JP2015036154A - Flying thickness change method in cold tandem rolling mill - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a flying thickness changing method in a cold tandem rolling mill which can accurately perform a flying thickness change when performing continuous rolling by the cold tandem rolling mill.SOLUTION: When performing a flying thickness change when welding a plurality of rolled materials different in conditions and continuously rolling them by a cold tandem rolling mill, a roll rolling distance in each rolling stand of the cold tandem rolling mill is divided into a plurality of rolling distance zones, an actual value of a friction coefficient or a rolling load is collected at each material quality and/or plate thickness of the rolled material with respect to a plurality of the rolling distance zones, a representative value of the friction coefficient or the rolling load in a plurality of the rolling distance zones at each material quality and/or plate thickness of the rolled material is obtained, an approximate curve is obtained on the basis of the respective representative value, the rolling load of each rolling mill is predicted by using the approximate curve, and the flying thickness change is performed by using the predicted rolling load.

Description

  The present invention relates to a method for changing a running plate thickness in a cold tandem rolling mill in which a cold rolled steel plate such as a tin plate or an automobile steel plate is constituted by two or more continuous rolling stands.

  In continuous cold rolling, multiple rolled materials with different conditions such as material and sheet thickness are welded and rolled continuously with a cold tandem rolling mill. Therefore, when the welding point passes between the rolls of each rolling stand, the plate thickness change is performed to change the plate thickness by changing the work roll reduction position in the rolling stand. At that time, by executing appropriate control with a process computer and its lower-level PLC (Programmable Logic Controller), a technology that suppresses troubles such as plate breakage and efficiently changes the running plate thickness is applied. .

  To do so, first, the process computer accurately predicts the rolling load and advance rate when changing the running plate thickness, calculates the rolling position of each rolling stand, and based on that, the PLC sends the rolling position change timing to the motor, etc. It is necessary to issue a command for the amount of change in the reduction position. As a technique for improving the prediction accuracy such as rolling load at that time, Patent Document 1 describes a friction coefficient using the actual data of the rolling mill when the friction coefficient between the roll and the material to be rolled is calculated and predicted. It shows a method of improving the prediction accuracy such as rolling load by learning the coefficient.

  Next, it is necessary to appropriately control the roll gap and roll peripheral speed when changing the running plate thickness by PLC, but in Patent Document 2, by tracking the running plate thickness change point by the flow rate of the plate This shows how to control the rolling position and tension of each stand in a timely manner.

  Further, in Patent Document 3, according to a mathematical model obtained by mathematically modeling the characteristics of a rolling mill, each setting value of the rolling mill is determined, and when rolling while correcting an error of the mathematical model by feeding back a rolling result, A method and apparatus for improving the accuracy of predictive control such as rolling load using both long-term learning and short-term learning are shown.

  Various approximate formulas are used for factors that have a large effect on rolling load. As a typical example, the effect of the reduction in the friction coefficient between the roll and the material to be rolled due to rolling roll wear is shown after the work roll is replaced. The rolling distance (roll rolling distance) is approximated to an exponential function or the like (for example, Patent Document 4, Formula (15)). Such an approximate expression is obtained by setting calculation (tuning) based on past performance data at the time of replacement with a new roll, and normally the approximate expression is continuously used unless there is a large change in operating conditions. .

JP 2006-122980 A Japanese Patent No. 2981797 Japanese Patent Laid-Open No. 11-707 JP-A-62-72409

  However, in actuality, the friction coefficient does not change according to the approximate expression due to various factors, and an error occurs between the actual friction coefficient and the friction coefficient predicted from the approximate expression as shown in FIG. The error increases as the roll rolling distance increases. As a result, the rolling load prediction accuracy at the time of changing the running plate thickness is deteriorated, and it becomes difficult to change the running plate thickness with high accuracy.

