Automatic slagging system for continuous casting machine

Abstract

The application discloses an automatic slag adding system of a continuous casting machine, which comprises a space detection module, a processor calculation module and a slag outlet module; the space detection module is used for detecting the thickness of the protective slag in the continuous casting machine; the processor calculation module can calculate the amount of the protective slag to be added according to the thickness of the protective slag in the continuous casting machine; and the slag outlet module can accurately add the corresponding amount of the protective slag into the continuous casting machine; the system has low use cost and can accurately determine the amount of the protective slag to be added.

Description

Technical Field

[0001] This invention relates to the field of metal casting technology, and in particular to an automatic slag feeding system for a continuous casting machine. Background Technology

[0002] As is well known, continuous casting mold flux plays a crucial protective role in the continuous casting process. It effectively protects the interface between the slag and the billet, reducing oxidation reactions and preventing metal oxidation and solidification. Before casting begins, a certain amount of mold flux is added to the slag trough of the continuous casting machine. As the casting process commences, the flux melts, forming a slag layer. During continuous casting, the mold flux continuously circulates, covering the space between the billet and the slag to form a protective layer. Finally, to maintain the performance and stability of the mold flux, the composition of the slag layer needs to be replenished and adjusted as needed. Traditional replenishment relies on the worker's experience to determine when and how much mold flux to add. However, this method lacks rigorous calculation and is prone to adding too much or too little mold flux. Insufficient mold flux can lead to a thinner slag layer, affecting the quality of the billet. Excessive mold flux, on the other hand, can cause the following problems:

[0003] 1. It can cause blockage of the slag chamber, thereby affecting the flow of molten steel and potentially leading to safety issues such as molten steel overflow or splashing.

[0004] 2. It causes uneven distribution of temperature and composition in the molten steel, resulting in areas of abnormally high or low temperature in the molten steel, leading to poor cooling conditions and poor crystallization in the crystallizer, which in turn has an adverse effect on the quality of the continuously cast billet.

[0005] 3. Excessive protective slag increases resistance during the casting process, restricting the flow of molten steel; this leads to a slower casting speed, longer casting time, and reduced continuous casting production efficiency.

[0006] Chinese patent (CN201310355697.X) discloses an automatic slag addition device for continuous casting machines. This device, via a feeding hopper positioned above the crystallizer, enables the automatic, timed, and quantitative addition of slag to the crystallizer on the steelmaking production line. This improves the uniformity of material distribution, reduces slag waste, enhances billet quality, and reduces the workload of operators. Furthermore, the overall equipment structure is simpler, occupies less space, and is less expensive. However, because the state of the slag in the continuous casting machine still requires manual judgment, the timing and amount of slag addition cannot be precisely determined, leading to the potential for excessive slag addition.

[0007] In addition, Chinese patent (CN202111506905.2) discloses a slag thickness adaptive control method and device based on the slag addition mechanism of continuous casting machine. It sends a specific control signal to the slag addition robot through a corresponding algorithm, thereby controlling the slag addition robot to add protective slag quantitatively. However, the cost of using the slag addition robot is relatively high.

[0008] Therefore, in summary, there is a need in the market for a slag addition system that is low in operating costs and can accurately determine the amount of protective slag to be added. Summary of the Invention

[0009] To overcome the shortcomings of the prior art, the present invention discloses an automatic slag feeding system for a continuous casting machine.

[0010] To achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0011] An automatic slag addition system for a continuous casting machine includes a space detection module, a processor calculation module, and a slag discharge module. The space detection module is used to detect the thickness of the protective slag inside the continuous casting machine. The processor calculation module can calculate the amount of protective slag to be added based on the thickness of the protective slag inside the continuous casting machine. The slag discharge module can accurately add the calculated amount of protective slag into the continuous casting machine.

[0012] Preferably, the space detection module consists of multiple laser scanners, which are located at the bottom of the slag discharge module, with the scanning end of the laser scanner facing the slag trough of the continuous casting machine.

