A method and apparatus for controlling solidification of high carbon steel round billets at high pulling speed

Abstract

This application relates to the technical field of continuous casting, and in particular to a method and apparatus for solidification control of high-carbon steel round billets produced at high casting speeds. Addressing defects such as central carbon segregation, shrinkage cavities, and porosity that are prone to occur in high-carbon steel during high-speed continuous casting, this application establishes a V-Q-P three-dimensional lookup table to monitor key parameters such as casting speed, superheat, billet shell thickness, and surface temperature difference in real time. It dynamically adjusts the PMO pulse voltage, secondary cooling water volume, and straightener reduction to achieve precise control of the solidification process. Simultaneously, a defect risk index R is introduced, and its value is used to determine whether the straightener should be pressed down, further optimizing the internal quality of the billet. This method significantly suppresses the central carbon segregation index (≤1.10) and shrinkage cavities (rating ≤0.5), increases the proportion of equiaxed crystal regions (≥50%), and effectively improves the uniformity and stability of the internal structure of the high-carbon steel round billet. Through a closed-loop feedback mechanism, process parameters are continuously optimized to ensure stable and high-quality production of high-carbon steel under high casting speed conditions.

Description

Technical Field

[0001] This application relates to the technical field of continuous casting, and in particular to a solidification control method and apparatus for high-speed (≥1.3 m / min) high-carbon steel (C≥0.65%) round billets (Φ200–350 mm). Background Technology

[0002] High-carbon steel (such as SWRH82B, GCr15, 10B21, etc.) round billets are prone to defects such as central carbon segregation, shrinkage cavities, and central porosity during high-speed continuous casting (≥1.3 m / min) due to the wide solidification zone and long liquidus cavity. Existing technologies typically employ a single electromagnetic stirrer or a single straightener reduction, which is insufficient to simultaneously address segregation and shrinkage cavities. While related technologies propose PMO to refine the solidification structure, their process windows are general-purpose and lack a closed-loop control strategy for high-speed casting of high-carbon steel. In actual use, the segregation index fluctuates greatly, and shrinkage cavities still occur frequently. Therefore, there is an urgent need for a device and method that can coordinately control PMO energy, secondary cooling strength, and straightener reduction online to stably obtain high-carbon steel round billets with low segregation and no shrinkage cavities under high casting speed conditions. Summary of the Invention

[0003] In order to achieve simultaneous suppression of segregation and shrinkage cavities under closed-loop synergistic conditions of PMO-secondary cooling-straightening machine pressure, this application provides a solidification control method and device for high-speed high-carbon steel round billets.

[0004] In a first aspect, this application provides a method for solidification control of high-speed high-carbon steel round billets, employing the following technical solution: A method for solidification control of high-speed high-carbon steel round billets includes the following steps: S1. Establish a three-dimensional lookup table for VQP, where V is the PMO voltage, Q is the secondary cooling water flow rate, and P is the pressure reduction; S2. Real-time acquisition of v, ΔT, h and Δθ, where v is the casting speed, ΔT is the tundish superheat, h is the billet shell thickness and Δθ is the surface temperature difference of the billet. Based on the first judgment condition, the value of V is changed and the value of Q is changed synchronously with reference to the VQP three-dimensional lookup table. S3. Change the PMO pulse frequency according to the second judgment condition; S4. Establish the formula for calculating the defect risk index R, calculate the defect risk index R, determine the defect risk index R, and judge whether the straightening machine should be pressed down based on the risk index R; S5. After the billet is removed from the production line, 25-point carbon segregation detection is performed, and the data is fed back to update the VQP three-dimensional lookup table; S6. Formula for calculating the periodic update defect risk index R.

