bubbles. Although this method can intuitively identify the leakage, it is limited to qualitative evaluation and it is difficult to accurately quantify the severity of the leakage. In addition, the water detection method relies on manual observation, which is time-consuming and inefficient, and it is difficult to meet the needs of large-scale production for rapid detection. At the same time, the gas meter must be dried after the water detection, which not only increases the production cost, but also makes the production process more complicated. More importantly, the water detection method may cause corrosion or damage to the inside of the gas meter, thereby affecting its service life and safety performance.

[0003] In view of the limitations of the water inspection method, a Chinese patent (patent number: CN201821312070) in the existing technical literature of my country proposes a dry gas meter leak detection device. The technical solution includes filling the gas meter with high-pressure gas, and judging the leakage by monitoring the liquid level change of the U-type pressure gauge connected to the pipeline of the gas meter under test during the pressure holding stage, so as to realize the leakage detection of the gas meter in a dry environment. The core of the technical solution is to detect the leakage phenomenon through the pressure fluctuation during the pressure holding stage. However, in practical applications, pressure fluctuations are extremely sensitive to environmental factors, which may cause the technical solution to misjudge qualified workpieces as unqualified. In order to avoid misjudgment, the technical solution usually needs to add a certain amount of pressure change redundancy during the actual implementation process. However, this may cause the workpiece with a small leakage to fail to accumulate enough pressure change in a short time, making it difficult to accurately judge the existence of leakage. In addition, the scheme can only determine whether there is a leak in the gas meter under test, but cannot accurately locate the specific location of the leak, which is not conducive to subsequent maintenance and rectification work. Summary of the invention

[0004] In view of the problems existing in the prior art, the object of the present invention is to provide a technical solution of a gas meter leakage detection device and a detection method thereof.

[0005] The gas meter leakage detection device is characterized by comprising a vacuum box, a vacuum pumping device, a gas charging device, a helium detection device, a gas infrared thermal imaging device, a control valve, a pressure sensor and a pipe for connection; The vacuum box is connected to the vacuum pump, the gas charging device and the helium detection device through a pipeline to form a complete detection circuit; a positioning fixture for fixing the workpiece to be inspected is provided inside the vacuum box, and a gas infrared thermal imaging device is provided outside the vacuum box; The vacuum pump is composed of a box vacuum pump and a workpiece vacuum pump. The box vacuum pump is connected to the vacuum box through a pipeline to achieve the vacuum degree requirement of the vacuum box during the detection process; the workpiece vacuum pump is connected to the workpiece to be detected through a pipeline to achieve the vacuum degree requirement of the workpiece to be detected during the detection process; The gas charging device includes a nitrogen gas source, a helium gas source, a methane gas source and corresponding pressure regulating devices, and the gas charging device is connected to the vacuum box and the workpiece to be inspected at the same time; the nitrogen gas source plays the role of pipeline cleaning and protective gas during the inspection process; the helium gas source is used in the medium leak detection and micro leak detection process to provide helium as a tracer gas for the helium detection device; the methane gas source is used in the leak location process to provide methane as a tracer gas for the gas infrared thermal imaging device; the pressure regulating device is provided with a pressure reducing valve with controllable output pressure, and can adjust each gas source to output a suitable pressure according to the requirements of different inspection stages; The helium detection device is composed of a helium mass spectrometer leak detector and its support frame. The helium detection device is connected to a vacuum box. By utilizing the high permeability and easy-to-detect characteristics of helium, the change in helium concentration in the vacuum box during the leak detection process is analyzed, thereby accurately quantifying the amount of helium leaking from the workpiece to be inspected into the vacuum box, so as to determine whether there is a leak in the workpiece to be inspected. The gas infrared thermal imaging device is composed of a gas infrared thermal imager probe and an imager display. The gas infrared thermal imager probe is firmly installed on the observation window of the vacuum box to capture the thermal image of the workpiece to be inspected during the leakage process. By using image enhancement and instance segmentation technology, the methane gas plume in the thermal image during the leakage process can be separated from the background, and the contour boundary of the gas plume can be depicted. The leakage position of each point source can also be accurately located, and the leakage degree can be graded; the imager display is responsible for displaying the gas plume changes during the leakage process of the workpiece to be inspected, so that the technicians can more conveniently identify the leakage position and analyze the leakage cause, so as to carry out subsequent corresponding investigation and improvement; The control valve is used to control the opening and closing state of the pipeline and is connected in series in each pipeline; the control valve adopts a pneumatic control design to reduce the influence of the heat generated by the control valve during operation on the detection process; The pressure sensor uses two pressure transmitters, which are respectively installed on the connecting pipelines of the vacuum box and the workpiece to be inspected, and are used to detect the pressure inside the vacuum box and the pressure inside the workpiece to be inspected; The vacuum pumping device, the gas charging device, the helium detection device, the gas infrared thermal imaging device, the control valve and the pressure sensor are all electrically connected to the upper computer.

[0006] The detection method of the gas meter leakage detection device is characterized by comprising the following steps: S1, the test starts, the workpiece data is input, the workpiece to be tested is installed in the vacuum box, and the vacuum box door is closed, and then the nitrogen pipeline is cleaned; the test parameters of large leakage, medium leakage and slight leakage are configured; S2, perform a major leak test on the workpiece to be tested. If the test result is qualified, proceed to step S3; if the test result is unqualified, proceed to step S6; S3, perform leakage detection on the workpiece to be inspected. If the detection result is qualified, continue to step S4; if the detection result is unqualified, proceed to step S5; S4, perform micro-leakage detection on the workpiece to be inspected. If the detection result is qualified, continue to step S7; if the detection result is unqualified, go to step S5; S5, performing leakage location detection on the workpiece to be inspected, and recording the leakage position image, and then directly executing step S7; S6, clean the air pipeline, open the vacuum box door, take out the workpiece to be inspected, and do not execute step S7, and the inspection ends directly; S7, clean the nitrogen pipeline, open the vacuum box door, take out the workpiece to be inspected, and the inspection is completed.

[0007] The detection method of a gas meter leak detection device is characterized in that the nitrogen pipeline cleaning operation in step S1 is specifically performed as follows: a vacuum pumping device is used to perform a vacuum treatment on the vacuum box and the workpiece to be inspected; after the vacuum degree meets the predetermined requirements, the nitrogen gas source in the inflation device is started, and the pipeline is flushed with high-purity nitrogen, which can minimize the initial helium background concentration in the vacuum box, thereby reducing possible misjudgments in the subsequent helium detection process; at the same time, the residual air in the workpiece to be inspected is replaced to avoid the risk of explosion caused by the mixing of methane and air in the subsequent leak location process.

[0008] The detection method of a gas meter leak detection device is characterized in that step S2 performs a large leak detection operation, and the specific process is as follows: use the box evacuation pump to pre-vacuum the vacuum box until a predetermined vacuum degree is reached; then, use the pressure sensor to perform a pressure holding test on the workpiece to be inspected. If the workpiece to be inspected can maintain pressure within the specified time, continue to step S3; if the workpiece fails to maintain pressure within the specified time, the system will directly execute step S6, issue an alarm and prompt the user to remove the workpiece to be inspected.

[0009] The detection method of a gas meter leakage detection device is characterized in that the step S3 performs a mid-leak detection operation, and the specific process is as follows: a box vacuum pump is used to vacuum the vacuum box, and a workpiece vacuum pump is used to vacuum the workpiece to be inspected; when the vacuum box pressure and the workpiece pressure to be inspected reach the preset target pressure, the helium detection device will be activated to detect the helium background in the vacuum box and record the corresponding background measurement value; then, the charging device turns on the helium gas source to charge the workpiece to be inspected with low-pressure helium, and the helium detection device is activated again to detect the helium concentration in the vacuum box and record the corresponding leak rate measurement value; finally, the actual detected leak rate is determined by subtracting the background measurement value from the measured leak rate; if the actual detected leak rate is less than the standard requirement, step S4 is continued to be executed; if the actual detected leak rate is greater than the standard requirement, step S5 is executed to find the leak position.

