Converter one-key tapping method based on fusion of visual perception and big data model

Abstract

The invention belongs to the technical field of converter tapping, and provides a converter one-key tapping method based on fusion of visual perception and a big data model, which comprises the following steps: on the basis of tapping historical data, dividing molten steel content intervals, and determining candidate uniform inclination angles of the intervals through inclination angle effective frequency analysis; a potential optimal dip angle pool is screened out through uniformity stability verification, and then the optimal dip angle of each interval is determined according to the molten steel flow uniformity. Meanwhile, the optimal dip angle maintaining time of multiple effective heat times is counted, an optimal dip angle-maintaining time length corresponding table of each molten steel content interval is formed, molten steel pouring is simulated based on the optimal dip angle-maintaining time length corresponding table, and a molten steel drop point position sequence in the optimal dip angle maintaining time length of each interval is output; and further determining the limited and optimal moving target point of the buggy ladle, and calculating the limited and optimal moving speed of the buggy ladle by combining the falling moments of the molten steel at the two points so as to guarantee the quality of the molten steel converter.

Description

Technical Field

[0001] This invention belongs to the field of converter steel tapping technology, specifically a one-click converter steel tapping method based on the integration of visual perception and big data model. Background Technology

[0002] In converter steelmaking, the control of molten steel flow uniformity and the coordination of ladle car positioning are key to ensuring steel quality. However, existing technologies have significant shortcomings and are difficult to meet the needs of precision production.

[0003] On the one hand, the matching of furnace tilt angle and molten steel content lacks scientific basis. Existing schemes mostly set fixed tilt angles based on a wide range of molten steel weights, without subdividing the molten steel content range to optimize parameters. Using the same tilt angle for different molten steel contents can easily lead to uneven flow, resulting in flow interruption or splashing. Moreover, the tilt angle maintenance time depends on experience and does not combine the optimal time with the statistics of historical effective heats. Inappropriate duration often causes process fluctuations and affects the stability of steel tapping.

[0004] On the other hand, the coordination precision between the molten steel drop point and the ladle car movement is insufficient. Existing technology lacks dynamic simulation analysis of "molten steel content - tilt angle - holding time - drop point position," making it impossible to obtain a precise drop point sequence. When the drop point deviation exceeds the standard, there is neither an "optimal moving target point" to make the drop point coincide with the center of the ladle, nor a "limited moving target point" and corresponding speed threshold to avoid accumulation. The ladle car movement is prone to being insufficient or excessive, leading to local accumulation of molten steel, uneven composition and temperature, and may also aggravate the wear of ladle lining bricks, thus restricting the improvement of molten steel quality and the optimization of production efficiency.

[0005] Therefore, this invention provides a one-click steel tapping method for converters based on the integration of visual perception and big data models. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0007] The technical solution adopted by this invention to solve its technical problem is: a one-click steel tapping method for converters based on the integration of visual perception and big data models, comprising:

[0008] Step 1: Based on historical steel tapping data, conduct effective frequency analysis of furnace tilt angles corresponding to different steel content ranges, determine candidate uniform tilt angles for each steel content range, and combine with uniformity stability verification to determine the potential optimal tilt angle pool for each steel content range.

[0009] Step 2: Based on the uniformity of molten steel flow, determine the optimal tilt angle for each steel content range in the potential optimal tilt angle pool, and determine the optimal tilt angle maintenance time for multiple effective heats to establish a corresponding table of optimal tilt angle and maintenance time for each steel content range.

[0010] Step 3: Simulate the pouring of molten steel according to the optimal tilt angle-holding time correspondence table for each molten steel content range, and determine the sequence of molten steel landing point positions within the optimal tilt angle holding time for each molten steel content range based on the simulation results;

[0011] Step 4: Based on the sequence of molten steel drop points, determine the limited target point and the optimal target point for the ladle car, and determine the limited speed and the optimal speed for the ladle car based on the time when the molten steel falls at the limited target point and the optimal target point.

