Method for predicting warpage shape of metal strip, method for controlling warpage shape of metal strip, method for manufacturing metal strip, method for generating warpage shape prediction model, and device for controlling warpage shape of metal strip

Abstract

Provided is a method for predicting a warpage shape of a metal strip, the method being capable of rapidly predicting an output-side warpage shape in consideration of an input-side warpage shape of the metal strip in a continuous annealing facility for the metal strip, the continuous annealing facility including a cooling zone for cooling the metal strip and a surface finishing mill for correcting the shape of the cooled metal strip. A method for predicting a warpage shape of a metal strip, which predicts an output-side warpage shape in a continuous annealing apparatus for the metal strip, the continuous annealing apparatus for the metal strip comprising: a heating apparatus for heating the metal strip; a cooling device (30) for cooling the metal strip (1) heated by the heating zone; a surface finishing mill (40) for correcting the shape of the metal strip (1) cooled by the cooling device; and an input-side warpage shape measuring device (16) that measures, between the cooling device (30) and the finishing mill (40), an input-side warpage shape, which is the warpage shape of the metal strip (1) on the upstream side of the finishing mill (40), and an output-side warpage shape, which is the warpage shape of the metal strip (1) on the downstream side of the finishing mill (40), the warpage shape prediction method of the metal strip predicts an output-side warpage shape based on at least one of an input-side warpage shape and an operating parameter of a surface finishing mill (40).

Description

Technical Field

[0001] The present invention relates to a method for predicting the warping shape of a metal strip, a method for controlling the warping shape of a metal strip, a method for manufacturing a metal strip, a method for generating a warping shape prediction model, and a device for controlling the warping shape of a metal strip. Background Art

[0002] In continuous annealing equipment for metal strips, it is important to control heat treatment conditions such as heating and cooling in order to give the metal strips to be treated the desired mechanical properties. In the manufacture of high-strength steel sheets, as a control of heat treatment conditions, tempering treatment is often performed on the metal strips after reheating and cooling in order to increase the cooling rate of the cooled metal strips and improve the stamping formability.

[0003] The continuous annealing equipment used to manufacture high-strength steel sheets has a heating zone, a soaking zone, and a cooling zone. The cooling methods in the cooling zone include liquid quenching, roller cooling, water-gas mixing (mist) cooling, gas jet cooling, etc., and an appropriate cooling method is appropriately selected to control the material of the metal strip. For example, in the case of manufacturing high-strength steel sheets with high tensile strength, it is effective to increase the cooling rate of the metal strip in the cooling zone. However, if the cooling rate of the metal strip is increased, there is a problem of poor shape of the metal strip due to the influence of the volume expansion caused by the thermal contraction of the metal strip and the phase change of the metal structure.

[0004] In the continuous annealing equipment, a temper rolling mill is arranged to correct the shape defects remaining after the metal strip is cooled. The shape defects of the metal strip after cooling can be corrected by temper rolling the metal strip using the temper rolling mill. On the other hand, if the tensile strength of the metal strip is high, the yield stress after cooling becomes very high, so even if the metal strip is temper rolled, the elongation required for shape correction may not be given. Therefore, in the case of a high tensile strength of the metal strip, the shape defects may remain in the metal strip that has become a product after temper rolling.

[0005] In this regard, Patent Document 1 discloses a technique in which, when a metal strip is cooled by a liquid quenching method, a metal strip during rapid cooling and quenching is restrained by restraining rollers in a temperature range determined by the temperature at which the martensitic transformation of the metal strip starts and ends. According to Patent Document 1, by restraining the metal strip in the above-mentioned temperature range, the shape change of the metal strip caused by rapid cooling can be suppressed.

[0006] Patent Document 2 discloses a method for controlling the shape of a metal strip by a rolling mill such as a skin-pass rolling mill, wherein the shape of the metal strip is measured using a shape meter, and a shape control actuator of the rolling mill is operated based on the deviation between the measured shape and the target shape. In the technology disclosed in Patent Document 2, a contact shape meter having a plurality of load cells embedded in a shape measuring roll is used to estimate the difference in elongation in the plate width direction of the metal strip based on the distribution of the contact load in the plate width direction.

[0007] Patent Document 3 discloses a technique of using work rolls having an average surface roughness exceeding 10.0 μm as a method of reducing the warpage height of a metal strip after temper rolling when temper rolling a high-strength steel sheet.

[0008] Prior art literature

[0009] Patent Literature

[0010] Patent Document 1: Japanese Patent No. 6094722

[0011] Patent Document 2: Japanese Patent No. 6673285

[0012] Patent Document 3: Japanese Patent Application Publication No. 2013-176802 Summary of the invention

[0013] Problems to be solved by the invention

[0014] The technology disclosed in Patent Document 1 requires determining the temperature region where the metal strip undergoes phase transformation and constraining the metal strip using constraining rollers in the temperature region. If the conveying speed of the metal strip changes as the operating conditions of the continuous annealing equipment, the cooling start temperature also changes, and the shape of the metal strip after cooling also changes. There is room for improvement in this regard. In particular, when the tensile strength of the metal strip is high, it is sometimes impossible to flatten the shape of the metal strip, and sometimes a portion with a poor shape remains in the longitudinal direction of the metal strip.

[0015] The technology disclosed in Patent Document 2 is a technology for flattening the shape of a metal strip by skin-pass rolling. It is possible to dynamically control the shape of the metal strip based on the shape of the metal strip measured at the output side of the skin-pass rolling mill, thereby stabilizing the shape of the metal strip in the length direction. However, the shape meter described in Patent Document 2 is a device for estimating the distribution of the extension in the plate width direction of the metal strip, and cannot measure the cross-sectional shape in the plate width direction that may occur in the high-strength steel plate after rapid cooling, which is a W-shaped or further approximated by a high-order function. Therefore, in the technology of Patent Document 2, it is impossible to control the cross-sectional shape in the plate width direction to be a W-shaped or further approximated by a high-order function.

[0016] The technology disclosed in Patent Document 3 is a technology for performing skin-pass rolling using work rolls having a specific surface roughness, and does not take into account the shape of the metal strip on the upstream side of the skin-pass rolling mill and the time-dependent change in the surface roughness of the work rolls. Therefore, when the warpage height of the metal strip on the upstream side of the skin-pass rolling mill increases, there is a case where the warpage shape cannot be sufficiently suppressed by following the shape, and the warpage height of the metal strip is maintained at a high state even after skin-pass rolling.

[0017] In order to solve these problems, it is effective to quickly predict the warping shape of the metal strip on the downstream side of the skin-pass rolling mill (hereinafter, this warping shape is described as the output-side warping shape) by considering the warping shape of the metal strip on the upstream side of the skin-pass rolling mill (hereinafter, this warping shape is described as the output-side warping shape), and control the operating conditions of the continuous annealing equipment based on the predicted output-side warping shape. However, there is a problem that there is no method that can quickly predict the warping shape of the metal strip on the downstream side of the skin-pass rolling mill by considering the input-side warping shape.

[0018] The present invention is completed to solve the above-mentioned problems, and its purpose is to provide a method for predicting the warping shape of a metal strip, a method for controlling the warping shape of a metal strip using the warping shape prediction method, a warping shape control device for a metal strip, and a method for manufacturing a metal strip, and a method for generating a warping shape prediction model for the warping shape prediction method for a metal strip, so that in a continuous annealing device for a metal strip including a cooling zone for cooling the metal strip and a surface finishing mill for correcting the shape of the metal strip after cooling, the warping shape of the output side can be quickly predicted based on the warping shape of the input side of the metal strip.

[0019] Means for solving problems

[0020] The gist of the present invention that can solve the above-mentioned problems is as follows.

[0021] [1] A method for predicting the warping shape of a metal strip, predicting the output side warping shape in a continuous annealing device for the metal strip, the continuous annealing device for the metal strip comprising: a heating device for heating the metal strip; a cooling device for cooling the metal strip after being heated by the heating device; a surface smoothing mill for correcting the shape of the metal strip after being cooled by the cooling device; and an input side warping shape measuring device, between the cooling device and the surface smoothing mill, measuring the warping shape of the metal strip on the upstream side of the surface smoothing mill, i.e., the input side warping shape, wherein the output side warping shape is the warping shape of the metal strip on the downstream side of the surface smoothing mill, wherein the method for predicting the warping shape of the metal strip predicts the output side warping shape based on the input side warping shape and at least one of the operating parameters of the surface smoothing mill.

[0022] [2] A method for predicting the warping shape of a metal strip according to [1], wherein the method for predicting the warping shape of a metal strip uses a warping shape prediction model to predict the output side warping shape, the warping shape prediction model takes as input data including the input side warping shape and at least one of the operating parameters of the surface smooth rolling mill, and takes as output the output side warping shape.

[0023] [3] According to the method for predicting the warping shape of a metal strip described in [1], the continuous annealing equipment further comprises a pre-cooling warping shape measuring device, and the pre-cooling warping shape measuring device is arranged on the upstream side of the cooling equipment. The method for predicting the warping shape of a metal strip predicts the output side warping shape based on the pre-cooling warping shape measured by the pre-cooling warping shape measuring device, the input side warping shape measured by the input side warping shape measuring device, and at least one of the operating parameters of the surface smooth rolling mill.

[0024] [4] A method for predicting the warping shape of a metal strip according to [3], wherein the method for predicting the warping shape of a metal strip uses a warping shape prediction model to predict the output side warping shape, the warping shape prediction model takes as input data including the warping shape before cooling, the input side warping shape and at least one of the operating parameters of the surface smooth rolling mill, and takes as output the output side warping shape.

[0025] [5] A method for controlling the warping shape of a metal strip, wherein, when the output side warping shape predicted by the warping shape prediction method of a metal strip described in any one of [1] to [4] exceeds a target value, the operating parameters of the cooling device are set in such a way as to reduce the output side warping shape.

[0026] [6] A method for controlling the warping shape of a metal strip according to [5], wherein the cooling device comprises: a cooling liquid injection device having a plurality of nozzles for injecting cooling fluid from both sides of the metal strip; and at least one pair of constraint rollers for constraining the metal strip, and the operating parameters of the cooling device are at least one of the amount of pressure of the pair of constraint rollers into the metal strip, the offset between the pair of constraint rollers, and the displacement of the pair of constraint rollers.

[0027] [7] A method for manufacturing a metal strip, using the warping shape control method of a metal strip described in [5] or [6] to manufacture a metal strip having a tensile strength of 980 MPa or more.

