Liquefied natural gas unloading pipe seal gasket judgment and disposal system

Abstract

A kind of liquefied natural gas unloading pipe seal pad judgment waste system, comprising: mobile module, prompt module, seal pad containing module, seal pad judging waste module;Mobile module carries prompt module, seal pad containing module to reach detection site, seal pad containing module is internally provided with light-tight upper light sensing cavity, light-tight lower light sensing cavity, there is annular light channel for light to pass between upper light sensing cavity and lower light sensing cavity, seal pad can be embedded in annular light channel, first light intensity sensor senses the brightness in lower light sensing cavity and generates first light intensity value, second light intensity sensor senses the brightness in upper light sensing cavity and generates second light intensity value, seal pad judging waste module calculates light intensity attenuation rate according to predetermined formula, and light intensity attenuation rate is compared with preset light intensity attenuation threshold value, when comparing that calculated light intensity attenuation rate is greater than preset light intensity attenuation threshold value, it is judged that current detected seal pad is scrapped.

Description

Technical fields:

[0001] This invention relates to the field of gasket testing technology for liquefied natural gas (LNG), and particularly to a LNG unloading pipe gasket rejection system. Background technology:

[0002] As a clean energy source, liquefied natural gas (LNG) is accounting for an increasing share of the global energy market year by year. LNG storage and transportation require extremely high safety standards, especially during the unloading process, where the sealing performance of the gaskets is crucial to ensuring the overall safety of the system.

[0003] Regarding the sealing performance testing of gaskets, different technical solutions are presented in the prior art: Patent publication number CN210665270U describes a performance testing device for composite gaskets, including a base, upper support plate, first support plate, second support plate, gasket loading device, temperature control device for adjusting the ambient temperature of the gasket testing environment, leakage rate detection device for testing the sealing performance of the gasket, and controller; Patent publication number CN111174715A describes a device and method for detecting the thickness of a turbocharger gasket, which includes a mounting base plate, positioning mechanism, in-situ sensor, displacement sensor assembly, etc. A laser displacement sensor detects the single-sided distance of the corresponding surface of the turbocharger gasket under test, a data transmission unit transmits this single-sided distance, and a data processing unit receives and processes the single-sided distance to obtain the actual thickness of the turbocharger gasket under test, and stores the thickness of the turbocharger gasket under test; Patent publication number CN210321676U describes a continuous testing mechanism for high wear-resistant gaskets. It mainly consists of a support rod and multiple components connected to it. A support block is fixedly connected to the top of the support rod, and the support block is fixedly connected to the middle of the cross plate of the U-shaped plate. One end of the U-shaped plate is connected to a square plate. An electric push rod is fixed on the top of the square plate for pushing the detection process. In addition, a detection mechanism is provided on the U-shaped plate. The automatic gasket warpage detection device with patent publication number CN210293106U includes a machine base. The gasket to be tested is transported to the detection device through a first conveyor belt. Next to the detection device is a second conveyor belt. The detection device contains a carrier platform and a third conveyor belt for continuing to transport the gasket forward to a fourth conveyor belt. An infrared scanner is installed on the top of the machine base for scanning and detecting the gasket during the transport process. The visual inspection device and method for combined gaskets with patent publication number CN117269184A include an automatic feeding mechanism, a rotating mechanism, a posture adjustment mechanism, a camera mechanism, a control system, and a unloading mechanism. It detects defects on each surface of the combined gasket through fully automatic feeding and unloading.

[0004] The aforementioned patents are all testing solutions for newly manufactured gaskets, covering multiple aspects such as gasket performance, thickness, assembly status, wear resistance, warpage, and appearance defects. However, they all have the following technical problems: they are only applicable to specific types or specifications of gaskets, and may not be able to effectively test other types or specifications of gaskets. For example, they cannot effectively test sealing gaskets used in liquefied natural gas. Summary of the Invention:

[0005] In view of this, the present invention discloses a liquefied natural gas unloading pipe sealing gasket rejection system.