[0013] Preferably, the calculation process of the processor computing module is as follows:

[0014] 1) Collect data including casting parameters, material properties, and production requirements;

[0015] 2) Based on the collected data and the detected thickness of the protective slag, a mathematical model was established using a multiple regression model;

[0016] 3) Based on the established mathematical model, optimization algorithms are used to find the optimal combination of slag addition rate, protective slag thickness, and slag density parameters;

[0017] 4) Set certain process constraints when optimizing parameters;

[0018] 5) The processor controls the amount of protective slag added in real time based on the final data obtained from the mathematical model under constraints.

[0019] Preferably, the specific formula of the mathematical model in step 2) is:

[0020] ;

[0021] Where y is the thickness of the protective slag. For regression coefficients, For pouring speed, For slag flow rate, For the density of slag, The error term is represented, and the specific formula for calculating the regression coefficient is as follows:

[0022]

[0023]

[0024]

[0025] ;

[0026] Here, mean() represents the average value, and sum() represents the summation.

[0027] Preferably, the optimization algorithm in step 3) adopts the gradient descent method, and its calculation process is as follows:

[0028] a. Define the objective function: The objective function is the thickness of the protective slag, i.e. ;

[0029] b. Initialize parameters: Select , , and As parameters, and assigning them initial values, denoted as... , , and ;

[0030] c. Define the loss function: The loss function is the squared difference of the objective function, used to measure the difference between the model's predicted value and the true value. The loss function can be defined as:

[0031] ;

[0032] in, This is the actual thickness of the protective slag. , and These are the corresponding actual pouring speed, actual slag flow rate, and actual slag density;

[0033] d. Calculate the gradient: for each parameter Calculate the gradient of the loss function with respect to this parameter. ;

[0034] e. Update parameters: According to the update rule of gradient descent, the learning rate is multiplied by the gradient to obtain the updated parameter values, and the update formula is as follows:

[0035] ;

[0036] ;

[0037] ;

[0038] ;

[0039] in, The learning rate determines the step size for parameter updates;

[0040] f. Repeat steps d and e until the stopping condition is met: Stopping conditions can be set, such as reaching the maximum number of iterations, the change in the loss function being less than a certain threshold, or reaching a predetermined accuracy requirement;

[0041] g. Output the optimal parameter combination: When the stopping condition is met, output the optimal combination. , , and The combination of these is used as the optimization result;

[0042] h. Repeat steps c and d until the stopping condition is met: You can set stopping conditions, such as reaching the maximum number of iterations, the objective function changing less than a certain threshold, or reaching a predetermined accuracy requirement;

[0043] i. Output optimal parameters: When the stopping condition is met, output the optimal parameter values ​​as the optimization result.

[0044] Preferably, the specific process of gradient calculation in step d is as follows:

[0045] ;

[0046] ;

[0047] ;

[0048] ;

[0049] Where Σ represents summing over all samples.

[0050] Preferably, the slag discharge module includes a storage tank, a solenoid valve, and a slag discharge pipe. The storage tank is used to store protective slag, and the bottom of the storage tank is provided with a discharge port. The discharge port is connected to the slag discharge pipe through the solenoid valve, and the slag discharge pipe discharges material into the slag trough.

[0051] Preferably, the storage box is provided with a baffle and a vibrating plate mechanism. The baffle is inclined, with its lower end hinged to the storage box and corresponding to the discharge port of the storage box. The upper end of the baffle abuts against the inner wall of the storage box, and the bottom of the baffle has an arc-shaped boss. The vibrating plate mechanism includes a support column, a motor, and a cam. The support column is installed on the inner bottom surface of the storage box, and the top of the support column has a cam and a motor for driving the cam to rotate. The wheel surface of the cam can abut against the boss.

[0052] Preferably, the high end face of the baffle is provided with an extension plate, and the inner side of the storage box is provided with a groove that cooperates with the extension plate.

[0053] By employing the technical solution described above, the present invention has the following beneficial effects:

[0054] The present invention discloses an automatic slag feeding system for a continuous casting machine. Through the processor calculation module, the system can accurately calculate the amount of slag to be added based on the thickness of the protective slag, thereby ensuring the smooth operation of subsequent continuous casting operations and the quality of the produced billets.

