JP7644340B2 - Meandering control method and device - Google Patents

Description

The present invention relates to a meandering control method and a meandering control device for controlling the meandering of a rolled material during continuous rolling.

In continuous rolling, where the material to be rolled (e.g., steel) passes through multiple rolling mills in succession, the center of the width of the material to be rolled S may shift from the mill center (the center position C of the rolling mill in the width direction, i.e., the center position in the direction of the rotation axis of the work roll 11) as shown in FIG. 15, and the material to be rolled S may move toward the end of the work roll 11, a phenomenon known as meandering. When meandering occurs, the flatness of the material to be rolled decreases, which may lead to a decrease in product quality. In addition, if the amount of meandering is large, the tail end of the material to be rolled may come into contact with the side guide and bend, causing the material to be bitten into the subsequent rolling mill in a double-folded state, resulting in a defect known as squeezing. When squeezing occurs, the surface of the work roll is damaged by the bent material to be rolled, so it is necessary to stop the production line and perform maintenance work such as inspection, maintenance, or replacement of the work roll, which may reduce the operating rate of the production line.

In continuous rolling, when the material being rolled is bitten between the work rolls of both the front and rear rolling mills, tension acts on the material in the longitudinal direction and the material is restrained, making it difficult for large meandering to occur. Large meandering that would cause squeezing is likely to occur when the tail end of the material being rolled leaves the front rolling mill. It has also been known that the greater the reduction rate in the rolling mills, the more likely meandering is to occur.

Therefore, for example, Patent Document 1 below discloses a technology that detects the passage (tail drop) of the tail end of the rolled material through the previous (N-1) stage rolling mill, and opens the reduction position of the work rolls of the next (N) stage rolling mill (i.e., the roll gap between the upper and lower work rolls) at the timing when the tail drop is detected. According to the technology described in Patent Document 1, when a region of a predetermined length including the tail end of the rolled material (hereinafter also referred to as the tail end) passes through the Nth stage rolling mill, the reduction rate of the Nth stage rolling mill is reduced, so that the amount of meandering of the rolled material can be reduced.

Patent Document 2 also discloses a technique for simultaneously opening the roll gaps of the final rolling mill among multiple rolling mills or multiple consecutive rolling mills including the final rolling mill in accordance with the timing at which the tail end of the rolled material passes through. According to the technique described in Patent Document 2, the tail end of the rolled material is no longer rolled by the final rolling mill or multiple consecutive rolling mills including the final rolling mill, so that meandering and squeezing that occur during rolling can be suppressed.

Japanese Patent Application Publication No. 55-161505 JP 2001-105005 A Japanese Patent Application Publication No. 08-108207 Japanese Patent Application Publication No. 05-272959 Patent No. 3396428

Nakajima, Kikuma, Matsumoto, Kajiwara, Kimura, and Tagawa, "Study on the method of controlling meandering in hot strip rolling (first report)", Spring Conference on Plastic Processing, 1980, pp. 61-64 Edited by The Rolling Theory Committee, Production Technology Division, The Iron and Steel Institute of Japan, "Theory and Practice of Plate Rolling (Revised Edition)", September 30, 2010, published by The Iron and Steel Institute of Japan, page 94

However, in the method of controlling meandering by opening the roll gap when the tail end of the rolled material passes through the rolling mill, as in the techniques described in Patent Documents 1 and 2, the tail end of the rolled material is not rolled to the desired thickness. Therefore, the tail end of the rolled material is cut off as a defective product, which causes a decrease in production efficiency. In addition, in the technique described in Patent Document 2, the gap-up amount for opening the roll gap is set the same regardless of the type of material being rolled. For this reason, the roll gap is opened even when meandering does not actually occur, which may cause unnecessary thickness discrepancies in the tail end of the rolled material.

The present invention was made in consideration of the above problems, and the object of the present invention is to provide a new and improved meandering control method and meandering control device that can suppress meandering when the tail end of the rolled material is threaded, while suppressing the decrease in yield caused by thickness mismatch.

In order to solve the above problem, according to one aspect of the present invention, there is provided a meandering control method for controlling the meandering of a material being rolled by a rolling mill, the meandering control method including: a parallel stiffness target value setting step for setting a parallel stiffness target value for the rolling of the material being rolled based on rolling performance information of the material being rolled before the start of rolling of the material being rolled; a parallel stiffness calculation value calculation step for calculating a parallel stiffness calculation value for the material being rolled before the tail end of the material being rolled passes through the rolling mill; a gap-up amount calculation step for calculating a gap-up amount for increasing the roll gap of the rolling mill at the timing when the tail end of the material being rolled passes through so that the parallel stiffness calculation value is smaller than the parallel stiffness target value; and a control step for controlling the rolling mill so that the roll gap is increased by the determined gap-up amount at the timing when the tail end of the material being rolled passes through.

The gap-up amount calculation step includes a comparison step of comparing the parallel stiffness calculation value with the parallel stiffness target value, a gap-up amount change step of increasing the gap-up amount by a preset unit gap-up amount when the parallel stiffness calculation value is equal to or greater than the parallel stiffness target value, and a determination step of determining the gap-up amount when the parallel stiffness calculation value becomes smaller than the parallel stiffness target value as the gap-up amount of the roll gap of the rolling mill to be set at the timing when the tail end of the rolled material passes through. Each time the gap-up amount is changed in the gap-up amount change step, the parallel stiffness calculation value is calculated again in the parallel stiffness calculation step with the changed gap-up amount until the parallel stiffness calculation value becomes smaller than the parallel stiffness target value, and the parallel stiffness calculation value is compared with the parallel stiffness target value in the comparison step.

In the parallel stiffness calculation step, the parallel stiffness calculation value of the rolled material may be calculated using the rolling results of the bottom portion at the tail end of the rolled material.

The meandering control method may further include a learning step of learning the parallel stiffness target value. In the learning step, a parallel stiffness target value that minimizes the total value of the trouble occurrence cost due to the occurrence of rolling trouble and the quality defect cost due to the quality defect of the product is learned based on past manufacturing results.

In the learning step, the trouble occurrence cost is calculated from the occurrence rate of rolling troubles in past production records and the trouble occurrence cost unit price, and the quality non-conformity cost may be calculated from the number of quality non-conformities in products in past production records and the thickness quality non-conformity cost unit price due to thickness quality non-conformity.

In addition, in order to solve the above problem, according to another aspect of the present invention, there is provided a meandering control device that controls the meandering of a material being rolled by a rolling mill, the meandering control device including a manufacturing condition receiving unit that acquires the manufacturing conditions of the material being rolled, a performance collecting unit that acquires rolling performance information of the bottom part of the material being rolled by the rolling mill at the timing when the tail end of the material being rolled passes a predetermined position set upstream in the sheet passing direction relative to the rolling mill, and a setting calculation unit that calculates a parallel stiffness calculation value of the material being rolled before the tail end of the material passes through the rolling mill based on the manufacturing conditions of the material being rolled, the rolling performance information, and a parallel stiffness target value set according to the steel type of the material being rolled, and determines a gap-up amount that increases the roll gap of the rolling mill at the timing when the tail end of the material being rolled passes through so that the parallel stiffness calculation value is smaller than the parallel stiffness target value.

The meandering control device may further include a learning unit that learns the parallel stiffness target value for each steel type. Based on past manufacturing records, the learning unit learns the parallel stiffness target value for each steel type that minimizes the total value of the trouble occurrence cost due to the occurrence of rolling trouble and the quality defect cost due to the quality defect of the product.

