METHOD FOR COLD ROLLING STEEL SHEET, MANUFACTURING METHOD FOR COLD-ROLLED STEEL SHEET, AND MANUFACTURING FACILITY FOR COLD-ROLLED STEEL SHEET

Description

CROSS REFERENCE TO RELATED APPLICATIONS

This is the U.S. National Phase application of PCT/JP2023/034283, filed Sep. 21, 2023 which claims priority to Japanese Patent Application No. 2022-185786, filed Nov. 21, 2022, the disclosures of these applications being incorporated herein by reference in their entireties for all purposes.

FIELD OF THE INVENTION

The present invention relates to a method for cold rolling a steel sheet, a manufacturing method for a cold-rolled steel sheet, and a manufacturing facility for a cold-rolled steel sheet.

BACKGROUND OF THE INVENTION

In a cold rolling process, a defect called edge cracking, which is minute cracks, may occur in both end surfaces of a steel sheet in a width direction during rolling. In recent years, there has been an increase in demand for a high-tensile steel sheet that contributes to higher strength and weight reduction of members and an electrical steel sheet that is essential for electrification of automobiles. However, in a rolling process for these steel sheets, the edge cracking is likely to occur due to a high rolling load in addition to material brittleness. The edge cracking is a serious defect that not only decreases the quality and yield of the steel sheet but also leads to a decrease in productivity and damage to a facility due to breakage of the steel sheet. In view of such a background, a cold rolling method has been proposed to suppress the edge cracking. Specifically, Patent Literature 1 describes a method of rolling a steel sheet by setting a draft at an end portion in a width direction and a work roll shape in each pass so that a ratio ε_ye/ε_yc of distortion ε_ye in a sheet thickness direction corresponding to a draft at the end portion in the width direction of a steel sheet, to distortion ε_yc in the sheet thickness direction corresponding to a draft at a center portion in the width direction of the steel sheet is equal to or larger than an edge cracking limit value that increases as the pass progresses.

Meanwhile, in the cold rolling process, regardless of the type of rolling mill such as a tandem mill having a plurality of rolling stands and a reverse mill of a single rolling stand, rolling speeds of front and tail end portions of a coil are slower than the rolling speed of a steady portion of the coil, and a rolling condition such as a rolling load changes depending on a rolling time (see FIG. 7). Therefore, in order to suppress the edge cracking in the entire cold rolling region, an operation coping with a fluctuation in the rolling condition during rolling is required. Note that the entire cold rolling region means a rolling area over the entire length of the coil, including both a steady rolling region in which rolling is performed at a constant speed and an unsteady rolling region which is an acceleration/deceleration region. Therefore, a method has been proposed to set a control amount of a shape control unit, a forward tension, and a backward tension so that the shape of a steel sheet during rolling matches a target shape (see Patent Literature 2). This method uses a mathematical model representing a percentage elongation difference with a rolling load, the control amount of the shape control means, an amount of material crown, a shape before rolling, and a work roll crown, as variables, and a mathematical model representing the rolling load with the forward tension and the backward tension as variables. In addition, in order to suppress edge cracking in the entire cold rolling region, a method has been proposed in which a prediction formula representing the tension of the sheet end portion on the delivery side of the mill is used to adjust a control amount of a shape control actuator so that the tension of the sheet on the delivery side is equal to or less than a fracture limit value indicating occurrence of fracture of the steel sheet (see Patent Literature 3).

PATENT LITERATURE

• Patent Literature 1: JP 2015-100796 A
• Patent Literature 2: JP 2009-22985 A
• Patent Literature 3: JP 2017-164796 A

SUMMARY OF THE INVENTION

However, when the method described in Patent Literature 1 is applied to a rolling mill having no work roll shift function, an optimized tapered work roll cannot be applied to a steel sheet other than a steel sheet having a specific sheet width. In the cold rolling process, it is common that there are steel sheets of various sheet widths even in the same steel grade. Therefore, the method described in Patent Literature 1 lacks versatility. When the draft at the end portion in the width direction of the steel sheet is small, elongation of the steel sheet in a longitudinal direction at the end portion in the width direction is smaller than elongation of the steel sheet in a longitudinal direction at the center portion in the width direction. Then, a tensile stress is generated at the end portion in the width direction of the steel sheet due to restraint from the center portion in the width direction of the steel sheet, and when the elongation at the end portion in the width direction of the steel sheet due to the tensile stress exceeds a tensile elongation at break, the edge cracking occurs. However, it is considered that the tensile stress generated at the end portion in the width direction of the steel sheet is affected not only by a difference in elongation between the center portion in the width direction and the end portion in the width direction of the steel sheet but also by the magnitude of total tension. Furthermore, during rolling, the rolling condition such as a rolling load changes from moment to moment. Therefore, it can be said that the method described in Patent Literature 1 is insufficient to stably suppress the occurrence of the edge cracking over the entire length of the coil.

Meanwhile, in the method described in Patent Literature 2, the mathematical model using, as variables, the rolling load, the control amount of the shape control means, the amount of the material crown, the shape before rolling, and the work roll crown, which can be factors of changing the shape after rolling, is used for controlling, and the target shape is maintained over the entire length of the coil. However, when the purpose is to suppress the occurrence of the edge cracking, it is more appropriate to use, as a variable, tension (hereinafter, referred to as edge tension) acting on the end portion in the width direction of the steel sheet, instead of the sheet shape. In addition, in the method described in Patent Literature 3, for the rolling load constituting the prediction formula representing the edge tension, a predicted value based on a correlation between the rolling speed and a friction coefficient is used for at least the unsteady rolling region. However, in the unsteady rolling region where the rolling load significantly changes, a fluctuation in edge tension also increases, and the edge cracking is likely to occur. Therefore, it is desirable to use an actually measured value rather than the predicted value, as the rolling load used for prediction of the edge tension in the unsteady rolling region, from the viewpoint of accuracy.

Aspects of the present invention have been made to solve the above problems, and an object of aspects of the present invention is to provide a method for cold rolling a steel sheet, a manufacturing method for a cold-rolled steel sheet, and a manufacturing facility for a cold-rolled steel sheet that are configured to stably suppress occurrence of edge cracking over the entire length of a coil regardless of the type of a rolling mill or a shape control actuator.

A method for cold rolling a steel sheet according to aspects of the present invention controls a control amount of a shape control actuator of a cold mill during cold rolling of the steel sheet by using a prediction model for prediction of tension at an end portion in a width direction of the steel sheet on a delivery side of the cold mill, the prediction model is generated with actual rolling data obtained in cold rolling of a steel sheet in a past, as an explanatory variable, and an estimated value of the tension at the end portion in the width direction of the steel sheet on the delivery side of the cold mill, as an objective variable, the actual rolling data includes actual rolling load data appropriately acquired in an entire cold rolling region, and the method includes a step of: continuously predicting tension at an end portion in a width direction of a steel sheet to be rolled on the delivery side of the cold mill by inputting a rolling condition for the steel sheet to be rolled into the prediction model; and controlling the control amount of the shape control actuator such that the predicted tension has a preset target tension.

