A visual judgment automatic argon blowing system and method for a refining furnace

Abstract

The application discloses a kind of refining furnace visual determination automatic argon blowing system and method, including camera, instrument cabinet, host computer, L2 system and bottom blowing argon control cabinet;The instrument cabinet includes gigabit switch, PLC controller and man-machine interface;The camera is connected with the first interface of gigabit switch;The first interface of the PLC controller is connected with the second interface of gigabit switch;The host computer is connected with the third interface of gigabit switch;The second interface of the PLC controller is connected with the L2 system;The third interface of the PLC controller is connected with the man-machine interface;The L2 system is connected with the bottom blowing argon control cabinet, the application not only can improve the operation efficiency of refining furnace, reduce labor cost, but also can improve product quality, reduce energy consumption, it has important significance to promote the intelligent development of industrial manufacturing field.

Description

Technical Field

[0001] This invention relates to the field of refining furnace smelting technology, and more specifically to a visual judgment automatic argon blowing system and method for refining furnaces. Background Technology

[0002] The refining furnace is the main ladle refining equipment in steel production. During the refining process, argon gas is blown into the molten steel through a pipe installed on the ladle and through the permeable bricks at the bottom to agitate the molten steel. Argon gas forms a large number of argon bubbles in the molten steel. These bubbles absorb and carry away harmful gases in the molten steel, while gases dissolved in the molten steel rise to the surface with the bubbles and escape. The rising of the argon bubbles agitates the molten steel, thereby removing impurities and accelerating the homogenization of the steel's temperature and composition, achieving the effect of precisely adjusting the composition of the molten steel.

[0003] Currently, both domestic and international argon blowing refining processes rely on manual adjustment of the argon blowing stirring flow rate. This involves manually opening or closing the flow control valve based on the actual stirring intensity of the molten steel. This method suffers from several drawbacks: first, the intensity itself lacks quantifiable indicators and is entirely determined manually, resulting in poor stability and precision; second, manual adjustment is time-consuming, impacting efficiency; and third, the actual stirring effect depends entirely on standardized human operation, leading to insufficient reliability. Therefore, existing manual argon blowing control schemes suffer from poor reliability, low stability, high labor intensity for workers, potential impact on molten steel quality, and increased argon energy costs.

[0004] Therefore, there is an urgent need to propose an automated technology that is highly efficient, has high control precision, and good stability to solve the problems in existing argon blowing and stirring processes. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide an automatic argon blowing system and method for visual judgment in refining furnaces, so as to solve the problems existing in the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] An automatic argon blowing system for visual judgment in a refining furnace includes a camera, an instrument cabinet, a host computer, an L2 system, and a bottom-blowing argon control cabinet;

[0008] The instrument cabinet includes a gigabit switch, a PLC controller, and a human-machine interface; the PLC controller is electrically connected to the gigabit switch and the human-machine interface respectively.

[0009] The camera is connected to a gigabit switch; the gigabit switch is connected to the host computer.

[0010] The PLC controller is electrically connected to the L2 system;

[0011] The L2 system is electrically connected to the bottom-blown argon control cabinet;

[0012] The camera is used to capture images of the distribution of molten steel and slag layers in the ladle, and uploads them to the host computer via the gigabit switch.

[0013] The host computer includes an image processing module and a data management module;

[0014] The image processing module is used to preprocess the image of the distribution of molten steel and slag layer in the ladle liquid surface, and calculate the proportion of the exposed area of ​​molten steel in the current ladle liquid surface based on image recognition technology.

[0015] The data management module has established expert model data, which is used to obtain the control target value of the bottom blowing argon control cabinet under the current stirring intensity from the expert model data, and to determine the argon blowing control amount under the current stirring intensity based on the proportion of exposed steel area in the ladle liquid surface and the control target value.

[0016] The PLC controller acquires the argon blowing control quantity under the current stirring intensity and transmits it to the L2 system. The L2 system determines the timing and quantity of argon blowing from the bottom blowing argon control cabinet to the ladle based on the argon blowing control quantity.

[0017] The L2 system uploads basic information about the ladle and argon blowing information to the host computer via a PLC controller.

[0018] Furthermore, the host computer also includes: a human-machine interface and an expert system module;

[0019] The human-machine interface is used to display images of molten steel and slag in the ladle in real time, as well as control information on the argon gas flow rate.

