A real-time slag quantity identification method and system for automatic slag dumping of a converter

Abstract

The application discloses a slag quantity real-time identification method and system for automatic slag pouring of a converter, and comprises the following steps: S1, obtaining the real-time thickness of a slag layer at any horizontal position of a slag pouring area; S2, collecting a real-time planar image of the slag layer in the vertical direction of the thickness, multiplying the thickness of the slag layer by the area of the slag layer in the planar area within a time interval to obtain a micro-layer slag volume; S3, multiplying the micro-layer slag volume by the density and conversion coefficient of the slag layer to obtain a micro-layer slag quantity value, continuously accumulating the real-time slag quantity value of the micro-layer to obtain a total slag quantity value of the slag poured into a slag tank. The application can rapidly, timely and accurately measure, calculate and feedback the weight of the slag poured out.

Description

Technical Field

[0001] This invention relates to steelmaking technology, and more specifically, to a method and system for real-time identification of slag quantity in an automatic converter slag dumping system. Background Technology

[0002] In the field of steelmaking technology, intelligent converter technology is constantly developing. The essence of the converter steelmaking process is the slag-making process, or slag-forming process. Generally, there are three operating methods for slag forming: single-slag method, double-slag method, and double-slag retention method.

[0003] Among them, the single-slag method involves slag formation during the blowing process without dumping or removing slag midway. This method is simple to operate, but the dephosphorization effect is relatively low, and it is only used for smelting general carbon steel and low alloy steel with low requirements for sulfur and phosphorus. The double-slag method involves dumping about 50%-65% of the slag during the blowing process and then adding slag material to form new slag. This method has a better dephosphorization and desulfurization effect and can be used when the phosphorus content in the molten iron is high and when blowing high carbon steel or molten iron with high silicon content to prevent splashing, or when blowing low manganese steel to prevent manganese reversion. Finally, the double-slag retention method refers to leaving part of the high-basicity, high-temperature, and high-FeO content final slag from the previous furnace in the furnace to accelerate the formation of the initial slag for the next furnace. During the blowing process, part of the slag is dumped out to form new slag. This method has high desulfurization and dephosphorization efficiency and is used in the smelting of steel grades with high requirements for steel quality.

[0004] Both the double-slag method and the double-slag retention method mentioned above require a mid-process slag dumping operation, that is, a certain amount of slag is dumped during the blowing process. In actual on-site production, the total amount of slag to be dumped can generally be determined by real-time weighing of the slag. However, the slag car weighing method has the following two shortcomings:

[0005] 1) Adding weighing equipment to slag trucks requires significant on-site modifications, which takes a long time and is costly;

[0006] 2) Real-time weighing of slag cars has a time lag and a significant impact. Because the slag dumping speed is very fast, the weighing data can only be fed back after the slag has fallen into the slag hopper. Therefore, in the later stages of slag dumping, the slag dumping process will have serious deviations. However, the present invention can determine the size of the slag about to fall from above the slag hopper, which can effectively improve the accuracy of slag identification.

[0007] The existing publication CN103397134A discloses a method for calculating the amount of slag left based on the tilting angle. This method involves measuring the molten steel level using a secondary lance, and then calculating the difference between the total volume of the molten steel inside the furnace and the remaining total volume of the molten steel using the ladle dimensions, furnace lining erosion data, and the tilting angle. This difference represents the volume of slag left in the converter. Multiplying this volume by the slag density yields the amount of slag left.

[0008] The existing publication number CN107502698A is applicable to an automated steelmaking method for low-slag smelting. It discloses a slag dumping and slag retention calculation model. The slag retention model is based on slag car weighing technology and calculates the difference between the slag car's weight before and after collection in real time, i.e., the slag dumping weight, thereby realizing the slag retention smelting model control method.