[0005] By adopting the above technical solutions, dynamic matching of PMO, secondary cooling strength and straightening pressure was achieved under high casting speed conditions, effectively shortening the liquid phase cavity length, promoting equiaxed crystal formation, and significantly suppressing central carbon segregation and shrinkage defects. Through a closed-loop feedback mechanism, process parameters were continuously optimized, keeping the segregation index stably controlled within 1.10 and the shrinkage rating ≤0.5. At the same time, the consistency and stability of the internal quality of the billet were improved, providing reliable technical support for the industrialized and stable production of high carbon steel under high-speed continuous casting conditions.

[0006] Optionally, in S1, the horizontal axis of the VQP three-dimensional lookup table is Q, the vertical axis is v, and the cells contain recommended values ​​for V and P.

[0007] By adopting the above technical solution, the PMO voltage, secondary cooling water volume and pulling speed are coupled and correlated to achieve precise matching of process parameters, improve control accuracy and response speed; and the reduction amount is dynamically adjusted in combination with the solidification characteristics of high carbon steel.

[0008] Optionally, the first determination condition in S2 is that when ΔT is higher than the preset threshold for tundish superheat or when h is lower than the preset threshold for billet shell thickness, the change method is to increase V and simultaneously increase Q by referring to the VQP three-dimensional lookup table.

[0009] By adopting the above technical solutions, when the overheating or billet thickness is abnormal, the electromagnetic stirring intensity and secondary cooling intensity can be increased in time to strengthen the formation of the primary billet shell and inhibit the excessive growth of columnar crystals; combined with dynamic adjustment of the cooling ratio by the pulling speed, the risk of cracking is avoided due to excessive surface reheat.

[0010] Optionally, the second determination condition in S3 is that when Δθ is greater than the preset threshold of the surface temperature difference of the billet, the change method is to trigger a high-frequency pulse mode.

[0011] By adopting the above technical solution, when the temperature difference on the surface of the billet is too large, high-frequency pulse electromagnetic stirring is started, which effectively breaks the dendritic structure in the front liquid core, promotes uniform flow of liquid phase, reduces central segregation, and expands the equiaxed crystal region; at the same time, magnetostrictive oscillation is generated on the surface of the molten metal, which acts on the metal solidification process until solidification, improves the metal solidification structure and refines it.

[0012] Optionally, the formula for calculating the defect risk index R in S4 is as follows: R = Weighting coefficient 1 × (superheat parameter) + Weighting coefficient 2 × (pulling speed parameter) + Weighting coefficient 3 × (carbon segregation parameter) in The superheat parameter is (ΔT-20) / 20. The pulling speed parameter is (v-1.3) / 0.2. The carbon segregation parameter is (1.1 - C_index) / 0.15. C_index is the ratio of the maximum to the average carbon segregation value at 25 points of the cast billet that has been completed. The sum of weight coefficient 1, weight coefficient 2, and weight coefficient 3 is 1. The initial weight coefficient 1 is 0.4, the initial weight coefficient 2 is 0.3, and the initial weight coefficient 3 is 0.3.

[0013] By adopting the above technical solution, a formula for calculating the defect risk index R is constructed to achieve real-time quantitative monitoring of the quality trend of the high carbon steel continuous casting process; when the R value exceeds the preset warning threshold, the process parameter control mechanism is automatically triggered.

[0014] Optionally, the straightener is activated to press down when R>0.3.

[0015] By adopting the above technical solution, the straightening machine starts pressing when R>0.3, and combines the dynamic light pressing strategy to compensate for solidification shrinkage and reduce central porosity and segregation; the pressing amount is adjusted according to the VQP three-dimensional lookup table to achieve coordinated matching of pressing amount, casting speed and cooling intensity, and ensure that the billet is in the best compaction state at the end of solidification.

[0016] Optionally, the update method of the VQP three-dimensional lookup table in S5 is to adjust it through the VQP model. The VQP model is a pre-trained deep learning model. The VQP model is pre-trained in the following way: the values ​​of V, Q, and P in the VQP three-dimensional lookup table are changed proportionally; based on different VQP three-dimensional lookup tables, a C_index sample set for model training is obtained; the training sample set is input into the preset VQP model for training to obtain the VQP model.

