Oxygen generator system
By collecting and analyzing semipermeable membrane images in the oxygen generator in real time, identifying and automatically cleaning dirt, the problem of semipermeable membrane surface detection is solved, and the efficiency and oxygen quality of the oxygen generator are improved.
Patent Information
- Application Number
- CN202410972700.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-19
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the existing oxygen generator system, it is difficult to detect dirt and impurities on the semi-permeable membrane surface, which affects separation efficiency and performance, and is poor in practicality.
Using semi-permeable membrane image acquisition module, image preprocessing module, feature extraction module and detection model training module, semi-permeable membrane images are captured in real time through the camera, preprocessing, feature extraction and dirt recognition, and automatically cleaned with the cleaning treatment system to ensure the membrane Clean surface.
Real-time detection and automatic cleaning of the semi-permeable membrane surface is realized, the separation efficiency and performance of the oxygen generator is improved, and the safety and reliability of oxygen quality is ensured.
Smart Images

Figure CN118904033A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of oxygen concentrators, and more specifically, to an oxygen concentrator system. Background Art
[0002] An oxygen concentrator (also called an oxygen generator) is a device used to separate oxygen from other gases in the air to produce pure oxygen. Its principle is usually based on molecular sieve technology or membrane separation technology.
[0003] Molecular sieve technology: This technology uses molecular sieves (also called adsorbents) to selectively adsorb nitrogen, water vapor and other impurity gases, allowing oxygen to pass through and produce pure oxygen. The process generally includes the following steps:
[0004] Adsorption stage: Air passes through the molecular sieve, where nitrogen, water vapor, etc. are adsorbed, while oxygen passes through.
[0005] Desorption stage: By adjusting the temperature or pressure, the adsorbent releases the adsorbed nitrogen and other impurity gases, thereby restoring its adsorption capacity again.
[0006] Membrane separation technology: This technology uses special semi-permeable membranes to selectively separate gases. Oxygen molecules are smaller than nitrogen molecules and can therefore pass through the membrane, resulting in pure oxygen. The process generally includes the following steps:
[0007] Membrane separation stage: air is passed through a semi-permeable membrane, through which oxygen molecules are able to pass while larger molecules such as nitrogen are blocked.
[0008] Collection stage: collects pure oxygen that passes through the membrane.
[0009] The working principle of oxygen concentrators is based on one of these two technologies, and the corresponding technology is usually selected according to the specific application requirements and the design of the oxygen concentrator. Oxygen concentrators can be used in medical, industrial, laboratory and other fields to provide oxygen with different purities. For example, medical oxygen concentrators are used to provide high-purity oxygen to meet the needs of medical purposes.
[0010] The document with the prior art publication number CN110732213A provides an air filter device for a medical oxygen concentrator. The device is provided with a secondary filtering mechanism. When the air enters the filter tube through the first corrugated tube, the HEPA plate, the semi-permeable membrane plate and the activated carbon plate inside the filter tube will filter the air layer by layer. When the filter tube needs to be replaced, the filter tube can be removed by twisting the connecting tubes at both ends of the filter tube in turn, and then the prepared new filter tube is installed between the two corrugated tubes. The secondary filtering mechanism is updated, the structure is simple, and the operation is more convenient. Compared with the prior art, the filtering structure is optimized and the operating efficiency is improved.
[0011] Although the above-mentioned prior art solution can achieve the relevant beneficial effects through the structure of the prior art, it still has the following defects: when the device is used, dirt and impurities will be generated on one side of the semipermeable membrane. Since the device is not convenient for detecting the surface of the semipermeable membrane, the semipermeable membrane with dirt and impurities will affect the separation efficiency and performance, and the practicality is poor.
[0012] In view of this, we propose an oxygen concentrator system. Summary of the invention
[0013] 1. Technical issues to be solved
[0014] The purpose of the present application is to provide an oxygen generator system, which solves the technical problems raised in the above-mentioned background technology and achieves the technical effect of facilitating the detection of the surface of the semipermeable membrane.
[0015] 2. Technical solution
[0016] The embodiment of the present application provides an oxygen concentrator system, including: a semipermeable membrane image acquisition module, a semipermeable membrane image preprocessing module, a semipermeable membrane image feature extraction module, a semipermeable membrane surface detection model training module, and a cleaning system module.