[0006] A liquefied natural gas (LNG) unloading pipe gasket rejection system includes: a moving module, a notification module, a gasket receiving module, and a gasket rejection module. The notification module and the gasket receiving module are mounted on the moving module, which carries them. The gasket rejection module is electrically connected to the notification module and the gasket receiving module. The gasket receiving module receives and positions the gasket. Inside the gasket receiving module are an opaque upper light-sensing cavity and an opaque lower light-sensing cavity. The lower light-sensing cavity contains a strong light source and a first light intensity sensor. The upper light-sensing cavity contains... A second light intensity sensor is provided. An annular light channel for light to pass through is located between the upper and lower light-sensing cavities. The shape and size of the annular light channel are adapted to the size of the sealing gasket, and the sealing gasket can be embedded within the annular light channel. The sealing gasket rejection module is electrically connected to the first and second light intensity sensors. The first light intensity sensor senses the brightness in the lower light-sensing cavity and generates a corresponding first light intensity value. The second light intensity sensor senses the brightness in the upper light-sensing cavity and generates a corresponding second light intensity value. The sealing gasket rejection module is used to acquire the first light intensity value (I...). A ), second light intensity value (I) B And calculate the light intensity attenuation rate according to Formula 1;

[0007]

[0008] The sealing gasket rejection module is also used to compare the calculated light intensity attenuation rate with the preset light intensity attenuation threshold. When the calculated light intensity attenuation rate is greater than the preset light intensity attenuation threshold, the module outputs a preset first prompt message "The currently tested sealing gasket is rejected" to the prompt module. The prompt module is used to convert the first prompt message "The currently tested sealing gasket is rejected" into information that can be recognized by the testing personnel.

[0009] In the aforementioned liquefied natural gas unloading pipe gasket rejection system, a mobile module carrying a prompting module and a gasket receiving module arrives at the testing site. The gasket receiving module contains an opaque upper light-sensing cavity and an opaque lower light-sensing cavity. Between the upper and lower light-sensing cavities is an annular light channel through which light passes. The shape and size of the annular light channel are adapted to the size of the gasket, allowing the gasket to be embedded within it. A first light intensity sensor senses the brightness within the lower light-sensing cavity and generates a corresponding first light intensity value. A second light intensity sensor senses the brightness within the upper light-sensing cavity... The brightness is measured and a corresponding second light intensity value is generated. The gasket rejection module calculates the light intensity attenuation rate according to the above formula one, and compares the calculated light intensity attenuation rate with the preset light intensity attenuation threshold. When the calculated light intensity attenuation rate is greater than the preset light intensity attenuation threshold, the gasket being tested is rejected, and the preset first prompt message "The gasket being tested is rejected" is output to the prompt module. The prompt module converts the first prompt message "The gasket being tested is rejected" into information that can be recognized by the testing personnel, thereby quickly completing the rejection test of the gasket of the liquefied natural gas unloading pipe. Attached image description:

[0010] Figure 1 This is a schematic diagram of the liquefied natural gas unloading pipe sealing gasket rejection system of this application.

[0011] Figure 2 This is a structural schematic diagram of the liquefied natural gas unloading pipe sealing gasket rejection system of this application from another angle.

[0012] Figure 3 This is a schematic diagram of the functional modules of the liquefied natural gas unloading pipe sealing gasket rejection system of this application.

[0013] Figure 4 This is a flowchart illustrating the risk assessment process for the reusable sealing gaskets in the AI ​​detection and analysis module of this application.

[0014] The diagram shows: 10 LNG unloading pipe gasket rejection system, 20 moving module, 21 moving guide rod, 30 prompting module, 40 gasket housing module, 41 high-intensity light source, 42 first light intensity sensor, 43 second light intensity sensor, 44 plate placement stage, 440 plate box, 441 top cover, 442 light blocking component, 45 dark chamber, 50 gasket rejection module, 60 gasket image acquisition module, 70 AI detection and analysis module, and 80 power supply. Detailed implementation method:

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0016] like Figures 1 to 3 As shown, the liquefied natural gas unloading pipe sealing gasket rejection system 10 provided by the present invention includes: a moving module 20, a prompting module 30, a sealing gasket receiving module 40, and a sealing gasket rejection module 50; the prompting module 30 and the sealing gasket receiving module 40 are disposed on the moving module 20, and the moving module 20 is used to carry the prompting module 30 and the sealing gasket receiving module 40; the sealing gasket rejection module 50 is electrically connected to the prompting module 30 and the sealing gasket receiving module 40; the sealing gasket receiving module 40 is used to receive and position the sealing gasket, and the sealing gasket receiving module 40 has an opaque upper light-sensing cavity and an opaque lower light-sensing cavity inside, and a strong light source 41 and a first light source are provided in the lower light-sensing cavity. A light intensity sensor 42 is included, and a second light intensity sensor 43 is disposed within the upper light-sensing cavity. An annular light channel for light transmission exists between the upper and lower light-sensing cavities. The shape and size of the annular light channel are adapted to the size of the sealing gasket, and the sealing gasket can be embedded within the annular light channel. A sealing gasket rejection module 50 is electrically connected to the first light intensity sensor 42 and the second light intensity sensor 43. The first light intensity sensor 42 senses the brightness within the lower light-sensing cavity and generates a corresponding first illuminance value. The second light intensity sensor 43 senses the brightness within the upper light-sensing cavity and generates a corresponding second illuminance value. The sealing gasket rejection module 50 is used to acquire the first illuminance value (I1). A ), second light intensity value (I) B And calculate the light intensity attenuation rate according to Formula 1;

[0017]

[0018] The sealing gasket rejection module 50 is also used to compare the calculated light intensity attenuation rate with a preset light intensity attenuation threshold. When the calculated light intensity attenuation rate is greater than the preset light intensity attenuation threshold, it outputs a preset first prompt message "The currently tested sealing gasket is rejected" to the prompt module 30. The prompt module 30 is used to convert the first prompt message "The currently tested sealing gasket is rejected" into information that can be recognized by the testing personnel. The sealing gasket may be made of tetrafluoroethylene (PTFE).

[0019] The mobile module 20 can be a mobile frame, with wheels installed underneath to enhance the mobility of the entire device, facilitating the transport of the LNG unloading pipe gasket rejection system 10 to the production site for gasket inspection by testing personnel. The gasket rejection module 50 is a data processing device using an STM32 microcontroller; other models can be selected in other embodiments. It reads the analog signals from the first light intensity sensor 42 and the second light intensity sensor 43 via an ADC (analog-to-digital converter), converts these analog signals into digital signals, and the STM32 microcontroller executes a programmed sequence to periodically read the ADC values ​​from the first and second light intensity sensors 42 and 43, directly calculating the aforementioned attenuation rate formula after reading the ADC values. The prompt module 30 can be a display panel that displays the information provided by the gasket rejection module 50 in real time, including but not limited to light intensity attenuation rate and related warning information. A rechargeable power supply 80 is placed in the space below the prompt module 30. The power supply provides power to functional components such as the display panel, strong light source, first light intensity sensor 42, second light intensity sensor 43, and sealing gasket rejection module 50.

[0020] In this embodiment, the sealing gasket receiving module 40 includes a high-intensity light source 41, a first light intensity sensor 42, a second light intensity sensor 43, a plate placement stage 44, and a dark chamber 45. The plate placement stage 44 and the dark chamber 45 are arranged adjacent to each other vertically. The dark chamber 45 is the lower photosensitive cavity, and the high-intensity light source 41 and the first light intensity sensor 42 are disposed inside the dark chamber 45. The plate placement stage 44 is the upper photosensitive cavity and is used to place the sealing gasket to be tested. The plate placement stage 44 includes a hollow gasket box 440 and a top cover 441. The dark chamber 45 is a sealed space located directly below the gasket box 440. The walls of the dark chamber 45 are black. The high-intensity light source 41 is installed at the bottom of the dark chamber 45. One end of the top cover 441 of the gasket box 440 is movably connected to a moving guide rod 21 disposed on the moving module 20. The top cover 441 moves along the moving guide rod 21. 1. Move vertically upward to open the gasket box 440, allowing the sealing gasket to be placed inside; 2. The second light intensity sensor 43 is installed on the lower surface of the upper cover 441. The upper cover 441 moves vertically downward along the moving guide rod 21 to cover the gasket box 440, providing a sealed detection environment for the seal detection; 3. The bottom of the gasket box 440 is made of glass with good light transmission. A light-blocking element 442 is provided inside the gasket box 440. The light-blocking element 442 is fixed to the bottom of the gasket box 440 and located at the center of the bottom of the gasket box 440. The light-blocking element 442 and the side wall of the gasket box 440 form the annular light channel. The height of the light-blocking element 442 is greater than the thickness of the sealing gasket and less than the height of the side wall of the gasket box, to avoid the upper cover 441 from not being able to cover the gasket box 440 tightly due to the light-blocking element 442 being too high. Among these advantages, glass has high light transmittance, and since the transmittance is a known value, its impact on the detection is controllable. The high transmittance of glass allows for clear observation of the light transmission of the gasket. Furthermore, because the transmittance of the glass is known during the detection process, the influence of transmittance variations on the test results can be eliminated, making the results more reliable. Secondly, the glass base is easy to clean, ensuring good light transmission performance over long-term use. Finally, this design offers flexibility; depending on different testing needs, glass bases of different thicknesses, colors, or transmittances can be selected to adapt to the testing requirements of different gasket types.