[0017] By adopting the above technical solution, the VQP model is iteratively optimized using historical process parameters and measured carbon segregation data, thereby improving the accuracy and adaptability of the lookup table update. The three-dimensional lookup table is constructed based on a large amount of production data and simulation results, and is continuously iteratively optimized with actual quality feedback to ensure that the optimal parameter combination can be output under different working conditions, thereby enhancing the system's adaptability and providing a core guarantee for the stability of the internal quality of the billet under high casting speed.

[0018] Optionally, the method for updating the calculation formula of the defect risk index R in S6 is to adjust it through a weight coefficient change model. The weight coefficient change model is a pre-trained deep learning model. The weight coefficient change model is pre-trained in the following way: periodically changing weight coefficient 1, weight coefficient 2 and weight coefficient 3 to obtain a C_index sample set for model training; inputting the training sample set into the preset weight coefficient change model for model training to obtain the weight coefficient change model.

[0019] By adopting the above technical solutions, the weight allocation is dynamically optimized using historical quality data, thereby improving the accuracy of the defect risk index R in representing the actual degree of segregation. The weight coefficient change model is continuously iterated in combination with real-time process fluctuations and final acceptance composition data, enhancing the adaptability and predictive reliability of the R value under different steel grades, casting speeds and cooling regimes, and ensuring the precise triggering of control strategies and process stability.

[0020] Secondly, this application provides a solidification control device for high-speed high-carbon steel round billets, employing the following technical solution: A solidification control device for high-speed high-carbon steel round billets includes a PMO pulse coil, which is installed at 0.8–1.2m in the secondary cooling zone. The PMO pulse coil outputs a pulse voltage range of 200–240V, a frequency range of 2–4Hz, and a pulse width range of 120–140μs. The cooling mechanism includes multiple nozzles located in section 3 of the secondary cooling zone. The nozzles are arranged circumferentially along section 3 of the secondary cooling zone. The number of nozzles is at least 6. The water pressure range of the cooling mechanism is 0.5-0.8 MPa. The flow rate of the cooling mechanism is continuously adjustable. A tension speed sensor is used to detect the tension speed v in real time. Superheat detector, used to detect the superheat ΔT of the intermediate ladle; The billet shell thickness calculation module calculates the billet shell thickness h in real time based on v, ΔT, and the current liquidus temperature of the steel grade. The pulse power controller outputs the PMO voltage V online based on h, ΔT, and v. The secondary cooling flow controller is used to control and adjust the secondary cooling water flow rate Q of the cooling mechanism; The pressure roller controller is used to adjust the pressure amount P of the end straightener of the round billet continuous casting machine and to control the pressure of the end straightener; The processor is simultaneously connected to the drawing speed sensor, the superheat detector, the billet shell thickness calculation module, the pulse power controller, the secondary cooling flow controller, and the pressing roll controller. The processor is used to execute the solidification control method for high-drawing-speed high-carbon steel round billets described above.

[0021] By adopting the above technical solutions, dynamic and coordinated control of the solidification process of round billets under high casting speed is achieved, effectively suppressing center segregation and porosity defects; the PMO pulsed magnetic field improves the flow of molten steel on the inner arc side and promotes the formation of equiaxed crystals. Combined with a closed-loop control strategy based on real-time process feedback, the intensity of secondary cooling and the amount of final reduction are precisely adjusted to improve the internal quality uniformity of the billet; the system forms a closed-loop control through data and models from multiple sensors to adapt to changes in different production conditions and ensure stable and high-quality production of high-carbon steel at high casting speed.