[0017] Semipermeable membrane image acquisition module: The semipermeable membrane image acquisition module is used to capture the image of the semipermeable membrane surface in real time through a camera installed in the oxygen concentrator;
[0018] Semi-permeable membrane image preprocessing module: The semi-permeable membrane image preprocessing module preprocesses the collected image, including adjusting the brightness and contrast, etc., to prepare the image for subsequent analysis and processing;
[0019] Semi-permeable membrane image feature extraction module: The semi-permeable membrane image feature extraction module extracts features about dirt and impurities from the pre-processed image, such as color, texture, shape, etc.;
[0020] Semipermeable membrane surface detection model training module: The semipermeable membrane surface detection model training module uses a labeled image data set to train machine learning for identifying dirt and impurities on the surface of the semipermeable membrane;
[0021] Cleaning system module: The cleaning system module automatically starts the cleaning process according to the type and degree of dirt and impurities detected.
[0022] As an optional solution of the technical solution of this application document, the semipermeable membrane image acquisition module is used to capture the image of the semipermeable membrane surface in real time through a camera installed in the oxygen generator, ensuring that the camera can capture all important details of the membrane surface, and the collected image data will be stored in a local database.
[0023] As an optional solution of the technical solution of the present application document, the semi-permeable membrane image preprocessing module preprocesses the collected image, including adjusting the brightness and contrast, etc., to prepare the image for subsequent analysis and processing, and the brightness adjustment: the brightness of the image is enhanced or weakened by the histogram equalization method to improve the visual effect of the image;
[0024] The contrast adjustment: increasing the contrast of an image can make the details in the image clearer. Contrast enhancement can be achieved by stretching the range of pixel values, such as using a linear stretching method.
[0025] As an optional solution of the technical solution of the present application document, the semi-permeable membrane image feature extraction module extracts features about dirt and impurities from the pre-processed image, such as color, texture, shape, etc. The color feature: through color space conversion: convert the image from RGB space to other color spaces (such as HSV, YUV, etc.) to better extract information about specific colors;
[0026] The texture features: through Local Binary Pattern (LBP): encode local texture information by comparing the grayscale values of a pixel with its neighboring pixels;
[0027] The shape feature: through the boundary feature: extract the outline of dirt or impurities, and analyze the smoothness and complexity of the boundary.
[0028] As an optional solution to the technical solution of the present application document, the semi-permeable membrane surface detection model training module uses a labeled image data set to train a convolutional neural network machine learning model. The model learns to extract features from the image for identifying dirt and impurities on the surface of the semi-permeable membrane. The trained model is deployed in a real-time monitoring system to continuously monitor the image of the semi-permeable membrane surface, and to perform early warning and processing based on the results of the model output.
[0029] As an optional solution of the technical solution of the present application document, the cleaning treatment system module automatically starts the cleaning treatment process according to the type and degree of dirt and impurities detected, thereby improving the production efficiency and stability of the oxygen generator, reducing the impact of pollutants on oxygen quality, and ensuring the safety and reliability of oxygen production and supply.
[0030] 3. Beneficial effects
[0031] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:
[0032] 1. The present application adopts technical means such as a semipermeable membrane image acquisition module, thereby effectively solving the technical problems in the prior art, and further realizing that when the oxygen generator is producing oxygen, the semipermeable membrane image acquisition module is used to collect the image data in real time, and the semipermeable membrane surface detection model is used to analyze the image data, so as to timely detect the type and degree of dirt and impurities on the semipermeable membrane, and the system automatically starts the cleaning process to ensure the separation efficiency and performance of the semipermeable membrane.
[0033] 2. The present application sets a cleaning seat and a transmission mechanism. After the oxygen filter mechanism is placed in the cleaning seat, the driving wheel can be used to drive the semipermeable membrane body to rotate in the cleaning liquid. At the same time, under the action of the reciprocating screw, the pushing frame drives the semipermeable membrane body to reciprocate, thereby realizing the automatic cleaning of the semipermeable membrane body, replacing manual cleaning, and improving the practicality of the device.
