JPH08267111A - Method for controlling thickness in tip part in hot rolling - Google Patents

Abstract

PURPOSE: To improve accuracy of thickness in the tip part of a coil by elevating the accuracy of thickness in the tip part at intermediate stands using setup learning. CONSTITUTION: In a controlling method of thickness in the tip part in hot rolling in which the amount of correction of thickness is determined using thickness deviation ΔhM in the tip part of a rolled stock B measured with an intermediate stand thickness meter A and the correction of thickness at downstream stands i+1, ..., N with a feedforward controller C is executed, the thickness deviation Ah. in the tip part is measured with the thickness meter E provided on the outlet side of the final stand N and also the amount Δhest ' of correction of thickness on the outlet side of the final stand which is executed by feed-forward control is determined. The setup of stand for the next rolled stock is executed based on the thickness deviation ΔhL in the tip part on the outlet side of the final stand and the amount Δhest ' of correction of thickness on the outlet side of the final stand.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a strip thickness control method for improving strip thickness precision of the tip portion of a hot coil in hot rolling.

[0002]

2. Description of the Related Art The sheet thickness accuracy of a hot coil, especially the tip of the coil, depends largely on the initial setup accuracy of the finish rolling mill. It is no exaggeration to say that the improvement of the plate thickness accuracy over the entire length of the hot coil can be achieved by improving the accuracy of the coil tip. This is because the AGC (automatic plate thickness control) function works for the plate thickness adjustment during rolling, and it is a problem to improve the plate thickness accuracy of the coil tip.

Therefore, in order to improve the plate thickness accuracy of the hot coil, especially the plate thickness accuracy of the coil tip, feedforward control using a plate thickness gauge between finishing stands has been conventionally performed. This feed-forward control measures the thickness deviation of the coil tip part during rolling, predicts the output side thickness deviation from this coil tip thickness deviation, and adjusts the gap and the rotation speed of the downstream stand so that this value becomes zero. Is to correct. This feed-forward control is characterized in that there is no control delay because the plate thickness control is performed in advance by predicting the plate thickness deviation. However, in this feedforward control, the plate thickness is controlled not based on the actual plate thickness deviation but based on the predicted plate thickness deviation. Therefore, when the prediction accuracy is low, the plate thickness accuracy is also low. .

Further, as disclosed in Japanese Patent Laid-Open No. 3-291107, in order to improve the accuracy of the coil tip plate thickness, one or more units are provided between the finishing rolling mill stands in addition to the output side plate thickness gauge. The intermediate plate thickness gauge is installed, and the plate thickness at the coil tip is measured during the finish rolling using this intermediate plate thickness gauge, and based on the result, between the rolls of the stand downstream from the plate thickness gauge in a feedforward manner. A method is known in which the gap and the number of revolutions are dynamically corrected before the tip of the coil is narrowed down to obtain a high plate thickness accuracy.

Further, as a general method for executing the plate thickness control using the intermediate stand plate thickness gauge, an error from the target plate thickness at the intermediate stand at the tip of the coil and, as a result, at the exit side of the final stand The relationship between the target product thickness and the difference between the actual values is statistically calculated, the finish side error of the coil is predicted from the error in the intermediate stand measured for a certain coil, and the intermediate error is absorbed. Japanese Patent Laid-Open No. 3-151109 discloses a method for correcting the roll gap and roll peripheral speed of the stands after the plate thickness gauge installation stand before the coil tip is caught.

However, in the tip end plate thickness control method using the intermediate plate thickness gauge, when the plate thickness deviation in the intermediate stand plate thickness gauge is large, or when the plastic coefficient is large like stainless Ni steel, However, there is a problem that the gap of the downstream stand and the correction amount of the number of revolutions become larger than the intermediate thickness gauge necessary to eliminate the thickness deviation on the delivery side of the final stand.
That is, it is considered that the correction is performed only in the latter stage and that the correction amount is large, which is considered to be a cause of the variation in the plate thickness accuracy, and it is not preferable from the viewpoint of ensuring the sheet passing property. Further, when the correction amount is large, the limiter of each preset stand gap is applied, and there is a risk that control will be insufficient.