[0022] In summary, this application includes at least one of the following beneficial technical effects: 1. By establishing a VQP three-dimensional lookup table and changing the PMO voltage and secondary cooling water volume based on real-time data acquisition, the PMO energy and secondary cooling intensity can be controlled online in a coordinated manner, which helps to stably obtain high carbon steel round billets with a central carbon segregation index ≤1.10, a shrinkage cavity rating ≤0.5, and an equiaxed crystal region area ratio ≥50% under high drawing speed conditions. 2. By changing the PMO pulse frequency according to the temperature difference on the surface of the billet, a high-frequency pulse mode can be triggered, which can further optimize the solidification process and reduce defects such as central carbon segregation, shrinkage cavities and central porosity. 3. Establish a formula for calculating the defect risk index R and use it to determine whether the straightening machine should be pressed down. Combine the carbon segregation detection data after the billet is removed from the production line to update the VQP three-dimensional lookup table and periodically update the formula for calculating the defect risk index R. This can achieve closed-loop control, effectively reduce the fluctuation of the segregation index, and reduce the occurrence of shrinkage defects. Attached Figure Description

[0023] Figure 1 This is a logic block diagram of the solidification control method for high-speed high-carbon steel round billets provided in the embodiments of this application.

[0024] Figure 2 This is a low-magnification photograph of the round blank corresponding to experimental parameter 1 provided in the embodiments of this application. Detailed Implementation

[0025] The following is in conjunction with the appendix Figure 1 This application will be described in further detail.

[0026] This application discloses a solidification control method for high-speed high-carbon steel round billets, used to obtain carbon steel round billets with a carbon content of not less than 0.65% and a central carbon segregation index ≤1.10, a shrinkage cavity rating ≤0.5, and an equiaxed crystal region area ratio ≥50% at a drawing speed ≥1.3m / min. The method is characterized by the following steps: S1. Establish a VQP three-dimensional lookup table, where V is the PMO voltage, Q is the secondary cooling water flow rate, and P is the reduction. The horizontal axis of the VQP three-dimensional lookup table is Q, the vertical axis is v, and the cells contain recommended values ​​for V and P. Couple the PMO voltage, secondary cooling water flow rate, and casting speed to achieve precise matching of process parameters, improving control accuracy and response speed; dynamically adjust the reduction based on the solidification characteristics of high-carbon steel.

[0027] S2. Real-time acquisition of v, ΔT, h, and Δθ, where v is the casting speed, ΔT is the tundish superheat, h is the billet shell thickness, and Δθ is the surface temperature difference of the billet. Based on the first judgment condition, the value of V is changed with reference to the VQP three-dimensional lookup table, and the value of Q is changed simultaneously. In actual production, when the billet shell thickness h is less than 25mm or the tundish superheat ΔT is greater than 25℃, the PMO voltage V can be increased to 235V, and the secondary cooling water volume Q can be increased simultaneously. When the superheat or billet shell thickness is abnormal, the electromagnetic stirring intensity and secondary cooling intensity are enhanced to promote the formation of the primary billet shell and inhibit the excessive growth of columnar crystals; at the same time, the cooling ratio is dynamically adjusted in conjunction with the casting speed to avoid excessive surface reheat and increased risk of cracking.

[0028] S3. According to the second judgment condition, change the PMO pulse frequency. In actual production, when Δθ is greater than 15℃, start the high-frequency pulse mode, adjust the PMO pulse frequency to 4Hz, and keep it for 30s. When the temperature difference on the surface of the billet is too large, applying high-frequency pulse electromagnetic stirring can effectively break the dendritic structure in the front liquid core, promote uniform flow of the liquid phase, significantly reduce central segregation, and expand the equiaxed crystal region; at the same time, this technology can excite magnetostrictive oscillation on the surface of the molten metal, which will continue to act on the entire solidification process, significantly improving and refining the metal solidification structure.

[0029] S4. Establish the formula for calculating the defect risk index R, calculate the defect risk index R, determine the defect risk index R, and start the straightening machine to press down when R>0.3.