[0034] 3. The present application provides a baffle to facilitate the overflow of cleaning liquid during cleaning, and provides an adjustment seat to facilitate the adjustment of the position of the oxygen filter mechanism and clean it, thereby further improving the practicality of the device. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 A schematic diagram of the overall structure of the semipermeable membrane surface quality detection process of the oxygen concentrator system disclosed in a preferred embodiment of the present application;
[0036] Figure 2 This is a schematic diagram of the overall structure of an oxygen concentrator system disclosed in a preferred embodiment of the present application;
[0037] Figure 3 It is a schematic cross-sectional view of the overall structure of an oxygen concentrator system disclosed in a preferred embodiment of the present application;
[0038] Figure 4 A schematic cross-sectional view of the adjustment seat structure of an oxygen concentrator system disclosed in a preferred embodiment of the present application;
[0039] Figure 5 A schematic cross-sectional view of the oxygen filter structure of an oxygen concentrator system disclosed in a preferred embodiment of the present application;
[0040] Figure 6 A schematic cross-sectional expansion diagram of the cleaning seat structure of the oxygen concentrator system disclosed in a preferred embodiment of the present application;
[0041] Figure 7 This is a schematic diagram of the transmission mechanism structure of an oxygen concentrator system disclosed in a preferred embodiment of the present application;
[0042] Explanation of the numbers in the figure: 1. Oxygen generator body; 2. Adjustment seat; 3. Oxygen filtering mechanism; 4. Cleaning seat; 5. Transmission mechanism; 201. Electric push rod; 202. Motor A; 203. Connecting rod group; 301. Mounting frame; 302. Gear ring; 303. Semipermeable membrane body; 401. Baffle; 402. Spring; 403. Liquid inlet pipe; 404. Liquid discharge pipe; 501. Motor B; 502. Reciprocating screw rod; 503. Push frame; 504. Universal joint; 505. Driving wheel. DETAILED DESCRIPTION
[0043] The present application is further described in detail below in conjunction with the accompanying drawings.
[0044] Reference Figure 1 , oxygen concentrator system, including: semipermeable membrane image acquisition module, semipermeable membrane image preprocessing module, semipermeable membrane image feature extraction module, semipermeable membrane surface detection model training module, cleaning treatment system module,
[0045] Semipermeable membrane image acquisition module: The semipermeable membrane image acquisition module is used to capture the image of the semipermeable membrane surface in real time through a camera installed in the oxygen concentrator;
[0046] Semi-permeable membrane image preprocessing module: The semi-permeable membrane image preprocessing module preprocesses the collected images, including adjusting the brightness and contrast, etc., to prepare the images for subsequent analysis and processing;
[0047] Semi-permeable membrane image feature extraction module: The semi-permeable membrane image feature extraction module extracts features about dirt and impurities from the pre-processed image, such as color, texture, shape, etc.;
[0048] Semipermeable membrane surface detection model training module: The semipermeable membrane surface detection model training module uses a labeled image dataset to train machine learning to identify dirt and impurities on the surface of the semipermeable membrane;
[0049] Cleaning system module: The cleaning system module automatically starts the cleaning process according to the type and degree of dirt and impurities detected.
[0050] The semipermeable membrane image acquisition module is used to capture the image of the semipermeable membrane surface in real time through the camera installed in the oxygen concentrator, ensuring that the camera can capture important details on the membrane surface. The collected image data will be stored in the local database.
[0051] The semi-permeable membrane image preprocessing module preprocesses the collected images, including adjusting the brightness and contrast, etc., to prepare the images for subsequent analysis and processing. Brightness adjustment: the brightness of the image is enhanced or weakened by the histogram equalization method to improve the visual effect of the image;
[0052] Contrast adjustment: Increasing the contrast of an image can make details in the image clearer. Contrast enhancement can be achieved by stretching the range of pixel values, for example using a linear stretch method.
[0053] The semi-permeable membrane image feature extraction module extracts features about dirt and impurities from the pre-processed image, such as color, texture, shape, etc. Color features: through color space conversion: convert the image from RGB space to other color spaces such as HSV, YUV, etc., so as to better extract information about specific colors;
[0054] Texture features: through local binary pattern LBP: encode local texture information by comparing the grayscale values of a pixel with its neighboring pixels;
[0055] Shape features: Through boundary features: extract the outline of dirt or impurities, and analyze the smoothness and complexity of the boundary.
[0056] The semipermeable membrane surface detection model training module uses labeled image datasets to train the convolutional neural network machine learning model. The model will learn to extract features from images to identify dirt and impurities on the surface of the semipermeable membrane. The trained model will be deployed to the real-time monitoring system to continuously monitor the images of the semipermeable membrane surface, and provide early warnings and processing based on the results of the model output.
[0057] The cleaning system module automatically starts the cleaning process according to the type and degree of dirt and impurities detected, thereby improving the production efficiency and stability of the oxygen generator, reducing the impact of pollutants on oxygen quality, and ensuring the safety and reliability of oxygen production and supply.