The actual rolling data may include at least one of: data about change in hot rolling crown in a longitudinal direction of the steel sheet; and data about temporal change in thermal crown of a work roll.

The tension at the end portion in the width direction of the steel sheet on the delivery side of the cold mill predicted by the prediction model may be a tension at a position set in advance within a range of 2 to 15 mm inward from the end portion in the width direction of the steel sheet.

A manufacturing method for a cold-rolled steel sheet according to aspects of the present invention includes a step of manufacturing a cold-rolled steel sheet by using the method for cold rolling a steel sheet according to aspects of the present invention.

A manufacturing facility for a cold-rolled steel sheet according to aspects of the present invention includes: a cold mill configured to perform cold rolling of a steel sheet; and a control device configured to control a control amount of a shape control actuator of the cold mill by using the method for cold rolling a steel sheet according to aspects of the present invention.

The method for cold rolling a steel sheet, the manufacturing method for a cold-rolled steel sheet, and the manufacturing facility for a cold-rolled steel sheet according to aspects of the present invention enable to stably suppress occurrence of edge cracking over the entire length of a coil, regardless of the type of the rolling mill or the shape control actuator.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram illustrating a configuration of a manufacturing facility for a cold-rolled steel sheet according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a configuration of an arithmetic unit.

FIG. 3 is a diagram illustrating a function of a prediction model illustrated in FIG. 1.

FIG. 4 is a graph illustrating an example of crown shapes in hot rolling at front and tail end portions and a steady portion of a steel sheet.

FIG. 5 is a graph illustrating the change in thermal crown with the passage of a rolling time.

FIG. 6 is a flowchart illustrating a procedure of a bender condition control process according to an embodiment of the present invention.

FIG. 7 is a graph illustrating an example of the change in rolling speed and rolling load with the passage of a rolling time.

DETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION

Hereinafter, a method for cold rolling a steel sheet, a manufacturing method for a cold-rolled steel sheet, and a manufacturing facility for a cold-rolled steel sheet according to an embodiment of the present invention will be described in detail with reference to the drawings. Note that component elements in the embodiments described below include components that can be readily replaced by a person skilled in the art or components that are substantially the same.

[Configuration]

First, a configuration of the manufacturing facility for a cold-rolled steel sheet according to an embodiment of the present invention will be described with reference to FIG. 1.

FIG. 1 is a schematic diagram illustrating a configuration of the manufacturing facility for a cold-rolled steel sheet according to an embodiment of the present invention. As illustrated in FIG. 1, the manufacturing facility for a cold-rolled steel sheet (hereinafter, abbreviated as manufacturing facility) according to an embodiment of the present invention is a continuous tandem rolling mill having a plurality of rolling stands. The manufacturing facility includes a pay-off reel 1, a joining device 2, a looper 3, a tandem cold mill 4, a cutting machine 5, a tension reel 6, and an arithmetic unit 7. In the present embodiment, the arithmetic unit 7 and a work roll bender control device, which is described later, function as a control device according to aspects of the present invention.

The pay-off reel 1 is a device for delivering a steel sheet S. Note that the manufacturing facility may include a plurality of the pay-off reels 1. In this configuration, the plurality of the pay-off reels 1 dispenses different steel sheets S.

The joining device 2 is a device that joins a tail end portion of a steel sheet (preceding material) previously delivered from the pay-off reel 1 and a front end portion of the steel sheet (following material) delivered later from the pay-off reel 1 to form a joined steel sheet. For the joining device 2, a laser welding machine is preferably used.

The looper 3 is a device that stores the steel sheet S so as to enable cold rolling of the steel sheet S by the tandem cold mill 4 until the end portions of the steel sheets S are joined by the joining device 2 (until completion of the joining).

The tandem cold mill 4 is a device for cold rolling of the steel sheet S so that the steel sheet S has a sheet thickness as a target sheet thickness. In the present embodiment, the tandem cold mill 4 includes five rolling stands of a first rolling stand to a fifth rolling stand (#1std to #5std) in order from an entry side (right side of FIG. 1) toward a delivery side (left side of FIG. 1) of the steel sheet S. In the tandem cold mill 4, a tension roll, a deflector roll, and a thickness gauge, which are not illustrated, are appropriately installed between adjacent rolling stands.

In the present embodiment, a shape control actuator of the tandem cold mill 4 is only a work roll bender, and no other shape control actuator such as work roll shift is provided. However, the type of the shape control actuator is not particularly limited. In addition, the configuration of each of the rolling stands, a conveyor for the steel sheet S, and the like are not particularly limited. Furthermore, in the present embodiment, the tandem cold mill 4 is a 4Hi rolling mill having four rolls in one rolling stand, but is not limited to this form, and another form such as 6Hi rolling stand can also be applied.

The cutting machine 5 is a device that cuts the steel sheet S after cold rolling.

The tension reel 6 is a device that winds up the steel sheet S cut by the cutting machine 5. The form of the tension reel 6 is not limited and may be, for example, a carousel tension reel. Furthermore, the manufacturing facility may include a plurality of the tension reels 6. In this configuration, the plurality of the tension reels 6 continuously wind up a plurality of the steel sheets S.

The arithmetic unit 7 uses a prediction model to predict a tension (edge tension) acting on an end portion in a width direction of a steel sheet S to be rolled at the delivery side of the cold mill 4, and controls a work roll bender condition for the tandem cold mill 4 upon cold rolling of the steel sheet S to be rolled on the basis of the predicted edge tension. The edge tension predicted using the prediction model may be a tension at the end portion in the width direction of the steel sheet S, but is preferably a tension at a position set in advance within a range of 2 to 15 mm inward from the end portion in the width direction of the steel sheet S, and more preferably within a range of 5 to 10 mm inward from the end portion in the width direction of the steel sheet S. Originally, the tension affecting the edge cracking is considered to be a tension acting on the end portion (outermost edge portion) in the width direction of the steel sheet S, but a result of analysis of the outermost edge portion in a shape analysis for estimation of the tension may be unstable. Therefore, the reliability of the tension or the like is further increased when the range of 2 to 15 mm inward from the end portion in the width direction of the steel sheet S is used. For further reliability of the tension, it is desirable to use the tension within the range of 5 to 10 mm inward from the end portion in the width direction of the steel sheet S, in consideration of instability of the result of the analysis for the tension in the vicinity of the end portion in the width direction of the steel sheet S and a fluctuation in edge tension that is smaller than an actual fluctuation in edge tension when a distance from the end portion in the width direction of the steel sheet S is increased. In addition, it is desirable to use, as the edge tension, a tension at a position in the vicinity of the end portion in the width direction as much as possible within a range in which the shape analysis data has reliability.