[0012] As a further technical solution of the present invention, the process of analyzing the effective frequency of furnace tilt angles corresponding to different steel content ranges is as follows:

[0013] Based on any range of molten steel content;

[0014] Extract the tilt angle time sequence data of all valid heats, count the frequency of each tilt angle throughout the entire steel tapping process, and use the tilt angles with frequencies higher than the preset frequency as candidate uniformity tilt angles for the steel content range.

[0015] As a further technical solution of the present invention, the process of determining the potential optimal tilt angle pool for each steel content range is as follows:

[0016] Based on any candidate uniformity tilt angle within the steel content range;

[0017] The uniformity index of molten steel flow at different time points under the candidate uniformity inclination angle was obtained, and the uniformity maintenance value and uniform stability value were obtained by processing and analysis.

[0018] The deviation between the uniform maintenance value and the uniform stable value is calculated to obtain the uniform performance value.

[0019] By summing up the candidate uniformity tilt angles with uniformity performance values ​​greater than or equal to the uniformity performance threshold, the potential optimal tilt angle pool for the steel content range is obtained.

[0020] As a further technical solution of the present invention, the method for obtaining the uniformity maintenance value is as follows:

[0021] The time period in which the uniformity index of molten steel flow meets the normal range (the fluctuation range of the cross-sectional diameter of the molten steel flow is less than or equal to 5%) is marked as the uniform period.

[0022] The ratio between the uniform time period and the total steel tapping time is calculated to obtain the uniformity maintenance value.

[0023] As a further technical solution of the present invention, the method for obtaining the uniform stable value is as follows:

[0024] The uniformity index of molten steel flow within a uniform time period is summarized into a molten steel flow uniformity index sequence according to the time sequence. The coefficient of variation of the molten steel flow uniformity index sequence is calculated to obtain the uniform stable value.

[0025] As a further technical solution of the present invention, the process of determining the optimal tilt angle-maintenance duration correspondence table for each steel content range is as follows:

[0026] Based on any range of molten steel content;

[0027] Extract the holding time of all effective heats at the optimal tilt angle, and calculate the average holding time to obtain the optimal tilt angle holding time corresponding to the steel content range;

[0028] The optimal tilt angle and the duration of maintenance of the optimal tilt angle corresponding to different steel content ranges are integrated to obtain the corresponding table of optimal tilt angle and duration of maintenance for each steel content range.

[0029] As a further technical solution of the present invention, the process of determining the sequence of molten steel landing point positions within the optimal tilt angle maintenance time for each molten steel content range is as follows:

[0030] A three-dimensional geometric model of the steel furnace was established, and the flow of molten steel in different steel content ranges was simulated in segments according to the optimal tilt angle and optimal tilt angle maintenance time corresponding to different steel content ranges in the optimal tilt angle-maintenance time correspondence table.

[0031] The coordinates of the molten steel landing point per unit time within the optimal tilt angle maintenance duration for each molten steel content range are extracted from the simulation results and integrated into a sequence of molten steel landing point positions according to the time sequence.

[0032] A further technical solution of the present invention is as follows: the process of determining the limited target point and the optimal target point of the ladle car is as follows:

[0033] In the sequence of molten steel drop point positions, the deviation distance between the coordinates of the molten steel drop point and the center position coordinates of the current ladle car is calculated in order;

[0034] If the deviation distance is greater than the preset deviation distance, the position of the molten steel landing point will be marked as the optimal moving target point;

[0035] If the deviation distance is equal to the preset deviation distance, then mark the location of the molten steel drop point as the limited moving target point;

[0036] If the deviation distance is less than the preset deviation distance, the analysis of the next molten steel landing point will continue.

[0037] As a further technical solution of the present invention, the process of determining the limited moving speed of the ladle car is as follows:

[0038] Obtain the time when the molten steel falls at the target point and perform a difference operation with the current time to obtain the limited movement time of the ladle car. Calculate the ratio between the deviation distance corresponding to the target point and the limited movement time of the ladle car to obtain the limited movement speed of the ladle car.