[0020] While the image processing module calculates the image of the proportion of molten steel area in the ladle in real time and controls the timing and amount of argon blowing in the bottom blowing argon control cabinet of the L2 system, the expert system module identifies the image of the proportion of molten steel area, extracts the feature parameters corresponding to the current image, and saves the feature parameters of the current image in the expert model data.

[0021] Furthermore, the system is also equipped with a camera protection device: the camera protection device includes a dust removal and cleaning device;

[0022] A slight positive pressure is formed inside the camera protection device. High-pressure gas is used to evenly blow the camera through the tilt angle of the front aperture lens, thus achieving uniform blowing of the camera.

[0023] Furthermore, the human-machine interface and human-machine interface are also used for users to set manual control parameters to achieve manual mode.

[0024] Furthermore, the camera is an infrared camera, and the operating parameters of the infrared camera are: the detector element is 640*480; the operating wavelength is 8-14um; and the frame rate is 50Hz.

[0025] Furthermore, the human-computer interaction interface is specifically used to display images of the ladle surface in real time, use different colors to represent molten steel and slag, display historical changes in the proportion of molten steel through trend curves, replay historical videos in real time, and query historical curves of changes in the area of ​​molten steel during argon blowing.

[0026] Furthermore, the expert model data includes: the target proportion of exposed molten steel area under different stirring intensities, the qualified proportion of exposed molten steel area, the opening degree of the flow valve, and the flow range;

[0027] The human-machine interface is also equipped with an automatic alarm function, which will provide alarm reminders when a fault occurs or the argon gas flow reaches the upper and lower limits.

[0028] The image processing module has a processing time of within 50ms.

[0029] Furthermore, the ladle information acquired by the L2 system is transmitted back to the system control module in real time. The ladle information includes: ladle number, steel type, argon blowing start, and argon flow rate.

[0030] This invention also provides a method for a visual judgment automatic argon blowing system for a refining furnace. The method includes: a camera capturing images of the distribution of molten steel and slag layers on the ladle surface, transmitting these images to a host computer system and a PLC controller via a gigabit switch; the image processing module in the host computer system preprocessing the images of the distribution of molten steel and slag layers on the ladle surface and calculating the proportion of exposed molten steel area on the ladle surface; then, based on the proportion of exposed molten steel area on the ladle surface and the control target value, determining the argon blowing control amount under the current stirring intensity using expert model data in the data management module; simultaneously determining whether the ladle has entered a predetermined position; when the ladle enters the predetermined position, the PLC controller acquiring the argon blowing control amount under the current stirring intensity and transmitting it to the L2 system; the L2 system controlling the flow regulating valve in the bottom blowing argon control cabinet to open according to the argon blowing control amount and blowing argon into the ladle.

[0031] When the proportion of exposed molten steel in the ladle reaches the control target value, the L2 system controls the flow regulating valve in the bottom blowing argon control cabinet to close until the argon blowing ends;

[0032] The L2 system uploads basic information about the ladle and argon blowing information to the host computer via a PLC controller.

[0033] Furthermore, the image of the distribution of molten steel and slag layer on the ladle surface is preprocessed as follows: the features in the image of the distribution of molten steel and slag layer on the ladle surface are preprocessed by increasing contrast and low-pass filtering, or one or more of these methods. Then, the proportion of exposed molten steel area is calculated by one or more methods, such as binary method and color histogram. The contour information of molten steel in the image is automatically extracted to determine the analysis area of ​​the image, and the molten steel and slag layer are identified in the analysis area of ​​the image using automatic threshold segmentation technology.

[0034] This invention also provides a visual judgment automatic argon blowing method for refining furnaces. The method includes: a camera capturing an image of the distribution of molten steel and slag layer on the ladle surface, transmitting it to a host computer system and a PLC controller via a gigabit switch; the image processing module in the host computer system preprocesses the image of the distribution of molten steel and slag layer on the ladle surface and calculates the proportion of exposed molten steel area on the ladle surface; then, based on the proportion of exposed molten steel area on the ladle surface and the control target value, the expert model data in the data management module determines the argon blowing control amount under the current stirring intensity, and simultaneously determines whether the ladle has entered a predetermined position; when the ladle enters the predetermined position, the PLC controller acquires the argon blowing control amount under the current stirring intensity and transmits it to the L2 system; the L2 system controls the flow regulating valve in the bottom blowing argon control cabinet to open according to the argon blowing control amount and blows argon into the ladle.