FIG. 2 is a block diagram illustrating a configuration of the arithmetic unit 7. As illustrated in FIG. 2, the arithmetic unit 7 includes an arithmetic device 71, an input device 72, a storage device 73, and an output device 74.

The arithmetic device 71 has a wired connection to the input device 72, the storage device 73, and the output device 74 via bus wiring 75. However, a connection form between the arithmetic device 71, the input device 72, the storage device 73, and the output device 74 is not limited to the wired connection, wireless connection may be used, and a form of combination between the wired connection and the wireless connection may be used.

The input device 72 functions as an input port to which information about control by the work roll bender control device and information from an operational information device 8 and an HOT information device 9 are input. The information from the operational information device 8 includes information about the steel sheet S to be rolled (steel grade, dimensions), cold rolling condition information set by a process computer or an operator before the cold rolling (numerical information, character information, and image information), and rolling condition information during the cold rolling (numerical information, character information, and image information). The HOT information device 9 includes information about hot rolling crown (sheet crown) of the steel sheet S to be rolled. The information about the sheet crown includes measurement data at a plurality of positions in the longitudinal direction of the steel sheet S, but it is more preferable for a measured pitch to be shorter.

The storage device 73 is a device that includes, for example, a known storage device such as a hard disk drive, a semiconductor drive, or an optical drive and that stores information necessary in the present system (information necessary for implementing a function of an arithmetic processing unit 78, which is described later).

The output device 74 functions as an output port that outputs a control signal from the arithmetic device 71 to the work roll bender control device.

The arithmetic device 71 includes RAM 76, ROM 77, and the arithmetic processing unit 78. The RAM 76, the ROM 77, and the arithmetic processing unit 78 are connected to the input device 72, the storage device 73, and the output device 74 via the bus wiring 75.

The RAM 76 is a working main memory used when the arithmetic device 71 performs any processing.

The ROM 77 stores a prediction model 77A and a prediction model execution program 77B. As illustrated in FIG. 3, when the prediction model 77A is generated, a shape analysis model is used first to estimate the edge tension (objective variable), on the basis of actual rolling data (explanatory variable) about the steel sheet S to be rolled. For the shape analysis model, a division model in which a roll barrel is divided into fine segments and a rolling load distribution and a contact pressure distribution between rolls are considered to be uniform in each segment may be used, or another model may be used. Then, the edge tension is estimated using a large possible amount of actual rolling data, and machine learning is performed using a set of the actual rolling data and an estimated value of the edge tension as training data, thereby generating the prediction model 77A to predict the edge tension on the basis of the rolling conditions for the steel sheet S to be rolled. The machine learning method can use a neural network model such as deep learning, a regression tree model such as random forest, a gradient boosting model such as XGBOOST, or the like, but is not particularly limited thereto, and another known machine learning method may be adopted.

In the present embodiment, a neural network is adopted as a machine learning method, and therefore, the prediction model 77A is a neural network model. The neural network model is represented by, for example, a function formula. Specifically, hyperparameters used for a machine learning model are set, and training by using the neural network model that uses the hyperparameters is performed. As optimization calculation for the hyperparameters, the neural network model in which some of the hyperparameters are changed stepwise is generated for the training data, selecting the hyperparameters providing the highest prediction accuracy for verification data is selected. The number of hidden layers, the number of neurons in each of the hidden layers, a dropout rate (blocking transmission through neurons with a certain probability) in each hidden layer, an activation function in each hidden layer, and the number of outputs are usually set as the hyperparameters, but are not limited thereto. Furthermore, a hyperparameter optimization method is not particularly limited, but grid research for stepwise change of parameters, random search for random selection of parameters, or search using Bayesian optimization can be used. Note that, in the present embodiment, the prediction model 77A takes the form of being stored in the ROM 77 in advance, but the form is not limited thereto, and for example, the arithmetic processing unit 78 may have a function of generating the prediction model 77A to generate the prediction model 77A online.

Here, parameters used for the explanatory variables of the prediction model 77A were examined. Specifically, in addition to parameters having been conventionally used in estimation of the edge tension such as draft distribution, rolling load, tension, roll diameter, and the control amount of the shape control actuator, influences of change in hot rolling crown in the longitudinal direction of the steel sheet and temporal change in thermal crown of the work roll were evaluated. First, the influence of the change in hot rolling crown in the longitudinal direction of the steel sheet on the fluctuation in edge tension was evaluated. A crown shape in hot rolling in the longitudinal direction of the steel sheet S was investigated at a plurality of points. As a result, in an investigated coil, it was confirmed that the crown shape in hot rolling is often different between the front and tail end portions and a steady portion. An example is illustrated in FIG. 4. Therefore, the edge tension on the delivery side of each rolling stand of the tandem cold mill 4 was estimated using, as a parameter used for the shape analysis, an example of the crown shapes in hot rolling at the front and tail end portions and the steady portion of the coil. As a result, a change in edge tension of approximately 30% at maximum could be confirmed from a difference in crown shape in hot rolling. From the past experience, this change in edge tension cannot be ignored as an influence on the occurrence of edge cracking. Therefore, it is considered that adding the change in hot rolling crown in the longitudinal direction of the steel sheet to the explanatory variables of the prediction model 77A provides a more effective prediction model to suppress occurrence of the edge cracking. In the present embodiment, information about the hot rolling crown in the steel sheet S is collected from the HOT information device 9 that stores data actually measured in a hot rolling mill, but the data collection method is not limited thereto. For example, data measured by a profilometer installed on the entry side of the tandem cold mill 4 may be used. When the data actually measured in the hot rolling mill is used, the edge tension can be estimated more accurately in consideration of a difference (trimmed amount) between a sheet width of the steel sheet (delivery side of hot rolling finishing) and a sheet width of the steel sheet on the entry side of the tandem cold mill 4, which are actually measured.

Next, the influence of the temporal change in thermal crown of the work roll on the fluctuation in edge tension was evaluated. The thermal crown is a phenomenon of thermal expansion of a roll R caused by processing heat during rolling, as illustrated in FIG. 5. The thermal crown grows rapidly as the temperature of the roll R is lower, that is, immediately after start of rolling. Therefore, a method of using a work roll in which the thermal crown is saturated by performing dummy rolling in advance may be adopted, but in some cases, rolling may be started for a steel sheet being unused due to steel sheet breakage during rolling or the like. The change in edge tension during rolling in this case was estimated by the shape analysis. As a result of calculation assuming that the conditions other than the thermal crown were constant, it was confirmed that the fluctuation in edge tension is approximately 60% at maximum during rolling. From the past experience, this change in edge tension cannot be ignored as an influence on the occurrence of edge cracking. Therefore, it is considered that adding the temporal change in thermal crown of the work roll to the explanatory variables of the prediction model 77A provides a more effective prediction model to suppress occurrence of the edge cracking. In the present embodiment, information about the thermal crown during rolling of the steel sheet S is collected from a thermal crown information device 10 in which a thermal crown prediction model based on actual measurement data is stored, but the prediction method for the thermal crown is not limited thereto, and a prediction model that is generated on the basis of a numerical analysis or the actual measurement data and numerical analysis results may be used. In addition, upon prediction of the thermal crown, consideration of the roughness of the work roll is considered to more accurately predict the thermal crown.