[0039] As a further technical solution of the present invention, the process of determining the optimal moving speed of the ladle car is as follows:

[0040] Obtain the moment when the molten steel falls at the optimal target point and perform a difference calculation with the current moment to obtain the optimal travel time of the ladle car. Calculate the ratio between the deviation distance corresponding to the optimal target point and the optimal travel time of the ladle car to obtain the optimal travel speed of the ladle car.

[0041] The beneficial effects of this invention are as follows: Based on historical steelmaking data, the steel content range is first divided. Candidate uniform tilt angles for each range are determined through effective tilt angle frequency analysis. Potential optimal tilt angle pools are screened after uniformity stability verification. Then, the optimal tilt angle for each range is determined from the pool based on the uniformity of steel flow. Simultaneously, the optimal tilt angle maintenance time for multiple effective heats is statistically analyzed, forming a table corresponding to the optimal tilt angle and maintenance time for each steel content range. Subsequently, based on this table, molten steel pouring is simulated, outputting the sequence of molten steel landing points within the optimal tilt angle maintenance time for each range. This determines the constraints and optimal movement target points for the ladle car (the former being the minimum movement target to avoid localized molten steel accumulation, and the latter ensuring the molten steel landing point coincides with the center of the ladle car). Combining the molten steel landing times at these two points, the constraints and optimal movement speed of the ladle car are calculated. Through historical data modeling and simulation, precise matching of steel content, tilt angle, maintenance time, and ladle car movement is achieved, effectively controlling the deviation of the molten steel landing point, preventing composition and temperature unevenness caused by localized molten steel accumulation, and ensuring steel quality. Attached Figure Description

[0042] The invention will now be further described with reference to the accompanying drawings.

[0043] Figure 1 This is a flowchart of the steps of a converter one-click steel tapping method based on the fusion of visual perception and big data model as described in an embodiment of the present invention.

[0044] Figure 2 This is a diagram illustrating the method for determining the moving target point of the ladle car in a converter one-click tapping method based on the fusion of visual perception and big data model, as described in an embodiment of the present invention. Detailed Implementation

[0045] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0046] Example:

[0047] Please see Figure 1-2 As shown in the figure, the converter one-click tapping method based on the fusion of visual perception and big data model according to an embodiment of the present invention includes the following steps:

[0048] Step 1: Based on historical steel tapping data, conduct effective frequency analysis of furnace tilt angles corresponding to different steel content ranges, determine candidate uniform tilt angles for each steel content range, and combine with uniformity stability verification to determine the potential optimal tilt angle pool for each steel content range.

[0049] In step one, the historical steel tapping data includes the actual weight of molten steel in the furnace, the time sequence data of the tilting angle, and the uniformity index of molten steel flow at the time of historical steel tapping.

[0050] Among them, the uniformity index of molten steel flow represents the fluctuation range of the diameter of the steel flow section during historical tapping (obtained by visually detecting the maximum and minimum diameters of the steel flow section during molten steel flow and calculating the deviation), and the time series data of tilting angle represents the tilting angle of the steel furnace at different times.

[0051] In step one, the steel content range is divided by the actual weight of the molten steel in the furnace. For example, the range width is set to 5t (such as 200-205t, 205-210t...) to reduce the difference in the pressure of molten steel on the furnace wall and the initial flow velocity within the same range, and to ensure higher consistency of the flow characteristics of molten steel within the range.

[0052] In step one, the process of analyzing the effective frequency of furnace tilt angles corresponding to different steel content ranges is as follows:

[0053] Based on any range of molten steel content;

[0054] Extract the tilt angle time series data of all valid heats, and count the frequency of each tilt angle (such as 28°, 29°, 30°...) throughout the entire steel tapping process. The tilt angles with frequencies higher than the preset frequency are used as candidate uniformity tilt angles for the steel content range.

[0055] Among them, the effective heats refer to the percentage of time during which the uniformity of molten steel flow meets the normal range of the index, which is higher than the preset time (more than 80% of the total tapping time). Specifically, the uniformity of molten steel flow meeting the normal range of the index means that the fluctuation range of the steel flow cross-section diameter is less than or equal to 5%.