[0021] The cross-sectional shape of the light-blocking element 442 is adapted to the inner ring of the sealing gasket. The light-blocking element 442 is also used to determine whether the sealing gasket is deformed. Specifically, when the sealing gasket is fitted onto the light-blocking element 442, if the inspector observes deformation, there is no need to continue operating the liquefied natural gas unloading pipe sealing gasket rejection system 10; the sealing gasket is then deemed unusable, thus saving inspection time. The light-blocking element 442 can be made of materials with poor light transmittance, such as rubber or silicone.

[0022] In the aforementioned liquefied natural gas unloading pipe gasket scrapping system 10, the gasket scrapping module 50 calculates the light intensity attenuation rate according to the above formula one, and compares the calculated light intensity attenuation rate with the preset light intensity attenuation threshold. When the calculated light intensity attenuation rate is greater than the preset light intensity attenuation threshold, the gasket being tested is determined to be scrapped. This system can quickly complete the scrapping detection of liquefied natural gas unloading pipe gaskets. However, the applicant discovered the following problem during the testing process: For example, after gasket A, numbered A, was removed from the pipeline and subjected to the first scrapping test, it was determined that gasket A could still be used. Therefore, gasket A was reinstalled on the pipeline. During use, gasket A showed a noticeable gas leak, or it was determined to be scrapped during a second test. At this point, the gas leak was irreversible and the safety hazard already existed. After studying and analyzing the scrapped gaskets, the applicant found that cracks, indentations, and deformation of the gasket surface were the main reasons for the scrapping.

[0023] To avoid the aforementioned adverse situations, please also refer to Figure 4 The LNG unloading pipe gasket rejection system 10 also includes a gasket image acquisition module 60 and an AI detection and analysis module 70. When the gasket rejection module 50 determines that the calculated light intensity attenuation rate is not greater than a preset light intensity attenuation threshold, it outputs a preset second prompt message, "The currently detected gasket can be used," to the prompt module 30 and sends a shooting control signal to the gasket image acquisition module 60. The gasket image acquisition module 60 responds to the shooting control signal provided by the gasket rejection module 50 by taking a picture of the currently detected gasket to obtain image data of the currently detected gasket. The AI ​​detection and analysis module 70 uses... Based on the image data and gasket identification information in the pre-stored image database, and the image data and identification information of the currently inspected gasket obtained by the gasket image acquisition module 60, the system determines whether the surface of the currently inspected gasket has cracks, indentations, or deformation. If the determination result is any of the above three determination results, a preset second prompt message, "The currently usable gasket is at risk, and replacement is recommended," is output to the prompt module 30. The prompt module 30 is used to convert the second prompt message, "The currently usable gasket is at risk, and replacement is recommended," into information that can be recognized by the inspection personnel. The gasket image acquisition module 60 is a high-resolution camera capable of capturing high-quality images. An industrial-grade CCD or CMOS camera can be considered, as these cameras can provide sufficient resolution to identify minute cracks and damage. The gasket image acquisition module 60 is installed on the inner surface of the upper cover 441 to ensure full coverage of the observation area of ​​the gasket and to capture images of the gasket from multiple angles.