Returning to FIG. 2. The arithmetic processing unit 78 has an arithmetic processing function, and has a wired connection to the RAM 76 and the ROM 77 via the bus wiring 75. The arithmetic processing unit 78 determines the rolling conditions (work roll bender conditions) for the steel sheet S to be rolled. In order to perform the above processing, when receiving a signal notifying of performance of cold rolling from the work roll bender control device via the input device 72, the arithmetic processing unit 78 executes the prediction model execution program 77B stored in the ROM 77 to function as an information reading unit 78A, a data preprocessing unit 78B, a rolling state prediction unit 78C, a bender condition determination unit 78D, and a result output unit 78E.

The information reading unit 78A reads the information about control by the work roll bender control device and the information from the operational information device 8, which are obtained from the input device 72.

The data preprocessing unit 78B performs a process of generating data to be input to the rolling state prediction unit 78C. Specifically, in order to cause the prediction model 77A to read the actual rolling data, and the rolling conditions for the steel sheet S to be rolled, the data preprocessing unit 78B executes processing for conversion of units, deletion of coil data including abnormal data and unacquired data, compensation for data, or the like.

The rolling state prediction unit 78C inputs the input data generated in the data preprocessing unit 78B to the prediction model 77A to predict a sheet thickness distribution and tension distribution in the steel sheet S.

The bender condition determination unit 78D changes the work roll bender conditions in the rolling conditions for the steel sheet S to be rolled to have a value equal to or less than a predetermined edge cracking occurrence threshold, the edge cracking occurrence threshold being set in advance to indicate occurrence of edge cracking which differs depending on the steel grade and the dimensions of the steel sheet S, to repeat processing of the information reading unit 78A, the data preprocessing unit 78B, and the rolling state prediction unit 78C. In other words, the bender condition determination unit 78D uses the prediction model 77A to continuously predict the edge tension of the steel sheet S to be rolled, on the basis of the rolling conditions for the steel sheet S to be rolled, and controls each unit to determine the work roll bender conditions under which the predicted edge tension is equal to or less than the edge cracking occurrence threshold. The edge cracking occurrence threshold is a value that can be quantitatively represented using the edge tension being one of the objective variables of the prediction model 77A, and is determined on the basis of experimental results from an experiment in the tandem cold mill 4.

When the edge tension of the steel sheet S to be rolled predicted by the prediction model 77A is equal to or less than the edge cracking occurrence threshold, the result output unit 78E operates to output the determined work roll bender conditions to the work roll bender control device.

[Bender Condition Control Process]

Next, a procedure of a bender condition control process according to an embodiment of the present invention will be described with reference to FIG. 6.

FIG. 6 is a flowchart illustrating the procedure of the bender condition control process according to an embodiment of the present invention. The flowchart illustrated in FIG. 6 starts at the timing when a signal notifying of performance of cold rolling is input from the work roll bender control device via the input device 72, and the bender condition control process proceeds to processing of Step S1.

In the processing of Step S1, the information reading unit 78A reads the prediction model 77A stored in the ROM 77. Therefore, the processing of Step S1 is completed, and the bender condition control process proceeds to the processing of Step S2.

In the processing of Step S2, the information reading unit 78A reads data about the edge cracking occurrence threshold of the steel sheet S to be rolled, stored in the storage device 73. Therefore, the processing of Step S2 is completed, and the bender condition control process proceeds to the processing of Step S3.

In the processing of Step S3, the information reading unit 78A reads the current rolling conditions for the steel sheet S to be rolled, via the input device 72. Therefore, the processing of Step S3 is completed, and the bender condition control process proceeds to the processing of Step S4.

In the processing of Step S4, the rolling state prediction unit 78C inputs the rolling conditions read in the processing of Step S3, to the prediction model 77A read in the processing of Step S1, thereby predicting the edge tension of the steel sheet S to be rolled. Therefore, the processing of Step S4 is completed, and the bender condition control process proceeds to the processing of Step S5.

In the processing of Step S5, the bender condition determination unit 78D determines whether the edge tension predicted in the processing of Step S4 is equal to or less than the edge cracking occurrence threshold read in the processing of Step S2. As a result of the determination, when the edge tension is equal to or less than the edge cracking occurrence threshold (Step S5: Yes), the bender condition determination unit 78D finishes a series of steps of the bender condition control process. Meanwhile, when the edge tension is larger than the edge cracking occurrence threshold (Step S5: No), the bender condition determination unit 78D advances the bender condition control process to the processing of Step S6.

In the process of Step S6, the bender condition determination unit 78D changes the work roll bender conditions in the rolling conditions read in the processing of Step S3. Specifically, the bender condition determination unit 78D calculates an appropriate value for the work roll bender, on the basis of a difference between the edge tension predicted in the processing of Step S4 and the edge cracking occurrence threshold read in the processing of Step S2. Therefore, the processing of Step S6 is completed, and the bender condition determination unit 78D returns the bender condition control process to the processing of Step S3.

Returning to the processing of Step S3, the rolling state prediction unit 78C reads the rolling conditions in which the work roll bender conditions are changed. When the rolling conditions are read, in the processing of Step S4, the rolling state prediction unit 78C inputs the rolling conditions in which only the value for the work roll bender is changed, to the prediction model 77A to predict the edge tension. When the edge tension is predicted, in the process of Step S5, the bender condition determination unit 78D determines whether the predicted edge tension is equal to or less than the edge cracking occurrence threshold. Then, a series of processing of Steps S3 to S6 is repeatedly performed until the predicted edge tension becomes equal to or less than the edge cracking occurrence threshold. As described above, the arithmetic processing unit 78 uses the prediction model 77A to continuously predict the edge tension of the steel sheet S to be rolled, on the basis of the rolling conditions for the steel sheet S to be rolled, and determines the value for the work roll bender so that the predicted edge tension is equal to or less than the edge cracking occurrence threshold.

As is apparent from the above description, in the present embodiment, the prediction model 77A is generated first in which the actual rolling data obtained upon cold rolling of the steel sheet S in the past is set as the explanatory variable and the edge tension is set as the objective variable. Then, the arithmetic processing unit 78 uses the prediction model 77A to continuously predict the edge tension of the steel sheet S to be rolled, on the basis of the rolling conditions for the steel sheet S to be rolled. Then, the arithmetic processing unit 78 determines the work roll bender condition so that the predicted edge tension is equal to or less than the edge cracking occurrence threshold set in advance on the basis of the experimental results.