[0010] The detection method of a gas meter leakage detection device is characterized in that step S4 performs a micro-leak detection operation, and the specific process is as follows: on the basis of medium leakage detection, the helium gas source in the charging device is further used to fill the workpiece to be inspected with high-pressure helium, the helium detection device is activated again, the helium concentration in the vacuum box is detected, and the corresponding leak rate measurement value is recorded. At this time, the actual detected leak rate is determined by subtracting the background measurement value from the measured leak rate. If the actual detected leak rate is less than the standard requirement, step S7 is continued to be executed; if the actual detected leak rate is greater than the standard requirement, step S5 is executed to find the leak position.

[0011] The detection method of a gas meter leakage detection device is characterized in that step S5 performs a leakage positioning operation, and the specific process is as follows: a vacuum box is evacuated using a box evacuation pump, and a workpiece evacuation pump is used to evacuate the workpiece at the same time; when the vacuum box pressure and the pressure of the workpiece to be inspected reach the preset target pressure, the methane gas source in the inflation device will be started to inject high-pressure methane gas into the workpiece to be inspected; at the same time, the gas infrared thermal imaging device is activated, and the gas infrared thermal imaging probe begins to record the infrared thermal image of the workpiece to be inspected during the leakage process; the infrared thermal image processed by the upper computer image enhancement and instance segmentation technology will display the development of the gas plume during the leakage process on the imager display, thereby accurately determining the specific location of the leakage.

[0012] The detection method of a gas meter leakage detection device is characterized in that step S7 performs a nitrogen pipeline cleaning operation, which is the same as the nitrogen pipeline cleaning process in step S1 but has a different purpose. The main purpose of step S7 is to replace the gas in the pipeline to prevent leakage of the detection gas when the workpiece to be inspected is subsequently unpacked and taken out.

[0013] The detection method of a gas meter leakage detection device is characterized by further comprising a gas meter leakage detection auxiliary system, wherein the gas meter leakage detection auxiliary system comprises a model library module, a data recording and analysis module, a prediction module and a technical improvement reminder module; Model library module: stores gas meter data of multiple specifications and dimensions, including the size, interface specifications, material properties, and historical test data of gas meters of different specifications, verifies whether the workpiece to be inspected is classified, and provides the required relevant information for other modules; Data acquisition and analysis module: acquires test data in real time and performs preliminary analysis, determines whether technical defects occur, and generates a test report based on a preset template; Prediction module: contains the optimal risk detection environment prediction model, which is updated with each test result, predicts the optimal risk detection environment configuration for each type of gas meter, and optimizes the test process; Technology Improvement Reminder Module: Generates technology improvement reminders.

[0014] The detection method of the gas meter leakage detection device is characterized in that the working process of the gas meter leakage detection auxiliary system at the beginning of the detection in step S1 includes the following steps: S101: Detection starts, obtaining input workpiece data, including size, interface specifications, and material properties, and verifying whether it is an existing classification of the model library module. If so, obtaining the workpiece number corresponding to the workpiece data; if it is verified to be an existing classification of a non-model library module, a new number is created, and the size, interface specifications, and material properties of the new product are entered; S102: The prediction module obtains the size, interface specification, material characteristics and number through the model library module, predicts the best risk detection environment for the numbered workpiece, and outputs it to the detection device to automatically configure the detection environment; The working process of the gas meter leakage detection auxiliary system at the end of the detection in step S7 includes the following steps: S701: After the detection is completed, the data acquisition and analysis module obtains the environmental parameters and detection results of the workpiece detection in real time through the leakage detection device. After the data is acquired and marked, it is stored in the data set of the corresponding category of workpieces in the model library module; the prediction module obtains new detection data through the model library module and automatically updates the model through incremental learning; S702: The data acquisition and analysis module determines the leakage situation according to the detection results. If there is no leakage after three detections, the process proceeds to S705; if there is leakage, the process proceeds to S703; S703: The data acquisition and analysis module acquires the leak-prone point coordinate group corresponding to the gas meter of the model library module, and determines whether there is a design defect. If so, the technical improvement reminder module is activated and the process goes to S704-1; if not, the process goes to S704-2; S704-1: The technology improvement reminder module generates a technology improvement reminder; S704-2: The data acquisition and analysis module enters the coordinates of the leakage point as new leakage point coordinates into the leakage point coordinate group in the model library module; S705: The data acquisition and analysis module generates a test report according to a preset template.

[0015] The present invention adopts helium detection technology, and utilizes the high permeability and easy-to-detect characteristics of helium to achieve accurate measurement and quantification of tiny leaks. At the same time, based on gas infrared thermal imaging and image enhancement technology, the present invention can fully record the leakage process and achieve accurate positioning of the leakage point. In addition, the present invention also reduces manual operations through highly automated control components and control methods, and achieves efficient detection of the gas tightness of the gas meter. Finally, the present invention also automatically predicts and sets detection parameters through an auxiliary detection system based on the above-mentioned detection device, provides an optimal experimental environment, monitors detection results, and issues technical improvement reminders in a timely manner, thereby optimizing the test process and improving test efficiency.

[0016] Beneficial effects of the present invention: (1) Using helium as the tracer gas and combining it with highly automated control components, accurate and efficient detection of the gas meter's air tightness is achieved; (2) The device uses helium as the tracer gas, taking advantage of its high permeability and easy detectability, so that even tiny leaks can be identified quickly and accurately, greatly improving detection efficiency and sensitivity compared to traditional methods; (3) The use of gas infrared thermal imaging technology and image enhancement technology can fully record the leakage process and accurately locate the leakage point; (4) Through the integrated control components, the entire detection process, including vacuuming, inflation, leak detection, leak location and other steps, has achieved a high degree of automation, which not only reduces the dependence on manual operation and reduces human errors, but also greatly improves work efficiency; (5) Learn and predict historical data, provide the best experimental environment, monitor and adjust the test process in real time, optimize the test process, and improve test efficiency and accuracy; (6) If leakage is detected at the same location multiple times, the system will initiate a technical improvement reminder, which can help discover design defects or manufacturing problems in the gas meter and avoid greater safety hazards. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a structural schematic diagram of a gas meter leakage detection device of the present invention; Figure 2 A flow chart of a gas meter leakage detection method according to the present invention; Figure 3 This is a diagram showing the control valve status during the nitrogen pipeline cleaning operation in the gas meter leak detection method. Figure 1 ; Figure 4 This is a diagram showing the control valve status during the nitrogen pipeline cleaning operation in the gas meter leak detection method. Figure 2 ; Figure 5 It is a schematic diagram of the control valve status in the large leak detection operation of the gas meter leak detection method; Figure 6 The control valve status diagram in the leak detection and micro-leak detection operation of the gas meter leak detection method is shown in FIG. Figure 1 ; Figure 7 The control valve status diagram in the leak detection and micro-leak detection operation of the gas meter leak detection method is shown in FIG. Figure 2 ; Figure 8 The control valve status diagram in the leak detection and micro-leak detection operation of the gas meter leak detection method is shown in FIG. Figure 3 ; Fig. 9 It is a schematic diagram of the control valve status in the leakage location operation of the gas meter leakage detection method; Fig.10 It is a schematic diagram of the control valve status in the air pipeline cleaning operation in the gas meter leakage detection method; Fig.11 A schematic diagram of a gas meter leakage detection auxiliary system module based on the above detection device; Fig.12 This is a working flow chart of the gas meter leakage detection auxiliary system based on the above detection device at the beginning of S1 detection; Fig.13 The following is a flowchart of the working process of the gas meter leakage detection auxiliary system based on the above detection device at the end of S7 detection. DETAILED DESCRIPTION

[0018] The present invention will be further described below in conjunction with the accompanying drawings: In view of the problems existing in the existing technical solutions, such as the difficulty in accurately quantifying the severity of the leakage, the detection process being easily affected by environmental factors, the high misjudgment rate in the case of small leaks, the difficulty in accurately determining the exact location of the leak, and the low detection efficiency, the present invention proposes a gas meter leakage detection device and a detection method thereof.

[0019] A gas meter leakage detection device comprises a vacuum box, a vacuum pumping device, a gas charging device, a helium detection device, a gas infrared thermal imaging device, and corresponding control valves, a pressure sensor and a pipeline for connection.