As described above, the present invention can suppress meandering when threading the tail end of the rolled material, while also suppressing yield reductions caused by thickness mismatches.

1 is an explanatory diagram showing a configuration of a meandering control system according to an embodiment of the present invention; FIG. 2 is an explanatory diagram showing an example of a roll profile. FIG. 11 is an explanatory diagram showing an example of a parallel stiffness target value table. 10 is an explanatory diagram illustrating an example of manufacturing performance information stored in a manufacturing performance storage unit; FIG. FIG. 13 is an explanatory diagram showing an example of a configuration of a trouble occurrence rate table; FIG. 13 is an explanatory diagram showing an example of a configuration of a gap-up amount table. FIG. 13 is an explanatory diagram illustrating an example of a configuration of a cost unit price table. 6 is a flowchart showing a gap-up amount setting process according to the embodiment; FIG. 1 is a diagram illustrating a schematic view of a deformation region in a rolled material in which meandering and camber have occurred. FIG. 13 is a diagram showing deformation of a rolling mill when a material to be rolled meanders. FIG. 13 is a diagram showing the distribution of line loads. 5 is a flowchart showing a parallel stiffness target value learning process according to the present embodiment. 13 is a flowchart showing a parallel stiffness target value search process in the parallel stiffness target value learning process. FIG. 13 is an explanatory diagram showing one relationship between parallel stiffness and cost. FIG. 4 is an explanatory diagram showing the meandering of a rolled material.

The preferred embodiment of the present invention will be described in detail below with reference to the attached drawings. Note that in this specification and the drawings, components having substantially the same functional configuration are designated by the same reference numerals to avoid redundant description.

＜1. Overview＞
When rolling a material to be rolled by a rolling mill, it is effective to reduce the reduction ratio in order to suppress the occurrence of meandering and squeezing of the material to be rolled, and conventionally, control is performed to open the roll gap at the timing when the tail end of the material to be rolled passes through the rolling mill (for example, Patent Document 2). However, while opening the roll gap can suppress the occurrence of meandering and squeezing of the material to be rolled, it causes the plate thickness at the tail end to deviate significantly from the target plate thickness, which becomes a cause of yield loss. In addition, depending on the steel type of the material to be rolled, meandering and squeezing may not occur even if the roll gap is not opened when the tail end passes. In such a case, opening the roll gap will cause unnecessary yield loss.

In this embodiment, the necessity of opening the roll gap when the tail end passes through the rolling mill during rolling is determined for each material to be rolled, and the optimal roll gap is set for each material. This reduces yield loss while suppressing the occurrence of meandering and squeezing. Below, the configuration of the meandering control system and meandering control method for controlling the meandering of the rolling mill according to this embodiment will be described. Note that, below, opening the roll gap is also referred to as gap up, and the amount of roll gap opening is also referred to as the gap up amount. In addition, in the following description, for simplicity, the rolling mill installed at the most downstream in the sheet passing direction among the multiple rolling mills that make up the rolling facility will be described as the control target for controlling the gap up amount.

<2. Wobble control system>
First, the configuration of a meandering control system according to an embodiment of the present invention will be described with reference to Figs. 1 to 7. Fig. 1 is an explanatory diagram showing the configuration of a meandering control system 1 according to this embodiment. Fig. 2 is an explanatory diagram showing an example of a roll profile. Fig. 3 is an explanatory diagram showing an example of a parallel stiffness target value table 170. Fig. 4 is an explanatory diagram showing an example of manufacturing performance information stored in the manufacturing performance storage unit 210. Fig. 5 is an explanatory diagram showing an example of a configuration of a trouble occurrence rate table 221. Fig. 6 is an explanatory diagram showing an example of a configuration of a gap-up amount table 223. Fig. 7 is an explanatory diagram showing an example of a configuration of a cost unit price table 225.

The meandering control system 1 according to this embodiment is a system for opening the roll gaps of the rolling mills at the timing when the tail end of the rolled material S passes through in a rolling facility 10 equipped with multiple rolling mills (rolling mills f1 to f6 in the example of FIG. 1), and controlling the meandering of the rolled material S. As shown in FIG. 1, the meandering control system 1 is composed of a meandering control device 100 and a memory device 200. Here, the meandering control device 100 controls the rolling mill f6 installed at the most downstream position in the sheet passing direction.

[2-1. Meandering control device]
As shown in FIG. 1 , the meandering control device 100 includes a manufacturing condition receiving unit 110, a material tracking unit 120, a performance collecting unit 130, a setting calculation unit 140, a transmitting unit 150, a learning unit 160, and a parallel stiffness target value table 170.

The manufacturing condition receiving unit 110 receives the manufacturing conditions of the rolled material S to be rolled by the rolling equipment 10. The manufacturing conditions are managed by an external production management computer 30. The manufacturing condition receiving unit 110 receives the manufacturing conditions of the rolled material S from the production management computer 30 and outputs them to the setting calculation unit 140 as a manufacturing condition file.

The material tracking unit 120 detects the timing when the tail end of the rolled material S passes through the rolling equipment 10 during rolling. A plate passing sensor 50 is provided at a position a predetermined distance upstream of the rolling mill f1, which is installed at the most upstream side in the plate passing direction among the rolling mills f1 to f6 that make up the rolling equipment 10, to detect whether the rolled material S is on the line. The plate passing sensor 50 may be, for example, a laser sensor or an imaging device such as a DDC camera. The plate passing sensor 50 transmits the detection results to the material tracking unit 120 at any time. The material tracking unit 120 detects that the tail end of the rolled material S has passed based on the detection results from the plate passing sensor 50. For example, when the plate passing sensor 50 outputs an ON signal when the rolled material S is on the line and outputs an OFF signal when the rolled material S is not on the line, the material tracking unit 120 determines that the tail end of the rolled material S has passed the position of the plate passing sensor 50 when the detection result received from the plate passing sensor 50 changes from an ON signal to an OFF signal. When the material tracking unit 120 detects the passage of the tail end of the rolled material S, it instructs the performance information collection unit 130 to collect performance information.

The performance data collection unit 130 acquires rolling performance information before the tail end of the rolled material S passes through the rolling mill to be controlled. The performance data collection unit 130 acquires, for example, the rolling load [ton], the roll position (i.e., roll gap) [mm], and the bender load [ton] when rolling down the bottom part of the rolled material S as the rolling performance information. The rolling performance information is continuously acquired at a predetermined timing (e.g., every 10 μsec). Upon receiving an acquisition instruction, the performance data collection unit 130 may output the most recently acquired rolling performance information, or may output the average value of the most recent several times (e.g., 10 times) as the rolling performance information.

The performance collection unit 130 also acquires the roll profile calculation value [μm] of the work rolls 11 and 12 of the rolling mill as rolling performance information. An example of the roll profile of the work rolls 11 and 12 is shown in FIG. 2. The roll profile shows the change in the roll diameter in the roll width direction, from which the shape of the outer circumferential surface of the roll can be grasped. In FIG. 2, the horizontal axis shows the position in the roll width direction, and the vertical axis shows the roll profile amount. The roll profile amount represents the change in the roll radius relative to the initial roll radius, and is set to 0 mm when the roll radius is the same as the initial roll radius. When the roll radius is larger, the value becomes positive, and when the roll radius is smaller, the value becomes negative. As shown in FIG. 2, the roll profile of the work rolls 11 and 12 changes due to wear due to use, thermal expansion due to contact with the high-temperature rolled material S, and the like. Since such changes in the roll profile of the work rolls 11 and 12 affect the occurrence of meandering of the rolled material S, in this embodiment, the gap-up amount is set taking into account the roll profile of the work rolls 11 and 12, and the meandering of the rolled material S is controlled by changing the roll gap. The performance collection unit 130 outputs the collected information to the setting calculation unit 140 as a performance data file.