On the other hand, it is disclosed, for example, in Japanese Patent Laid-Open No. 60-15010 that learning is performed in order to improve the accuracy of the mathematical model for predicting the coil finishing side error. When the tip portion thickness control in hot rolling is performed using this learning function, generally, first,
Create a mathematical model to predict the final stand outlet thickness deviation from various rolling conditions, and measure the final stand outlet thickness deviation with a final stand outlet thickness gauge while actually rolling various rolled materials. Then, the mathematical model is modified so that the difference between the actual result of the final stand output side plate thickness deviation and the predicted final stand output side plate thickness deviation becomes smaller. That is, by performing learning, it is possible to accurately predict the final stand outlet side plate thickness deviation with high accuracy. Then, after the learning is completed, the final stand outlet side plate thickness deviation is predicted based on the corrected mathematical model to control each stand. However, the actual output thickness deviation of the final stand is only used for learning the feedforward control,
There was no idea to further improve the control accuracy by using this final stand output side plate thickness deviation record.

[0008]

Therefore, in the present invention, in order to maximize the effect of the feedforward control, the initial set-up accuracy of the finishing stand, especially the tip end plate thickness accuracy of the intermediate stand, is set-up learned. The purpose is to improve the accuracy of the coil tip plate thickness by improving the use.

[0009]

According to the present invention, a plate thickness correction amount is obtained by using a plate thickness deviation of a front end portion of a rolled material measured by an intermediate stand plate thickness gauge, and a plate thickness correction at a downstream stand is performed by feedforward control. In the method for controlling the thickness of the front end portion in hot rolling, the deviation of the thickness of the front end portion is measured by a thickness gauge provided on the output side of the final stand, and the thickness correction on the output side of the final stand performed by the feedforward control is performed. An amount is obtained, and the stand is set up for the next rolled material on the basis of the deviation of the plate thickness at the tip of the final stand exit side and the correction amount of the final stand exit side plate thickness.

[0010]

[Function] The plate thickness control using the intermediate plate thickness gauge predicts the outlet plate thickness deviation from the coil tip part plate thickness deviation at the intermediate stand,
In order to make this value zero, the gap and the rotation speed of the thickness gauge downstream stand are corrected. Therefore, it is preferable from the viewpoint of plate thickness accuracy and plate passing property to reduce the plate thickness deviation in the intermediate stand and reduce the correction amount of the latter stand.

In view of this, in the present invention, the learning function for the final stand exit side plate thickness, which has been conventionally performed, is applied to the plate thickness at the intermediate stand. That is, the final stand outlet side thickness deviation used for conventional learning is replaced with the estimated final stand outlet side thickness deviation when feedforward control using the intermediate thickness gauge is not performed, and the sheet thickness for the next rolled material is replaced. The correction coefficient is obtained as a feedback value to reduce the intermediate stand plate thickness deviation.

Further, in order to further reduce the disturbance due to the control residual of the feedforward control by the intermediate thickness gauge, the correction amount and the actual value of the final stand exit side plate thickness by the feedforward control (final stand exit side actual plate thickness deviation ), Each gain is obtained, and the above-mentioned feedback value is multiplied to obtain an appropriate correction coefficient for the next rolled material.

By the above measures, the precision of the coil tip plate thickness can be improved.

[0014]

EXAMPLES The present invention will be specifically described below with reference to examples.

FIG. 1 is a block diagram showing a configuration for carrying out the plate thickness control method of the present invention. In the figure, 1
˜N is a stand, A is an intermediate thickness gauge, B is a rolled material, C is a feedforward control device, D is a finish setup computing device, and E is a final stand delivery side thickness gauge.

First, the outline of the operation of the block shown in FIG. 1 will be described. Using the tip thickness deviation of the rolled material B measured by the intermediate thickness gauge A installed between the intermediate stands,
The feed-forward control device C has a downstream stand i +.
The gaps and speeds in 1, ..., N are corrected. In addition to this, the finish setup computing device D performs the next rolling based on the tip thickness deviation measured by the final stand outlet thickness gauge E and the final stand outlet thickness correction amount performed by the feedforward controller. Set up all stands for the material.

Next, details of the operation of the blocks shown in FIG. 1 will be described.