[0028] [8] A method for generating a warpage shape prediction model, predicting the output side warpage shape in a continuous annealing device for a metal strip, the continuous annealing device for the metal strip comprising: a heating device for heating the metal strip; a cooling device for cooling the metal strip after being heated by the heating device; a surface smoothing mill for correcting the shape of the metal strip after being cooled by the cooling device; an input side warpage shape measuring device, between the cooling device and the surface smoothing mill, for measuring the warpage shape of the metal strip on the upstream side of the surface smoothing mill, i.e., the input side warpage shape; and an output side warpage shape measuring device, at the downstream side of the surface smoothing mill, for measuring the warpage shape of the metal strip on the downstream side of the surface smoothing mill, i.e., the output side warpage shape. The output side warping shape is the warping shape of the metal strip at the downstream side of the surface smooth rolling mill, wherein the method for generating the warping shape prediction model obtains multiple data groups, and generates the warping shape prediction model by machine learning using the obtained multiple data groups as training data, the data group takes the input side warping shape measured by the input side warping shape measuring device, at least one of the operating parameters of the surface smooth rolling mill, and the output side warping shape measured by the output side warping shape measuring device as a group, the warping shape prediction model takes the input data including the input side warping shape and at least one of the operating parameters of the surface smooth rolling mill as input, and takes the output side warping shape as output.

[0029] [9] A method for generating a warp shape prediction model according to [8], wherein the continuous annealing equipment further comprises a pre-cooling warp shape measuring device, the pre-cooling warp shape measuring device being arranged on the upstream side of the cooling equipment, the method for generating a warp shape prediction model acquiring a plurality of data groups, and generating a warp shape prediction model by machine learning using the acquired plurality of data groups as training data, the data group comprising the pre-cooling warp shape measured by the pre-cooling warp shape measuring device, the input side warp shape, at least one of the operating parameters of the surface smooth rolling mill, and the output side warp shape as a group, the warp shape prediction model taking as input the input data including the pre-cooling warp shape, the input side warp shape, and at least one of the operating parameters of the surface smooth rolling mill, and taking as output the output side warp shape.

[0030]

[10] A warping shape control device for a metal strip, which controls the warping shape on the output side in a continuous annealing device for the metal strip, wherein the continuous annealing device for the metal strip comprises: a heating device for heating the metal strip; a cooling device for cooling the metal strip after being heated by the heating device; a surface smoothing mill for correcting the shape of the metal strip after being cooled by the cooling device; and an input side warping shape measuring device, which measures the warping shape of the metal strip on the upstream side of the surface smoothing mill, i.e., the input side warping shape, between the cooling device and the surface smoothing mill, wherein the output side warping shape is the warping shape of the metal strip on the downstream side of the surface smoothing mill, wherein the warping shape of the metal strip The control device comprises: an acquisition unit, which acquires the input side warping shape measured by the input side warping shape measuring device and at least one of the operating parameters of the surface smooth rolling mill; a warping shape prediction unit, which predicts the output side warping shape using a warping shape prediction model, the warping shape prediction model takes as input input data including the input side warping shape and at least one of the operating parameters of the surface smooth rolling mill, and takes as output the output side warping shape; and an operating parameter determination unit, which determines the operating parameters of the cooling device in a manner that reduces the output side warping shape when the output side warping shape predicted by the warping shape prediction unit exceeds a target value.

[0031]

[11] A warping shape control device for a metal strip according to

[10] , wherein the continuous annealing equipment further comprises a pre-cooling warping shape measuring device, the pre-cooling warping shape measuring device is arranged on the upstream side of the cooling equipment, the acquisition unit further acquires the pre-cooling warping shape measured by the pre-cooling warping shape measuring device, the warping shape prediction unit uses a warping shape prediction model to predict the output side warping shape, the warping shape prediction model takes as input data including the pre-cooling warping shape, the input side warping shape and at least one of the operating parameters of the surface smooth rolling mill, and takes as output the output side warping shape.

[0032] Effects of the Invention

[0033] According to the present invention, in a continuous annealing equipment for a metal strip including a heating device for heating the metal strip, a cooling device for cooling the metal strip, and a skin-pass rolling mill for correcting the shape of the cooled metal strip, the output-side warping shape of the skin-pass rolling mill can be rapidly predicted based on the input-side warping shape. Furthermore, by determining the operating parameters of the cooling device so as to reduce the predicted output-side warping shape, the operating parameters can be used as operating conditions of the cooling device, thereby realizing the production of a metal strip with a reduced warping height in the plate width direction. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] [ Figure 1 ] Figure 1 It is a schematic diagram showing a continuous annealing facility which is an example of a metal strip manufacturing facility capable of implementing the metal strip warping shape prediction method and the metal strip warping shape control method of the present embodiment.

[0035] [ Figure 2 ] Figure 2 This is a schematic diagram showing an example of a water-cooling type cooling device constituting a cooling zone.

[0036] [ Figure 3 ] Figure 3 This is a schematic diagram showing an example of arrangement of a pair of constraining rollers arranged on the front and back sides of a metal strip so as to be offset with respect to the running direction of the metal strip.

[0037] [ Figure 4 ] Figure 4 It is a schematic diagram showing an example of skin temper rolling equipment including a skin temper rolling mill.

[0038] [ Figure 5 ] Figure 5 It is a schematic diagram showing an example of the warping shape of the metal strip in the plate width direction.

[0039] [ Figure 6 ] Figure 6 This is a schematic diagram showing a state in which the warping shape of a metal strip is measured using a laser scanning laser rangefinder.

[0040] [ Figure 7 ] Figure 7 This is a schematic diagram showing a state where a warpage shape is measured using a plurality of laser scanning laser rangefinders.

[0041] [ Figure 8 ] Figure 8 It is a schematic diagram showing a structural example of a warping shape control device.

[0042] [ Fig. 9 ] Fig. 9 is a diagram showing an example of a machine learning model using a neural network.

[0043] [ Fig.10 ] Fig.10 This is a flowchart showing the flow of the warping shape control method of the metal strip according to the present embodiment. DETAILED DESCRIPTION

[0044] Hereinafter, the present invention will be specifically described by way of the embodiment of the present invention. The following embodiment is a preferred example of the present invention, and the present invention is not limited to these embodiments at all.

[0045] The warping shape prediction method and the warping shape control method of the metal strip according to the embodiment of the present invention can be implemented using a continuous annealing device, which includes: a heating device for heating a continuously conveyed metal strip; a cooling device for cooling the metal strip heated by the heating device; a skin-pass rolling mill for correcting the shape of the metal strip cooled by the cooling device; and an input-side warping shape measuring device for measuring the warping shape of the metal strip on the upstream side of the skin-pass rolling mill, i.e., the input-side warping shape, between the cooling device and the skin-pass rolling mill. The continuous annealing device is a device for manufacturing a metal strip by continuously annealing a metal strip having a plate thickness of 0.4 to 3.2 mm, a plate width of 700 to 1800 mm, and a length of 600 to 4000 m.

[0046] Figure 1 1 is a schematic diagram showing an example of a continuous annealing device 100 as a manufacturing device for a metal strip capable of implementing the warpage shape prediction method and the warpage shape control method of the metal strip of the present embodiment. The continuous annealing device 100 is roughly divided into an input side device 20, a furnace body device 21, and an output side device 24. The input side device 20 includes an unwinder 2, a welding machine 3, and an input side looper 4. The furnace body device 21 is composed of an annealing device 22 and a reheating device 23. The annealing device 22 includes a heating zone 6, a soaking zone 7, and a cooling zone 8, and sometimes includes a preheating zone 5 on the upstream side of the heating zone 6. The reheating device 23 includes a reheating zone 9, an over-aging zone 10, and a final cooling zone 11, and an induction heating device is arranged in the reheating zone 9. The output side device 24 includes an output side looper 12, an inspection table 14, and a tension reel 15.

[0047] The heating zone 6 is provided with a heating device for heating the metal strip 1, and the metal strip 1 is heated to a preset temperature in the range of about 600 to 900°C according to the composition of the metal strip 1. In the heating zone 6, a direct fire or radiation burner is used. The treatment performed in the heating zone 6 is called a heating process. The soaking zone 7 is provided with a device for maintaining the metal strip 1 at a predetermined temperature. The device for maintaining the metal strip 1 at a predetermined temperature is a device for supplementing the heating capacity of the furnace body heat dissipation and the like.

[0048] The cooling zone 8 is provided with a cooling device for cooling the metal strip 1 to a predetermined temperature. As cooling means in the cooling device, liquid cooling, gas jet cooling, roller cooling, spray cooling (gas-liquid mixed cooling), etc. are used. Liquid cooling is mostly performed by water cooling (water quenching) using water. Water cooling is a cooling means for cooling the metal strip 1 by immersing it in an immersion water tank provided on the downstream side of the uniform heating zone 7. Gas jet cooling is a cooling means for blowing gas from a nozzle onto the surface of the metal strip 1. Roller cooling is a cooling means for cooling the metal strip 1 by contacting it with a water-cooled roller. Spray cooling is a cooling means for spraying water into a fine mist and cooling it by absorbing its vaporization heat. In spray cooling, the size of the sprayed water droplets is mostly about 0.1 to 1 mm.

[0049] The reheating device 23 is arranged on the downstream side of the cooling zone 8. After the metal strip 1 is cooled to a predetermined temperature in the cooling zone 8, the metal strip 1 is reheated to a temperature of about 300 to 400°C using an induction heating device or the like arranged in the reheating zone 9. The over-aging zone 10 is a device for performing an over-aging treatment by holding the reheated metal strip 1 for a predetermined time. The final cooling zone 11 is a device for finally cooling the over-aged metal strip 1 to near room temperature. However, the reheating device 23 is not a necessary device for the continuous annealing device, and some continuous annealing devices may not have the reheating device 23.

[0050] The output side looper 12 is a device for temporarily storing the metal strip 1 in order to adjust the conveying speed of the metal strip 1 in the furnace body device 21 and the processing speed in the output side device 24. The inspection station 14 performs inspections on the dimensional accuracy and surface quality of the metal strip 1. The tension reel 15 is a device for winding the metal strip 1 into a coil. The metal strip 1 that is wound into a coil by the tension reel 15 and judged to be qualified by the quality inspection at the inspection station 14 may be shipped as a product coil or sent to a surface treatment device for plating the metal strip 1 and subjected to surface treatment. On the other hand, the metal strip 1 that is wound into a coil by the tension reel 15 and judged to be unqualified or retained by the quality inspection at the inspection station 14 is sent to a recoiling line to adjust the size or weight of the metal strip 1, collect samples for quality inspection, inspect the shape / size, rewind the coil, etc.

[0051] The skin-pass rolling equipment 13 is arranged between the output-side looper 12 and the inspection table 14. The work rolls used in the skin-pass rolling equipment 13 need to be replaced at a predetermined timing. By arranging the skin-pass rolling equipment 13 on the downstream side of the output-side looper 12, the output-side looper 12 can be used to suppress the speed change of the metal strip 1 in the furnace equipment 21 when the work rolls are replaced.