The setting calculation unit 140 calculates the amount of gap-up by which the roll gap is opened when the tail end of the rolled material S passes through the rolling mill to be controlled. The setting calculation unit 140 calculates the parallel stiffness (first type parallel stiffness), which is a physical quantity that is an index of the degree of meandering of the rolled material S, based on the manufacturing condition file input from the manufacturing condition receiving unit 110 and the actual data file input from the actual data collecting unit 130. The details of the calculation process of the parallel stiffness will be described later. Then, the setting calculation unit 140 compares the calculated parallel stiffness with a parallel stiffness target value to be described later, and calculates the amount of gap-up when the tail end of the rolled material S currently being rolled passes through the rolling mill so as to suppress the occurrence of meandering and reduce the yield. The details of the gap-up amount setting process by the setting calculation unit 140 will be described later. The setting calculation unit 140 outputs the calculated gap-up amount to the transmission unit 150.

The transmission unit 150 outputs the gap-up amount of the rolling mill calculated by the setting calculation unit 140 to the setting control device 70, which controls the rolling setting conditions of the rolling equipment 10. The setting control device 70 controls the roll gap of the rolling mill to be controlled so that the roll gap of the rolling mill is opened by the gap-up amount input from the transmission unit 150 at the timing when the rolled material S passes through the rolling mill.

The learning unit 160 learns and updates the parallel stiffness target value used when the setting calculation unit 140 determines the gap-up amount at a predetermined timing. The parallel stiffness target value is a value acquired based on past manufacturing results, and is set for each steel type of the rolled material S. The learning unit 160 calculates the optimal parallel stiffness that suppresses the occurrence of meandering of the rolled material S and also reduces the yield based on various data stored in the storage device 200 described later, and sets it as the parallel stiffness target value. In this embodiment, the parallel stiffness target value is periodically learned and updated, so that the gap-up amount can always be set based on the optimal parallel stiffness target value. Details of the learning process of the parallel stiffness target value by the learning unit 160 will be described later. The learning unit 160 records the acquired parallel stiffness target value for each steel type in the parallel stiffness target value table 170.

The parallel stiffness target value table 170 stores the parallel stiffness target value for each steel type learned by the learning unit 160. The parallel stiffness target value table 170 stores the steel type and the parallel stiffness target value in association with each other, for example, as shown in FIG. 3.

[2-2. Storage device]
The storage device 200 is configured as a storage medium for holding information used by the learning unit 160 of the meandering control device 100, and is configured, for example, of a ROM, a RAM, etc. As shown in FIG.

The manufacturing history storage unit 210 stores the manufacturing history information of the rolled material S in the past. As the manufacturing history information, the manufacturing history storage unit 210 stores information such as the coil number, plate thickness, plate width, steel type, plate thickness tolerance, parallel stiffness target value, parallel stiffness calculation value, gap-up amount, presence or absence of thickness mismatch (quality mismatch), and presence or absence of shrinkage (problem occurrence), as shown in FIG. 4, for example. For thickness mismatch and shrinkage occurrence, "0" indicates no, and "1" indicates yes. The manufacturing history information is recorded in the manufacturing history storage unit 210 each time rolling of the rolled material S is completed, or each time rolling of a predetermined number of rolled materials S is completed.

The table storage unit 220 stores information used to learn the parallel stiffness target value, and stores, for example, various tables set based on past performance. The table storage unit 220 stores, for example, a trouble occurrence rate table 221, a gap-up amount table 223, and a cost unit price table 225.

The trouble occurrence rate table 221 shows the relationship between the parallel stiffness calculated in past production and the trouble occurrence rate, and is generated based on the past production record information of the rolled material S stored in the production record storage unit 210. As shown in FIG. 5, for example, the trouble occurrence rate table 221 is information in which the trouble occurrence rate is calculated from the number of troubles that occurred for the total number of rolled pieces, for each calculated parallel stiffness category. In this embodiment, the number of rolled pieces in which narrowing occurred due to meandering is regarded as the number of troubles as rolling troubles, and the trouble occurrence rate is calculated.

The gap-up amount table 223 shows the relationship between the parallel stiffness difference, which is the difference between the parallel stiffness calculated in past production and the parallel stiffness target value, and the set gap-up amount, as shown in FIG. 6, for example. The gap-up amount table 223 is generated based on the past production performance information of the rolled material S stored in the production performance memory unit 210.

The cost unit price table 225 stores the unit price of costs incurred due to the yield of the rolled material S and rolling troubles. For example, as shown in FIG. 7, the cost unit price table 225 stores the unit price of thickness quality deviation costs caused by deviations in the thickness quality of a single rolled material, the unit price of trouble occurrence costs caused by the occurrence of rolling trouble causing operation to stop and the operating rate of the production line to decrease, and the like. Each cost unit price may be set, for example, as a unit price per rolled material. Each cost unit price is set to a value (for example, an average value, etc.) estimated from past performance.

<3. Meandering control method>
The meandering control system 1 according to this embodiment calculates and sets the roll-up opening amount (i.e., gap-up amount) during rolling when the tail end of the rolled material passes through the rolling mill to be controlled. This makes it possible to set an optimal gap-up amount for each rolled material, preventing the occurrence of meandering of the rolled material, and reducing unnecessary roll gap opening to prevent a drop in yield. A method for controlling meandering of the rolled material by the meandering control system 1 according to this embodiment will be described in detail below.

[3-1. Gap-up amount setting process]
First, the gap-up amount setting process during rolling by the meandering control system 1 according to this embodiment will be described with reference to Fig. 8. Fig. 8 is a flowchart showing the gap-up amount setting process according to this embodiment. The gap-up amount setting process according to this embodiment is executed by the setting calculation unit 140 of the meandering control device 100 of the meandering control system 1. The gap-up amount setting process is started at the timing when the material tracking unit 120 detects that the rolled material has passed the position of the plate passing sensor 50. Note that before the gap-up amount setting process is executed, the setting calculation unit 140 may receive a manufacturing condition file including the manufacturing conditions of the rolled material currently being rolled from the manufacturing condition receiving unit 110.

In the gap-up amount setting process, first, information required for calculating the gap-up amount is acquired (S102 to 108). As shown in FIG. 8, first, the manufacturing conditions of the material being rolled that is currently being rolled are acquired (S102). The setting calculation unit 140 acquires at least the target plate thickness and steel type from the manufacturing conditions of the material being rolled that is currently being rolled received from the manufacturing condition receiving unit 110. The target plate thickness is the target value of the plate thickness of the material being rolled after being rolled by the rolling mill, and is set according to the plate thickness required for the product. The roll gap during rolling is set based on the target plate thickness. The steel type represents the type of steel, such as general material, tinplate material, electromagnetic material, etc., and is determined based on the material components (for example, components contained in iron (slab) (for example, carbon, silicon, etc.)). The manufacturing method is determined based on the determined steel type. In step S102, the target plate thickness and steel type used in the gap-up amount setting process are acquired.

