A bottom blowing mode control method and system in LF refining furnace

Abstract

The present invention relates to the field of LF refining, and specifically discloses a method and system for controlling a bottom blowing mode in an LF refining furnace. The method predicts the temperature of exposed molten steel in a ladle based on a temperature prediction model; obtains the actual infrared temperature of the ladle; generates a two-dimensional point cloud data map of the ladle infrared image based on the predicted temperature of the exposed molten steel in the ladle and the actual infrared temperature of the ladle, and based on defined temperature threshold boundary conditions, including a cloud data set of an exposed molten steel area and a cloud data set of an area covered by a slag layer; calculates the area of ​​the exposed molten steel area and the area of ​​the slag layer covered area according to a convex hull algorithm; calculates the area ratio of the exposed molten steel area; matches a bottom blowing mode according to the area ratio of the exposed molten steel area, and dynamically associates the matched bottom blowing mode with different operation processes of LF refining. The present invention automatically selects a bottom blowing mode and dynamically associates the matched mode with different operation processes of LF refining, thereby realizing automatic control of the bottom blowing mode.

Description

Technical Field

[0001] The present invention relates to the field of LF refining, and in particular to a method and system for controlling a bottom blowing mode in an LF refining furnace. Background Art

[0002] The molten steel LF (Ladle Furnace) refining unit has functions such as adjusting the molten steel composition and temperature, desulfurization, and controlling the morphology of inclusions. It is one of the main production units in the steelmaking process for producing high-value-added steel varieties, an important part of the clean steel production platform, and a key process for achieving efficient connection in steelmaking production organization.

[0003] As molten steel enters the ladle, argon gas continuously enters the liquid steel through the ladle, forming numerous argon bubbles. These bubbles, surrounded by nitrogen, oxygen, and hydrogen, create a nearly zero-pressure, enclosed space. As the nitrogen and hydrogen increase, the pressure within the bubbles increases, causing the bubbles to rise. However, the bubbles, expanding due to heat in the molten steel, remain at a relatively low pressure, continuing their upward buoyancy. As they rise, they collide with impurities in the molten steel, causing the impurities to merge into the bubbles, which then remove them from the molten steel. For this reason, argon refining at the bottom of the LF refining furnace is widely used, achieving significant results in reducing harmful gases in the molten steel and improving the steel's purity and reducing impurities.

[0004] Currently, LF refining furnaces typically feature a variety of bottom-blowing argon modes, including strong stirring, medium stirring, light stirring, weak stirring, and light blowing. The selection of bottom-blowing modes for different operating procedures is typically based on experience, with operators manually selecting the mode. However, LF refining processes are sealed and covered, and with the advancement of remote, centralized, and intelligent control, dynamic monitoring of the ladle liquid level to provide real-time visibility into the ladle's bottom-blowing status has become increasingly urgent. Summary of the Invention

[0005] In order to solve the above problems, the present invention provides a method and system for controlling the bottom blowing mode in an LF refining furnace, so as to realize automatic control of the bottom blowing mode.

[0006] In a first aspect, the technical solution of the present invention provides a method for controlling a bottom blowing mode in an LF refining furnace, comprising the following steps:

[0007] Predict the temperature of exposed molten steel in the ladle based on the temperature prediction model;

[0008] Get the actual infrared temperature of the ladle;

[0009] Based on the predicted temperature of the exposed molten steel in the ladle and the actual infrared temperature of the ladle, and based on the defined temperature threshold boundary conditions, a two-dimensional point cloud data map of the ladle infrared image is generated, including a cloud data set of the exposed molten steel area and a cloud data set of the slag layer covered area;

[0010] Based on the cloud dataset of the exposed molten steel area and the cloud dataset of the slag layer covered area, the areas of the exposed molten steel area and the slag layer covered area are calculated using the convex hull algorithm.

[0011] Calculate the area ratio of exposed molten steel area;

[0012] The bottom blowing mode is matched according to the area ratio of the exposed molten steel area, and the matched bottom blowing mode is dynamically associated with different operation processes of LF refining.