[0052] The continuous annealing equipment 100 includes an input side warpage shape measuring device 16. The input side warpage shape measuring device 16 is disposed between the cooling device of the cooling zone 8 and the surface temper rolling device 13, and measures the input side warpage shape of the metal strip 1. In addition, the continuous annealing equipment 100 may also include a pre-cooling warpage shape measuring device 17 and an output side warpage shape measuring device 18. The pre-cooling warpage shape measuring device 17 is disposed on the upstream side of the cooling device disposed in the cooling zone 8, and measures the warpage shape of the upstream side of the cooling device disposed in the cooling zone 8 of the metal strip 1 (hereinafter, this warpage shape is recorded as the pre-cooling warpage shape). The pre-cooling warpage shape measuring device 17 is preferably disposed between the soaking zone 7 and the cooling zone 8. Thus, the warpage shape of the metal strip 1 immediately before cooling can be measured.

[0053] The output-side warping shape measuring device 18 is provided on the downstream side of the surface-pass rolling equipment 13 to measure the output-side warping shape of the metal strip 1 after surface-pass rolling. The output-side warping shape measuring device 18 is preferably arranged near the inspection table 14. The output-side warping shape of the metal strip 1 sometimes becomes an inspection item for product quality. Therefore, by arranging the output-side warping shape measuring device 18 near the inspection table 14, it is easy to establish correspondence with the results of other quality inspections. The conveying distance of the metal strip 1 from the surface-pass rolling equipment 13 to the inspection table 14 is generally 20 to 100 m, and the conveying time from the surface-pass rolling equipment 13 to the inspection table 14 is about 2 to 120 seconds.

[0054] The continuous annealing equipment 100 includes a control computer 50 for managing the operation of the continuous annealing equipment 100, a control controller 51, and a warp shape control device 52. The control computer 50 and the warp shape control device 52 are both composed of general-purpose computers such as workstations and personal computers. For example, the control controller 51 uses a dedicated computer for executing sequence control such as a PLC (Programmable Logic Controller).

[0055] The control computer 51 sets the operating conditions of each device and equipment constituting the continuous annealing equipment 100, and obtains sensor information from various sensors installed in the continuous annealing equipment 100. The control computer 50 determines the manufacturing conditions of each device and equipment of the continuous annealing equipment 100 for manufacturing the metal strip 1 of the desired product quality. In addition, the control computer 50 collects the sensor information obtained by the control controller 51, obtains and saves various information related to the operating performance of the metal strip 1. In addition, the control computer 50 tracks the welded portion between the preceding metal strip 1 and the succeeding metal strip 1 connected by a welding machine or the like on the upstream side of the continuous annealing equipment 100, and determines the current position of the welded portion in the continuous annealing equipment 100.

[0056] The warp shape control device 52 predicts the output-side warp shape of the metal strip 1 based on the input-side warp shape measured by the input-side warp shape measuring device 16 and at least one of the operating parameters of the skin-pass mill 40. The warp shape control device 52 predicts the output-side warp shape of the metal strip 1 using a warp shape prediction model that takes as input input data including at least one of the input-side warp shape and the operating parameters of the skin-pass mill 40 and outputs the output-side warp shape. The warp shape control device 52 uses, for example, a machine learning model that has been learned by machine learning as the warp shape prediction model.

[0057] Next, the cooling zone 8 will be described. The cooling zone 8 in the continuous annealing equipment 100 includes a cooling device for cooling the metal strip 1 heated by the heating zone 6. The process performed in the cooling zone 8 is called a cooling process. The cooling method applicable to the cooling zone 8 is preferably liquid cooling using water as a refrigerant, that is, water cooling (water quenching). When the metal strip 1 heated by the heating zone 6 is cooled to a temperature lower than the martensitic transformation start temperature (Ms) or the bainitic transformation start temperature (Bf) by water cooling, the metal strip 1 is deformed and the warping shape of the metal strip 1 changes. In the case where the warping shape of the metal strip 1 changes like this, it often changes in the direction in which the warping height becomes higher.

[0058] Figure 2 Schematic diagram showing an example of a water-cooling cooling device 30 constituting the cooling zone 8. The cooling device 30 includes a cooling liquid spraying device 32 having a plurality of water spraying nozzles 64, an immersion water tank 34, and deflection rollers 36 and 37. In addition, the cooling device 30 may also include a pressing roller 31 for preventing the metal strip 1 from shaking. The metal strip 1 that has passed through the immersion water tank 34 is dewatered by a dehydrating roller 38 to remove water attached to the metal strip 1, and is dried in a drying furnace 39 before being sent to the reheating zone 9.

[0059] The immersion water tank 34 is configured in a container shape. The immersion water tank 34 stores a cooling fluid 35 for immersion cooling the metal strip 1. In order to suppress the temperature rise of the cooling fluid 35 stored in the immersion water tank 34 and maintain the water temperature of the cooling fluid 35 within a predetermined range, a cooling machine (not shown) that circulates the cooling fluid 35 in the immersion water tank 34 for cooling may be provided.

[0060] The cooling liquid spraying device 32 sprays the cooling fluid 35 onto the metal strip 1 to cool it. The cooling liquid spraying device 32 is provided with a pair of water spraying nozzles 64 that spray the cooling fluid 35 onto both sides of the running metal strip 1 in multiple stages. By spraying the cooling fluid onto both sides of the front and back sides of the metal strip 1 from these multiple water spraying nozzles 64, the metal strip 1 passing through the soaking zone 7 can be effectively cooled.

[0061] The cooling liquid injection device 32 has a pair of restraining rollers 33 for suppressing deformation of the metal strip 1 and preventing deterioration of the shape of the metal strip 1. The restraining rollers 33 are arranged so as to clamp the metal strip 1 from the front and back sides. The pair of restraining rollers 33 can be arranged to be staggered with respect to the running direction of the metal strip 1 on the front and back sides of the metal strip 1, respectively.

[0062] Figure 3 It is a schematic diagram showing an example of arrangement of a pair of constraining rollers 33 a and 33 b arranged on the front and back surfaces of the metal strip 1 so as to be offset with respect to the running direction of the metal strip 1 . Figure 3 (a) is a schematic diagram illustrating the amount of press-in and the amount of displacement. Figure 3 (b) is a schematic diagram illustrating the displacement amount of the restraining roller.

[0063] like Figure 3 As shown in (a), the deviation amount of the constraint rollers 33a and 33b arranged on the front and back sides in the running direction of the metal strip 1 is called the offset amount (inter-constraint roller offset). In addition, it is preferred that at least one of the constraint rollers 33a and 33b arranged on the front and back sides has a position adjustment function for shifting the constraint roller 33 toward the metal strip 1 in a manner that the amount of pressure on the metal strip 1 is variable.

[0064] The amount of deviation of the restraining rollers 33a and 33b disposed on the front and back of the metal strip 1 in the direction perpendicular to the running direction of the metal strip 1 is referred to as the pressing amount or the restraining roller gap. Figure 3 In the example shown in (a), the constraining roller 33a is fixed, and the constraining roller 33b moves to press the constraining roller 33b into the metal strip 1. In this case, the amount by which the moving constraining roller 33b protrudes horizontally from the rolling line of the metal strip 1 is referred to as the pressing amount of the constraining roller.

[0065] Furthermore, a plurality of pairs of restraining rollers 33 may be arranged in the cooling liquid injection device 32 in the traveling direction of the metal strip 1. By arranging a plurality of pairs of restraining rollers 33 together with the cooling liquid injection device 32, the deformation of the metal strip 1 can be more effectively restrained between the martensitic transformation start temperature (Ms) and the end temperature (Mf) of the metal strip 1. Thus, it is possible to suppress the shape defects of the metal strip 1 caused by the volume change caused by the thermal contraction and phase transformation generated when the metal strip 1 is rapidly cooled. In this case, the pressing amount of the restraining roller is defined for each restraining roller that can perform pressing.

[0066] Alternatively, the pair of restraining rollers 33a and 33b may be displaced toward the surface or back side of the metal strip 1 (restraining roller displacement) while maintaining the spacing between the pair of restraining rollers 33a and 33b in a direction perpendicular to the surface of the metal strip 1 at a fixed distance. Figure 3In the example shown in (b), the metal strip 1 is displaced in one direction while maintaining the deviation (pressing amount) of the restraining rollers 33a and 33b arranged on the front and back sides in the direction perpendicular to the running direction of the metal strip 1. The average of the displacement amount of the restraining roller 33a and the displacement amount of the restraining roller 33b is referred to as the displacement amount of the restraining roller. By displacing the restraining roller, the distance between the water jet nozzle 64 and the metal strip 1 becomes shorter on one side and longer on the other side, so that a difference in cooling capacity is generated on the front and back sides of the metal strip 1, and a difference in thermal strain is generated on the front and back sides of the metal strip 1, thereby changing the warping shape of the metal strip 1.

[0067] Next, the temper rolling equipment 13 will be described. Figure 4 1 is a schematic diagram showing an example of the temper rolling equipment 13 including the temper rolling mill 40. The temper rolling mill 40 is a device that provides a stretch of about 0.1 to 3.0% to the metal strip 1 to flatten the shape of the metal strip 1.

[0068] The temper rolling equipment 13 includes a temper rolling mill 40, an input side tensioning roll 46, an input side tension meter 47, an output side tension meter 48, and an output side tensioning roll 49. The temper rolling mill 40 includes a pair of working rolls 41a and 41b that directly contact and press down the metal strip 1 as a rolled material, and support rolls 42a and 42b that support the pair of working rolls 41a and 41b from above and below and suppress the bending of the working rolls 41a and 41b. At least one of the working rolls 41a and 41b and the support rolls 42a and 42b is connected to a driving motor via a coupling or a speed reducer at one end, and the driving motor rotates the roll. The support roll 42a is supported by a bearing box (support roll holder) 44 at the end of the shaft. The press-down device 43 adjusts the gap between the working roll 41a and the working roll 41b, that is, the roll gap, by displacing the bearing box 44 up and down. The load detector 45 detects the rolling load of the temper rolling mill 40.

[0069] An input side tension roller 46, an input side tension meter 47, an output side tension meter 48, and an output side tension roller 49 are arranged before and after the skin-pass rolling mill 40. The input side tension roller 46 and the output side tension roller 49 apply appropriate tension to the metal strip 1. The input side tension meter 47 and the output side tension meter 48 are respectively arranged between the skin-pass rolling mill 40 and the input side tension roller 46, and between the skin-pass rolling mill 40 and the output side tension roller 49. The input side tension meter 47 and the output side tension meter 48 measure the tension of the metal strip 1 on the input side and the output side of the skin-pass rolling mill 40.

[0070] The skin-pass rolling mill 40 rolls the metal strip 1 at a predetermined elongation. The elongation refers to the extension of the metal strip 1 before and after skin-pass rolling, and is defined by the increase rate of the length of the metal strip 1 before and after skin-pass rolling. The elongation is measured by the circumferential speed difference between the circumferential speed of the input side tension roller 46 and the circumferential speed of the output side tension roller 49.

