Steel ladle monitoring method and system

Abstract

The invention relates to the field of steel ladle intelligent management systems, in particular to a steel ladle monitoring method and system. The steel ladle monitoring method comprises the steps that an on-site steel ladle point cloud model is obtained; the site steel ladle point cloud model is compared with a preset template model through a point cloud registration algorithm to obtain steel ladle lining residual thickness information of different areas, and the residual thickness information comprises slag line residual thickness, ladle body residual thickness, ladle bottom residual thickness and breathable core information; numbering the steel ladles, obtaining the corresponding used ladle age and lining temperature, and predicting the remaining ladle age at least according to the steel ladle lining residual thickness information and the used ladle age; outputting radar early warning information at least according to the residual thickness information of the steel ladle lining; through a point cloud registration algorithm, the residual thickness information of the steel ladle lining in different areas can be accurately obtained, including key parts such as a slag line, a ladle body, a ladle bottom and a breathable core; and compared with a detection system for counting according to times, a more detailed, more accurate and credible steel ladle wear condition is provided.

Description

Technical Field

[0001] The present invention relates to the field of intelligent management systems for steel ladles, and in particular to a method and system for monitoring steel ladles. Background Art

[0002] In the steelmaking industry, the ladle is an important container for carrying and transporting high-temperature molten steel. Its stability and safety are directly related to the continuity and safety of steelmaking production. Traditional ladle management methods mainly rely on recording the age of the ladle and the operator's visual inspection and experience of the degree of erosion of the ladle lining. This method has many shortcomings, which seriously restricts the improvement of steelmaking production efficiency and the guarantee of safe production level.

[0003] In view of this, the present invention is provided. Summary of the invention

[0004] To solve the above problems, the ladle monitoring method and system provided in the present application are intended to solve the problems of inaccurate lining erosion assessment, low safety and unreasonable refractory consumption in traditional ladle management methods.

[0005] In the first aspect of the present application, a ladle monitoring method is provided, comprising: obtaining an on-site ladle point cloud model; comparing the on-site ladle point cloud model with a preset template model through a point cloud registration algorithm to obtain the residual thickness information of the ladle lining in different areas, wherein the residual thickness information includes the residual thickness of the slag line, the residual thickness of the ladle body, the residual thickness of the ladle bottom and the air core information; numbering the ladle and obtaining the corresponding used ladle age and lining temperature, and calculating the erosion rate of different areas at least based on the residual thickness information of the ladle lining and the used ladle age, and predicting the remaining ladle age based on the erosion rate of different areas and the used ladle age.

[0006] In a further scheme of the present application, the on-site point cloud model of the ladle and the preset template model are compared through a point cloud registration algorithm to obtain the residual thickness information of the ladle lining in different areas, including: stratifying the ladle along the depth direction, and calibrating the data of each refractory brick of the slag line and the ladle body, each refractory brick includes a number and a corresponding azimuth; aligning the on-site point cloud model of the ladle and the template model based on the same central axis; obtaining the measurement radius according to the shortest distance from each refractory brick to the central axis in the on-site point cloud model of the ladle, and comparing the measured radius with the actual radius of the corresponding position to calculate the residual thickness of the slag line and the residual thickness of the ladle body.

[0007] In a further scheme of the present application, the on-site ladle point cloud model and the preset template model are compared through a point cloud registration algorithm to obtain the residual thickness information of the ladle lining in different areas, including: data calibration of each refractory brick of the ladle bottom and the breathable core, each refractory brick includes a number and a corresponding azimuth; the on-site ladle point cloud model and the template model are aligned based on the same central axis; a straight line is fitted on the on-site ladle point cloud model to extract the outer bottom line of the on-site ladle point cloud model; translation is performed to align the outer bottom line of the on-site ladle point cloud model with the template model, and the inner bottom of the ladle of the on-site ladle point cloud model and the actual inner bottom are compared to calculate the residual thickness of the ladle bottom of the corresponding refractory brick; the breathable core information is obtained based on the inner wall diameter of the breathable core of the on-site ladle point cloud model and the actual breathable core diameter.

[0008] In a further scheme of the present application, the ladles are numbered and the corresponding used ladle ages and lining temperatures are obtained, and the remaining ladle age is predicted at least based on the ladle lining residual thickness information and the used ladle age, including: obtaining the minimum slag line residual thickness among the slag line residual thicknesses of several refractory bricks, the minimum ladle body residual thickness among the ladle body residual thicknesses, the minimum ladle bottom residual thickness among the ladle bottom residual thicknesses, and the minimum permeable core residual thickness in the permeable core information by sorting; obtaining the slag line residual thickness erosion rate of the slag line per unit time, the ladle body residual thickness erosion rate of the ladle body per unit time, the ladle bottom residual thickness erosion rate of the ladle bottom per unit time, and The erosion rate of the breathable core per unit time; the first duration is calculated according to the minimum slag line residual thickness, the erosion rate of the slag line residual thickness and the preset slag line safety residual thickness; the second duration is calculated according to the minimum package body residual thickness, the erosion rate of the package body residual thickness and the preset package body safety residual thickness; the third duration is calculated according to the minimum package bottom residual thickness, the erosion rate of the package bottom residual thickness and the package bottom safety residual thickness; the fourth duration is calculated according to the minimum breathable core residual thickness, the erosion rate of the breathable core and the safety residual thickness of the breathable core; the minimum value among the first duration, the second duration, the third duration and the fourth duration is the predicted remaining package age.

