KR19980020024A - Outer plate thickness control method in hot finishing mill - Google Patents

Abstract

The present invention relates to a method for controlling the exit plate thickness in a hot finishing mill.

The present invention calculates the thickness deviation correction amount according to the control program with the hot finishing mill 30, the thickness measuring device 40 for measuring the thickness of the plate through the hot finishing rolling, and the plate thickness information of the thickness measuring device 40 Including a calculation controller 50 for controlling the roll gap adjustment push-up cylinder device 34 mechanically coupled to the finishing rolling roll 32 of the finishing mill 30, to improve control response in hot finishing rolling To improve the control hit rate and the real rate by designing a filter that can estimate the original sheet thickness deviation.

Description

Outer plate thickness control method in hot finishing mill

The present invention relates to a method for controlling the exit plate thickness in a hot finishing mill, and in particular, to improve the control hit ratio and the real rate by designing a filter capable of estimating the original sheet thickness variation in order to improve the control performance in the hot finishing mill. The present invention relates to a method for controlling the exit plate thickness in hot rolling.

1 is a configuration diagram of a plate thickness control device of a general hot rolling mill. Referring to FIG. 1, the hot rolling process is briefly described. A rough rolling mill is rolled to an approximate thickness by using a bar extracted from a heating furnace. After passing through 10, the finishing mill 30 is rolled to the desired thickness. The rough rolling mill 10 is a device for rolling a bar to a substantially constant thickness so as to easily roll the bar before finishing rolling, the finishing rolling mill 30 is rough rolling to a thickness easy to roll in the rough rolling mill 10 It is a device for rolling to maintain a more precise plate thickness after being made. In addition, there is a sensor, which is a plate thickness meter 40, at the exit of the final rolling stand to measure the exit plate thickness, and a predetermined control method in the calculation controller 50 based on the plate thickness information by the thickness gauge 40. According to this, the plate thickness is controlled through the roll gap adjustment push-up cylinder device 34 and the finishing rolling roll 32.

FIG. 2 is a block diagram illustrating a conventional plate thickness control method. Referring to FIG. 2, a conventional plate thickness control method will be described. When the measured plate thickness deviation (Δhnoise) is larger than a desired size, FIG. Gain control (AGC) is used to control the stand-side plate thickness deviation as small as possible. The automatic gain control (AGC) operation principle senses the stand exit thickness with a sensor, which is a thickness gauge, to control the deviation from the desired thickness using the roll gap method. This control method is called monitor automatic gain control (monitor AGC), the roll gap control equation of the monitor automatic gain control (monitor AGC) can be expressed as shown in the following equation (1).

△ S = (M + Q) / M × △ hnoise --------------------- (1)

Where M is the MILL constant,

Q: plasticity factor,

△ hnoise: deviation of the thickness of the exit plate.

In the above formula (1), M and Q are values determined by rolling characteristics, so in order to increase the control performance of monitor AGC, whether or not the measurement of the feed plate thickness is precisely measured and fed back. It depends.

On the other hand, the plate thickness control method in the finishing rolling fixed is largely composed of a monitor AGC and a roll force AGC, the monitor AGC is a method of controlling the feedback thickness deviation by feeding back. In the hot rolling process, a sensor, which is usually a thickness gauge, is installed at the end of the last stand (stand 7). Because it is not easy. Therefore, the thickness deviation sensed from the thickness sensor at the rear end of the 7th stand is fed back to the monitor AGC to calculate the roll gap amount of the 4th, 5th, 6th and 7th stand and then distributed to each stand.

However, there are various problems in controlling the 4, 5, and 6 stand roll gaps by feeding back the 7th thickness deviation. Controlling the Roll Gap of Stands 4, 5, and 6 by using the thickness deviation of stand 7 is problematic because it controls the thickness variation of the past, and at this time, various sensor noises and sensors Due to its own error, it is difficult to feed back the original thickness deviation. This makes it difficult to accurately control the roll gaps of stands 4, 5, and 6 even when feeding back the thickness deviation. If the thickness deviation is fed back with error, the AGC control performance will be further deteriorated.

Therefore, in the conventional control method, enhancement noises (such as substrate thickness disturbances, temperature disturbances, rule eccentric disturbances, etc.) are generated in the sensor measuring the plate thickness, and the feedback of the exact thickness deviation due to the error of the thickness gauge itself is performed. Due to the difficult relationship, there was a problem that the automatic gain control (AGC) hit ratio was lowered due to noise or error, and the real rate was also lowered.