Therefore, rolling control that satisfies various restrictions in rolling operation regardless of the experience or subjective view of the operator is performed, enabling stable suppression of occurrence of edge cracking during rolling.

[Modifications]

Although the embodiments of the present invention have been described above, the present invention is not limited thereto, and various modifications and improvements can be made. For example, the devices included in the manufacturing facility are not limited to the above descriptions. Therefore, the rolling mill may be a reverse rolling mill instead of a tandem rolling mill. In addition, it is also possible to continue the cold rolling process and a pickling process which is a previous process to the cold rolling process, and a pickling device that pickles the steel sheet S may be arranged between the looper 3 and the tandem cold mill 4. In addition, in the present embodiment, the rolling mill is employed that includes the shape control actuator only including the work roll bender, and therefore, edge tension control by the work roll bender control is performed. However, in a case where a rolling mill including other shape control actuators is employed, the edge tension control may be also performed by fully using the shape control actuators. The tandem cold mill 4 is not limited to the 4Hi rolling mill, and may be a multi-stage rolling mill such as a 6Hi rolling mill, and the number of rolling stands is not particularly limited. In addition, the rolling mill may be a cluster rolling mill or a Sendzimir rolling mill.

In addition, when the arithmetic unit 7 calculates the control amount exceeding the upper or lower limit of a facility specification in an amount of change in the work roll bender or cannot calculate the control amount, it is preferable to display a warning screen on a monitor and issue an alarm, without performing the present implementation. Furthermore, in the present embodiment, the work roll bender control is automatically performed, but an appropriate amount of change in the work roll bender output from the arithmetic device 71 may be displayed on the monitor or the like so that the operator may manually change the work roll bender. Furthermore, aspects of the present invention can also be used only for setup, and in such a case, the value for the work roll bender during rolling has a constant value. For the data used as the explanatory variable, the cold rolling condition information set by the process computer or the operator before cold rolling is preferably used, and for hot rolling crown information, data of the steady portion (near the center portion in the longitudinal direction) of a hot-rolled coil is preferably used, and for the thermal crown information, a numerical value in a state in which the thermal crown is saturated is preferably used.

EXAMPLES

In the present example, a cold rolling experiment for a steel grade A (electrical steel sheet) having a base material thickness of 2.2 mm, a finished thickness of 0.35 mm, and a sheet width of 1300 mm was performed using the tandem cold mill 4 illustrated in FIG. 1. The used prediction model and a state of occurrence of edge cracking are shown in Tables 1-1 and 1-2. In the present example, as pretraining, learning using a machine learning model (deep learning) was performed with training data (past actual rolling data about approximately 3000 steel sheets) to generate a learning model that predicts edge tension on the basis of the rolling conditions for the steel sheet to be rolled. In the following inventive examples and comparative examples, this learning model was used. In addition, as a result of consideration of the edge cracking occurrence threshold of the steel grade A based on the actual rolling data about the steel grade A rolled in the past, it was estimated that edge cracking can be suppressed when the rolling stands after #4std have an edge tension of 180 MPa or less. Therefore, the edge cracking occurrence threshold was set to 180 MPa (on the delivery side of #4std and #5std).

In Inventive Example 1, for the actual rolling data as the explanatory variables, the deformation resistance, sheet thickness, sheet width, tension, load, work roll diameter, work roll shape, value for the work roll bender, and shape change coefficient (extension of the steel sheet in the width direction), which was experimentally adjusted in advance, were used for the steel grade A. For each work roll, a new work roll was used, and the surface temperature of the work roll immediately after starting rolling was the same as the room temperature. As a result of rolling 20 coils, minute edge cracking (to 1 mm) was confirmed at the front and tail end portions of three coils, but the remaining 17 coils had no occurrence of edge cracking.

In Inventive Example 2, for the actual rolling data as the explanatory variables, in addition to the actual rolling data used in Inventive Example 1, the change in hot rolling crown in the longitudinal direction of the steel sheet was used. For each work roll, a new work roll was used, and the surface temperature of the work roll immediately after starting rolling was the same as the room temperature. As a result of rolling 20 coils, minute edge cracking (to 1 mm) was confirmed in one coil immediately after starting the rolling, but the remaining 19 coils had no occurrence of edge cracking.

In Inventive Example 3, for the actual rolling data as the explanatory variables, in addition to the actual rolling data used in Inventive Example 1, the change in hot rolling crown in the longitudinal direction of the steel sheet was used. For each work roll, a work roll in which thermal crown was saturated by dummy rolling was used. In other words, the states of the work rolls used in Inventive Example 2 and Inventive Example 3 are different from each other. As a result of rolling 20 coils, edge cracking was able to be suppressed over the entire length in all the coils.

From the experimental results in Inventive Example 2 and Inventive Example 3, it was confirmed that the growth of the thermal crown during rolling may cause the occurrence of edge cracking. Therefore, it was confirmed that use of the temporal change in thermal crown as the explanatory variable in the learning model is effective for suppressing occurrence of the edge cracking, when there is a possibility of using a new work roll.

In Inventive Example 4, for the actual rolling data as the explanatory variables, in addition to the actual rolling data used in Inventive Example 1, the temporal change in thermal crown of the work roll was used. For each work roll, a new work roll was used, and the surface temperature of the work roll immediately after starting rolling was the same as the room temperature. In addition, the change in hot rolling crown in the longitudinal direction of the steel sheet is not used for the explanatory variable. As a result of rolling 20 coils, minute edge cracking (to 1 mm) was confirmed in two coils. In one coil, edge cracking was confirmed only at the front and tail end portions of the coil, and in the other one coil, edge cracking was confirmed over the entire length of the coil. In the investigation after rolling, in the two coils having occurrence of the edge cracking, one coil had significantly different crown shape in hot rolling at the front and tail end portions and the other one coil had significantly different crown shape in hot rolling over the entire length, as compared with a coil in which no edge cracking occurred.

In Inventive Example 3, edge cracking was able to be suppressed in all the coils rolled using the change in hot rolling crown in the longitudinal direction of the steel sheet as the explanatory variable. Therefore, it was confirmed that the change in hot rolling crown in the longitudinal direction of the steel sheet is desirably used as the explanatory variable, in order to cope with variation in the crown shape in hot rolling and stably suppress edge cracking.

In Inventive Example 5, for the actual rolling data as the explanatory variables, in addition to the actual rolling data used in Inventive Example 1, the change in hot rolling crown in the longitudinal direction of the steel sheet and the temporal change in thermal crown of the work roll were used. For each work roll, a new work roll was used, and the surface temperature of the work roll immediately after starting rolling was the same as the room temperature. As a result of rolling 20 coils, edge cracking was able to be suppressed over the entire length in all the coils.

