Efficient and energy-saving intelligent control method for slab continuous casting

Abstract

The invention belongs to the technical field of metallurgy, and particularly relates to an efficient and energy-saving intelligent control method for slab continuous casting. Comprising the following steps that S1, secondary cooling control is conducted, specifically, casting blank surface temperature field information is collected in real time, and partition cooling control is conducted on a casting blank based on a dynamic water distribution model; s2, vibration control is conducted, specifically, according to the real-time pulling speed, a crystallizer is controlled to conduct non-sinusoidal vibration through an online vibration parameter adjusting model; s3, multi-mode electromagnetic field control, wherein electromagnetic fields of different modes are applied to the crystallizer area, the secondary cooling area and the tail end area correspondingly; s4, multi-parameter breakout early warning is conducted, specifically, real-time diagnosis and early warning are conducted on breakout based on the multi-source signal fusion and early warning model; and S5, dynamically matching the performance of the casting powder: dynamically matching the viscosity of the casting powder according to the real-time pulling speed. According to the method, a fixed water distribution model in the prior art is corrected through the dynamic water distribution model, and the secondary cooling area is independently controlled in a partitioned mode, so that the cooling effect and uniformity are improved, and the temperature difference is reduced.

Description

Technical Field

[0001] This invention belongs to the field of metallurgical technology, and in particular relates to an intelligent control method for high-efficiency and energy-saving slab continuous casting. Background Technology

[0002] Slab continuous casting is a critical process in steel production, and its control level directly affects slab quality, production efficiency, and energy consumption. Traditional continuous casting control methods have the following technical problems:

[0003] 1. The secondary cooling control uses a fixed water distribution model, which cannot be adjusted in real time according to the surface temperature of the billet. This results in uneven cooling of the wide and narrow sides of the billet, with a temperature difference of more than 60°C, which easily leads to surface cracks.

[0004] 2. The vibration waveform of the crystallizer vibration system is severely distorted, with the skewness fluctuating by more than 15%, which affects the surface quality of the cast billet.

[0005] 3. The electromagnetic stirring has a dead zone, the equiaxed crystal ratio is less than 35% at 1 / 4 thickness of the billet, and the center segregation is serious.

[0006] 4. Relying on a single temperature signal for steel leakage prediction results in a false alarm rate as high as 25%, affecting production stability.

[0007] 5. The consumption of protective slag is greater than 0.4 kg / t steel, the lubrication effect is unstable, and the surface quality of the cast billet is affected.

[0008] The aforementioned problems result in a high surface defect rate in cast billets, a low proportion of equiaxed crystals, high energy consumption, and low production efficiency, necessitating an intelligent control method capable of achieving multi-parameter collaborative optimization. Summary of the Invention

[0009] The purpose of this invention is to provide an intelligent control method for high-efficiency and energy-saving slab continuous casting, so as to solve the problems existing in the prior art.

[0010] The technical solution adopted by this invention to solve its technical problem is:

[0011] An intelligent control method for high-efficiency and energy-saving slab continuous casting includes the following steps:

[0012] S1, Secondary Cooling Control: Real-time acquisition of surface temperature field information of the billet, and zoned cooling control of the billet based on dynamic water distribution model;

[0013] S2. Vibration control: Based on the real-time pulling speed, the crystallizer is controlled to perform non-sinusoidal vibration through an online vibration parameter adjustment model;

[0014] S3, Multi-mode electromagnetic field control: Different modes of electromagnetic fields are applied in the crystallizer region, the secondary cooling region and the end region respectively;

[0015] S4. Multi-parameter steel leakage early warning: Real-time diagnosis and early warning of steel leakage based on multi-source signal fusion and early warning model;

[0016] S5. Dynamic matching of protective slag performance: Dynamically match the viscosity of the protective slag according to the real-time casting speed.

[0017] Furthermore, in step S1, the zoned cooling control involves dividing the secondary cooling zone of the billet into at least 16 independent control zones on the wide side and at least 6 independent control zones on the narrow side.

[0018] The dynamic water distribution model is: Q i =K×v 1.2 ×ΔT i 0.8 ;

[0019] Among them, Q i Here, K is the adjustment value for the spray water volume in zone i, K is the steel grade coefficient, v is the real-time casting speed, and ΔT is the value for the water volume adjustment in zone i. i It is the difference between the actual temperature and the target temperature of the billet surface in zone i.

