Converter mouth flame AI analysis automatic steelmaking system

Abstract

The invention discloses a converter mouth flame AI analysis automatic steelmaking system. The converter, the front-end AI device and the rear-end monitoring device are arranged on the converter platform; the front-end AI device comprises a flame collection unit and an AI temperature analysis unit, the AI temperature analysis unit is electrically connected with the flame collection unit, and the flame collection unit is arranged right in front of the converter and used for collecting flame data of the converter in real time; the field controller and the flame acquisition unit are respectively in communication connection with the AI server, and the AI server is also in communication connection with the steel mill PLC control system; the AI server performs preprocessing based on the flame data acquired by the flame acquisition unit, AI analyzes the temperature and the steelmaking process, judges the reaction condition of molten steel in the converter, determines the control parameters of the converter, and feeds back the control parameters to the steel mill PLC control system; the intelligent steel-making system achieves intelligentization and automation of steel-making, and plays an important role in improving steel-making quality and efficiency development.

Description

Technical Field

[0001] The invention relates to the technical field of metallurgy and ironmaking, and more particularly to an automatic steelmaking system for AI analysis of converter mouth flame. Background Art

[0002] In the traditional converter steelmaking process, steel workers need to observe the flame at the furnace mouth with their naked eyes, and make empirical judgments on the reaction of the molten steel in the furnace based on the characteristics and changes of the flame, so as to adjust the steelmaking parameters. This method not only has safety hazards, but also requires high skills of workers, and long-term observation can easily damage eyesight. In addition, this method that relies on manual judgment often has errors, making it difficult to ensure the quality and efficiency of steelmaking.

[0003] Therefore, it is urgent to propose a flame AI analysis automatic steelmaking system that has high reliability, real-time performance and high safety to solve the problems in existing automatic steelmaking. Summary of the invention

[0004] In view of this, the purpose of the present invention is to propose an automatic steelmaking system for converter mouth flame AI analysis to solve the problems existing in the above-mentioned prior art.

[0005] To achieve the above object, the present invention provides the following technical solutions: A converter mouth flame AI analysis automatic steelmaking system, comprising: a converter arranged on a converter platform, a front-end AI device and a back-end monitoring device; The front-end AI device includes a flame collection unit and an AI temperature analysis unit, wherein the AI ​​temperature analysis unit is electrically connected to the flame collection unit, and the AI ​​temperature analysis unit is connected to the field controller; the flame collection unit is arranged in front of the converter and is used to collect flame data of the converter in real time; The back-end monitoring device includes an AI server, the field controller and the flame collection unit are respectively connected to the AI ​​server in communication, and the AI ​​server is also connected to the steel plant PLC control system in communication; The AI ​​server performs preprocessing based on the flame data collected by the flame collection unit. The AI ​​analyzes the temperature and steelmaking process, determines the reaction of the molten steel in the converter, determines the control parameters of the converter, and feeds back to the steel plant PLC control system.

[0006] Furthermore, the back-end monitoring device also includes a gigabit switch, the flame collection units are respectively communicatively connected to the gigabit switch, and the gigabit switch is communicatively connected to the AI ​​server.

[0007] Furthermore, the AI ​​server includes a data management module, a data processing module and an AI algorithm module; The data management module acquires the steel plant PLC signal sent by the PLC control system in real time; The data processing module is used to pre-process the flame change data in the converter, and perform real-time analysis on the pre-processed flame change data based on the AI ​​algorithm module to obtain the reaction of the molten steel in the converter and send it to the data management module; The data management module also establishes a standard database, which is used to obtain the target value of the converter parameter adjustment under the current process based on the reaction of the molten steel in the converter from the standard database; based on the target value of the converter parameter adjustment under the current process, it is compared with the steel plant PLC signal to determine the control parameters of the converter under the current process, and fed back to the steel plant PLC control system; The steel plant PLC signal includes an oxygen lance position signal and a slag addition amount signal.

[0008] Furthermore, the flame data collected by the flame collection unit is preprocessed, and AI analyzes the temperature and steelmaking process, specifically: The AI ​​server performs AI analysis on the flame temperature and steelmaking process based on the flame color, shape, grayscale, and brightness change data collected by the flame collection unit, and determines the reaction of the molten steel in the converter. Specifically, the AI ​​server uses a deep learning algorithm to extract and process the features of the flame color, brightness, and shape, and determines the splashing, slag condition, carbon content, molten steel temperature, and endpoint temperature prediction in the converter.