  The present invention has been made in view of such circumstances, and it is possible to change the running plate thickness with high accuracy when continuously rolling with a cold tandem rolling mill. It is an object of the present invention to provide a method for changing the thickness of an intermediate sheet.

  In order to solve the above-described problem, in the first aspect of the present invention, when a plurality of rolled materials having different conditions are welded and continuously rolled by a cold tandem rolling mill constituted by two or more continuous rolling stands. Is a method of changing the thickness of the rolling plate, and the roll rolling distance of the rolling roll in each rolling stand of the cold tandem rolling mill is divided into a plurality of rolling distance categories, and the material of the material to be rolled and For each of the sheet thicknesses, for each of the plurality of rolling distance sections, collect the friction coefficient between the rolling roll and the material to be rolled or the actual value of the rolling load to collect the material and / or sheet thickness of the material to be rolled. For each, obtain a representative value of the friction coefficient or the rolling load in the plurality of rolling distance sections, obtain an approximate curve based on the representative value, predict the rolling load of each rolling mill using the approximate curve, Predict Providing-fly thickness changing in tandem cold rolling mill and performing-fly thickness change with rolling load.

  As the representative value in the plurality of rolling distance sections, the friction coefficient collected in each rolling distance section or the average value of the actual values of the rolling load can be used.

  According to the present invention, the roll rolling distance is divided into a plurality of rolling distance sections, and for each material and / or sheet thickness of the material to be rolled, the friction coefficient between the rolling roll and the material to be rolled for the plurality of rolling distance sections. Or, by collecting the rolling load value and approximating the influence of the friction coefficient or rolling load based on it, even if the roll rolling distance increases, it can be avoided that the error of the friction coefficient becomes large as before. . For this reason, it is possible to predict the rolling load with high accuracy, and it is possible to accurately change the running plate thickness based on the prediction.

It is a schematic block diagram which shows an example of the cold tandem rolling mill with which this invention is applied. It is a flowchart which shows the 1st example of the plate | board thickness change method in a cold tandem rolling mill. It is a flowchart which shows the 2nd example of the plate | board thickness change method in a cold tandem rolling mill. It is a figure which compares and shows the approximate curve which shows the relationship between the roll rolling distance and friction coefficient obtained based on this invention, and the conventional approximate expression. It is a figure which shows the difference | error between the actual friction coefficient with respect to a roll rolling distance, and the friction coefficient estimated from the approximate expression.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.
[Configuration of cold tandem rolling mill]
FIG. 1 is a schematic configuration diagram showing an example of a cold tandem rolling mill to which the present invention is applied.

  The cold tandem rolling mill 1 according to the present embodiment welds a plurality of rolled materials having different materials (steel types) and / or plate thicknesses and continuously cold-rolls them. Has two rolling stands. Each rolling stand has a pair of work rolls (rolling rolls) 11 for rolling the material to be rolled 2 and a pair of backup rolls 12 for supporting them. The work roll 11 is driven by a roll driving device (not shown) constituted by an electric motor or the like, and continuously feeds the material 2 to be rolled while conveying the material 2 to be rolled from the first rolling stand to the fifth rolling stand. Cold rolling. Typical examples of the material 2 to be rolled include a tin plate and an automobile steel plate. Moreover, it may consist of not only a steel plate but another metal plate.

  The backup roll 12 on the upper side of each rolling stand is provided with a hydraulic or electric motor type reduction device 13. The reduction device 13 adjusts the reduction position of the work roll 11. A reduction position detector 14 is provided in the vicinity of the reduction apparatus 13. The reduction position detector 14 may detect the reduction position directly from the position of the reduction apparatus 13, or in the case of an electric motor type, may detect based on the number of rotations.

  A rolling load detector 15 composed of a load cell is provided under the backup roll 12 below each rolling stand. Further, a tension meter 16 for detecting the tension of the rolled material 2 is provided between the stands. Furthermore, a thickness gauge 17 for detecting the thickness of the rolled material 2 is provided on the exit side of the first rolling stand and the exit side of the fifth rolling stand.

