Cold rolling method for steel sheet, production method for cold rolled steel sheet and production facility for cold rolled steel sheet

Abstract

To provide a cold rolling method for a steel sheet capable of suppressing fluctuation of sheet thickness over a coil entire length including a rolling acceleration / deceleration region and satisfying a severe requirement of sheet thickness accuracy on the basis of a friction coefficient prediction method enabling highly accurate prediction of a friction coefficient between a work roll and a steel sheet and an obtained friction coefficient, a production method for a cold rolled steel sheet and a production facility for the cold rolled steel sheet.SOLUTION: In a cold rolling method for a steel sheet, roll circumferential speed of a cold rolling mill is controlled by using a prediction model for predicting a friction coefficient between a work roll and a steel sheet of the cold rolling mill. The prediction model is generated by using rolling record data as an explanatory variable and an estimation value of a friction coefficient as an objective variable. The rolling record data includes deformation resistance of the steel sheet and a work roll diameter. The cold rolling method for the steel sheet includes a step of predicting the friction coefficient between the work roll and the steel sheet of the cold rolling mill by inputting a rolling condition of the steel sheet to the prediction model and controlling the roll circumferential speed of the cold rolling mill on the basis of sheet thickness change on the cold rolling mill outlet side.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a cold rolling method for steel sheets, a method for manufacturing cold-rolled steel sheets, and manufacturing equipment for cold-rolled steel sheets.

Background Art

[0002] In the cold rolling process, the technical difficulty of rolling has increased due to the increase in thin and hard materials, and there is an increasing demand for strict sheet thickness accuracy. As a result, the conventional sheet thickness control technology may cause the sheet thickness tolerance to deviate. Particularly in the acceleration and deceleration regions of rolling including the low-speed region, the rolling load fluctuates greatly, and sheet thickness fluctuations are likely to occur. This is affected by the change in the friction coefficient due to the change in the amount of rolling oil introduced between the work roll and the steel sheet. Therefore, accurate prediction of the friction coefficient is important for suppressing sheet thickness fluctuations in the acceleration and deceleration regions of rolling.

[0003] As a technique for predicting the friction coefficient during rolling, for example, in Patent Document 1, the following method has been proposed for predicting the friction coefficient generated between the work roll of a rolling mill and a steel sheet during rolling of the steel sheet. In this method, the friction coefficient is predicted by a pre-created friction coefficient model formula, and the roll gap setting value of the rolling mill is corrected according to the change in the predicted friction coefficient. Also, at that time, the friction coefficient μ A inversely calculated by the rolling load formula and the friction coefficient μ C predicted by the friction coefficient model formula, the friction coefficient change amount ε A which is the difference between them, is predicted by machine learning (specifically, a neural network) to correct the friction coefficient change amount ε C for correction.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, it is difficult to accurately predict the rapid change in the friction coefficient in the low-speed rolling region only by optimizing the coefficients of the friction coefficient model formula proposed in Patent Document 1. In addition, there is a problem that the variables used as explanatory variables in the machine learning model do not sufficiently cover the variables related to the variation of the friction coefficient, and the prediction accuracy of the friction coefficient is low.

[0006] Therefore, the present invention has been made in view of the above, and an object thereof is to provide a friction coefficient prediction method capable of predicting the friction coefficient between a work roll and a steel sheet with high accuracy, and based on the obtained friction coefficient, suppressing the sheet thickness variation over the entire length of the coil including the acceleration and deceleration regions of rolling, and providing a cold rolling method for a steel sheet, a method for manufacturing a cold-rolled steel sheet, and a manufacturing facility for a cold-rolled steel sheet that can satisfy strict required sheet thickness accuracy.

Means for Solving the Problems

[0007] The cold rolling method for a steel sheet according to the present invention that advantageously solves the above problems is configured as follows. [1] A cold rolling method for a steel sheet in which the steel sheet is cold-rolled by controlling the roll peripheral speed in the stand of the cold rolling mill when cold-rolling the steel sheet, using a prediction model for predicting the friction coefficient between the work roll of the cold rolling mill and the steel sheet, wherein the prediction model is generated using, as explanatory variables, rolling performance data and, as the objective variable, an estimated value of the friction coefficient when the steel sheet was cold-rolled in the past. The rolling performance data includes the deformation resistance of the steel sheet and the work roll diameter. By inputting the rolling conditions of the steel sheet to be rolled into the prediction model, the friction coefficient between the work roll of the cold rolling mill and the steel sheet is continuously predicted, and based on the change in the predicted friction coefficient, the step of controlling the roll peripheral speed of the cold rolling mill is included.