[0009] Furthermore, the control parameters of the converter are determined, specifically including: adjusting the oxygen lance position, the amount of slag added, the temperature and chemical composition of the molten steel in the converter, and automatically performing the steel tapping operation when the preset indicators are reached.

[0010] Furthermore, the AI ​​server is also provided with a human-computer interaction interface, which is used to display the flame change trend, oxygen filling amount, re-blowing time, gun position adjustment information and feeding information in real time.

[0011] Furthermore, the AI ​​server is also provided with an alarm control unit, which is used to analyze the temperature data in the converter and compare it with the temperature data under the current process in the standard database; when the real-time temperature distribution data exceeds the preset value of the temperature data under the current process, an alarm is issued.

[0012] Furthermore, the flame collection unit is specifically a high-definition area array camera.

[0013] Furthermore, the AI ​​temperature analysis unit is specifically a GPU server.

[0014] Furthermore, the flame collection unit is provided with a high-temperature air-cooled protective cover.

[0015] According to the specific embodiments provided by the present invention, the present invention solves the key details of steelmaking such as temperature, slag condition, splashing, carbon content, etc., changes the traditional steelmaking mode of viewing through an observation window, and adopts advanced AI algorithms to provide operators with more realistic flame forms and artificial intelligence-assisted decision-making, which is convenient for production control to save energy and reduce consumption in steelmaking, and to speed up the pace of steelmaking, reduce accidents, eliminate safety hazards, and realize the intelligence and automation of steelmaking, which plays an important role in improving the quality and efficiency of steelmaking. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0017] The automatic steelmaking system for converter mouth flame AI analysis of the present invention will be further described below in conjunction with the accompanying drawings; Figure 1 It is the overall structure diagram of the converter mouth flame AI analysis automatic steelmaking system provided by the present invention; Figure 2 It is a schematic diagram of the screen display of a server in the automatic steelmaking system for AI analysis of converter mouth flame provided by the present invention.

[0018] In the figure: 1. Converter; 2. Converter platform; 3. Field controller; 4. AI temperature analysis unit; 5. Flame collection unit; 6. Gigabit switch; 7. AI server; 8. Human-computer interaction interface. DETAILED DESCRIPTION

[0019] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0020] In order to better understand the purpose, structure and function of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings.

[0021] like Figure 1 As shown, the present invention provides an automatic steelmaking system for AI analysis of converter mouth flame, a converter 1, a front-end AI device and a back-end monitoring device arranged on a converter platform 2; The front-end AI device includes a flame collection unit 5 and an AI temperature analysis unit 4, wherein the AI ​​temperature analysis unit 4 is electrically connected to the flame collection unit 5, and the AI ​​temperature analysis unit 4 is connected to the field controller 3; the flame collection unit 5 is arranged in front of the converter 1, and is used to collect flame data of the converter in real time; The back-end monitoring device includes an AI server 7, the field controller 3 and the flame collection unit 5 are respectively connected to the AI ​​server 7 for communication, and the AI ​​server 7 is also connected to the steel plant PLC control system for communication; The AI ​​server 7 performs preprocessing based on the flame data collected by the flame collection unit 5, and the AI ​​analyzes the temperature and steelmaking process, determines the reaction of the molten steel in the converter 1, determines the control parameters of the converter 1, and feeds back to the steel plant PLC control system; It should be noted that: AI analyzes temperature and process through feature processing and python algorithm training on the AI ​​server.

[0022] The back-end monitoring device further includes a gigabit switch 6, the flame collection units 5 are respectively connected to the gigabit switch 6 for communication, and the gigabit switch 6 is connected to the AI ​​server 7 for communication; The AI ​​server 7 includes a data management module, a data processing module and an AI algorithm module; The data management module acquires the steel plant PLC signal sent by the PLC control system in real time; The data processing module is used to preprocess the flame change data in the converter 1, and perform real-time analysis on the preprocessed flame change data based on the AI ​​algorithm module to obtain the reaction of the molten steel in the converter 1 and send it to the data management module; The data management module also establishes a standard database, which is used to obtain the target value of the converter parameter adjustment under the current process based on the reaction of the molten steel in the converter 1 from the standard database; based on the target value of the converter parameter adjustment under the current process, it is compared with the steel plant PLC signal to determine the control parameters of the converter 1 under the current process, and fed back to the steel plant PLC control system; The steel plant PLC signal includes an oxygen lance position signal and a slag addition amount signal.

[0023] It should be noted that the system is connected to the steel plant PLC signal, and the oxygen filling amount, re-blowing time and other information can be superimposed on the screen. At the same time, the system can automatically adjust the camera exposure time according to different processes to achieve the best display effect and provide an operating basis for operators.

