Audio upper limit automatic adjustment method and system for driving sonar slagging through intelligent flame recognition

Abstract

The invention discloses an audio upper limit automatic adjustment method and system for driving sonar slagging through intelligent flame recognition. The system comprises a flame recognition module and a splashing line position adjustment module. The flame recognition module recognizes whether a splashing alarm exists or not, and the splashing line position adjusting module recognizes whether the splashing alarm exists or not according to flames and adjusts the position of a splashing line. By introducing the flame recognition technology, automatic correction and adjustment of the audio splashing alarm line are achieved, the accuracy and practicability of the sonar slagging system are remarkably improved, and the adjusted audio interval can accurately reflect the actual performance effect of slag during splashing and drying in the smelting process of the current heat; through the application of the method and the device, an on-site operator can obtain accurate slag state information in time, so that the state of the oxygen lance is adjusted more accurately, the phenomena of splashing and drying are effectively avoided, the steelmaking efficiency is improved, and potential safety hazards and environmental pollution risks are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of converter steelmaking, and in particular to a method and system for automatically adjusting the upper limit of an audio frequency for driving sonar slagging through intelligent flame recognition. Background Art

[0002] In the modern steel production process, converter steelmaking technology occupies a vital position. In order to improve steelmaking efficiency, reduce production costs, and meet increasingly stringent environmental protection requirements, automatic steelmaking and smart steelmaking technologies have been widely studied and applied. As one of the important technologies, sonar slag removal technology evaluates the state of slag by monitoring the audio signal of slag, thereby guiding the steelmaking process.

[0003] The existing converter audio slagging range is composed of two clearly defined curves, such as Figure 1 As shown. The upper blue solid line is defined as the splash warning line. When the actual measured audio signal line approaches or exceeds this line, it indicates that the converter has entered a splash state, and the area above the blue solid line is considered a splash risk area. The lower red dotted line represents the re-drying warning line. When the real-time audio signal line approaches or crosses this line, it indicates that the converter is in a re-drying state, and the area below the red dotted line is the re-drying risk area. Through these two clear warning lines, the audio slagging technology in the converter steelmaking process can more accurately guide the work of operators, thereby optimizing the steelmaking process, improving production efficiency and reducing safety hazards.

[0004] In actual applications, the effect of sonar slag removal technology is not good, mainly because the traditional splash warning line is a fixed curve (pre-set by the system), while in the actual smelting process, the splash warning line may be different from the setting, resulting in the inability to accurately determine whether it is splashing, which brings great interference to on-site use. In severe cases, it may even cause the sonar slag removal to be unusable and cause converter splash accidents.

[0005] The position of the splash line is restricted by many factors. First, the complexity of the converter conditions poses a challenge to the application of sonar slag removal technology. The high temperature, high pressure and complex chemical reactions inside the converter will interfere with the audio signal, seriously affecting the accuracy of the audio signal. Secondly, changes in scrap steel and slag conditions will also have a significant impact on the audio signal. Factors such as the shape, size and composition of the scrap steel, as well as the composition and fluidity of the slag will cause the audio signal to fluctuate, thereby affecting the interpretation and judgment of the audio signal. Finally, the size and height of the furnace mouth will also affect the collection of the audio signal, further increasing the difficulty of applying sonar slag removal technology.

[0006] Due to the influence of the above factors, sonar slag removal technology often faces the problem of inaccurate audio representation of slag conditions in practical applications. The theoretical splash warning line is inconsistent with the actual one, resulting in misjudgment and misoperation of the slag status, which will not only affect the stability and efficiency of the steelmaking process, but may also cause safety accidents and environmental pollution problems. This greatly limits the promotion and application of sonar slag removal technology and also affects the development of smart steelmaking technology. Summary of the invention

[0007] Purpose of the invention: The present invention provides a method and system for automatically adjusting the audio upper limit of sonar slagging driven by intelligent flame recognition, which can ensure the accuracy of audio characterization of slag conditions, improve the stability and efficiency of the steelmaking process, and promote the advancement of smart steelmaking technology.

[0008] Technical solution: The method for automatically adjusting the upper limit of the audio frequency of sonar slagging driven by intelligent flame recognition according to the present invention comprises the following steps: Step 1: Make initial adjustments within 40% of the total oxygen blowing volume; Step 2: If the audio continues to occur after adjustment in step 1 If the value reaches the splash warning line and there is no splash alarm in the flame recognition, continue to execute step 1 and adjust the splash line position to ; Step 3: If audio has been occurring since the start of refining until the current time period If the value reaches the splash warning line and the flame recognition has no splash alarm, the process of step 1 is repeated, and the final splash line position is adjusted to ; Step 4: After flame recognition splashing occurs, the adjustment measures in step 1 are no longer executed, and the loop of step 1 is exited; Step 5: When flame recognition splash occurs and the current audio When the value is less than the minimum audio value before this moment, the splash alarm line needs to be adjusted. The splash line adjustment position is the current Value position; Step 6. When the situation described in step 5 continues to occur, continue the adjustment method of step 5. Each subsequent adjustment is made once the situation described in step 5 occurs, and the splash alarm line will no longer be adjusted after the oxygen blowing volume reaches 80% of the total oxygen blowing volume.