[0024] The pre-stored image database includes a first image database and a second image database. The first image database stores image data and identification information of previously detected and deemed usable gaskets. The second image database stores image data and identification information of previously detected and deemed unusable gaskets. The image data includes surface crack data, surface indentation data, and ellipticity data of the gaskets. The AI ​​detection and analysis module 70 searches the second image database based on the image data and identification information of the currently detected gasket obtained by the gasket image acquisition module 60. If the corresponding gasket identification information and surface crack data, surface indentation data, and ellipticity data are found, it is determined that the surface of the currently detected gasket has cracks, indentations, or deformation. It is understood that the image data in the first and second image databases covers all stages of gasket development, from undamaged to severely damaged.

[0025] In this embodiment, when the gasket rejection module 50 determines that the calculated light intensity attenuation rate is not greater than the preset light intensity attenuation threshold, it generates first storage information based on the identification information of the currently detected gasket and sends a shooting control signal to the gasket image acquisition module 60. For example, when the gasket rejection module 50 determines that the calculated light intensity attenuation rate is not greater than the preset light intensity attenuation threshold, it outputs "Please enter the identification information of the currently detected gasket". The inspection personnel enter the identification information of the currently detected gasket according to the prompt. When the gasket rejection module 50 provides the first storage information to the AI ​​detection and analysis module 70, the AI ​​detection and analysis module 70 updates the first image database based on the first storage information and the image data of the currently detected gasket obtained by the gasket image acquisition module 60.

[0026] When the calculated light intensity attenuation rate is greater than a preset light intensity attenuation threshold, the sealing gasket rejection module 50 generates second storage information based on the acquired identification information of the currently detected sealing gasket and sends a shooting control signal to the sealing gasket image acquisition module 60. The sealing gasket rejection module 50 provides the second storage information to the AI ​​detection and analysis module 70. When the AI ​​detection and analysis module 70 receives the second storage information and the image data of the currently detected sealing gasket obtained by the sealing gasket image acquisition module 60, it determines whether the surface of the currently detected sealing gasket has cracks, indentations, or deformation. If the determination result is any of the above three determination results, it searches for the corresponding image data in the first image database based on the second storage information and the image data of the currently detected sealing gasket obtained by the sealing gasket image acquisition module 60. It adds the found corresponding image data and the corresponding sealing gasket identification information to the second image database and deletes the found corresponding image data and sealing gasket identification information from the first image database to complete the update of the first and second image databases. In this way, the AI ​​detection and analysis module 70 can be continuously optimized, so that the AI ​​detection and analysis module 70 improves the accuracy of rejection year by year. The AI ​​detection and analysis module 70 can transmit data with the gasket rejection module 50 and the gasket image acquisition module 60 via wired or wireless means.

[0027] The AI ​​detection and analysis module 70 is a microcomputer with data processing capabilities. During each inspection, in conjunction with the sealing gasket rejection module 50, the AI ​​detection and analysis module 70 stores the image data and identification information of sealing gaskets deemed unusable in the second database, and the image data and identification information of sealing gaskets deemed usable in the first database. Alternatively, the image data in the second and first databases can be processed to improve their accuracy. For example, when a year's worth of data has accumulated or a total of 500 photos has been collected, the data from the second and first databases can be imported into the EasyDL image recognition and classification module configured on the computer, and EasyDL can be started. The syDL model is trained to generate a waste classification model. The AI ​​detection and analysis module 70 then uses the waste classification model to determine the cracks, indentations, and deformations of the sealing gasket. Furthermore, the AI ​​detection and analysis module 70 processes and controls the data flow of the received image data of the currently detected sealing gasket. Through image preprocessing algorithms such as grayscale conversion, noise removal, and edge enhancement, the recognizability of the sealing gasket features in the image is improved. For example, the OpenCV library in Python is used to convert the color image to a grayscale image; median filtering or Gaussian filtering is used to remove random noise in the image to help improve image quality; and Sobel operator or Canny edge detection method is used to enhance the edges in the image and increase the visual sharpness of the image.

[0028] The AI ​​detection and analysis module 70 processes each image and then provides detailed annotations for each image, including parameters such as the grayscale value of the indentation, the severity of the crack, and the ellipticity.