[0013] In an optional embodiment, the temperature prediction model is:

[0014]

[0015] Among them, ∑ t Ar is the argon blowing temperature loss during this prediction period, ∑ t H is the heating temperature contribution in this prediction period, ∑ t M is the impact of feed temperature during this prediction period, α is the significant influencing factor, and t is the process time.

[0016] In an optional embodiment, the method further includes the step of adjusting the significant impact factor α, including:

[0017] If temperature measurement occurs, adjust α according to the prediction deviation, so that α 新 =(T 测温 -T 当前 ) / (t 测温时间间隔 );

[0018] If α 新 ≥α down And α 新 ≤α up , then let α=α 新 ;

[0019] If α 新 <α down , then let α 新 =α down ;

[0020] If α 新 >α up , then α 新 =α up

[0021] Among them, α down and α upis the maximum mean and minimum mean of the α value statistics of multiple furnaces, T 测温 It is the actual temperature measured on the exposed molten steel.

[0022] In an optional embodiment, the method further includes the step of defining a temperature threshold boundary condition, comprising:

[0023] T' is the infrared measured temperature of the slag layer coverage area;

[0024] Calculate ΔT = T 当前 -T';

[0025] The area where ΔT is between 0-15℃ is defined as the steel slag mixing zone;

[0026] The temperature threshold boundary condition of the exposed molten steel area is defined as [T 当前 -5℃,T 当前 ];

[0027] The boundary conditions of the slag layer coverage area are regularly distributed in a circle according to the ladle lining, with the current slag layer temperature T' as its boundary temperature, and △T'=T' n -T' n-1 ≦5℃, where n is the point cloud data value of the lining distribution at the contact point of the steel slag in the ladle.

[0028] In an optional embodiment, based on the cloud dataset of the exposed molten steel area and the cloud dataset of the slag layer covered area, the area of ​​the exposed molten steel area and the area of ​​the slag layer covered area are calculated according to a convex hull algorithm, specifically including:

[0029] According to the temperature threshold boundary condition, the intersection S0 of all convex sets containing the temperature T in the same area is positioned as the convex hull of temperature T. The convex hulls of different temperatures T can be represented by all points in T (x1, x2, ..., x m ) is constructed by a linear combination of:

[0030]

[0031] According to the construction formula, the area of ​​different cloud data sets is calculated.

[0032]

[0033] Where m is the number of convex hull vertices, and the determination of m is based on the number of temperature field distribution grids obtained by infrared photography;

[0034] x i is the x coordinate of the i-th vertex, y i is the y-coordinate of the i-th vertex; the x-coordinate and y-coordinate represent the values ​​of the two coordinate axes respectively and can be calculated by the following formula:

[0035]

[0036]

[0037] Among them, Pixel x Pixel is the horizontal axis pixel. y The vertical axis is pixels; dpi is resolution.

[0038] In an optional embodiment, calculating the area ratio of the exposed molten steel region specifically includes calculating the area ratio τ of the exposed molten steel region by the following formula:

[0039]

[0040] In an optional embodiment, the method further includes the step of adjusting the temperature threshold boundary and the number of infrared image grids according to the infrared image.

[0041] In an optional embodiment, the step of adjusting the temperature threshold boundary and the number of infrared image grids according to the infrared image specifically includes:

[0042] Convert the infrared image into a grayscale image using the first channel; wherein the infrared image refers to the image of the actual infrared temperature of the ladle;

[0043] Set pixel thresholds for the ladle edge, exposed molten steel area, slag layer covered area, and steel-slag mixed area;

[0044] Traverse each pixel point p and divide the area to which the pixel point belongs according to the following formula:

[0045]

[0046] Calculate the pixel ratio τ' of the exposed molten steel area,

[0047]

[0048] Set τ / τ' = λ;

[0049] If λ is within the preset range, the currently defined temperature threshold boundary and the number of infrared image grids are judged to be appropriate;

[0050] If λ is within the preset range, it is determined that the currently defined temperature threshold boundary and the number of infrared image grids need to be adjusted.