[0009] In a further scheme of the present application, the slag line residual thickness erosion rate per unit time, the ladle body residual thickness erosion rate per unit time, the ladle bottom residual thickness erosion rate per unit time and the breathable core erosion rate per unit time are obtained, including: obtaining a first linear graph of the minimum slag line residual thickness associated with unit time according to the age of the used ladle, and obtaining the slag line residual thickness erosion rate by fitting the linear relationship of the first linear graph; obtaining a second linear graph of the minimum ladle body residual thickness associated with unit time according to the age of the used ladle, and obtaining the ladle body residual thickness erosion rate by fitting the linear relationship of the second linear graph; obtaining a third linear graph of the minimum ladle bottom residual thickness associated with unit time according to the age of the used ladle, and obtaining the ladle bottom residual thickness erosion rate by fitting the linear relationship of the third linear graph; obtaining a fourth linear graph of the minimum breathable core residual thickness associated with unit time according to the age of the used ladle, and obtaining the breathable core erosion rate by fitting the linear relationship of the fourth linear graph.

[0010] In a further scheme of the present application, the ladles are numbered and the corresponding used ladle ages and lining temperatures are obtained, and the remaining ladle age is predicted at least based on the ladle lining residual thickness information and the used ladle age, and also includes: when the lining temperature is higher than a preset temperature threshold, obtaining the temperature difference between the lining temperature and the temperature threshold; and weighting the slag line residual thickness erosion rate, the ladle body residual thickness erosion rate, the ladle bottom residual thickness erosion rate and the air core erosion rate with different coefficients according to the temperature difference.

[0011] In a further scheme of the present application, the ladle monitoring method also includes outputting radar warning information at least according to the residual thickness information of the ladle lining, wherein the outputting radar warning information at least according to the residual thickness information of the ladle lining includes: outputting any one of slag line erosion, slag line penetration or slag line hanging slag according to the comparison between the minimum slag line residual thickness, the safe residual thickness of the slag line and the original thickness of the slag line; outputting any one of ladle body erosion, ladle penetration or ladle body hanging slag according to the comparison between the minimum ladle body residual thickness, the safe residual thickness of the ladle body and the original thickness of the ladle body; outputting any one of ladle bottom erosion, ladle bottom penetration or ladle bottom slag according to the comparison between the minimum ladle bottom residual thickness, the safe residual thickness of the ladle bottom and the original thickness of the ladle bottom; outputting any one of breathable core erosion, breathable core penetration or breathable core hanging slag according to the comparison between the minimum breathable core residual thickness, the safe residual thickness of the breathable core and the original thickness of the breathable core.

[0012] In a further scheme of the present application, before obtaining the on-site ladle point cloud model, it also includes: obtaining the on-site ladle bottom image; performing brightness statistics based on the ladle bottom image to obtain the pixel number distribution at multiple brightness levels; when the number of pixels in the brightness level higher than the brightness threshold is higher than the preset number threshold, outputting the ladle bottom molten steel impact.

[0013] In a further scheme of the present application, after obtaining the on-site ladle point cloud model, it also includes: fitting the first central axis of the on-site ladle point cloud model according to the edge of the ladle in at least the x-direction and the y-direction respectively; comparing the first central axis with the second central axis of the template model in the x-direction and the y-direction respectively, and if the offset of the first central axis and the second central axis in either the x-direction or the y-direction exceeds a preset deviation threshold, outputting a detection position abnormality.

[0014] The second aspect of the present application also provides a ladle monitoring system, including: an image detection module, used to obtain an on-site ladle point cloud model; an image analysis module, used to compare the on-site ladle point cloud model with a preset template model through a point cloud registration algorithm to obtain the residual thickness information of the ladle lining in different areas, wherein the residual thickness information includes the residual thickness of the slag line, the residual thickness of the ladle body, the residual thickness of the ladle bottom and the air core information; a data display and analysis module, which numbers the ladle and obtains the corresponding used ladle age and lining temperature, and predicts the remaining ladle age based on at least the residual thickness information of the ladle lining and the used ladle age; an early warning information output module, used to output radar early warning information based on at least the residual thickness information of the ladle lining.

[0015] The present application also provides a controller device having a computer-readable storage medium storing program instructions thereon, wherein the program instructions, when executed by a processor in the controller device, enable the controller device to execute the above method.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] Through the point cloud registration algorithm, the residual thickness information of the ladle lining in different areas can be accurately obtained, including key parts such as slag line, ladle body, ladle bottom and breathable core; compared with the detection system that counts the number of times, it provides more detailed, accurate and reliable ladle wear information, which can avoid individual differences and is suitable for continuous flow production. It can discover potential safety hazards, improve ladle utilization, and reduce the workload of manual inspection and recording; and combined with multi-dimensional data such as the age of the used ladle and the lining temperature, a more comprehensive ladle status assessment system can be constructed to provide effective reference information; this monitoring method is not limited by the type, size and material of the ladle, and can be widely used in the monitoring of different types of ladles. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0019] Figure 1 The overall flow chart of the ladle monitoring method for the ladle outlet provided by the embodiment of the present invention;

[0020] Figure 2 A schematic diagram of the structure of a steel ladle according to an embodiment of the present invention;

[0021] Figure 3 A comparison diagram of the on-site ladle point cloud model (blue) and the template model (red) provided in an embodiment of the present invention:

[0022] Figure 4A comparison diagram of a point cloud model of a steel ladle at a certain layer under a cross section provided by an embodiment of the present invention (blue) and a corresponding layer of a template model (red);

[0023] Figure 5 A grid display diagram of erosion degree provided in an embodiment of the present invention, wherein the horizontal axis is the azimuth angle from 0 to 360 degrees, and the vertical axis is the ladle number of the corresponding layer;

[0024] Figure 6 The radar warning information and its corresponding operation interface shown in the embodiment of the present invention; and

[0025] Figure 7 A line graph showing the association of a ladle and a slag line with time, and a fan graph showing the age prediction of a ladle, according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0027] The current existing technology evaluates the life and status of the ladle by recording the number of times and time the ladle is used, thereby achieving ladle management; this system estimates the life cycle of the ladle based on historical experience, and then counts the number of times based on time and area, thereby achieving ladle management.