[0008] The method for manufacturing a cold-rolled steel sheet according to the present invention that advantageously solves the above problems is configured as follows. [2] A method for manufacturing a cold-rolled steel sheet, including the step of manufacturing a cold-rolled steel sheet using the cold rolling method for a steel sheet described in [1] above.

[0009] The manufacturing equipment for cold-rolled steel sheets according to the present invention that advantageously solves the above problems is configured as follows. [3] It includes a cold rolling mill for cold rolling a steel sheet and a control device for controlling the control amount of the roll peripheral speed of the cold rolling mill. The control device uses a prediction model for predicting the friction coefficient between the work roll of the cold rolling mill and the steel sheet to control the roll peripheral speed at the stand of the cold rolling mill when cold rolling the steel sheet. The prediction model is generated with the rolling performance data as the explanatory variable and the estimated value of the friction coefficient as the objective variable when the steel sheet was cold rolled in the past. The rolling performance data includes the deformation resistance of the steel sheet and the work roll diameter. By inputting the rolling conditions of the steel sheet to be rolled into the prediction model, the friction coefficient between the work roll of the cold rolling mill and the steel sheet is continuously predicted, and based on the change in the predicted friction coefficient, the roll peripheral speed of the cold rolling mill is controlled. It is the manufacturing equipment for cold-rolled steel sheets.

Effect of the Invention

[0010] According to the present invention, by using a prediction model that adds the deformation resistance of the steel sheet and the work roll diameter as input data and performs machine learning, the change amount of the friction coefficient used when correcting the roll peripheral speed setting value of the rolling mill can be predicted with high accuracy.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Mode for Carrying Out the Invention

[0012] Hereinafter, with reference to the drawings, a cold rolling method of a steel sheet, a method for manufacturing a cold-rolled steel sheet, and a manufacturing facility for a cold-rolled steel sheet, which are embodiments of the present invention, will be described in detail. Note that the constituent elements in the embodiments shown below include those that can be replaced by those skilled in the art, those that are easy to replace, or those that are substantially the same.

[0013] <Manufacturing Facility for Cold-Rolled Steel Sheet> The manufacturing facility for a cold-rolled steel sheet according to the present embodiment includes a cold rolling mill for cold rolling a steel sheet and a control device for controlling the control amount of the roll peripheral speed of the cold rolling mill. The control device uses a prediction model for predicting the friction coefficient between the work roll of the cold rolling mill and the steel sheet to control the roll peripheral speed in the stand of the cold rolling mill when cold rolling the steel sheet. The prediction model is generated with the rolling performance data as the explanatory variable and the estimated value of the friction coefficient as the objective variable when the steel sheet was cold rolled in the past. The rolling performance data includes the deformation resistance of the steel sheet and the work roll diameter. By inputting the rolling conditions of the steel sheet to be rolled into the prediction model, the friction coefficient between the work roll of the cold rolling mill and the steel sheet is continuously predicted, and based on the change in the predicted friction coefficient, the roll peripheral speed of the cold rolling mill is controlled.

[0014] Figure 1 is a schematic diagram showing the configuration of a cold-rolled steel sheet manufacturing facility according to an embodiment of the present invention. As shown in Figure 1, a cold-rolled steel sheet manufacturing facility (hereinafter abbreviated as "manufacturing facility") according to an embodiment of the present invention is a continuous tandem rolling line having a plurality of stands, and includes an uncoiler 1, a joining device 2, a looper 3, a cold tandem rolling mill 4, a cutter 5, a tension reel 6, a tensiometer 7, and a thickness gauge 8.

[0015] The uncoiler 1 is a device for paying out the steel sheet S. The manufacturing facility may include a plurality of uncoilers 1. In this case, the plurality of uncoilers pay out different steel sheets S.

[0016] The joining device 2 is a device for joining the tail end of the steel sheet (preceding material) paid out earlier from the uncoiler 1 and the tip end of the steel sheet (following material) paid out later from the uncoiler 1 to form a joined steel sheet. As the joining device 2, a laser welding machine is preferably used.

[0017] The looper 3 is a device for storing the steel sheet S so that cold rolling by the cold tandem rolling mill 4 can be continued until the steel sheets are joined by the joining device 2 (until the joining is completed).

[0018] The cold tandem rolling mill 4 is a device for cold rolling the steel sheet S to make the thickness of the steel sheet S the target thickness. In the present embodiment, the cold tandem rolling mill 4 includes five rolling stands of the first rolling stand to the fifth rolling stand (#1std to #5std) in order from the inlet side (right side toward the paper surface of Figure 2) to the outlet side (left side toward the paper surface of Figure 2) of the steel sheet S. The configuration of the rolling stand and the conveying device of the steel sheet S are not particularly limited. Further, in the present embodiment, the cold tandem rolling mill 4 has a form called 4Hi having four rolls in one stand, but is not limited thereto, and other forms such as 6Hi can also be applied.