[0009] However, the accuracy of slag quantity control and slag retention control in the two patented technologies mentioned above is not high. This is partly due to the complexity of on-site conditions and environment, and partly due to the low accuracy of slag car weighing control technology. From domestic and international technology research, foreign companies such as Nippon Steel and POSCO do not have real-time online measurement methods for converter slag thickness. Conventional image recognition technology can only calculate the slag quantity from a fuzzy planar image, resulting in large errors. To date, there are no publicly available documents or reports on accurately measuring the amount of converter slag. Summary of the Invention

[0010] In view of the above-mentioned defects in the existing technology, the purpose of this invention is to provide a method and system for real-time identification of slag quantity in automatic slag dumping of converters, which can quickly, timely and accurately measure, calculate and feedback the weight of dumped slag in real time.

[0011] To achieve the above objectives, the present invention adopts the following technical solution:

[0012] On the one hand, a method for real-time identification of slag quantity in an automatic converter slag discharge includes the following steps:

[0013] S1. Obtain the real-time thickness of the slag layer at any horizontal position in the dumping slag area;

[0014] S2. Acquire a real-time planar image of the slag layer in the direction perpendicular to its thickness, and multiply the thickness of the slag layer by the area of ​​the slag layer in the planar region within the time interval to obtain the volume of the micro-layer slag.

[0015] S3. Multiply the volume of the micro-layer slag by the density and conversion factor of the slag layer to obtain the slag quantity value of the micro-layer. Continuously accumulate the real-time slag quantity values ​​of the micro-layer to obtain the total slag quantity value that has been poured into the slag tank.

[0016] Preferably, in step S1, a laser profile sensor is arranged at the same horizontal position on both sides of the dumping slag area to measure the front and rear waveform curves of the dumping slag area in real time, and the real-time thickness of the slag layer at the horizontal position is obtained by the distance difference between the two waveform curves.

[0017] Preferably, the laser profile sensor is located between 1000mm and 8000mm away from the dumping area, the laser sampling frequency is greater than 100Hz, and the distance measurement repeatability is less than 50μm.

[0018] Preferably, in step S2, a free fall model and / or a free fall motion model with an initial velocity are used, and real-time planar images are acquired by a high-speed camera.

[0019] The high-speed camera has a frame rate greater than 30 FPS.

[0020] Preferably, the calculation of the area of ​​the slag layer specifically includes the following steps:

[0021] S21. Acquire a real-time video stream using the high-speed camera;

[0022] S22. Extract an image of the slag flow region from the video stream;

[0023] S23. The image of the slag flow area is binarized, with the grayscale value of the background pixels being 0 and the grayscale value of the pixels in the slag flow area being 1.

[0024] S24. Accumulate the pixel grayscale values ​​of the slag flow micro-layer region, and the sum is the characterization value of the real-time micro-slag layer area.

[0025] Preferably, when the ladle is nearly horizontally inverted (0-10 degrees), the free fall model is used; when the ladle is tilted or inverted at a large angle (greater than 10 degrees), the free fall motion model with an initial velocity is used, with the initial velocity being 0.01-0.3 m / s, which is related to the ladle's tilt angle.

[0026] Preferably, in step S3, the total slag volume is calculated using the following formula:

[0027]

[0028] On the other hand, a real-time slag quantity identification system for automatic slag removal in a converter includes:

[0029] Converter;

[0030] A converter tilting drive device is connected to the converter and is used to drive the converter to tilt and perform slag dumping operation;

[0031] A converter tilt angle measuring device is used to measure the tilt angle of the converter;

[0032] A thickness gauge is used to measure the real-time thickness of the slag layer at any horizontal position in the slag dumping area when the converter is performing slag dumping.

[0033] An image acquisition device is used to acquire a real-time video stream of the dumped slag layer in the direction perpendicular to its thickness;

[0034] The image processing module processes the video stream into a planar image;

[0035] A storage module is used to store the video stream and the planar image;

[0036] An image recognition and calculation module is used to identify the slag flow area image from the planar image;

[0037] The control server system is connected to the converter, the converter tilting drive device, the converter tilt angle measuring device, the thickness gauge, the image acquisition device, the image processing module, the storage module, and the image recognition and calculation module via network communication. The thickness of the slag layer is multiplied by the area of ​​the slag layer in the planar region within the time interval using a free fall model or a free fall motion model with an initial velocity to obtain the micro-layer slag volume. The micro-layer slag volume is then multiplied by the density and conversion factor of the slag layer to obtain the slag quantity value of the micro-layer. The real-time slag quantity values ​​of the micro-layer are continuously accumulated to obtain the total slag quantity value that has been poured into the slag pot.