[0035] When the proportion of exposed molten steel in the ladle reaches the control target value, the L2 system controls the flow regulating valve in the bottom blowing argon control cabinet to close until the argon blowing ends;

[0036] The L2 system uploads basic information about the ladle and argon blowing information to the host computer via a PLC controller.

[0037] Furthermore, the image of the distribution of molten steel and slag layer on the ladle surface is preprocessed as follows: the features in the image of the distribution of molten steel and slag layer on the ladle surface are preprocessed by increasing contrast and low-pass filtering, or one or more of these methods. Then, the proportion of exposed molten steel area is calculated by one or more methods, such as binary method and color histogram. The contour information of molten steel in the image is automatically extracted to determine the analysis area of ​​the image, and the molten steel and slag layer are identified in the analysis area of ​​the image using automatic threshold segmentation technology.

[0038] According to specific embodiments of the present invention, the invention captures the melting state of the molten steel in the ladle within the refining furnace in real time and uses an expert system module to precisely analyze the images to determine the optimal timing and amount of argon blowing. When the expert system module determines that argon blowing is necessary, it automatically triggers the argon blowing device for precise control, ensuring that the melting process within the refining furnace reaches its optimal state. The implementation of this invention not only improves the operating efficiency of the refining furnace and reduces labor costs but also enhances product quality and reduces energy consumption, thus playing a significant role in promoting the intelligent development of the industrial manufacturing sector. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0040] The following description, in conjunction with the accompanying drawings, further illustrates the automatic argon blowing system and method for visual judgment in refining furnaces according to the present invention;

[0041] Figure 1 This is an overall structural diagram of the intelligent system of the bottom blowing argon device in the automatic argon blowing system for visual judgment of refining furnace provided by the present invention;

[0042] Figure 2 This is a schematic diagram of the area ratio of molten steel and slag in the automatic argon blowing system for visual judgment of refining furnace provided by the present invention.

[0043] Figure 3 This is a diagram showing the radiation coefficients of molten steel and slag in the automatic argon blowing system for visual judgment of refining furnaces provided by this invention.

[0044] Figure 4 This invention provides a system flow chart and process flow diagram for the automatic argon blowing system for visual judgment in refining furnaces.

[0045] In the diagram: 1. Camera; 2. Instrument cabinet; 3. Gigabit switch; 4. PLC controller; 5. Human-machine interface; 6. Host computer; 7. L2 system; 8. Bottom-blown argon control cabinet; 9. Steel ladle. Detailed Implementation

[0046] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0047] To better understand the purpose, structure, and function of this invention, the invention will be described in further detail below with reference to the accompanying drawings.

[0048] like Figure 1 As shown, the present invention provides an automatic argon blowing system for visual judgment of refining furnace, including a camera 1, an instrument cabinet 2, a host computer 6, an L2 system 7, and a bottom blowing argon control cabinet 8;

[0049] The instrument cabinet 2 includes a gigabit switch 3, a PLC controller 4, and a human-machine interface 5; the PLC controller 4 is electrically connected to the gigabit switch 3 and the human-machine interface 5 respectively.

[0050] The camera 1 is communicatively connected to the gigabit switch 3; the gigabit switch 3 is communicatively connected to the host computer 6;

[0051] The PLC controller 4 is electrically connected to the L2 system 7;

[0052] The L2 system 7 is electrically connected to the bottom-blown argon control cabinet 8;

[0053] The camera 1 is used to collect images of the distribution of molten steel and slag layer in the ladle 9, and upload them to the host computer 6 through the gigabit switch 3.

[0054] The host computer 6 includes an image processing module and a data management module;

[0055] The image processing module is used to preprocess the image of the distribution of molten steel and slag layer in the ladle 9, and calculate the proportion of the exposed area of ​​molten steel in the ladle 9 based on image recognition technology.