[0051] In an optional embodiment, before setting the thresholds for the lap edge, the exposed molten steel area, the slag layer covered area, and the steel-slag mixed area, the method further includes:

[0052] Filter noise from grayscale images.

[0053] In a second aspect, the technical solution of the present invention provides a bottom blowing mode control system in an LF refining furnace, comprising:

[0054] Temperature prediction module: predicts the temperature of exposed molten steel in the ladle based on the temperature prediction model;

[0055] Measured temperature acquisition module: obtains the actual infrared temperature of the ladle;

[0056] Cloud dataset acquisition module: Based on the predicted temperature of the exposed molten steel in the ladle and the actual infrared temperature of the ladle, and based on the defined temperature threshold boundary conditions, it generates a two-dimensional point cloud data map of the ladle infrared image, including a cloud dataset of the exposed molten steel area and a cloud dataset of the slag layer covered area;

[0057] Area calculation module: Based on the cloud dataset of exposed molten steel area and the cloud dataset of slag layer covered area, the area of ​​exposed molten steel area and the area of ​​slag layer covered area are calculated according to the convex hull algorithm;

[0058] Exposed molten steel area ratio calculation module: calculates the exposed molten steel area ratio;

[0059] Bottom blowing mode selection module: matches the bottom blowing mode according to the area ratio of the exposed molten steel area, and dynamically associates the matching bottom blowing mode with different LF refining operation processes.

[0060] The present invention provides a method and system for controlling the bottom blowing mode in an LF refining furnace. Compared with the prior art, the present invention has the following beneficial effects: the temperature of the exposed molten steel is predicted, and regions are divided based on the actual temperature. The area is calculated based on the regional division, and then the area ratio of the exposed molten steel region is calculated. The bottom blowing mode is automatically selected according to the ratio, and the matching bottom blowing mode is dynamically associated with different LF refining operation processes to achieve automatic control of the bottom blowing mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0062] Figure 1 It is a flow chart of a method for controlling the bottom blowing mode in an LF refining furnace provided by an embodiment of the present invention.

[0063] Figure 2 This is a schematic diagram of the LF refining ladle liquid level viewed from above.

[0064] Figure 3 It is a schematic diagram of the grid distribution of the ladle liquid surface temperature field captured by infrared photography.

[0065] Figure 4It is a grayscale image after the infrared camera is converted into grayscale.

[0066] Figure 5 yes Figure 4 Pixel distribution map after noise filtering of the grayscale image.

[0067] Figure 6 This is a schematic block diagram of the structure of a bottom blowing mode control system in an LF refining furnace provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to enable those skilled in the art to better understand the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the embodiments described are only some of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0069] Figure 1 This is a flow chart of a method for controlling the bottom blowing mode in a LF refining furnace provided by an embodiment of the present invention. Figure 1 As shown, the method includes the following steps.

[0070] S1, predict the temperature of exposed molten steel in the ladle based on the temperature prediction model.

[0071] S2, obtain the actual infrared temperature of the ladle.

[0072] S3, based on the predicted temperature of the exposed molten steel in the ladle and the actual infrared temperature of the ladle, and based on the defined temperature threshold boundary conditions, generates a two-dimensional point cloud data map of the ladle infrared image, including a cloud data set of the exposed molten steel area and a cloud data set of the slag layer covered area.

[0073] S4, based on the cloud dataset of the exposed molten steel area and the cloud dataset of the slag layer covered area, calculate the area of ​​the exposed molten steel area and the area of ​​the slag layer covered area according to the convex hull algorithm.

[0074] S5, calculate the area ratio of the exposed molten steel area.

[0075] S6, matching the bottom blowing mode according to the area ratio of the exposed molten steel area, and dynamically associating the matching bottom blowing mode with different operation processes of LF refining.

[0076] The present invention predicts the temperature of exposed molten steel, divides the area into regions based on the measured temperature, calculates the area according to the regional division, and then calculates the area ratio of the exposed molten steel region, automatically selects the bottom blowing mode according to the ratio, and dynamically associates the matching bottom blowing mode with different LF refining operation processes to achieve automatic control of the bottom blowing mode.

[0077] For a further understanding of the present invention, a specific example is provided below to further illustrate the present invention in detail.

