JP2022059387A - Plate thickness control method, plate material manufacturing method, and plate thickness control device - Google Patents

Abstract

To provide a plate thickness control method, a plate material manufacturing method, and a plate thickness control device that can set a roll gap to an appropriate value at the time of roll biting-in.SOLUTION: In a plate thickness control method, a plate thickness of a plate material is controlled by controlling a position of a rolling reduction cylinder according to a target plate thickness of the plate material. The plate thickness control method includes: a prediction step in which a second parameter is predicted for a rolling object plate material, by inputting a first parameter of the rolling object plate material to a prediction model created from past first parameters and past second parameters therefor, by using a first parameter associated with a plate thickness of a previous-pass rolling-out side end of the plate material as an input variable, and by using a second parameter associated with a present-pass biting side end of the plate material as an output variable; and a roll gap setting step for setting a roll gap when the plate material is bitten between rolling rolls, on the basis of the predicted second parameter.SELECTED DRAWING: Figure 1

Description

The present invention relates to a plate thickness control method, a plate material manufacturing method, and a plate thickness control device.

A rolling method is known in which the rolling method for controlling the plate thickness of a plate material (for example, a thick steel plate) by adjusting the rolling amount of a hot rolling mill (hereinafter, simply referred to as “rolling machine”) is known. The load behavior of the end portion (tip portion, tail end portion) of the plate material during rolling is different from that of the stationary portion (part other than the end portion) of the plate material due to supercooling or the like. If the load fluctuates abruptly, the hydraulic cylinder of AGC (automatic plate thickness control) cannot follow, and the plate thickness at the end becomes thicker than that at the stationary portion.

In order to prevent the plate thickness at the end of the plate from becoming thicker than that at the stationary portion, conventionally, in addition to correcting the sinking of the hydraulic cylinder when the plate bites into the work roll of the rolling mill, the plate is rolled. The roll opening is controlled to be closed by the correction amount γ from the set roll opening before being bitten by the work roll of the machine and before a predetermined time α elapses after being bitten by the work roll (). For example, see Patent Document 1).

Japanese Unexamined Patent Publication No. 2019-18237

However, the correction amount γ is given as a table value according to the dimensions of the plate material, and the maintenance cost is large. Further, if the set correction amount γ is inappropriate, there is a possibility that a plate thickness defect may occur.

The present invention has been made in view of the above, and provides a plate thickness control method, a plate material manufacturing method, and a plate thickness control device capable of setting a roll opening degree at the time of roll engagement to an appropriate value. The purpose is.

In order to solve the above-mentioned problems and achieve the object, in the method of rolling a plate material according to the present invention, after the plate material is bitten between the rolling rolls, the position of the rolling cylinder is based on the target plate thickness of the plate material. This is a plate thickness control method for controlling the plate thickness of the plate material by controlling the above plate material, with the first parameter related to the plate thickness of the bite-through side end of the front path of the plate material as an input variable. For the prediction model generated from the first parameter in the past and the second parameter in the past, with the second parameter related to the plate thickness at the biting side end of the current path as the output variable. By inputting the first parameter of the plate material to be rolled, the plate material is the rolling roll based on the prediction step of predicting the second parameter for the plate material to be rolled and the predicted second parameter. Includes a roll opening setting step for setting the roll opening when bitten in between.

Further, in the method for rolling a plate material according to the present invention, in the above invention, the first parameter is the deformation resistance of the bite-through side end of the front pass, and the second parameter is the bite-in side of the current pass. Deformation resistance at the end.

Further, in the method for rolling a plate material according to the present invention, in the above invention, the first parameter is the deformation resistance of the bite-through side end of the front pass, and the second parameter is the bite load of the current pass. Is.

Further, in the method for rolling a plate material according to the present invention, in the above invention, the first parameter is the bite-through load of the front pass, and the second parameter is the deformation resistance of the bite-side end of the current pass. be.

Further, in the method for rolling a plate material according to the present invention, in the above invention, the first parameter is the bite-through load of the front pass, and the second parameter is the bite load of the current pass.