[0056] The data management module establishes expert model data, which is used to obtain the control target value of the bottom blowing argon control cabinet 8 under the current stirring intensity from the expert model data, and to determine the argon blowing control amount under the current stirring intensity based on the proportion of the exposed area of ​​molten steel in the ladle 9 and the control target value.

[0057] The PLC controller 4 acquires the argon blowing control quantity under the current stirring intensity and transmits it to the L2 system 7. The L2 system 7 determines the timing and quantity of argon blowing from the bottom blowing argon control cabinet 8 to the ladle 9 based on the argon blowing control quantity.

[0058] The L2 system 7 uploads the basic information of the ladle and the argon blowing information to the host computer 6 through the PLC controller 4.

[0059] The host computer also includes: a human-machine interface and an expert system module;

[0060] The human-machine interface is used to display images of molten steel and slag in the ladle 9 in real time, as well as control information on the argon gas flow rate.

[0061] While the image processing module calculates the proportion of molten steel area in the ladle 9 in real time and the L2 system 7 controls the timing and amount of argon blowing in the bottom blowing argon control cabinet 8, the expert system module identifies the proportion of molten steel area in the image, extracts the feature parameters corresponding to the current image, and saves the feature parameters of the current image in the expert model data.

[0062] The system is also equipped with a camera protection device, which includes a dust removal device and a temperature transmitter.

[0063] A slight positive pressure is formed inside the camera protection device. High-pressure gas is used to evenly blow the camera through the tilt angle of the front aperture lens, thus achieving uniform blowing of the camera.

[0064] The system also includes a temperature transmitter for monitoring the temperature inside the camera protection device. When the temperature inside the camera protection device exceeds the set value, it automatically alarms and sends a signal to the PLC in the control box to execute the extension and retraction.

[0065] The temperature transmitter is used to monitor the temperature inside the camera protection device. When the temperature inside the camera protection device exceeds the set value, an alarm is automatically triggered.

[0066] The human-machine interface 5 and the human-machine interface are also used for users to set manual control parameters to realize manual mode.

[0067] The camera 1 is an infrared camera, and the working parameters of the infrared camera are: the detector element is 640*480; the working wavelength is 8-14um; and the frame rate is 50Hz.

[0068] The human-computer interaction interface is specifically used to display the liquid surface image of the ladle 9 in real time, use different colors to represent molten steel and slag, display the historical changes in the proportion of molten steel through trend curves, replay historical videos in real time, and query the historical curves of changes in the area of ​​molten steel during argon blowing.

[0069] The expert model data includes: the target proportion of exposed molten steel area under different stirring intensities, the qualified proportion of exposed molten steel area, the opening degree of the flow valve, and the flow range.

[0070] The human-machine interface is also equipped with an automatic alarm function, which will provide alarm reminders when a fault occurs or the argon gas flow reaches the upper and lower limits.

[0071] The image processing module has a processing time of within 50ms.

[0072] The information of ladle 9 obtained by the L2 system 7 is transmitted back to the system control module in real time. The information of ladle 9 includes: ladle number, steel type, argon blowing start and argon flow rate.

[0073] This invention further provides a visual judgment automatic argon blowing method for refining furnaces. The method includes: a camera capturing an image of the distribution of molten steel and slag layer on the surface of ladle 9, transmitting it to a host computer system and a PLC controller via a gigabit switch; the image processing module in the host computer system preprocesses the image of the distribution of molten steel and slag layer on the surface of ladle 9, and calculates the proportion of exposed molten steel area on the surface of ladle 9; then, based on the expert model data in the data management module and the proportion of exposed molten steel area on the surface of ladle 9 and the control target value, the argon blowing control amount under the current stirring intensity is determined, and it is determined whether the ladle has entered the predetermined position. When the ladle enters the predetermined position, the PLC controller 4) acquires the argon blowing control amount under the current stirring intensity and transmits it to the L2 system 7; the L2 system 7 controls the flow regulating valve in the bottom blowing argon control cabinet 8 to open according to the argon blowing control amount, and blows argon onto the ladle.

[0074] When the proportion of exposed molten steel in the ladle reaches the control target value, the L2 system controls the flow regulating valve in the bottom blowing argon control cabinet to close until the argon blowing ends;

[0075] The L2 system 7 uploads the basic information of the ladle and the argon blowing information to the host computer via the PLC controller 4.