[0028] However, the existing management methods are highly dependent on historical experience and estimated cycles. Due to the differences in the use environment, workload and maintenance conditions of different ladles, it is difficult to use a unified standard to accurately evaluate the life of all ladles. In addition, ladles are affected by many factors during use, such as the temperature of the molten steel, the chemical composition, the impact during transportation, etc., and the individual differences are large. In addition, for safety reasons, ladles are often hot-repaired or replaced when the specified number of times is reached, which to a certain extent leads to insufficient ladle utilization, reduces the utilization rate of the ladle and the efficiency of steelmaking production; and this type of ladle management system cannot provide users with reference information. In simple terms, users cannot obtain useful information about the ladle when hot-repairing the ladle, and only know that it needs to be replaced; therefore, when entering the hot repair, they still need to rely on manpower for detailed inspection work; this not only increases the complexity and cost of hot repair, but also limits users' comprehensive understanding of the ladle status and improvement of management efficiency.

[0029] See also Figures 1 to 6 In order to solve the problems in the prior art, the present invention provides a method for monitoring a ladle, comprising at least the following steps:

[0030] Step S10, obtaining the on-site ladle point cloud model;

[0031] Step S20, comparing the on-site ladle point cloud model with a preset template model through a point cloud registration algorithm to obtain ladle lining residual thickness information in different areas, wherein the residual thickness information includes slag line residual thickness, ladle body residual thickness, ladle bottom residual thickness and air core information;

[0032] Step S30, number the ladles and obtain the corresponding used ladle age and lining temperature, and calculate the erosion rate of different areas based on at least the residual thickness information of the ladle lining and the used ladle age, and predict the remaining ladle age based on the erosion rate of different areas and the used ladle age.

[0033] Based on the general concept of the present application, a three-dimensional scanning device (such as a lidar) is used to scan the on-site ladle to obtain its point cloud data; the purpose is to obtain accurate three-dimensional shape information of the on-site ladle and provide basic data for subsequent analysis.

[0034] like Figure 2 The slag line, ladle body, ladle bottom, air core, etc. shown in the figure are used to align the point cloud model obtained on site with the preset template model representing the ideal or standard state; and by comparison, the residual thickness information of the ladle lining in different areas is identified; the purpose is to obtain the detailed wear condition of the ladle lining on site, and provide a basis for subsequent prediction and early warning.

[0035] This application assigns a unique number to each ladle, records its used ladle age and lining temperature, and uses appropriate algorithms or models (such as machine learning models, empirical formulas, etc.) to predict the remaining service life of the ladle based on the residual thickness information of the ladle lining and the used ladle age. This allows the life of the ladle to be obtained in advance, providing a basis for production planning and maintenance arrangements.

[0036] Each ladle is uniquely numbered, and its used ladle age and lining temperature are recorded; the erosion rate of each area is calculated based on the residual thickness information and used ladle age of different areas. The erosion rate is calculated by deducing the relationship between the residual thickness reduction and time, which reflects the wear rate of the ladle lining; further based on the erosion rate and the used ladle age, an appropriate mathematical model or algorithm is used to predict the remaining ladle age of the entire ladle; that is, the current erosion rate is extrapolated to the future to obtain the remaining life of the ladle.

[0037] Based on the residual thickness information of the ladle lining and the used ladle age, the erosion rate in different areas can be calculated, and the remaining service life of the ladle can be predicted accordingly. The maintenance or replacement plan of the ladle can be planned in advance to avoid production interruptions and safety accidents caused by sudden equipment failure, and production plans and maintenance strategies can be formulated in advance to improve the overall operation level. Especially compared with the traditional monitoring method based on the number of times, by accurately measuring the residual thickness information of the ladle lining and considering multiple factors such as the used ladle age and lining temperature, the current state of the ladle can be more accurately evaluated and its remaining life can be predicted; providing more reliable data support and reducing misjudgments caused by inaccurate estimates.

[0038] Optionally, based on the above-predicted remaining ladle age, as well as the detected residual thickness information, used ladle age, ladle phase data (including but not limited to lining temperature, number), etc., the present application can also selectively output warning information.

[0039] For example, warning conditions can be set, and when the warning conditions are met, warning information is automatically output; the output carrier may be an industrial computer, a mobile terminal, or a remote platform.

[0040] The warning information can be output selectively, and may include the number of the ladle, the specific value of the residual thickness of the lining, the task status, the maintenance post number, etc. The warning information can be transmitted through the interface of the mobile terminal or the industrial control terminal; and relevant personnel can be reminded in time to pay attention to the wear of the ladle to avoid the occurrence of safety accidents.

[0041] Through the point cloud registration algorithm, the residual thickness information of the ladle lining in different areas can be accurately obtained, including key parts such as slag line, ladle body, ladle bottom and breathable core; compared with the detection system based on the number of statistics, it provides more detailed, accurate and reliable ladle wear information, which can discover potential safety hazards, improve ladle utilization, and reduce the workload of manual inspection and recording; and combined with multi-dimensional data such as the age of the used ladle and the lining temperature, a more comprehensive ladle status assessment system can be constructed to provide effective reference information; this monitoring method is not limited by the type, size and material of the ladle, and can be widely used in the monitoring of different types of ladles.

[0042] Combination Figure 2 Based on the above general concept, a specific embodiment is provided below:

[0043] In one implementation of step S20, the on-site ladle point cloud model is compared with the preset template model through a point cloud registration algorithm to obtain the ladle lining residual thickness information in different areas, including:

[0044] Step S210, the ladle is layered along the depth direction, and data calibration is performed on each refractory brick of the slag line and the ladle body, each refractory brick includes a number and a corresponding azimuth;

[0045] The ladle is evenly divided into 10 layers along the depth direction, and each layer contains several refractory bricks. Data calibration is performed on each refractory brick of the slag line and the ladle body, including the number and the corresponding azimuth. The refractory bricks are numbered from 1 to N (N is the total number of refractory bricks, Figure 2 36), for example, the number of layers is 14 and the azimuth angle is 60; since the main part of the lining of the ladle is in the shape of a body of revolution, the azimuth angle is based on the central axis of the ladle from top to bottom in a counterclockwise direction.