[0030] The formula for calculating the defect risk index R is as follows: R = Weighting coefficient 1 × (superheat parameter) + Weighting coefficient 2 × (pulling speed parameter) + Weighting coefficient 3 × (carbon segregation parameter) in The superheat parameter is (ΔT-20) / 20. The pulling speed parameter is (v-1.3) / 0.2. The carbon segregation parameter is (1.1 - C_index) / 0.15. C_index is the ratio of the maximum to the average carbon segregation value at 25 points of the cast billet that has been completed. The sum of weight coefficient 1, weight coefficient 2, and weight coefficient 3 is 1. The initial weight coefficient 1 is 0.4, the initial weight coefficient 2 is 0.3, and the initial weight coefficient 3 is 0.3.

[0031] A formula for calculating the defect risk index R in the continuous casting process of high carbon steel is established to achieve real-time quantitative monitoring of quality trends; when the R value exceeds the preset warning threshold, the system immediately presses down the straightening machine.

[0032] S5. After the billet is removed from the production line, a 25-point carbon segregation test is performed, and the data is fed back to update the VQP 3D lookup table. The update method is to adjust the VQP model, which is a pre-trained deep learning model. The VQP model is pre-trained in the following way: the values ​​of V, Q, and P in the VQP 3D lookup table are changed proportionally, and a C_index sample set for model training is obtained based on different VQP 3D lookup tables. The training sample set is then input into the preset VQP model for training to obtain the VQP model.

[0033] Iterative optimization of the VQP model is carried out based on historical process parameters and measured carbon segregation data to improve its update accuracy and adaptability. The three-dimensional lookup table is constructed based on a large amount of production data and simulation results, and is continuously and dynamically updated with actual quality feedback to ensure that the optimal parameter combination can be output under different working conditions, significantly enhancing the system's adaptability and providing a solid guarantee for the stability of the internal quality of the billet under high casting speed conditions.

[0034] S6. The formula for periodically updating the defect risk index R; the update method is to adjust it through the weight coefficient change model, which is a pre-trained deep learning model. The weight coefficient change model is pre-trained in the following way: periodically change weight coefficient 1, weight coefficient 2 and weight coefficient 3 to obtain the C_index sample set for model training; input the training sample set into the preset weight coefficient change model to train the model and obtain the weight coefficient change model.

[0035] By dynamically optimizing the weight allocation strategy using historical quality data, the accuracy of the defect risk index R in representing the actual degree of segregation is improved. Through continuous iteration using a weight coefficient change model combined with real-time process fluctuations and final acceptance composition data, the adaptability and predictive reliability of the R value under different steel grades, casting speeds, and cooling regimes are improved, ensuring the precise triggering capability of the control strategy and the stability of the process.

[0036] The following specific experimental parameters were collected during the actual production process in this application: Experimental parameter 1 SWRH82B (C=0.82%, Mn=0.82%) was produced on a Φ250mm round billet continuous casting machine at a casting speed of 1.35m / min and a superheat of 22℃.

[0037] –PMO coil 220V / 3Hz / 130μs; – Secondary cooling section 3 water flow rate: 1.05L / kg; – The gap between the rollers pressed down by the straightening machine is 0.6mm; Results: equiaxed crystal region 175 mm, central carbon segregation index 1.08, shrinkage pore grade 0, central porosity grade 1.0.

[0038] Experimental parameter 2 GCr15 (C=1.0%), Φ300mm, stretching speed 1.30m / min, superheat 25℃.

[0039] –PMO230V / 4Hz / 140μs; – Secondary cooling: 1.10L / kg; – The straightening machine pressed down 0.7mm; Results: equiaxed crystal region 170 mm, central carbon segregation index 1.05, shrinkage pore grade 0.

[0040] Experimental parameter 3 10B21 (C=0.80%), Φ200mm, pulling speed 1.45m / min, superheat 18℃, using an adaptive closed-loop model.

[0041] Results: Central carbon segregation index 1.02, shrinkage cavity grade 0, drawing speed increased by 12%.

[0042] This application also discloses a solidification control device for high-speed high-carbon steel round billets.