[0078] In this specific embodiment, the ladle temperature is collected by an infrared camera. In order to obtain a complete and good infrared image and ensure the reliable and stable application of the camera equipment, a suitable position is selected on the LF refining furnace cover. In view of the high temperature and slag splashing environment of LF refining, the infrared camera body is protected by water cooling and nitrogen sealing in front of the lens.

[0079] (1) Obtain the temperature infrared image inside the ladle, perform regularized gradient processing based on the model predicted temperature or point measured temperature, realize the equal temperature threshold boundary condition, and generate a two-dimensional point cloud data map.

[0080] According to the heat balance of LF refining process, set up a real-time temperature prediction model:

[0081]

[0082] Among them, ∑ t Ar is the argon blowing temperature loss during this prediction period, which is calculated by multiplying the flow coefficient by time.

[0083] Table 1 Temperature drop coefficient at different flow ranges

[0084]

[0085]

[0086] ∑ t H is the heating temperature contribution during this prediction period, which is calculated by multiplying the gear temperature rise coefficient (see the table below for temperature rise coefficients at different gears) by time.

[0087] Table 2 Temperature rise coefficient at different gears

[0088] gear Temperature rise coefficient (℃ / min) 1 2 6 3 5 4 4 5 3 6 2 7 The temperature of molten steel at the station is high, slag insulation 8 Insulation 9 Insulation 10 Insulation

[0089] ∑ t M is the impact of feed temperature during this forecast period, which is the sum of the temperature influence coefficient of the feed material type (see the table below, temperature drop coefficient of different materials) multiplied by the feed amount.

[0090] Table 3 Temperature drop coefficient of different materials

[0091]

[0092]

[0093] Where t is the process time. The model predicts the temperature every 10 seconds and updates the temperature loss caused by argon blowing, heating, and adding materials every 1 second during this 10 seconds.

[0094] The model was used to linearly regress the significant impact factors to α.

[0095] If temperature measurement occurs, adjust α according to the prediction deviation, so that α 新 =(T 测温 -T 当前 ) / (t 测温时间间隔 );

[0096] If α 新 ≥α down And α 新 ≤α up , then let α=α 新 ;

[0097] If α 新 <α down , then let α 新 =α down ;

[0098] If α 新 >α up , then α 新 =α up

[0099] Among them, α down and α up is the maximum mean and minimum mean of the α value statistics of multiple furnaces, T 测温 It is the actual temperature measured on the exposed molten steel.

[0100] Let T' be the infrared measured temperature of the slag layer coverage area; calculate △T = T 当前 -T'; define the area where △T is between 0-15℃ as the slag mixing area; define the temperature threshold boundary condition of the exposed molten steel area as [T 当前 -5℃,T 当前 ]; The boundary conditions of the slag layer coverage area are regularly distributed in a circular pattern according to the ladle lining, with the current slag layer temperature T' as its boundary temperature, and △T'=T' n -T' n-1 ≦5℃.

[0101] Specifically, the exposed molten steel temperature T 当前 The highest temperature of the liquid surface is the temperature predicted by the model or the point-measured temperature. Based on this temperature, the temperature threshold range is determined using the temperature gradient, as shown in the attached figure. Figure 1 , combined with the predicted temperature T of the on-site molten steel temperature 当前 The infrared measured temperature T' of the slag layer coverage area is obtained by △T=T 当前 -T'≥15℃, T 当前The distribution area is the temperature of the exposed surface of the molten steel, that is, the bright surface, T' is the temperature of the slag layer covered area, and the area between △T 0-15℃ is the steel slag mixing area. According to this rule, the temperature threshold boundary conditions can be further determined and the following is formed: Figure 2 Point cloud data diagram, in the figure, label 1 is the exposed area of ​​molten steel, label 2 is the steel slag mixed area, label 3 is the slag layer covered area, label 4 is the ladle edge: S1, S2 (under normal conditions, the bright surface area of ​​LF refined steel ladle is 2 pieces, because the ladle bottom blowing has two air bricks with asymmetric single-sided design) are the bright surface area, S is the slag layer covered area, in order to further improve the calculation accuracy of the bright surface area and the molten steel temperature has certain fluctuations, the temperature threshold boundary conditions of the exposed surface of molten steel S1 and S2 are set as [T 当前 -5℃,T 当前 ], that is, the measured temperature is within the range [T 当前 -5℃,T 当前 ] is the bright surface area, S is the slag layer coverage area, and its boundary conditions are regularly distributed in a circular pattern according to the ladle lining. The current slag layer temperature T' is its boundary temperature, and △T'=T' n -T' n-1 ≤5℃, where n is the point cloud data value of the lining distribution at the slag contact point in the ladle. According to the rule, the maximum value can be determined to be 360.