[0046] Step S220, aligning the on-site ladle point cloud model and the template model based on the same central axis;

[0047] The on-site ladle point cloud model and the template model are aligned based on the same center axis; that is, it is achieved through a point cloud registration algorithm, and the on-site ladle point cloud model is moved through matrix transformation to ensure that the two models are in the same coordinate system.

[0048] Step S230: obtaining a measured radius according to the shortest distance from each refractory brick to the central axis in the on-site ladle point cloud model, and comparing the measured radius with the actual radius of the corresponding position to calculate the slag line residual thickness and the ladle body residual thickness.

[0049] like Figure 3 As shown in the figure, the comparison between the on-site ladle point cloud model (blue) and the template model (red): Figure 4 As shown, it is a comparison diagram of the on-site ladle point cloud model (blue) of a certain layer under the cross section and the corresponding layer of the template model (red):

[0050] In the on-site ladle point cloud model, a separate calculation is performed on each layer, that is, the maximum distance from each refractory brick to the center axis of the circle is found in each layer as the measurement radius; this is achieved by calculating the vertical distance from each point in the point cloud to the center axis; and the measured radius is compared with the actual radius of the corresponding position (that is, the radius in the template model).

[0051] In specific calculations, the center axis of each layer is simplified to a center point coordinate; for each refractory brick in each layer, it is necessary to find the maximum distance from the inner edge of the refractory brick to the center point within the angle range corresponding to the layer. This distance is defined as the measurement radius.

[0052] For each layer and each refractory brick, traverse the edge point cloud located on the inner side and calculate the distance from each point to the center point; after the traversal is completed, the point farthest from the center point is the point with the largest distance recorded, that is, the measured radius; the calculation formula can be the Euclidean distance formula.

[0053] In this application, the actual radius is the original size of the refractory brick in the template model; the slag line residual thickness and the ladle residual thickness are the measured actual radius, and the erosion value can be calculated by the following formula:

[0054] Erosion value = actual radius - measured radius;

[0055] For example, assuming that the actual radius of the first refractory brick (No. 1) of the first layer of the slag line is R1=200mm, and the measured radius is r1=180mm, then the residual thickness of the first refractory brick of the first layer of the slag line is: Δ1=R1-r1=200mm-180mm=20mm.

[0056] Thus, we can obtain Figure 5 A grid display of erosion degree, with the horizontal axis being the azimuth angle from 0 to 360 degrees and the vertical axis being the ladle number of the corresponding layer.

[0057] In one implementation of step S20, the on-site ladle point cloud model is compared with the preset template model through a point cloud registration algorithm to obtain the ladle lining residual thickness information in different areas, including:

[0058] Step S210', calibrate the data of each refractory brick of the bottom and the air-permeable core, each refractory brick includes a number and a corresponding azimuth;

[0059] This step is the same as above and will not be repeated here.

[0060] Step S220', aligning the on-site ladle point cloud model and the template model based on the same central axis;

[0061] As described above, the central axes of the on-site ladle point cloud model and the template model are identified and determined, and the on-site ladle point cloud model is converted to align with the template model.

[0062] Step S230', fitting a straight line to the on-site ladle point cloud model to extract the outer bottom line of the on-site ladle point cloud model;

[0063] On the on-site ladle point cloud model, a straight line is fitted at the bottom of the outer side of the ladle using the least square method or other curve fitting algorithms, and the straight line represents the outer bottom line of the ladle.

[0064] Step S240', translate to align the outer bottom line of the on-site ladle point cloud model with the template model, compare the inner bottom of the ladle based on the on-site ladle point cloud model and the actual inner bottom to calculate the residual thickness of the ladle bottom; obtain the air core information based on the inner wall diameter of the air core of the on-site ladle point cloud model and the actual air core diameter.

[0065] The outer bottom line of the on-site ladle point cloud model is translated to align with the outer bottom line of the template model. It is assumed that the outer bottom line equation of the on-site ladle point cloud model is obtained by least squares fitting (simplified on the yz or xz two-dimensional plane, which should actually be a plane in three-dimensional space). In this example, for simplicity, the outer bottom line equation is z=b. By translating up and down, the outer bottom line is aligned with the bottom of the actual template model.

[0066] After aligning the bottom and the center axis, the residual thickness of the ladle bottom and the ladle bottom erosion can be accurately obtained by comparing the measured inner bottom of the ladle with the actual inner bottom. Specifically, the residual thickness of the ladle bottom is the lowest point of the inner bottom of the ladle, and the ladle bottom erosion is the difference between the lowest point of the inner bottom of the ladle corresponding to the refractory brick and the actual inner bottom.

[0067] In an implementation manner corresponding to step S30, the ladles are numbered and the corresponding used ladles age and lining temperature are obtained, and the remaining ladles age is predicted at least based on the residual thickness information of the lining of the ladles and the used ladles age, including:

[0068] S310, obtaining respectively by sorting the minimum slag line residual thickness among the slag line residual thicknesses of several refractory bricks, the minimum ladle body residual thickness among the ladle body residual thicknesses, the minimum ladle bottom residual thickness among the ladle bottom residual thicknesses, and the minimum permeable core residual thickness among the permeable core information;

[0069] S320, obtaining the slag line residual thickness erosion rate per unit time, the ladle body residual thickness erosion rate per unit time, the ladle bottom residual thickness erosion rate per unit time, and the air core erosion rate per unit time;

[0070] S330, calculating a first duration according to the minimum slag line residual thickness, the slag line residual thickness erosion rate and the preset slag line safety residual thickness, calculating a second duration according to the minimum package body residual thickness, the package body residual thickness erosion rate and the preset package body safety residual thickness, calculating a third duration according to the minimum package bottom residual thickness, the package bottom residual thickness erosion rate and the package bottom safety residual thickness, and calculating a fourth duration according to the minimum air permeable core residual thickness, the air permeable core erosion rate and the air permeable core safety residual thickness;

[0071] S340: Take the minimum value among the first duration, the second duration, the third duration and the fourth duration as the predicted remaining packet age.