The present invention has been made to solve the above problems and to achieve an improvement.

Accordingly, an object of the present invention is to improve the automatic gain control (AGC) hit ratio by designing a filter that can estimate the original plate thickness deviation in order to reduce the sensor noise that increases the plate thickness variation and the error of the sensor itself. In addition, the present invention provides a method for controlling the exit plate thickness in a hot finishing mill that increases the real rate.

As a technical means for achieving the above object of the present invention, the method for controlling the exit plate thickness in the hot finishing mill of the present invention includes an annual finishing mill, a thickness gauge for measuring the thickness of the plate through hot finishing rolling, In the hot finishing mill, including an arithmetic controller for controlling the roll gap adjustment push-up cylinder device mechanically coupled to the finishing mill roll of the finishing mill by calculating the thickness deviation correction amount according to the control program with the thickness information of the thickness measuring instrument. A method for controlling the exit plate thickness, comprising: a first step of receiving a plate thickness (hnoise) from the thickness meter and calculating a plate thickness deviation (Δhnoise); A second step of calculating an estimated deviation [Delta] hest with the plate thickness deviation [Delta] hnoise through an estimation filtering process and repeatedly performing the estimated deviation [Delta] hest until the estimated deviation [Delta] hest is smaller than a prescribed deviation [epsilon]; And a third step of calculating the roll gap with the estimated deviation Δhest through the monitor AGC process and outputting the roll gap to the push-up cylinder device for adjusting the roll gap of the finishing mill.

1 is a configuration diagram of a plate thickness control device of a general hot rolling mill.

2 is a block diagram for explaining a conventional plate thickness control method.

3 is a block diagram for explaining a plate thickness control method according to the present invention.

4 is a flowchart showing a plate thickness control method according to the present invention.

5 is a graph for comparison between the original thickness deviation Δhref and the noise applied thickness deviation Δhest according to the present invention.

6 is a graph for comparing the thickness deviation? Hnoise with the noise applied to the measurement thickness deviation? Hest according to the present invention.

7 is a graph for comparison of the original thickness deviation Δhref and the measured thickness deviation Δhest according to the present invention.

* Description of the symbols for the main parts of the drawings *

1: plate 10: crude rolling mill

20: cutting machine 30: rolling mill

32: rolling mill 34: push-up cylinder device for roll gap adjustment

40: thickness meter 50: calculation controller

Hereinafter, with reference to the accompanying drawings, the configuration of the apparatus for performing the exit plate thickness control method in the hot finishing mill according to the present invention.

FIG. 3 is a block diagram illustrating a plate thickness control method according to the present invention. Referring to FIG. 3, an apparatus for performing the exit plate thickness control method in a hot finishing mill according to the present invention may include a plate (1). The operation controller controls the roughing mill 10 for coarsening to a predetermined thickness, the cutting machine 20 for cutting the front end of the plate 1 through the roughing mill 10, and the plate 1 passing through the cutting machine 20. According to the control of 50, the finishing mill 30 performs rolling, the thickness measuring device 40 for measuring the thickness of the plate through hot finishing rolling, and the plate thickness information of the thickness measuring device 40. According to the calculation of the thickness deviation correction amount is composed of a calculation controller 50 for controlling the roll gap adjustment push-up cylinder device 34 mechanically coupled to the finishing rolling roll 32 of the finishing mill (30).

4 is a flowchart showing a plate thickness control method according to the present invention, and FIG. 5 is a graph for comparing the original thickness deviation Δhref and the noise applied thickness deviation Δhnoise according to the present invention. 6 is a graph for comparing the thickness deviation? Hnoise with the noise applied to the measurement thickness deviation? Hest according to the present invention. 7 is a graph for comparison between the original thickness deviation Δhref and the measured thickness deviation Δhest according to the present invention.

The control method of the present invention according to the apparatus configured as described above will be described in detail below with reference to the accompanying drawings.

In the first steps 110 and 120, a plate thickness deviation Δhnoise is calculated by receiving the plate thickness hnoise from the thickness meter 40.

In the second step 130, the estimated deviation Δhest is calculated through the estimation filtering process using the plate thickness deviation Δhnoise, and is repeated until the estimated deviation Δhest is smaller than the prescribed deviation ε. In detail, first, the estimated deviation Δhest fed back from the plate thickness deviation Δhnoise from the first step 120 is corrected to correct the deviation, and the filtering gain is adjusted to the corrected deviation. do.