In Inventive Examples 6 to 12, only the definition of the edge tension was changed in the actual rolling data of Inventive Example 5. Specifically, in Inventive Example 6, tension (edge 0 mm) at an end portion in the width direction of the steel sheet is defined as the edge tension. Note that in Inventive Examples 1 to 5, tension at a position 10 mm inward from the end portion in the width direction of the steel sheet is defined as the edge tension. As a result of rolling 20 coils, edge cracking of approximately 3 mm at maximum occurred in three coils. In Inventive Example 5, an amount of variation in the work roll bender during rolling was 13 ton/chock, whereas in Inventive Example 6, the amount of variation in the work roll bender was 15 ton/chock. It is considered that this is because the fluctuation in tension larger than the actual fluctuation in tension was predicted due to the end portion in the width direction of the steel sheet that is an unstable region to predict the tension, in the shape analysis model used for generating the prediction model.

In Inventive Example 7, tension at a position 2 mm inward from an end portion in the width direction of the steel sheet was defined as the edge tension. As a result of rolling 20 coils, minute edge cracking (to 1 mm) was confirmed in one coil.

In Inventive Example 8, tension at a position 5 mm inward from the end portion in the width direction of the steel sheet is defined as the edge tension. As a result of rolling 20 coils, edge cracking was able to be suppressed over the entire length in all the coils.

In Inventive Example 9, tension at a position 7 mm inward from the end portion in the width direction of the steel sheet was defined as the edge tension. As a result of rolling 20 coils, edge cracking was able to be suppressed over the entire length in all the coils.

In Inventive Example 10, tension at a position 13 mm inward from the end portion in the width direction of the steel sheet is defined as the edge tension. As a result of rolling 20 coils, minute edge cracking (to 1 mm) was confirmed in one coil.

In Inventive Example 11, tension at a position 15 mm inward from the end portion in the width direction of the steel sheet is defined as the edge tension. As a result of rolling 20 coils, minute edge cracking (to 1 mm) was confirmed in one coil.

In Inventive Example 12, tension at a position 17 mm inward from the end portion in the width direction of the steel sheet is defined as the edge tension. As a result of rolling 20 coils, edge cracking of 2 mm at maximum was confirmed in one coil. The edge cracking occurred in one coil, but the depth of the edge cracking was 2 mm at maximum, and therefore, it can be determined that the effect of suppressing the edge cracking is low as compared with Inventive Example 10 and Inventive Example 11.

From the experimental results of Inventive Examples 5 to 12, it was confirmed that when the position of the tension used for the edge tension is too close to the end portion in the width direction, the predicted value of the tension has an unstable value in the shape analysis model used this time, and appropriate bender control cannot be performed. On the other hand, it was confirmed that as the position of the tension used for the edge tension is apart from the end portion in the width direction, the fluctuation becomes smaller than the actual edge tension fluctuation, and appropriate bender control cannot be performed. In this condition, it was confirmed that the position of the tension used for the edge tension is preferably 2 to 15 mm (more preferably, 5 to 10 mm) inward from the end portion in the width direction of the steel sheet.

In Comparative Example 1, the edge tension control of aspects of the present invention was used only in the steady rolling region, and not used in the unsteady rolling region (acceleration/deceleration section). As a result of rolling 20 coils, edge cracking of 2 mm at maximum was confirmed in 13 coils. Generally, in rolling of the cold-rolled steel sheet, the steel sheet is rolled at a low speed immediately after starting the rolling, and then rolling is accelerated to high speed. Near the tail end of the coil, rolling is decelerated, and the steel sheet is rolled at low speed again. The rolling speed and the rolling load have a correlation, and the rolling load significantly changes in the acceleration/deceleration section. When the rolling load changes, the tension distribution in the steel sheet in the width direction also changes. As a result of investigation of a longitudinal position of the edge cracking in the coil in which the edge cracking occurred, after rolling, the position matched the unsteady rolling region. Therefore, it can be considered that when the edge tension control is not used in the unsteady rolling region, maintaining appropriate edge tension in the unsteady rolling region is made difficult, and the edge cracking occurred.

In Comparative Example 2, the edge tension control of aspects of the present invention was used only in the unsteady rolling region, and not in the steady rolling region. As a result of rolling 20 coils, edge cracking of 2 mm at maximum was confirmed in 14 coils. For the value for the work roll bender upon starting the rolling, an appropriate value for the bender was predicted and determined on the basis of past actual rolling of the steel grade A. As a result of confirmation of the longitudinal position of edge cracking, from the coil after rolling, some edge cracking occurred only in the front end portion of the coil and other edge cracking occurred also in the steady portion. It can be considered that the former was caused by the value for bender predicted from the past actual rolling was not appropriate and the latter was caused by operation not coping with the fluctuation in the rolling load in the longitudinal direction of the coil.

From Comparative Example 1 and Comparative Example 2, it was confirmed that application of edge tension control of aspects of the present invention to the entire cold rolling region (entire length of the coil) stably suppresses the edge cracking over the entire length of the coil.

In Comparative Example 3, shape control described in Patent Literature 2 was used without using the machine learning model, for the work roll bender control. As a result of rolling 20 coils, edge cracking of 2 mm at maximum occurred in 10 coils. It is considered that this is because the shape control was not able to cope with the change in the sheet thickness or crown shape in hot rolling, of the steel sheet during rolling, and the edge cracking occurred.

In Comparative Example 4, the shape control described in Patent Literature 3 was used without using the machine learning model, for the work roll bender control. For the rolling load in the steady rolling region, an actual value was used, and for the rolling load in the unsteady rolling region, a value obtained according to the prediction formula described in Patent Literature 3 was used. As a result of rolling 20 coils, edge cracking of 2 mm at maximum occurred in seven coils. As a result of investigation of the longitudinal position of the edge cracking from the coil after rolling, the edge cracking frequently occurred in the unsteady rolling region. It is considered as follows: the predicted value was used instead of the actual value for the rolling load in the unsteady rolling region, so that the edge tension could not be accurately predicted and the edge cracking occurred.

In Comparative Example 5, a setup value for the bender was determined by visually confirming the shape of the steel sheet by the operator without using the machine learning model for the work roll bender control, and a constant value for the work roll bender during rolling was used. As a result of rolling 20 coils, edge cracking of approximately 3 mm at maximum occurred in 18 coils.

In Comparative Example 6, a setup value for the work roll bender was determined on the basis of a result of numerical calculation without using the machine learning model for the work roll bender control, and a constant value for the work roll bender during rolling was used. As a result of rolling 20 coils, edge cracking of 4 mm at maximum occurred in all the coils. It is considered that this is because a result calculated before rolling under various assumptions was not able to suppress occurrence of edge cracking due to different points from actual rolling conditions.

In Comparative Example 7, an appropriate setup value for the work roll bender was predicted and determined from the past actual rolling of the steel grade A, without using the machine learning model for the work roll bender control. A constant value for the work roll bender during rolling was used. As a result of rolling 20 coils, edge cracking of approximately 2 mm at maximum occurred in 15 coils. It is considered that this is because occurrence of edge cracking was not able to be suppressed due to different points from actual rolling conditions, as in Comparative Example 6.