[0072] In order to avoid local penetration, the minimum values ​​of the residual thickness of several refractory bricks in different areas (such as slag line, ladle body, ladle bottom and breathable core) are obtained respectively. The residual thickness of the refractory bricks in each area needs to be sorted and their respective minimum values ​​extracted; specifically, they can be divided according to the above-mentioned numbering.

[0073] For example, the erosion rate of each area is divided into: slag line residual thickness erosion rate: R slagIndicates: the residual thickness erosion rate of the package body: R body Indicates: the residual thickness erosion rate of the bottom of the package: R bottom Indicates; breathable core erosion rate: R vent express.

[0074] Initial thickness:

[0075] Initial thickness of slag line: T slag,initial Indicates; initial thickness of the bag body: T body,initial Indicates; initial thickness of the bottom of the package: T bottom,initial Indicates; initial thickness of the breathable core: T vent,initial express.

[0076] Safety residual thickness:

[0077] Safe residual thickness of slag line: use T slag,safe Indicates; bag body safety thickness: T body,safe Indicates: Safety residual thickness of the bottom of the bag: T bottom,safe Indicates; breathable core safety residual thickness: T vent,safe express.

[0078] Used package age: Indicated by n, which indicates the usage time from the initial to the current time.

[0079] In one approach, the erosion rate R i It can be calculated by the following formula:

[0080]

[0081] ΔTi represents the erosion value of a certain area i (such as the slag line, the ladle body, the ladle bottom and one of the air-permeable core) after the ladle has been used for a long time. The specific calculation method can be found above.

[0082] For each area, its remaining useful life (i.e. the remaining package age that can still be used) can be calculated:

[0083]

[0084] In the obtained multiple regions n remaining,i In the above example, select the smallest n remaining,i As the maximum predicted package age, it means that area i will reach its safety residual thickness threshold first:

[0085] n remaining,i =min(n remaining,slag , n remaining,body , n remaining,botten , n remaining,vent )

[0086] The predicted remaining packet age is then output.

[0087] In the above step S330, the erosion rate R i You can also use the following methods:

[0088] Obtain the slag line residual thickness erosion rate per unit time, the ladle body residual thickness erosion rate per unit time, the ladle bottom residual thickness erosion rate per unit time, and the permeable core erosion rate per unit time, including:

[0089] Step S331, obtaining a first line graph of the minimum slag line residual thickness associated with unit time according to the used package age, and obtaining the slag line residual thickness erosion rate by fitting the linear relationship of the first line graph;

[0090] Step S332, obtaining a second linear graph of the minimum ladle body residual thickness associated with unit time according to the used ladle age, and obtaining the ladle body residual thickness erosion rate by fitting the linear relationship of the second linear graph;

[0091] Step S333, obtaining a third linear graph of the minimum lap bottom residual thickness associated with unit time according to the used lap age, and obtaining the lap bottom residual thickness erosion rate by fitting the linear relationship of the third linear graph;

[0092] Step S334: obtaining a fourth linear graph of the minimum air permeable core residual thickness and unit time according to the used package age, and obtaining the air permeable core erosion rate by fitting the linear relationship of the fourth linear graph.

[0093] The first line graph, the second line graph, the third line graph and the fourth line graph may be straight line graphs or curved line graphs, for example:

[0094] like Figure 6 and Figure 7 , plot these data into a line graph (i.e., the first line graph, as shown in the upper right corner of the figure), with the horizontal axis being the age of the used package and the vertical axis being the minimum residual thickness of the slag line; try to fit a linear line graph (the first line graph), in the form of y=mx+b, where y is the minimum residual thickness of the slag line, x is the age of the used package, m is the slope (i.e., the erosion rate of the residual thickness of the slag line), and b is the intercept, and fit through the points on the first linear graph to output the first line graph with the highest fitting degree.

[0095] In some embodiments, if the erosion rate gradually increases with the increase of the used package age (i.e., accelerated erosion), an exponential function can be used to describe this relationship. On the contrary, if the erosion rate gradually slows down with the increase of the used package age (i.e., decelerated erosion), a logarithmic function or a power function can be used to describe it. Until a first line graph that best fits the corresponding point on the first line graph is fitted.

[0096] By fitting the model to determine the erosion rate with multiple data points, the results are more robust and reliable. Direct calculation only considers the terminal data points. The fitting model can capture the long-term trend of the erosion rate and reduce the impact of accidental errors.

[0097] It should be noted that the present application is applicable to the monitoring of ladles in continuous production. If the ladle stops production, that is, stops being used on the timeline, the ladle with the corresponding number will go offline to stop the calculation (to avoid erroneous interference in the calculation of the erosion rate), and the calculation will resume when the time scale for the continuous production state begins.

[0098] Furthermore, in step S30, the ladles are numbered and the corresponding used ladles age and lining temperature are obtained, and the remaining ladles age is predicted at least based on the residual thickness information of the lining of the ladles and the used ladles age, and further includes:

[0099] Step S341: when the lining temperature is higher than a preset temperature threshold, obtaining a temperature difference between the lining temperature and the temperature threshold;

[0100] Step S342, weighting the slag line residual thickness erosion rate, the ladle body residual thickness erosion rate, the ladle bottom residual thickness erosion rate and the air permeable core erosion rate with different coefficients according to the temperature difference.

[0101] In order to weight the lining temperature T into the calculation of the erosion rate, we need to assume that there is a certain functional relationship between the erosion rate and the temperature. This relationship may be linear, exponential or other forms, but it usually needs to be determined by experimental data. For example, a simplified model is used to assume that there is a linear relationship between the erosion rate and the temperature, that is, the erosion rate increases linearly with the increase of temperature.