[0020] Furthermore, in step S2, the waveform skewness of the non-sinusoidal vibration is adjustable within the range of 10% to 35%, and the online frequency adjustment model includes:

[0021] Vibration frequency adjustment formula: f=28 / (1+1.5e) -0.4v ), where f is the vibration frequency and v is the real-time pulling speed;

[0022] Amplitude adjustment formula: A = 5.5 - 0.8v + 0.12v 2 Where A is the amplitude and v is the real-time pulling speed.

[0023] Furthermore, in step S3, the multi-mode electromagnetic field control is as follows: a rotating magnetic field is applied in the crystallizer region, a static magnetic field and a pulsed magnetic field are applied simultaneously in the secondary cooling region, and electromagnetic braking is applied in the end region.

[0024] Furthermore, the magnetic field strength in the multi-mode electromagnetic field control satisfies the gradient control strategy: crystallizer zone > secondary cooling zone > end zone; the end zone adopts an end electromagnetic brake with a gradient intensity distribution, the magnetic field strength is relatively weak in the central region of the billet width, and gradually increases in the two sides near the narrow face.

[0025] Furthermore, in step S3, the frequency of the rotating magnetic field in the crystallizer zone is adjusted within the range of 0~20Hz; the static magnetic field strength in the secondary cooling zone is 0.5T, and the pulse magnetic field frequency is 2Hz.

[0026] Furthermore, in step S4, the multi-source signals include, but are not limited to, the temperature, vibration, friction, and acoustic emission signals of the crystallizer; the early warning model is a convolutional neural network model.

[0027] The diagnosis and early warning system involves real-time acquisition and fusion of temperature, vibration, friction, and acoustic emission signals from the crystallizer. The fused multi-source signals are then input into the early warning model, which outputs the probability of steel leakage. Based on the probability level, a three-level response is executed: early warning, speed reduction, or shutdown.

[0028] Furthermore, in step S5, the dynamic matching of the protective slag viscosity is calculated based on the real-time casting speed to determine the target value of the protective slag viscosity, and the amount or composition of the protective slag added is adjusted accordingly.

[0029] The formula for calculating the target viscosity value of the protective slag is: η = 0.25 + 0.1e -0.6v Where η is the viscosity of the protective slag and v is the real-time casting speed.

[0030] The present invention has the following beneficial effects:

[0031] 1. This invention modifies the existing fixed water distribution model by using a dynamic water distribution model and independently controls the secondary cooling zones, thereby improving the cooling effect and uniformity and reducing temperature differences.

[0032] 2. The vibration of the crystallizer is dynamically adjusted by using an online vibration parameter adjustment model, so that the servo hydraulic drive system matches the pulling speed and ensures vibration accuracy.

[0033] 3. The viscosity of the protective slag matching the current working conditions is obtained by calculating the target value of the protective slag viscosity, providing a basis for the addition of the protective slag. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0035] Example 1:

[0036] An intelligent control method for high-efficiency and energy-saving slab continuous casting, taking the production of high-carbon steel as an example, includes the following steps:

[0037] S1. Secondary Cooling Control: Infrared thermal imagers are deployed in the secondary cooling zone to collect real-time temperature field information on the surface of the billet. The secondary cooling zone is divided into 16 wide independent control zones and 6 narrow independent control zones. The target temperature is set at 930℃. First, the basic water volume of each control zone is calculated using the existing fixed water distribution model. Then, ΔT is calculated based on the surface temperature of each control zone collected by the infrared thermal imager. iTaking Zone 5 as an example, the basic water volume is 50L / min. Due to equipment status, water pressure fluctuations, and other reasons, a temperature deviation occurs, with the measured temperature being 935℃. Therefore, ΔT5 = 5℃. The current pulling speed is v = 1.8m / min, and the steel grade coefficient K is taken as 0.85. According to the dynamic water distribution model formula, the spray water volume adjustment value Q5 for Zone 5 is calculated to be 6.21L / min, which means an increase of 6.21L / min on the basis of the basic water volume. The control system automatically adjusts the spray water volume of Zone 5 to 56.21L / min. The spray water volume adjustment method for other zones is the same.