[0102] (2) Use the infrared camera to set the number of temperature field distribution grids, determine the vertex values ​​of the point cloud data graph, and calculate the area of ​​the relevant region of the point cloud data graph based on the convex hull algorithm.

[0103] Specifically, the plane point cloud dataset is extracted with the above boundary conditions. In a real vector space V, according to the temperature threshold boundary condition, the intersection S0 of all convex sets containing the temperature T in the same area is located as the convex hull of the temperature T. The convex hulls of different temperatures T can be represented by all points in T (x1, x2, ..., x m ) is constructed by a linear combination of:

[0104]

[0105] According to the construction formula, the area of ​​different cloud data sets is calculated.

[0106]

[0107] Where m is the number of convex hull vertices. The determination of m is based on the number of temperature field distribution grids obtained by infrared photography, such as Figure 3 According to the calculation accuracy and calculation time, it can be manually set in the camera system.

[0108] x i is the x-coordinate of the i-th vertex, in meters;

[0109] y iis the y coordinate of the i-th vertex, in meters;

[0110] x i+1 is the x-coordinate of the i+1th vertex, in meters;

[0111] y i+1 is the y-coordinate of the i+1th vertex, in meters.

[0112] The x-coordinate and y-coordinate represent the values ​​of the two coordinate axes, which are also the values ​​in the length and width directions, and can be calculated by the following formula:

[0113]

[0114]

[0115] Among them, Pixel x Pixel is the horizontal axis pixel. y The vertical axis is pixels; dpi is resolution.

[0116] (3) The exposed area of ​​the molten steel and the total area of ​​the ladle liquid surface are compared, and the intensity of the argon blowing in the ladle is quantified based on the comparison of the two. The exposed ratio of the molten steel is matched with different bottom blowing modes, and the matching mode is dynamically associated with different LF refining operation processes to achieve visual real-time control of the LF refining bottom blowing process.

[0117] The ratio of exposed molten steel area τ is calculated by the following formula:

[0118]

[0119] Specifically, the areas of different temperature regions of the image captured by the infrared camera at a certain moment, such as S1, S2, and S, can be calculated respectively from the above, and a calculation model for the exposed proportion of molten steel in the ladle can be constructed based on the known areas:

[0120]

[0121] Where S1 and S2 represent the exposed surface area of ​​molten steel in different regions, which are obtained by the area formula of point cloud dataset and the unit is m 2 ;

[0122] S represents the non-exposed molten steel area, unit is m 2 ;

[0123] τ is the ratio of exposed molten steel area on the ladle surface, in %.

[0124] Based on this, the bottom blowing mode at different stirring degrees is defined according to the size range of τ, see Table 4 below.

[0125] Table 4: Definition of the relationship between LF refining bottom blowing mode and exposed steel surface ratio

[0126] Argon blowing mode Strong stirring Medium stirring Stir gently Weak stirring Soft Blow τ 50-80 30-50 10-15 5-10 1-5

[0127] Through the above definition, and after calculating and analyzing the current ladle liquid surface temperature distribution image obtained by the infrared camera, the value is provided to the LF refining bottom blowing control model; for the exposed steel surface ratio defined under different operation steps in the LF refining operation process, the bottom blowing mode is automatically matched when the LF refining sequence moves to different operation steps, as shown in Table 5 below, for real-time dynamic matching.