[0020] The vacuum box is connected to the vacuum pump, the gas charging device and the helium detection device through pipes, forming a complete detection circuit. The vacuum box is equipped with a positioning fixture for fixing the workpiece to be inspected, and the outside is equipped with a gas infrared thermal imaging device. During the detection process, the vacuum box serves as an external pressure-maintaining container for the workpiece to be inspected, and its door adopts a vertical pull-up design. This design is not only convenient for the installation of the workpiece to be inspected, but also can withstand large air pressure fluctuations and is not prone to deformation or damage, thereby ensuring stability during the detection process. In addition, a transparent observation window is provided on the wall of the vacuum box to meet the observation requirements of the gas infrared thermal imaging device.

[0021] The vacuum pump consists of a box vacuum pump and a workpiece vacuum pump. The box vacuum pump is connected to the vacuum box through a pipeline to achieve the vacuum degree requirement of the vacuum box during the inspection process. The workpiece vacuum pump is connected to the workpiece to be inspected through a pipeline to achieve the vacuum degree requirement of the workpiece to be inspected during the inspection process. To ensure safety, both the box vacuum pump and the workpiece vacuum pump use explosion-proof vacuum pumps, and the gases extracted during the vacuum pumping process are discharged to the waste gas recovery and treatment system.

[0022] The charging device includes a nitrogen gas source, a helium gas source, a methane gas source and a corresponding pressure regulating device. The charging device is connected to the vacuum box and the workpiece to be inspected at the same time, in order to provide the required gas source during the charging stage. Specifically, the nitrogen gas source plays the role of pipeline cleaning and protective gas during the detection process; the helium gas source is used in the medium leak detection and micro leak detection process to provide helium as a tracer gas for the helium detection device; the methane gas source is used in the leak location process to provide methane as a tracer gas for the gas infrared thermal imaging device. The pressure regulating device is equipped with a pressure reducing valve with controllable output pressure. Its function is to adjust each gas source to output a suitable pressure according to the needs of different detection stages.

[0023] The helium detection device is composed of a helium mass spectrometer leak detector and its support frame. The helium detection device is connected to the vacuum box. By utilizing the high permeability and easy detection characteristics of helium, the helium concentration change in the vacuum box during the leak detection process is analyzed, thereby accurately quantifying the amount of helium leakage from the workpiece to be inspected into the vacuum box to determine whether the workpiece to be inspected has leakage.

[0024] The gas infrared thermal imaging device consists of a gas infrared thermal imager probe and an imager display. The gas infrared thermal imager probe is firmly installed on the observation window of the vacuum box, and is designed to capture the thermal image of the workpiece to be inspected during the leakage process. Using image enhancement and instance segmentation technology, the methane plume in the thermal image during the leakage process can be separated from the background, and the contour boundary of the plume can be depicted. In addition, the leakage position of each point source can be accurately located, and the degree of leakage can be graded. The imager display is responsible for displaying the changes in the plume during the leakage process of the workpiece to be inspected, so that technicians can more easily identify the leakage location and analyze the cause of the leakage, so as to carry out subsequent corresponding investigations and improvements.

[0025] The control valve is used to control the opening and closing state of the pipeline. The control valve is always designed with pneumatic control, which can reduce the influence of the heat generated by the control valve during operation on the detection process.

[0026] The pressure sensor is specifically composed of two pressure transmitters, which are respectively installed on the connecting pipelines of the vacuum box and the workpiece to be inspected, and are used to detect the pressure inside the vacuum box and the pressure inside the workpiece to be inspected.

[0027] The vacuum device, the gas charging device, the helium detection device, the gas infrared thermal imaging device, the control valve and the pressure sensor are all electrically connected to the upper computer.

[0028] A detection method for a gas meter leakage detection device specifically comprises the following steps: S1, the detection starts, the workpiece data is input, the workpiece is installed in the vacuum box, and the vacuum box door is closed, and then the nitrogen pipeline is purged, and the detection parameters of large leakage, medium leakage and slight leakage are automatically configured; S2, perform a major leak test on the workpiece to be tested. If the test result is qualified, proceed to step S3; if the test result is unqualified, proceed to step S6; S3, perform leakage detection on the workpiece to be inspected. If the detection result is qualified, continue to step S4; if the detection result is unqualified, proceed to step S5; S4, perform micro-leakage detection on the workpiece to be inspected. If the detection result is qualified, continue to step S7; if the detection result is unqualified, go to step S5; S5, performing leakage location detection on the workpiece to be inspected, and recording the leakage position image, and then directly executing step S7; S6, clean the air pipeline, open the vacuum box door, take out the workpiece to be inspected, and do not execute step S7, and the inspection ends directly; S7, clean the nitrogen pipeline, open the vacuum box door, take out the workpiece to be inspected, and the inspection is completed.

[0029] Step S1: Cleaning of nitrogen pipelines. The specific process includes using a vacuum pump to vacuum the vacuum box and the workpiece to be inspected; after the vacuum degree meets the predetermined requirements, start the nitrogen gas source in the gas charging device and flush the pipeline with high-purity nitrogen. This is to minimize the initial helium background concentration in the vacuum box, thereby reducing possible misjudgments in the subsequent helium detection process; at the same time, replace the residual air in the workpiece to be inspected to avoid the risk of explosion caused by the mixing of methane and air in the subsequent leak location process.

[0030] Step S2 performs a large leak detection operation, and its specific process involves using a box evacuation pump to pre-evacuate the vacuum box until a predetermined vacuum degree is reached. Subsequently, a pressure sensor is used to perform a pressure holding test on the workpiece to be inspected. If the workpiece to be inspected can maintain pressure within the specified time, step S3 will be continued; if the workpiece to be inspected fails to maintain pressure within the specified time, the system will directly execute step S6, issue an alarm and prompt the user to remove the workpiece. In this technical solution, by introducing a large leak detection step, it is ensured that only after the workpiece passes the large leak detection will the subsequent medium leak and micro leak detection be continued. This measure effectively avoids invalid detection and improves the overall detection efficiency. When the workpiece to be inspected fails to pass the large leak detection stage, the system will automatically perform the air pipeline cleaning, and immediately issue an alarm and prompt the user to remove the workpiece. This instant feedback mechanism enables the user to quickly learn the test results and take corresponding treatment measures, thereby avoiding unnecessary waiting and waste of resources. If the workpiece to be inspected is identified as having a serious leak in the large leak detection stage, it will not enter the subsequent detection links, thereby avoiding unnecessary consumption of equipment and resources in these links.

[0031] Step S3 performs a medium leak detection operation, specifically including using a box evacuation pump to evacuate the vacuum box, and using the workpiece evacuation pump to evacuate the workpiece to be inspected. When the vacuum box pressure and the pressure of the workpiece to be inspected reach the preset target pressure, the helium detection device will be activated to detect the helium background in the vacuum box and record the corresponding background measurement value. Subsequently, the inflation device turns on the helium gas source to fill the workpiece with low-pressure helium, and the helium detection device is activated again to detect the helium concentration in the vacuum box and record the corresponding leak rate measurement value. Finally, the actual detected leak rate is determined by subtracting the background measurement value from the measured leak rate. If the actual detected leak rate is less than the standard requirement, continue to step S4; if the actual detected leak rate is greater than the standard requirement, execute step S5 to find the leak location. In this technical solution, it can be ensured that only those workpieces that perform well in the medium leak detection will enter the micro leak detection stage. For workpieces that fail the medium leakage test, leak location detection is directly carried out without entering the micro leakage detection stage. This can reduce excessive use and wear of the detection equipment and help extend the service life of the equipment.

[0032] Step S4 performs a micro-leak detection operation, which specifically includes further using the helium gas source in the inflation device to fill the workpiece with high-pressure helium on the basis of the previous medium leak detection. The helium detection device is activated again to detect the helium concentration in the vacuum box and record the corresponding leak rate measurement value. At this time, the actual detected leak rate is determined by subtracting the background measurement value from the measured leak rate. If the actual detected leak rate is less than the standard requirement, continue to step S7; if the actual detected leak rate is greater than the standard requirement, execute step S5 to find the leak location. In this technical solution, micro-leak detection is an important part of the detection process, which can capture tiny leaks in the workpiece and ensure that all workpieces to be inspected have undergone comprehensive inspections, whether they are large leaks, medium leaks or micro leaks, they have been strictly examined. Only workpieces that pass the micro-leak detection will enter the subsequent helium discharge step, which avoids misjudgment caused by undetected tiny leaks in the workpiece, thereby improving the accuracy of the entire detection process.

