Method for measuring opening degrees of smoke hood and furnace mouth of converter

Abstract

The invention relates to a method for measuring opening degrees of a converter smoke hood and a furnace mouth, and belongs to the technical field of dust removal in the metallurgical industry. According to the technical scheme, image data of the converter are transmitted to an identification computer (3), and the identification computer (3) reads blowing signals of the converter and judges whether the converter is in a blowing state or not; if the converter is in the blowing state, the recognition computer starts to read image data of the camera and carries out image analysis; and through an image algorithm, the system calculates the height difference between the converter and the converter smoke hood, namely the opening degree H of the converter smoke hood and the converter mouth, and the measurement result is displayed on a picture in a character form. The method has the beneficial effects that the investment cost is low, and workers at operation posts can clearly know the height of the converter smoke hood only by observing the monitoring screen, so that the labor intensity of workers is reduced; and meanwhile, whole-course online detection is realized, the problem of misjudgment caused by manual observation is reduced, and the intelligent control level is remarkably improved.

Description

Technical Field

[0001] The invention relates to a method for measuring the opening of a converter smoke hood and a furnace mouth, and belongs to the technical field of dust removal in the metallurgical industry. Background Art

[0002] In the blowing process of converter steelmaking, the height setting of the furnace mouth and the converter hood is a key factor, which directly affects the safety of operation, the dust removal effect and the production efficiency of the converter. The height setting of the converter hood is intended to effectively capture the dust, smoke and other waste gases generated during the converter steelmaking process. The converter hood is usually set above the furnace mouth, and its height setting should consider the following factors: Effective capture of waste gas: The converter hood needs to be in a position that can effectively capture waste gas, usually maintaining a certain distance from the furnace mouth to form a reasonable airflow channel. Avoid interference with blowing operations: The height of the converter hood should be moderate and not too low, so as not to interfere with blowing operations and furnace mouth observation. Maintenance convenience: The height of the converter hood also needs to take into account daily maintenance and overhaul. Summary of the invention

[0003] The purpose of the present invention is to provide a method for measuring the converter smoke hood and furnace mouth opening. The operating staff only needs to observe the monitoring screen to clearly understand the height of the converter smoke hood, thereby reducing the intensity of manual labor; at the same time, it realizes full-process online detection, reduces the misjudgment problem caused by manual observation, significantly improves the level of intelligent control, and solves the above-mentioned technical problems existing in the existing technology.

[0004] The technical solution of the present invention is: A method for measuring the opening of a converter smoke hood and a furnace mouth, characterized in that it includes the following steps: a monitoring system controls a camera to collect an image of a dust removal chimney of the converter, the monitoring system transmits the image data of the converter to a recognition computer through a switch, the recognition computer reads a blowing signal of the converter, and determines whether it is in a blowing state; if it is in a blowing state, the recognition computer starts to read the image data of the camera and performs image analysis; through an image algorithm, the system calculates the height difference between the converter and the converter smoke hood, that is, the opening H of the converter smoke hood and the converter furnace mouth, and displays the measurement result in text form on a screen.

[0005] The specific steps include: The image of the dust removal chimney collected by the camera is transmitted to the recognition computer through the switch; the recognition computer reads the converter blowing signal of the dust removal PLC to determine whether it is in the blowing state; if it is not in the blowing state, the recognition computer will not read the image of the camera; if it is in the blowing state, the recognition computer will read the image of the camera; After reading the image from the camera, the recognition computer uses pixel comparison to analyze the recognition area of ​​the image, which is divided into two steps: furnace mouth image recognition and converter hood image recognition. During the furnace mouth image recognition process, the converter hood image recognition program will not be entered; only after the furnace mouth image recognition is completed will the converter hood image recognition program be entered. The recognition program of the recognition computer first searches for the position of the converter mouth in the blowing process image; sets the pixel coordinate range of the recognition area in the captured image (x1, y1), (x1, y2), (x2, y1), (x2, y2), and the program presets multiple x0 values, and calculates the average brightness of two adjacent sliders on the image along the straight line x=x0 to make a gradient judgment. The slider is a vertical rectangle composed of several pixel points; the point with the largest brightness gradient difference is identified as the edge of the target; after successfully identifying multiple furnace mouth points, the program will start to identify the middle edge of the converter hood; the program presets the column where the converter hood edge is located, and scans along the column. The point with the largest brightness gradient difference is identified as the edge point of the converter hood; When the positions of the converter mouth and the converter hood are identified, the height between the converter mouth and the converter hood is calculated and the result is printed on the screen.