  The cold tandem rolling mill 1 includes a process computer 20 that controls the rolling process and a PLC 30 that is a lower-level controller. The process computer 20 calculates the pass schedule, the predicted rolling load value of each rolling stand, and the roll gap target value according to the upper direction of the material slab size, product target size, etc. given from the host computer by the setting calculation means, and calculates these values. Set to lower PLC30. The PLC 30 continuously collects rolling data such as the rolling load detected by the rolling load detector 15 and the tension from the tension meter 16 and outputs it to the process computer 20.

  In the present embodiment, a rolling load model 21 is incorporated in the process computer 20, thereby predicting a rolling load per unit width when the running plate thickness is changed, and the running plate thickness is based on the predicted value. The reduction position of the work roll 11 in each rolling stand at the time of change is calculated. Further, the PLC 30 incorporates a running plate thickness change program 31, and gives a command of the reduction position change timing and change amount to the reduction device 13 from the reduction position calculated by the process computer 20.

[First example of a method for changing the running plate thickness in a cold tandem rolling mill]
Next, the 1st example of the running plate | board thickness change method in the cold tandem rolling mill 1 comprised as mentioned above is demonstrated.

  In the cold tandem rolling mill 1 configured as described above, a plurality of rolled materials having different materials (steel types) and / or plate thicknesses are welded and continuously rolled. At this time, since the roll-down position of the work roll in each rolling stand is different before and after the welded portion, the running plate thickness is changed to change the roll-down position of the work roll in the rolling stand.

  In order to change the running plate thickness, the rolling load model 21 as a premise is constructed in consideration of the effect of the reduction of the friction coefficient between the roll and the material to be rolled due to roll wear, and the running plate thickness is changed based on that. Predict the rolling load per unit width of time.

Therefore, conventionally, an approximate expression expressed by an exponential function as shown in the following equation (1) is used to tune based on past performance data and to influence the decrease in the friction coefficient μ between the roll and the material to be rolled. Is approximated using
μ = μ ₀ + A · exp (B · W) (1)
However, A and B are constants and W is a roll rolling distance.

  However, as shown in FIG. 5 described above, an error occurs between the actual friction coefficient and the friction coefficient predicted from the approximate expression, and the error increases as the roll rolling distance increases. As a result, the prediction accuracy of the rolling load is low, and it is difficult to change the running plate thickness with high accuracy.

Therefore, in the first example, the running plate thickness is changed as follows based on the flowchart of FIG.
First, the roll rolling distance in each rolling stand is divided into a plurality of rolling distance sections, and the roll / rolled material is divided into a plurality of rolling distance sections for each material (steel type) and / or sheet thickness of the material to be rolled. The actual value of the friction coefficient is collected for a plurality of coils (for example, 100 coils) (step 1).

  The friction coefficient is based on the actual rolling load using, for example, a well-known Hill approximation formula (for example, “Theory and Practice of Sheet Rolling” edited by Japan Iron and Steel Institute, page 35, issued on September 1, 1984). It can be obtained by calculating backwards. That is, since the Hill approximation formula shows the relationship between the rolling load and the friction coefficient, the friction coefficient can be obtained from the actual rolling load.

  Next, based on the collected actual values, a representative value of the friction coefficient in a plurality of rolling distance sections for each material and / or thickness of the material to be rolled is obtained (step 2).

  As a representative value of the friction coefficient, an average value for each rolling distance section can be used. The method for calculating the average value is not particularly limited, and various methods such as arithmetic average and moving average can be used. The representative value of the roll rolling distance in each rolling distance section may be an average value or a median value (median).

  Thus, after obtaining a representative value for each rolling distance section, an approximation curve is obtained by approximating the friction coefficient with respect to the roll rolling distance based on the representative value (step 3). The approximation method at this time is not particularly limited, and various methods such as linear approximation, logarithmic approximation, exponential approximation, polynomial approximation, and power approximation can be used.