In Comparative Example 8, load interlocking control was used during rolling, in addition to the set up method for the work roll bender of Comparative Example 7. For a preliminary study, an amount of correction in the work roll bender that enables offsetting of the shape change due to fluctuation in load during rolling of the steel grade A was estimated using the shape analysis model, and used for the load interlocking control. As a result of rolling 20 coils, edge cracking of 2 mm at maximum occurred in 12 coils. It is considered that this is because when the setup value for the work roll bender predicted from the past actual rolling was not an appropriate setup value for the work roll bender, to the steel sheet to be rolled, and the edge cracking occurred. In addition, the load interlocking control is performed on the basis of the value, and therefore, the load linkage control cannot be appropriately used unless the setup value is correct. Furthermore, the bender value for correcting the shape change due to the change in rolling load constituting the load interlocking control is an estimation result under various assumptions, and therefore, it is considered that the accuracy is reduced as compared with the prediction model according to aspects of the present invention.

Here, the effect of suppressing edge cracking by the prediction model according to aspects of the present invention was verified for a coil having a very large amount of longitudinal fluctuation of the hot rolling crown generated with a probability of approximately 1%. The coils described in Table 1-1 and Table 1-2 did not include such a rare coil having a special shape. The used prediction model and the state of occurrence of edge cracking are shown in Table 1-3. In Inventive Examples 3-A, 3-B, 5-A, and 5-B shown below, the prediction model used in the rolling experiments shown in Table 1-1 and Table 1-2 was used.

In Inventive Example 3-A, for the actual rolling data as the explanatory variables, in addition to the actual rolling data used in Inventive Example 1, the change in hot rolling crown in the longitudinal direction of the steel sheet (longitudinal change in hot rolling crown) was used. Here, for the data about the longitudinal change in hot rolling crown, a value was used that was obtained by predicting a sheet crown shape on the entry side of the tandem cold mill, assuming a difference between an average value of a sheet width on the delivery side of hot rolling finishing and a process width on the entry side of the tandem cold mill, as the trimmed amount. This is the same model as the prediction model used in Inventive Example 3 shown in Table 1-1. Furthermore, for each work roll (WR), a work roll in which thermal crown was saturated by dummy rolling was used. As a result of rolling 100 coils, minute edge cracking (to 1 mm) was confirmed in one coil. When the crown shape in hot rolling of the rolled 100 coils was investigated, only one coil in which edge cracking occurred was a rare coil having a very large amount of longitudinal fluctuation of the hot rolling crown.

In Inventive Example 3-B, for the actual rolling data as the explanatory variables, in addition to the actual rolling data used in Inventive Example 1, the longitudinal change in hot rolling crown was used. Here, for the data about the longitudinal change in hot rolling crown, a value was used that accurately grasps the trimmed amount in the longitudinal direction of the coil and enables accurate estimation of the coil shape on the entry side of the tandem cold mill, for trimming performed between the delivery side of hot rolling finishing and the entry side of the tandem cold mill. In Inventive Example 3-A, the average value of sheet widths on the delivery side of hot rolling finishing was used, but the average value of sheet widths on the delivery side of hot rolling finishing has a fluctuation of several ten millimeters in the longitudinal direction, and therefore, the average value of sheet widths on the delivery side of hot rolling finishing is insufficiently used to predict the sheet crown shape on the entry side of the tandem cold mill. In the present example, the sheet crown shape on the entry side of the tandem cold mil was predicted, from the data of the sheet width measured on the delivery side of hot rolling finishing, the sheet width measured on the entry side of the tandem cold mill, and cut lengths at the front and tail end portions of the coil between the delivery side of hot rolling finishing and the entry side of the tandem cold mill. However, the prediction method for the sheet crown shape on the entry side of the tandem cold mill is not limited thereto, and for example, the sheet crown shape may be actually measured on the entry side of the tandem cold mill. Furthermore, for each work roll, a work roll in which thermal crown was saturated by dummy rolling was used. As a result of rolling 100 coils, edge cracking was able to be suppressed in all the coils. When the crown shape in hot rolling of the rolled 100 coils was investigated, one coil was a rare coil having a very large amount of longitudinal fluctuation of the hot rolling crown. However, accurate prediction of the sheet crown shape on the entry side of the tandem cold mill was able to suppress edge cracking, in response to this situation.

In Inventive Example 5-A, for the actual rolling data as the explanatory variables, in addition to the actual rolling data used in Inventive Example 1, the longitudinal change in hot rolling crown and the temporal change in thermal crown of the work roll were used. In addition, the method of Inventive Example 3-B was used for prediction of the sheet crown shape on the entry side of the tandem cold mill. For each work roll, a new work roll was used, and the surface temperature of the work roll immediately after starting rolling was the same as the room temperature. The same data as that used in Inventive Example 5 was used for prediction of the thermal crown. As a result of rolling 100 coils, minute edge cracking (to 1 mm) was confirmed in one coil. When the crown shape in hot rolling of the rolled 100 coils was investigated, only one coil in which edge cracking occurred was a rare coil having a very large amount of longitudinal fluctuation of the hot rolling crown.

In Inventive Example 5-B, for the actual rolling data as the explanatory variables, in addition to the actual rolling data used in Inventive Example 1, the longitudinal change in hot rolling crown and the temporal change in thermal crown of the work roll were used. In addition, the method of Inventive Example 3-B was used for prediction of the sheet crown shape on the entry side of the tandem cold mill. For each work roll, a new work roll was used, and the surface temperature of the work roll immediately after starting rolling was the same as the room temperature. For the prediction of the thermal crown, a model configured to predict the thermal crown in consideration of the roughness of the work roll was used. In the model used in Inventive Example 5-A, the roughness of the work roll was not taken into consideration to predict the thermal crown. As a result of rolling 100 coils, edge cracking was able to be suppressed in all the coils. When the crown shape in hot rolling of the rolled 100 coils was investigated, one coil was a rare coil having a very large amount of longitudinal fluctuation of the hot rolling crown. However, accurate prediction of the thermal crown enabled accurate prediction of the edge tension, for appropriate work roll bender control. As a result, edge cracking was suppressed.

From the above description, it was confirmed that accurate prediction of the sheet crown shape on the entry side of the tandem cold mill or the thermal crown by using the above method enables suppression of edge cracking, for the rare coil having a very large amount of longitudinal fluctuation of the hot rolling crown too.

From the above description, it was confirmed that appropriate prediction of edge tension of the steel sheet during rolling by using the manufacturing method and manufacturing facility for a cold-rolled steel sheet according to aspects of the present invention, and control of the work roll bender so as to be equal to or less than the edge cracking occurrence threshold are effective to suppressing edge cracking. In addition, it was confirmed that application of aspects of the present invention makes it possible not only to prevent a decrease in productivity and damage to the facility due to breakage of the steel sheet during rolling, but also to significantly contribute to improvement in quality and yield. Furthermore, a portion where edge cracking has occurred is cut on another line before a next process, and therefore, the edge cracking is also suppressed, contributing to reduction of the amount of energy used in the process of manufacturing.