[0019] The cutter 5 is a device for cutting the cold-rolled steel sheet S into coils.

[0020] The tension reel 6 is a device for winding the steel sheet S cut by the cutting machine 5. The type of the tension reel 6 is not limited, and for example, it may be a carousel tension reel. Further, the manufacturing facility may be provided with a plurality of tension reels 6. In this case, the plurality of tension reels 6 continuously wind a plurality of steel sheets S.

[0021] The tensiometer 7 measures the tension of the steel sheet S on the inlet side and the outlet side of the rolling mill (each rolling stand).

[0022] The thickness gauge 8 measures the thickness of the steel sheet S on the inlet side and the outlet side of the rolling mill (each rolling stand).

[0023] During the rolling of the steel sheet S, the arithmetic unit 9 predicts the friction coefficient between the work roll of the rolling mill and the steel sheet S by a prediction model, and controls the roll peripheral speed of the rolling mill so as to suppress the change in the sheet thickness on the outlet side of the rolling mill due to the change in the predicted friction coefficient. In this embodiment, the explanatory variables and the objective variables of the prediction model are input with the values of each stand of the cold tandem rolling mill 4, and the friction coefficient of each stand can be predicted by the same prediction model. Further, roll peripheral speed control is performed at each stand where the friction coefficient is predicted. The roll peripheral speed control of this embodiment is not limited to this, and a prediction model may be created for each stand of the cold tandem rolling mill 4, or the roll peripheral speed control may not be applied to all stands but only to some stands.

[0024] FIG. 2 is a block diagram showing the configuration of the arithmetic unit 9. As shown in FIG. 2, the arithmetic unit 9 includes an arithmetic device 91, an input device 92, a storage device 93, and an output device 94.

[0025] The arithmetic device 91 is wired-connected to the input device 92, the storage device 93, and the output device 94 via a bus wiring 95. However, the connection form of the arithmetic device 91, the input device 92, the storage device 93, and the output device 94 is not limited to wired connection, and they may be wirelessly connected or connected in a combination of wired and wireless forms.

[0026] The input device 92 functions as an input port to which control information from the roll peripheral speed control device and information from the operation information device 10 are input. The information from the operation information device 10 may include information regarding the steel sheet S to be rolled (steel type, dimensions), cold rolling condition information (numerical information, character information, and image information) set by the process computer or the operator before cold rolling, and rolling condition information (numerical information, character information, and image information) during cold rolling, etc.

[0027] The storage device 93 is composed of a known storage device such as a hard disk drive, a semiconductor drive, an optical drive, etc., and is a device that stores information necessary for this system (information necessary for realizing the functions of the arithmetic processing unit 98 described later).

[0028] The output device 94 functions as an output port that outputs a control signal from the arithmetic unit 91 to the roll peripheral speed control device.

[0029] The arithmetic unit 91 includes a RAM 96, a ROM 97, and an arithmetic processing unit 98. The RAM 96, the ROM 97, and the arithmetic processing unit 98 are connected to the input device 92, the storage device 93, and the output device 94 via a bus wiring 95.

[0030] The RAM 96 is a working main memory used when performing some processing in the arithmetic unit 91.

[0031] The ROM 97 stores a prediction model 97A and a prediction model execution program 97B. As shown in FIG. 3, when creating the prediction model 97A, the rolling performance data of the steel sheet S to be rolled is taken as an explanatory variable, and the difference between the friction coefficient calculated from the following formula (1) and the friction coefficient calculated from the following formula (2) is taken as an objective variable, and machine learning is performed. Here, when predicting the friction coefficient between the steel sheet S to be rolled and the work roll, it is common to estimate the friction coefficient by inverse calculation from a known rolling load formula as shown in the following formula.

[0032]

Equation

[0033]

number

[0034] The friction coefficients calculated from equation (1) and equation (2) are shown in Figure 4. A large difference can be seen when comparing the friction coefficient estimated using equation (1) from the actual values ​​when rolling the steel plate S to be rolled with the friction coefficient estimated using equation (2) prepared in advance from the relationship between the rolling speed v and reduction ratio r. Furthermore, looking at the plot of the friction coefficient estimated from equation (1), it can be seen that the friction coefficient decreases as the rolling speed increases. This is thought to be because the higher the rolling speed, the more rolling oil gets caught in the roll bite, causing the friction coefficient to decrease. In addition, the rate of change of the friction coefficient with respect to the rolling speed is greater at lower speeds; in the region where the rolling speed changes from low to high at the start of rolling of each coil (the tip of the coil) and in the region where the rolling speed changes from high to low near the end of rolling of each coil (the tail end of the coil), the rate of change of the friction coefficient is high and the rate of change of the rolling load is also high, requiring highly accurate thickness control to suppress fluctuations in the delivery thickness.