Further, in the method for rolling a plate material according to the present invention, in the above invention, in addition to the deformation resistance of the bite-through side end portion of the front pass, the material characteristics of the plate material and the bite-through load of the front pass are used as the input variables. Including at least one of them.

Further, in the above invention, the method for rolling a plate material according to the present invention includes at least one of the dimensions, temperature, actual components and rolling settings of the plate material as the material characteristics of the plate material.

In order to solve the above-mentioned problems and achieve the object, the plate material manufacturing method according to the present invention rolls the plate material by using the above-mentioned plate material rolling method.

In order to solve the above-mentioned problems and achieve the object, the plate rolling apparatus according to the present invention is a rolling machine having a rolling roll that bites and rolls the plate, and the plate is bitten between the rolling rolls. After that, a control unit for controlling the plate thickness of the plate material by controlling the position of the reduction cylinder based on the target plate thickness of the plate material is provided, and the control unit is engaged with the front pass of the plate material. The first parameter related to the plate thickness at the exit side end is used as an input variable, and the second parameter related to the plate thickness at the biting side end of the current pass of the plate material is used as an output variable. By inputting the first parameter of the plate material to be rolled into the prediction model generated from the parameter of and the second parameter in the past, the second parameter is predicted for the plate material to be rolled. Then, based on the predicted second parameter, the roll opening degree when the plate material is bitten between the rolling rolls is set.

According to the plate thickness control method and the plate thickness control device according to the present invention, the behavior of the end load different for each material is predicted in the roll opening setting during rolling of the plate material, and the roll bite is predicted based on the prediction result. The roll opening at the time of crowding can be set to an appropriate value. Further, according to the method for producing a plate material according to the present invention, it is possible to produce an excellent plate material without a defective plate thickness.

FIG. 1 is a diagram showing a schematic configuration of a plate thickness control device according to an embodiment of the present invention. FIG. 2 is a diagram showing an example of a hardware configuration for realizing a control unit of the plate thickness control device according to the embodiment of the present invention. FIG. 3 is a graph showing an example of input variables input to the prediction model in the plate thickness control device according to the embodiment of the present invention. FIG. 4 is an embodiment of the present invention, and is a graph showing the prediction accuracy of the deformation resistance of the end portion of the plate material in the conventional method. FIG. 5 is an embodiment of the present invention, and is a graph showing the prediction accuracy of the deformation resistance of the end portion of the plate material in the method of the present invention. FIG. 6 is an embodiment of the present invention, and is a graph comparing the biting load when the conventional method is used and the biting load when the method of the present invention is used. FIG. 7 is an embodiment of the present invention, and is a graph comparing the plate thickness of the end portion of the plate material when the conventional method is used and the plate thickness of the end portion of the plate material when the method of the present invention is used. ..

The plate thickness control method, the plate material manufacturing method, and the plate thickness control device according to the embodiment of the present invention will be described with reference to the drawings.

[Plate thickness control device]
The configuration of the plate thickness control device according to the embodiment of the present invention will be described with reference to FIG. The plate thickness control device 1 includes a rolling mill 10, a hydraulic cylinder (compression cylinder) 20, a load cell 30, a plate thickness control unit 50, a servo amplifier 60, and a servo valve 70.

The rolling mill 10 includes a pair of upper and lower work rolls (rolling rolls) 11 for rolling the plate material S, and a pair of upper and lower backup rolls 12 for reinforcing the work roll 11. Examples of the plate material S in the embodiment include thick steel plates and the like.

The hydraulic cylinder 20 is for adjusting the roll opening degree (roll gap) of the work roll 11. The hydraulic cylinder 20 adjusts the roll opening degree by moving the positions of the work roll 11 and the backup roll 12 based on the control of the servo valve 70.

The load cell 30 is for detecting the rolling load of the rolling mill 10. The load cell 30 is installed on the work roll 11 and the backup roll 12, respectively. Further, the load cell 30 outputs the detected rolling load to the plate thickness control unit 50.