  The rolling load model 21 is constructed using the approximate curve thus obtained (step 4). And the rolling load per unit width of each rolling stand is predicted by the rolling load model 21 constructed in this way (Step 5), and the roll reduction position of each rolling stand is determined using the predicted rolling load per unit width. Set and change the plate thickness between runs (step 6).

  In this way, the actual value of the friction coefficient is collected and learned for each roll rolling distance, and the effect of the friction coefficient is approximated based on the collected value. An increase in size can be avoided. For this reason, it is possible to predict the rolling load with high accuracy, and based on this, it is possible to change the running plate thickness with high accuracy.

  In addition, the conventional approximate expression has to be tuned each time when there is a large change in the operation status such as replacement with a new roll or rolling oil. Moreover, it was difficult to cope with the difference in the influence of roll wear due to temperature changes of rolling oil and coolant due to seasonal factors. Further, when such an approximate expression is managed using a fine mesh table for each steel type or plate thickness, tuning is required for each mesh, and thus enormous manpower is required. For these reasons, it has been extremely difficult to maintain an optimal tuning state. On the other hand, when the method of collecting the actual value of the coefficient of friction for each roll rolling distance as in the present embodiment is adopted, the conventional tuning is unnecessary, and thus the manpower problem is also solved. can do.

[Second example of a method for changing the running plate thickness in a cold tandem rolling mill]
The target for collecting the actual value for each rolling distance section is not limited to the coefficient of friction between the roll and the material to be rolled, but may be the rolling load (rolling load per unit width) itself.
A second example of the running plate thickness changing method will be described with reference to the flowchart of FIG. 3 as an example in which a rolling load is used as a target for collecting actual values for each rolling distance section.
In this example, first, the roll rolling distance in each rolling stand is divided into a plurality of rolling distance sections, and the rolling load of the plurality of rolling distance sections is divided for each material (steel type) and / or sheet thickness of the material to be rolled. Collect actual values for multiple coils (for example, 100 coils) (step 11).

  Next, based on the collected actual values of rolling loads, representative values of rolling loads in a plurality of rolling distance sections for each material and / or sheet thickness of the material to be rolled are obtained (step 12).

  Next, an approximate expression for the rolling load is constructed using these representative values (step 13). The approximate expression at this time is not particularly limited, and various expressions such as linear approximation, logarithmic approximation, exponential approximation, polynomial approximation, and power approximation can be used.

  Then, the rolling load per unit width of each rolling stand using the constructed approximate expression is predicted (step 14), and the roll reduction position of each rolling stand is set using the predicted rolling load per unit width. Then, the running plate thickness is changed (step 15).

  Thereby, rolling load prediction can be performed by a simpler method of using the existing rolling load itself without constructing a new program. And, by collecting and learning the actual value of rolling load for each roll rolling distance and approximating the influence of rolling load based on it, avoiding the increase in error as before even if the roll rolling distance increases can do. For this reason, it is possible to predict the rolling load with high accuracy, and it is possible to accurately change the running plate thickness based on the prediction.

Next, examples of the present invention will be described.
Here, a cold tandem rolling mill having five rolling stands shown in FIG. 1 was used. A cold tandem rolling mill having the following specifications was used.
Maximum line speed: 2000 mpm
Work roll diameter: 500-600mm ^φ
Backup roll diameter: 1300-1400mm ^φ
In addition, the rolling load detector, the reduction position detector, the tensiometer, and the plate thickness meter were provided at the positions shown in FIG.

  And according to this invention, the rolling load prediction at the time of a welding point passage was performed on the following conditions.

・ Results data collection The timing at which the process computer collects data was before and after passing through the welding point, and the collected items were rolling load, tension, plate thickness, and roll rolling distance.