[0056] In step one, the process of determining the potential optimal tilt angle pool for each steel content range, in conjunction with the uniformity and stability verification, is as follows:

[0057] Based on any candidate uniformity tilt angle within the steel content range;

[0058] Obtain the uniformity index of molten steel flow at different time points under the candidate uniformity inclination angle, and mark the time period when the uniformity index of molten steel flow meets the normal range (the fluctuation range of the diameter of the steel flow section is less than ≤5%) as the uniform time period.

[0059] Calculate the ratio between the uniform time period and the total steel tapping time to obtain the uniformity maintenance value;

[0060] The uniformity index of molten steel flow within a uniform time period is summarized into a molten steel flow uniformity index sequence according to the time sequence. The coefficient of variation of the molten steel flow uniformity index sequence is calculated to obtain the uniform stable value.

[0061] The deviation between the uniform maintenance value and the uniform stable value is calculated to obtain the uniform performance value.

[0062] In some embodiments, the uniform performance value is compared with the uniform performance threshold;

[0063] If the uniformity performance value is greater than or equal to the uniformity performance threshold, the candidate uniformity tilt angle is retained.

[0064] If the uniformity performance value is less than the uniformity performance threshold, the candidate uniformity tilt angle will be eliminated.

[0065] By summing up all the retained candidate uniformity inclination angles, the potential optimal inclination angle pool for the steel content range is obtained;

[0066] Understandably, the uniform performance value is calculated using the uniform maintenance value and the uniform stability value. The uniform maintenance value reflects the proportion of the duration of uniform flow of molten steel under the candidate uniformity angle corresponding to the molten steel content range. The larger the proportion of the duration, the higher the preference of the angle. Similarly, the uniform stability value reflects the stability of the uniform flow of molten steel under the candidate uniformity angle corresponding to the molten steel content range. Although the molten steel indicators meet the uniformity requirements within the uniformity period, stable uniformity ensures that the molten steel not only flows uniformly but also that the degree of uniformity is stable, which is beneficial to the subsequent production and application of molten steel. Therefore, the higher the uniformity stability, the higher the preference of the angle.

[0067] Step 2: Based on the uniformity of molten steel flow, determine the optimal tilt angle for each steel content range in the potential optimal tilt angle pool, and determine the optimal tilt angle maintenance time for multiple effective heats to establish a corresponding table of optimal tilt angle and maintenance time for each steel content range.

[0068] In step two, the optimal tilt angle for each steel content range in the potential optimal tilt angle pool is determined by...

[0069] In the pool of potential optimal tilt angles, the candidate uniformity tilt angle with the largest uniformity performance value is selected as the optimal tilt angle;

[0070] In step two, the process of determining the optimal tilt angle-holding time correspondence table for each steel content range is as follows:

[0071] Based on any range of molten steel content;

[0072] Extract the holding time of all effective heats at the optimal tilt angle, and calculate the average holding time to obtain the optimal tilt angle holding time corresponding to the steel content range;

[0073] By integrating the optimal tilt angle and the duration of maintenance of the optimal tilt angle for different steel content ranges, a corresponding table of optimal tilt angle and maintenance duration for each steel content range is obtained.

[0074] Step 3: Simulate the pouring of molten steel according to the optimal tilt angle-holding time correspondence table for each molten steel content range, and determine the sequence of molten steel landing point positions within the optimal tilt angle holding time for each molten steel content range based on the simulation results;

[0075] In step three, the process of determining the sequence of molten steel landing points within the optimal tilt angle maintenance duration for each molten steel content range is as follows:

[0076] A1. Establish a 1:1 three-dimensional geometric model of the steel furnace, and simulate the flow of molten steel in different steel content ranges according to the optimal tilt angle and optimal tilt angle maintenance time corresponding to different steel content ranges in the optimal tilt angle-maintenance time correspondence table.

[0077] A2. Extract the coordinates of the molten steel landing point per unit time (e.g., 1 second) within the optimal tilt angle maintenance duration for each molten steel content range from the simulation results, and integrate them into a sequence of molten steel landing point positions according to the time sequence.