[0033] Step S5 performs the leakage locating operation, and the specific operation includes using the box evacuation pump to evacuate the vacuum box, and using the workpiece evacuation pump to evacuate the workpiece. When the vacuum box pressure and the pressure of the workpiece to be inspected reach the preset target pressure, the methane gas source in the inflation device will be started to inject high-pressure methane gas into the workpiece to be inspected. At the same time, the gas infrared thermal imaging device is activated, and the gas infrared thermal imager probe begins to record the infrared thermal image of the workpiece to be inspected during the leakage process. The infrared thermal image processed by the upper computer image enhancement and instance segmentation technology will display the development of the gas plume during the leakage process on the imager display, so as to accurately determine the specific location of the leak. In this technical solution, the leak locating detection process is a key link, which can accurately locate the specific location of the leak, which is helpful for subsequent analysis and rectification work.

[0034] Step S7 performs a nitrogen pipeline cleaning operation, and its specific operation is the same as the nitrogen pipeline cleaning process in step S1. However, unlike the purpose of step S1 to reduce the impact of the detection gas on the detection results, the main purpose of step S7 is to replace the gas in the pipeline to prevent the leakage of the detection gas when the workpiece is subsequently unpacked and taken out.

[0035] In this technical solution, the detection method of the gas meter leakage detection device includes all-round detection from large-scale leakage to small leakage, ensuring the accuracy of the gas meter under different leakage degrees. Large-scale leakage is identified by pre-vacuuming and pressure holding test, and then low-pressure and high-pressure helium are filled to simulate the actual working environment to detect medium-scale and small leakage, so that the detection results are more comprehensive and reliable. With the high permeability of helium, combined with high-precision detection equipment and automatic control system, it is ensured that each step in the detection process reaches extremely high accuracy, whether it is vacuum control, helium filling amount, or leakage rate detection, the results can be guaranteed to be accurate. At the same time, the use of gas infrared thermal imaging technology and image enhancement technology can fully record the leakage process and realize the precise positioning of the leakage point. By filling high-purity nitrogen to remove the residual helium background, the false alarm or pollution problem caused by residual helium in the equipment is effectively avoided, which not only ensures the long-term stable operation of the equipment, but also extends the service life of the equipment.

[0036] A detection method for a gas meter leakage detection device also includes a gas meter leakage detection auxiliary system, including a model library module, a data recording and analysis module, a prediction module, and a technical improvement reminder module. The gas meter leakage detection auxiliary system sets detection parameters, and the detection parameters include vacuum degree, inflation pressure, and detection time.

[0037] Model library module: stores gas meter data of multiple specifications and dimensions, including the size, interface specifications, material properties, historical test data, etc. of gas meters of different specifications, verifies whether the inspected workpiece is classified, and provides the required relevant information for other modules; Data acquisition and analysis module: acquires test data in real time and performs preliminary analysis, determines whether technical defects occur, and generates a test report based on a preset template; Prediction module: contains the optimal risk detection environment prediction model, which is updated with each test result, predicts the optimal risk detection environment configuration for each type of gas meter, and optimizes the test process; Technology Improvement Reminder Module: Generates technology improvement reminders. Embodiment 1

[0038] A gas meter leakage detection device comprises a vacuum box 1, a vacuum pumping device 2, a charging device 3, a helium detection device 4, a gas infrared thermal imaging device 5, a workpiece to be inspected 6, a control valve F1, a control valve F2, a control valve F3, a control valve F4, a control valve F5, a control valve F6, a control valve F7, a control valve F8, a control valve F9, a pressure sensor P1, a pressure sensor P2, a nitrogen pressure regulating device T1, a helium pressure regulating device T2, a methane pressure regulating device T3, a tee 3T1, a tee 3T2, a cross 4T1, a tee 4T2 and a pipeline for connection.

[0039] The vacuum device 2 is composed of a box vacuum pump 21 and a workpiece vacuum pump 22. The gas charging device 3 includes a nitrogen gas source 31 and a nitrogen pressure regulating device T1, a helium gas source 32 and a helium pressure regulating device T2, and a methane gas source 33 and a methane pressure regulating device T3. The gas infrared thermal imaging device 5 includes a gas infrared thermal imaging probe 51 and an imaging display 52.

[0040] The vacuum box 1 is connected to the box body evacuation pump 21 in the vacuum device 2 through a pipeline via a control valve F1; the vacuum box 1 is connected to the nitrogen pressure regulating device T1 and the nitrogen gas source 31 in the charging device 3 through a pipeline via a control valve F4 and a tee 3T1; the vacuum box 1 is connected to the helium detection device 4 through a pipeline via a control valve F3; in addition, the vacuum box 1 is connected to the air through a pipeline via a control valve F9 and a tee 3T2. A positioning fixture for fixing the workpiece 6 to be inspected is provided inside the vacuum box 1, and a gas infrared thermal imaging device 5 is equipped on the outside. The vacuum box 1 is also equipped with a pressure sensor P1 for detecting the pressure of the vacuum box 1. During the inspection process, the vacuum box 1 serves as an external pressure-maintaining container for the workpiece to be inspected, and its door adopts a vertical pull-up design, which is not only convenient for the installation of the workpiece to be inspected, but also can withstand large air pressure fluctuations and is not easily deformed or damaged, thereby ensuring stability during the inspection process. In addition, a transparent observation window is provided on the wall of the vacuum box 1 to meet the observation requirements of the gas infrared thermal imaging device 5.

[0041] The vacuum pump 2 is composed of a box vacuum pump 21 and a workpiece vacuum pump 22. The box vacuum pump 21 is connected to the vacuum box 1 through the control valve F1, and is used to achieve the vacuum degree requirement of the vacuum box 1 during the detection process. The workpiece vacuum pump 22 is connected to the workpiece 6 to be inspected through the control valve F2, and is used to achieve the vacuum degree requirement of the workpiece 6 to be inspected during the detection process. To ensure safety, both the box vacuum pump 21 and the workpiece vacuum pump 22 use explosion-proof vacuum pumps, and the gases extracted during the vacuum process are discharged to the waste gas recovery and treatment system.

[0042] The inflating device 3 includes a nitrogen gas source 31 and a nitrogen pressure regulating device T1, a helium gas source 32 and a helium pressure regulating device T2, and a methane gas source 33 and a methane pressure regulating device T3. Among them, the nitrogen gas source 31 and the nitrogen pressure regulating device T1 are connected to the vacuum box 1 through the three-way 3T1 and the control valve F4; the nitrogen gas source 31 and the nitrogen pressure regulating device T1 are connected to the workpiece 6 to be inspected through the three-way 3T1, the control valve F5, and the four-way 4T1. Specifically, the nitrogen gas source 31 plays the role of pipeline cleaning and as a protective gas during the detection process. The helium gas source 32 and the helium pressure regulating device T2 are connected to the workpiece 6 to be inspected through the control valve F6 and the four-way 4T1. The helium source is dedicated to micro-leak detection and provides the necessary helium. The methane gas source 33 and the methane pressure regulating device T3 are connected to the workpiece 6 to be inspected via the control valve F7, the three-way 4T2, and the four-way 4T1. The methane gas source 33 and the methane pressure regulating device T3 are connected to the air via the control valve F7, the three-way 4T2, the control valve F8, and the three-way 3T21. The methane gas source is used for the methane required in the leak location process. The nitrogen pressure regulating device T1, the helium pressure regulating device T2, and the methane pressure regulating device T3 are all equipped with a pressure reducing valve with controllable output pressure. Its function is to adjust each gas source to output a suitable pressure according to the requirements of different detection stages.

[0043] The interior of the helium detection device 4 is actually composed of a helium mass spectrometer leak detector and its support frame. The helium detection device 4 is connected to the vacuum box 1 via a control valve F3. By utilizing the high permeability and easy-to-detect characteristics of helium, the change in helium concentration in the vacuum box 1 during the leak detection process is analyzed, thereby accurately quantifying the amount of helium leakage from the workpiece 6 to the vacuum box 1 to determine whether the workpiece 6 to be inspected has a leak.

[0044] The gas infrared thermal imaging device 5 is composed of a gas infrared thermal imaging probe 51 and an imaging display 52. ​​The gas infrared thermal imaging probe 51 is firmly mounted on the observation window of the vacuum box 1, and is intended to capture the infrared thermal image of the workpiece 6 to be inspected during the leakage location process. The imaging display 52 is responsible for displaying the gas plume changes of the workpiece 6 to be inspected during the leakage location process, so that technicians can more easily identify the leakage location and analyze the cause of the leakage, and then implement corresponding troubleshooting and improvements.