The plate thickness control unit 50 performs automatic plate thickness control (AGC) after the plate material S is bitten into the work roll 11. That is, based on the rolling load detected by the load cell 30, the position of the hydraulic cylinder 20 is controlled so that the plate thickness deviation from the target plate thickness of the plate material S becomes zero, and the plate thickness of the plate material S is controlled.

Specifically, the plate thickness control unit 50 calculates the roll opening degree based on the information such as the rolling load input from the load cell 30. Then, the plate thickness control unit 50 outputs the calculated roll opening degree to the servo amplifier 60. As a result, a signal based on the roll opening degree is output from the servo amplifier 60 to the servo valve 70, and the pressure oil amount of the hydraulic cylinder 20 is controlled by the servo valve 70, so that the roll opening degree is adjusted.

An example of the hardware configuration for realizing the plate thickness control unit 50 will be described with reference to FIG. The plate thickness control unit 50 includes an information processing device 51, an input device 52, and an output device 53.

The information processing device 51 is composed of a general-purpose device such as a personal computer or a workstation, and includes a RAM 511, a ROM 512, and a CPU 513. The RAM 511 temporarily stores a processing program and processing data related to the processing executed by the CPU 513, and functions as a working area of the CPU 513.

The ROM 512 stores a control program 512a that executes the plate thickness control method according to the embodiment of the present invention, and a processing program and processing data that control the operation of the entire information processing apparatus 51.

The CPU 513 controls the operation of the entire information processing apparatus 51 according to the control program 512a and the processing program stored in the ROM 512.

The input device 52 is composed of devices such as a keyboard, a mouse pointer, and a numeric keypad, and is operated when inputting various information to the information processing device 51. The output device 53 is composed of a display device, a printing device, and the like, and outputs various processing information of the information processing device 51.

The plate thickness control device 1 according to the embodiment also functions as an apparatus for manufacturing the plate material S. In this case, the plate thickness control device 1 rolls the plate material S by the rolling mill 10 while controlling the plate thickness of the plate material S by using the plate thickness control method described above.

[Plate thickness control method]
The plate thickness control method by the plate thickness control device 1 described above will be described. Here, the roll opening degree of the finish rolling pass by the rolling mill 10 that affects the final shape (plate thickness) of the plate material can be determined by the following formula (1).

In the above formula (1), S is the roll opening, h is the thickness on the exit side, Sm'is the vertical extension of the mill, Se is the lateral extension of the mill, Rw is the roll wear, Scr is the crown, and S ₀ is the roll open at the time of zero adjustment. Degree, Off is an offset term, Wdte is a roll wear, exp is a roll expansion, gmho is a GM correction term, β is a bender correction term, Pb is a bender pressure, and Kb is a bender mill constant. Further, "Sm'-Se" in the above formula (1) is called mill elongation. This mill elongation is calculated from the biting load when the plate material S bites into the work roll 11. Further, the biting load can be obtained by the following formula (2).

In the above equation (2), P0 is the biting load, b is the plate width, ld is the contact arc length, kf is the deformation resistance at the end, and Qp is the rolling force function.

Conventionally, the deformation resistance kf of the above equation (2) has been manually calculated by simple regression from the conventional actual data, so that the maintenance cost is large. Further, if the set value of the deformation resistance kf is inappropriate, the roll opening degree may be set improperly, and the plate thickness of the plate material may be defective.

Therefore, in the plate thickness control method according to the embodiment, the deformation resistance kf of the biting side end portion of the current pass according to the actual biting load is predicted by a prediction model constructed in advance. Further, in addition to the deformation resistance of the bite-through side end of the front pass, the material characteristics of the plate material S, the bite-through load of the front pass, and the like are added to the input variables to improve the prediction accuracy of the bite load. Then, by calculating the roll opening using the prediction result, the roll opening for each material can be appropriately set. The "now pass" refers to the rolling path subject to plate thickness control, and the "previous pass" refers to the rolling path immediately before the rolling path subject to plate thickness control.