[0043] The solidification control device for high-speed high-carbon steel round billets includes a PMO pulse coil, which is installed at 0.8–1.2m in the secondary cooling zone. The PMO pulse coil outputs a pulse voltage range of 200–240V, a frequency range of 2–4Hz, and a pulse width range of 120–140μs. The cooling mechanism includes multiple nozzles located in section 3 of the secondary cooling zone. The nozzles are arranged circumferentially along section 3 of the secondary cooling zone. The number of nozzles is at least 6. The water pressure range of the cooling mechanism is 0.5-0.8 MPa. The flow rate of the cooling mechanism is continuously adjustable. A tension speed sensor is used to detect the tension speed v in real time. Superheat detector, used to detect the superheat ΔT of the intermediate ladle; The billet shell thickness calculation module calculates the billet shell thickness h in real time based on v, ΔT, and the current liquidus temperature of the steel grade. The pulse power controller outputs the PMO voltage V online based on h, ΔT, and v. The secondary cooling flow controller is used to control and adjust the secondary cooling water flow rate Q of the cooling mechanism; The pressure roller controller is used to adjust the pressure amount P of the end straightener of the round billet continuous casting machine and to control the pressure of the end straightener; The processor is simultaneously connected to the drawing speed sensor, the superheat detector, the billet shell thickness calculation module, the pulse power controller, the secondary cooling flow controller, and the pressing roll controller. The processor is used to execute the solidification control method for high-drawing-speed high-carbon steel round billets described above.

[0044] The implementation principle of the solidification control device for high-speed high-carbon steel round billets in this application embodiment is as follows: By real-time monitoring of casting speed and superheat, and dynamic calculation of billet shell thickness based on the liquidus temperature of the steel grade, dynamic and coordinated control of the solidification process of round billets under high casting speed conditions is successfully achieved, thereby effectively suppressing center segregation and porosity defects; the PMO pulsed magnetic field is used to significantly improve the flow of molten steel on the inner arc side, promote the formation of equiaxed crystals, and combined with a closed-loop control strategy based on real-time process feedback, the secondary cooling intensity and end reduction are precisely controlled, significantly improving the internal quality uniformity of the billet; the system forms a closed-loop control by integrating data from multiple sensors and models, flexibly adapting to changes in different production conditions, and effectively ensuring the stable and high-quality production of high-carbon steel under high casting speed.

[0045] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A solidification control method for high-speed high-carbon steel round billets, used to obtain carbon steel round billets with a carbon content of not less than 0.65% and a central carbon segregation index ≤1.10, a shrinkage cavity rating ≤0.5, and an equiaxed crystal region area ratio ≥50% at a drawing speed ≥1.3m / min, characterized in that, Includes the following steps: S1. Establish a three-dimensional lookup table for VQP, where V is the PMO voltage, Q is the secondary cooling water flow rate, and P is the pressure reduction; S2. Real-time acquisition of v, ΔT, h and Δθ, where v is the casting speed, ΔT is the tundish superheat, h is the billet shell thickness and Δθ is the surface temperature difference of the billet. Based on the first judgment condition, the value of V is changed and the value of Q is changed synchronously with reference to the VQP three-dimensional lookup table. S3. Change the PMO pulse frequency according to the second judgment condition; S4. Establish the formula for calculating the defect risk index R, calculate the defect risk index R, determine the defect risk index R, and judge whether the straightening machine should be pressed down based on the risk index R; S5. After the billet is removed from the production line, 25-point carbon segregation detection is performed, and the data is fed back to update the VQP three-dimensional lookup table; S6. Formula for calculating the periodic update defect risk index R.

2. The solidification control method for high-speed high-carbon steel round billets according to claim 1, characterized in that, In S1, the VQP three-dimensional lookup table has Q as the horizontal axis and v as the vertical axis, with the recommended values ​​for V and P in each cell.

3. The solidification control method for high-speed high-carbon steel round billets according to claim 2, characterized in that, The first determination condition in S2 is that when ΔT is higher than the preset threshold for tundish superheat or when h is lower than the preset threshold for billet shell thickness, the change method is to increase V and simultaneously increase Q by referring to the VQP three-dimensional lookup table.