[0024] It should be noted that the flame acquisition unit obtains the image of the converter mouth flame in real time, and then uses the AI ​​algorithm model to identify and analyze the flame image. By extracting and processing the color, grayscale, brightness, shape and other characteristics of the flame, the system can judge the reaction of the molten steel in the furnace, such as abnormal splashing and slag jumping, and changes in the hardness and softness of the flame.

[0025] Based on the above analysis, the system can automatically adjust the production parameters in the steelmaking process, such as the oxygen lance position, slag addition amount, etc., to ensure the stability and efficiency of the steelmaking process. At the same time, the system can also monitor the temperature and chemical composition of the molten steel in the furnace in real time, and automatically perform the steel tapping operation when the preset indicators are reached.

[0026] It should be noted that this system can realize functions such as dynamic image adjustment, temperature curve calculation, converter status monitoring, etc., and also supports multi-level user authority management to reduce the risk of misoperation and production data leakage.

[0027] Through the implementation of the present invention, the steel smelting industry can realize the automation and intelligence of the steelmaking process, which has important practical application value for improving the quality and efficiency of steelmaking and reducing the labor intensity and safety risks of workers.

[0028] The flame data collected by the flame collection unit 5 is preprocessed, and AI analyzes the temperature and steelmaking process, specifically: The AI ​​server 7 performs AI analysis on the temperature and steelmaking process of the flame based on the color, shape, grayscale and brightness change data of the flame collected by the flame collection unit 5, and determines the reaction of the molten steel in the converter 1. Specifically, the AI ​​server 7 uses a deep learning algorithm to extract and process the color, brightness and shape of the flame, and determines the splashing, slag condition, carbon content, molten steel temperature and terminal temperature prediction in the converter 1. It should be noted that the flame AI analysis based on the flame change data to determine the reaction of the molten steel in the converter includes: using a deep learning algorithm to extract and process the color, brightness, and shape of the flame to determine the abnormal state of slag splashing and jumping in the converter 1 and the changes in the softness and hardness of the flame.

[0029] The control parameters of the converter 1 are determined, specifically including: adjusting the oxygen lance position, the amount of slag added, the temperature and chemical composition of the molten steel in the converter 1, and automatically performing the steel tapping operation when the preset indicators are reached.

[0030] like Figure 2 As shown, the AI ​​server 7 is also provided with a human-computer interaction interface 8, and the human-computer interaction interface 8 is used to display the flame change trend, oxygen filling amount, supplementary blowing time, gun position adjustment information and feeding information in real time.

[0031] The flame collection unit 5 is specifically a flame collection high-definition area array camera.

[0032] It should be noted that: the present invention overcomes the shortcomings of ordinary cameras that are difficult to reflect the details and temperature of the furnace mouth flame. It dynamically controls the camera parameters through AI algorithms, accurately displays the flame intensity, color and shape, analyzes the temperature according to the flame color AI, and provides steel plant operators with decarbonization, desulfurization, dephosphorization, deoxidation, slag splashing, splashing temperature control and other operations. It plays an auxiliary role, allowing operators to adjust the gun position in real time, and the addition of materials and other controls plays a key role, which has good economic benefits for increasing efficiency and reducing losses and eliminates safety hazards.

[0033] The server is also provided with an alarm control unit, which is used to analyze the real-time temperature data in the converter and compare it with the temperature data under the current process in the standard database; when the real-time temperature distribution data exceeds the preset value of the temperature data under the current process, an alarm is issued.

[0034] The present invention plays a key role in energy saving and consumption reduction, speeding up the steelmaking rhythm and reducing accidents by changing the traditional steelmaking mode of watching through an observation window. It also has the following technical solutions and effects: With advanced machine vision systems, operators can see flames more realistically: The system uses an advanced AI vision system to provide operators with more realistic flame forms, facilitating production control.

[0035] Using industrial-grade cameras with higher signal-to-noise ratio and stronger anti-interference ability: Using industrial-grade chips, the camera has enhanced anti-electromagnetic interference performance in view of the harsh electromagnetic environment in the industrial field.

[0036] Using advanced AI color intelligent temperature algorithm Intelligent analysis: The system dynamically controls camera parameters through software program algorithms, accurately displays flame intensity, color and shape, intelligently analyzes temperature based on flame color grayscale values, and provides steel plant operators with details such as slag overflow, slag pile, decarbonization, splashing, slag splashing, and carbon content.

[0037] The system is reliable, real-time, and has good economic benefits. It has a high production guidance function and plays a key role in energy conservation and consumption reduction in steelmaking, accelerating the pace of steelmaking, and reducing accidents. It provides strong support for steel mills to save costs, reduce losses, and improve production efficiency.