[0035] Therefore, in this embodiment, instead of the conventionally used formula (2), a prediction model that can predict with high accuracy the friction coefficient between the steel sheet S and the work roll during rolling is used to suppress the delivery thickness fluctuation. When creating the prediction model 97A, it is desirable to perform machine learning using as much rolling performance data as possible for the steel plate S to be thickness-controlled, but the method is not particularly limited. Neural network models such as deep learning, regression tree models such as random forest, gradient boosting such as XGBOOST, etc. can be used, but other known machine learning methods may be adopted.

[0036] In this embodiment, since a neural network is adopted as the machine learning method, the prediction model 97A is a neural network model. The neural network model is expressed by a function formula, for example. Specifically, the hyperparameters used in the machine learning model are set, and learning is performed using the neural network model with these hyperparameters.

[0037] As the optimization calculation of the hyperparameters, for the learning data, a neural network model with some of the hyperparameters changed step by step is created, and the hyperparameters that maximize the prediction accuracy for the verification data are selected. As hyperparameters, usually, the number of hidden layers, the number of neurons in each hidden layer, the dropout rate in each hidden layer (blocking the transmission of neurons with a certain probability), the activation function in each hidden layer, and the number of outputs are set, but it is not limited to this. Also, the optimization method of the hyperparameters is not particularly limited, but grid search for changing the parameters step by step, random search for randomly selecting the parameters, or search by Bayesian optimization can be used. In this embodiment, the prediction model 97A is in a form pre-stored in the ROM 97, but the form is not limited to this. For example, the arithmetic processing unit 98 may be provided with a function for creating the prediction model 97A to create the prediction model 97A online.

[0038] Here, we examined the parameters used as explanatory variables in the prediction model 97A. Specifically, in addition to the conventionally used parameters such as the concentration of rolling oil, the temperature of rolling oil, the flow rate of rolling oil, the rolling speed, the coil length, the reduction ratio, the inlet plate thickness, and the outlet plate thickness (used in Patent Document 1), we evaluated the influence of the work roll diameter, which is a variable related to the change in the deformation resistance of the steel plate S and the contact area between the steel plate S and the work roll. Generally, the frictional force between the steel plate S and the work roll is related to the true contact area. Assuming formulas such as Hitchcock's formula and using known theoretical formulas such as the approximate formula of Bland & Ford, the smaller the work roll diameter and the lower the hardness (deformation resistance) of the steel plate, the higher the friction coefficient.

[0039] Therefore, for example, when rolling a steel plate with high hardness properties, if the friction coefficient is predicted without considering the hardness of the steel plate and the work roll diameter, there may be a large difference between the actual friction coefficient and the predicted value of the friction coefficient. As a result, the correction of the roll peripheral speed setting value of the rolling mill calculated based on the change amount of the friction coefficient becomes insufficient. For example, as shown in FIG. 5, depending on the steel plate, a plate thickness deviation may occur when the rolling speed is accelerated or decelerated. In the figure, (a) is a graph showing the rolling speed in the time direction, and (b) is a graph showing the plate thickness deviation in the time direction.

[0040] As an explanatory variable of the prediction model, the deformation resistance of the steel plate S is considered to be newly considered because it affects the rolling load. It is not a constant, but the actual performance values are obtained in advance and sorted and applied for each steel type, for example.

[0041] In addition, the work roll diameter decreases by about several tens of μm due to changes over time (wear) during use. The work roll diameter may also increase by about several tens of μm due to thermal expansion during rolling. These wear and thermal expansion can be estimated based on the rolling distance of the steel plate. A work roll that has been used to a certain extent (for example, a work roll worn by several tens of μm) is taken out from the rolling mill, its surface is polished by a polishing machine, and then reused within the operating range. When the operating range diameter (the difference between the upper limit value and the lower limit value) is 50 mm, the work roll used for #1std of the rolling mill shown in Fig. 1 may be, for example, initially Φ600 mm, and after polishing, it may be Φ580 when loaded, or Φ550 in some cases. This value (the value of the work roll diameter after roll polishing) is recorded on the system.