[0045] The workpiece 6 to be inspected is connected to the workpiece evacuation pump 22 in the vacuum device 2 via the control valve F2. The workpiece 6 to be inspected is connected to the nitrogen pressure regulating device T1 and the nitrogen gas source 31 inside the gas charging device 3 via the cross-way 4T1, the control valve F5, the three-way 3T1. The workpiece 6 to be inspected is connected to the helium pressure regulating device T2 and the helium gas source 32 inside the gas charging device 3 via the cross-way 4T1, the control valve F6. The workpiece 6 to be inspected is connected to the methane pressure regulating device T3 and the methane gas source 33 inside the gas charging device 3 via the cross-way 4T1, the three-way 4T2, the control valve F7. The workpiece 6 to be inspected is connected to the air via the cross-way 4T1, the three-way 4T2, the control valve F8, the three-way 3T21. A pressure sensor P2 is installed between the workpiece 6 to be inspected and the cross-way 4T1 to detect the pressure inside the workpiece 6 to be inspected.

[0046] The vacuum device 2, the charging device 3, the helium detection device 4, the gas infrared thermal imaging device 5, the workpiece to be inspected 6, the control valve F1, the control valve F2, the control valve F3, the control valve F4, the control valve F5, the control valve F6, the control valve F7, the control valve F8, the control valve F9, the nitrogen pressure regulating device T1, the helium pressure regulating device T2, the methane pressure regulating device T3, the pressure sensor P1, and the pressure sensor P2 are all electrically connected to the upper computer. Embodiment 2

[0047] A detection method of a gas meter leakage detection device comprises the following steps: S1, the detection starts, the workpiece 6 to be inspected is installed in the vacuum box 1, and the vacuum box door is closed, and then the nitrogen pipeline is cleaned by the vacuum device 2 and the charging device 3, and the detection device automatically configures the detection parameters of large leakage, medium leakage and slight leakage through the auxiliary system; S2, perform a major leak test on the workpiece 6 to be tested. If the test result is qualified, proceed to step S3; if the test result is unqualified, proceed to step S6; S3, perform a leakage test on the workpiece 6 to be tested. If the test result is qualified, proceed to step S4; if the test result is unqualified, proceed to step S5; S4, perform micro-leakage detection on the workpiece 6 to be inspected. If the detection result is qualified, continue to step S7; if the detection result is unqualified, proceed to step S5; S5, performing leakage location detection on the workpiece 6 to be inspected, and recording the leakage position image, and then directly executing step S7; S6, clean the air pipeline, open the vacuum box door, take out the workpiece 6 to be inspected, and do not execute step S7 anymore, and the inspection ends directly.

[0048] S7, clean the nitrogen pipeline, open the vacuum box door, take out the workpiece 6 to be inspected, and the inspection is completed.

[0049] The specific control process of the nitrogen pipeline cleaning operation in step S1 includes: STEP 1: Open control valve F1 and control valve F2, close control valve F3, control valve F4, control valve F5, control valve F6, control valve F7, control valve F8, and control valve F9, start the box body evacuation pump 21 in the vacuum pumping device 2 to evacuate the vacuum box 1, and at the same time, use the workpiece evacuation pump 22 in the vacuum pumping device 2 to evacuate the workpiece 6 to be inspected. Figure 3 As shown; STEP2: When the pressure sensor P1 detects that the pressure of the vacuum box 1 is less than 30Pa, the control valve F1 is closed and the box body evacuation pump 21 is stopped; when the pressure sensor P2 detects that the pressure of the workpiece 6 to be inspected is less than 2000Pa, the control valve F2 is closed and the workpiece evacuation pump 22 is stopped; STEP3: Open control valves F4 and F5, close control valves F1, F2, F3, F6, F7, F8, and F9, start the nitrogen source 31 and nitrogen pressure regulating device T1 in the charging device 3, and use high-purity nitrogen to flush the pipeline. Figure 4 shown.

[0050] The above measures are intended to minimize the initial helium background concentration in the vacuum box 1, thereby reducing possible misjudgments during subsequent helium inspections; at the same time, the air remaining in the workpiece 6 to be inspected is replaced to avoid the risk of explosion caused by the mixing of methane and air during the subsequent leak location process.

[0051] Step S2 performs a major leak detection operation, and its specific control process includes: STEP 1: Open control valve F1, close control valves F2, F3, F4, F5, F6, F7, F8, and F9, and start the box evacuation pump 21 in the vacuum device 2 to evacuate the vacuum box 1. Figure 5 As shown; STEP2: When the pressure sensor P1 detects that the pressure of vacuum box 1 reaches the set maximum leak vacuum When the control valve F1 is closed, the box evacuation pump 21 is stopped; STEP3: Close all control valves, and the pressure sensor P2 performs a pressure-maintaining test on the workpiece 6 to be inspected; if the workpiece 6 to be inspected can maintain pressure within the specified time, continue to step S3; if the workpiece 6 to be inspected cannot maintain pressure within the specified time, the system will directly execute step S6, issue an alarm and prompt the user to remove the workpiece.

[0052] By introducing the large leak detection step, it is ensured that only after the workpiece passes the large leak detection, the subsequent medium leak and micro leak detection will be performed. This measure effectively avoids invalid detection and improves the overall detection efficiency. When the workpiece fails to pass the large leak detection stage, the system will automatically perform the air pipeline cleaning, and immediately issue an alarm and prompt the user to remove the workpiece. This instant feedback mechanism enables users to quickly learn the test results and take corresponding treatment measures, thereby avoiding unnecessary waiting and waste of resources. If the workpiece is identified as having a serious leak in the large leak detection stage, it will not enter the subsequent detection links, thereby avoiding unnecessary consumption of equipment and resources in these links.

[0053] Step S3 performs a leakage detection operation, and its specific control process includes: STEP1: Open control valve F2 and control valve F3, close control valve F1, control valve F4, control valve F5, control valve F6, control valve F7, control valve F8 and control valve F9, and use the workpiece vacuum pump 22 in the vacuum device 2 to vacuum the workpiece 6 to be inspected; when the pressure sensor P2 detects that the pressure of the workpiece 6 to be inspected is less than 2000Pa, close the control valve F2 and stop the workpiece vacuum pump 22; at the same time, start the helium detection device 4 to detect the helium background of the vacuum box 1, and record the corresponding background measurement value v 0 ,like Figure 6 As shown; STEP2: Open control valve F6, close control valves F1, F2, F3, F4, F5, F7, F8, and F9, and activate the helium gas source 32 and helium pressure regulating device T2 in the charging device to start charging the workpiece 6 to be inspected with low-pressure helium. Figure 7 As shown; STEP3: When the pressure sensor P2 detects that the pressure of the workpiece 6 to be inspected reaches the medium leakage inflation pressure At the same time, the control valve F6 is closed; at the same time, the control valve F3 is opened, and the helium detection device 4 is started again to detect the helium concentration of the vacuum box 1 and record the corresponding leakage rate measurement value v t ,like Figure 8 As shown; STEP4: At this point, the final actual leak rate v f It is the leak rate value measured by STEP3 v t Subtract the background value measured in STEP1 v 0 To determine, that is ; If the final actual leakage rate v fLess than 1×10 -6 Pa.m 3 / s, then continue to step S4; if the final actual leakage rate v f Greater than 1×10 -6 Pa.m 3 / s, then execute step S5.

[0054] This ensures that only those workpieces that pass the medium leak test will enter the micro leak test stage; for workpieces that fail the medium leak test, leak location testing will be carried out directly without entering the micro leak test stage. This can reduce excessive use and wear of the testing equipment and help extend the service life of the equipment.