[0055] In addition, the use of a simple slag discharge module makes the system inexpensive. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of one structure of the present invention;

[0057] Figure 2 This is a schematic diagram of the slag discharge module.

[0058] In the diagram: 1. Space detection module; 2. Processor calculation module; 3. Slag discharge module; 301. Storage tank; 302. Solenoid valve; 303. Slag discharge pipe; 304. Baffle; 305. Vibrating plate mechanism; 3051. Support column; 3052. Motor; 3053. Cam; 306. Protrusion; 307. Extension plate. Detailed Implementation

[0059] The technical solution of the present invention will now be described with reference to the accompanying drawings of the embodiments of the present invention. In the description, it should be understood that the terms "upper," "lower," "front," "rear," "left," and "right," etc., indicating directions or positional relationships, are only used to correspond to the accompanying drawings of the present invention for the purpose of facilitating the description of the present invention, and are not intended to indicate or imply that the device or element referred to must have a specific orientation.

[0060] Combined with appendix Figure 1-2The aforementioned automatic slag addition system for a continuous casting machine includes a space detection module 1, a processor calculation module 2, and a slag discharge module 3. The space detection module 1 is used to detect the thickness of the protective slag inside the continuous casting machine. The processor calculation module 2 can calculate the amount of protective slag to be added based on the thickness of the protective slag inside the continuous casting machine. The slag discharge module 3 can accurately add the calculated amount of protective slag into the continuous casting machine.

[0061] Example 1: The space detection module 1 consists of multiple laser scanners. The laser scanners are located at the bottom of the slag removal module 3, and the scanning end of the laser scanners faces the slag trough of the continuous casting machine. The thickness of the protective slag of the billet can be detected in real time through the laser scanners.

[0062] Example 2 is as follows: The calculation process of the processor computing module 2 is as follows:

[0063] 1) Collect data including casting parameters, material properties and production requirements, mainly slag addition rate, protective slag thickness and slag density;

[0064] 2) Based on the collected data and the detected thickness of the protective slag, a mathematical model was established using a multiple regression model;

[0065] 3) Based on the established mathematical model, optimization algorithms are used to find the optimal combination of slag addition rate, protective slag thickness, and slag density parameters;

[0066] 4) Set certain process constraints when optimizing parameters;

[0067] 5) The processor controls the amount of protective slag added in real time based on the final data obtained from the mathematical model under constraints.

[0068] As needed, the specific formula for the mathematical model in step 2) is as follows:

[0069] ;

[0070] Where y is the thickness of the protective slag. , where represents the regression coefficients for the intercept term, pouring speed, slag flow rate, and slag density, respectively. For pouring speed, For slag flow rate, For the density of slag, The error term represents the part of the data that the model cannot perfectly fit. The specific formula for calculating the regression coefficients is as follows:

[0071] ;

[0072] ;

[0073] ;

[0074] ;

[0075] Here, mean() represents the average value, and sum() represents the summation.

[0076] Furthermore, the optimization algorithm in step 3) employs the gradient descent method, and its calculation process is as follows:

[0077] a. Define the objective function: The objective function is the thickness of the protective slag, i.e. ;

[0078] b. Initialize parameters: Select , , and As parameters, and assigning them initial values, denoted as... , , and ;

[0079] c. Define the loss function: The loss function is the squared difference of the objective function, used to measure the difference between the model's predicted value and the true value. The loss function can be defined as:

[0080] ;

[0081] in, This is the actual thickness of the protective slag. , and These are the corresponding actual pouring speed, actual slag flow rate, and actual slag density;

[0082] d. Calculate the gradient: for each parameter Calculate the gradient of the loss function with respect to this parameter. ;

[0083] e. Update parameters: According to the update rule of gradient descent, the learning rate is multiplied by the gradient to obtain the updated parameter values, and the update formula is as follows:

[0084] ;

[0085] ;

[0086] ;

[0087] ;

[0088] in, The learning rate determines the step size for parameter updates;

[0089] f. Repeat steps d and e until the stopping condition is met: Stopping conditions can be set, such as reaching the maximum number of iterations, the change in the loss function being less than a certain threshold, or reaching a predetermined accuracy requirement;

[0090] g. Output the optimal parameter combination: When the stopping condition is met, output the optimal combination. , , and The combination of these is used as the optimization result;

[0091] h. Repeat steps c and d until the stopping condition is met: You can set stopping conditions, such as reaching the maximum number of iterations, the objective function changing less than a certain threshold, or reaching a predetermined accuracy requirement;

[0092] i. Output optimal parameters: When the stopping condition is met, output the optimal parameter values ​​as the optimization result.

