JPH0698367B2 - Plate flatness control device - Google Patents

Description

The present invention relates to a rolling mill for rolling a plate material, and particularly to a plate flatness control for removing a left-right asymmetric component of a widthwise distribution of plate flatness. Regarding the device.

(Prior Art) In this type of plate flatness control, a plate flatness meter that detects a plate surface at each position in the plate width direction and obtains a width direction distribution based on signals from these plate detectors is provided. Is often used.

However, this plate flatness meter has a problem in response speed due to the complexity of signal processing, and various other plate flatness controls that do not use this have been proposed.

One of them is a tensiometer that detects the tension of a plate. This is to remove the left-right asymmetric component of the widthwise distribution by utilizing the deviation between the output of the tension meter on the driving side of the rolling mill and the output of the tension meter on the driven side.

(Problems to be Solved by the Invention) The above-described method of utilizing the tension difference is advantageous in that a tension meter having a quicker response than the plate flatness meter is used, but the tension difference is converted into a leveling difference. The proper method was not established and it was still in the search stage.

The present invention has been made to solve the above-mentioned problems, and can effectively correct the leveling of the roll even when using the deviations of the tension meter outputs on the driving side and the driven side, respectively, which results in excellent responsiveness. It is an object of the present invention to provide a plate flatness control device.

[Structure of Invention]

(Means for Solving Problems) The present invention is provided on a driving side and a driven side of a rolling mill for rolling a plate material, and a tensiometer for detecting tension of each plate, and a plurality of positions in a width direction of the rolling machine. The effect coefficient vector determining means for experimentally determining the effect coefficient vector of the leveling amount with respect to the tension stress at each position of the driving side, and the driving side and the driven side based on the driving side tension and the driven side tension detected by each tensiometer. A first calculation means for calculating the difference in tension stress on the side;
Based on the calculated leveling correction amount of the rolling mill, based on the influence coefficient vector determined by the tension stress difference stress influence vector determination unit And a leveling amount correction means for correcting the leveling amount.

(Operation) Here, after explaining the leveling amount and the leveling correction amount, a method of converting the tension meter output difference between the driving side and the driven side, which is the basis of the present invention, into the leveling correction amount will be described.

A rolling mill is usually provided with a pair of work rolls for directly sandwiching and rolling a material to be rolled, and a pair of retaining rolls for backing up these work rolls. In some cases, an intermediate roll may be arranged between the work roll and the backup roll, but here, a normal rolling mill without the intermediate roll is targeted.

The roll gap between the work rolls is adjusted by moving the bearing position of the retaining roll up and down. The bearings of the backup roll are on the drive side (the side connected to the electric motor) and the driven side (the side not connected to the electric motor). In order to roll the rolling mill symmetrically when viewed from the front, the inter-bearing distance S _d on the drive side of the retaining roll and the inter-bearing distance S _w on the driven side may be equal. However, in practice,
Due to the unevenness of the thermal expansion of the roll on the left and right, the unevenness of the material on the left and right of the material to be rolled, etc., the bearing-side distance S _d on the driving side and the bearing-side distance S _w on the driven side were set equal. A left-right uneven distribution of elongation of the material (corresponding to a left-right uneven distribution of tension applied to the material to be rolled) occurs.

Therefore, in such a case, it is possible to cause a difference between the bearing-side distance S _d on the driving side and the bearing-side distance S _w on the driven side, and as a result it is possible to control the elongation of the rolled material unevenly. Is. The difference between the driving-side bearing distance S _d and the driven-side bearing distance S _w is called the leveling amount.

When actually correcting the inter-bearing distance S _d and the driven inter-bearing distance S _w , when moving both the bearing position of the lower retaining roll and the bearing position of the upper retaining roll in conjunction with each other Alternatively, the other may be moved based on either one of them.

Further, the correction amount of the difference between the above-mentioned bearing distance S _d and the driven-side bearing distance S _w is called the leveling correction amount.

Next, a method for converting the difference between the tension meter outputs on the driving side and the driven side into the leveling correction amount will be described.

Normally, the leveling amount for removing the left-right asymmetric component of the plate flatness is obtained as follows using the output of the plate flatness meter.

Now, tension stress f (x _j ) (kg) at each position x _j (mm) in the plate width direction (j = 1, 2, ..., N, numbering is performed from the driven side)
/ mm ² ) is obtained from the plate flatness meter. This is expressed in vector notation as F = [f (x _j ), f (x ₂ ), ..., f (x _n )] ^T (1) (T represents transposition). At this time, the average tensile stress is f _M (kg / m
As m ² ), the following error vector F _E = [{f _M −f (x ₁ )}, {f _M −f (x ₂ )}, ... {f _M −
f (x _n )}] ^T (2), the purpose of plate flatness control is to make the distribution of tensile stress uniform in the width direction. Is reduced to the minimum.