[0071] Next, the warping shape predicted and controlled by the warping shape prediction method and the warping shape control method of the metal strip of the present embodiment is described. As the warping of the metal strip 1, L warping (long side warping) and C warping (width warping) are known. Among them, C warping is included in the warping shape predicted and controlled by the warping shape prediction method and the warping shape control method of the metal strip of the present embodiment, but L warping is not included. This is because L warping is constrained by the line tension when the metal strip 1 is continuously conveyed, and it is difficult to measure L warping online. The warping shape predicted and controlled by the warping shape prediction method and the warping shape control method of the metal strip of the present embodiment includes not only C warping in which the cross-sectional shape of the metal strip 1 in the plate width direction is approximated by a quadratic function or an arc, but also includes warping in which the cross-sectional shape in the plate width direction becomes a W-shaped warping and warping approximated by a higher-order function of more than three degrees.

[0072] The warping of the metal strip 1 in the plate width direction is caused by the inability to maintain the in-plane stress generated in the plate width direction of the metal strip 1 due to the cooling process, etc., and it is displaced out of the plane due to buckling. In contrast, the wave-shaped shape defect is caused by the extension of the length direction of the metal strip 1 distributed in the plate width direction, and thus the in-plane stress generated in the length direction of the metal strip 1 cannot be maintained in the plane, and it is displaced out of the plane due to buckling. The warping shape in the plate width direction and the wave shape are the same in that they are out-of-plane deformations caused by buckling, but there is a difference in whether the direction of the stress that causes the buckling is the plate width direction or the length direction.

[0073] Therefore, in the case of a corrugated shape, if tension is applied in the longitudinal direction of the metal strip 1, the difference in extension in the plate width direction (difference in extension in the plate width direction) is absorbed, and the shape defect of the metal strip 1 is latent (changes to a flat shape in appearance). On the other hand, in the case of a warped shape in the plate width direction, even if the tension applied in the longitudinal direction of the metal strip 1 is changed, the stress in the plate width direction is not greatly affected. Therefore, even if the tension applied in the longitudinal direction of the metal strip 1 is changed, the warped shape of the metal strip 1 in the plate width direction does not change.

[0074] The contact shape meter disclosed in Patent Document 2 applies a large tension (e.g., 100 MPa to 200 MPa) between the surface tempering mill and the tension roller on the output side to make the shape defects of the metal strip 1 latent, and measures the contact load distribution in the plate width direction. That is, this contact shape meter is used to measure the wave shape of the metal strip 1, not to measure the warping shape in the plate width direction. In addition, the displacement of the wave shape in the height direction of the metal strip 1 relative to the length direction of the metal strip 1 changes periodically, while there is a difference in the warping shape in which the periodic displacement relative to the length direction of the metal strip 1 is not observed. In addition, the period of the wave shape of the metal strip 1 is mostly a spacing of about 0.5 to 5 m in the length direction of the metal strip 1.

[0075] Figure 5 It is a schematic diagram showing an example of the warping shape of the metal strip 1 in the plate width direction. Figure 5 The warping shape shown is an example of the warping shape measured for a metal strip 1 composed of a high-strength steel plate having a plate thickness of 1.2 mm and a plate width of 1200 mm after a cooling process using water quenching and after rolling by a skin-pass rolling mill 40. However, Figure 5 The warpage heights shown are normalized to the maximum height across the width of the board. Figure 5 As shown, when the metal strip 1 composed of a high-strength steel plate is rapidly cooled by water quenching, the warping shape in the plate width direction becomes a warping shape approximated by a high-order function.

[0076] Next, the input-side warping shape measuring device 16 will be described. The input-side warping shape measuring device 16 measures the input-side warping shape of the conveyed metal strip 1, and obtains information related to the height of the metal strip 1 in the plate width direction of the metal strip 1. The information related to the height of the metal strip 1 may be, for example, an average value, a maximum height, or a standard deviation of the warping height in the plate width direction of the metal strip 1, or may be information for determining the distribution of the warping height.

[0077] The information determining the distribution of the warpage height of the metal strip 1 is, for example, Figure 5 The information of the distribution of the warpage height at each position in the plate width direction of the metal strip 1 is shown. As the input side warpage shape of the metal strip 1, the input side warpage shape measuring device 16 can also use an arbitrary function to approximate the curve representing the distribution of the warpage height in the plate width direction, and obtain the coefficient of the function that determines the approximation.

[0078] The input side warpage shape measuring device 16 can also obtain Figure 5 The distribution of curvature determined by the curve showing the relationship between the position in the plate width direction and the warpage height is taken as the input side warpage shape of the metal strip 1. In addition, the input side warpage shape measuring device 16 can also measure the input side warpage shape of the metal strip 1. Figure 5The position in the board width direction and the warpage height shown are subjected to discrete Fourier transform, the intensity of each concavity and convexity in the board width direction is extracted, and the relationship between the pitch of the concavity and convexity and the spectral intensity calculated by discrete Fourier transform is obtained.

[0079] In addition, the input-side warping shape measuring device 16 may average the distribution of the warping height at each position in the plate width direction of the metal strip 1 in a range of about 1 to 5 m in the length direction of the metal strip 1. In this way, by averaging the warping height in a range of about 1 to 5 m in the length direction of the metal strip 1, even if the metal strip 1 includes both a warping shape and a wavy shape, the wavy shape can be removed and the warping shape can be determined.

[0080] As the input-side warping shape measuring device 16, any device can be used as long as it is a device that can measure the height distribution in the plate width direction of the metal strip 1. As the input-side warping shape measuring device 16, for example, a laser rangefinder can be used. A plurality of laser rangefinders as non-contact rangefinders can be arranged at different positions in the plate width direction of the conveyed metal strip 1, and the warping height at each position can be measured using the laser rangefinder (distance measurement method). For example, when a laser rangefinder is used to measure the warping height of a metal strip 1 with a plate width of 800 mm to 1900 mm, the number of laser rangefinders arranged at different positions in the plate width direction is preferably 3 or more and 100 or less. More preferably, it is 10 to 20. If the number of laser rangefinders is less than 3, it is difficult to determine the warping shape in the plate width direction of the metal strip 1, so it is not preferred. In addition, even if the number of laser rangefinders is more than 100, the measurement accuracy of the warping shape will not be improved, and the equipment cost will increase, so it is not preferred.

[0081] Figure 6 Schematic diagram showing the state of measuring the warping shape of the metal strip 1 using a laser scanning laser rangefinder. Figure 6 As shown, a laser scanning laser rangefinder 70 may be used as the input-side warpage shape measuring device 16. The laser scanning laser rangefinder measures the distance to each position in the plate width direction of the metal strip 1 by scanning the laser in a fan shape, and calculates the height at each position in the plate width direction based on the measured distance. By using the laser scanning laser rangefinder 70, the warpage height in the plate width direction of the metal strip 1 can be measured by scanning the laser in the plate width direction from above the conveyed metal strip 1.

[0082] Figure 7 Schematic diagram showing the state of measuring the warpage shape using multiple laser scanning laser rangefinders. Figure 7As shown in FIG. 1 , a plurality of laser scanning laser distance meters 70 may be used as the input-side warpage shape measuring device 16. In this case, a plurality of laser scanning laser distance meters 70 are arranged in parallel in the plate width direction of the metal strip 1 to measure the warpage height in the plate width direction of the metal strip 1. Figure 6 The device shown measures the input side warpage shape in a short time.

[0083] In addition, as the input-side warpage shape measuring device 16, a device that measures the input-side warpage shape by a light section method may also be used. In the light section method, a laser beam that expands in a fan shape relative to the plate width direction is irradiated from above the metal strip 1, and reflected light from the metal strip 1 is received, and the received reflected light is photographed by an image sensor, and the photographed image is processed to measure the warpage height of the metal strip 1. Devices that measure the input-side warpage shape by the light section method are, for example, the LJ-X8000 series manufactured by Keyence Corporation and the Gocator manufactured by LMT Technologies Corporation.

[0084] The input-side warping shape measuring device 16 preferably has a measurement accuracy of 1 mm or less, and more preferably has a measurement accuracy of 0.5 mm or less, for the warping height of the metal strip 1. By using the input-side warping shape measuring device 16 with a measurement accuracy of 1 mm or less, even if the upper limit of the warping height of the metal strip 1 is set to a range of 5 to 10 mm, the warping height of the metal strip 1 can be controlled within the range.

[0085] The measurement of the warpage height of the metal strip 1 by the input-side warpage shape measuring device 16 is preferably performed at intervals of 1 m or less in the length direction of the metal strip 1. The measurement frequency of the warpage height of the continuously conveyed metal strip 1 is preferably 10 Hz or more, and more preferably 20 Hz or more. For example, in the output-side device 24 of the continuous annealing device 100, the metal strip 1 is sometimes conveyed at a speed of 600 m / min or more. Even if the metal strip 1 is conveyed at this conveying speed, as long as the measurement frequency of the warpage shape is 10 Hz or more, the warpage shape can be measured at intervals of 1 m or less in the length direction of the metal strip 1. In addition, the above description of the input-side warpage shape measuring device 16 is applicable to the pre-cooling warpage shape measuring device 17 and the output-side warpage shape measuring device 18.

[0086] Next, the warp shape control device 52 will be described. Figure 8: is a schematic diagram showing a structural example of a warping shape control device 52. The warping shape control device 52 includes a control unit 54, an input unit 55, an output unit 56, and a storage unit 57. The control unit 54 is, for example, a CPU, and by executing a program read from the storage unit 57, the control unit 54 functions as an acquisition unit 58, a warping shape prediction unit 59, an operation parameter determination unit 60, and a warping shape prediction model generation unit 61. The storage unit 57 is, for example, an information recording medium such as a flash memory that can update records, a hard disk built in or connected via a data communication terminal, or a memory card, and a read / write device thereof. The storage unit 57 stores programs for the control unit 54 to execute various functions, data used by the programs, and the like.

[0087] The storage unit 57 also stores a database 62 and a warp shape prediction model 63. The database 62 stores a data set that is a set of at least one of the measured values ​​of the input-side warp shape of the metal strip 1 manufactured by the skin-pass rolling equipment 13 in the past, the measured values ​​of the output-side warp shape, and the operating parameters of the skin-pass rolling mill 40. These measured values ​​and operating parameters are collected and stored in the control computer 50, and are obtained from the control computer 50 via the control controller 51. The warp shape prediction model 63 is prepared in advance by the warp shape prediction model generation unit 61 and stored in the storage unit 57.