4. The solidification control method for high-speed high-carbon steel round billets according to claim 2, characterized in that, The second determination condition in S3 is that when Δθ is greater than the preset threshold of the surface temperature difference of the billet, the change method is to trigger a high-frequency pulse mode.

5. The solidification control method for high-speed high-carbon steel round billets according to claim 2, characterized in that, The formula for calculating the defect risk index R in S4 is as follows: R = Weighting coefficient 1 × (superheat parameter) + Weighting coefficient 2 × (pulling speed parameter) + Weighting coefficient 3 × (carbon segregation parameter) in The superheat parameter is (ΔT-20) / 20. The pulling speed parameter is (v-1.3) / 0.

2. The carbon segregation parameter is (1.1 - C_index) / 0.

15. C_index is the ratio of the maximum to the average carbon segregation value at 25 points of the cast billet that has been completed. The sum of weight coefficient 1, weight coefficient 2, and weight coefficient 3 is 1. The initial weight coefficient 1 is 0.4, the initial weight coefficient 2 is 0.3, and the initial weight coefficient 3 is 0.

3.

6. The solidification control method for high-speed high-carbon steel round billets according to claim 5, characterized in that, When R>0.3, the straightening machine starts pressing down.

7. The solidification control method for high-speed high-carbon steel round billets according to claim 5, characterized in that, The update method of the VQP three-dimensional lookup table in S5 is to adjust it through the VQP model. The VQP model is a pre-trained deep learning model. The VQP model is pre-trained in the following way: the values ​​of V, Q and P in the VQP three-dimensional lookup table are changed proportionally, and a C_index sample set for model training is obtained based on different VQP three-dimensional lookup tables. The training sample set is input into a preset VQP model for training to obtain the VQP model.

8. The solidification control method for high-speed high-carbon steel round billets according to claim 5, characterized in that, The method for updating the calculation formula of the defect risk index R in S6 is to adjust it through a weight coefficient change model. The weight coefficient change model is a pre-trained deep learning model. The weight coefficient change model is pre-trained in the following way: periodically changing weight coefficient 1, weight coefficient 2 and weight coefficient 3 to obtain a C_index sample set for model training; inputting the training sample set into the preset weight coefficient change model for model training to obtain the weight coefficient change model.

9. A solidification control device for high-speed high-carbon steel round billets, used in continuous casting machines suitable for Φ200–350mm round billets, characterized in that, include: The PMO pulse coil is installed at a distance of 0.8–1.2m in the second cooling zone. The PMO pulse coil outputs a pulse voltage range of 200–240V, a frequency range of 2–4Hz, and a pulse width range of 120–140μs. The cooling mechanism includes multiple nozzles located in section 3 of the secondary cooling zone. The nozzles are arranged circumferentially along section 3 of the secondary cooling zone. The number of nozzles is at least 6. The water pressure range of the cooling mechanism is 0.5-0.8 MPa. The flow rate of the cooling mechanism is continuously adjustable. A tension speed sensor is used to detect the tension speed v in real time. Superheat detector, used to detect the superheat ΔT of the intermediate ladle; The billet shell thickness calculation module calculates the billet shell thickness h in real time based on v, ΔT, and the current liquidus temperature of the steel grade. The pulse power controller outputs the PMO voltage V online based on h, ΔT, and v. The secondary cooling flow controller is used to control and adjust the secondary cooling water flow rate Q of the cooling mechanism; The pressure roller controller is used to adjust the pressure amount P of the end straightener of the round billet continuous casting machine and to control the pressure of the end straightener; The processor is simultaneously connected to the drawing speed sensor, the superheat detector, the billet shell thickness calculation module, the pulse power controller, the secondary cooling flow controller, and the pressing roll controller. The processor is used to execute the solidification control method for high-speed high-carbon steel round billets according to any one of claims 1-8.

Images