[0078] For example, the process of simulating molten steel pouring based on the optimal tilt angle-holding time correspondence table for each molten steel content range can be as follows:

[0079] A 1:1 three-dimensional geometric model of the steel furnace is established based on the actual dimensions of the converter, furnace mouth, ladle car, and ladle (such as converter inner diameter, furnace mouth inclination angle, ladle diameter / height, and initial positioning coordinates of the ladle car).

[0080] Based on the actual physical properties of molten steel (such as density, viscosity, surface tension at tapping temperatures of 1500-1600℃, and the heat exchange coefficient between molten steel and furnace lining / air), two core physical fields are defined:

[0081] Fluid dynamics (CFD): Simulates the flow state of molten steel in a converter affected by the tilting angle (such as the liquid level height and outflow velocity of the molten steel in the furnace), as well as its trajectory in the air after outflow (affected by gravity and air resistance).

[0082] Kinematics field: The parameters of optimal tilt angle and optimal tilt angle maintenance time are converted into the tilting motion law of the converter (such as the time for the tilt angle to be uniformly adjusted from the initial value (such as 0°) to the optimal tilt angle, and the angular stability (deviation ≤ 0.1°) during the maintenance of the optimal tilt angle), and synchronously associated with the initial position of the ladle car (fixed value, which serves as the basis for subsequent landing point calculation).

[0083] The converter mouth is the steel outlet boundary, allowing molten steel to flow out only in the tilting direction; the inner wall of the ladle is the steel impact boundary, and the impact rebound coefficient when molten steel falls in needs to be calculated (to avoid ignoring the impact point offset).

[0084] The air environment is a no-slip boundary, used to calculate the slight resistance of the air to the steel flow trajectory (especially significant for long-distance steel tapping scenarios).

[0085] The simulated converter maintains the optimal tilt angle for the full duration (e.g., 80 seconds), and key data is recorded at 1-second / step intervals (to ensure data accuracy):

[0086] Steel flow trajectory coordinates every second: aerial trajectory (x / y / z axis coordinates) from furnace outlet to ladle inlet. The focus is on extracting the landing point coordinates when the steel flow falls into the ladle, that is, the coordinates of the first point where the molten steel contacts the inner wall of the ladle. The landing point coordinates of the molten steel every second within the optimal tilt angle maintenance time are extracted from the simulation results. The filtered time-landing point coordinate data are organized into a sequence of molten steel landing point positions.

[0087] Step 4: Based on the sequence of molten steel drop points, determine the limited target point and the optimal target point for the ladle car, and determine the limited speed and the optimal speed for the ladle car based on the time when the molten steel falls at the limited target point and the optimal target point.

[0088] In step four, the process of determining the limited target point and the optimal target point of the ladle car based on the sequence of molten steel drop positions is as follows:

[0089] In the sequence of molten steel landing point positions, the deviation distance PL between the coordinates of the molten steel landing point and the center position coordinates of the current ladle car is calculated sequentially. The specific calculation formula is as follows:

[0090] , where xi and yi represent the x and y coordinates of the i-th molten steel landing point in the sequence of molten steel landing points, respectively, and x0 and y0 represent the x and y coordinates of the center position of the current ladle car.

[0091] Compare the deviation distance PL with the preset deviation distance;

[0092] If the deviation distance PL is greater than the preset deviation distance, the position of the molten steel landing point will be marked as the optimal moving target point;

[0093] If the deviation distance PL is equal to the preset deviation distance, then mark the position of the molten steel landing point as the limited moving target point;

[0094] If the deviation distance PL is less than the preset deviation distance, continue to calculate and analyze the deviation distance PL between the molten steel landing point coordinates and the center position coordinates of the current ladle car in sequence (analysis of the next molten steel landing point position).

[0095] In step four, the process of determining the limiting and optimal moving speed of the ladle car is as follows:

[0096] Obtain the moment when the molten steel falls at the optimal target point and perform a difference calculation with the current moment to obtain the optimal moving time of the ladle car. Calculate the ratio between the deviation distance corresponding to the optimal target point and the optimal moving time of the ladle car to obtain the optimal moving speed of the ladle car.