[0088] In the warp shape prediction method and the warp shape control method of the metal strip of the present embodiment, the warp shape prediction model 63 is a machine learning model that has been learned and takes as input input data including at least one of the input side warp shape and the operating parameters of the skin-pass rolling mill 40, and takes as output the output side warp shape. In the case where the input side warp shape is the distribution of the warp height in the plate width direction of the metal strip 1, the input data input to the learning model includes the value of the warp height at each position in the plate width direction and at least one of the operating parameters of the skin-pass rolling mill 40. In addition, in the case where the input side warp shape is a function obtained by approximating the distribution of the warp height in the plate width direction of the metal strip 1, the input data input to the learning model includes each coefficient that determines the function and at least one of the operating parameters of the skin-pass rolling mill 40.

[0089] In addition, when the input-side warping shape is a distribution of curvatures determined from a curve indicating a relationship between a position in the sheet width direction of the metal strip 1 and a warping height, the input data input to the learning model includes at least one of a value of the curvature at each position in the sheet width direction and an operating parameter of the skin-pass rolling mill 40. Furthermore, when the input-side warping shape is a relationship between a pitch of concavities and convexities and a spectral intensity, the input data input to the learning model includes at least one of a spectral intensity of a pitch of concavities and convexities and an operating parameter of the skin-pass rolling mill 40.

[0090] In addition, when the distribution of the warpage height in the plate width direction of the metal strip 1 is output as the output side warpage shape, a warpage shape prediction model is prepared to output the warpage height at each position in the plate width direction. Similarly, when a function that approximates the warpage height is output as the output side warpage shape, a warpage shape prediction model that outputs each coefficient of the function is prepared in advance. In addition, when the distribution of the curvature is output as the output side warpage shape, a warpage shape prediction model that outputs the curvature is prepared in advance. Furthermore, when the spectral intensity of each concave-convex pitch is output as the output side warpage shape, a warpage shape prediction model that outputs the spectral intensity of each concave-convex pitch is prepared.

[0091] Next, the processing performed by the acquisition unit 58, the warpage shape prediction unit 59, the operation parameter determination unit 60, and the warpage shape prediction model generation unit 61 will be described. The acquisition unit 58 acquires the input-side warpage shape of the metal strip 1 from the input-side warpage shape measurement device 16. In addition, the acquisition unit 58 acquires at least one of the operation parameters of the skin-pass rolling mill 40 from the control computer 50 via the control controller 51. At this time, the acquisition unit 58 acquires at least one of the operation parameters set for the skin-pass rolling mill 40 at the time when the position where the input-side warpage shape is measured by the input-side warpage shape measurement device 16 reaches the skin-pass rolling mill 40. The acquisition unit 58 outputs at least one of the input-side warpage shape and the operation parameter of the skin-pass rolling mill 40 to the warpage shape prediction unit 59.

[0092] When obtaining at least one of the input-side warp shape and the operating parameters of the skin-pass rolling mill 40, the warp shape prediction unit 59 reads the warp shape prediction model 63 from the storage unit 57, inputs at least one of the input-side warp shape and the operating parameters of the skin-pass rolling mill 40 to the warp shape prediction model, and outputs the output-side warp shape. In this way, the warp shape prediction unit 59 predicts the output-side warp shape.

[0093] Here, the operating parameters of the temper rolling mill 40 acquired by the acquisition unit 58 will be described. The operating parameters of the temper rolling mill 40 used for prediction of the output-side warp shape are set values ​​of operating conditions of the temper rolling mill 40 that affect the warp shape of the metal strip 1.

[0094] As operating parameters of such a temper rolling mill 40, the elongation (extension), rolling load, and roll gap between the work rolls 41a and 41b during temper rolling of the metal strip 1 can be used. In addition, the tension (input side tension) applied to the metal strip 1 between the input side tension roll 46 and the temper rolling mill 40 and the tension (output side tension) applied to the metal strip 1 between the output side tension roll 49 and the temper rolling mill 40 can also be used. This is because these tensions affect the size of the plastic deformation applied to the metal strip 1 by the pressing of the work rolls 41a and 41b, and the warping shape changes according to the size of the plastic deformation applied to the metal strip 1.

[0095] In addition, as an operating parameter of the skin-pass rolling mill 40, the bending force of the work roll bender, which is a shape control actuator of the skin-pass rolling mill 40, may be used. The work roll bender of the skin-pass rolling mill 40 is a shape control actuator that applies a bending force to the work rolls 41a and 41b to adjust the deflection of the work rolls 41a and 41b. By adjusting the bending force of the work roll bender, the deflection of the work rolls 41a and 41b is adjusted, and a distribution of a reduction ratio or an elongation ratio is formed in the sheet width direction of the metal strip 1. By performing skin-pass rolling using the work rolls 41a and 41b with the distribution of the reduction ratio or the elongation ratio, the warping shape in the sheet width direction of the metal strip 1 can be controlled.

[0096] The warping shape prediction unit 59 outputs the predicted output-side warping shape to the operating parameter determination unit 60. The operating parameter determination unit 60 first determines whether the predicted output-side warping shape is less than a preset target value (warping target value). The target value may be obtained from the control computer 50 via the acquisition unit 58 and the control controller 51, or may be input by an operator from the input unit 55. In addition, the target value may be stored in the storage unit 57 in advance and read from the storage unit 57. When it is determined that the output-side warping shape predicted by the warping shape prediction unit 59 exceeds the target value, the operating parameter determination unit 60 determines the operating parameters of the cooling device 30 that can reduce the output-side warping shape of the metal strip 1.

[0097] The operating parameters of the cooling device 30 determined by the operating parameter determination unit 60 are preferably operating parameters that have a large influence on the change in the warping shape of the metal strip 1 due to cooling. As such operating parameters of the cooling device 30, any one of the pressing amount of the restraining roller 33 of the cooling device 30, the offset amount between the restraining rollers 33, and the displacement amount of the restraining rollers can be used.

[0098] For example, when the amount of pressure of the restraining roller 33 is used, by reducing the amount of pressure (closing the restraining roller), the warping of the metal strip 1 caused by the strain due to the heat shrinkage and phase change generated in the cooling metal strip 1 can be reduced. As a result, the warping shape of the metal strip 1 on the output side can be reduced.

[0099] When the offset between the constraining rollers 33 is used, the effect of constraining the deformation of the metal strip 1 during cooling becomes larger by reducing the offset, and the warping of the metal strip 1 is reduced. Thus, the warping shape of the metal strip 1 on the outlet side can be reduced.

[0100] Furthermore, when the displacement of the restraining roller 33 is used, the distance between the surface of the metal strip 1 and the liquid injection nozzle is changed by changing the displacement, thereby making a difference in the cooling strength between the front and back of the metal strip 1, thereby changing the warping shape. By controlling the change in the warping shape in the direction of reducing the warping, the warping shape on the output side of the metal strip 1 can be reduced.

[0101] In addition, the amount of water sprayed from the liquid spray nozzle may be used as an operating parameter of the cooling device 30. The warping shape can also be changed by providing a difference in the amount of water between the front and back sides of the metal strip 1. However, the responsiveness of the control of the amount of water sprayed from the liquid spray nozzle is not necessarily high, so it is preferable to use the displacement amount of the restraining roller 33 as an operating parameter of the cooling device 30.

[0102] The operating parameter determination unit 60 outputs the determined operating parameters of the cooling device 30 to the control controller 51. The control controller 51 changes the operating conditions of the cooling device 30 using the operating parameters obtained from the operating parameter determination unit 60. In addition, the operating parameter determination unit 60 may cause the output unit 56 to display the determined operating parameters of the cooling device 30, and the operator may change the operating conditions of the cooling device 30 using the operating parameters.

[0103] In this way, the output-side warping shape of the metal strip 1 is predicted immediately after passing through the skin-pass rolling mill 40, and the operating conditions of the cooling device 30 are controlled in such a manner that the predicted output-side warping shape becomes less than the target value. By controlling the cooling device 30 in this way, the output-side warping shape is actually measured at the inspection station 14 disposed on the downstream side of the continuous annealing device 100, and based on the output-side warping shape, control can be performed more quickly than by controlling the operating conditions of the cooling device 30, and in the case where a defective portion of the output-side warping shape of the metal strip 1 occurs, the defective portion can be reduced.

[0104] Such control is preferably performed over the entire length from the front end to the rear end of the metal strip 1. As a result, the warping shape at the output side becomes less than the target value over the entire length of the metal strip 1, and the metal strip 1 having an excellent warping shape can be manufactured. In the continuous annealing device 100 including a heating device and a cooling device, the metal strip 1 is rapidly cooled, and the warping shape of the metal strip 1 is easily deteriorated. In particular, the warping shape of the metal strip 1 having a tensile strength of 980 MPa or more is easily deteriorated due to rapid cooling. Therefore, by applying the warping shape control method of the metal strip of the present embodiment, it is possible to suppress the decrease in yield caused by poor warping shape, and to manufacture the metal strip 1 having an excellent warping shape.

[0105] In the warp shape control method of the metal strip of the present embodiment, the operating parameters of the cooling device 30 are determined and set because, in the continuous annealing device 100, it is the cooling process in the cooling device 30 that greatly changes the warp shape of the metal strip 1. On the other hand, in order to control the warp shape on the delivery side, it is also conceivable to control the operating conditions of the skin-pass rolling mill 40, but the skin-pass rolling mill 40 has not only the function of correcting the shape of the metal strip 1 but also the function of adjusting the mechanical properties of the metal strip 1, so the range of the elongation of the metal strip 1 that can be changed in the skin-pass rolling mill 40 is limited. Therefore, the range of the operating conditions that can be changed to correct the warp shape on the delivery side is limited, and it is not necessarily possible to become an operating condition that can make the warp height less than the target value. Therefore, in the warp shape control method of the metal strip of the present embodiment, the operating parameters of the cooling device 30 that can reduce the warp shape on the delivery side are determined and set.

[0106] Next, a description will be given of a method for generating the warp shape prediction model 63 by the warp shape prediction model generating unit 61. When generating the warp shape prediction model 63, the input-side warp shape measuring device 16 for measuring the input-side warp shape of the metal strip 1 on the upstream side of the skin-pass rolling mill 40 and the output-side warp shape measuring device 18 for measuring the output-side warp shape of the metal strip 1 on the downstream side of the skin-pass rolling mill 40 are used to obtain measurement data of the input-side warp shape and the output-side warp shape.

[0107] The control computer 50 generates tracking information of the welded portion of the metal strip 1. By using the information related to the conveying speed of the metal strip 1 and the tracking information, the control computer 50 obtains and stores the input side warping shape and the output side warping shape in correspondence with the distance from the front end welded portion of the metal strip 1. At this time, since the metal strip 1 is extended in the skin-pass rolling mill 40, the control computer 50 preferably corrects the position from the front end welded portion measured by the output side warping shape measuring device 18 according to the elongation of the skin-pass rolling mill 40. In addition, the control computer 50 obtains and stores the operating parameters of the skin-pass rolling mill 40 in correspondence with the distance from the front end welded portion of the metal strip 1.