[0093] As needed, the specific process of gradient calculation in step d is as follows:

[0094] ;

[0095] ;

[0096] ;

[0097] ;

[0098] Where Σ represents summing over all samples.

[0099] Example 4: The slag discharge module 3 includes a storage tank 301, a solenoid valve 302, and a slag discharge pipe 303. The storage tank 301 is used to store protective slag. The bottom of the storage tank 301 is provided with a discharge port. The discharge port is connected to the slag discharge pipe 303 through the solenoid valve 302. The slag discharge pipe 303 discharges material into the slag trough. The opening and closing of the solenoid valve 302 can control the discharge action of the protective slag in the storage tank 301. The action of the solenoid valve 302 is controlled by the processor calculation module 2.

[0100] In addition, the storage tank 301 is equipped with a baffle 304 and a vibrating plate mechanism 305. The baffle 304 is inclined, with its lower end hinged to the storage tank 301 and corresponding to the discharge port of the storage tank 301. The upper end of the baffle 304 abuts against the inner wall of the storage tank 301. The bottom of the baffle 304 has an arc-shaped protrusion 306. The baffle 304 allows the protective slag inside the tank to move along the inclined plate towards the discharge port of the storage tank 301, achieving automatic feeding. The vibrating plate mechanism 305 includes a support column 3051, a motor 3052, and a cam 3053. The support column 3051 is installed on the inner bottom surface of the storage tank 301, and the top of the support column 3051 has a protrusion. The cam 3053 is a wheel 3053, and the motor 3052 is used to drive the cam 3053 to rotate. The wheel surface of the cam 3053 can abut against the boss 306. When the baffle 304 is in the normal state, the boss 306 at the bottom of the baffle 304 abuts against the short end of the cam 3053. After the motor 3052 drives the cam 3053 to rotate, the long end of the cam 3053 will push the boss 306, thereby driving the baffle 304 to move upward. After the long end of the cam 3053 no longer abuts against the boss 306, the baffle 304 will move downward and return to its original position. That is, the rotation of the cam 3053 can drive the baffle 304 to move up and down intermittently, realizing the vibration of the baffle 304, making the process of feeding protective slag on the baffle 304 smoother.

[0101] As needed, the high end face of the baffle 304 is provided with an extension plate 307, and the inner side of the storage box 301 is provided with a groove that cooperates with the extension plate 307. That is, during the up and down movement of the baffle 304, the extension plate 307 is always in the groove, thereby preventing the leakage of protective slag.

[0102] The parts of this invention not described in detail are prior art. It will be apparent to those skilled in the art that this invention is not limited to the details of the above exemplary embodiments, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be regarded as exemplary and non-limiting in all respects. The scope of this invention is defined by the appended claims rather than the foregoing description. Therefore, it is intended to include all changes that fall within the meaning and scope of the equivalents of the claims within this invention, and no reference numerals in the claims should be regarded as limiting the scope of the claims.