[0076] The image of the distribution of molten steel and slag layer in the ladle 9 is preprocessed as follows: the features in the image of the distribution of molten steel and slag layer in the ladle 9 are preprocessed by increasing contrast and low-pass filtering, or one or more of these methods. Then, the proportion of exposed molten steel area is calculated by one or more methods, such as binary method and color histogram. The contour information of molten steel in the image is automatically extracted to determine the analysis area of ​​the image, and the molten steel and slag layer are identified in the analysis area of ​​the image using automatic threshold segmentation technology.

[0077] It should be noted that the image processing module in this invention adopts proprietary optimized image processing technology, making full use of the parallel processing capabilities of multi-core computers to perform real-time analysis of the images from camera 1, with the processing time controlled within 50ms.

[0078] Because of factors such as optical system distortion and atmospheric flow during photography, the quality of the acquired image may be reduced. Therefore, image augmentation techniques must be used to preprocess certain features of the captured image, such as enhancing contrast and low-pass filtering. Subsequently, the proportion of exposed steel area is calculated using methods such as binarization and color histogram.

[0079] After image acquisition, filtering is performed to automatically extract the contour information of molten steel in the image, determine the analysis area of ​​the image, and use automatic threshold segmentation technology to identify molten steel and slag in the area.

[0080] The host computer 6 is equipped with a synchronous display in the control room and a corresponding integrated display and operation screen in the outdoor field. It displays the liquid level image in the ladle in real time, so that the operator can monitor the whole process more intuitively. Different colors are used to represent molten steel and slag, and trend curves are used to display the historical changes in the molten steel ratio. Users can also play back historical videos in real time and query historical curves to analyze the changes in the molten steel area during the argon blowing process.

[0081] The image processing process of this invention includes an efficient image processing algorithm and has the following characteristics:

[0082] (1) Image acquisition, filtering, and extraction of image contour information;

[0083] (2) Determine the analysis area;

[0084] (3) Threshold segmentation technology is used to automatically identify molten steel and slag in the analysis area;

[0085] (4) Identify the characteristics of steel slag and prevent interference caused by factors such as flames and liquid surface tumbling during argon blowing;

[0086] (5) Color treatment: molten steel and slag are contrasted with different colors.

[0087] The operating parameters of the camera 1 are as follows: the detector element is 640*480; the working wavelength is 8-14um; the frame rate is 50Hz; and the maximum temperature measurement is 2000℃.

[0088] The process of color processing to analyze the distribution of molten steel and slag layers on the ladle surface involves: preprocessing certain features in the image using one or more methods, such as increasing contrast and low-pass filtering; calculating the proportion of the molten steel area using one or more methods, such as binary analysis and color histogram analysis; automatically extracting the contour information of the molten steel in the image to determine the analysis area; and using automatic threshold segmentation technology to identify the molten steel and slag layers within the analysis area.

[0089] The human-machine interface is specifically designed to display real-time images of the ladle surface, using different colors to represent molten steel and slag, and displaying historical changes in the proportion of molten steel through trend curves. It can also play back historical videos in real time and query historical curves showing changes in the area of ​​molten steel during argon blowing.

[0090] The data management module also stores a standard database, which includes the target proportion of exposed molten steel area under different stirring intensities, the qualified proportion of exposed molten steel area, the opening degree of the flow valve, and the flow range. After the image processing module calculates the proportion of exposed molten steel area, it obtains the corresponding argon blowing control amount based on the control target value under the current stirring intensity.

[0091] A standard database is established, containing constant data such as the target proportion of exposed molten steel area under various stirring intensities, the qualified proportion of exposed molten steel area, flow valve opening, and flow range. These parameters can be viewed and modified online within the software. After the image recognition function calculates the proportion of exposed molten steel area, the corresponding control quantity is obtained from an expert data model based on the control target value under the current stirring intensity.

[0092] Data management primarily involves storing and retrieving field application data and saving images of weak stirring. The data is indexed by furnace number and stored in a SQL Server database, allowing for future queries of historical data by furnace number. The images of weak stirring are saved as images on the computer's hard drive.