[0097] Obtain the time when the molten steel falls at the target point, and perform difference processing with the current time to obtain the limited movement time of the ladle car. Calculate the ratio between the deviation distance corresponding to the target point and the limited movement time of the ladle car to obtain the limited movement speed of the ladle car.

[0098] Understandably, the optimal target point refers to the point where the deviation between the molten steel's landing point and the center of the ladle car exceeds the allowable range. Moving the ladle car to the optimal target point ensures that the molten steel's landing point coincides with the center of the ladle car, preventing localized accumulation of molten steel within the ladle and thus avoiding uneven composition or temperature.

[0099] Therefore, the optimal target point is the ladle car's movement target point. The limited target point refers to the maximum allowable distance between the point where the molten steel falls and the center of the ladle car. To ensure that the molten steel does not accumulate locally in the ladle, the ladle car must at least move to the limited target point. That is, the ladle car's movement speed must not be lower than the limited movement speed. Therefore, by finding the optimal and limited movement speeds of the ladle car through the optimal and limited target points, the quality of the molten steel converter can be guaranteed.

[0100] Once the ladle car has moved to the specified or optimal target point at the specified or optimal speed, the above operation is repeated until the steel converter is finished.

[0101] The technical solution of this invention is as follows: Based on historical steel tapping data, an effective frequency analysis of the furnace tilt angle corresponding to different steel content ranges is performed to determine the candidate uniform tilt angle for each steel content range. Combined with uniformity stability verification, a potential optimal tilt angle pool for each steel content range is determined. Based on the uniformity of molten steel flow, the optimal tilt angle for each steel content range is determined within the potential optimal tilt angle pool. The optimal tilt angle maintenance time for multiple effective heat cycles is determined, and a correspondence table of optimal tilt angle and maintenance time for each steel content range is established. Molten steel pouring is simulated based on this table. The sequence of molten steel landing points within the optimal tilt angle maintenance time for each steel content range is determined based on the simulation results. Based on this sequence, the limited target point and optimal target point for the ladle car are determined. Finally, based on the molten steel landing time at the limited target point and optimal target point, the limited moving speed and optimal moving speed of the ladle car are determined to ensure the quality of the molten steel converter.

[0102] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A one-click steel tapping method for converters based on the fusion of visual perception and big data models, characterized in that: include: Step 1: Based on historical steel tapping data, conduct effective frequency analysis of furnace tilt angles corresponding to different steel content ranges, determine candidate uniform tilt angles for each steel content range, and combine with uniformity stability verification to determine the potential optimal tilt angle pool for each steel content range. Step 2: Based on the uniformity of molten steel flow, determine the optimal tilt angle for each steel content range in the potential optimal tilt angle pool, and determine the optimal tilt angle maintenance time for multiple effective heats to establish a corresponding table of optimal tilt angle and maintenance time for each steel content range. Step 3: Simulate the pouring of molten steel according to the optimal tilt angle-holding time correspondence table for each molten steel content range, and determine the sequence of molten steel landing point positions within the optimal tilt angle holding time for each molten steel content range based on the simulation results; Step 4: Based on the sequence of molten steel drop points, determine the limited target point and the optimal target point for the ladle car, and determine the limited speed and the optimal speed for the ladle car based on the time when the molten steel falls at the limited target point and the optimal target point.

2. The converter one-click tapping method based on the fusion of visual perception and big data model as described in claim 1, characterized in that: The process of analyzing the effective frequency of furnace tilt angles corresponding to different steel content ranges is as follows: Based on any range of molten steel content; Extract the tilt angle time sequence data of all valid heats, count the frequency of each tilt angle throughout the entire steel tapping process, and use the tilt angles with frequencies higher than the preset frequency as candidate uniformity tilt angles for the steel content range.