[0006] The more specific steps are as follows: The recognition computer analyzes the image data collected by the monitoring system frame by frame; the analysis program is written in Python and contains two threads, thread one and thread two; thread one is responsible for analyzing the image data collected by the monitoring system, while thread two is responsible for reading in real time whether the converter is currently in the blowing state; this process is carried out through the Modbus server to obtain the current status of the converter in real time.

[0007] The monitoring system is a Modbus server. Modbus is a protocol used for communication between electronic devices in industrial automation systems. It is based on a master-slave / client-server communication model and is mainly used for data exchange between programmable logic controllers (PLCs). The two threads share a global variable. When the converter is in a blowing state, the global variable is set to True, thereby allowing thread one to perform image data analysis.

[0008] The image data acquisition process is implemented by OpenCV, which is a computer vision library based on BSD license and is widely used in the fields of image processing and computer vision. After the recognition computer reads the data of the monitoring system, the image is analyzed by image pixel comparison, which is divided into two steps: image recognition of the furnace mouth and image recognition of the converter hood. After the image recognition of the furnace mouth is successful, the image recognition program of the converter hood is entered.

[0009] If the furnace is currently in the blowing state and the brightness of the current screen exceeds the preset value, the program will enter the furnace mouth identification process.

[0010] The program first searches for the position of the furnace mouth in the captured image; in the program, the pixel coordinate range of the recognition area is set to (x1, y1), (x1, y2), (x2, y1), (x2, y2); in the image, the horizontal axis is the x-axis and the vertical axis is the y-axis. The program sets multiple x-axis points, and scans each preset x-axis point along the y-axis direction in each frame; two sliders are set to scan the average brightness of the image; the average brightness of each of the two adjacent sliders is calculated, and the difference between the two is calculated, and the difference is the average brightness of the upper slider minus the average brightness of the lower slider; the point with the largest brightness difference is called the furnace mouth point of interest in the line, and is recorded as the current furnace mouth position. After the line is scanned, the recorded furnace mouth points are added to the furnace mouth position set of the line; when the number of elements in the furnace mouth position set exceeds the preset value, the majority in the set is selected as the vertical coordinate of the furnace mouth position of the scan line, and the furnace mouth height on the scan line can be determined.

[0011] If the furnace mouth heights on all scanning lines are not determined within a certain period of time, it means that the brightness difference in the undetermined scanning line area is small, there is no flame between the converter hood and the furnace mouth, and the height difference between the converter hood and the furnace mouth is considered to be 0.

[0012] After the y values ​​of all preset x-axis points are determined, the program will stop identifying the furnace mouth height and turn to identifying the converter hood height.

[0013] In order to prevent the converter hood from being blocked by the observation window and thus affecting the results, the program will preset an x0 value as the horizontal coordinate value of the converter hood, which is in the middle of the observation window, and then scan the slider on the straight line x=x0 where the value is located. When the brightness difference between the two sliders is the largest, the point is called the converter hood point of interest and is recorded as the current converter hood position. After each scan, the current converter hood coordinates are added to the converter hood position set. When the number of elements in the converter hood position set exceeds the preset value, the mode of the set is selected as the vertical coordinate of the converter hood position, and the converter hood coordinates can be determined. When selecting the mode of the set again after the first selection, it is necessary to judge the range of the set. If the range of the set exceeds the preset range, the set will be cleared and rescanned.

[0014] After successfully identifying the furnace mouth position and converter smoke hood position, select the point with the smallest ordinate from all the identified furnace mouth position coordinates as the highest furnace mouth coordinate. If the ordinate value of the highest furnace mouth is greater than the ordinate value of the converter smoke hood, start calculating the height between the furnace mouth and the converter smoke hood, that is, the opening H between the converter smoke hood and the converter furnace mouth. The calculation formula is as follows: H = (y炉口位置 -y 转炉烟罩位置 )×W y 炉口位置 Represents the ordinate of the pixel point at the blast furnace mouth obtained by real-time calculation of the image, y 转炉烟罩位置 is the ordinate of the converter hood point; W is a constant, indicating the actual distance corresponding to each pixel coordinate in the image; after the calculation is completed, y 炉口位置 ,y 转炉烟罩位置 And the height between the furnace mouth and the converter hood is printed on the image.