The setting calculation unit 140 also acquires a preset unit gap-up amount ΔS (S104). The unit gap-up amount ΔS is the amount of gap-up that is changed each time the parallel stiffness is calculated when determining the gap-up amount. The unit gap-up amount ΔS is set based on past performance. For example, the unit gap-up amount ΔS may be set to a value between 50 and 100 μm.

Then, the setting calculation unit 140 acquires the parallel stiffness target value HG_ref for the current rolling of the material to be rolled (S106). The parallel stiffness target value HG_ref is acquired from the parallel stiffness target value table 170 in which the parallel stiffness target value generated by the learning unit 160 is recorded. The setting calculation unit 140 sets the optimal parallel stiffness target value for the material to be rolled from the parallel stiffness target value table 170 based on the manufacturing conditions of the material to be rolled acquired in step S102. For example, when the parallel stiffness target value for each steel type is learned by the learning unit 160, the parallel stiffness target value table 170 describes the parallel stiffness target value for each steel type, as shown in FIG. 3. At this time, the setting calculation unit 140 acquires and sets the parallel stiffness target value HG_ref for the current rolling based on the steel type of the material to be rolled.

The setting calculation unit 140 also acquires rolling performance information and roll profile when the bottom part of the current rolled material passes through the rolling mill to be controlled, acquired by the performance collection unit 130 (S108). Here, the bottom part of the rolled material refers to a range of 10 to 20% from the tail end of the total length of the rolled material. In meandering control of the rolled material, the behavior of the bottom part of the rolled material as it passes through the rolling mill is important, and this behavior changes depending on, for example, the rolling load [ton], reduction position (i.e., roll gap) [mm], and bender load [ton] when rolling down the bottom part of the rolled material S.

The occurrence of meandering of the rolled material is also influenced by the roll profile of the work roll of the rolling mill. The roll profile is estimated and updated by a roll profile calculation process executed at a predetermined timing. The roll profile calculation process is executed by a separate arithmetic processing device (not shown). The roll profile changes mainly due to roll wear and thermal expansion of the roll. The roll wear can be understood, for example, by comparing the roll profile before and after rolling, and is updated every time the rolled material passes through the rolling mill. In addition, the roll shape that changes over time due to thermal expansion can be understood by estimating the roll profile at a fixed period (for example, every 10 seconds). For example, at the timing when the change in the roll profile amount due to roll wear or the change in the roll profile amount due to roll thermal expansion is calculated, the arithmetic processing device combines the latest change in the roll profile amount due to roll wear and the change in the roll profile amount due to roll thermal expansion to obtain the current roll profile.

The roll profile calculation process may use known roll profile calculation methods described in, for example, Patent Documents 3 and 4. Patent Document 3 describes a method for determining a correction coefficient for the roll thermal expansion amount by learning regarding the thermal expansion of the roll, correcting the error between the estimated value of the roll thermal expansion amount by a roll thermal expansion amount calculation model and the actual value during rolling operation, and estimating the roll thermal expansion amount with high accuracy. Patent Document 4 describes a method for measuring the change in the amount of distance between the roll axis center and the displacement detector arrangement point by moving three or more displacement detectors arranged in the circumferential direction of the roll approximately parallel to the roll axis direction, and acquiring the roll profile based on the measured value.

To calculate the gap-up amount, the setting calculation unit 140 obtains rolling performance information that is set when the bottom part of the rolled material passes through the rolling mill to be controlled during rolling, and the roll profile of the rolling mill calculated by the calculation processing device from the performance data file.

Once the information required to calculate the gap-up amount is acquired in steps S102 to S108, calculation of the gap-up amount begins (S110 to S120). In this embodiment, the gap-up amount is gradually increased, and the gap-up amount when the calculated parallel stiffness (also called the "parallel stiffness calculation value") becomes smaller than the parallel stiffness target value is adopted. This prevents the gap-up amount from being set unnecessarily large, and is expected to reduce thickness discrepancies in the rolled material.

First, an initial value of the gap-up amount in the calculation of parallel stiffness is set (S110). The initial value of the gap-up amount is set to 0. In other words, it is assumed that the tail end of the rolled material passes through the roll gap as it was during rolling. Therefore, in step S110, the setting variable n of the gap-up amount is set to 0. Then, the setting calculation unit 140 sets the gap-up amount when passing through the bottom part (S112). The gap-up amount is expressed by the product (ΔS × n) of the unit gap-up amount ΔS and the setting variable n.

Next, the setting calculation unit 140 calculates the parallel stiffness calculation value HG_bot when rolling the bottom part of the rolled material (S114). The parallel stiffness in this embodiment is a parameter defined as the first type parallel stiffness in the above-mentioned Non-Patent Document 1. As described in Patent Document 5, the first type parallel stiffness is a plate thickness wedge that can occur from the perspective of mill deformation when the rolled material meanders from the mill center by a unit amount, and can be easily calculated together with the deformation calculation of the roll system if the deformation characteristics of the rolling machine other than the roll system can be accurately understood. The setting calculation unit 140 calculates the parallel stiffness calculation value HG_bot based on the various information acquired in steps S102 to S108.

(Parallel stiffness calculation example)
Here, we will explain the parallel stiffness (first type parallel stiffness), the relationship between the parallel stiffness and the meandering amount of the rolled material, and the dependency of the parallel stiffness on the line load and bending force. The parallel stiffness is a constant that represents the difference in plate thickness in the width direction (wedge amount) when the center of the rolled material in the plate width direction is shifted by a unit amount from the mill center while the line load applied to the rolled material from the work rolls is constant in the left-right direction as described above. Therefore, first, we derive the relational equation between the meandering amount and the wedge amount, and then we derive the relational equation between the parallel stiffness and the wedge amount. Finally, we derive the relational equation between the meandering amount and the parallel stiffness by combining these relational equations.

(a. Derivation of the Relational Formula Between the Amount of Meander and the Amount of Wedge)
First, a relational expression between the amount of meandering and the amount of wedge is derived. As shown in Fig. 9, a model is considered in which meandering and camber occur in the rolled material in a predetermined region including directly below the work roll. Fig. 9 is a diagram showing a schematic view of the deformation region in the rolled material in which meandering and camber occur. In Fig. 9, the x-y coordinate system is taken in a horizontal plane (in a plane parallel to the plate surface of the rolled material), and the origin O is set to the center of the barrel of the work roll. The direction parallel to the rotation axis of the work roll (left-right direction) is set to the y-axis direction.

The motion of the rolled material can be considered as rigid body motion in a horizontal plane. If the position of a specific point in the rolled material at time t is x and y, then the velocities v and u of that point in the x and y directions can be expressed by the following equations (1) and (2).

Here, the feed direction of the rolled material immediately below the work roll is assumed to coincide with the circumferential direction of the work roll. Also, ω(t) and V(t) are the angular velocity of the rotation of the rolled material in the horizontal plane and the velocity at the origin, respectively. ω(t) and V(t) are functions of time only, but since the rolled material deforms at x=0, they can generally be different functions at the entry side and the exit side. Therefore, in the following, when ω(t) and V(t) are particularly distinguished between the entry side and the exit side, they will be represented by different symbols. That is, the angular velocity and velocity of the rotation of the rolled material at the entry side are respectively ω ₁ (t) and v ₁ (t), and the angular velocity and velocity of the rotation of the rolled material at the exit side are respectively ω ₂ (t) and v ₂ (t).