[0029] For crack and damage recognition, commonly used CNN architectures include LeNet, AlexNet, VGG, and ResNet for image recognition problems. The collected image data is used to train the CNN model. The dataset is split into training, validation, and test sets to evaluate model performance. Hyperparameters (such as learning rate, batch size, and number of training epochs) are tuned to optimize the performance of the aforementioned defect classification model. Techniques such as data augmentation and dropout are applied to prevent overfitting.

[0030] For indentations on gaskets, collected images were analyzed to extract grayscale information. Based on the extracted data, grayscale levels were divided according to the data's range and distribution, and a corresponding color chart was designed. A program was written using programming languages ​​(such as Python and MATLAB) to map the extracted data onto the corresponding color chart and generate a visualization, resulting in a color chart containing nine grayscale levels. Each level corresponds to a different degree of gasket wear, which requires combining professional knowledge and statistical analysis to determine. Machine learning algorithms, such as linear regression or logistic regression, were used to establish a model relating grayscale values ​​to the degree of gasket damage. The color chart model was validated and adjusted using labeled image data to ensure its accuracy and practicality.

[0031] To address the ellipticity of the gaskets, the shape of the inner ring of each gasket was analyzed, and its minimum ellipticity was collected. The relationship between ellipticity and gasket damage was analyzed to determine the critical ellipticity value before failure.

Claims

1. A liquefied natural gas unloading pipe sealing gasket rejection system, characterized in that, The system includes a moving module, a prompting module, a sealing gasket receiving module, and a sealing gasket rejection module. The prompting module and the sealing gasket receiving module are mounted on the moving module, which carries them. The sealing gasket rejection module is electrically connected to the prompting module and the sealing gasket receiving module. The sealing gasket receiving module is used to receive and position the sealing gasket. Inside the sealing gasket receiving module, there are two opaque light-sensing cavities: an upper opaque light-sensing cavity and a lower opaque light-sensing cavity. The lower light-sensing cavity contains a strong light source and a first light intensity sensor, while the upper light-sensing cavity contains a second light intensity sensor. There is an annular light channel between the upper and lower light-sensing cavities, through which light passes. The shape and size of the annular light channel are adapted to the size of the sealing gasket, and the sealing gasket can be embedded in the annular light channel. The sealing gasket rejection module is electrically connected to a first light intensity sensor and a second light intensity sensor. The first light intensity sensor senses the brightness in the lower light-sensing cavity and generates a corresponding first light intensity value. The second light intensity sensor senses the brightness in the upper light-sensing cavity and generates a corresponding second light intensity value. The sealing gasket rejection module is used to acquire the first light intensity value I. A Second light intensity value I B And calculate the light intensity attenuation rate according to Formula 1; Light intensity attenuation rate (%) = Formula 1 The sealing gasket rejection module is also used to compare the calculated light intensity attenuation rate with a preset light intensity attenuation threshold. When the calculated light intensity attenuation rate is greater than the preset light intensity attenuation threshold, a preset first prompt message "The currently tested sealing gasket is rejected" is output to the prompt module. The prompt module is used to convert the first prompt message "The currently tested sealing gasket is rejected" into information that can be recognized by the testing personnel. The sealing gasket receiving module includes a placement stage, a darkroom, the high-intensity light source, the first light intensity sensor, and the second light intensity sensor. The placement stage and the darkroom are arranged adjacent to each other. The lower photosensitive cavity contains a strong light source and a first light intensity sensor, both housed within a dark chamber. The upper photosensitive cavity contains a placement stage for placing the sealing gasket to be tested. The placement stage includes a hollow gasket box and a top cover. One end of the top cover is movably connected to a guide rod mounted on a moving module. The top cover moves vertically upward along the guide rod to open the gasket box, allowing the sealing gasket to be placed inside. A second light intensity sensor is mounted on the lower surface of the top cover. The top cover moves vertically downward along the guide rod to cover the gasket box, providing a sealed testing environment for the sealing test. The bottom of the gasket box is made of glass with good light transmission. A light-blocking component is installed inside the gasket box. The light-blocking component is fixed to the bottom of the gasket box and located at the center of the bottom of the gasket box. The light-blocking component and the side wall of the gasket box form the annular light channel. The height of the light-blocking component is greater than the thickness of the sealing gasket and less than the height of the side wall of the gasket box.