Now, when the leveling amount S (mm) is corrected by a fixed amount δS, the tension stress fluctuation amounts at the width-direction positions x ₁ , x ₂ , ..., X _n are δf (x ₁ ), δf (x ₂ ),…, Δf (x _n ) (kg / m
m ² ), the degree of influence of δS on the tension is expressed by the ratio of δf (x ₁ ) and δS, the ratio of δf (x ₂ ) and δS, ..., δf (x _n ) and δS, respectively. Will be expressed as a ratio.

If all of these ratios are expressed mathematically using a matrix as the influence coefficient vector Y {(kg / mm ² ) / mm}, the leveling correction amount ΔS that minimizes J in equation (3).
(Mm) is ΔS = Z ^T · F _E (4) (where Z = Y · [Y ^T · Y] ⁻¹ ) by the method of least squares.

On the other hand, when the leveling correction amount ΔS = C · e (5) is obtained by multiplying the tension stress difference e (kg / mm ₂ ) on the driving side and the driven side by a certain coefficient C, ideally, equation (4) is used. And ΔS in equation (5) must be equal.

By the way, the tension stress difference e is However, Therefore, from equations (4), (5), and (6), Becomes Therefore, in order for ΔS in equations (4) and (5) to be approximately equal to each other, the error of the left side and the right side of equation (7) must be 2
It suffices to find the coefficient C that minimizes the sum of multiplications. Such coefficient C is calculated by the method of least squares. Becomes

Here, the derivation of equation (7) and the relationship between equations (7) and (8) will be described in detail.

The purpose of the plate flatness control is to make the distribution of the tensile stress expressed by the equation (1) in the plate width direction as flat as possible. In order to formulate this, in the present embodiment, the control purpose is defined as minimizing the squared error represented by J in equation (3). J in the equation (3) represents the sum of squares of the error between the tensile stress at the position in the plate width direction and the average tensile stress.

The above-mentioned "error with tension stress at each position in the plate width direction" is represented by a vector, which is the error vector of the equation (2). When this error vector is given, the expression (4) shows the leveling correction amount for minimizing J in the expression (3).

However, in actual rolling, the tension stress distribution (kg / mm ² ) at each position in the strip width direction is not obtained as a sensor output, and the total tension on the driving side (kg) and the total tension on the driven side (kg) are not obtained. can get. Therefore, the average tension stress on the driving side (kg / mm ² ) and the average tension stress on the driven side (kg / mm ² ) can be obtained.

From the above, in actual rolling, instead of obtaining the leveling correction amount from the "error vector between the tension stress at each position in the strip width direction and the average tension stress", the "average tension stress on the driving side (kg / mm ² ) And the average tension stress on the driven side (kg / mm ² ) ”
Therefore, the leveling correction amount must be obtained in the form of equation (5).

In fact, both equations (4) and (5) aim to minimize J in equation (3), so both equations should essentially be equal. Therefore, it is understood that the coefficient C of the equation (5) should be set so that the equation (4) and the equation (5) are equal.

The expression (7) is a rewrite of the expression that the expressions (4) and (5) are equal, and the left side of the expression (7) is the right side of the expression (4) and the right side of the expression (7) is The right side of equation (5) is rewritten using equation (6).

Looking at equation (7), ideally, If C can be obtained so that However, since Z and U are vectors and C is a scalar, it is not possible to obtain the coefficient C for which the above equation (9) is strictly satisfied. Therefore, in the present embodiment, C is calculated by the approximate solution (8) using the least square method.

From the above, in order to find the leveling correction amount ΔS (mm) for correcting the left-right asymmetric component of the plate flatness width direction distribution using the tension meter output instead of the plate flatness meter output, the drive side and driven It can be seen that the tensile stress difference e (kg / mm ² ) on the side can be obtained by multiplying by C shown in the equation (8).

Therefore, in the present invention, first, while the influence coefficient vector determining means experimentally obtains the influence coefficient vector, the output difference of the fast-response tensiometers provided on each of the driving side and the driven side is calculated, and The leveling correction amount is calculated based on the tension stress difference and the influence coefficient vector. As a result, it is possible to obtain a control accuracy comparable to the case where a plate flatness meter is used, and to configure a control system having excellent responsiveness.

〔Example〕

The drawing is a block diagram showing the construction of an embodiment of the present invention.
In the figure, a driving-side tensiometer 1D and a driven-side tensiometer 1W are connected to the input end of the adder 2. The adder 2 is connected to an arithmetic unit 3 that calculates a tension stress difference e based on the output ΔT and the plate width B and the plate thickness h.
On the other hand, a simulation device 4 is provided as an influence coefficient vector determination means for obtaining the influence coefficient vector Y, and the simulation device 4 further includes a calculation device 5 for calculating a vector Z necessary for calculation of the equation (8). Are connected. In addition, the vector Z of the arithmetic unit 5
An arithmetic unit 6 for calculating the leveling correction amount ΔS based on the tension stress difference e of the arithmetic unit 3 and the conversion vector U is provided.

The operation of this embodiment configured as described above will be described below.

First, adder 2 outputs a tension deviation ΔT by subtracting the output T _W tensiometer 1W of the driven from the output T _D tensiometer 1D of the drive side. The arithmetic unit 3 performs the following arithmetic operation in order to convert the tension deviation ΔT into a difference e (kg / mm ² ) between the driving side tension stress and the driven side tension stress.