[0038] The real-time slag quantity identification system for automatic slag removal in converters is used to realize the real-time slag quantity identification method for automatic slag removal in converters.

[0039] Preferably, the thickness gauge is a laser profile sensor, and two of them are respectively located at the same horizontal position on both sides of the dumping slag area;

[0040] The laser profile sensor is located between 1000mm and 8000mm away from the dumping area, with a laser sampling frequency greater than 100Hz and a ranging repeatability of less than 50μm.

[0041] The laser profile sensor is equipped with a high-temperature protective cover that is either air-cooled or water-cooled, and a dust removal device that blows air to remove smoke.

[0042] Preferably, the image acquisition device is a high-speed camera;

[0043] The high-speed camera has a frame rate greater than 30 FPS.

[0044] The present invention provides a real-time slag quantity identification method and system for automatic slag dumping in converters. This method employs a stereoscopic recognition technology combining machine vision recognition of two-dimensional slag image area with laser measurement of third-dimensional thickness. This effectively solves the time lag problem in slag car weighing and the challenge of measuring the thickness of rapidly flowing slag. It is operable, effective, and practical in the actual implementation of automatic slag dumping technology in converters, and can effectively improve the accuracy and control precision of intelligent recognition technology. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating the real-time slag quantity identification method for automatic slag removal in converters according to the present invention.

[0046] Figure 2 This is a schematic diagram of the arrangement of laser profile sensors in the real-time slag quantity identification method for automatic slag removal in converters of the present invention.

[0047] Figure 3 This is a schematic diagram illustrating the principle of the real-time slag quantity identification method for automatic slag removal in converters according to the present invention. Detailed Implementation

[0048] To better understand the above-mentioned technical solutions of the present invention, the technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0049] Combination Figure 1 As shown, the present invention provides a method for real-time identification of slag quantity in an automatic converter slag discharge system, comprising the following steps:

[0050] S1. Obtain the real-time thickness of the slag layer at any horizontal position in the dumping slag area;

[0051] S2. Acquire a real-time planar image of the slag layer in the direction perpendicular to its thickness, and multiply the thickness of the slag layer by the area of ​​the slag layer in a small time interval within a planar micro-region to obtain the volume of the micro-layer slag.

[0052] S3. Multiply the volume of the micro-layer slag by the density and conversion factor of the slag layer to obtain the slag quantity value of the micro-layer. Continuously accumulate the real-time slag quantity values ​​of the micro-layer to obtain the total slag quantity value that has been poured into the slag pot.

[0053] Combination Figure 2 As shown, in step S1 above, when the ladle 100 pours slag into the slag pot 200, a laser profile sensor 1 is arranged at the same horizontal position on both sides of the slag pouring area to measure the front and back waveform curves of the slag pouring area in real time. The real-time thickness of the slag layer at that horizontal position is obtained by subtracting the distance difference between the two waveform curves.

[0054] The distance between the laser profile sensor 1 and the dumping slag area is between 1000mm and 8000mm, the laser sampling frequency is greater than 100Hz, and the distance measurement repeatability is less than 50μm.

[0055] Combination Figure 2 As shown, in step S2 above, a high-speed camera is used to acquire real-time planar images, and the thickness of the slag layer is multiplied by the area of ​​the slag layer in the planar region within the time interval using a free fall model or a free fall motion model with an initial velocity to obtain the micro-layer slag volume. The frame rate of the high-speed camera is greater than 30 FPS.

[0056] The calculation of the slag layer area includes the following steps:

[0057] S21. Acquire real-time video streams using a high-speed camera;

[0058] S22. Extract an image of the slag flow region from the video stream (e.g., ... Figure 3 The area within the dashed box (the range of this area is adjustable);

[0059] S23. Binarize the image of the slag flow area (remove the background and leave only the slag flow. Alternatively, dynamic thresholding, gradient sharpening, morphological transformation, edge detection and other algorithms can be used). The grayscale value of the background pixels is 0, and the grayscale value of the pixels in the slag flow area is 1.