First, in the finish setup computing device D, the load predicting formula, the mill extension predicting formula. Calculation is performed based on a rolling prediction formula or the like, and initial setting is performed for each of the stands 1 to N based on the calculation result.

The rolled material B is rolled while sequentially passing through the stands 1 to N. An intermediate thickness gauge A is arranged between the i-th stand i and the i + 1-th stand i + 1,
The plate thickness deviation Δh _M on the exit side of the i-th stand i is detected, and this plate thickness deviation Δh _M is supplied to the feedforward controller C. Further, a final stand outlet side plate thickness gauge E is provided on the downstream side of the N-th stand N which is the final stand, and the final stand outlet side tip end plate thickness deviation Δh _L is detected, and this plate thickness deviation Δh _L is also detected. It is supplied to the feedforward controller C.

Now, feedforward control using the intermediate stand thickness gauge A will be described.

In the feedforward controller C, the final stand outgoing side plate thickness estimated value Δh is calculated based on the following equation.
_est is required.

Δh _est = f (Δh _M , P _i , P _i−1 , ...) Next, based on the following equation, the gap correction amount at the j-th stand arranged downstream of the intermediate thickness gauge A: ΔS _j is obtained.

[0023]

[Equation 1] Here, M _j is the jth stand mill constant, Q _j is the jth stand plasticity coefficient, and G _Sj is the jth plate thickness correction distribution ratio.

The gaps of the stands i + 1 to N are corrected on the basis of the gap correction amount ΔS _j . The upper and lower limits of the gap correction amount ΔS _j are limited by the limiter.

The speed correction amount Δ at the j-th stand
V _j is _calculated based on the following equation, for example.

[0026]

[Equation 2] Here, Δh _j is the plate thickness error of the j-th stand, h _j is the outgoing target value of the j-th stand, and V _Rj is the initial roll peripheral speed of the j-th stand.

The speed of each stand _i to N + 1 is corrected based on this speed correction amount ΔV _j .

As described above, the feed-forward control is performed on the final stand outgoing side plate thickness estimated value Δh _est obtained from the tip side plate thickness deviation Δh _M on the outgoing side of the i-th stand i detected by the intermediate plate thickness gauge A. Based on this, the control of each stand i + 1 to N is performed.

Next, the thickness control of the intermediate stand will be described.

Here, the calculated load is first obtained, and the gap value is calculated from this calculated load.

The calculated load P _cal is obtained from the following equation.

P _cal = 1.15 × C ₁ × C ₂ × kfm ×
b × √ (R′Δh) × Q _P (μ) where C ₁ is a plate thickness correction coefficient (feedback value), C
₂ is a coefficient determined for each characteristic value / plate thickness, kfm is a restrained average deformation resistance, b is a plate width, √ (R′Δh) is a contact arc length with a roll, and Q _P (μ) is a rolling force function.

The plate thickness correction coefficient (feedback value) C
₁ is calculated from the following formula.

C ₁ = C ₃ × (Δh _l + Δh _est ′) × α _i where Δh _est ′ is the Nth stand N, that is, the final stand exit side plate thickness correction amount, and C ₃ is a coefficient for feedback control, α _i is the gain determined by the feedforward control and the actual output side plate thickness deviation. Note that Δh _est 'is
If the limiter is not applied, it is equal to Δh _est .

Final stand delivery side plate thickness correction amount Δh _est '
Is supplied to the finish setup computing device D, and initial setting is performed for all stands based on the final stand outlet side plate thickness correction amount Δh _est ′ and the final stand outlet side tip part plate thickness deviation Δh _L.

The gains α _i determined by the above-mentioned feedforward control are classified by the method shown in Table 1, for example.

[0037]

[Table 1] The relationship between the measured delivery side plate thickness and the plate thickness correction amount will be described.

For example, when Δh _est calculated from Δh ₄ is a large positive value, that is, when the thickness is large, the gaps of the 6th stand and the 7th stand are closed to correct the plate thickness. Then, as a result, if the measured delivery side plate thickness is large and is too thick (α _{1 in} Table ₁ ), it is determined that the control is insufficient. The opposite is α ₉ , and the gain of the intermediate thickness learning is increased for these two.