・ Learning items The items to be learned by collecting the actual measurement values were the friction coefficient between the roll and the material to be rolled and the actual rolling load values. The coefficient of friction was obtained by calculating backwards using an approximate Hill formula with respect to the measured rolling load. In addition, collection of actual values of the friction coefficient was performed for each steel type and base plate thickness (0.5 mm unit).

・ Data editing for roll rolling distance The roll rolling distance is divided into five rolling distance categories of less than 0 to less than 10 km, less than 10 km to less than 30 km, less than 30 to 50 km, less than 50 km to less than 70 km, and more than 70 km. It was assigned to the rolling distance section, and the friction coefficient for each rolling distance section was averaged to obtain a representative value for the rolling distance section. A moving average was used as a calculation method. In addition, for the roll rolling distance, an average value was calculated by moving average for each rolling distance section, thereby obtaining a representative point for each rolling distance section.

-Approximation with respect to average With respect to the representative point data obtained by taking the average value of the friction coefficient for each rolling distance section as described above, an approximation curve was obtained by approximating the friction coefficient with respect to the roll rolling distance. A quadratic polynomial approximation was used as an approximation method.

-Result As mentioned above, the approximate curve (approximation formula) which shows the relationship between the roll rolling distance and friction coefficient obtained based on this invention, and the conventional approximation formula are compared, and it shows in FIG. As shown in FIG. 4, it can be seen that the approximate curve created based on the present invention is greatly different from the conventional approximate expression. This is because the approximate expression of the present invention collects and learns the actual value of the friction coefficient for each roll rolling distance, and approximates the influence of the friction coefficient based on it, so the friction coefficient is optimized for the roll rolling distance. On the other hand, the conventional approximate expression is only tuned based on past data at the beginning of rolling, and is not optimized in consideration of the roll rolling distance.

  In the case of the present invention example, the actual value of the friction coefficient is collected and learned for each roll rolling distance, and the result is used to predict the rolling load for changing the running plate thickness. The book prediction error (actual / prediction) was extremely high accuracy with an average of 1.002 and a standard deviation of 0.026. On the other hand, as a result of performing rolling load prediction using the approximate expression obtained by the conventional setting calculation, the prediction error of 100 similar steel plates was 1.031 on average and 0.058 on standard deviation. From this result, it was confirmed that the rolling load prediction accuracy for changing the strip thickness was significantly improved by the present invention.

DESCRIPTION OF SYMBOLS 1 Cold tandem rolling mill 2 Material to be rolled 11 Work roll (rolling roll)
DESCRIPTION OF SYMBOLS 12 Backup roll 13 Rolling apparatus 14 Rolling position detector 15 Rolling load detector 16 Tension meter 17 Sheet thickness meter 20 Process computer 21 Rolling load model 30 PLC
31 Running thickness change program

Claims (2)

This is a running plate thickness change method that changes the running plate thickness when a plurality of rolled materials with different conditions are welded and continuously rolled by a cold tandem rolling mill composed of two or more continuous rolling stands. And
The roll rolling distance of the rolling roll in each rolling stand of the cold tandem rolling mill is divided into a plurality of rolling distance sections, and for each of the plurality of rolling distance sections, for each material and / or sheet thickness of the material to be rolled, Collect actual values of the friction coefficient or rolling load between the rolling roll and the material to be rolled, and the friction coefficient or the rolling load in the plurality of rolling distance sections for each material and / or sheet thickness of the material to be rolled. A representative value is obtained, an approximate curve is obtained based on the representative value, a rolling load of each rolling mill is predicted using the approximate curve, and a running plate thickness is changed using the predicted rolling load. A method for changing the running plate thickness in a cold tandem rolling mill.   The representative value in the plurality of rolling distance sections is an average value of the friction coefficient collected in each of the rolling distance sections or the actual value of the rolling load. A method for changing the sheet thickness in a cold tandem rolling mill.

Images