[0055] Step S4 performs a micro-leakage detection operation, and its specific control process includes: STEP1: Open control valve F6, close control valves F1, F2, F3, F4, F5, F7, F8, and F9, and activate the helium source 32 in the charging device 3 and the helium pressure regulating device T2 to start charging the workpiece 6 to be inspected with high-pressure helium. Figure 7 As shown; STEP2: When the pressure sensor P2 detects that the pressure of the workpiece 6 to be inspected reaches the slight leakage inflation pressure At the same time, the control valve F6 is closed; at the same time, the control valve F3 is opened, and the helium detection device 4 is started again to detect the helium concentration of the vacuum box 1 and record the corresponding leakage rate measurement value ,like Figure 8 As shown; STEP3: At this point, the final actual leak rate The leak rate value measured by STEP2 Subtract the background measurement value measured in step S3 in the leak detection step STEP1 v 0 To determine, that is ; If the final actual leakage rate Less than 1×10 -6 Pa.m 3 / s, then continue to step S7; if the final actual leakage rate Greater than 1×10 -6 Pa.m 3 / s, then execute step S5.

[0056] Micro-leak detection is an important part of the inspection process. It can capture tiny leaks in the workpiece and ensure that all workpieces have been fully inspected. Whether it is a large leak, a medium leak or a micro leak, it has been strictly examined. Only workpieces that pass the micro-leak test will enter the subsequent helium discharge step, which avoids misjudgment due to undetected micro-leakage in the workpiece, thereby improving the accuracy of the entire inspection process.

[0057] Step S5 performs the leakage location operation, and its specific control process includes: STEP 1: Open control valve F1 and control valve F2, close control valve F3, control valve F4, control valve F5, control valve F6, control valve F7, control valve F8, and control valve F9, start the box body evacuation pump 21 in the vacuum pumping device 2 to evacuate the vacuum box 1, and at the same time, use the workpiece evacuation pump 22 in the vacuum pumping device 2 to evacuate the workpiece 6 to be inspected. Figure 3 As shown; STEP2: When the pressure sensor P1 detects that the pressure of the vacuum box 1 reaches the set vacuum degree (depending on the step of triggering the leakage, select the vacuum degree of large leakage, medium leakage or slight leakage) ), close the control valve F1 and stop the box evacuation pump 21; when the pressure sensor P2 detects that the pressure of the workpiece 6 to be inspected is less than 2000Pa, close the control valve F2 and stop the workpiece evacuation pump 22; STEP3: Open control valve F7, close control valves F1, F2, F3, F4, F5, F6, F8 and F9, and activate methane gas source 33 and methane pressure regulating device T3 in the charging device to start charging high-pressure methane into the workpiece 6 to be inspected. Fig. 9 As shown; STEP4: When the pressure sensor P2 detects that the pressure of the workpiece 6 to be inspected reaches the inflation pressure (select the inflation pressure of large leakage, medium leakage or slight leakage according to the step of triggering leakage ), close the control valve F7; at the same time, the gas infrared thermal imaging device 5 is activated, and the gas infrared thermal imaging probe 51 starts to record the infrared thermal image of the workpiece 6 to be inspected during the leakage process; the infrared thermal image processed by the host computer image enhancement and instance segmentation technology will display the development of the gas plume during the leakage process on the imager display 52, so as to accurately determine the specific location of the leakage.

[0058] The leak location detection process is a key link, which can accurately locate the specific location of the leak and help subsequent analysis and rectification work.

[0059] Step S6 performs a nitrogen pipeline cleaning operation, and its specific control process includes: Open control valves F8 and F9, and close control valves F1, F2, F3, F4, F5, F6, and F7. Then wait for the indications of pressure sensors P1 and P2 to return to normal atmospheric pressure before unpacking and taking out the workpiece to be inspected. Fig.10 shown.

[0060] Step S7 executes a nitrogen pipeline cleaning process, and its specific operation is the same as the nitrogen pipeline cleaning process in step S1; the difference is that the nitrogen pipeline cleaning operation in step S1 is intended to reduce the impact of the detection gas on the detection results, while the main purpose of step S7 is to replace the gas in the pipeline with nitrogen to prevent leakage of the detection gas when unpacking and taking out the workpiece.

[0061] In the present invention, the detection method of the gas meter leakage detection device includes all-round detection from large-scale leakage to small leakage, ensuring the accuracy of the gas meter under different leakage degrees. Large-scale leakage is identified by pre-vacuuming and pressure holding test, and then low-pressure and high-pressure helium are filled to simulate the actual working environment to detect medium-scale and small leakage, so that the detection results are more comprehensive and reliable. With the high permeability of helium, combined with high-precision detection equipment and automatic control system, it is ensured that each step in the detection process reaches extremely high accuracy, whether it is vacuum control, helium filling amount, or leakage rate detection, the results can be guaranteed to be accurate. At the same time, the use of gas infrared thermal imaging technology and image enhancement technology can fully record the leakage process and realize the precise positioning of the leakage point. By filling high-purity nitrogen to remove the residual helium background, the false alarm or pollution problem caused by the residual helium of the equipment is effectively avoided, which not only ensures the long-term stable operation of the equipment, but also extends the service life of the equipment.

[0062] The detection method of the gas meter leakage detection device also includes a leakage detection auxiliary system based on the above detection device, which is used to optimize the detection process of the gas meter. The auxiliary system includes a model library module, a data recording and analysis module, a prediction module, and a technical improvement reminder module.

[0063] Among them, the model library module is connected with the data acquisition and analysis module and the prediction module, and the data acquisition and analysis module is connected with the technology improvement reminder module.

[0064] Model library module stores various specifications of workpiece GM i , where i is the number of each workpiece.

[0065] In the model library module, workpiece data is divided into various categories according to size, interface specifications, and material properties, and each type of workpiece data is numbered and classified for easy storage and training.

[0066] The specific division method of workpiece number i (i=1, 2, 3...n) is: According to the size, interface specifications and material properties, the workpieces are divided into various types. Workpieces with the same size, interface specifications and material properties are grouped together, and the various workpieces are marked with numbers 1, 2, 3...n in sequence.

[0067] Specifically, the meter body data and historical test data of various types of gas meters are stored together in the model library module according to the category number.

[0068] Preferably, the workpiece GM i The data set includes the following multi-dimensional arrays and corresponding parameter data: Size i = [L, W, H], where L, W, and H are the length, width, and height of the gas meter, respectively, in centimeters; Interface Specification i = [I, S], where I and S are the interface type and interface size respectively; Material properties i = [M, σ], where M and σ are material type and material tensile strength value respectively; Historical test data

[0069] in, They are the vacuum degree (unit: Pa), inflation pressure (unit: Pa), inspection time (unit: seconds), inflation time (unit: seconds), decompression time (unit: seconds), and inspection result at the time of the nth inspection. i_n The values ​​include large leakage (marked as 1), medium leakage (marked as 2), slight leakage (marked as 3), and no leakage (marked as 4); Coordinate group D of leak-prone points i , D i It is composed of the coordinates of the leakage points recorded when leakage occurred in the above historical tests.

[0070] The model library module is also used to verify whether the artifact belongs to an existing classification, as follows: At the beginning of the inspection, the input workpiece data (size, interface specifications, material properties) is obtained to verify whether it is an existing classification of the model library module. If so, the workpiece number corresponding to the workpiece data is obtained; if it is verified to be an existing classification of a non-model library module, a new number is created and the size, interface specifications, and material properties of the new product are entered.

[0071] The prediction module is used to obtain the workpiece data of the corresponding number through the model library module, namely the size, interface specification, material characteristics, and number, and predict the optimal risk detection environment OptEnv of the workpiece with the number through the trained optimal risk detection environment prediction model. i , and output to the leak detection device to automatically configure the detection environment.

[0072] Specifically, the output optimal risk detection environment data is: ,in, The predicted vacuum degree for large leak detection, the predicted inflation pressure for large leak detection, the predicted detection time for large leak detection, the predicted inflation time for large leak detection, and the predicted depressurization time for large leak detection are; the predicted vacuum degree for medium leak detection, the predicted inflation pressure for medium leak detection, the predicted detection time for medium leak detection, the predicted inflation time for medium leak detection, and the predicted depressurization time for medium leak detection are; It is the predicted vacuum degree of micro-leak detection, the predicted inflation pressure of micro-leak detection, the predicted detection time of micro-leak detection, the predicted filling time of micro-leak detection, and the predicted depressurization time of micro-leak detection.

[0073] The training process of the optimal risk detection environment prediction model is as follows.

[0074] One possible optimization method is to forward propagate the model in small batches to calculate the loss of each batch, optimize the model parameters according to the loss function, and gradually adjust the model to improve the prediction performance. The loss function is , where Loss is the total loss value; TimeLoss is the time loss; SafeLoss is the safety loss; RateLoss is the rate loss; LeakLoss is the leakage loss; They are the weight coefficients of TimeLoss, SafeLoss, RateLoss, and LeakLoss respectively.