[0042] The relationship between the friction coefficient calculated inversely from Equation (1), the friction coefficient predicted from the prior art (for example, Patent Document 1), and the friction coefficient predicted from the prediction model used in this embodiment (adding the deformation resistance of the steel sheet S and the work roll diameter to the explanatory variables) and the rolling speed is shown in Fig. 6. It can be seen that the friction coefficient predicted from the prediction model used in this embodiment is almost the same as the friction coefficient calculated inversely from Equation (1) in the entire speed range, indicating that the prediction accuracy of the friction coefficient has been improved compared with the prior art. As a result, for example, as shown in Fig. 7, it becomes possible to suppress the sheet thickness deviation when the rolling speed is accelerating or decelerating. In the same figure, (a) is a graph showing the rolling speed in the time direction, and (b) is a graph showing the sheet thickness deviation in the time direction.

[0043] Also, in Fig. 2, the arithmetic processing unit 98 has an arithmetic processing function and is wired-connected to the RAM 96 and the ROM 97 via the bus wiring 95. The arithmetic processing unit 98 determines the rolling conditions (roll peripheral speed conditions) of the steel sheet S to be rolled. To perform the above processing, when the arithmetic processing unit 98 receives a signal notifying that cold rolling is being executed from the roll peripheral speed control device via the input device 92, it executes the prediction model execution program 97B stored in the ROM 97, thereby functioning as the information reading unit 98A, the data preprocessing unit 98B, the roll peripheral speed condition determination unit 98C, and the result output unit 98D.

[0044] The information reading unit 98A reads the control information by the roll peripheral speed control device obtained from the input device 92 and the information from the operation information device 10.

[0045] The data preprocessing unit 98B executes a creation process of data to be input to the roll peripheral speed determination unit 98C. Specifically, the data preprocessing unit 98B performs processes such as unit conversion, deletion of coil data including abnormal data or unacquired data, or data filling in order to cause the prediction model 97A to read the rolling result data and the rolling conditions of the steel sheet S to be rolled.

[0046] The roll peripheral speed condition determination unit 98C controls each unit to determine the roll peripheral speed condition of the steel sheet S to be rolled by causing the prediction model 97A to read the information obtained from the data preprocessing unit 98B. The result output unit 98D outputs the determined roll peripheral speed condition to the roll peripheral speed control device.

[0047] [Modification Example] Although the embodiments of the present invention have been described above, the present embodiment is not limited thereto, and various changes and improvements can be made. For example, the devices included in the manufacturing equipment are not limited to the devices described above. Therefore, it is also possible to make the cold rolling process and the pickling process, which is the previous process thereof, continuous, and an acid pickling device for pickling the steel sheet S may be arranged between the looper 3 and the cold tandem rolling mill 4. Further, the cold tandem rolling mill 4 is not limited to 4Hi, and may be a multi-rolling mill such as 5Hi or 6Hi, and there is no particular limitation on the number of rolling stands. When creating the prediction model 97A, instead of the difference in the friction coefficient predicted from the theoretical formula and the model formula, the friction coefficient may be directly estimated (used as the target variable). Further, the present embodiment does not need to be applied to all steel types rolled on the rolling line, and it is also possible to apply it only to specific steel types such as steel sheets with strict thickness requirements. Further, in the embodiment, the roll peripheral speed control of the rolling mill has been described as the control target for suppressing the change in the sheet thickness on the exit side of the rolling mill due to the change in the predicted friction coefficient. Instead of this, or at the same time as this, roll gap control may be executed.

[0048] <Cold Rolling Method of Cold Rolled Steel Sheet> The cold rolling method of the cold rolled steel sheet according to this embodiment is a cold rolling method of a steel sheet by controlling the roll peripheral speed in the stand of a cold rolling mill when cold rolling the steel sheet, using a prediction model for predicting the friction coefficient between the work roll of the cold rolling mill and the steel sheet. The prediction model is generated with the rolling performance data as the explanatory variable and the estimated value of the friction coefficient as the objective variable when the steel sheet was cold rolled in the past. The rolling performance data includes the deformation resistance of the steel sheet and the work roll diameter. By inputting the rolling conditions of the steel sheet to be rolled into the prediction model, the friction coefficient between the work roll of the cold rolling mill and the steel sheet is continuously predicted, and based on the change in the predicted friction coefficient, the step of controlling the roll peripheral speed of the cold rolling mill is included, which is a cold rolling method of the steel sheet.

[0049] Roll peripheral speed control Referring to FIG. 8, the flow of the roll peripheral speed condition control process, which is an embodiment of the present invention, will be described. FIG. 8 is a flowchart showing the flow of the roll peripheral speed condition control process, which is an embodiment of the present invention. The flowchart shown in FIG. 8 starts at the timing when a signal notifying that cold rolling is being executed is input from the roll peripheral speed control device via the input device 92, and the roll peripheral speed condition control process proceeds to the process of step S1.