In the plate thickness control method according to the embodiment, the explanatory variables as shown in Table 1 below are used as learning data, and the bite of the current pass of the plate S with respect to the deformation resistance of the bite-through side end of the front pass of the plate S. A prediction model of the deformation resistance of the embedded side end is constructed, and the prediction result is used as the deformation resistance kf of the above equation (2). Then, the roll opening degree is calculated based on the above (1).

In Table 1 above, the "explanatory variable" indicates an input variable input when training the prediction model, and the "classification" indicates an attribute of the explanatory variable. Further, the "category: component" indicates the actual amount of the component contained in the plate material S.

FIG. 3 shows a more specific example of the explanatory variables in Table 1, in which the vertical axis shows the items of the explanatory variables and the horizontal axis shows the effect of each explanatory variable, that is, the degree of influence on the accuracy of the prediction model. Among the explanatory variables shown in the figure, "slab dimension_length, slab dimension_width, slab dimension_thickness" corresponds to "slab dimension" in Table 1. Further, "rolling command dimension_length, rolling command dimension_width, rolling command dimension_thickness" corresponds to "rolling dimension" in Table 1. Further, the "slab actual weight" corresponds to the "weight" in Table 1.

Further, among the explanatory variables shown in FIG. 3, "bender pressure setting_top, vendor pressure setting_bottom" corresponds to "bender pressure setting" in Table 1. Further, the "lower WR_diameter" corresponds to the "WR diameter" in Table 1. Further, the "front pass exit side thickness edge" corresponds to the "extrusion side thickness" in Table 1.

Hereinafter, a specific processing procedure of the plate thickness control method according to the embodiment will be described. In the plate thickness control method, a model generation step, a prediction step, and a roll opening degree setting step are performed. Of the above steps, the model generation step may be performed only once in advance. That is, once the prediction model is generated by the model generation step, only the prediction step and the roll opening degree setting step need to be performed as the plate thickness control method. Further, each step of the plate thickness control method according to the embodiment is mainly executed by the CPU 513 of the information processing apparatus 51 described above.

In the model generation step, the deformation resistance of the bite-through side end of the front pass of the plate material S is used as an input variable, and the deformation resistance of the bite-through side end of the current pass of the plate material S is used as an output variable. A predictive model is generated from the deformation resistance of the side end and the deformation resistance of the biting side end of the past now path.

In the model generation step, in addition to the deformation resistance of the bite-through side end of the front pass, the material characteristics of the plate material and the bite-through load of the front pass may be used as input variables. Examples of the above-mentioned "material properties of the plate material S" include the dimensions, temperature, actual components, rolling settings, etc. of the plate material S shown in Tables 1 and 3 described above.

In the model generation step, for example, regression forest, regression tree, random forest, DBM (Deep Boltzmann Machine), neural network (especially deep learning), gradient boosting, gradient boosting regression tree (Extreme Gradient Boosted Trees Regressor with early stopping). ), Gradient boosting regression tree AVG Blender, Elastic Net regression, and other machine learning methods to generate predictive models. It is preferable to use a regression forest as a learning method of the prediction model.

Subsequently, in the prediction step, the deformation resistance of the bite-through side end of the front pass of the plate material S to be rolled is input to the prediction model generated in the model generation step, so that the plate material to be rolled is now passed. Predict the deformation resistance of the biting side end.

Subsequently, in the roll opening degree setting step, the roll opening degree when the plate material S is bitten between the work rolls 11 is set based on the deformation resistance of the biting side end portion of the current pass predicted in the prediction step. Specifically, in the roll opening degree setting step, after the biting load P0 is calculated by the above formula (2), the roll opening degree is calculated by the above formula (1).

〔Example〕
Examples of the plate thickness control method according to the present invention will be described with reference to FIGS. 4 to 7.