[0075] The above loss calculations ensure that the output meets the following goals at the same time, and can generate the most leak-prone detection environment parameters for this model of gas meter without damaging the meter body due to excessive vacuum or excessive pressure filling and pressure relief rate: (1) The time loss is calculated by the mean square error to minimize the prediction detection time; (2) Safety loss ,in, is the pressure difference between inside and outside of the table, and the vacuum degree V of the test data is j , inflation pressure P j The absolute value of the difference is calculated, N represents the number of samples in the mini-batch, j represents the jth sample in the current mini-batch, and the safety threshold Calculated by the following formula: ,in, is the tensile strength of the material, is the tensile strength of the connection, is the material safety factor, is the safety factor of the connection port, The maximum pressure difference that the internal metering module of the gas meter can withstand; (3) Rate loss The calculation is used to avoid damage to the workpiece due to excessive filling and release rate, where the filling rate is determined by The pressure relief rate is calculated from Calculated, is the pressure difference between the initial pressure of the charge and discharge and the target pressure, N is the number of small batch samples, j is the jth sample in the current small batch, The preset maximum safe speed; (4) Leakage loss It is used to improve the accuracy of leak identification, that is, to prioritize the risk test environment where leaks are more likely to occur. j represents the actual label of the jth test data sample, indicating whether leakage occurs (y j =1 indicates leakage occurs, y j =0 means no leakage occurs); represents the model prediction probability of the jth test data sample, ranging from [0,1]; N represents the number of mini-batch samples.

[0076] One possible training method is to first preprocess the data, extract features and normalize them, then use the random forest algorithm for training, integrate the results of multiple decision trees, and continuously optimize to minimize the total loss. During the training process, use random sampling methods to ensure the diversity of training data and avoid overfitting. Use cross-validation and other methods to evaluate the trained model to ensure that it has high accuracy and generalization ability.

[0077] The data acquisition and analysis module is used to receive test data in real time and perform preliminary analysis, and generate a test report based on a preset template. The specific acquisition and analysis methods are as follows: The new detection environmental parameters and detection results are obtained through the gas meter leak detection device, including the vacuum degree of the large leak detection, the inflation pressure of the large leak detection, the detection time of the large leak detection, the pressure charging time of the large leak detection, the pressure relief time of the large leak detection and the detection result of the large leak detection, which are marked as ; The vacuum degree of the middle leak detection, the inflation pressure of the middle leak detection, the detection time of the middle leak detection, the pressure charging time of the middle leak detection, the pressure relief time of the middle leak detection and the detection result of the middle leak detection are marked as ; The vacuum degree of micro-leak detection, the inflation pressure of micro-leak detection, the detection time of micro-leak detection, the pressure-charging time of micro-leak detection, the pressure-releasing time of micro-leak detection and the detection result of micro-leak detection are marked as , where i=1, 2, 3...n, i represents the workpiece number. If no medium leakage or micro leakage detection occurs, the experimental data will not be recorded. After the data is marked, it is stored in the model library module corresponding to the category of workpiece GM i in the data set; The data acquisition and analysis module determines the leakage situation based on the detection results.

[0078] Data acquisition and analysis module obtains the model library module corresponding to the gas meter GM i The coordinate group D of the leak-prone points i , calculate the new leakage point coordinates and the historical leakage point coordinate group D i The Euclidean distance of each point. If there is a distance less than or equal to the preset distance threshold, it is judged that the leakage parts overlap, there is a design defect, and the technical improvement reminder module is activated; if all distances are greater than the preset distance threshold, it is judged as a new leak point, and the coordinates are entered into the leak point coordinate group D i .

[0079] The technology improvement reminder module is used to generate technology improvement reminders.

[0080] like Fig.12 As shown, specifically, the working process of the leakage detection auxiliary system at the beginning of S1 detection includes the following steps: S101: Detection starts, obtaining input workpiece data (size, interface specifications, material properties), verifying whether it is an existing classification of the model library module, and if so, obtaining the workpiece number corresponding to the workpiece data; if it is verified to be an existing classification of a non-model library module, a new number is created, and the size, interface specifications, and material properties of the new product are entered; S102: The prediction module obtains the size, interface specifications, material characteristics and number through the model library module, and predicts the optimal risk detection environment OptEnv for the numbered workpiece i , and output to the leak detection device to automatically configure the detection environment.

[0081] like Fig.13 Specifically, the working process of the leakage detection auxiliary system at the end of S7 detection includes the following steps: S701: After the detection is completed, the data acquisition and analysis module obtains the environmental parameters and detection results of the workpiece detection in real time through the leakage detection device. After the data is acquired and marked, it is stored in the corresponding category of the workpiece GM of the model library module i The prediction module obtains new detection data through the model library module and automatically updates the model through incremental learning; S702: The data acquisition and analysis module determines the leakage situation according to the detection results. If there is no leakage after three detections, the process proceeds to S705; if there is leakage, the process proceeds to S703; S703: The data acquisition and analysis module acquires the gas meter GM corresponding to the model library module i The coordinate group D of the leak-prone points i , it is determined whether there is a design defect, if so, the technical improvement reminder module is activated and the process goes to S704-1; if not, the process goes to S704-2; S704-1: The technology improvement reminder module generates a technology improvement reminder; S704-2: The data acquisition and analysis module enters the coordinates of the leakage point as the new leakage point coordinates into the leakage point coordinate group D in the model library module i ; S705: The data acquisition and analysis module generates a test report according to a preset template.

[0082] The specific description of the present invention in the above embodiments is only used to further illustrate the present invention and cannot be understood as limiting the scope of protection of the present invention. Technical engineers in this field may make some non-essential improvements and adjustments to the present invention based on the contents of the above invention, which fall within the scope of protection of the present invention.

Claims

1. A gas meter leakage detection device, characterized in that It includes a vacuum box, a vacuum pumping device, a gas charging device, a helium detection device, a gas infrared thermal imaging device, a control valve, a pressure sensor and pipes for connection; The vacuum box is connected to the vacuum pump, the gas charging device and the helium detection device through a pipeline to form a complete detection circuit; a positioning fixture for fixing the workpiece to be inspected is provided inside the vacuum box, and a gas infrared thermal imaging device is provided outside the vacuum box; The vacuum pump is composed of a box vacuum pump and a workpiece vacuum pump. The box vacuum pump is connected to the vacuum box through a pipeline to achieve the vacuum degree requirement of the vacuum box during the detection process; the workpiece vacuum pump is connected to the workpiece to be detected through a pipeline to achieve the vacuum degree requirement of the workpiece to be detected during the detection process; The gas charging device includes a nitrogen gas source, a helium gas source, a methane gas source and corresponding pressure regulating devices, and the gas charging device is connected to the vacuum box and the workpiece to be inspected at the same time; the nitrogen gas source plays the role of pipeline cleaning and protective gas during the inspection process; the helium gas source is used in the medium leak detection and micro leak detection process to provide helium as a tracer gas for the helium detection device; the methane gas source is used in the leak location process to provide methane as a tracer gas for the gas infrared thermal imaging device; the pressure regulating device is provided with a pressure reducing valve with controllable output pressure, and can adjust each gas source to output a suitable pressure according to the requirements of different inspection stages; The helium detection device is composed of a helium mass spectrometer leak detector and its support frame. The helium detection device is connected to a vacuum box. By utilizing the high permeability and easy-to-detect characteristics of helium, the change in helium concentration in the vacuum box during the leak detection process is analyzed, thereby accurately quantifying the amount of helium leaking from the workpiece to be inspected into the vacuum box, so as to determine whether there is a leak in the workpiece to be inspected. The gas infrared thermal imaging device is composed of a gas infrared thermal imager probe and an imager display. The gas infrared thermal imager probe is firmly installed on the observation window of the vacuum box to capture the thermal image of the workpiece to be inspected during the leakage process. By using image enhancement and instance segmentation technology, the methane gas plume in the thermal image during the leakage process can be separated from the background, and the contour boundary of the gas plume can be depicted. The leakage position of each point source can also be accurately located, and the leakage degree can be graded; the imager display is responsible for displaying the gas plume changes during the leakage process of the workpiece to be inspected, so that the technicians can more conveniently identify the leakage position and analyze the leakage cause, so as to carry out subsequent corresponding investigation and improvement; The control valve is used to control the opening and closing state of the pipeline and is connected in series in each pipeline; the control valve adopts a pneumatic control design to reduce the influence of the heat generated by the control valve during operation on the detection process; The pressure sensor uses two pressure transmitters, which are respectively installed on the connecting pipelines of the vacuum box and the workpiece to be inspected, and are used to detect the pressure inside the vacuum box and the pressure inside the workpiece to be inspected; The vacuum pumping device, the gas charging device, the helium detection device, the gas infrared thermal imaging device, the control valve and the pressure sensor are all electrically connected to the upper computer.