A certain time (for example, the time when the rolling mill bites into the leading edge of the rolled material) is taken as the reference (t=0), and the centerline shape in the width direction of the inlet material to be rolled at t=0 is expressed as _y0 = _f0 ( _x0 ) ( _x0 ≦ 0) using the coordinates ( _x0 , _y0 ) of a point on the centerline. The position of the point ( _x0 , _y0 ) on the centerline at time t is given by the following equations (3) and (4) by integrating the above equations (1) and (2).

Here, assuming that the left-right asymmetry of the meandering and camber is a first-order slight amount, it is sufficient to use the value of the formula (3) obtained by approximating the formula (3) to the zeroth order as x in the formula (4). Therefore, the x in the formula (4) is regarded as x(t)= _x0 + _v1t . By substituting the value of x(t)= _x0 + _v1t into the formula (4), the following formula (5) is obtained.

Furthermore, by eliminating x ₀ and y ₀ from the above equation (5) and rearranging it, the following equation (6) can be obtained, which represents the shape of the center line on the entry side at time t.

In the above formula (6), the first term on the right-hand side represents the influence of the translation of the rolled material in the x direction, the second term represents the influence of the rotation of the rolled material in a horizontal plane, and the third term represents the influence of the translation of the rolled material in the y direction. Since the amount of meandering _yc of the rolled material is defined by the y coordinate of the plate center line directly below the work roll (x=0), the following formula (7) is obtained by substituting x=0 into the above formula (6). The following formula (7) represents the relationship between the amount of meandering and the inlet rotational speed.

Except for the initial conditions, the above equation (7) is equivalent to the differential equation expressed by the following equation (8).

Here, the elongation λ of the rolled material can be expressed as λ = _v2 / _v1 . If the difference between the elongation λ on the work side and the elongation λ on the drive side (left-right difference in elongation λ) is _λdf , the change in the left-right sheet thickness difference and sheet thickness ratio (wedge ratio) from the inlet side to the outlet side (wedge ratio change) is ΔΨ, and the sheet width is b, the following equation (9) can be obtained from the law of conservation of volume in the rolled material.

Furthermore, since the forward and reverse rates of the rolled material are approximately proportional, it is assumed that the relationship expressed by the following formula (10) holds.

From the above equations (9) and (10), the relationship expressed by the following equation (11) can be obtained.

By combining the above equations (8) and (11), a relational expression between the amount of meandering y _c and the change in wedge ratio ΔΨ can be obtained.

(b. Deriving the Relational Equation Between Parallel Stiffness and Wedge Amount)
Next, a relational expression between the parallel stiffness and the wedge amount is derived. Consider a model in which the rolling mill is deformed due to the meandering of the material to be rolled as shown in Fig. 10. Fig. 10 is a diagram showing the state of deformation of the rolling mill when the material to be rolled meanders. The model shown in Fig. 10 is known as a parallel stiffness model of the rolling mill.

As shown in Fig. 10, when the rolled material meanders, the rolling mill deforms asymmetrically and a wedge is generated in the rolled material. Here, points A and A' shown in Fig. 10 are the set points of the reduction position, points B and B' are the positions of the axial centers of the backup rolls at the reduction points, Σ is the distance between the axial centers of the upper and lower backup rolls at the reduction point (BB'), a is the distance between the left and right reduction points, b is the plate width, _yc is the amount of meandering, P is the load on each side, and _K0 is the spring stiffness between A and B.

To simply represent the characteristics of the rolling mill in the model shown in Figure 10, we assume that the line load applied to the work rolls of the rolling mill has a linear distribution as shown in Figure 11. Figure 11 shows the distribution of line load.

As shown in Fig. 11, the strip width center is located at a position shifted from the origin by the amount of meandering _yc . Assuming that the line load is linearly distributed, the line load at the strip width center is _pc , and the difference between the line loads at both ends in the strip width direction is _pdf , the line load at one end in the strip width direction, y = _yc + b/2 (e.g., the end on the work side), can be expressed as p = _pc + (1/2) x _pdf . The line load at the other end in the strip width direction, y = _yc - b/2 (e.g., the end on the drive side), can be expressed as p = _pc - (1/2) x _pdf .

The difference in thickness between the left and right sides (i.e., the wedge amount) h _df that occurs when a line load as shown in Fig. 11 is applied can be expressed by a known calculation method such as a roll deformation calculation used when calculating a sheet crown. For example, the wedge amount h _df can generally be expressed by the following formula (12).

Here, as shown in Fig. 10, a is the distance between the left and right rolling points, i.e., the distance between the supports of the work rolls, and b is the plate width. Also, _Sdf is the difference in the opening degree (leveling value) between the left and right work rolls. Also, _gdf ( _yc , _pdf ) is a term that depends on the amount of meandering _yc and the left and right difference _pdf in line load, and is a term that is derived by a known calculation method such as roll deformation calculation as described above, but to avoid complicating the explanation, detailed explanation is omitted. By taking a first approximation of the above formula (12), the following formula (13) is obtained.

Here, E and D are the basic constants of the rolling mill related to the meandering phenomenon, and are called the first type parallel stiffness and the second type parallel stiffness, respectively. The first type parallel stiffness and the second type parallel stiffness have the following physical meanings.

The first type parallel stiffness E represents the wedge amount when the center of the rolled material in the width direction is shifted by a unit amount from the mill center when the line load applied to the rolled material from the work roll is constant in the width direction of the rolled material (i.e., there is no difference between the left and right line loads). The second type parallel stiffness D represents the wedge amount when the center of the rolled material in the width direction is located at the mill center and there is a difference in the line load applied to the rolled material from the work roll in the width direction of the rolled material (i.e., there is a difference between the left and right line loads). Both the first type parallel stiffness E and the second type parallel stiffness D are constants that depend on the width, line load, bending force, etc. of the rolled material.

Here, by differentiating the rolling load equation when there is no tension and no constraint on either the entry or exit side with respect to the entry plate thickness H and the exit plate thickness h, the following equation (14) can be obtained.

m is the plasticity coefficient per unit width, and _p0df is a disturbance term due to the difference in hardness between the left and right sides, etc. Furthermore, _Hdf is the wedge amount on the entry side, and _hdf is the wedge amount on the exit side.

By eliminating p _df from the above equations (13) and (14), the following equation (15) which expresses the relationship between the parallel stiffness (first type parallel stiffness) and the wedge amount can be obtained.

(c. Deriving the Relational Expression Between the Amount of Wobble and the Parallel Stiffness)
The relational expression between the amount of meandering and the parallel stiffness (first type parallel stiffness) can be obtained by eliminating h _df from the above expressions (8), (11), and (15), where the disturbance terms are combined into one term.

It can be said that the differential equation shown in the above formula (16) represents a control system for controlling the meandering amount _yc . From the above formula (16), it can be seen that whether the system is stable (i.e., the meandering amount _yc remains within a finite range) or unstable (i.e., the meandering amount _yc diverges) against a slight disturbance depends on the value of the parallel stiffness E. That is, it can be seen that when the value of the parallel stiffness E is sufficiently small (including negative values), the system is stable, and when the value of the parallel stiffness E is large, the system diverges.

For example, the parallel stiffness E can be calculated using the following equation (17) which is obtained by setting the difference in the left and right work roll gaps S _df =0 and the left and right line load difference p _df =0 in the above equation (13).