[0058] Reference Figure 2 and Figure 3 The oxygen concentrator system includes an oxygen concentrator body 1, an adjustment seat 2, an oxygen filter mechanism 3, a cleaning seat 4, and a transmission mechanism 5.
[0059] An adjustment seat 2 is slidably arranged in the oxygen concentrator body 1, and two oxygen filtering mechanisms 3 are rotatably connected to the adjustment seat 2. A cleaning seat 4 is fixedly connected to the oxygen concentrator body 1, and a transmission mechanism 5 is fixedly connected to the cleaning seat 4.
[0060] Reference Figure 4 An electric push rod 201 is connected between the adjustment seat 2 and the oxygen concentrator body 1. Two motors A202 are fixedly connected in the adjustment seat 2. A connecting rod group 203 is fixedly connected to the output end of the motor A202. A plurality of cameras are installed on the adjustment seat 2. The connecting rod group 203 enables the oxygen filter mechanism 3 to be placed in the cleaning seat 4.
[0061] Reference Figure 5The oxygen filtering mechanism 3 includes a mounting frame 301, which is rotatably connected to the other end of the connecting rod assembly 203. A gear ring 302 is slidably and rotatably connected in the mounting frame 301, and a semipermeable membrane body 303 is connected in the gear ring 302.
[0062] Reference Figure 6 The inner walls on both sides of the opening of the cleaning seat 4 are slidably provided with baffles 401, a spring 402 is fixedly provided between the baffle 401 and the cleaning seat 4, and a liquid inlet pipe 403 and a liquid discharge pipe 404 are connected to one side of the cleaning seat 4. The baffle 401 realizes the function of blocking the cleaning liquid.
[0063] Reference Figure 7 The transmission mechanism 5 includes a motor B501, the output end of the motor B501 extends to the inside of the cleaning seat 4 and is connected and fixed with a reciprocating screw 502, the outer wall of the reciprocating screw 502 is slidably matched with a pushing frame 503, the pushing frame 503 is movably contacted with the outer wall of the gear ring 302, the other end of the reciprocating screw 502 is connected and fixed with a universal joint 504, the other end of the universal joint 504 is connected and fixed with a driving wheel 505, and the driving wheel 505 is meshed with the gear ring 302 for transmission.
[0064] The implementation principle of the oxygen concentrator system of the embodiment of the present application is as follows: when oxygen production is required, the semipermeable membrane body 303 is used to filter the air so that oxygen can pass through to complete oxygen production. During the oxygen production process, the semipermeable membrane image acquisition module is used to capture the image of the semipermeable membrane surface in real time through the camera installed in the oxygen concentrator to ensure that the camera can capture important details on the membrane surface. The collected image data will be stored in the local database;
[0065] Then the semi-permeable membrane image preprocessing module preprocesses the collected images, including adjusting the brightness and contrast, etc., to prepare the images for subsequent analysis and processing. Brightness adjustment: The brightness of the image is enhanced or weakened by the histogram equalization method to improve the visual effect of the image;
[0066] Contrast adjustment: Increasing the contrast of an image can make the details in the image clearer. Contrast enhancement can be achieved by stretching the range of pixel values, such as using a linear stretch method;
[0067] The semi-permeable membrane image feature extraction module is used to extract features about dirt and impurities from the pre-processed image, such as color, texture, shape, etc. Color features: through color space conversion: the image is converted from RGB space to other color spaces such as HSV, YUV, etc., so as to better extract information about specific colors;
[0068] Texture features: through local binary pattern LBP: encode local texture information by comparing the grayscale values of a pixel with its neighboring pixels;
[0069] Shape features: Through boundary features: extract the outline of dirt or impurities, analyze the smoothness and complexity of the boundary, etc.;
[0070] The semi-permeable membrane surface detection model training module uses the labeled image data set to train the convolutional neural network machine learning model. The model will learn to extract features from the image to identify dirt and impurities on the surface of the semi-permeable membrane. The trained model is deployed to the real-time monitoring system to continuously monitor the images of the semi-permeable membrane surface, and to issue warnings and process according to the results output by the model.