[0009] Furthermore, in step 1, it is assumed that the initial splash warning line position is , flame recognition without splash alarm, if the audio is detected If the value is less than or equal to 0, that is, it reaches or exceeds the splash warning line position, the splash warning line needs to be adjusted. The splash warning line position is the position before adjustment plus , as shown in the following formula, the result is that the overall curve is The splash warning line and the back-drying warning line move in the same direction. (1).

[0010] Further, in step 2, the splash line position is adjusted to , as shown below: (2) in, is the splash line position after adjustment in the initial stage, The value to translate.

[0011] Furthermore, in step 3, the final splash line position is adjusted to , as shown below: (3) in, For the Adjust the front splash line position, The value to translate.

[0012] Further, in step 5, when the audio is actually If the value is less than or equal to 0, it reaches or even exceeds the splash warning line position. (4).

[0013] Correspondingly, an audio upper limit automatic adjustment system for sonar slagging driven by intelligent flame recognition includes: a flame recognition module and a splash line position adjustment module; the flame recognition module recognizes whether there is a splash alarm, and the splash line position adjustment module adjusts the splash line position according to whether there is a splash alarm according to the flame recognition.

[0014] Furthermore, when the oxygen blowing amount reaches more than 80% of the total oxygen blowing amount, the splash line position adjustment module no longer adjusts the splash alarm line.

[0015] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: the present invention introduces flame recognition technology to achieve automatic correction and adjustment of the audio splash alarm line, significantly improving the accuracy and practicality of the sonar slag system, so that the adjusted audio range can accurately reflect the actual performance of the slag during splashing and drying during the current furnace smelting process; through the application of the present invention, on-site operators can obtain accurate slag status information in a timely manner, thereby more accurately adjusting the oxygen lance status and effectively avoiding the occurrence of splashing and drying, which not only improves steelmaking efficiency, but also reduces safety hazards and environmental pollution risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic diagram of the audio slagging area of ​​the converter.

[0017] Figure 2 It is a schematic diagram of the method flow of the present invention.

[0018] Figure 3 This is a schematic diagram of the adjustment of the splash warning line of the present invention. DETAILED DESCRIPTION

[0019] like Figure 2 As shown, a method and system for automatically adjusting the audio upper limit of sonar slagging driven by intelligent flame recognition comprises the following steps: Step 1: Initial stage adjustment, which usually occurs within 40% of the total oxygen blowing volume. Assume that the initial splash warning line position is , flame recognition without splash alarm, if the audio is detected If the value is less than or equal to 0 (reaches or exceeds the splash warning line position), the splash warning line needs to be adjusted. The splash warning line position is the position before adjustment plus , as shown in formula (1), the result is that the overall curve is The splash warning line and the back-drying warning line move in the same direction. Figure 3 As shown, after translation, the splash warning line becomes a blue dotted line, and the back-drying warning line becomes a red double dotted line.

[0020] (1).

[0021] Step 2: If the audio continues to occur after step 1 is adjusted once If the value reaches the splash warning line and there is no splash alarm in the flame recognition, continue to execute step 1 and adjust the splash line position to , as shown in formula (2): (2).

[0022] Step 3: If audio has been occurring since the start of blowing until the current time period If the value reaches the splash warning line and the flame recognition has no splash alarm, the process of step 1 is repeated, and the final splash line position is adjusted to , as shown in formula (3).

[0023] (3).

[0024] in ≥0, is the splash line position after the i-th adjustment, For the Adjust the front splash line position again.

[0025] Step 4: After flame recognition splashing occurs, the adjustment measures in step 1 are no longer executed and the step 1 loop is exited.

[0026] Step 5: When flame recognition splashing occurs and the current audio p value is less than the minimum audio value before this moment, as shown in formula (4), the splashing alarm line needs to be adjusted. The splashing line adjustment position is the current Value position.

[0027] (4).

[0028] Step 6: When the situation described in step 5 continues to occur, continue with the adjustment method of step 5, and adjust once each time the situation in step 5 occurs subsequently, until the oxygen blowing volume reaches 80% of the total oxygen blowing volume and then the splash alarm line is no longer adjusted.

[0029] Example 1: Adjustment of the warning line during normal smelting process: During the normal smelting process of the converter, it is assumed that the initial splash warning line position is Set to 2. As smelting progresses, the system continuously monitors the actual audio value At some point, the system detects audio When the value reaches 0 (0 is the splash warning threshold), the flame recognition system does not alarm splashing. At this time, the system automatically adjusts the splash warning line position according to the algorithm and moves it up by 0.5. After adjustment, the splash warning line and the back-drying warning line are moved up accordingly. Figure 3 If the audio frequency continues during the smelting process When the value reaches the new splash warning line position again, the system will continue to adjust the warning line position according to the above rules until a stable smelting state is reached.