Here, the difference in plate thickness h _df between the left and right can be obtained by inputting the difference in opening angle S _df =0 between the left and right work rolls, the difference in line load p _df =0 between the left and right, and the amount of meandering y _c in a method (split model) in which the roll barrel is split in the barrel length direction and roll deformation is calculated, as described in, for example, Non-Patent Document 2. In this case, the amount of meandering y _c in the above formula (17) may be the same value as the amount of meandering y _c used as an input value for the split model.

Returning to the explanation of FIG. 8, the setting calculation unit 140 may calculate the parallel stiffness (first type parallel stiffness) E obtained by the above formula (17) as the parallel stiffness calculation value HG_bot based on the various information acquired in steps S102 to S108. Note that the method of calculating the parallel stiffness by the setting calculation unit 140 is not limited to the above example.

Then, the setting calculation unit 140 compares the parallel stiffness calculation value HG_bot calculated in step S114 with the parallel stiffness target value HG_ref obtained in step S106 (S116). The larger the parallel stiffness value, the more likely it is that meandering will occur, and if the parallel stiffness calculation value is smaller than the parallel stiffness target value, the possibility of the rolled material meandering is extremely low. Therefore, the setting calculation unit 140 changes the gap-up amount and sets the minimum gap-up amount at which the parallel stiffness calculation value HG_bot is smaller than the parallel stiffness target value HG_ref.

If the parallel stiffness calculation value HG_bot is equal to or greater than the parallel stiffness target value HG_ref in step S116, the gap-up amount setting variable n is increased by 1 (S118), the gap-up amount is set again (S112), and the parallel stiffness calculation value HG_bot is calculated with the updated gap-up amount (S114). The processing of steps S112 to S118 is repeated, and when the parallel stiffness calculation value HG_bot becomes smaller than the parallel stiffness target value HG_ref in step S116, the setting calculation unit 140 determines the gap-up amount at that time as the gap-up amount of the rolling mill to be controlled (S120). The determined gap-up amount is recorded in the calculation result storage file and transmitted to the setting control device 70 via the transmission unit 150, and the setting of the rolling mill is changed. Then, at the timing when the tail end of the rolled material passes through the rolling mill, the work rolls of the rolling mill are moved so that the roll gap of the rolling mill is increased by the determined gap-up amount.

The above describes the gap-up amount setting process according to this embodiment. According to this embodiment, an optimal parallel stiffness target value is set according to the material being rolled, and the gap-up amount is set so that the parallel stiffness is below the parallel stiffness target value. This makes it possible to set an optimal gap-up amount for each individual material.

[3-2. Parallel stiffness target value learning process]
(1) Processing Contents The parallel stiffness target value HG_ref acquired in step S106 in FIG. 8 is learned by the learning unit 160 at a predetermined timing when the manufacturing performance data is accumulated to a certain extent. For example, the learning process may be executed when the manufacturing performance data of 1000 or more rolled materials is accumulated, or the learning process may be executed once a month. In this way, by periodically learning and updating the parallel stiffness target value, the gap-up amount can be set more appropriately in the gap-up amount setting process shown in FIG. 8. The parallel stiffness target value learning process will be described in detail below with reference to FIGS. 12 to 14. FIG. 12 is a flowchart showing the parallel stiffness target value learning process according to this embodiment. FIG. 13 is a flowchart showing the parallel stiffness target value search process in the parallel stiffness target value learning process. FIG. 14 is an explanatory diagram showing one relationship between parallel stiffness and cost.

First, the learning unit 160 calculates the trouble occurrence rate for each category of the parallel stiffness target value for each steel type using past manufacturing performance data stored in the manufacturing performance memory unit 210 of the storage device 200 (S200). In this embodiment, the occurrence of a rolling trouble is assumed to be a reduction in the material being rolled. The occurrence or non-occurrence of a rolling trouble such as a reduction is recorded as manufacturing performance data, for example, as shown in FIG. 4. The learning unit 160 classifies the parallel stiffness target value into a range of predetermined values from the manufacturing performance data, and calculates the trouble occurrence rate by tallying up the number of rolled pieces of the material being rolled and the number of pieces that have experienced a reduction for each category. The learning unit 160 updates the trouble occurrence rate table 221 of the storage device 200 shown in FIG. 5 to the calculated trouble occurrence rate for each category of the parallel stiffness target value.

Then, the learning unit 160 learns the parallel stiffness target value for each steel type (S210s to S210e). First, the learning unit 160 performs initial setting of values required to execute the learning process (S220). Specifically, the initial parallel stiffness target value HG_ref, the initial total cost, the target parallel stiffness correction amount ΔHG_ref, and the parallel stiffness search maximum value HG_max are set.

The initial value of the target parallel stiffness HG_ref is set to 0. The initial total cost is the initial value of the total cost Tcost, which is an evaluation index of the learning result, and is appropriately set based on past manufacturing results. In this learning, in order to obtain the parallel stiffness target value when the total cost is minimum, the initial total cost may be set high so that the minimum value of the total cost (hereinafter referred to as the "minimum total cost Tcost_min") is updated to a small value based on the learning result. By setting the initial total cost high, it is possible to reliably update the minimum total cost Tcost_min. The initial total cost is set as the current minimum total cost Tcost_min. The parallel stiffness target value correction amount ΔHG_ref represents the increment for changing the parallel stiffness target value in the learning process. In addition, the parallel stiffness search maximum value HG_max is the maximum value of the parallel stiffness target value set during learning, and is used to determine the end of learning. The parallel stiffness target value correction amount ΔHG_ref and the parallel stiffness search maximum value HG_max are set in advance.

After completing the initial settings, the learning unit 160 executes a parallel stiffness target value search process to determine the parallel stiffness target value (S230). As shown in FIG. 13, the parallel stiffness target value search process learns the parallel stiffness target value from 0 to the parallel stiffness search maximum value HG_max in increments of the parallel stiffness target value correction amount ΔHG_ref (S230s to S230e).

First, the learning unit 160 calculates the parallel stiffness difference from the parallel stiffness target value and the parallel stiffness calculation value (S231). The parallel stiffness difference is the difference between the parallel stiffness target value and the parallel stiffness calculation value, and is calculated for each data item using the parallel stiffness target value and the parallel stiffness calculation value included in the manufacturing performance data.

Next, the learning unit 160 acquires the corresponding gap-up amount from the gap-up amount table 223 based on the calculated parallel stiffness difference (S232). The learning unit 160 also acquires the corresponding trouble occurrence rate calculated in step S200 from the trouble occurrence rate table 221 based on the parallel stiffness target value of each manufacturing performance data (S233). Furthermore, the learning unit 160 acquires the thickness quality non-conformity cost unit price and the trouble occurrence cost unit price from the cost unit price table 225 (S234). The thickness quality non-conformity cost unit price is the cost per rolled material that occurs when the plate thickness of the product falls outside the tolerance. In this embodiment, the cost unit price that occurs when drawing occurs is taken into consideration as the trouble occurrence cost unit price. At least the thickness quality non-conformity cost unit price is set for each steel type.

The learning unit 160 also calculates the number of rolled materials that have a thickness quality that does not meet the criteria based on the gap-up amount acquired in step S232 and the thickness tolerance that is set in advance as a product specification (S235). The larger the gap-up amount, the more the thickness of the rolled material after rolling deviates from the target thickness of the product. Therefore, the learning unit 160 determines whether the thickness is outside the tolerance based on the set gap-up amount, and counts the number of rolled materials that are determined to be outside the tolerance (hereinafter also referred to as the "number of thickness deviations"), which is the number of rolled materials that have a thickness quality that does not meet the criteria.