However, B: plate width (mm) h: plate thickness (mm).

The tension stress difference e thus obtained is input to the arithmetic unit 6. On the other hand, in the rolling mill to which the present invention is applied,
Sensors are installed at each of the widthwise positions x ₁ , x ₂ , ..., X _n of the material to be rolled to detect the tension applied to the material. Therefore, when the leveling amount is corrected by a certain amount δS during the experimental rolling, the amount of tension variation δf (x ₁ ), δf (x ₂ ) at each position x ₁ , x ₂ , ..., X _{n in} the width direction, …, Δf (x _n )
Can be measured. The simulation device 4 is
From the detection values of the tension detection sensors during the experimental rolling, the fluctuation amounts δf (x ₁ ), δf (x ₂ ), ..., δf (x _n ) of tension are simulated,
Furthermore, the fluctuation amounts δf (x ₁ ) and δf of these tensions
The ratio between (x ₂ ), ..., δf (x _n ) and the correction amount δS of the leveling amount is calculated and the influence coefficient vector Y is output.

Therefore, the arithmetic unit 5 calculates the vector Z according to the following equation.

Z = Y · [Y ^T · Y] ^-1 (11) Next, the vector U applied to the arithmetic unit 6 is given as follows.

Therefore, the computing device 6 is The leveling correction amount ΔS is calculated in accordance with the above and added to the leveling amount correction device 7.

Here, e is the tension measurement value f (x ₁ ), f (x ₂ ), ..., f (x _n ) at each position in the width direction x ₁ , x ₂ , ..., X _n (6) Calculated by the formula.

The leveling amount correction device 7 corrects the leveling amount as follows.

Now, when correcting the inter-bearing distance S _d on the driving side and the inter-bearing distance S _w on the driven side by moving the upper retaining roll with reference to the lower retaining roll, the driven side of the upper retaining roll is A hydraulic cylinder is installed in each of the bearing and the drive-side bearing. Oil is sent from the oil tank into these hydraulic cylinders, and the pressure of this oil causes the bearings to move up and down. Therefore, an oil column is formed in the hydraulic cylinder, and the length of the oil column affects the distance between the bearings of the retaining roll. This pillar of oil is called an oil pillar.

A hydraulic pressure reduction device is provided to control the oil column. The hydraulic pressure reduction device is a well-known automatic plate thickness controller (AGC).
_{Input the} drive-side oil column reference O _DREH and the driven-side oil column reference O _WREF output from, and also input the drive-side oil column detection value and the driven-side oil column detection value converted from the information of the bearing position sensor, The drive side oil column reference and the drive side oil column detection value match,
In addition, feedback control of the oil column is performed so that the driven-side oil column reference and the driven-side oil column detection value match.

A plate flatness control device is provided in association with the automatic plate thickness control device in order to control the shape of the plate. This plate flatness control device outputs a signal corresponding to the leveling amount S and applies it to the automatic plate thickness control device. The automatic plate thickness control device multiplies the signal of the leveling amount S by a coefficient 1/2 to obtain a correction value S / 2 based on the oil column. Next, the drive side oil column reference O _DREF is _corrected to O _DREF + S / 2, the driven side oil column reference O _WREF is _corrected to O _WREF -S / 2, and each is added to the hydraulic pressure reduction device. There is.

When the leveling correction amount ΔS is calculated, the leveling amount correction device 7 adds the leveling correction amount ΔS to the leveling amount S and applies a signal corresponding to S + ΔS to the automatic plate thickness control device. This corrects the leveling amount.
Thus, according to this embodiment, it is possible to control the plate flatness with excellent responsiveness by using the output of the tensiometer.

〔The invention's effect〕

As is apparent from the above description, according to the present invention, when removing the left-right asymmetric component of the plate flatness width direction distribution, the leveling correction amount is calculated using the output of the tensiometer having excellent responsiveness, and As compared with the case of using the flatness meter, it is possible to perform control with excellent responsiveness without lowering control accuracy.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention. 1D …… Tensometer on drive side, 1W …… Tensometer on driven side, 2 …… Adder,
3,5,6 …… Computing device, 4 …… Simulation device.

Claims (1)

[Claims] 1. A tensiometer, which is provided on each of a driving side and a driven side of a rolling mill for rolling a sheet material, for detecting the tension of the sheet, and a leveling amount with respect to a tensile stress at a plurality of positions in a sheet width direction of the rolling machine. Influence coefficient vector determining means for empirically determining the influence coefficient vector, and a tension stress difference between the driving side and the driven side is calculated based on the tension on the driving side and the tension on the driven side detected by each of the tensiometers. First computing means and second computing means for computing the leveling correction amount of the rolling mill based on the tension stress difference calculated by the first computing means and the influence coefficient vector determined by the influence vector determining means. A plate flatness control device comprising: a leveling amount correction unit that corrects the leveling amount based on the calculated leveling correction amount.