[0093] Expert system module:

[0094] As artificial intelligence is increasingly applied across various industries, intelligent systems need to possess relearning capabilities. As mentioned above, automatic argon blowing of molten steel is an intelligent system based on image recognition. The images originate from actual on-site conditions, and the core of this invention lies in accurately matching image features with the actual state. Therefore, while performing image recognition and control output, the system software can also recognize images with known results, extract the feature parameters corresponding to the current image, and store these parameters in expert model data, forming an expert system with deep learning capabilities.

[0095] Data communication module:

[0096] (1) Obtain information transmitted from the secondary system of the steel plant, such as: ladle number, steel type, argon blowing start, argon flow rate, etc.;

[0097] (2) Information transmitted back to the steel plant's secondary system: argon blowing control permission, argon flow rate increase or decrease, etc.;

[0098] (3) Obtain furnace information from steelmaking L2, process the data, and transmit it back to the steelmaking L2 system. (Host computer)

[0099] Human-machine interfaces have the following characteristics:

[0100] (1) An LCD monitor is installed in the control room, with an image area on the main screen for real-time display of the current molten steel and slag. The operator can observe the changes in the surface area of ​​the molten steel in the ladle during the argon blowing process through this monitor;

[0101] (2) During the automatic control process, the operation screen will display the ongoing operation, such as increasing or decreasing the argon flow rate.

[0102] (3) It is equipped with an alarm function. When the system malfunctions, such as when the argon flow rate reaches the upper or lower limit, it will prompt for handling.

[0103] The system in this invention also has the following technical effects:

[0104] a. It can automatically confirm whether the steel ladle has arrived;

[0105] b. The camera lens protection device is designed to be shockproof, dustproof, and high-temperature resistant;

[0106] c. Image transmission is equipped with anti-electromagnetic interference processing to avoid noise interference during image transmission;

[0107] d. Enable communication with the existing secondary systems of the steel plant;

[0108] e. The system uses color display to distinguish molten steel and slag by distinct colors, and creates a color transition effect based on the slag layer thickness. Operators can clearly distinguish the distribution of slag by looking at the color effect diagram.

[0109] f. The system can calculate the proportion of molten steel in the ladle in real time and automatically adjust the argon flow rate according to the set percentage.

[0110] g. Real-time monitoring of the change in the surface area of ​​molten steel in the ladle during the argon blowing process and automatic real-time video recording function;

[0111] h. A synchronous real-time display screen is installed in the control room so that operators can observe the changes in molten iron and slag in the ladle in real time;

[0112] i. The system can freely switch between manual and automatic argon blowing control;

[0113] j. Implement historical data query function. Operators can query historical data based on key information such as ladle number, argon blowing time, and steel type. After selecting the query conditions, the system will automatically load the corresponding slag removal historical video, historical curve, etc. The system has an automatic historical data deletion function and the historical data retention period can be set.

[0114] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A visually-based automatic argon blowing system for a refining furnace, characterized in that, Includes camera (1), instrument cabinet (2), host computer (6), L2 system (7) and bottom blowing argon control cabinet (8); The instrument cabinet (2) includes a gigabit switch (3), a PLC controller (4), and a human-machine interface (5); the PLC controller (4) is electrically connected to the gigabit switch (3) and the human-machine interface (5) respectively. The camera (1) is communicatively connected to the gigabit switch (3); the gigabit switch (3) is communicatively connected to the host computer (6); The PLC controller (4) is electrically connected to the L2 system (7); The L2 system (7) is electrically connected to the bottom-blown argon control cabinet (8); The camera (1) is used to collect images of the distribution of molten steel and slag layer in the ladle (9) and upload them to the host computer (6) through the gigabit switch (3). The host computer (6) includes an image processing module and a data management module; The image processing module is used to preprocess the image of the distribution of molten steel and slag layer in the ladle (9) liquid surface, and calculate the proportion of the exposed area of ​​molten steel in the current ladle (9) liquid surface based on image recognition technology. The data management module establishes expert model data, which is used to obtain the control target value of the bottom blowing argon control cabinet (8) under the current stirring intensity from the expert model data, and determine the argon blowing control amount under the current stirring intensity based on the proportion of exposed steel area in the ladle (9) liquid surface and the control target value. The PLC controller (4) acquires the argon blowing control quantity under the current stirring intensity and transmits it to the L2 system (7). The L2 system (7) determines the timing and amount of argon blowing from the bottom blowing argon control cabinet (8) to the ladle (9) based on the argon blowing control quantity. The L2 system (7) uploads the basic information of the ladle and the argon blowing information to the host computer (6) through the PLC controller (4); The host computer also includes: a human-machine interface and an expert system module; The human-machine interface is used to display images of molten steel and slag in the ladle (9) in real time, as well as control information on the flow rate of argon gas. The expert system module, while the image processing module calculates the image of the proportion of molten steel area in the ladle (9) in real time and the L2 system (7) controls the timing and amount of argon blowing in the bottom blowing argon control cabinet (8), identifies the image of the proportion of molten steel area, extracts the feature parameters corresponding to the current image, and saves the feature parameters of the current image in the expert model data. The system is also equipped with a camera protection device, which includes a dust removal and cleaning device. A slight positive pressure is formed inside the camera protection device. High-pressure gas is used to evenly blow the camera through the tilt angle of the front aperture lens, thus achieving uniform blowing of the camera. The human-machine interface (5) and the human-machine interface are also used for users to set manual control parameters to realize manual mode; The camera (1) is an infrared camera. The working parameters of the infrared camera are: the detector element is 640*480; the working wavelength is 8-14um; and the frame rate is 50Hz.