3. The converter one-click tapping method based on the fusion of visual perception and big data model as described in claim 2, characterized in that: The process for determining the potential optimal tilt angle pool for each steel content range is as follows: Based on any candidate uniformity tilt angle within the steel content range; The uniformity index of molten steel flow at different time points under the candidate uniformity inclination angle was obtained, and the uniformity maintenance value and uniform stability value were obtained by processing and analysis. The deviation between the uniform maintenance value and the uniform stable value is calculated to obtain the uniform performance value. By summing up the candidate uniformity tilt angles with uniformity performance values ​​greater than or equal to the uniformity performance threshold, the potential optimal tilt angle pool for the steel content range is obtained.

4. The converter one-click tapping method based on the fusion of visual perception and big data model as described in claim 3, characterized in that: The method for obtaining the uniformity maintenance value is as follows: The time period in which the uniformity index of molten steel flow meets the normal range (the fluctuation range of the cross-sectional diameter of the molten steel flow is less than or equal to 5%) is marked as the uniform time period. The ratio between the uniform time period and the total tapping time is calculated to obtain the uniformity maintenance value.

5. The converter one-click tapping method based on the fusion of visual perception and big data model as described in claim 3, characterized in that: The method for obtaining the uniform stable value is as follows: The uniformity index of molten steel flow within a uniform time period is summarized into a molten steel flow uniformity index sequence according to the time sequence. The coefficient of variation of the molten steel flow uniformity index sequence is calculated to obtain the uniform stable value.

6. The converter one-click tapping method based on the fusion of visual perception and big data model as described in claim 5, characterized in that: The process of determining the optimal tilt angle-maintenance duration correspondence table for each steel content range is as follows: Based on any range of molten steel content; Extract the holding time of all effective heats at the optimal tilt angle, and calculate the average holding time to obtain the optimal tilt angle holding time corresponding to the steel content range; The optimal tilt angle and the duration of maintenance of the optimal tilt angle corresponding to different steel content ranges are integrated to obtain the corresponding table of optimal tilt angle and duration of maintenance for each steel content range.

7. The converter one-click tapping method based on the fusion of visual perception and big data model as described in claim 6, characterized in that: The process of determining the sequence of molten steel landing point positions for each molten steel content range within the optimal tilt angle maintenance duration is as follows: A three-dimensional geometric model of the steel furnace was established, and the flow of molten steel in different steel content ranges was simulated in segments according to the optimal tilt angle and optimal tilt angle maintenance time corresponding to different steel content ranges in the optimal tilt angle-maintenance time correspondence table. The coordinates of the molten steel landing point per unit time within the optimal tilt angle maintenance duration for each molten steel content range are extracted from the simulation results and integrated into a sequence of molten steel landing point positions according to the time sequence.

8. The converter one-click tapping method based on the fusion of visual perception and big data model as described in claim 7, characterized in that: The process of determining the constrained target point and the optimal target point for the ladle car is as follows: In the sequence of molten steel drop point positions, the deviation distance between the coordinates of the molten steel drop point and the center position coordinates of the current ladle car is calculated in order; If the deviation distance is greater than the preset deviation distance, the position of the molten steel landing point will be marked as the optimal moving target point; If the deviation distance is equal to the preset deviation distance, then mark the location of the molten steel drop point as the limited moving target point; If the deviation distance is less than the preset deviation distance, the analysis of the next molten steel landing point will continue.

9. The converter one-click tapping method based on the fusion of visual perception and big data model as described in claim 8, characterized in that: The process of determining the limited moving speed of the ladle car is as follows: Obtain the time when the molten steel falls at the target point and perform a difference operation with the current time to obtain the limited movement time of the ladle car. Calculate the ratio between the deviation distance corresponding to the target point and the limited movement time of the ladle car to obtain the limited movement speed of the ladle car.

10. The converter one-click tapping method based on the fusion of visual perception and big data model as described in claim 8, characterized in that: The process for determining the optimal moving speed of the ladle car is as follows: Obtain the moment when the molten steel falls at the optimal target point and perform a difference calculation with the current moment to obtain the optimal travel time of the ladle car. Calculate the ratio between the deviation distance corresponding to the optimal target point and the optimal travel time of the ladle car to obtain the optimal travel speed of the ladle car.

Images