Then, the learning unit 160 calculates the total cost Tcost, which is the sum of the quality non-compliance cost and the trouble occurrence cost that will occur when the current parallel stiffness target value is set, for the manufacturing history data of the rolled material used in the current learning (for example, manufacturing history data of 1,000 or more rolled materials or manufacturing history data for one month) (S236). The total cost Tcost is expressed by the following formula (18).

Tcost = quality non-conformity cost + trouble occurrence cost
= Number of thickness deviations x unit cost of thickness quality deviations
+ Trouble occurrence rate × number of rolled pieces × trouble occurrence cost per unit ... (18)

That is, the quality non-compliance cost is obtained by multiplying the number of thickness deviations calculated in step S235 by the unit cost of thickness non-compliance obtained in step S234. The trouble occurrence cost is obtained by multiplying the number of rolled pieces by the trouble occurrence rate obtained in step S233, the number of rolled pieces, and the trouble occurrence cost obtained in step S234, with the number of production performance data used for learning being the number of rolled pieces.

Then, the learning unit 160 compares the total cost Tcost calculated in step S236 with the current minimum total cost Tcost_min (S237). If the total cost Tcost is smaller than the minimum total cost Tcost_min, the minimum total cost Tcost_min is updated to the total cost Tcost calculated in step S236, and the parallel stiffness target value HG_ref at this time is set as the total cost minimum parallel stiffness target value HG_ref_min (S238). On the other hand, if the total cost Tcost is equal to or greater than the minimum total cost Tcost_min, the minimum total cost Tcost_min and the total cost minimum parallel stiffness target value HG_ref_min are not updated.

Then, the parallel stiffness target value HG_ref is updated by adding the parallel stiffness target value correction amount ΔHG_ref to the parallel stiffness target value HG_ref, and the process of FIG. 13 is repeated until the parallel stiffness target value HG_ref exceeds the parallel stiffness search maximum value HG_max.

Returning to the explanation of FIG. 12, when the parallel stiffness target value search process is completed, the learning unit 160 stores the final total cost minimum parallel stiffness target value HG_ref_min as the parallel stiffness target value for the steel type in the parallel stiffness target value table 170 (S240). In the parallel stiffness target value search process of FIG. 13, a parallel stiffness target value that minimizes the total cost is searched for. The total cost minimum parallel stiffness target value HG_ref_min finally obtained in the parallel stiffness target value search process represents the parallel stiffness target value that minimizes the total cost.

Here, FIG. 14 shows a relationship between the parallel stiffness target value and the quality non-conformity cost, the trouble occurrence cost, and the total cost. As described above, the parallel stiffness represents a plate thickness wedge that may occur from the viewpoint of mill deformation when the rolled material meanders from the mill center by a unit amount. Therefore, the larger the parallel stiffness value, the more likely the rolled material is to meander. Also, the larger the gap-up amount, the more the occurrence of meandering is suppressed, but the plate thickness of the product is out of tolerance, resulting in quality non-conformity. That is, as shown in FIG. 14, the larger the parallel stiffness target value set, the more the quality non-conformity cost is suppressed, while the trouble occurrence cost increases, and there is a trade-off between the quality non-conformity cost and the trouble occurrence cost. Therefore, in this embodiment, the case where the total cost, which is the sum of the quality non-conformity cost and the trouble occurrence cost, is minimized is considered as the optimal manufacturing condition, and the parallel stiffness target value at this time, i.e., the total cost minimum parallel stiffness target value HG_ref_min, is determined as the parallel stiffness target value HG_ref for the steel type. In the example of Figure 14, the total cost is smallest when the parallel stiffness is 2, so the parallel stiffness target value is set to 2.

The learning unit 160 executes the above process for all steel types to obtain the parallel stiffness target value for each steel type. In this manner, in this embodiment, the parallel stiffness target value is periodically updated for each steel type, so that the gap-up amount can be set more appropriately in the gap-up amount setting process shown in FIG. 8.

(2) Specific Example Below, as an example of the parallel stiffness target value learning process shown in Fig. 12 and Fig. 13, an example of learning the parallel stiffness target value for steel type A will be shown. In this example, it is assumed that the number of rolled products of steel type A in one month is 430, of which 50 products have had a rolling trouble caused by a reduction. The parallel stiffness target value is set in the range of 0 to 5, and when the parallel stiffness target value is divided into five categories, 0 or more and less than 1, 1 or more and less than 2, 2 or more and less than 3, 3 or more and less than 4, and 4 or more and less than 5, the number of rolled products, the number of products with a reduction, and the trouble occurrence rate for each category are as shown in Table 1 below. The following information is acquired in step S200 of Fig. 12.

Next, after initial settings are made in step S220, a search is made for the optimal parallel stiffness target value for steel type A in step S230. First, the parallel stiffness difference of each rolled material is calculated from the production performance data of steel type A produced over a one-month period, and the gap-up amount corresponding to the calculated parallel stiffness difference is obtained from the gap-up amount table. Here, it is assumed that the gap-up amount table is set as shown in Table 2 below.

The trouble occurrence rate table is also referenced to obtain the trouble occurrence rate corresponding to the parallel stiffness target value of each production performance data. Here, the trouble occurrence rate used is the trouble occurrence rate where the rolled material is squeezed. The obtained trouble occurrence rate is calculated for each parallel stiffness target value, for example, as shown in Table 3 below.

Furthermore, the cost unit price for thickness quality non-compliance and the cost unit price for trouble occurrence are obtained from the cost unit price table. Each cost unit price is set, for example, as shown in Table 4 below. Each cost unit price is expressed as the cost incurred per piece of rolled material, for example, as shown in Table 4. Note that in Table 4, each cost unit price is expressed as a ratio; in this example, if the cost when one piece of thickness quality non-compliance occurs is set to 1, the cost incurred when a trouble occurs will be 5 times that amount.

The number of thickness deviations is also calculated for each piece of manufacturing performance data. For example, when the plate thickness tolerance is 200 μm, materials with gap-up amounts set to greater than 200 μm are outside the tolerance range, resulting in thickness quality failure. The number of thickness deviations obtained is calculated for each parallel stiffness target value, for example, as shown in Table 5 below.

Then, using the above acquired values, the thickness quality non-conformity cost, trouble occurrence cost, and total cost are calculated. For example, when the number of rolled pieces is 430 and the parallel stiffness target value HG_ref is 0, the trouble occurrence rate is 0.02% and the number of thickness deviations is 340. Therefore, the total cost Tcost is 3.83 (= 340 x 0.01 + 2% x 430 x 0.05) from the above formula (18). Also, when the parallel stiffness target value HG_ref is 1, the trouble occurrence rate is 0.06% and the number of thickness deviations is 250. Therefore, the total cost Tcost is 3.79 (= 250 x 0.01 + 6% x 430 x 0.05). Similarly, when the parallel stiffness target value is 5, the calculation is repeated, and the results are as shown in Table 6 below. From Table 6, when the parallel stiffness target value is 2, the total cost is the smallest. Therefore, the parallel stiffness target value of steel type A is set to 2.