On the other hand, the mapping mill of the control target system is represented by a discrete state equation (Discrete State Equation) as shown in Equation (2) below.

x (k) = Ax (k-1) + w (k), k = 0, 1, 2,... -----------------(2)

Where x (k) is the state variable at time k.

x (k-1): State variable at time k-1

A: State Transition Matrix

W: process noise sequence

Next, the measurement equation (Measurement Equation) is as shown in the following equation (3).

y (k) = Cx (k) + v (k) -------------------------- (3)

Where y is the output variable

C: Measurement matrix

v: sensor noise sequence

In addition, the conditions of a process noise and a sensor noise are the same as following formula (4)-(8).

E [w (i)] = 0 ------------------------- (4)

E [v (i)] = 0 ------------------------- (5)

E [w (i) w (j)] = Q (i) ------------------------- (6)

E [v (i) v (j)] = R (i) ------------------------- (7)

E [w (i) v (j)] = 0 ------------------------- (8)

Where E [xx] is the mean of xx

Q (k): Covariance of W (k)

R (k): Mutual dispersion of V (k)

A state estimation filter is constructed for the state equations and the measurement equations expressed in the above equations (2) and (3) as described below. First, the filtering gain calculation is as shown in the following equations (9) (10) (11).

P1 (k) = AP (k-1) A ^T + Q (k-1) ----------------- (9)

P (k) = P1 (k) -K (k) C (k) P1 (k) --------------------- (10)

K (k) = P1 (k) C ^T + [CP1 (k) C ^T + R (k)] ^-1 ----------------- (11)

Where P1 (k) is the error covariance of the predicted estimate, P (k) is the deviation covariance of the filtered estimate, and K (k) is the filter gain.

Next, the model disturbance is added to the gain-adjusted deviation. After calculating the state estimation vector with the deviation of the model disturbances, subtracting the actual deviation Δhact from the estimated deviation Δhest of the state estimation vector, and when the absolute value is larger than the specified deviation ε. In the case of the deviation correction step, and if it is small proceeds to the third step, the calculation equation of the state estimation vector is the same as the following equation (12) including the equation (11).

By using Equation (12), AGx control performance is improved by obtaining an estimated value (x) by using an estimation filter and feeding it back to the monitor AGC using the estimated filter.

In the third step 140, the roll gap is calculated through the monitor AGC process using the estimated deviation Δhest and outputted to the roll gap adjustment push-up cylinder device 34 of the finishing mill 30.

EXAMPLE

The following describes the effects of the present invention with examples.

First, Figure 2 shows a simplified block diagram of a conventional plate thickness deviation control system.

In this figure, the plant is a finishing mill, and the process noise applied to the plant is usually the thickness of the substrate, the temperature disturbance, and the roll eccentric disturbance.

Further, the actual plate thickness deviation after the worker desired plate thickness deviation? Href is rolled through the rolling mill is? Href. However, since the deviation thickness of the sensor measuring the plate thickness deviation is improved and fed back is Δhref, the monitor AGC calculates the roll gap using this deviation. In addition, this thickness deviation includes high frequency noise that is not related to the outgoing plate thickness variation, making it difficult to obtain the original plate thickness variation, making it difficult to accurately sense the AGC.

Therefore, the present invention proposes a system as shown in FIG. 3 in order to measure the original sheet thickness deviation without sensor noise from the thickness deviation? Href to which noise is applied. Here, [Delta] href is applied to the estimation filter and estimated as the original plate thickness deviation, so that this estimated value [Delta] href is fed back to the monitor AGC.

At this time, the estimate of the original state is calculated using Equation (2-12).

Next, a simulation is performed to verify the effectiveness of the estimation filter from data extracted directly from the field. For simulation, the finishing mill is modeled as shown in Equation (2). The noise and sensor noise used in the simulation are shown in Table 1.

Where C _H : amplitude of substrate thickness disturbance (0.4mm)

f _H : Frequency of substrate thickness disturbance (0.1Hz)

C _T : amplitude of temperature disturbance (0.1㎜)

f _T : Frequency of temperature disturbance (0.8Hz)

C _L : Amplitude of roll single-phase disturbance (0.15㎜)

f _L : Frequency of roll eccentric disturbance (V / R)

V: roll speed of each stand

R: Backup roll radius of each stand

Process noises used in the simulation include substrate thickness deviation, disturbance, temperature disturbance, roll eccentric disturbance, and the like. Most process noises are characterized by repetition of a certain magnitude and period. Therefore, for simulation, the disturbance terms are approximated in the form of a constant sine wave. Table 1 below shows the values used in this simulation. In addition, sensor noise is a component whose normality is 0 and the variance is 0.5.