[0038] S2. Vibration Control: The current real-time tension speed v = 1.8 m / min. Vibration parameters are calculated using an online vibration parameter adjustment model: Vibration frequency f = 28 / (1 + 1.5e) -0.4v ≈2.1Hz; Amplitude A = 5.5 - 0.8V + 0.12V 2 ≈4.2mm. With a waveform skew rate set at 20%, the servo hydraulic drive system drives the crystallizer to perform non-sinusoidal vibration according to the above parameters, ensuring vibration accuracy. Mechanical backlash control can be achieved by using a zero-backlash guide mechanism to control the mechanical backlash to <0.05mm.

[0039] S3. Multi-mode electromagnetic field control: A rotating magnetic field with a set frequency of 8Hz is applied in the crystallizer zone; a static magnetic field of 0.5T and a pulsed magnetic field of 2Hz are simultaneously applied in the secondary cooling zone; electromagnetic braking is applied in the end zone using an end electromagnetic brake with a gradient intensity distribution. The magnetic field strength is relatively weak in the central region of the billet width, and gradually increases in the regions near the narrow side. The magnetic field strength in the multi-mode electromagnetic field control satisfies the gradient control strategy: crystallizer zone > secondary cooling zone > end zone.

[0040] S4. Multi-parameter Steel Leakage Early Warning: Real-time acquisition of copper plate temperature, vibration acceleration, friction force, and acoustic emission signals from the crystallizer. These four signals are time-series aligned and feature-fused, then input into a pre-trained convolutional neural network model. A one-dimensional convolutional neural network (1D-CNN) is used as the early warning model, containing three convolutional layers with kernel sizes of 3, 5, and 3, and a stride of 1. Each convolutional layer is followed by a batch normalization layer and a ReLU activation function; two max-pooling layers with a pooling window size of 2; and finally, two fully connected layers with 128 and 64 neurons respectively. The output layer uses the Softmax function to calculate the steel leakage probability. Historical production data is collected, and normal samples and steel leakage samples are labeled by experts, divided into training and test sets at an 8:2 ratio. The cross-entropy loss function, Adam optimizer, initial learning rate of 0.001, and 100 training epochs are used. The trained model achieves an accuracy of 98.5% on the test set.

[0041] The system sets leakage probability thresholds P1=60%, P2=80%, and P3=95%. Based on the probability P output by the model, the system makes real-time decisions: normal operation when P<60%; warning when 60%≤P<80%; automatic speed reduction when 80%≤P<95%; and emergency shutdown when P≥95%. This mechanism ensures accurate warnings while avoiding the impact of false alarms on production.

[0042] S5. Dynamic matching of protective slag performance: Current real-time casting speed v = 1.8 m / min, calculate the target viscosity value of the protective slag η = 0.25 + 0.1e -0.6v The viscosity is approximately 0.28 Pa·s. The continuous casting system adjusts the amount of protective slag added through an automatic slag adding device to make the actual viscosity approach 0.28 Pa·s, thus ensuring stable lubrication.

[0043] Example 2:

[0044] This embodiment provides an intelligent control method for high-efficiency and energy-saving slab continuous casting. The method steps are basically the same as those in Embodiment 1. The difference is that when the early warning model detects that the leakage probability threshold is 85%, which exceeds the speed reduction threshold of 80%, the continuous casting system automatically executes a speed reduction response, reducing the casting speed to 1.2 m / min.

[0045] Secondary cooling control: After the drawing speed is reduced, the measured temperature in the fifth zone drops to 915℃, so ΔT5=-15℃. The steel grade coefficient K is taken as 0.85. According to the dynamic water distribution model formula, Q5=-9.16L / min is calculated, which means that the water volume needs to be reduced by 9.16L / min based on the basic water volume.

[0046] Vibration control: The current real-time tension v = 1.2 m / min, the calculated f ≈ 1.9 Hz, A ≈ 4.6 mm, and the waveform skew rate is set to 15%.

[0047] Dynamic matching of protective slag performance: The target viscosity value of the protective slag is η≈0.28Pa·s. The system adjusts the amount of protective slag added to match the target value.

[0048] The above embodiments are merely descriptions of preferred embodiments of the present invention and are not intended to limit the concept and scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the design concept of the present invention should fall within the protection scope of the present invention.