[0050] In the process of step S1, the information reading unit 98A reads the prediction model 97A stored in the ROM 97. Thereby, the process of step S1 is completed, and the roll peripheral speed condition control process proceeds to the process of step S2.

[0051] In the process of step S2, the information reading unit 98A reads the current rolling conditions of the steel sheet S to be rolled via the input device 92. Thereby, the process of step S2 is completed, and the roll peripheral speed condition control process proceeds to the process of step S3.

[0052] In the process of step S3, the roll peripheral speed determination unit 98C inputs the rolling conditions read in the process of step S2 into the prediction model 97A read in the process of step S1, and predicts the change amount of the friction coefficient. Thereby, the process of step S3 is completed, and the roll peripheral speed condition control process proceeds to the process of step S4.

[0053] In the process of step S4, the roll peripheral speed determination unit 98C predicts the change amount of the rolling load from the change amount of the friction coefficient continuously predicted in the process of step S3. The change amount of the rolling load from the change amount of the friction coefficient is predicted using the following formula. Thereby, the process of step S4 is completed, and the roll peripheral speed condition control process proceeds to the process of step S5.

[0054]

Equation

[0055] Here, ΔP represents the change amount of the rolling load, a represents the influence coefficient of the change amount of the friction coefficient on the change amount of the rolling load, and Δμ represents the change amount of the friction coefficient. In the process of step S5, the roll peripheral speed determination unit 98C predicts the change amount of the roll peripheral speed from the change amount of the rolling load predicted in the process of step S4. The change amount of the roll peripheral speed from the change amount of the rolling load is predicted using the following formula.

[0056]

Equation

[0057] Here, ΔS represents the change amount of the roll peripheral speed, b represents the influence coefficient of the change amount of the rolling load on the change amount of the roll peripheral speed, and ΔP represents the change amount of the rolling load. Through the above processing, the roll peripheral speed condition determination unit 98C ends a series of roll peripheral speed condition control processes. In this way, the arithmetic processing unit 98 determines the roll peripheral speed conditions using the prediction model 97A from the rolling conditions of the steel plate S to be rolled.

[0058] As is clear from the above description, in the present embodiment, first, a prediction model 97A is created with the rolling performance data when the steel sheet S was cold-rolled in the past as the explanatory variable and the difference between the formulas (1) and (2) for predicting the friction coefficient as the objective variable. Then, the arithmetic processing unit 98 determines the roll peripheral speed condition of the steel sheet S to be rolled from the rolling conditions of the steel sheet S to be rolled using the prediction model 97A. Thereby, it is possible to suppress the sheet thickness variation over the entire coil length including the acceleration / deceleration region of rolling and to improve the sheet thickness accuracy.

Example

[0059] Hereinafter, the present embodiment will be specifically described using examples, but the present invention is not limited to these examples.

[0060] Using the cold tandem rolling mill consisting of all 5 rolling stands of the embodiment shown in FIG. 1, cold rolling experiments were carried out on steel types A and B (with different deformation resistances) with a base material thickness of 2.2 mm, a finish thickness of 0.35 mm, and a plate width of 1300 mm. The effects of the present invention were verified in multiple chances with different work roll diameters and acceleration / deceleration rates during rolling. Table 1 shows the steel types, work roll diameters, acceleration / deceleration rates, the sheet thickness control systems used, and the maximum sheet thickness deviations. Here, as shown in Table 1, the acceleration / deceleration rate is lower for Pattern I and higher for Pattern II. Note that the required sheet thickness deviation for both steel types A and B is within ±5 μm.

[0061] First, as a preliminary preparation, prediction models for steel types A and B are created. Learning is carried out using a machine learning model (deep learning) with learning data (rolling performance data of about 3000 pieces for each steel type). This prediction model was used in the following inventive examples.

[0062] In Inventive Example 1, the steel sheet is steel type A with high deformation resistance, the work roll diameter is φ530, and the acceleration / deceleration pattern is I. The maximum sheet thickness deviation over the entire coil length is 4.0 μm, achieving the required sheet thickness accuracy. In Invention Example 2, the steel sheet is of steel type A with high deformation resistance, the work roll diameter is φ530, and the acceleration / deceleration pattern is II. The acceleration / deceleration rate is higher than that in Invention Example 1, and it is a condition where rapid fluctuations in rolling load are likely to occur. However, the maximum sheet thickness deviation over the entire coil length is 4.5 μm, achieving the required sheet thickness accuracy. In Invention Example 3, the steel sheet is of steel type A with high deformation resistance, the work roll diameter is φ560, and the acceleration / deceleration pattern is I. The maximum sheet thickness deviation over the entire coil length is 4.0 μm, achieving the required sheet thickness accuracy. In Invention Example 4, the steel sheet is of steel type A with high deformation resistance, the work roll diameter is φ560, and the acceleration / deceleration pattern is II. The maximum sheet thickness deviation over the entire coil length is 5.0 μm, achieving the required sheet thickness accuracy.