[0060] S24. Accumulate the pixel grayscale values ​​of the slag flow micro-layer region, and the sum is the real-time characterization value of the micro-layer slag area.

[0061] The total amount of slag can be calculated by multiplying the characteristic value of the area of ​​the micro-layer slag by the measured value of the thickness of the slag layer and then summing them.

[0062] In addition, the conversion factor of the slag quantity calculation model is corrected by the actual slag weight. The conversion factor ranges from 0.01 to 15.0.

[0063] Continue to refer to Figure 3 Image recognition technology can be used to accomplish the following:

[0064] 1) The slag width distribution during the slag dumping process can be obtained in real time. When the slag width is less than a certain threshold, it can be considered that the slag dumping end point has been reached.

[0065] 2) The amount of slag to be dumped can be calculated by accumulating the results of real-time identification by a high-speed camera based on the free fall model, and the amount of slag to be dumped can be estimated to meet the requirements for the amount of slag to be retained.

[0066] When the ladle is nearly horizontally inverted (0-10 degrees), a free fall model is used; when the ladle is tilted or inverted at a large angle (greater than 10 degrees), a free fall model with an initial velocity is used, with the initial velocity ranging from 0.01 to 0.3 m / s, depending on the ladle's tilt angle.

[0067] Therefore, the real-time values ​​of the slag volume entering the image recognition area per unit time can be calculated and accumulated to obtain the final total area value of the slag volume.

[0068] Continue to refer to Figure 3 Depth is the accumulated region depth calculated based on the time interval between camera shots. It is the difference between the height value h1 when entering the accumulated region and the height value h2 when exiting the accumulated region. V1 and V2 are the speed values ​​when entering and exiting the accumulated region, respectively. DT is the time interval value calculated based on the camera frame rate (FPS).

[0069] In step S3, the total slag volume is calculated using the following formula (1):

[0070]

[0071] This invention also discloses a real-time slag quantity identification system for automatic slag removal in converters, comprising:

[0072] Converter;

[0073] A converter tilting drive device is connected to the converter and is used to drive the converter to tilt and perform slag dumping operations;

[0074] A converter tilt angle measuring device is used to measure the tilt angle of a converter.

[0075] The thickness gauge is used to measure the real-time thickness of the slag layer at any horizontal position in the slag dumping area during the slag dumping process in the converter.

[0076] An image acquisition device is used to acquire real-time video streams of the dumped slag layer in the direction perpendicular to its thickness;

[0077] The image processing module processes the video stream into a planar image;

[0078] Storage module, used to store video streams and 2D images;

[0079] The image recognition and calculation module is used to identify slag flow area images from planar images;

[0080] The control server system is connected to the converter, converter tilting drive device, converter tilt angle measuring device, thickness gauge, image acquisition device, image processing module, storage module and image recognition and calculation module via network communication. The thickness of the slag layer is multiplied by the area of ​​the slag layer in the plane region within the time interval using a free fall model or a free fall motion model with initial velocity to obtain the micro-layer slag volume. The micro-layer slag volume is then multiplied by the density and conversion factor of the slag layer to obtain the slag quantity value of the micro-layer. The real-time slag quantity values ​​of the micro-layer are continuously accumulated to obtain the total slag quantity value that has been poured into the slag pot.

[0081] The present invention utilizes a real-time slag quantity identification system for automatic slag removal in converters to realize the real-time slag quantity identification method for automatic slag removal in converters.

[0082] The thickness gauge uses a laser profile sensor 1, which is provided in two places, respectively located on the same horizontal position on both sides of the dumping slag area; the distance between the laser profile sensor and the dumping slag area is between 1000mm and 8000mm, the laser sampling frequency is greater than 100Hz, and the distance measurement repeatability is less than 50μm; the laser profile sensor is equipped with an air-cooled or water-cooled high-temperature protective cover and an air blowing dust removal device.