However, as a result of the fact that Δh _est is large and the gap is closed and the measured output side plate thickness deviates greatly and thinly (α _{7 in} Table 1), reverse control is performed and the gain for learning the intermediate plate thickness is small. , For example, approximately 0. The same applies to the case of α ₃ .

In addition, α ₄ and α ₆ are cases where the plate thicknesses match as a result of the large plate thickness modification, and the modification amount is reduced by intermediate learning. alpha _2, alpha _5, when it correction amount is small side thickness out largely deviated, to reduce Delta] h _l at intermediate training. Since α ₅ has a small correction amount and the result is good,
There is no need for intermediate learning, and the gain may be small.
The magnitude relationship between the above-mentioned gains is shown below.

100% ≧ α ₁ , α ₉ > α ₄ , α ₆ ≧
α ₂ , α ₈ > α ₃ , α ₅ , α ₇ ≧ 0% Next, the gap value is calculated from the calculated load P _cal obtained as described above.

The gap value can be obtained from the load, the extension of the mill, and the entry side plate thickness. The relationship between these variables is approximated by a straight line for simplicity. FIG. 3 is a graph showing the relationship between the extension of the mill and the load, and the slope of the straight line indicates the mill constant (M). FIG. 4 is a graph showing the relationship between the entry side plate thickness and the load, and the slope of the straight line shows the plasticity coefficient (Q). FIG. 5 is a graph obtained by combining FIG. 3 and FIG. As shown in FIG. 5, if the entrance side plate thickness H and the load P _cal are found, the exit side plate thickness h and the gap length s are found.

By setting each stand based on the gaps and velocities of all stands obtained as described above, the deviation of the thickness of the intermediate stand plate is reduced. As a result, the correction amount in the latter stand can be reduced, so that the correction amount is not applied to the limiter, and the accuracy of the Nth stand N output side plate thickness which is the final stand is improved.

FIG. 2 is a graph showing the effect of the present invention. In the case of rolling a stainless Ni steel having a thickness of 2.4 to 8.0 mm, the ± 50 μm on-gauge ratio was 73% (the number of samples was 300) by the conventional method using only feedforward, whereas the method of the present invention was used. With the method that uses the intermediate plate thickness learning together, the on-gauge ratio of ± 50 μm is 81
% (The number of samples is 273), and the tip plate thickness accuracy is 8%
Improved.

In the DSU, the same material is controlled, whereas in the intermediate plate pressure learning in this embodiment, the learning result is reflected in the setup of the same steel type and the same size material to be rolled thereafter. . In the setup, the gap is set from the load for each stand calculated by P _cal described above. For example, when the front material is displaced in the thick direction, the coefficient of P _cal is increased by learning and the gap is increased by increasing the load. For example, the coefficient for calculating the calculation load is calculated by learning, and the plate thickness is adjusted.

[0046]

According to the present invention, since the deviation of the plate thickness on the delivery side of the intermediate stand is reduced, the correction amount in the latter stand is reduced. As a result, the correction amount in the latter stand is not applied to the limiter and the control is not insufficient. Therefore, the accuracy of the coil tip plate thickness can be increased.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration for carrying out a plate thickness control method of the present invention.

FIG. 2 is a graph showing the effect of the present invention.

FIG. 3 is a graph showing the relationship between the extension of the mill and the load.

FIG. 4 is a graph showing the relationship between the entrance side plate thickness and the load.

5 is a graph in which FIG. 3 and FIG. 4 are combined.

[Explanation of symbols]

1 to N ... Finishing stand, A ... Intermediate thickness gauge, B ... Rolled material,
C ... Feedforward control device, D ... Finishing setup computing device, E ... Final stand exit side plate thickness gauge

Claims (1)

[Claims] 1. A front end plate in hot rolling in which a plate thickness correction amount is obtained by using a front plate thickness deviation of a rolled material measured by an intermediate stand plate thickness gauge, and the plate thickness is corrected in a downstream stand by feedforward control. In the thickness control method, the tip thickness deviation is measured by a thickness gauge provided on the exit side of the final stand, and the plate thickness correction amount on the exit side of the final stand performed by the feed-forward control is obtained. A method for controlling a front end plate thickness in hot rolling, comprising setting up a stand for a next rolled material on the basis of a side front end plate thickness deviation and the final stand exit side plate thickness correction amount.