2. The liquefied natural gas unloading pipe sealing gasket rejection system as described in claim 1, characterized in that: The cross-sectional shape of the light-blocking component is adapted to the inner ring of the sealing gasket. The light-blocking component is also used to determine whether the sealing gasket is deformed. Specifically, when the sealing gasket is fitted onto the light-blocking component, if the inspector observes deformation, the sealing gasket is deemed unusable.

3. The liquefied natural gas unloading pipe sealing gasket rejection system as described in claim 1, characterized in that: The darkroom is a sealed space located directly below the gasket box. The walls of the darkroom are black, and the strong light source is installed at the bottom of the darkroom.

4. The liquefied natural gas unloading pipe sealing gasket rejection system as described in claim 2 or 3, characterized in that: The mobile module is a mobile frame with casters installed below the frame and the prompting module and sealing gasket receiving module installed above the frame.

5. The liquefied natural gas unloading pipe sealing gasket rejection system as described in claim 1, characterized in that: The liquefied natural gas unloading pipe gasket rejection system also includes a gasket image acquisition module and an AI detection and analysis module. When the gasket rejection module determines that the calculated light intensity attenuation rate is not greater than the preset light intensity attenuation threshold, it outputs a preset second prompt message "The currently detected gasket can be used" to the prompt module and sends a shooting control signal to the gasket image acquisition module. The gasket image acquisition module is used to respond to the shooting control signal provided by the gasket rejection module and take a picture of the currently detected gasket to obtain the image data of the currently detected gasket. The AI ​​detection and analysis module is used to determine whether the surface of the currently inspected gasket has cracks, indentations, or deformation based on image data and gasket identification information in the pre-stored image database, and image data and identification information of the currently inspected gasket obtained by the gasket image acquisition module. When the judgment result is any of the above three judgment results, a preset second prompt message "The currently usable gasket is at risk and it is recommended to replace it" is output to the prompt module. The prompt module is used to convert the second prompt message "The currently usable gasket is at risk and it is recommended to replace it" into information that can be recognized by the inspection personnel.

6. The liquefied natural gas unloading pipe sealing gasket rejection system as described in claim 5, characterized in that: The pre-stored image database includes a first image database and a second image database. The first image database stores image data and identification information of previously detected and deemed usable gaskets. The second image database stores image data and identification information of previously detected and deemed unusable gaskets. The image data includes surface crack data, surface indentation data, and ellipticity data of the gasket. The AI ​​detection and analysis module searches the second image database based on the image data of the currently detected gasket and its identification information obtained by the gasket image acquisition module. If the corresponding gasket identification information, surface crack data, surface indentation data, and ellipticity data are found, it is determined that the surface of the currently detected gasket has cracks, indentations, or deformation.

7. The liquefied natural gas unloading pipe sealing gasket rejection system as described in claim 6, characterized in that: When the gasket rejection module determines that the calculated light intensity attenuation rate is not greater than the preset light intensity attenuation threshold, it generates first storage information based on the identification information of the currently detected gasket and sends a shooting control signal to the gasket image acquisition module. When the gasket rejection module provides the first storage information to the AI ​​detection and analysis module, the AI ​​detection and analysis module updates the first image database based on the first storage information and the image data of the currently detected gasket obtained by the gasket image acquisition module. When the gasket rejection module determines that the calculated light intensity attenuation rate is greater than the preset light intensity attenuation threshold, it generates second storage information based on the identification information of the currently detected gasket and sends a shooting control signal to the gasket image acquisition module. The gasket rejection module then provides the second storage information to the AI ​​detection and analysis module. When the AI ​​detection and analysis module receives the second storage information and the image data of the currently detected gasket obtained by the gasket image acquisition module, it determines whether the surface of the currently detected gasket has cracks, indentations, or deformation. If the determination result is any of the above three determination results, it searches for the corresponding image data in the first image database based on the second storage information and the image data of the currently detected gasket obtained by the gasket image acquisition module. It adds the searched image data and gasket identification information to the second image database and deletes the searched image data and gasket identification information from the first image database to complete the update of the first and second image databases.

Images