FIG. 4 shows the prediction accuracy of the deformation resistance of the end portion of the plate material S (hereinafter referred to as “end deformation resistance”) when the conventional method (method of calculating the deformation resistance by simple regression from actual data) is carried out. It is a graph which shows. In the figure, the horizontal axis is the actual value of the end deformation resistance, and the vertical axis is the calculated value of the end deformation resistance calculated by the conventional method. As shown in the figure, the standard deviation σ in the conventional method was “10.7%”.

FIG. 5 is a graph showing the prediction accuracy of the end deformation resistance of the plate material S when the plate thickness control method according to the present invention (hereinafter referred to as “the method of the present invention”) is carried out. In the figure, the horizontal axis is the actual value of the end deformation resistance, and the vertical axis is the predicted value of the end deformation resistance predicted by the prediction model of the method of the present invention. As shown in the figure, the standard deviation σ in the conventional method was “3.8%”. Therefore, it can be seen that the prediction accuracy of the end deformation resistance is improved by using the method of the present invention.

Subsequently, FIG. 6 is an embodiment of the present invention, and is a graph comparing the biting load when the conventional method is used and the biting load when the method of the present invention is used. As shown in the figure, the biting load when the conventional method was used was "4211 ton", whereas the biting load when the method of the present invention was used was "4612 ton". Therefore, by using the method of the present invention, it is possible to close the roll opening degree by 0.29 mm as compared with the conventional method.

Subsequently, FIG. 7 is a graph comparing the plate thickness of the end portion of the plate material S when the conventional method is used and the plate thickness of the end portion of the plate material S when the method of the present invention is used. In the figure, the horizontal axis is the length of the plate material S (distance from the end of the plate material S), and the vertical axis is the plate thickness of the plate material S. Further, the solid line graph in the figure is the actual plate thickness of the plate material S measured by the laser plate thickness gauge after rolling. As shown in the figure, by closing the roll opening degree by 0.29 mm, it is possible to control the plate thickness at the end of the plate material S to be smaller.

According to the plate thickness control method and the plate thickness calculation device according to the above-described embodiment, the behavior of the end load different for each material is predicted in the roll opening setting during rolling of the plate material S, and based on the prediction result. Therefore, the roll opening degree at the time of rolling biting can be set to an appropriate value. That is, according to the plate thickness control method and the plate thickness calculation device according to the embodiment, the roll opening is corrected by the deformation resistance of the bite-through side end of the front pass, the material characteristics of the plate material S, and the bite-through load of the front pass. Since the amount (deformation resistance kf) is learned, it is possible to accurately set the appropriate roll opening without giving detailed instructions for each component system, temperature, and dimension, and the plate thickness of the plate material S is defective. It is possible to reduce it. Further, according to the method for producing a plate material S according to the present invention, it is possible to produce an excellent plate material S without a defective plate thickness.

Although the plate thickness control method, the plate material manufacturing method, and the plate thickness control device according to the present invention have been specifically described with reference to the embodiments and examples for carrying out the invention, the gist of the present invention is limited to these descriptions. It must be broadly interpreted based on the description of the scope of claims. Needless to say, various changes, modifications, etc. based on these descriptions are also included in the gist of the present invention.

For example, in the plate thickness control method, the plate material manufacturing method, and the plate thickness control device according to the embodiment, the deformation resistance kf of the above equation (2) was predicted by the prediction model, but even if the bite load P0 itself is predicted. good.

Further, in the plate thickness control method, the plate material manufacturing method, and the plate thickness control device according to the embodiment, "deformation resistance of the bite-through side end of the front pass" is used as an input variable, and "the bite-side end of the current pass" is used as an input variable. The prediction model was trained with "deformation resistance" as the output variable, but training may be performed by combining input variables and output variables other than the above.

For example, in the plate thickness control method, the plate material manufacturing method, and the plate thickness control device according to the embodiment, the deformation resistance of the bite-through side end of the front pass of the plate material is used as an input variable, and the bite load of the current pass of the plate material is used. As an output variable, the bite of the front pass of the plate material to be rolled with respect to the prediction model generated from the deformation resistance of the bite-through side end of the front pass in the past and the bite load of the current pass in the past. By inputting the deformation resistance of the pull-out side end portion, the biting load of the current pass may be predicted for the plate material to be rolled.