[0102] For each part (slag line, ladle body, ladle bottom, breathable core), a weighted erosion rate Ri'(T) can be defined, taking into account the impact of the lining temperature T exceeding the standard:

[0103] Ri'(T)=Ri·(1+ai·(T-Tref))

[0104] Where: Ri is the erosion rate before weighting; a i It is the coefficient of the influence of temperature on the erosion rate, which needs to be determined through experimental data; Tref is a reference temperature, which can be a certain base temperature or average temperature; T is the current lining temperature, and T-Tref is the temperature difference.

[0105] Using the weighted erosion rate Ri'(T), the remaining useful life of each part can be calculated:

[0106]

[0107] Select the minimum weighted remaining useful life as the maximum predicted package age: where the coefficient α i It is determined that it needs to be fitted by experimental data. It may be necessary to conduct a series of experiments, measure the erosion rate at different temperatures, and use methods such as regression analysis to determine αi.

[0108] To apply this model, the real-time temperature of the liner needs to be accurately measured using a durable and high-temperature temperature sensor.

[0109] By monitoring the lining temperature and comparing it with the preset temperature threshold, it can respond quickly when the temperature rises abnormally; allowing the system or operator to dynamically adjust the calculation or prediction of the erosion rate according to the current actual temperature conditions; different lining areas (such as slag line, ladle body, ladle bottom and breathable core) may have different sensitivities to temperature changes. By weighting the erosion rates of different areas with different coefficients according to the temperature difference, this difference can be reflected more accurately. By monitoring the lining temperature and comparing it with the preset temperature threshold, when the temperature is higher than the threshold, the erosion rates of different areas are weighted and adjusted according to the temperature difference, the actual erosion situation can be predicted more accurately. This dynamic adjustment can reflect the accelerating effect of high temperature environment on erosion rate, thereby avoiding prediction deviation caused by over-reliance on raw data.

[0110] Furthermore, the method for monitoring the ladle also includes:

[0111] Step S40, outputting radar warning information at least according to the ladle lining residual thickness information;

[0112] Wherein, step S40 includes:

[0113] Step S410, outputting any one of slag line molten steel influence, slag line penetration or slag line hanging slag according to the comparison between the minimum slag line residual thickness, the slag line safety residual thickness and the slag line original thickness;

[0114] Step S420, outputting any one of the ladle body molten steel influence, ladle body penetration or ladle body slag hanging according to the comparison between the minimum ladle body residual thickness, the ladle body safety residual thickness and the ladle body original thickness;

[0115] Step S430, outputting any one of the influence of molten steel on the ladle bottom, ladle penetration or ladle slag on the ladle bottom according to the comparison between the minimum ladle bottom residual thickness, the ladle bottom safety residual thickness and the ladle bottom original thickness;

[0116] Step S440: output any one of the following: molten steel impact on the breathable core, penetration of the breathable core, or slag hanging on the breathable core, based on the comparison among the minimum residual thickness of the breathable core, the safe residual thickness of the breathable core, and the original thickness of the breathable core.

[0117] It can be understood that, taking step S420 as an example, the system compares the minimum bag body residual thickness with the safety residual thickness threshold.

[0118] If the minimum ladle residual thickness is less than the safety residual thickness threshold, but greater than or equal to zero (or a set minimum acceptable residual thickness value), it means that the ladle has begun to be corroded, but has not yet reached a level that endangers safety. At this time, the system may output an early warning message of "Ladle body molten steel impact", prompting the operator to pay attention to the ladle condition and prepare for necessary maintenance.

[0119] If the minimum ladle residual thickness is less than zero (or the set minimum acceptable residual thickness value), it indicates that the ladle has been penetrated and molten steel may leak; at this time, the system should urgently output the warning information of "ladle penetration" and immediately take safety measures such as shutdown and evacuation of personnel. In addition, if the system detects an abnormal increase in the ladle residual thickness (i.e., exceeding the original thickness), this may be caused by abnormal conditions such as slag. At this time, the system may output the warning information of "ladle slag", prompting the operator to clean and inspect.

[0120] According to the results of comparative analysis, the system outputs corresponding warning information through preset communication channels (such as industrial computer interface, sound alarm, etc.); so that the operator can quickly understand the package status and take corresponding countermeasures.

[0121] Before obtaining the on-site ladle point cloud model, it also includes:

[0122] Step S510, obtaining an image of the bottom of the steel ladle at the scene;

[0123] Step S520, performing brightness statistics according to the ladle bottom image to obtain pixel quantity distribution at multiple brightness levels;

[0124] Step S530: When the number of pixels in the brightness level higher than the brightness threshold is higher than the preset number threshold, the impact of the molten steel at the bottom of the ladle is output.

[0125] This step is triggered every time before step S10, that is, when the on-site ladle point cloud model is obtained, the on-site ladle bottom image (a two-dimensional image taken from a bird's-eye view) is obtained at the same time. The system performs brightness statistics on the image. If the number of pixels above a certain brightness threshold significantly exceeds the preset number threshold, it can be determined whether there is residual molten steel. For example, an RGB format ladle bottom image is obtained, and the RGB image is converted into a grayscale image for brightness statistics. The pixel value range of the grayscale image is between 0 (black) and 255 (white); the brightness of the grayscale image is divided into 10 levels, and the range of each level is as follows: Level 1: 0-25; Level 2: 26-50; ... Level 10: 231-255. Traverse each pixel of the grayscale image, count the number of pixels at each brightness level, and set the brightness threshold to level 8 (that is, the pixel value range is between 206-230), that is, this range usually corresponds to a brighter area, which may indicate light reflected by molten steel. Set the number threshold to 1000 pixels. This means that if the number of pixels with brightness level 8 and above exceeds 1000, the system will consider that there is molten steel influence on the bottom of the ladle and output the "molten steel influence on the bottom of the ladle" warning information.

[0126] After eliminating the influence of molten steel, the point cloud model can more truly reflect the structure and status of the ladle, thereby improving the accuracy of monitoring, avoiding misjudgment and notifying personnel in time to handle it; significantly improving the accuracy of monitoring, enhancing safety, optimizing production processes and reducing maintenance costs.