[0108] The acquisition unit 58 acquires a data set from the control computer 50 via the control controller 51, the data set including the measured values ​​of the input-side warpage shape and the output-side warpage shape at the position where the distance from the front end weld portion of the metal strip 1 is the same, and at least one of the operating parameters of the skin-pass rolling mill 40 when the position passes through the skin-pass rolling mill 40. The acquisition unit 58 stores the acquired data set in the database 62 of the storage unit 57. Preferably, more than 200 data sets are stored in the database 62, and more preferably, more than 1000 data sets are stored in the database 62.

[0109] In addition, the acquisition unit 58 preferably acquires the above data groups from the front end (e.g., within 20 m from the front end welding part), the stable part, and the tail end (e.g., within 20 m from the tail end welding part) of the metal strip 1. In metal products, it is usually determined whether the product can be shipped by inspecting the front end, the stable part, and the tail end. Therefore, in order to cope with the inspection, it is preferred to acquire the data groups of the front end, the stable part, and the tail end.

[0110] Furthermore, the acquisition unit 58 preferably acquires a data set of the entire length of the metal strip 1 at a predetermined interval from the front end to the rear end of the metal strip 1. The interval for acquiring the data set is preferably set within a range of 5 m to 20 m. In this way, the data set covering the entire length of the metal strip 1 is stored in the database 62, and the warping shape prediction model is created using the data set, thereby forming a warping shape prediction model that can predict the warping shape of the output side with high accuracy. In addition, the acquisition unit 58 may set a certain number of data sets as an upper limit, and appropriately update the data set stored in the database 62 within the upper limit.

[0111] The warping shape prediction model generation unit 61 uses the data set stored in the database 62 to generate a warping shape prediction model 63 that takes as input the input data including at least one of the input side warping shape and the operating parameters of the surface light machine 40 and takes as output the output side warping shape. In the case where the warping shape prediction model 63 is a machine learning model that has been learned, the warping shape prediction model generation unit 61 uses the data set stored in the database 62 as training data to cause the machine learning model to perform machine learning and generates a machine learning model that has been learned. As a machine learning model, commonly used neural networks (including deep learning, convolutional neural networks, etc.), decision tree learning, random forests, support vector regression, etc. can be used. In addition, an integrated model combining multiple models can also be used.

[0112] Fig. 9 is a diagram showing an example of a machine learning model using a neural network. The warpage shape prediction model 63 can be, for example, Fig. 9 It is generated by a general neural network machine learning model as shown. Fig. 9 L1, L2, and L3 are the input layer, the intermediate layer, and the output layer, respectively. In particular, if deep learning using a neural network with a multi-layer structure is used, the problem of multicollinearity can be ignored, and other operating parameters that are correlated with the warping shape of the metal strip 1 can be freely selected as inputs, thereby improving the prediction accuracy of the warping shape of the output side of the metal strip 1. Fig. 9 As shown, as a neural network, a neural network with 2 to 3 intermediate layers and 18 to 512 nodes can be used, and as an activation function, a Sigmoid function can be used. The number of nodes in the output layer can be set in a manner consistent with the number of numerical values ​​set as the output of the warping shape prediction model. For example, in the case where the warping shape prediction model outputs the warping height at each position in the board width direction, the number of output layers can be set to the number of positions in the board width direction that determine the warping height. In addition, in the case where the warping shape prediction model outputs a function that approximates the warping height, the number of output layers can be set to the number of coefficients used to determine the function.

[0113] In addition, the warping shape prediction model generation unit 61 may also perform machine learning by dividing the data set stored in the database 62 into training data and test data. In this way, by dividing the data set into training data and test data, the warping shape prediction model generation unit 61 can use the training data to learn the weight coefficients of the neural network, and generate the warping shape prediction model while changing the structure of the neural network (the number of intermediate layers, the number of nodes) so that the accuracy of the output side warping shape in the test data becomes higher. By generating such a warping shape prediction model 63, the estimation accuracy of the output side warping shape based on the warping shape prediction model 63 can be improved.

[0114] In addition, the weight coefficient of the warpage shape prediction model 63 may be updated, and the update of the weight coefficient may also use the error propagation method. In addition, the warpage shape prediction model 63 may be updated to a new warpage shape prediction model by performing machine learning every 6 months or every year, for example. By updating the warpage shape prediction model using a data set including the latest data, even if the state of the skin-pass rolling mill 40 changes over time, the latest state can be reflected in the warpage shape prediction model.

[0115] Fig.10 This is a flowchart showing the process of the warping shape control method of the metal strip according to the present embodiment. Fig.10 The process of the warping shape control method of the metal strip of the present embodiment is described below. The process of the warping shape control method of the metal strip of the present embodiment is started, for example, when the continuous annealing equipment 100, the control computer 50, the control controller 51 and the warping shape control device 52 are started.

[0116] First, the acquisition unit 58 acquires the input-side warpage shape from the input-side warpage shape measuring device 16. Furthermore, the acquisition unit 58 acquires at least one of the operating conditions set in the skin-pass rolling mill 40 from the control controller 51 when the position for measuring the input-side warpage shape reaches the skin-pass rolling mill 40 (step S101). The acquisition unit 58 outputs at least one of the acquired input-side warpage shape and the operating condition of the skin-pass rolling mill 40 to the warpage shape prediction unit 59.

[0117] The warp shape prediction unit 59 reads the warp shape prediction model 63 from the storage unit 57, inputs at least one of the operating conditions of the skin temper rolling mill 40 and the input side warp shape to the warp shape prediction model 63, outputs the output side warp shape, and predicts the output side warp shape (step S102). The warp shape prediction unit 59 outputs the predicted output side warp shape to the operating parameter determination unit 60.

[0118] After obtaining the output-side warping shape, the operating parameter determination unit 60 determines whether the output-side warping shape exceeds the target value (step S103). When it is determined that the output-side warping shape exceeds the target value (step S103: Yes), the operating parameter determination unit 60 determines the operating parameters of the cooling device 30 that can reduce the output-side warping shape (step S104). On the other hand, when it is determined that the output-side warping shape is below the target value (step S103: No), the operating parameter determination unit 60 returns the process to step S101 and repeatedly executes the process from step S101.

[0119] If the operating parameters of the cooling device 30 that can reduce the output side warping shape are determined, the operating parameter determination unit 60 outputs the operating parameters to the control controller 51, and the control controller 51 changes the operating conditions of the cooling device 30. After outputting the operating parameters, the operating parameter determination unit 60 determines whether an instruction to end the control is input from the control controller 51 or from the operator via the input unit 55 (step S106). When the operating parameter determination unit 60 determines that an input to end the control is input (step S106: yes), the processing of the warping shape control method of the metal strip is ended. In addition, when the operating parameter determination unit 60 determines that an input to end the control is not input, the processing returns to step S101 and the processing from step S101 is repeatedly executed. Then, for the output side warping shape predicted using the input side warping shape after the operating parameters of the cooling device 30 are changed, it is determined again in step S103 whether it exceeds the target value. By repeatedly executing this processing, the steel strip 1 with the output side warping shape being less than the target value can be manufactured.

[0120] Thus, by implementing the warpage shape control method of the metal strip of the present embodiment, the warpage shape of the output side can be quickly predicted based on the warpage shape of the input side of the metal strip 1 when the metal strip 1 reaches the skin-pass rolling mill 40. Furthermore, by determining the operating parameters of the cooling device 30 so as to reduce the predicted warpage shape of the output side, the operating parameters can be set as the operating conditions of the cooling device 30, thereby realizing the manufacture of the metal strip 1 with a reduced warpage height in the plate width direction.

[0121] <Other Implementation Methods>

[0122] Next, other embodiments of the shape prediction method of the metal strip of the present embodiment are described. As the warpage shape prediction method of the present embodiment, an example is shown in which the warpage shape prediction unit 59 predicts the output side warpage shape using a machine learning model that has been learned and takes the output side warpage shape as an output, which is input, including input data of at least one of the input side warpage shape and the operating parameters of the skin-pass rolling mill 40, but is not limited to this. For example, when the average value of the warpage height in the plate width direction of the metal strip 1, the maximum value of the warpage height, or the standard deviation of the warpage height is used as the input side warpage shape and the output side warpage shape, the warpage shape prediction unit 59 may also predict the output side warpage shape using a functional formula that uses at least one of the input side warpage shape and the operating parameters of the skin-pass rolling mill 40. In this case, the warpage shape prediction unit 59 predicts the output side warpage shape using the function f shown in the following formula (1).

[0123] HO=f(HI,g)···(1)

[0124] In the above formula (1), HO is the warping height (mm) on the output side, HI is the warping height (mm) on the input side, and g is the operating parameter of the skin-pass rolling mill 40 .

[0125] The operating parameter g of the temper rolling mill 40 is a set value of an operating condition that affects the warping shape of the metal strip 1 among the operating conditions of the temper rolling mill 40. For example, a single parameter or a plurality of parameters represented by scalar quantities such as the elongation applied to the metal strip 1 by the temper rolling mill 40 and the bending force of the work roll bender can be used. Generally, if the elongation increases, there is a tendency for the output side warping height HO to decrease. As the input side warping height HI increases, even if the same elongation is applied, there is a tendency for the output side warping height HO to be difficult to decrease. Therefore, as the function f, a function represented by a linear or quadratic expression of HI and g can be applied. The function f is generated by a multivariate regression analysis using a data set stored in the database 62 and is stored in the storage unit 57 in advance.

[0126] In addition, when the pre-cooling warp shape measuring device 17 is provided between the heating zone 6 and the cooling zone 8 of the continuous annealing equipment 100, the warp shape prediction unit 59 may also predict the output side warp shape using a learned machine learning model, which takes as input the input data including the pre-cooling warp shape, the input side warp shape, and at least one of the operating parameters of the skin-pass rolling mill 40, and takes as output the output side warp shape. In this case, a plurality of data sets are stored in the database 62, which take as a group the measured values ​​of the pre-cooling warp shape, the measured values ​​of the input side warp shape, at least one of the operating parameters of the skin-pass rolling mill 40, and the measured values ​​of the output side warp shape. The warp shape prediction model generation unit 61 generates as a warp shape prediction model 63 a learned machine learning model that performs machine learning using the data set as training data.

[0127] If the warp height of the metal strip 1 before cooling is high, the warp shape formed in the metal strip 1 by cooling in the cooling zone 8 also becomes larger, and as a result, the warp shape on the output side of the downstream side of the skin-pass rolling mill 40 also becomes larger. In this way, there is a correlation between the warp shape of the metal strip 1 before cooling and the warp shape on the output side, so by including the warp shape before cooling in the input data, the prediction accuracy of the warp shape on the output side using the learned machine learning model can be improved.