[0015] The beneficial effects of the present invention are as follows: the investment cost is low, and the operating staff only needs to observe the monitoring screen to clearly understand the height of the converter smoke hood, thereby reducing the labor intensity; at the same time, it realizes full-process online detection, reduces the misjudgment problem caused by manual observation, and significantly improves the level of intelligent control. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of an embodiment of the present invention; Figure 2 A schematic diagram of a system according to an embodiment of the present invention; Figure 3 This is a connection diagram of an embodiment of the present invention; Figure 4 This is a schematic diagram of the identification area according to an embodiment of the present invention; In the figure: monitoring system 1, dust removal PLC 2, identification computer 3, switch 4, identification area 5, converter 6, converter hood 7, scanning line 8, camera 9, observation window 10, converter hood and converter furnace opening H. DETAILED DESCRIPTION

[0017] The present invention will be further described below by way of embodiments in conjunction with the accompanying drawings.

[0018] A method for measuring the opening of a converter smoke hood and furnace mouth, characterized in that it includes the following steps: a monitoring system 1 controls a camera 9 to collect an image of a dust removal chimney of a converter 6, the monitoring system 1 transmits the image data of the converter to a recognition computer 3 through a switch 4, the recognition computer 3 reads a blowing signal of the converter to determine whether it is in a blowing state; if it is in a blowing state, the recognition computer will start to read the image data of the camera and perform image analysis; through an image algorithm, the system calculates the height difference between the converter and the converter smoke hood, that is, the opening H of the converter smoke hood and the converter furnace mouth, and displays the measurement result in text form on the screen.

[0019] The specific steps include: The image of the dust removal chimney collected by camera 9 is transmitted to the recognition computer through the switch; the recognition computer reads the converter blowing signal of the dust removal PLC2 to determine whether it is in the blowing state; if it is not in the blowing state, the recognition computer 3 will not read the image of camera 9; if it is in the blowing state, the recognition computer 3 will read the image of camera 9; After reading the image from the camera, the recognition computer analyzes the recognition area 5 of the image by pixel comparison, which is divided into two steps: furnace mouth image recognition and converter hood image recognition. During the furnace mouth image recognition process, the converter hood image recognition program will not be entered; the converter hood image recognition program will not be entered until the furnace mouth image recognition is completed. The recognition program of the recognition computer first searches for the position of the converter mouth in the blowing process image; sets the pixel coordinate range of the recognition area 5 in the captured image (x1, y1), (x1, y2), (x2, y1), (x2, y2), and the program presets multiple x0 values, and calculates the average brightness of two adjacent sliders on the image along the straight line x=x0 to make a gradient judgment, wherein the slider is a vertical rectangle composed of several pixel points; the point with the largest brightness gradient difference is identified as the edge of the target; after successfully identifying multiple furnace mouth points, the program will start to identify the middle edge of the converter hood; the program presets the column where the converter hood edge is located, and scans along the column, and the point with the largest brightness gradient difference is identified as the edge point of the converter hood; When the positions of the converter mouth and the converter hood are identified, the height between the converter mouth and the converter hood is calculated and the result is printed on the screen.

[0020] The more specific steps are as follows: The recognition computer 3 analyzes the image data collected by the monitoring system 1 frame by frame; the analysis program is written in Python and contains two threads, namely thread one and thread two; thread one is responsible for analyzing the image data collected by the monitoring system, while thread two is responsible for reading in real time whether the converter is currently in a blowing state; this process is carried out through the Modbus server to obtain the current state of the converter in real time.

[0021] The monitoring system 1 is a Modbus server. Modbus is a protocol used for communication between electronic devices in industrial automation systems. It is based on a master-slave / client-server communication model and is mainly used for data exchange between programmable logic controllers (PLCs). The two threads share a global variable. When the converter is in a blowing state, the global variable is set to True, thereby allowing thread 1 to perform image data analysis.

[0022] The image data acquisition process is implemented by OpenCV, which is a computer vision library based on BSD license and is widely used in the fields of image processing and computer vision. After the recognition computer reads the data of the monitoring system, the image is analyzed by image pixel comparison, which is divided into two steps: image recognition of the furnace mouth and image recognition of the converter hood. After the image recognition of the furnace mouth is successful, the image recognition program of the converter hood is entered.

[0023] If the furnace is currently in the blowing state and the brightness of the current screen exceeds the preset value, the program will enter the furnace mouth identification process.

[0024] The program first searches for the position of the furnace mouth in the captured image; in the program, the pixel coordinate range of the recognition area is set to (x1, y1), (x1, y2), (x2, y1), (x2, y2); in the image, the horizontal axis is the x-axis and the vertical axis is the y-axis. The program sets multiple x-axis points and scans each preset x-axis point along the y-axis direction in each frame; two sliders are set to scan the average brightness of the image; the brightness average of each of the two adjacent sliders is calculated, and the difference between the two is calculated, and the difference is the brightness average of the upper slider minus the brightness average of the lower slider; the point with the largest brightness difference is called the furnace mouth point of interest in the line and is recorded as the current furnace mouth position. After the line is scanned, the recorded furnace mouth points are added to the furnace mouth position set of the line; when the number of elements in the furnace mouth position set exceeds the preset value, the majority in the set is selected as the vertical coordinate of the furnace mouth position of the scan line, and the furnace mouth height on the scan line can be determined.