[0030] Example 2: Adjustment of the warning line after flame recognition splashing: In a certain smelting process, assuming that the splash warning line position is Set to 2, the system detects flame recognition splashing. At this time, the system stops executing the normal warning line adjustment mechanism (that is, it no longer adjusts according to the rules in Example 1). The value is 0.7, the position is 1, and the system finds the current audio The value is less than the minimum audio value before this moment (the minimum value is 1.2), indicating that the splash warning line position has not been determined before this point. The system adjusts the splash warning line position according to the algorithm, and the splash warning line position is adjusted from 2 to 1 to make it closer to the current furnace condition. This adjustment will continue until the oxygen blowing process reaches 80% progress, at which time the warning line adjustment stops to ensure the stability of the subsequent smelting process.

[0031] Converter flame recognition technology is an important application in the steelmaking industry. It uses AI vision technology to monitor and intelligently analyze the flame state in the converter smelting process in real time to improve steelmaking efficiency and safety. In the traditional steelmaking process, workers need to observe the flame in a high-temperature and dangerous environment. This work is not only labor-intensive, but also poses a threat to workers' health. With the development of industrial automation and intelligence, converter flame recognition technology has emerged, aiming to replace manual observation with a machine vision system to improve production efficiency and safety. It is mainly based on machine vision and deep learning algorithms. The image of the converter flame is captured by a high-definition industrial camera, and then the flame state is analyzed and identified using image processing technology and deep learning models (such as residual neural networks). These models can learn the characteristics of the flame at different smelting stages, so as to accurately judge whether the flame is normal, whether there is slag overflow, splashing, back drying, etc.

Claims

1. A method for automatically adjusting the upper limit of the audio frequency of sonar slagging driven by intelligent flame recognition, characterized in that: The steps include: Step 1: Make initial adjustments within 40% of the total oxygen blowing volume; Step 2: If the audio P value reaches the splash warning line after step 1 is adjusted once and there is no splash alarm in the flame identification, continue to execute step 1 and adjust the splash line position to ; Step 3: If the audio P value reaches the splash warning line and there is no splash alarm in the flame identification from the start of blowing to the current time period, the process of step 1 is repeated, and the splash line position is finally adjusted to L(i+1); Step 4: After flame recognition splashing occurs, the adjustment measures in step 1 are no longer executed, and the loop of step 1 is exited; Step 5: When flame recognition splashing occurs and the current audio p value is less than the minimum audio value before this moment, the splashing alarm line needs to be adjusted, and the splashing line adjustment position is the current p value position; Step 6. When the situation described in step 5 continues to occur, continue the adjustment method of step 5. Each subsequent adjustment is made once the situation described in step 5 occurs, and the splash alarm line will no longer be adjusted after the oxygen blowing volume reaches 80% of the total oxygen blowing volume.

2. The method for automatically adjusting the upper limit of the audio frequency of sonar slagging driven by intelligent flame recognition according to claim 1 is characterized in that: In step 1, it is assumed that the initial splash warning line position is , flame recognition without splash alarm, if the actual audio P value is detected to be less than or equal to 0, that is, it reaches or even exceeds the splash warning line position, then the splash warning line needs to be adjusted. The splash warning line position is the position before adjustment plus , as shown in the following formula, the result is that the overall curve is The splash warning line and the back-drying warning line move in the same direction. （1）。 3. The method for automatically adjusting the upper limit of the audio frequency of sonar slagging driven by intelligent flame recognition as claimed in claim 1, characterized in that: In step 2, the splash line position is adjusted to , as shown below: （2） in, is the splash line position after adjustment in the initial stage, The value to translate.

4. The method for automatically adjusting the upper limit of the audio frequency of sonar slagging driven by intelligent flame recognition according to claim 1, characterized in that: In step 3, the final splash line position is adjusted to , as shown below: （3） in, For the Adjust the front splash line position. The value to translate.

5. The method for automatically adjusting the upper limit of the audio frequency of sonar slagging driven by intelligent flame recognition as claimed in claim 1, characterized in that: In step 5, when the actual audio P value is less than or equal to 0, that is, it reaches or even exceeds the splash warning line position. （4）。 6. A system for realizing the method for automatically adjusting the upper limit of the audio frequency of sonar slagging driven by intelligent flame recognition as claimed in claim 1, characterized in that: include: Flame recognition module and splash line position adjustment module; The flame recognition module recognizes whether there is a splash alarm situation, and the splash line position adjustment module adjusts the splash line position according to whether there is a splash alarm situation according to the flame recognition.

7. The audio upper limit automatic adjustment system for sonar slagging driven by intelligent flame recognition as claimed in claim 6 is characterized in that: When the oxygen blowing amount reaches 80% of the total oxygen blowing amount, the splash line position adjustment module will no longer adjust the splash alarm line.

Images