[0127] After obtaining the on-site ladle point cloud model, it also includes:

[0128] Step S610, fitting a first central axis of the on-site ladle point cloud model according to the ladle edge in at least the x-direction and the y-direction respectively;

[0129] Step S620, compare the first central axis with the second central axis of the template model in the x direction and the y direction respectively, if the offset of the first central axis and the second central axis in either the x direction or the y direction exceeds a preset deviation threshold, output a detection position abnormality.

[0130] This step is also triggered every time before step S10. Specifically, for example, from the side perspective in the x-direction, select the point set at the edge of the on-site ladle point cloud model from the side perspective in the x-direction. These point sets should be evenly distributed on the left and right edges of the ladle; taking the two left and right edges of the ladle as the reference, select a target point at a fixed interval (for example, every 5 cm) to ensure that multiple target points are evenly distributed on the left and right edges. For each pair of left and right target points, calculate their midpoint coordinates, and use the least squares method or other fitting algorithms (such as linear regression, polynomial fitting, etc.) for all midpoint coordinates to obtain the position of the first central axis in the x-direction. This central axis can be represented as a straight line; according to the steps of fitting the central axis in the x-direction, similarly select the target point in the y-direction, calculate the midpoint coordinates, and use the fitting algorithm to obtain the position of the first central axis in the y-direction. Combining the x-direction and y-direction, the spatial orientation of the first central axis can be known.

[0131] The first center axis of the on-site ladle point cloud model has been fitted through step S610, and the position information of the center axis in the x-direction and the y-direction is known; at the same time, the position information of the second center axis of the template model has also been obtained, which is extracted from the standard ladle model and represents the center position of the ladle in an ideal state. The offset can be obtained by calculating the coordinate difference between the two center axes in the corresponding directions. For example, if the position of the first center axis in the x-direction is x1, and the position of the second center axis in the x-direction is x2, the offset is |x1-x2|; the calculated offset is compared with the preset deviation threshold. If any offset in the x-direction or y-direction exceeds the preset deviation threshold, it is considered that the position of the on-site ladle point cloud model has been abnormally offset. At this time, the user can recalibrate the radar equipment, adjust the detection configuration, etc. according to the direction and size of the error to improve the accuracy of the detection.

[0132] The second aspect of the present application further provides a ladle monitoring system 100 , including an image detection module 10 , an image analysis module 20 , a data display and analysis module 30 , and an early warning information output module 40 .

[0133] The image detection module 10 is used to obtain the on-site ladle point cloud model; the image analysis module 20 is used to compare the on-site ladle point cloud model with the preset template model through a point cloud registration algorithm to obtain the residual thickness information of the ladle lining in different areas, wherein the residual thickness information includes the residual thickness of the slag line, the residual thickness of the ladle body, the residual thickness of the ladle bottom and the air core information; the data display and analysis module 30 numbers the ladle and obtains the corresponding used ladle age and lining temperature, and predicts the remaining ladle age based on at least the residual thickness information of the ladle lining and the used ladle age; the warning information output module 40 is used to output radar warning information based on at least the residual thickness information of the ladle lining.

[0134] It can be understood that the above modules can infer their processes and implementation methods based on the aforementioned method embodiments, which will not be repeated here. Those skilled in the art should also know that the above methods can also be used in the corresponding modules of this system.

[0135] In order to solve the problems in the prior art, the present invention also provides a controller device, characterized in that it includes a computer-readable storage medium having program instructions stored thereon, and the program instructions, when executed by a processor in the controller device, enable the controller device to execute a method.

[0136] In the present application, unless otherwise expressly specified and limited, the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. The first feature being "up" or "down" the second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, the first feature being "above", "above" and "above" the second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. The first feature being "below", "below" and "below" the second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature, and therefore cannot be understood as limiting the present invention.

[0137] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited thereto. Within the technical concept of the present invention, the technical solution of the present invention can be subjected to a variety of simple modifications, including the combination of various specific technical features in any appropriate manner. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations. However, these simple modifications and combinations should also be regarded as the contents disclosed by the present invention and belong to the protection scope of the present invention.

[0138] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A ladle monitoring method, characterized in that: The ladle monitoring method comprises: Obtain the on-site ladle point cloud model; The on-site ladle point cloud model is compared with a preset template model by a point cloud registration algorithm to obtain the ladle lining residual thickness information in different areas, wherein the residual thickness information includes the slag line residual thickness, ladle body residual thickness, ladle bottom residual thickness and air permeable core information; The ladles are numbered and the corresponding used ladle ages and lining temperatures are obtained, and the erosion rates of different areas are calculated at least based on the residual thickness information of the ladle lining and the used ladle age, and the remaining ladle age is predicted based on the erosion rates of different areas and the used ladle age.

2. The ladle monitoring method according to claim 1, characterized in that: The on-site ladle point cloud model is compared with a preset template model by a point cloud registration algorithm to obtain the ladle lining residual thickness information in different areas, including: The ladle is layered along the depth direction, and data calibration is performed on each refractory brick of the slag line and the ladle body, each refractory brick includes a number and a corresponding azimuth; Aligning the on-site ladle point cloud model and the template model based on the same central axis; The measured radius is obtained according to the shortest distance from each refractory brick to the central axis in the on-site ladle point cloud model, and the measured radius is compared with the actual radius of the corresponding position to calculate the residual thickness of the slag line and the residual thickness of the ladle body.

3. The ladle monitoring method according to claim 2, characterized in that: The on-site ladle point cloud model is compared with a preset template model by a point cloud registration algorithm to obtain the ladle lining residual thickness information in different areas, including: Data calibration is performed on each refractory brick of the base and the air-permeable core, and each refractory brick includes a number and a corresponding azimuth angle; Aligning the on-site ladle point cloud model and the template model based on the same central axis; Fitting a straight line to the on-site ladle point cloud model to extract the outer bottom line of the on-site ladle point cloud model; The outer bottom line of the on-site ladle point cloud model is translated to align with the template model, and the inner bottom of the ladle of the on-site ladle point cloud model and the actual inner bottom are compared to calculate the residual thickness of the ladle bottom of the corresponding refractory bricks; the permeable core information is obtained according to the inner wall diameter of the permeable core of the on-site ladle point cloud model and the actual permeable core diameter.