The simulation results are shown in FIGS. 5 to 7, and referring to FIGS. 5 to 7, first, in FIG. 5, the original thickness deviation? Href (1) and the sensor where no sensor noise is applied are shown. The thickness deviation Δhnoise (2) to which noise is applied is shown. Since the thickness deviation 2 to which the sensor noise is applied is not only large in hunting compared to the original thickness deviation 1 but also includes a fairly large frequency component, if the signal 2 is fed back to the equation (1), If the roll is controlled by the calculated roll gap amount, the roll moves along the high frequency component, resulting in deeper chattering of the roll, which causes a problem of promoting aging of the roll.

Therefore, in FIG. 5, it can be seen that the feedback of the thickness deviation without noise is calculated by reducing the thickness deviation and improving the AGC control performance. In FIG. 6, the thickness deviation? Hnoise (1) to which sensor noise is applied and the thickness deviation? Href (2) after filtering the sensor noise with the measurement filter are respectively controlled by the monitor AGC. Indicates. As predicted above, it can be seen that the plate thickness deviation obtained by feeding back the estimated value of the original state is smaller than the feedback of the thickness deviation with the sensor noise, and the result shows that the thickness deviation of (1) is largely hunted. It was. In FIG. 7, the original thickness deviation DELTA href (1) and the thickness deviation DELTA href (2) in which the original thickness deviation is estimated are shown. In this figure, (1) and (2) have similar thickness deviation magnitudes and no roll chattering. Therefore, even when the estimated signal 2 was fed back, the result was similar to that of the original signal 1 fed back, and the thickness deviation was good at about 60 µm. Also, it can be seen that the absolute value of the two signal errors, that is, the average of ΔΔhref * Δhref, is about 20 μm, which is much better than the absolute value of the error of the two signals of FIG. 5 about 35 μm.

According to the present invention as described above, the automatic gain control (AGC) stacking rate by designing a filter that can estimate the original plate thickness deviation in order to reduce the sensor noise and the error of the sensor itself to increase the plate thickness variation This has the special effect of increasing the error rate and improving the error rate.

The above description is only a description of one embodiment of the present invention, the present invention is capable of various changes and modifications within the scope of the configuration. In addition, it will be apparent to those skilled in the art that the method of controlling the exit plate thickness in the hot finishing mill is not limited to the above-described embodiment.

Claims (2)

With the calculation finishing mill 30, the thickness measuring device 40 for measuring the thickness of the plate through hot finishing rolling, and the thickness information of the thickness measuring device 40, the thickness deviation correction amount is calculated according to a control program. In the method of controlling the exit plate thickness in a hot finishing mill, including an arithmetic controller 50 for controlling the roll gap adjusting push-up cylinder device 34 mechanically coupled to the finishing rolling roll 32 of the rolling mill 30, A first step (110, 120) of calculating a plate thickness deviation (Δhnoise) by receiving a plate thickness (hnoise) from the thickness gauge (40); A second step 130 of calculating the estimated deviation Δhest with the plate thickness deviation Δhnoise through an estimation filtering process and repeatedly performing the estimated deviation Δhest until the estimated deviation Δhest is smaller than the prescribed deviation ε. ); The hot mapping comprising the third step 140 of calculating the roll gap with the estimated deviation Δhest and outputting the roll gap through the monitor AGC process to the push-up cylinder device 34 for adjusting the roll gap of the finishing mill 30. Control of exit plate thickness in rolling mills. The method of claim 1, wherein the second step (130) comprises: a deviation correction step of correcting the deviation by subtracting the estimated deviation (Δhest) fed back from the plate thickness deviation (Δhnoise) from the first step (120); A filtering gain adjustment step of adjusting a filtering gain to the corrected deviation; A model disturbance summing step of adding a model disturbance to the gain-adjusted deviation; A state estimation vector calculation step of calculating a state estimation vector with the deviation of the model disturbance being summed up; After subtracting the actual deviation Δhest from the estimated deviation Δhest of the state estimation vector, if the absolute value is larger than the prescribed deviation ε, the process proceeds to the deviation correction step, and if the value is smaller, proceeds to the third step. Deviation determination step; Method of controlling the exit plate thickness in the hot finishing mill, characterized in that consisting of.