2. A detection method for a gas meter leakage detection device as claimed in claim 1, characterized in that The following steps are involved: S1, the test starts, the workpiece data is input, the workpiece to be tested is installed in the vacuum box, and the vacuum box door is closed, and then the nitrogen pipeline is cleaned; the test parameters of large leakage, medium leakage and slight leakage are configured; S2, perform a major leak test on the workpiece to be tested. If the test result is qualified, proceed to step S3; if the test result is unqualified, proceed to step S6; S3, perform leakage detection on the workpiece to be inspected. If the detection result is qualified, continue to step S4; if the detection result is unqualified, proceed to step S5; S4, perform micro-leakage detection on the workpiece to be inspected. If the detection result is qualified, continue to step S7; if the detection result is unqualified, go to step S5; S5, performing leakage location detection on the workpiece to be inspected, and recording the leakage position image, and then directly executing step S7; S6, clean the air pipeline, open the vacuum box door, take out the workpiece to be inspected, and do not execute step S7, and the inspection ends directly; S7, clean the nitrogen pipeline, open the vacuum box door, take out the workpiece to be inspected, and the inspection is completed.

3. A detection method for a gas meter leakage detection device according to claim 2, characterized in that The specific process of the nitrogen pipeline cleaning operation in step S1 is as follows: the vacuum box and the workpiece to be inspected are vacuumed by using a vacuum pumping device; after the vacuum degree meets the predetermined requirement, the nitrogen gas source in the inflation device is started, and the pipeline is flushed with high-purity nitrogen, which can minimize the initial helium background concentration in the vacuum box, thereby reducing possible misjudgments in the subsequent helium detection process; at the same time, the residual air in the workpiece to be inspected is replaced to avoid the risk of explosion caused by the mixing of methane and air in the subsequent leak location process.

4. A detection method for a gas meter leakage detection device according to claim 2, characterized in that The step S2 performs a large leak detection operation, and the specific process is as follows: the vacuum box is pre-evacuated using the box evacuation pump until a predetermined vacuum degree is reached; then, the pressure sensor is used to perform a pressure holding test on the workpiece to be inspected, and if the workpiece to be inspected can maintain pressure within a specified time, step S3 is continued; If the workpiece fails to maintain pressure within the specified time, the system will directly execute step S6, sound an alarm and prompt the user to remove the workpiece to be inspected.

5. A detection method for a gas meter leakage detection device according to claim 2, characterized in that The step S3 performs the leakage detection operation, and the specific process is as follows: the vacuum box is evacuated by the box evacuation pump, and the workpiece to be inspected is evacuated by the workpiece evacuation pump; when the pressure of the vacuum box and the pressure of the workpiece to be inspected reach the preset target pressure, the helium detection device will be activated to detect the helium background in the vacuum box and record the corresponding background measurement value; then, the charging device turns on the helium gas source to charge the workpiece to be inspected with low-pressure helium, and the helium detection device is activated again to detect the helium concentration in the vacuum box and record the corresponding leakage rate measurement value; Ultimately, the actual detected leak rate is determined by subtracting the background measurement from the measured leak rate; If the actually detected leakage rate is less than the standard requirement, then continue to execute step S4; if the actually detected leakage rate is greater than the standard requirement, then execute step S5 to find the leakage location.

6. A detection method for a gas meter leakage detection device according to claim 2, characterized in that The step S4 performs a micro-leak detection operation, and the specific process is as follows: on the basis of the medium leak detection, the helium source in the charging device is further used to charge the workpiece to be inspected with high-pressure helium, the helium detection device is activated again, the helium concentration in the vacuum box is detected, and the corresponding leak rate measurement value is recorded. At this time, the actual detected leak rate is determined by subtracting the background measurement value from the measured leak rate. If the actual detected leak rate is less than the standard requirement, step S7 is continued; If the actually detected leakage rate is greater than the standard requirement, step S5 is executed to find the leakage location.

7. A detection method for a gas meter leakage detection device according to claim 2, characterized in that The step S5 performs the leakage locating operation, and the specific process is as follows: the vacuum box is evacuated by a box evacuation pump, and the workpiece is evacuated by a workpiece evacuation pump; when the pressure of the vacuum box and the pressure of the workpiece to be inspected reach the preset target pressure, the methane gas source in the inflation device will be started to inject high-pressure methane gas into the workpiece to be inspected; at the same time, the gas infrared thermal imaging device is activated, and the gas infrared thermal imaging probe begins to record the infrared thermal image of the workpiece to be inspected during the leakage process; the infrared thermal image processed by the upper computer image enhancement and instance segmentation technology will display the development of the gas plume during the leakage process on the imager display, thereby accurately determining the specific location of the leakage.

8. A detection method for a gas meter leakage detection device according to claim 2, characterized in that The step S7 performs a nitrogen pipeline cleaning operation, which is the same as the nitrogen pipeline cleaning process in the step S1 but has a different purpose. The main purpose of step S7 is to replace the gas in the pipeline to prevent leakage of the detection gas when the workpiece to be inspected is subsequently unpacked and taken out.

9. A detection method for a gas meter leakage detection device according to claim 2, characterized in that It also includes a gas meter leakage detection auxiliary system, which includes a model library module, a data recording and analysis module, a prediction module and a technical improvement reminder module; Model library module: stores gas meter data of multiple specifications and dimensions, including the size, interface specifications, material properties, and historical test data of gas meters of different specifications, verifies whether the workpiece to be inspected is classified, and provides the required relevant information for other modules; Data acquisition and analysis module: acquires test data in real time and performs preliminary analysis, determines whether technical defects occur, and generates a test report based on a preset template; Prediction module: contains the optimal risk detection environment prediction model, which is updated with each test result, predicts the optimal risk detection environment configuration for each type of gas meter, and optimizes the test process; Technology Improvement Reminder Module: Generates technology improvement reminders.

10. A detection method for a gas meter leakage detection device according to claim 9, characterized in that The working process of the gas meter leakage detection auxiliary system at the beginning of the detection in step S1 includes the following steps: S101: Detection starts, obtaining input workpiece data, including size, interface specifications, and material properties, and verifying whether it is an existing classification of the model library module. If so, obtaining the workpiece number corresponding to the workpiece data; if it is verified to be an existing classification of a non-model library module, a new number is created, and the size, interface specifications, and material properties of the new product are entered; S102: The prediction module obtains the size, interface specification, material characteristics and number through the model library module, predicts the best risk detection environment for the numbered workpiece, and outputs it to the detection device to automatically configure the detection environment; The working process of the gas meter leakage detection auxiliary system at the end of the detection in step S7 includes the following steps: S701: After the detection is completed, the data acquisition and analysis module obtains the environmental parameters and detection results of the workpiece detection in real time through the leakage detection device. After the data is acquired and marked, it is stored in the data set of the corresponding category of workpieces in the model library module; the prediction module obtains new detection data through the model library module and automatically updates the model through incremental learning; S702: The data acquisition and analysis module determines the leakage situation according to the detection results. If there is no leakage after three detections, the process proceeds to S705; if there is leakage, the process proceeds to S703; S703: The data acquisition and analysis module acquires the leak-prone point coordinate group corresponding to the gas meter of the model library module, and determines whether there is a design defect. If so, the technical improvement reminder module is activated and the process goes to S704-1; if not, the process goes to S704-2; S704-1: The technology improvement reminder module generates a technology improvement reminder; S704-2: The data acquisition and analysis module enters the coordinates of the leakage point as new leakage point coordinates into the leakage point coordinate group in the model library module; S705: The data acquisition and analysis module generates a test report according to a preset template.

Images