[0049] The technologies, shapes, and structures not described in detail in this invention are all known technologies.

Claims

1. A highly efficient and energy-saving intelligent control method for slab continuous casting, characterized in that, Includes the following steps: S1, Secondary Cooling Control: Real-time acquisition of surface temperature field information of the billet, and zoned cooling control of the billet based on dynamic water distribution model; S2. Vibration control: Based on the real-time pulling speed, the crystallizer is controlled to perform non-sinusoidal vibration through an online vibration parameter adjustment model; S3, Multi-mode electromagnetic field control: Different modes of electromagnetic fields are applied in the crystallizer region, the secondary cooling region and the end region respectively; S4. Multi-parameter steel leakage early warning: Real-time diagnosis and early warning of steel leakage based on multi-source signal fusion and early warning model; S5. Dynamic matching of protective slag performance: Dynamically match the viscosity of the protective slag according to the real-time casting speed.

2. The intelligent control method for high-efficiency and energy-saving slab continuous casting according to claim 1, characterized in that, In step S1, the partitioned cooling control involves dividing the secondary cooling zone of the billet into at least 16 independent control zones on the wide side and at least 6 independent control zones on the narrow side. The dynamic water distribution model is: Q i =K×v 1.2 ×ΔT i 0.8 ; Among them, Q i Here, K is the adjustment value for the spray water volume in zone i, K is the steel grade coefficient, v is the real-time casting speed, and ΔT is the value for the water volume adjustment in zone i. i It is the difference between the actual temperature and the target temperature of the billet surface in zone i.

3. The intelligent control method for high-efficiency and energy-saving slab continuous casting according to claim 1, characterized in that, In step S2, the waveform skewness of the non-sinusoidal vibration is adjustable within the range of 10% to 35%, and the online frequency adjustment model includes: Vibration frequency adjustment formula: f=28 / (1+1.5e) -0.4v ), where f is the vibration frequency and v is the real-time pulling speed; Amplitude adjustment formula: A = 5.5 - 0.8v + 0.12v 2 Where A is the amplitude and v is the real-time pulling speed.

4. The intelligent control method for high-efficiency and energy-saving slab continuous casting according to claim 1, characterized in that, In step S3, the multi-mode electromagnetic field control is as follows: a rotating magnetic field is applied in the crystallizer region, a static magnetic field and a pulsed magnetic field are applied simultaneously in the secondary cooling region, and electromagnetic braking is applied in the end region.

5. The intelligent control method for high-efficiency and energy-saving slab continuous casting according to claim 4, characterized in that, The magnetic field strength in the multi-mode electromagnetic field control satisfies the gradient control strategy: crystallizer zone > secondary cooling zone > end zone; the end zone adopts an end electromagnetic brake with a gradient intensity distribution, the magnetic field strength is relatively weak in the central region of the billet width, and gradually increases in the two sides near the narrow face.

6. The intelligent control method for high-efficiency and energy-saving slab continuous casting according to claim 4, characterized in that, In step S3, the frequency of the rotating magnetic field in the crystallizer zone is adjusted within the range of 0~20Hz; the static magnetic field strength in the secondary cooling zone is 0.5T, and the pulse magnetic field frequency is 2Hz.

7. The intelligent control method for high-efficiency and energy-saving slab continuous casting according to claim 1, characterized in that, In step S4, the multi-source signals include, but are not limited to, the temperature, vibration, friction, and acoustic emission signals of the crystallizer; the early warning model is a convolutional neural network model. The diagnosis and early warning system involves real-time acquisition and fusion of temperature, vibration, friction, and acoustic emission signals from the crystallizer. The fused multi-source signals are then input into the early warning model, which outputs the probability of steel leakage. Based on the probability level, a three-level response is executed: early warning, speed reduction, or shutdown.

8. The intelligent control method for high-efficiency and energy-saving slab continuous casting according to claim 1, characterized in that, In step S5, the dynamic matching of the protective slag viscosity is calculated based on the real-time casting speed to determine the target value of the protective slag viscosity, and the amount or composition of the protective slag added is adjusted accordingly. The formula for calculating the target viscosity value of the protective slag is: η = 0.25 + 0.1e -0.6v Where η is the viscosity of the protective slag and v is the real-time casting speed.