[0038] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one 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 present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A converter mouth flame AI analysis automatic steelmaking system, characterized in that: include: A converter (1), a front-end AI device, and a back-end monitoring device arranged on a converter platform (2); The front-end AI device comprises a flame collection unit (5) and an AI temperature analysis unit (4); the AI ​​temperature analysis unit (4) is electrically connected to the flame collection unit (5); and the AI ​​temperature analysis unit (4) is connected to the field controller (3); the flame collection unit (5) is arranged in front of the converter (1) and is used to collect flame data of the converter in real time; The back-end monitoring device comprises an AI server (7), the field controller (3) and the flame collection unit (5) are respectively connected to the AI ​​server (7) for communication, and the AI ​​server (7) is also connected to the steel plant PLC control system for communication; The AI ​​server (7) performs preprocessing based on the flame data collected by the flame collection unit (5), and the AI ​​analyzes the temperature and steelmaking process, determines the reaction of the molten steel in the converter (1), determines the control parameters of the converter (1), and feeds back to the steel plant PLC control system.

2. The automatic steelmaking system for converter mouth flame AI analysis according to claim 1 is characterized in that: The back-end monitoring device further comprises a gigabit switch (6), the flame collection units (5) are respectively connected in communication with the gigabit switch (6), and the gigabit switch (6) is connected in communication with the AI ​​server (7).

3. The automatic steelmaking system for converter mouth flame AI analysis according to claim 1 is characterized in that: The AI ​​server (7) comprises a data management module, a data processing module and an AI algorithm module; The data management module acquires the steel plant PLC signal sent by the PLC control system in real time; The data processing module is used to pre-process the flame change data in the converter (1), and to perform real-time analysis on the pre-processed flame change data based on the AI ​​algorithm module, to obtain the reaction status of the molten steel in the converter (1), and to send the result to the data management module; The data management module also establishes a standard database for obtaining from the standard database target values ​​for adjusting parameters in the converter under the current process based on the reaction of the molten steel in the converter (1); based on the target values ​​for adjusting parameters in the converter under the current process, the target values ​​are compared with the steel plant PLC signal to determine the control parameters for the converter (1) under the current process, and the control parameters are fed back to the steel plant PLC control system; The steel plant PLC signal includes an oxygen lance position signal and a slag addition amount signal.

4. The automatic steelmaking system for converter mouth flame AI analysis according to claim 1 is characterized in that: The flame data collected by the flame collection unit (5) is preprocessed, and the AI ​​analyzes the temperature and steelmaking process, specifically: The AI ​​server (7) performs AI analysis on the flame temperature and steelmaking process based on the flame color, shape, grayscale, and brightness change data collected by the flame collection unit (5), and determines the reaction of the molten steel in the converter (1). Specifically, the AI ​​server (7) uses a deep learning algorithm to extract and process the features of the flame color, brightness, and shape, and determines the splashing, slag condition, carbon content, molten steel temperature, and terminal temperature prediction in the converter (1).

5. The automatic steelmaking system for converter mouth flame AI analysis according to claim 1 is characterized in that: The control parameters of the converter (1) are determined, specifically including: adjusting the oxygen lance position, the amount of slag added, the temperature and chemical composition of the molten steel in the converter (1), and automatically performing the steel tapping operation when the preset indicators are reached.

6. The automatic steelmaking system for converter mouth flame AI analysis according to claim 1 is characterized in that: The AI ​​server (7) is also provided with a human-computer interaction interface (8), and the human-computer interaction interface (8) is used to display the flame change trend, oxygen filling amount, supplementary blowing time, gun position adjustment information and feeding information in real time.

7. The automatic steelmaking system for converter mouth flame AI analysis according to claim 3 is characterized in that: The AI ​​server (7) is also provided with an alarm control unit, which is used to analyze the temperature data in the converter and compare it with the temperature data under the current process in the standard database; when the real-time temperature distribution data exceeds the preset value of the temperature data under the current process, an alarm is issued.

8. The automatic steelmaking system for converter mouth flame AI analysis according to claim 1 is characterized in that: The flame collection unit (5) is specifically a high-definition area array camera.

9. The automatic steelmaking system for converter mouth flame AI analysis according to claim 1, characterized in that: The AI ​​temperature analysis unit (4) is specifically a GPU server.

10. The automatic steelmaking system for converter mouth flame AI analysis according to claim 1, characterized in that: The flame collection unit (5) is provided with a high-temperature air-cooling protective cover.

Images