[0128] Table 5: Bottom blowing mode matching under different treatment modes of LF refining

[0129]

[0130]

[0131] In an optional embodiment, the method further includes adjusting the temperature threshold boundary according to the infrared image.

[0132] In order to further improve the accuracy of convex hull area calculation under different thresholds, the acquired infrared images are converted into grayscale synchronously, and the total area of ​​the bright surface and the ladle liquid surface is compared using the number of pixels, and then compared with the area ratio of the area divided by the temperature threshold. The accuracy of the temperature threshold determination is further verified based on the comparison results.

[0133] Specifically include the following steps.

[0134] Step 1: Convert the RGB three-channel color image obtained by the infrared camera into a grayscale image using the first channel, as shown in the following example: Figure 3 shown.

[0135] img_original=cv2.imread('5.jpg',0)

[0136] Step 2: Then perform Gaussian filtering to remove the noise of the image, forming Figure 4 The image shown.

[0137] cv2.GaussianBlur(img_original,(13,13),13)

[0138] Step 3: Set the Figure 2 The threshold pixels for the middle package edge, slag layer area, mixed area, and exposed area are: package edge = 50, slag layer area = 120, mixed area = 130, exposed area = 200

[0139] thresh=[50,120,130,200].

[0140] Step 4: According to experience, T 裸露区 >T 混合区 >T 覆盖区 >T 包沿 , based on each pixel point p in the image, assuming its pixel value is Pixel, its division rule is as follows:

[0141]

[0142] Y1 and Y2 are the slag layer covered area and the exposed area of ​​molten steel on the ladle surface, respectively. Traverse each pixel in the image and divide each pixel into its own area:

[0143]

[0144] Step 5: Calculate the exposed surface ratio of molten steel, τ', as follows:

[0145]

[0146] Step 6: Set τ / τ' = λ, and judge the analysis results based on the value of λ. If λ is within the preset range, the temperature threshold boundary and the number of infrared image grids are judged to be appropriate. If λ is not within the preset range, the temperature threshold boundary and the number of infrared image grids need to be adjusted.

[0147] Specifically, after the calculation, τ and τ' are compared synchronously. If τ / τ'=λ is set, and λ is within the range of 1±0.2, it means that the calculation and judgment results using the temperature threshold model are consistent with the pixel recognition analysis results, thereby effectively reflecting the intensity of the stirring of the liquid surface of the LF refining ladle; if λ is not within the range of 1±0.2, the temperature threshold judgment rule (bright surface boundary temperature and slag layer boundary temperature gradient range) is adjusted or the number of infrared camera temperature field distribution grids m is further set. The temperature threshold judgment rule and the number of infrared camera temperature field distribution grids are continuously adjusted to adapt to the actual working conditions on site.

[0148] The above describes in detail an embodiment of a method for controlling the bottom blowing mode in an LF refining furnace. Based on the method for controlling the bottom blowing mode in an LF refining furnace described in the above embodiment, an embodiment of the present invention also provides a system for controlling the bottom blowing mode in an LF refining furnace corresponding to the method.

[0149] Figure 6 This is a schematic block diagram of the structure of a bottom blowing mode control system in an LF refining furnace provided by an embodiment of the present invention. In this embodiment, the bottom blowing mode control system in an LF refining furnace can be divided into multiple functional modules according to the functions it performs, such as Figure 6The functional modules may include: a temperature prediction module, a measured temperature acquisition module, a cloud data set acquisition module, an area calculation module, an exposed molten steel area ratio calculation module, and a bottom blowing mode selection module. The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can perform fixed functions, and is stored in a memory.

[0150] Temperature prediction module: predicts the temperature of exposed molten steel in the ladle based on the temperature prediction model.

[0151] Measured temperature acquisition module: obtains the actual infrared measured temperature of the ladle.

[0152] Cloud dataset acquisition module: Based on the predicted temperature of the exposed molten steel in the ladle and the actual infrared temperature of the ladle, and based on the defined temperature threshold boundary conditions, a two-dimensional point cloud data map of the ladle infrared image is generated, including a cloud dataset of the exposed molten steel area and a cloud dataset of the slag layer covered area.