Further, in the plate thickness control method, the plate material manufacturing method, and the plate thickness control device according to the embodiment, the bite-through load of the front pass of the plate material is used as an input variable, and the deformation resistance of the bite-side end of the current pass of the plate material. As an output variable, for the prediction model generated from the past bite-through load of the previous pass and the deformation resistance of the past bite-side end of the current pass, the front pass of the plate material to be rolled By inputting the bite-through load, the deformation resistance of the bite-side end portion of the current pass may be predicted for the plate material to be rolled.

Further, in the plate thickness control method, the plate material manufacturing method, and the plate thickness control device according to the embodiment, the bite-through load of the front pass of the plate material is used as an input variable, and the bite load of the current pass of the plate material is used as an output variable. Rolling by inputting the bite-through load of the front pass of the plate material to be rolled into the prediction model generated from the bite-through load of the front pass in the past and the bite load of the present pass in the past. The biting load of the current pass may be predicted for the target plate material.

1 Plate thickness control device 10 Roller 11 Work roll 12 Backup roll 20 Hydraulic cylinder 30 Load cell 50 Plate thickness control unit 51 Information processing device 52 Input device 53 Output device 511 RAM
512 ROM
512a control program 513 CPU
60 Servo amplifier 70 Servo valve S Plate material

Claims (9)

A plate thickness control method for controlling the plate thickness of the plate material by controlling the position of the reduction cylinder based on the target plate thickness of the plate material after the plate material is bitten between the rolling rolls.
The first parameter related to the plate thickness at the bite-through side end of the front pass of the plate material is used as the input variable, and the second parameter related to the plate thickness at the bite-side end of the current pass of the plate material is used as the output variable. As for the plate material to be rolled, by inputting the first parameter of the plate material to be rolled into the prediction model generated from the first parameter in the past and the second parameter in the past. A prediction step for predicting the second parameter,
A roll opening setting step for setting the roll opening when the plate material is bitten between the rolling rolls based on the predicted second parameter, and a roll opening setting step.
Plate thickness control method including. The first parameter is the deformation resistance of the bite-through side end of the front pass.
The second parameter is the deformation resistance of the biting side end of the pass now.
The plate thickness control method according to claim 1. The first parameter is the deformation resistance of the bite-through side end of the front pass.
The second parameter is now the biting load of the pass,
The plate thickness control method according to claim 1. The first parameter is the bite-through load of the front pass.
The second parameter is the deformation resistance of the biting side end of the pass now.
The plate thickness control method according to claim 1. The first parameter is the bite-through load of the front pass.
The second parameter is now the biting load of the pass,
The plate thickness control method according to claim 1. The plate thickness according to claim 2, wherein the input variable includes at least one of the material characteristics of the plate material and the penetration load of the front pass, in addition to the deformation resistance of the bite-through side end of the front pass. Control method. The plate thickness control method according to claim 6, wherein the material properties of the plate material include at least one of the dimensions, temperature, actual components, and rolling settings of the plate material. A method for manufacturing a plate material, wherein the plate material is rolled by a rolling mill while controlling the plate thickness of the plate material by using the plate thickness control method according to any one of claims 1 to 7. A rolling mill having a rolling roll that bites and rolls a plate material,
After the plate material is bitten between the rolling rolls, a control unit that controls the plate thickness of the plate material by controlling the position of the reduction cylinder based on the target plate thickness of the plate material.
Equipped with
The control unit uses the first parameter related to the plate thickness of the bite-through side end of the front pass of the plate material as an input variable, and the second parameter related to the plate thickness of the bite-side end of the current pass of the plate material. By inputting the first parameter of the plate material to be rolled into the prediction model generated from the first parameter in the past and the second parameter in the past with the parameter of the above as an output variable. The second parameter is predicted for the plate material to be rolled, and the roll opening degree when the plate material is bitten between the rolling rolls is set based on the predicted second parameter.
Plate thickness control device.

Images