[0083] The image acquisition device uses a high-speed camera; the frame rate of the high-speed camera is greater than 30 FPS.

[0084] This invention combines machine vision recognition of two-dimensional slag image area with laser measurement of third-dimensional thickness, and utilizes high-speed communication network and stereo recognition technology to effectively solve the time lag problem in slag truck weighing. Furthermore, the use of relative laser profile measurement technology on the same horizontal plane effectively solves the challenge of measuring the thickness of rapidly flowing slag.

[0085] Example 1

[0086] Two laser profile sensors are used, each with a sampling frequency of 120Hz, installed 5000mm away from the slag outlet, with a high-speed communication interface and a wavelength of 405nm;

[0087] It uses a high-speed camera with a frame rate of 60fps, a black and white visible light camera with an infrared filter;

[0088] The converter steel weighs 250 tons. The automatic slag removal process takes 85 seconds. The camera actually samples one image every 20ms. Within 20ms, the image shifts down by 235 pixels. Cumulatively:

[0089] Total slag dumping area * slag thickness * conversion factor = total slag dumping volume = 8.304 (m³) 3 )

[0090] Calculate the weight of the slag dumped = total volume of slag dumped * average density of slag = 8.304 * 2.3 = 19.1 tons.

[0091] (Average slag density = 2.3 tons / cubic meter) The success rate and accuracy of slag dumping and retention are greater than 85%.

[0092] Example 2

[0093] Two laser profile sensors are used, each with a sampling frequency of 120Hz, installed 5000mm away from the slag outlet, with a high-speed communication interface and a wavelength of 405nm;

[0094] It uses a high-speed camera with a frame rate of 60fps, a black and white visible light camera with an infrared filter;

[0095] The converter molten steel weighed 255 tons. The automatic slag removal process took 79 seconds. The camera actually sampled one image every 20ms. In 20ms, the image shifted down by 235 pixels. Cumulatively:

[0096] Total slag dumping area * slag thickness * conversion factor = total slag dumping volume = 8.043 (m³) 3 )

[0097] Calculate the weight of the slag dumped = total volume of slag dumped * average density of slag = 8.043 * 2.3 = 18.5 tons.

[0098] (Average slag density = 2.3 tons / cubic meter) The success rate and accuracy of slag dumping and retention are greater than 85%.

[0099] Example 3

[0100] Two laser profile sensors are used, each with a sampling frequency of 120Hz, installed 5000mm away from the slag outlet, with a high-speed communication interface and a wavelength of 405nm;

[0101] It uses a high-speed camera with a frame rate of 60fps, a black and white visible light camera with an infrared filter;

[0102] The converter steel weighed 235 tons. The automatic slag removal process took 69 seconds. The camera actually sampled one image every 20ms. In 20ms, the image shifted down by 235 pixels. The cumulative total was:

[0103] Total slag dumping area * slag thickness * conversion factor = total slag dumping volume = 7.217 (m³) 3 )

[0104] Calculate the weight of the slag dumped = total volume of slag dumped * average density of slag = 7.217 * 2.3 = 16.6 tons.

[0105] (Average slag density = 2.3 tons / cubic meter) The success rate and accuracy of slag dumping and retention are greater than 85%.

[0106] Those skilled in the art should recognize that the above embodiments are merely illustrative of the present invention and are not intended to limit the present invention. Any variations or modifications to the above embodiments that are within the spirit and essence of the present invention will fall within the scope of the claims of the present invention.