[0025] If the furnace mouth heights on all scanning lines are not determined within a certain period of time, it means that the brightness difference in the undetermined scanning line area is small, there is no flame between the converter hood and the furnace mouth, and the height difference between the converter hood and the furnace mouth is considered to be 0.

[0026] After the y values ​​of all preset x-axis points are determined, the program will stop identifying the furnace mouth height and turn to identifying the converter hood height.

[0027] In order to prevent the converter hood from being blocked by the observation window and thus affecting the results, the program will preset an x0 value as the horizontal coordinate value of the converter hood, which is in the middle of the observation window, and then scan the slider on the straight line x=x0 where the value is located. When the brightness difference between the two sliders is the largest, the point is called the converter hood point of interest and is recorded as the current converter hood position. After each scan, the current converter hood coordinates are added to the converter hood position set. When the number of elements in the converter hood position set exceeds the preset value, the mode of the set is selected as the vertical coordinate of the converter hood position, and the converter hood coordinates can be determined. When selecting the mode of the set again after the first selection, it is necessary to judge the range of the set. If the range of the set exceeds the preset range, the set will be cleared and rescanned.

[0028] After successfully identifying the furnace mouth position and converter smoke hood position, select the point with the smallest ordinate from all the identified furnace mouth position coordinates as the highest furnace mouth coordinate. If the ordinate value of the highest furnace mouth is greater than the ordinate value of the converter smoke hood, start calculating the height between the furnace mouth and the converter smoke hood, that is, the opening H between the converter smoke hood and the converter furnace mouth. The calculation formula is as follows: H = (y 炉口位置 -y 转炉烟罩位置 )×W y 炉口位置 Represents the ordinate of the pixel point at the blast furnace mouth obtained by real-time calculation of the image, y 转炉烟罩位置 is the ordinate of the converter hood point; W is a constant, indicating the actual distance corresponding to each pixel coordinate in the image; after the calculation is completed, y 炉口位置 ,y 转炉烟罩位置 And the height between the furnace mouth and the converter hood is printed on the image.

[0029] In the embodiments, Figure 2 Schematic diagram of the system, the outermost layer is an iron door, and an observation window 10 is arranged on the upper part of the iron door. The camera 9 collects images of the observation window.

[0030] Attached Figure 3 Schematic diagram of the embodiment. Figure 2 and attached Figure 3 , the monitoring system 1 controls the camera 9 to collect the converter image data through the observation window on the door, and transmits the image data to the recognition computer 3 through the switch 4. The recognition computer 3 reads the information of the dust removal PLC 2 to determine whether the converter is in the blowing state. If the converter is in the blowing state, the recognition computer 3 will perform image recognition on the image data, calculate the height difference between the converter hood and the converter mouth in the blowing state in real time, and display the result on the screen.

[0031] After receiving the image data, the recognition computer analyzes the recognition area 5 where the converter and converter hood are located in the image. If the current state is blowing, the average brightness of the current picture will be identified. If the average brightness exceeds 2, the program will enter the furnace mouth recognition process.

[0032] The pixel coordinates of the area of ​​interest set in this embodiment are (510, 340), (510, 685), (1560, 340) and (1560, 685). The program uses a vertical scanning method from top to bottom, and sets two sliders to scan the average brightness of adjacent areas. In this embodiment, the length of the slider is 3 pixels, the width is 1 pixel, and the two sliders are connected end to end. When scanning, the program will calculate the average brightness of the two sliders and the difference between the two. Each scan will record the point with the largest brightness difference. After each calculation, the two sliders slide down 3 pixels at the same time. After each scan is completed, the recorded point ordinate is added to the furnace mouth ordinate set. When the number of elements in the set exceeds 100, the mode of the ordinate set is selected as the ordinate of the furnace mouth point, and then the coordinates of the furnace mouth point are determined. In order to improve the accuracy of the image, it is necessary to identify multiple points of the furnace mouth. In this embodiment, the horizontal coordinates of seven furnace mouth positions are set, namely [620, 710, 820, 1035, 1240, 1320, 1400].

[0033] After identifying 2000 frames, if the furnace mouth coordinates of the scan line have not been completely determined, it means that the brightness difference of the undetermined scan line area is small, there is no flame between the converter hood and the furnace mouth, and the height difference between the converter hood and the furnace mouth is considered to be 0.