[0153] Area calculation module: Based on the cloud dataset of exposed molten steel area and the cloud dataset of slag layer covered area, the area of ​​exposed molten steel area and the area of ​​slag layer covered area are calculated according to the convex hull algorithm.

[0154] Exposed molten steel area ratio calculation module: calculates the exposed molten steel area ratio.

[0155] Bottom blowing mode selection module: matches the bottom blowing mode according to the area ratio of the exposed molten steel area, and dynamically associates the matching bottom blowing mode with different LF refining operation processes.

[0156] The bottom blowing mode control system in the LF refining furnace of this embodiment is used to implement the aforementioned bottom blowing mode control method in the LF refining furnace. Therefore, the specific implementation method of this system can be seen in the embodiment part of the bottom blowing mode control method in the LF refining furnace in the previous text. Therefore, its specific implementation method can refer to the description of the corresponding embodiments of each part, and will not be elaborated here.

[0157] In addition, since the bottom blowing mode control system in the LF refining furnace of this embodiment is used to implement the aforementioned bottom blowing mode control method in the LF refining furnace, its function corresponds to that of the aforementioned method and will not be described in detail here.

[0158] The above disclosure is only a preferred embodiment of the present invention, but the present invention is not limited thereto. Any non-creative changes that can be thought of by those skilled in the art, as well as several improvements and modifications made without departing from the principles of the present invention, should fall within the scope of protection of the present invention.

Claims

1. A method for controlling the bottom blowing mode in a LF refining furnace, characterized in that: The following steps are involved: Predict the temperature of exposed molten steel in the ladle based on the temperature prediction model; Get the actual infrared temperature of the ladle; Based on the predicted temperature of the exposed molten steel in the ladle and the actual infrared temperature of the ladle, and based on the defined temperature threshold boundary conditions, a two-dimensional point cloud data map of the ladle infrared image is generated, including a cloud data set of the exposed molten steel area and a cloud data set of the slag layer covered area; Based on the cloud dataset of the exposed molten steel area and the cloud dataset of the slag layer covered area, the area of ​​the exposed molten steel area and the area of ​​the slag layer covered area are calculated according to the convex hull algorithm. Specifically, according to the temperature threshold boundary condition, the intersection S0 of all convex sets containing the same area temperature T is positioned as the convex hull of temperature T, and the convex hulls of different temperatures T are calculated using all points (x1, x2, ..., x m ) is constructed by a linear combination of: According to the construction formula, the area of ​​different cloud data sets is calculated. Where m is the number of convex hull vertices, and the determination of m is based on the number of temperature field distribution grids obtained by infrared photography; x i is the x coordinate of the i-th vertex, y i is the y coordinate of the i-th vertex; The x-coordinate and y-coordinate represent the values ​​of the two coordinate axes respectively and can be calculated by the following formula: Among them, Pixelx is the horizontal pixel, Pixely is the vertical pixel; dpi is the resolution; Calculate the area ratio of exposed molten steel area; The bottom blowing mode is matched according to the area ratio of the exposed molten steel area, and the matched bottom blowing mode is dynamically associated with different operation processes of LF refining.

2. The method for controlling the bottom blowing mode in the LF refining furnace according to claim 1, characterized in that: The temperature prediction model is: in, is the argon blowing temperature loss during this prediction period, Contribution to heating temperature during this forecast period, is the influence of feeding temperature during this prediction period, α is the significant influencing factor, and t is the process time.

3. The method for controlling the bottom blowing mode in the LF refining furnace according to claim 2, characterized in that: The method further includes the step of adjusting the significant impact factor α, including: If temperature measurement occurs, adjust α according to the prediction deviation, so that α 新 =(T 测温 -T 当前 ) / (t 测温时间间隔 ); if α 新 ≥α down And α 新 ≤α up , then let α=α 新 ; If α 新 <α down , then let α 新 =α down ; Young 新 >a up , a 新 =a up Among them, α down and α up is the maximum mean and minimum mean of the α value statistics of multiple furnaces, T 测温 It is the actual temperature measured on the exposed molten steel.