Claims

1. A method for real-time identification of slag quantity in an automatic converter slag discharge system, characterized in that, Includes the following steps: S1. Obtain the real-time thickness of the slag layer at any horizontal position in the dumping slag area. By placing a laser profile sensor at the same horizontal position on both sides of the dumping slag area, the front and rear waveform curves of the dumping slag area are measured in real time, and the real-time thickness of the slag layer at that horizontal position is obtained by the distance difference between the two waveform curves. S2. Acquire a real-time planar image of the slag layer in the direction perpendicular to its thickness. Multiply the thickness of the slag layer by the area of ​​the slag layer within the planar region during the time interval to obtain the volume of the micro-layer slag. Real-time planar images are acquired using a high-speed camera. The thickness of the slag layer is multiplied by the area of ​​the slag layer within a planar region during a time interval using a free-fall model or a falling motion model with an initial velocity, to obtain the volume of the micro-layer slag. The calculation of the area of ​​the slag layer specifically includes the following steps: S21. Acquire a real-time video stream using the high-speed camera; S22. Extract an image of the slag flow region from the video stream; S23. The image of the slag flow area is binarized, with the grayscale value of the background pixels being 0 and the grayscale value of the pixels in the slag flow area being 1. S24. Accumulate the pixel grayscale values ​​of the slag flow micro-layer region, and the sum is the characterization value of the real-time micro-slag layer area. S3. Multiply the volume of the micro-layer slag by the density and conversion factor of the slag layer to obtain the slag quantity value of the micro-layer. Continuously accumulate the real-time slag quantity values ​​of the micro-layer to obtain the total slag quantity value that has been poured into the slag tank.

2. The real-time slag quantity identification method for automatic slag removal in converters according to claim 1, characterized in that: The laser profile sensor is located between 1000mm and 8000mm away from the dumping area, with a laser sampling frequency greater than 100Hz and a ranging repeatability of less than 50μm.

3. The real-time slag quantity identification method for automatic slag removal in converters according to claim 1, characterized in that: The high-speed camera has a frame rate greater than 30 FPS.

4. The method for real-time identification of slag quantity in automatic converter slag removal according to claim 1, characterized in that: When the inversion angle of the converter is 0 to 10 degrees, the free fall model is used; when the inversion angle of the converter is greater than 10 degrees, the free fall motion model with initial velocity is used, with the initial velocity ranging from 0.01 to 0.3 m / s.

5. The method for real-time identification of slag quantity in automatic converter slag removal according to claim 1, characterized in that, In step S3, the total slag volume is calculated using the following formula: 。 6. A real-time slag quantity identification system for automatic slag removal in a converter, wherein the real-time slag quantity identification system for automatic slag removal in a converter is used to realize the real-time slag quantity identification method for automatic slag removal in a converter as described in any one of claims 1-5, characterized in that, include: Converter; A converter tilting drive device is connected to the converter and is used to drive the converter to tilt and perform slag dumping operation; A converter tilt angle measuring device is used to measure the tilt angle of the converter; A thickness gauge is used to measure the real-time thickness of the slag layer at any horizontal position in the slag dumping area when the converter is performing slag dumping. An image acquisition device is used to acquire a real-time video stream of the slag layer in the direction perpendicular to its thickness; The image processing module processes the video stream into a planar image; A storage module is used to store the video stream and the planar image; An image recognition and calculation module is used to identify the slag flow area image from the planar image; The control server system is connected to the converter, the converter tilting drive device, the converter tilt angle measuring device, the thickness gauge, the image acquisition device, the image processing module, the storage module, and the image recognition and calculation module via network communication. It obtains the micro-layer slag volume by multiplying the thickness of the slag layer by the area of ​​the slag layer within a planar region during a time interval using a free-fall model or a free-fall motion model with an initial velocity. Then, it multiplies the micro-layer slag volume by the density and conversion factor of the slag layer to obtain the slag quantity value of the micro-layer. The real-time slag quantity values ​​of the micro-layers are continuously accumulated to obtain the total slag quantity value that has been poured into the slag pot.

7. The real-time slag quantity identification system for automatic slag removal in converters according to claim 6, characterized in that: The thickness gauge uses a laser profile sensor, and there are two of them, which are respectively set at the same horizontal position on both sides of the dumping slag area; The laser profile sensor is located between 1000mm and 8000mm away from the dumping area, with a laser sampling frequency greater than 100Hz and a ranging repeatability of less than 50μm. The laser profile sensor is equipped with a high-temperature protective cover that is either air-cooled or water-cooled, and a dust removal device that blows air to remove smoke.

Images