[0034] After all the furnace mouth points are determined, the coordinates of the converter smoke hood are identified. In this embodiment, the preset horizontal coordinate of the converter smoke hood is 1035. Similarly, the two sliders scan the straight line x=1035 to find the point with a larger average brightness difference of the vertical coordinate corresponding to the horizontal coordinate, and record it in the vertical coordinate set of the converter smoke hood. When the number of elements in the set exceeds 50, if the range of the set is less than 40, select the mode of the vertical coordinate in the set as the vertical coordinate of the converter smoke hood point, and then determine the coordinates of the converter smoke hood point. Then clear the set for the next round of scanning.

[0035] After identifying the coordinates of the furnace mouth and converter hood, select the point with the smallest ordinate from all the furnace mouth coordinates as the highest furnace mouth coordinate. If the ordinate of the highest furnace mouth is greater than the ordinate of the converter hood, it means that the identification result is valid. Calculate the height difference between the two coordinate points, that is, the opening H of the converter hood and the converter mouth. The calculation formula is as follows: H = (y 炉口位置 -y 转炉烟罩位置 )×W In this embodiment, the length W represented by each pixel is 0.21 cm. After successfully identifying all the positioning points of the furnace mouth, the program only identifies the coordinate points of the converter hood. If the coordinates of the converter hood are successfully identified, the height difference of the coordinate point will be calculated and displayed in real time on the image screen.

Claims

1. A method for measuring the opening of a converter hood and furnace mouth, characterized in that The following steps are involved: The monitoring system (1) controls the camera (9) to collect the image of the dust removal chimney of the converter (6). The monitoring system (1) transmits the image data of the converter to the recognition computer (3) through the switch (4). The recognition computer (3) reads the blowing signal of the converter to determine whether it is in a blowing state. If it is in a blowing state, the recognition computer will start to read the image data of the camera and perform image analysis. Through the image algorithm, the system calculates the height difference between the converter and the converter hood, that is, the opening H of the converter hood and the converter mouth, and displays the measurement result in text form on the screen.

2. A method for measuring the opening of a converter hood and furnace mouth according to claim 1, characterized in that The specific steps include: The camera (9) collects an image of the dust removal chimney and transmits it to the recognition computer through the switch; the recognition computer reads the converter blowing signal of the dust removal PLC (2) to determine whether it is in a blowing state; if it is not in a blowing state, the recognition computer (3) will not read the image of the camera (9); if it is in a blowing state, the recognition computer (3) will read the image of the camera (9); After reading the image from the camera, the recognition computer uses pixel comparison to analyze the recognition area (5) of the image, which is divided into two steps: furnace mouth image recognition and converter hood image recognition. During the furnace mouth image recognition process, the converter hood image recognition program will not be entered; the converter hood image recognition program will not be entered until the furnace mouth image recognition is completed. The recognition program of the recognition computer first searches for the position of the converter mouth in the blowing process image; sets the pixel coordinate range of the recognition area (5) in the captured image (x1, y1), (x1, y2), (x2, y1), (x2, y2), the program presets multiple x0 values, calculates the average brightness of two adjacent sliders on the image along the straight line x=x0, and performs gradient judgment, wherein the slider is a vertical rectangle composed of several pixel points; the point with the largest brightness gradient difference is identified as the edge of the target; after successfully identifying multiple furnace mouth points, the program will start to identify the middle edge of the converter hood; the program presets the column where the converter hood edge is located, scans along the column, and the point with the largest brightness gradient difference is identified as the edge point of the converter hood; When the positions of the converter mouth and the converter hood are identified, the height between the converter mouth and the converter hood is calculated and the result is printed on the screen.

3. The method for measuring the converter hood and furnace opening according to claim 2, characterized in that: After successfully identifying the furnace mouth position and converter smoke hood position, select the point with the smallest ordinate from all the identified furnace mouth position coordinates as the highest furnace mouth coordinate. If the ordinate value of the highest furnace mouth is greater than the ordinate value of the converter smoke hood, start calculating the height between the furnace mouth and the converter smoke hood, that is, the opening H between the converter smoke hood and the converter furnace mouth. The calculation formula is as follows: H =( y 炉口位置 - y 转炉烟罩位置 )×W y 炉口位置 Represents the ordinate of the pixel point at the blast furnace mouth obtained by real-time calculation of the image, y 转炉烟罩位置 is the ordinate of the converter hood point; W is a constant, which represents the actual distance corresponding to each pixel coordinate in the image. 炉口位置 ,y 转炉烟罩位置 And the height between the furnace mouth and the converter hood is printed on the image.

Images