[0128] Furthermore, even when the pre-cooling warp shape measuring device 17 is provided between the heating zone 6 and the cooling zone 8 of the continuous annealing equipment 100, when the average value of the warp height in the plate width direction of the metal strip 1, the maximum value of the warp height, or the standard deviation of the warp height, etc., is used as the input-side warp shape and the output-side warp shape, the output-side warp shape can be predicted using a functional expression using the pre-cooling warp shape, the input-side warp shape, and the operating parameters of the skin-pass rolling mill 40. In this case, the warp shape prediction unit 59 predicts the output-side warp shape using the function f2 shown in the following formula (2).

[0129] HO=f2 (HI, HC, g)···(2)

[0130] In the above formula (2), HO is the output side warpage height (mm), HI is the input side warpage height (mm), HC is the pre-cooling warpage height (mm), and g is an operating parameter of the skin-pass rolling mill 40. Function f2 is generated by multivariate regression analysis using the data set accumulated in the database 62 in the same way as function f, and is stored in the storage unit 57 in advance.

[0131] In this way, even when other embodiments of the warp shape prediction method of a metal strip are used, the warp shape of the output side of the skin-pass rolling mill 40 can be quickly predicted when the metal strip 1 reaches the skin-pass rolling mill 40. Furthermore, by determining the operating parameters of the cooling device 30 so as to reduce the predicted warp shape on the output side, the operating parameters can be used as operating conditions of the cooling device 30, thereby realizing the manufacture of the metal strip 1 with a reduced warp height in the plate width direction.

[0132] In addition, Figure 1 In the example shown, the control computer 50, the control controller 51, and the warping shape control device 52 are shown as different devices, but the present invention is not limited thereto. The control computer 50 and the control controller 51 may be one device, or the control controller 51 and the warping shape control device 52 may be one device. In addition, the control computer 50, the control controller 51, and the warping shape control device 52 may be one device.

[0133] Example 1

[0134] As an embodiment of the present invention, Figure 1 The following describes Example 1 in which the continuous annealing equipment 100 shown predicts the warpage shape of the metal strip at the outlet. The metal strip used in the first example is a high-strength steel strip having a tensile strength TS of 980 MPa, a thickness of 1.4 mm, and a width of 1007 mm.

[0135] The continuous annealing equipment 100 is provided with a heating device for heating the metal strip in the heating zone 6, and a cooling device 30 for water cooling (water quenching) in the cooling zone 8. The metal strip is heated by the heating device, cooled to a temperature lower than the martensitic transformation start temperature (Ms) by the cooling device 30, and transported to the output side equipment 24. The surface temper rolling mill 40 is arranged at the output side equipment 24, the input side warpage shape measuring device 16 is arranged at the exit of the final cooling zone 11, and the output side warpage shape measuring device 18 is arranged at the inspection station 14.

[0136] The diameter of the working roll of the skin-pass rolling mill 40 is 470 mm, and the surface roughness of the working roll is Ra5.4 μm. The input side tension and the output side tension of the skin-pass rolling mill 40 are set to 70 kN, and the elongation of the skin-pass rolling is changed in the range of 0.1-0.2% by changing the roll gap.

[0137] In Example 1, the measurement data of the input-side warpage shape measured by the input-side warpage shape measuring device 16, the measurement data of the output-side warpage shape measured by the output-side warpage shape measuring device 18, and the actual operation data of the skin-pass rolling mill 40 are obtained as information of the same position in the longitudinal direction of the metal strip, and a data set consisting of these data is stored in the database 62 as a group. The data set stored in the database 62 is 50 to 200 for each position in the longitudinal direction of one metal strip. Then, at the stage where the data sets for 20 metal strips are stored in the database 62, a warpage shape prediction model is generated by the warpage shape prediction model generation unit 61. At this time, in the warpage shape prediction model generation unit 61, 2100 of the 3000 data sets stored in the database 62 are used as training data for the machine learning model, and the remaining 900 are used as test data to perform data segmentation of the data set, and evaluate the prediction accuracy of the output-side warpage shape.

[0138] The actual operation data of the skin-pass rolling mill 40 used in the input of the warp shape prediction model are the elongation given to the metal strip by the skin-pass rolling mill 40 and the actual value of the work roll bending machine. The measured data of the input side warp shape input to the warp shape prediction model selects both the maximum value of the warp height in the plate width direction and the average value of the warp height. In addition, for the actual data of the output side warp shape as the output of the warp shape prediction model, both the maximum value of the warp height in the plate width direction and the average value of the warp height are also selected.

[0139] In Example 1, using Fig. 9 The neural network shown is used as a warping shape prediction model. As the neural network, a neural network with 3 intermediate layers and 256 nodes is used, and as the activation function, a sigmoid function is used.

[0140] For the warp shape prediction model generated in this way, the maximum value and average height of the warp height in the plate width direction of the output side warp shape based on the warp shape prediction model were verified using 900 test data accumulated in the database. As a result, for the test data, the error between the maximum value of the output side warp shape output from the warp shape prediction model and the corresponding measured value of the maximum value of the output side warp shape was 0.5 mm on average, and the standard deviation was 0.4 mm. In addition, the error between the predicted value of the average height of the output side warp shape and the corresponding measured value was 0.3 mm on average, and the standard deviation was 0.2 mm. Based on these results, it was confirmed that the output side warp shape can be predicted with high accuracy by the warp shape prediction model.

[0141] Example 2

[0142] Next, the use of Figure 1The continuous annealing equipment 100 shown in the embodiment 2 is used to control the warpage shape of the metal strip at the outlet side. The metal strip used in this embodiment is a high-strength steel plate with a tensile strength TS of 1310 MPa, a plate thickness of 1.4 mm, and a plate width of 1000 mm.

[0143] The cooling start temperature of the metal strip in the cooling device 30 was 740° C., and the cooling end temperature was 30° C. The amount of cooling fluid sprayed from the cooling liquid spraying device 32 was 1000 m 3 / hr, and the cooling fluid water temperature is 30°C. The conveying speed of the metal strip passing through the cooling device 30 is 50 to 80 m / min. A pair of restraining rollers 33 are provided in the cooling liquid spraying device 32 to suppress the deformation of the metal strip and prevent the shape of the metal strip from deteriorating. The roller diameter of the restraining roller 33 is 150 mm, and the offset between the restraining rollers 33 is 80 mm.

[0144] The diameter of the working rolls 41a and 41b of the skin-pass rolling mill 40 is 470 mm, and the surface roughness of the working rolls 41a and 41b is Ra5.4 μm. The input side tension and the output side tension of the skin-pass rolling mill 40 are set to 70 kN, and the elongation of the skin-pass rolling is changed in the range of 0.1 to 0.2% by changing the roll gap.

[0145] In the input side warping shape measuring device 16 and the output side warping shape measuring device 18, as a means of measuring the warping height using the light sectioning method, two LJ-X8900s manufactured by Keyence are arranged side by side in the plate width direction of the metal strip, and the warping shape of the metal strip is measured at a spacing of 5m in the length direction.

[0146] In the second embodiment, first, the acquisition unit 58 acquires a data set including measurement data of the input-side warpage shape measured by the input-side warpage shape measuring device 16, measurement data of the output-side warpage shape measured by the output-side warpage shape measuring device 18, and operation parameters of the skin-pass rolling mill 40. A data set including measurement data of the same distance from the welded portion in the longitudinal direction of the metal strip is acquired and stored in the database 62. The database 62 stores data sets for 20 metal strips. The maximum value of the warpage height in the plate width direction of the metal strip is used as the input-side warpage shape and the output-side warpage shape included in the data set. In addition, as the operation parameters of the skin-pass rolling mill 40, the elongation given to the metal strip by the skin-pass rolling mill 40 and the setting value of the bending force of the work roll bender are used.

[0147] In the second embodiment, the output-side warpage height shape HO is approximated by a function f having the input-side warpage height HI and the skin-pass rolling operation parameter g as inputs, using the data set stored in the database 62. The function f is represented by a linear combination of HI, g1 (elongation), and g2 (bending force of the work roll bender) as variables, and the respective coefficients are calculated by the least square method. The output-side warpage shape is predicted using the function f calculated in this way. This embodiment is referred to as Invention Example 1 and Invention Example 2.

[0148] In addition, a warpage shape prediction model is generated by performing machine learning using the data set stored in the database 62 as training data, and the output side warpage shape is predicted using the warpage shape prediction model. This embodiment is referred to as Invention Example 3 to Invention Example 5. As the warpage shape prediction model of Invention Example 3 to Invention Example 5, Fig. 9 As the neural network, three intermediate layers and 256 nodes were used, and the Sigmoid function was used as the activation function.

[0149] Furthermore, in Example 2, the output-side warping shape of the newly manufactured metal strip is predicted using the above-mentioned warping shape prediction model and the warping shape prediction unit 59. In the operation parameter determination unit 60, the target value of the output-side warping shape (warping target value) is compared with the output-side warping shape predicted by the warping shape prediction unit 59. The warping target value is set to 6 mm. When the warping height of the output-side warping shape output from the warping shape prediction model is less than 6 mm, the operation parameter determination unit 60 determines it as OK (affirmative), and when the warping height of the output-side warping shape exceeds 6 mm, it is determined as NG (negative). When the output-side warping shape is determined to be NG by the operation parameter determination unit 60, the operation parameter determination unit 60 determines the operation parameter of the cooling device 30 that can reduce the warping height. The operation parameter determination unit 60 changes the operation condition of the cooling device 30 by outputting the operation parameter to the control controller 51. In this way, the operation parameter determined by the operation parameter determination unit 60 is reflected in the operation condition of the cooling device 30.

[0150] In Invention Examples 1 and 3, the amount of indentation of the constraining roller 33 is used as the operating parameter of the cooling device 30. In Invention Examples 2 and 4, the amount of displacement of the constraining roller 33 is used as the operating parameter of the cooling device 30. In Invention Example 5, both the amount of indentation of the constraining roller 33 and the amount of displacement of the constraining roller 33 are used as the operating parameters of the cooling device 30.

[0151] On the other hand, Comparative Example 1 is an example in which the metal strip is manufactured by releasing the restraining roller 33 in the cooling device 30 (in a state in which the restraining roller is separated by 160 mm in the horizontal direction) without predicting the warping shape on the output side. Comparative Example 2 is an example in which the metal strip is manufactured by using the restraining roller 33 in the cooling device 30 without predicting the warping shape on the output side and setting the pressing amount of the restraining roller to 0 mm in a manner consistent with the rolling line position. In addition, even if the pressing amount of the restraining roller is 0 mm, the metal strip is in a state of being pressed by an amount corresponding to its plate thickness.