Although the embodiments to which the invention made by the present inventors is applied have been described above, the present invention is not limited to the descriptions and drawings constituting part of the disclosure of the present invention according to the present embodiments. In other words, other embodiments, examples, operation techniques, and the like made by those skilled in the art and the like on the basis of the present embodiment are all included in the scope of the present invention.

TABLE 1-1
No.  1        2    3      4      5	Name     Inventive Example 1     Inventive Example 2 Inventive Example 3   Inventive Example 4   Inventive Example 5	Prediction model for edge tension       Used       Used   Used     Used     Used	Explanatory variable                  Deformation resistance, sheet thickness, sheet width, tension, load, work roll diameter, work roll shape, value for bender, shape change coefficient Inventive Example 1 + longitudinal change in hot rolling crown Inventive Example 1 + longitudinal change in hot rolling crown   Inventive Example 1 + temporal change in thermal crown of work roll Inventive Example 1 + longitudinal change in hot rolling crown + temporal change in thermal crown of work roll	Amount of variation in bender during rolling (#4std)     [ton/chock]      10       12   12     11     13	Number of coils in which edge cracking occurred 3/20       1/20   0/20     2/20     0/20	Edge cracking depth        to 1 mm       to 1 mm   —     to 1 mm     —	Remark                   Use new work roll       Use new work roll   Use work roll in which thermal crown is saturated by dummy rolling Use new work roll     Use new work roll
6    7    8    9   10   11   12	Inventive Example 6 Inventive Example 7 Inventive Example 8 Inventive Example 9 Inventive Example 10 Inventive Example 11 Inventive Example 12	Used (Edge tension: edge 0 mm) Used (Edge tension: edge 2 mm) Used (Edge tension: edge 5 mm) Used (Edge tension: edge 7 mm) Used (Edge tension: edge 13 mm) Used (Edge tension: edge 15 mm) Used (Edge tension: edge 17 mm)	Inventive Example 5   Inventive Example 5   Inventive Example 5   Inventive Example 5   Inventive Example 5   Inventive Example 5   Inventive Example 5	15   14   13   13   12   12   11	3/20   1/20   0/20   0/20   1/20   1/20   1/20	to 3 mm   to 1 mm   —   —   to 1 mm   to 1 mm   to 2 mm	Change definition of edge tension (Inventive Examples 1 to 5, Comparative Examples 1 and 2: edge 10 mm)

TABLE 1-2
				Amount of
				variation in	Number of
				bender during	coils in
		Prediction		rolling	which edge	Edge
		model for	Explanatory	(#4std)	cracking	cracking
No.	Name	edge tension	variable	[ton/chock]	occurred	depth	Remark

13	Comparative	Used	Inventive	2 (in each	13/20	to 2 mm
	Example 1	(only steady	Example 5	steady rolling
		rolling region)		region)
14	Comparative	Used	Inventive	10	14/20	to 2 mm	Predict setup
	Example 2	(only unsteady	Example 5				value from past
		rolling region)					actual rolling
15	Comparative	Not used	—	7	10/20	to 3 mm	shape control in
	Example 3						Patent Literature 2
16	Comparative	Not used	—	9	7/20	to 2 mm	shape control in
	Example 4						Patent Literature 3
17	Comparative	Not used	—	0 (only	18/20	to 3 mm	Visual
	Example 5			setup			confirmation
				control)			by operator
18	Comparative	Not used	—	0 (only	20/20	to 4 mm	Result of
	Example 6			setup			numerical
				control)			calculation
19	Comparative	Not used	—	0 (only	15/20	to 2 mm	Predicted from
	Example 7			setup			past actual
				control)			rolling
20	Comparative	Not used	—	9	12/20	to 2 mm	Comparative
	Example 8						Example 7 + Load
							interlocking
							control

TABLE 1-3
				Amount of
				variation in	Number of
		Prediction		bender during	coils in
		model		rolling	which edge	Edge
		for edge	Explanatory	(#4std)	cracking	cracking	WR to
No.	Name	tension	variable	[ton/chock]	occurred	depth	be used	Remark
1	Inventive	Used	Inventive Example 1	12	1/100	to 1 mm	WR in which
	Example		+longitudinal change				thermal crown
	3-A		in hot rolling crown				is saturated by
							dummy rolling
2	Inventive	Used	Inventive Example 1	13	0/100	—	WR in which	Consider accurate
	Example		+longitudinal change				thermal crown	trimmed amount
	3-B		in hot rolling crown				is saturated by
							dummy rolling
3	Inventive	Used	Inventive Example 1	13	1/100	to 1 mm	New WR	Consider accurate
	Example		+longitudinal change					trimmed amount
	5-A		in hot rolling crown
			+temporal change in
			thermal crown of
			work roll
4	Inventive	Used	Inventive Example 1	14	0/100	—	New WR	Consider accurate
	Example		+longitudinal change					trimmed amount,
	5-B		in hot rolling crown					and use “roughness
			+temporal change in					of work roll” as
			thermal crown of					parameter for
			work roll					thermal crown
								calculation

INDUSTRIAL APPLICABILITY

According to aspects of the present invention, it is possible to provide the method for cold rolling a steel sheet, the manufacturing method for a cold-rolled steel sheet, and the manufacturing facility for a cold-rolled steel sheet that are configured to stably suppress occurrence of edge cracking over the entire length of a coil regardless of the type of the rolling mill or the shape control actuator.

REFERENCE SIGNS LIST

• • 1 PAY-OFF REEL
  • 2 JOINING DEVICE
  • 3 LOOPER
  • 4 TANDEM COLD MILL
  • 5 CUTTING MACHINE
  • 6 TENSION REEL
  • 7 ARITHMETIC UNIT
  • 8 OPERATIONAL INFORMATION DEVICE
  • 9 HOT INFORMATION DEVICE
  • 10 THERMAL CROWN INFORMATION DEVICE
  • 71 ARITHMETIC DEVICE
  • 72 INPUT DEVICE
  • 73 STORAGE DEVICE
  • 74 OUTPUT DEVICE
  • 75 BUS WIRING
  • 76 RAM
  • 77 ROM
  • 77A PREDICTION MODEL
  • 77B PREDICTION MODEL EXECUTION PROGRAM
  • 78 ARITHMETIC PROCESSING UNIT
  • 78A INFORMATION READING UNIT
  • 78B DATA PREPROCESSING UNIT
  • 78C ROLLING STATE PREDICTION UNIT
  • 78D BENDER CONDITION DETERMINATION UNIT
  • 78E RESULT OUTPUT UNIT