4. The ladle monitoring method according to claim 1, characterized in that: The ladle is numbered and the corresponding used ladle age and lining temperature are obtained, and the erosion rate of different areas is calculated at least according to the residual thickness information of the ladle lining and the used ladle age, and the remaining ladle age is predicted based on the erosion rate of different areas and the used ladle age, including: By sorting, the minimum slag line residual thickness among the slag line residual thicknesses of several refractory bricks, the minimum ladle body residual thickness among the ladle body residual thicknesses, the minimum ladle bottom residual thickness among the ladle bottom residual thicknesses, and the minimum permeable core residual thickness among the permeable core information are obtained respectively; Obtain the residual thickness erosion rate of the slag line per unit time, the residual thickness erosion rate of the ladle body per unit time, the residual thickness erosion rate of the ladle bottom per unit time, and the permeable core erosion rate of the permeable core per unit time; The first duration is calculated according to the minimum slag line residual thickness, the slag line residual thickness erosion rate and the preset slag line safety residual thickness; the second duration is calculated according to the minimum package body residual thickness, the package body residual thickness erosion rate and the preset package body safety residual thickness; the third duration is calculated according to the minimum package bottom residual thickness, the package bottom residual thickness erosion rate and the package bottom safety residual thickness; the fourth duration is calculated according to the minimum air permeable core residual thickness, the air permeable core erosion rate and the air permeable core safety residual thickness; The minimum value among the first duration, the second duration, the third duration and the fourth duration is used as the predicted remaining package age.

5. The ladle monitoring method according to claim 4, characterized in that: The method of obtaining the residual thickness erosion rate of the slag line per unit time, the residual thickness erosion rate of the ladle body per unit time, the residual thickness erosion rate of the ladle bottom per unit time, and the erosion rate of the air permeable core per unit time comprises: Obtain a first linear graph of the minimum slag line residual thickness associated with unit time according to the used package age, and obtain the slag line residual thickness erosion rate by fitting the linear relationship of the first linear graph; Obtain a second linear graph of the minimum ladle body residual thickness associated with unit time according to the used ladle age, and obtain the ladle body residual thickness erosion rate by fitting the linear relationship of the second linear graph; Obtain a third linear graph of the minimum residual thickness of the bottom of the package and the unit time according to the used package age, and obtain the residual thickness erosion rate of the bottom of the package by fitting the linear relationship of the third linear graph; A fourth linear graph of the minimum air permeable core residual thickness and unit time is obtained according to the used package age, and the air permeable core erosion rate is obtained by fitting the linear relationship of the fourth linear graph.

6. The ladle monitoring method according to claim 4, characterized in that: The method of numbering the ladles and obtaining the corresponding used ladles age and lining temperature, and predicting the remaining ladles age at least based on the residual thickness information of the lining of the ladles and the used ladles age, further includes: When the lining temperature is higher than a preset temperature threshold, obtaining a temperature difference between the lining temperature and the temperature threshold; According to the temperature difference, the residual thickness erosion rate of the slag line, the residual thickness erosion rate of the ladle body, the residual thickness erosion rate of the ladle bottom and the erosion rate of the air permeable core are weighted with different coefficients.

7. The ladle monitoring method according to claim 4, characterized in that: The ladle monitoring method further comprises outputting radar warning information at least according to the ladle lining residual thickness information, wherein the outputting radar warning information at least according to the ladle lining residual thickness information comprises: According to the comparison between the minimum slag line residual thickness, the safe slag line residual thickness and the original slag line thickness, outputting any one of slag line erosion, slag line penetration or slag line hanging; Outputting any one of body erosion, body penetration or body dross according to the comparison between the minimum body residual thickness, body safety residual thickness and body original thickness; According to the comparison between the minimum ladle bottom residual thickness, the ladle bottom safety residual thickness and the ladle bottom original thickness, outputting any one of ladle bottom erosion, ladle bottom penetration or ladle bottom dross; According to the comparison among the minimum breathable core residual thickness, the breathable core safety residual thickness and the breathable core original thickness, any one of breathable core erosion, breathable core penetration or breathable core slag is output.

8. The ladle monitoring method according to claim 7, characterized in that: Before obtaining the on-site ladle point cloud model, it also includes: Get the on-site ladle bottom image; Perform brightness statistics according to the ladle bottom image to obtain pixel quantity distribution at multiple brightness levels; When the number of pixels in the brightness level higher than the brightness threshold is higher than the preset number threshold, the ladle bottom molten steel impact is output.

9. The ladle monitoring method according to claim 7, characterized in that: After obtaining the on-site ladle point cloud model, it also includes: In at least the x-direction and the y-direction, respectively, a first central axis of the on-site ladle point cloud model is fitted according to the ladle edge; The first central axis is compared with the second central axis of the template model in the x direction and the y direction respectively. If the offset of the first central axis and the second central axis in either the x direction or the y direction exceeds a preset deviation threshold, an abnormal detection position is output.

10. A ladle monitoring system, characterized in that: include: Image detection module, used to obtain the on-site ladle point cloud model; An image analysis module is used to compare the on-site ladle point cloud model with a preset template model through a point cloud registration algorithm to obtain the ladle lining residual thickness information in different areas, wherein the residual thickness information includes the slag line residual thickness, ladle body residual thickness, ladle bottom residual thickness and air core information; The data display and analysis module numbers the ladles and obtains the corresponding used ladles age and lining temperature, and predicts the remaining ladles age at least based on the residual thickness information of the ladles lining and the used ladles age; The warning information output module is used to output radar warning information at least according to the residual thickness information of the ladle lining.

Images