<4. Summary>
The configuration of the meandering control system according to the embodiment of the present invention and the meandering control method using the system have been described above. According to this embodiment, before the tail end of the rolled material passes through the rolling mill to be controlled, the gap-up amount of the work rolls when the tail end passes through is calculated and set. The gap-up amount is set using a parallel stiffness calculation value that indicates the ease of meandering of the rolled material so that the parallel stiffness calculation value is smaller than the parallel stiffness target value. The parallel stiffness target value is set for each steel type as a cost-optimal value for suppressing the occurrence of narrowing and thickness quality irregularities. Therefore, by setting the gap-up amount so that the parallel stiffness calculation value is smaller than the parallel stiffness target value, it is possible to realize optimal meandering control that suppresses the occurrence of meandering and narrowing of the rolled material and suppresses costs due to product quality and rolling troubles.

The parallel stiffness target value may also be periodically updated through learning. This allows the optimal parallel stiffness target value to always be set based on the latest manufacturing results, making it possible to achieve more optimal meandering control.

The above describes in detail preferred embodiments of the present invention with reference to the attached drawings, but the present invention is not limited to such examples. It is clear that a person with ordinary knowledge in the technical field to which the present invention pertains can conceive of various modified or revised examples within the scope of the technical ideas described in the claims, and it is understood that these also naturally fall within the technical scope of the present invention.

For example, in the above embodiment, the rolling mill to be controlled by the meandering control device is the rolling mill installed at the most downstream in the sheet passing direction among the multiple rolling mills that make up the rolling facility, but the present invention is not limited to this example. For example, one or multiple rolling mills that make up the rolling facility may be controlled. Since meandering is greatly affected by the rolling state of the rolling mill downstream in the sheet passing direction, it is preferable to control the gap-up amount of at least the rolling mill most downstream in the sheet passing direction. In addition, it is preferable to control the gap-up amount of the rolling mills in order from the downstream side in the sheet passing direction. When controlling the gap-up amount of multiple rolling mills, the gap-up amount is set according to the mass flow of the material to be rolled by each rolling mill.

In the above embodiment, the gap-up amount setting process is started when the material tracking unit detects that the material being rolled has passed the position of the plate passing sensor, but the present invention is not limited to this example. For example, the gap-up amount setting process may be executed when the tail end portion passes the rolling mill that is the most upstream in the plate passing direction in the rolling equipment. The timing when the tail end portion of the material being rolled has passed the rolling mill can be determined when the load detected by a load detection device that detects the load on the roll, such as a load cell, falls below a predetermined value.

REFERENCE SIGNS LIST 1 meandering control system 11, 12 work roll 30 production management computer 50 strip threading sensor 70 setting control device 100 meandering control device 110 manufacturing condition receiving section 120 material tracking section 130 performance collection section 140 setting calculation section 150 transmission section 160 learning section 170 parallel stiffness target value table 200 storage device 210 manufacturing performance storage section 220 table storage section 221 trouble occurrence rate table 223 gap-up amount table 225 cost unit table 230 gap-up amount table 240 cost unit table

Claims (7)

A meandering control method for controlling meandering of a rolled material being rolled by a rolling mill, comprising the steps of:
As the parallel stiffness, which is a physical quantity that is an index showing the degree of meandering of the rolled material, when the line load applied to the rolled material is constant in the plate width direction of the rolled material and the center of the rolled material in the plate width direction is shifted by a unit amount from the mill center, a first type parallel stiffness is used, which is indicated by a difference in plate thickness in the plate width direction when the center of the rolled material in the plate width direction is shifted by a unit amount from the mill center.
a parallel stiffness target value setting step of setting a first type parallel stiffness target value in rolling of the rolled material based on rolling performance information of the rolled material before starting rolling of the rolled material;
a parallel stiffness calculation step of calculating a first type parallel stiffness calculation value of the rolled material before the tail end portion of the rolled material passes through the rolling mill ;
a gap-up amount calculation step of calculating a gap-up amount for increasing the roll gap of the rolling mill at a timing when the tail end of the rolled material passes through the roll gap so that the first type parallel stiffness calculated value is smaller than the first type parallel stiffness target value;
a control step of controlling the rolling mill so that the roll gap is increased by the determined gap-up amount at a timing when the tail end of the rolled material passes through;
The meandering control method includes: The gap-up amount calculation step includes:
a comparison step of comparing the first type parallel stiffness calculation value with the first type parallel stiffness target value;
a gap-up amount changing step of increasing the gap-up amount by a preset unit gap-up amount when the first parallel stiffness calculation value is equal to or greater than the first parallel stiffness target value;
a determining step of determining the gap-up amount when the first type parallel stiffness calculated value becomes smaller than the first type parallel stiffness target value as a gap-up amount of the roll gap of the rolling mill to be set at a timing when the tail end of the rolled material passes through;
Including,
2. The meandering control method according to claim 1, wherein each time the gap-up amount is changed in the gap-up amount changing step, the first type parallel stiffness calculation value is calculated again in the parallel stiffness calculation step using the changed gap-up amount until the first type parallel stiffness calculation value becomes smaller than the first type parallel stiffness target value, and the first type parallel stiffness calculation value is compared with the first type parallel stiffness target value in the comparison step. The meandering control method according to claim 1 or 2, wherein in the parallel stiffness calculation step, a first type parallel stiffness calculation value of the rolled material is calculated using the rolling results of the bottom portion of the rolled material on the tail end side. The method further includes a learning step of learning the first type parallel stiffness target value,
The meandering control method according to any one of claims 1 to 3, wherein in the learning step, the first type parallel stiffness target value that minimizes a total value of a trouble occurrence cost due to the occurrence of a rolling trouble and a quality defect cost due to a quality defect of a product is learned based on a past manufacturing record. In the learning step,
The trouble occurrence cost is calculated from the occurrence rate of the rolling trouble in the past manufacturing results and a trouble occurrence cost unit price,
The meandering control method according to claim 4 , wherein the quality non-compliance cost is calculated from the number of quality non-compliances of the products in the past production records and a unit cost of thickness quality non-compliance due to thickness quality non-compliance. A meandering control device for controlling the meandering of a rolled material being rolled by a rolling mill,
As the parallel stiffness, which is a physical quantity that is an index showing the degree of meandering of the rolled material, when the line load applied to the rolled material is constant in the plate width direction of the rolled material and the center of the rolled material in the plate width direction is shifted by a unit amount from the mill center, a first type parallel stiffness is used, which is indicated by a difference in plate thickness in the plate width direction when the center of the rolled material in the plate width direction is shifted by a unit amount from the mill center.
A manufacturing condition receiving unit that acquires manufacturing conditions of the rolled material;
a performance collection unit that acquires rolling performance information of the bottom portion of the rolled material by the rolling mill at a timing when the tail end portion of the rolled material passes through a predetermined position set upstream in a sheet passing direction with respect to the rolling mill;
a setting calculation unit that calculates a first type parallel stiffness calculation value of the rolled material before the tail end of the rolled material passes through the rolling mill based on the manufacturing conditions of the rolled material, the rolling performance information, and a first type parallel stiffness target value set according to the steel type of the rolled material, and determines a gap-up amount for increasing the roll gap of the rolling mill at the timing when the tail end of the rolled material passes through so that the first type parallel stiffness calculation value is smaller than the first type parallel stiffness target value;
A meandering control device comprising: A learning unit that learns the first type parallel stiffness target value for each steel type is further provided,
The meandering control device according to claim 6, wherein the learning unit learns, for each steel type, the first type parallel stiffness target value that minimizes the total value of the trouble occurrence cost due to the occurrence of rolling trouble and the quality defect cost due to the quality defect of the product, based on past manufacturing results.

Images