[0063] In Invention Example 5, the steel sheet is of steel type B with low deformation resistance, the work roll diameter is φ530, and the acceleration / deceleration pattern is I. The maximum sheet thickness deviation over the entire coil length is 4.0 μm, achieving the required sheet thickness accuracy. In Invention Example 6, the steel sheet is of steel type B with low deformation resistance, the work roll diameter is φ530, and the acceleration / deceleration pattern is II. The maximum sheet thickness deviation over the entire coil length is 4.5 μm, achieving the required sheet thickness accuracy. In Invention Example 7, the steel sheet is of steel type B with low deformation resistance, the work roll diameter is φ560, and the acceleration / deceleration pattern is I. The maximum sheet thickness deviation over the entire coil length is 4.0 μm, achieving the required sheet thickness accuracy. In Invention Example 8, the steel sheet is of steel type B with low deformation resistance, the work roll diameter is φ560, and the acceleration / deceleration pattern is II. The maximum sheet thickness deviation over the entire coil length is 5.0 μm, achieving the required sheet thickness accuracy.

[0064] In Comparative Example 1, the control system of Patent Document 1 is used. The steel sheet is of steel type A with high deformation resistance, the work roll diameter is φ530, and the acceleration / deceleration pattern is I. The maximum sheet thickness deviation over the entire coil length is 7.5 μm, and the required sheet thickness accuracy cannot be satisfied. It is considered that the hardness of the steel sheet and the fluctuations in the contact area between the steel sheet and the work roll are not included in the explanatory variables of the prediction model, resulting in low prediction accuracy of the change amount of the friction coefficient. Comparative Example 2 has the same sheet thickness control as Comparative Example 1. The steel sheet is Steel Grade A with high deformation resistance, the work roll diameter is φ530, and the acceleration / deceleration pattern is II. The maximum sheet thickness deviation over the entire coil length is 10.0 μm, and the required sheet thickness accuracy cannot be satisfied. Compared with Comparative Example 1, the sheet thickness deviation is larger, which is considered to be due to the high acceleration / deceleration rate and the resulting rapid change in rolling load and friction coefficient in the acceleration / deceleration section. Comparative Example 3 has the same sheet thickness control as Comparative Example 1. The steel sheet is Steel Grade A with high deformation resistance, the work roll diameter is φ560, and the acceleration / deceleration pattern is I. The maximum sheet thickness deviation over the entire coil length is 7.5 μm, and the required sheet thickness accuracy cannot be satisfied. The cause is considered to be the same as that of Comparative Example 1. Comparative Example 4 has the same sheet thickness control as Comparative Example 1. The steel sheet is Steel Grade A with high deformation resistance, the work roll diameter is φ560, and the acceleration / deceleration pattern is II. The maximum sheet thickness deviation over the entire coil length is 10.0 μm, and the required sheet thickness accuracy cannot be satisfied. The cause is considered to be the same as that of Comparative Example 2.

[0065] Comparative Example 5 has the same sheet thickness control as Comparative Example 1. The steel sheet is Steel Grade B with low deformation resistance, the work roll diameter is φ530, and the acceleration / deceleration pattern is I. The maximum sheet thickness deviation over the entire coil length is 7.0 μm, and the required sheet thickness accuracy cannot be satisfied. The cause is considered to be the same as that of Comparative Example 1. Comparative Example 6 has the same sheet thickness control as Comparative Example 1. The steel sheet is Steel Grade B with low deformation resistance, the work roll diameter is φ530, and the acceleration / deceleration pattern is II. The maximum sheet thickness deviation over the entire coil length is 10.0 μm, and the required sheet thickness accuracy cannot be satisfied. The cause is considered to be the same as that of Comparative Example 2. Comparative Example 7 has the same sheet thickness control as Comparative Example 1. The steel sheet is Steel Grade B with low deformation resistance, the work roll diameter is φ560, and the acceleration / deceleration pattern is I. The maximum sheet thickness deviation over the entire coil length is 7.0 μm, and the required sheet thickness accuracy cannot be satisfied. The cause is considered to be the same as that of Comparative Example 1. Comparative Example 8 has the same sheet thickness control as Comparative Example 1. The steel sheet is Steel Type B with low deformation resistance, the work roll diameter is φ560, and the acceleration / deceleration pattern is II. The maximum sheet thickness deviation over the entire coil length is 10.0 μm, and the required sheet thickness accuracy cannot be satisfied. The cause is considered to be the same as in Comparative Example 2.