2. The automatic argon blowing system for visual judgment in a refining furnace according to claim 1, characterized in that, The human-machine interface is specifically used to display the liquid surface image of the ladle (9) in real time, use different colors to represent molten steel and slag, display the historical changes in the proportion of molten steel through trend curves, play back historical videos in real time, and query the historical curve of the change in the area of ​​molten steel during the argon blowing process.

3. The automatic argon blowing system for visual judgment in a refining furnace according to claim 1, characterized in that, The expert model data includes: the target proportion of exposed molten steel area under different stirring intensities, the qualified proportion of exposed molten steel area, the opening degree of the flow valve, and the flow range. The human-machine interface is also equipped with an automatic alarm function, which will provide an alarm reminder when a fault occurs or the argon gas flow reaches the upper and lower limits. The image processing module has a processing time of within 50ms.

4. The automatic argon blowing system for visual judgment in a refining furnace according to claim 1, characterized in that, The information of the ladle (9) obtained by the L2 system (7) is transmitted back to the system control module in real time. The information of the ladle (9) includes: ladle number, steel type, argon blowing start and argon flow rate.

5. A method for visually determining an automatic argon blowing system in a refining furnace according to any one of claims 1-4, characterized in that, The method includes: the camera (1) transmits the image of the distribution of molten steel and slag layer in the liquid surface of the ladle (9) to the host computer system and PLC controller through a gigabit switch; the image processing module in the host computer system preprocesses the image of the distribution of molten steel and slag layer in the liquid surface of the ladle (9) and calculates the proportion of the exposed area of ​​molten steel in the liquid surface of the ladle (9); then, based on the proportion of the exposed area of ​​molten steel in the liquid surface of the ladle (9) and the control target value, the expert model data in the data management module determines the current stirring intensity argon blowing control amount, and at the same time determines whether the ladle has entered the predetermined position. When the ladle enters the predetermined position, the PLC controller (4) obtains the current stirring intensity argon blowing control amount and transmits it to the L2 system (7); the L2 system (7) controls the flow regulating valve in the bottom blowing argon control cabinet (8) to open according to the argon blowing control amount and blows argon into the ladle. When the proportion of exposed steel surface in the ladle (9) reaches the control target value, the flow regulating valve in the bottom blowing argon control cabinet of the L2 system is closed until the argon blowing ends. The L2 system (7) uploads the basic information of the ladle and the argon blowing information to the host computer through the PLC controller (4).

6. The automatic argon blowing method for visual judgment in a refining furnace according to claim 5, characterized in that, The image of the distribution of molten steel and slag layer in the ladle (9) is preprocessed as follows: the features in the image of the distribution of molten steel and slag layer in the ladle (9) are preprocessed by increasing contrast and low-pass filtering, and then the proportion of exposed molten steel area is calculated by one or more methods such as binary method and color histogram; the contour information of molten steel in the image is automatically extracted, the analysis area of ​​the image is determined, and the molten steel and slag layer are identified in the analysis area of ​​the image using automatic threshold segmentation technology.

Images