[0152] In the invention examples 1 to 5, the comparative examples 1 and 2, 5 metal strips were manufactured respectively. At this time, the control of the operation parameter determination unit 60 and the control controller 51 was implemented at intervals of 10 m in the longitudinal direction of the metal strip during the rolling of the metal strip by the skin-pass rolling mill 40, and the output side warping shape in the longitudinal direction of the metal strip was controlled. The control conditions of the invention examples 1 to 5, the comparative examples 1 and 2 in the second embodiment and the measurement results of the maximum warping height of the metal strip are shown in Table 1.

[0153]

[0154] The maximum warping height shown in Table 1 indicates the maximum warping height of the five metal strips at various positions including the length direction. It should be noted that the values ​​of "constraint roller gap" and "constraint roller displacement" in the table indicate the range of values ​​set as the operating conditions of the cooling device 30. The displacement of the constraint roller 33 is the displacement of the constraint roller from the rolling line position when the back side of the metal strip is set to - and the surface side is set to +. As shown in Table 1, in Inventive Examples 1 to 5, the maximum warping height of the metal strip becomes smaller than the values ​​in Comparative Examples 1 and 2.

[0155] The maximum warping height of the metal strip in Comparative Example 1 is 35 mm, which is a very large value. In addition, in Comparative Example 2, the output side warping shape is not predicted as in Invention Examples 1 to 5, so the operating parameters of the cooling device 30 are fixed to a constant value. The maximum warping height of the metal strip in Comparative Example 2 is 9 mm, which is significantly smaller than that in Comparative Example 1, but the warping target value (6 mm) cannot be achieved in a part of the metal strip, resulting in an unqualified part. In this way, the following is confirmed: the output side warping shape of the metal strip is quickly predicted by the warping shape prediction method of the metal strip of the present embodiment, and the operating parameters of the cooling device are determined and set in a manner that the predicted output side warping shape becomes smaller, thereby the output side warping shape becomes less than the warping target value over the entire length of the metal strip in the longitudinal direction, and the warping shape control method of the present invention is effective for the manufacture of high-strength steel sheets.

[0156] Description of Reference Numerals

[0157] 1 Metal belt

[0158] 2 Unwinding machine

[0159] 3. Welding Machine

[0160] 4 Input side looper

[0161] 5 Preheating Zone

[0162] 6 Heating Zones

[0163] 7. Heat distribution area

[0164] 8 Cooling Zone

[0165] 9 Reheating Zone

[0166] 10 Expiration zone

[0167] 11 Final cooling zone

[0168] 12 Output side looper

[0169] 13 Surface polishing equipment

[0170] 14. Examination Table

[0171] 15 Tension reel

[0172] 16 Input side warpage shape measurement device

[0173] 17 Warp shape measurement device before cooling

[0174] 18 Output side warpage shape measurement device

[0175] 21 Furnace equipment

[0176] 22 Annealing equipment

[0177] 23 Reheating equipment

[0178] 24 Output side equipment

[0179] 30 Cooling equipment

[0180] 31 Pressing roller

[0181] 32 Cooling liquid injection device

[0182] 33 Constraint Roller

[0183] 34 Dipping Tank

[0184] 35 Cooling fluid

[0185] 36 Deflection roller

[0186] 37 Deflection roller

[0187] 38 Dewatering Roller

[0188] 39 Drying furnace

[0189] 40 Surface polishing mill

[0190] 41a, 41b working rolls

[0191] 42a, 42b Support rollers

[0192] 43 Pressing device

[0193] 44 Bearing box

[0194] 45 Load Detector

[0195] 46 Input side tension roller

[0196] 47 Input side tension meter

[0197] 48 Output side tension meter

[0198] 49 Output side tension roller

[0199] 50 Control computer

[0200] 51 Control Controller

[0201] 52 Warp shape control device

[0202] 54 Control Department

[0203] 55 Input

[0204] 56 Output

[0205] 57 Storage

[0206] 58 Acquisition Department

[0207] 59 Warpage shape prediction department

[0208] 60 Operation parameter determination unit

[0209] 61 Warpage shape prediction model generation unit

[0210] 62 Database

[0211] 63 Warpage Shape Prediction Model

[0212] 64 Water jet nozzles

[0213] 70 Laser rangefinder

Claims

1. A method for predicting the warpage shape of a metal strip, wherein the warpage shape of an output side is predicted in a continuous annealing device for a metal strip, wherein the continuous annealing device for a metal strip comprises: A heating device for heating the metal strip; a cooling device for cooling the metal strip heated by the heating device; A skin-pass rolling mill for correcting the shape of the metal strip after being cooled by the cooling device; and an input-side warping shape measuring device for measuring the warping shape of the metal strip on the upstream side of the skin-pass rolling mill, i.e., the input-side warping shape, between the cooling device and the skin-pass rolling mill, wherein the output-side warping shape is the warping shape of the metal strip on the downstream side of the skin-pass rolling mill, The warping shape prediction method of a metal strip predicts the output-side warping shape based on the input-side warping shape and at least one operating parameter of an operating parameter of the skin-pass rolling mill.

2. The method for predicting the warping shape of a metal strip according to claim 1, wherein: The warping shape prediction method of the metal strip uses a warping shape prediction model to predict the output side warping shape, and the warping shape prediction model takes as input data including the input side warping shape and at least one of the operating parameters of the surface rolling mill, and takes as output the output side warping shape.

3. The method for predicting the warping shape of a metal strip according to claim 1, wherein: The continuous annealing device further comprises a pre-cooling warpage shape measuring device, wherein the pre-cooling warpage shape measuring device is arranged on the upstream side of the cooling device. The warping shape prediction method of a metal strip predicts the output side warping shape based on the pre-cooling warping shape measured by the pre-cooling warping shape measuring device, the input side warping shape measured by the input side warping shape measuring device, and at least one operating parameter of the skin-pass rolling mill.

4. The method for predicting the warping shape of a metal strip according to claim 3, wherein: The warping shape prediction method of the metal strip uses a warping shape prediction model to predict the output side warping shape. The warping shape prediction model takes as input data including the warping shape before cooling, the input side warping shape and at least one of the operating parameters of the surface smooth rolling mill, and takes as output the output side warping shape.

5. A method for controlling the warping shape of a metal strip, wherein: When the output-side warping shape predicted by the warping shape prediction method of a metal strip according to any one of claims 1 to 4 exceeds a target value, the operating parameters of the cooling device are set so as to reduce the output-side warping shape.

6. The method for controlling the warping shape of a metal strip according to claim 5, wherein: The cooling device comprises: a cooling liquid spraying device having a plurality of nozzles for spraying cooling fluid from both sides of the metal strip; and at least one pair of restraining rollers for restraining the metal strip. The operating parameter of the cooling device is at least one of an amount of pressing of the pair of constraining rollers into the metal strip, an amount of offset between the pair of constraining rollers, and an amount of displacement of the pair of constraining rollers.

7. A method for manufacturing a metal strip, wherein: By using the warpage shape control method of a metal strip according to claim 5 or claim 6, a metal strip having a tensile strength of 980 MPa or more is manufactured.

8. A method for generating a warpage shape prediction model, for predicting an output-side warpage shape in a continuous annealing device for a metal strip, the continuous annealing device for a metal strip comprising: A heating device for heating the metal strip; a cooling device for cooling the metal strip heated by the heating device; A skin-pass rolling mill that corrects the shape of the metal strip after being cooled by the cooling device; an input-side warping shape measuring device that measures the warping shape of the metal strip on the upstream side of the skin-pass rolling mill, i.e., the input-side warping shape, between the cooling device and the skin-pass rolling mill; and an output-side warping shape measuring device that measures the warping shape of the metal strip on the downstream side of the skin-pass rolling mill, i.e., the output-side warping shape, on the downstream side of the skin-pass rolling mill, wherein: The method for generating the warp shape prediction model obtains a plurality of data groups, and generates the warp shape prediction model by machine learning using the obtained plurality of data groups as training data, wherein the data group takes as a group the input side warp shape measured by the input side warp shape measuring device, at least one of the operating parameters of the surface smooth rolling mill, and the output side warp shape measured by the output side warp shape measuring device, and the warp shape prediction model takes as input the input data including the input side warp shape and at least one of the operating parameters of the surface smooth rolling mill, and takes as output the output side warp shape.

9. The method for generating a warpage shape prediction model according to claim 8, wherein: The continuous annealing device further comprises a pre-cooling warpage shape measuring device, wherein the pre-cooling warpage shape measuring device is arranged on the upstream side of the cooling device. The method for generating the warp shape prediction model obtains a plurality of data groups, and generates the warp shape prediction model by machine learning using the obtained plurality of data groups as training data, wherein the data group takes the pre-cooling warp shape measured by the pre-cooling warp shape measuring device, the input side warp shape, at least one of the operating parameters of the surface smooth rolling mill, and the output side warp shape as a group, and the warp shape prediction model takes the input data including the pre-cooling warp shape, the input side warp shape, and at least one of the operating parameters of the surface smooth rolling mill as input, and takes the output side warp shape as output.

10. A warping shape control device for a metal strip, in a continuous annealing device for a metal strip, for controlling the warping shape at an output side, the continuous annealing device for a metal strip comprising: A heating device for heating the metal strip; a cooling device for cooling the metal strip heated by the heating device; A skin-pass rolling mill for correcting the shape of the metal strip after being cooled by the cooling device; and an input-side warping shape measuring device for measuring the warping shape of the metal strip on the upstream side of the skin-pass rolling mill, i.e., the input-side warping shape, between the cooling device and the skin-pass rolling mill, wherein the output-side warping shape is the warping shape of the metal strip on the downstream side of the skin-pass rolling mill, The warping shape control device of the metal strip comprises: an acquisition unit that acquires at least one operating parameter of the input-side warpage shape measured by the input-side warpage shape measuring device and the operating parameter of the skin-pass rolling mill; a warp shape prediction unit that predicts the output-side warp shape using a warp shape prediction model that takes as input input data including the input-side warp shape and at least one of the operating parameters of the skin-pass rolling mill and outputs the output-side warp shape; and The operation parameter determination unit determines the operation parameter of the cooling device so as to reduce the output-side warping shape when the output-side warping shape predicted by the warping shape prediction unit exceeds a target value.

11. The warping shape control device of a metal strip according to claim 10, wherein: The continuous annealing device further comprises a pre-cooling warpage shape measuring device, wherein the pre-cooling warpage shape measuring device is arranged on the upstream side of the cooling device. The acquisition unit further acquires the pre-cooling warpage shape measured by the pre-cooling warpage shape measurement device, The warping shape prediction unit predicts the output side warping shape using a warping shape prediction model, wherein the warping shape prediction model takes as input data the warping shape before cooling, the input side warping shape, and at least one of the operating parameters of the surface smooth rolling mill, and takes as output the output side warping shape.

Images