[0071] Then, the cleaning system module automatically starts the cleaning process according to the type and degree of dirt and impurities detected, thereby improving the production efficiency and stability of the oxygen generator, reducing the impact of pollutants on oxygen quality, and ensuring the safety and reliability of oxygen production and supply;
[0072] First, the electric push rod 201 is used to drive the adjustment seat 2 to move, so that the cleaned oxygen filter mechanism 3 is placed on the cleaning seat 4, and then the motor A202 is used to drive the connecting rod group 203 to rotate, so that the connecting rod group 203 drives the oxygen filter mechanism 3 to be inserted into the cleaning seat 4;
[0073] Then, the cleaning liquid is injected into the cleaning seat 4 through the liquid inlet pipe 403, and then the motor B501 is used to drive the reciprocating screw 502 to rotate. At this time, the reciprocating screw 502 will drive the pushing frame 503 to move back and forth, thereby driving the gear ring 302 to move back and forth, so that the semipermeable membrane body 303 moves back and forth in the cleaning liquid, and then under the action of the driving wheel 505 connected to the universal joint 504, the driving wheel 505 will drive the gear ring 302 to rotate, so that the semipermeable membrane body 303 rotates in the cleaning liquid to complete self-cleaning.
Claims
1. An oxygen concentrator system, comprising: a semipermeable membrane image acquisition module, a semipermeable membrane image preprocessing module, a semipermeable membrane image feature extraction module, a semipermeable membrane surface detection model training module, and a cleaning treatment system module, characterized in that: Semipermeable membrane image acquisition module: The semipermeable membrane image acquisition module is used to capture the image of the semipermeable membrane surface in real time through a camera installed in the oxygen concentrator; Semi-permeable membrane image preprocessing module: The semi-permeable membrane image preprocessing module preprocesses the collected image, including adjusting the brightness and contrast, etc., to prepare the image for subsequent analysis and processing; Semi-permeable membrane image feature extraction module: The semi-permeable membrane image feature extraction module extracts features about dirt and impurities from the pre-processed image, such as color, texture, shape, etc.; Semipermeable membrane surface detection model training module: The semipermeable membrane surface detection model training module uses a labeled image data set to train machine learning for identifying dirt and impurities on the surface of the semipermeable membrane; Cleaning system module: The cleaning system module automatically starts the cleaning process according to the type and degree of dirt and impurities detected.
2. The oxygen concentrator system according to claim 1, characterized in that: The semipermeable membrane image acquisition module is used to capture the image of the semipermeable membrane surface in real time through a camera installed in the oxygen concentrator, ensuring that the camera can capture all important details of the membrane surface, and the acquired image data will be stored in a local database.
3. The oxygen concentrator system according to claim 1, characterized in that: The semi-permeable membrane image preprocessing module preprocesses the collected image, including adjusting the brightness and contrast, etc., to prepare the image for subsequent analysis and processing. The brightness adjustment: enhances or reduces the brightness of the image by a histogram equalization method to improve the visual effect of the image; The contrast adjustment: increasing the contrast of an image can make the details in the image clearer. Contrast enhancement can be achieved by stretching the range of pixel values, such as using a linear stretching method.
4. The oxygen concentrator system according to claim 1, characterized in that: The semi-permeable membrane image feature extraction module extracts features about dirt and impurities from the pre-processed image, such as color, texture, shape, etc. The color feature: through color space conversion: converting the image from RGB space to other color spaces (such as HSV, YUV, etc.) to better extract information about specific colors; The texture features: through Local Binary Pattern (LBP): encode local texture information by comparing the grayscale values of a pixel with its neighboring pixels; The shape feature: through the boundary feature: extract the outline of dirt or impurities, and analyze the smoothness and complexity of the boundary.
5. The oxygen concentrator system according to claim 1, characterized in that: The semipermeable membrane surface detection model training module uses a labeled image data set to train a convolutional neural network machine learning model. The model learns to extract features from images to identify dirt and impurities on the surface of the semipermeable membrane. The trained model is deployed in a real-time monitoring system to continuously monitor images of the semipermeable membrane surface, and to provide early warnings and processing based on the results output by the model.
6. The oxygen concentrator system according to claim 1, characterized in that: The cleaning system module automatically starts the cleaning process according to the type and degree of dirt and impurities detected, thereby improving the production efficiency and stability of the oxygen generator, reducing the impact of pollutants on oxygen quality, and ensuring the safety and reliability of oxygen production and supply.
Citation Information
Cited By
Arduino-based image recognition fish tank filter cotton automatic replacement device
CN119453137A
Automatic replacement device for filter cotton of fish tank based on image recognition of arduino
CN119453137B