[0066] In Comparative Examples 9 to 12, when created without including the work roll diameter and including the deformation resistance as explanatory variables of the prediction model and applied to cold rolling, the maximum sheet thickness deviation is 5.5 to 7.0 μm, and the required sheet thickness accuracy cannot be satisfied. The reason is considered to be that the work roll diameter is not included in the explanatory variables of the prediction model, resulting in low prediction accuracy of the change amount of the friction coefficient.

[0067] In Comparative Examples 13 to 16, when created without including the deformation resistance and including the work roll diameter as explanatory variables of the prediction model and applied to cold rolling, the maximum sheet thickness deviation is 5.5 to 7.0 μm, and the required sheet thickness accuracy cannot be satisfied. The reason is considered to be that the deformation resistance is not included in the explanatory variables of the prediction model, resulting in low prediction accuracy of the change amount of the friction coefficient.

[0068]

Table 1

[0069] From the above, it was confirmed that predicting the change amount of the friction coefficient during rolling with high accuracy using the cold rolling method for steel sheets and the manufacturing method for cold rolled steel sheets according to the present invention is useful for realizing high sheet thickness accuracy. Moreover, it can not only contribute to improving quality and yield, but also contribute to reducing the amount of energy used in the manufacturing process by omitting operations such as cutting defective sheet thickness parts on a separate line.

[0070] The embodiments to which the invention made by the present inventors has been applied have been described above. However, the present invention is not limited by the description and drawings that form part of the disclosure of the present invention according to this embodiment. That is, all other embodiments, examples, and operation techniques made by those skilled in the art based on this embodiment are included in the scope of the present invention.

Explanation of Signs

[0071] S: Steel plate 1: Payoff reel 2: Joining device 3: Looper 4: Cold tandem rolling mill 5: Cutting machine 6: Tension reel 7: Tensiometer 8: Thickness gauge 9: Arithmetic unit 10: Operation information device 91: Arithmetic device 92: Input device 93: Storage device 94: Output device 95: Bus wiring 96: RAM 97: ROM 97A: Prediction model 97B: Prediction model execution program 98: Arithmetic processing unit 98A: Information reading unit 98B: Data preprocessing unit 98C: Roll peripheral speed determination unit 98D: Result output unit

Claims

1. A cold rolling method for a steel sheet, comprising controlling the roll peripheral speed in a stand of a cold rolling mill when cold rolling the steel sheet by using a prediction model for predicting the friction coefficient between a work roll of the cold rolling mill and the steel sheet, wherein the prediction model is generated by using, as explanatory variables, rolling performance data and, as the objective variable, an estimated value of the friction coefficient when the steel sheet was cold rolled in the past, the rolling performance data includes the deformation resistance of the steel sheet and the work roll diameter, including the step of continuously predicting the friction coefficient between the work roll of the cold rolling mill and the steel sheet by inputting the rolling conditions of the steel sheet to be rolled into the prediction model, and controlling the roll peripheral speed of the cold rolling mill based on the change in the predicted friction coefficient. A cold rolling method for a steel sheet.

2. A method for manufacturing a cold-rolled steel sheet, comprising the step of manufacturing the cold-rolled steel sheet by using the cold rolling method for a steel sheet according to claim 1.

3. A cold rolling mill for cold rolling a steel sheet, a control device for controlling the control amount of the roll peripheral speed of the cold rolling mill, provided with the control device is configured to control the roll peripheral speed in a stand of the cold rolling mill when cold rolling the steel sheet by using a prediction model for predicting the friction coefficient between a work roll of the cold rolling mill and the steel sheet, wherein the prediction model is generated by using, as explanatory variables, rolling performance data and, as the objective variable, an estimated value of the friction coefficient when the steel sheet was cold rolled in the past, the rolling performance data includes the deformation resistance of the steel sheet and the work roll diameter, and is configured to continuously predict the friction coefficient between the work roll of the cold rolling mill and the steel sheet by inputting the rolling conditions of the steel sheet to be rolled into the prediction model, and control the roll peripheral speed of the cold rolling mill based on the change in the predicted friction coefficient. Manufacturing equipment for cold-rolled steel sheets.

Images