Claims

1. An automatic slag feeding system for a continuous casting machine, characterized in that: It includes a space detection module (1), a processor calculation module (2), and a slag removal module (3); the space detection module (1) is used to detect the thickness of the protective slag in the continuous casting machine, the processor calculation module (2) can calculate the amount of protective slag to be added based on the thickness of the protective slag in the continuous casting machine, and the slag removal module (3) can accurately add the calculated amount of protective slag into the continuous casting machine. The slag discharge module (3) includes a storage tank (301), a solenoid valve (302), and a slag discharge pipe (303). The storage tank (301) is used to store protective slag. The bottom of the storage tank (301) is provided with a discharge port. The discharge port is connected to the slag discharge pipe (303) through the solenoid valve (302). The slag discharge pipe (303) discharges material into the slag trough. The storage box (301) is provided with a baffle (304) and a vibrating plate mechanism (305). The baffle (304) is inclined, and the lower end of the baffle (304) is hinged to the storage box (301). The lower end of the baffle (304) is corresponding to the discharge port of the storage box (301). The upper end of the baffle (304) is in contact with the inner wall of the storage box (301). The bottom of the baffle (304) is provided with an arc-shaped boss (306). The vibrating plate mechanism (305) includes a support column (3051), a motor (3052), and a cam (3053). The support column (3051) is installed on the inner bottom surface of the storage box (301). The top of the support column (3051) is provided with a cam (3053) and a motor (3052) for driving the cam (3053) to rotate. The wheel surface of the cam (3053) can contact the boss (306). The calculation process of the processor computing module (2) is as follows: 1) Collect data including casting parameters, material properties, and production requirements; 2) Based on the collected data and the detected thickness of the protective slag, a mathematical model was established using a multiple regression model; 3) Based on the established mathematical model, optimization algorithms are used to find the optimal combination of slag addition rate, protective slag thickness, and slag density parameters; 4) Set certain process constraints when optimizing parameters; 5) The processor controls the amount of protective slag added in real time based on the final data obtained from the mathematical model under constraints.

2. The automatic slag feeding system for continuous casting machines as described in claim 1, characterized in that: The space detection module (1) consists of multiple laser scanners. The laser scanners are located at the bottom of the slag discharge module (3), and the scanning end of the laser scanners faces the slag trough of the continuous casting machine.

3. The automatic slag feeding system for continuous casting machines as described in claim 1, characterized in that: The specific formula for the mathematical model in step 2) is: ； Where y is the thickness of the protective slag. For regression coefficients, For pouring speed, For slag flow rate, For the density of slag, The error term is represented, and the specific formula for calculating the regression coefficient is as follows: ； Here, mean() represents the average value, and sum() represents the summation.

4. The automatic slag feeding system for continuous casting machines as described in claim 3, characterized in that: The optimization algorithm in step 3) uses the gradient descent method, and its calculation process is as follows: a. Define the objective function: The objective function is the thickness of the protective slag, i.e. ; b. Initialize parameters: Select , , and As parameters, and assigning them initial values, denoted as... , , and ; c. Define the loss function: The loss function is the squared difference of the objective function, used to measure the difference between the model's predicted value and the true value. The loss function can be defined as: ； in, This is the actual thickness of the protective slag. , and These are the corresponding actual pouring speed, actual slag flow rate, and actual slag density; d. Calculate the gradient: for each parameter Calculate the gradient of the loss function with respect to this parameter. ; e. Update parameters: According to the update rule of gradient descent, the learning rate is multiplied by the gradient to obtain the updated parameter values, and the update formula is as follows: ； ； ； ； in, The learning rate determines the step size for parameter updates; f. Repeat steps d and e until the stopping condition is met: Stopping conditions can be set, such as reaching the maximum number of iterations, the change in the loss function being less than a certain threshold, or reaching a predetermined accuracy requirement; g. Output the optimal parameter combination: When the stopping condition is met, output the optimal combination. , , and The combination of these is used as the optimization result; h. Repeat steps c and d until the stopping condition is met: You can set stopping conditions, such as reaching the maximum number of iterations, the objective function changing less than a certain threshold, or reaching a predetermined accuracy requirement; i. Output optimal parameters: When the stopping condition is met, output the optimal parameter values ​​as the optimization result.

5. The automatic slag feeding system for continuous casting machines as described in claim 4, characterized in that: The specific process of gradient calculation in step d is as follows: ； ； ； ； Where Σ represents summing over all samples.

6. The automatic slag feeding system for continuous casting machines as described in claim 1, characterized in that: The high end face of the baffle (304) is provided with an extension plate (307), and the inner side of the storage box (301) is provided with a groove that cooperates with the extension plate (307).

Images