4. The method for controlling the bottom blowing mode in the LF refining furnace according to claim 3, characterized in that: The method further includes the step of defining a temperature threshold boundary condition, comprising: T' is the infrared measured temperature of the slag layer coverage area; Calculate ΔT = T 当前 -T'; The area where ΔT is between 0-15℃ is defined as the steel slag mixing zone; The temperature threshold boundary condition of the exposed molten steel area is defined as [T 当前 -5℃,T 当前 ]; The boundary conditions of the slag layer coverage area are regularly distributed in a circle according to the ladle lining, with the current slag layer temperature T' as its boundary temperature, and △T'=T' n -T' n-1 ≤5℃, where n is the point cloud data value of the lining distribution at the contact point of the steel slag in the ladle.

5. The method for controlling the bottom blowing mode in the LF refining furnace according to claim 1, characterized in that: Calculating the ratio of the exposed molten steel area, specifically including calculating the ratio of the exposed molten steel area τ by the following formula:

6. The method for controlling the bottom blowing mode in the LF refining furnace according to claim 5, characterized in that: The method further includes the step of adjusting the temperature threshold boundary and the number of infrared image grids according to the infrared image.

7. The method for controlling the bottom blowing mode in the LF refining furnace according to claim 6, characterized in that: The steps of adjusting the temperature threshold boundary and the number of infrared image grids according to the infrared image specifically include: Convert the infrared image into a grayscale image using the first channel; wherein the infrared image refers to an image of the actual infrared temperature of the ladle; Set pixel thresholds for the ladle edge, exposed molten steel area, slag layer covered area, and steel-slag mixed area; Traverse each pixel point p and divide the area to which the pixel point belongs according to the following formula: Calculate the pixel ratio τ' of the exposed molten steel area, Set τ / τ' = λ; If λ is within the preset range, the currently defined temperature threshold boundary and the number of infrared image grids are judged to be appropriate; If λ is within the preset range, it is determined that the currently defined temperature threshold boundary and the number of infrared image grids need to be adjusted.

8. The method for controlling the bottom blowing mode in the LF refining furnace according to claim 7, characterized in that: Before setting the thresholds for the ladle edge, exposed molten steel area, slag layer covered area and steel-slag mixed area, it also includes: Filter noise from grayscale images.

9. A bottom blowing mode control system in a LF refining furnace, characterized in that: Including, temperature prediction module: predict the temperature of exposed molten steel in the ladle based on the temperature prediction model; Measured temperature acquisition module: obtains the actual infrared temperature of the ladle; Cloud dataset acquisition module: Based on the predicted temperature of the exposed molten steel in the ladle and the actual infrared temperature of the ladle, and based on the defined temperature threshold boundary conditions, it generates a two-dimensional point cloud data map of the ladle infrared image, including a cloud dataset of the exposed molten steel area and a cloud dataset of the slag layer covered area; Area calculation module: Based on the cloud dataset of exposed molten steel area and the cloud dataset of slag layer covered area, the area of ​​exposed molten steel area and the area of ​​slag layer covered area are calculated according to the convex hull algorithm. Specifically, according to the temperature threshold boundary condition, the intersection S0 of all convex sets containing the same regional temperature T is positioned as the convex hull of temperature T, and the convex hulls of different temperatures T are calculated using all points in T (x1, x2, ..., x m ) is constructed by a linear combination of: According to the construction formula, the area of ​​different cloud data sets is calculated. Where m is the number of convex hull vertices, and the determination of m is based on the number of temperature field distribution grids obtained by infrared photography; x i is the x coordinate of the i-th vertex, y i is the y coordinate of the i-th vertex; The x-coordinate and y-coordinate represent the values ​​of the two coordinate axes respectively and can be calculated by the following formula: Among them, Pixelx is the horizontal pixel, Pixely is the vertical pixel; dpi is the resolution; Exposed molten steel area ratio calculation module: calculates the exposed molten steel area ratio; bottom blowing mode selection module: matches the bottom blowing mode according to the exposed molten steel area ratio, and dynamically associates the matched bottom blowing mode with different LF refining operation processes.

Images