JPS638848B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for controlling the thickness of a continuous rolling mill in which a plurality of stands are arranged in tandem, particularly for eliminating variations in thickness due to eccentricity of rolling rolls.

In continuous rolling mills that are generally used to manufacture hot-rolled steel sheets, a gauge meter method with excellent responsiveness is widely adopted as a thickness control method to make the thickness of hot-rolled steel sheets uniform in the rolling direction, that is, in the longitudinal direction. has been done. This gauge meter method uses the plate thickness variation Δh as the variation in the roll gap of the rolling roll,
In other words, (Δh
=ΔS+ΔP/M), and control is performed so that the plate thickness fluctuation amount becomes zero (Δh→0), that is, ΔS+ΔP/M→0. By the way, eccentricity of rolling rolls is known to be a factor in the thickness variation of hot-rolled steel sheets, particularly caused by the rolling mill itself.

In order to further improve plate thickness control accuracy, the present inventors focused on the effect of roll eccentricity on plate thickness, and conducted an experimental study on the effect of roll eccentricity under gauge meter type plate thickness control. As a result, the following facts were discovered.

For example, if the eccentricity of the rolling roll (hereinafter simply referred to as the roll eccentricity) is ΔRe, the roll gap will change by the sum of the reduction change ΔS and the roll eccentricity ΔRe, that is, ΔS + ΔRe, so the plate thickness variation Δh will be It can be expressed as the following equation (1).

Δh=ΔS+ΔRe+ΔP/M...(1) Assuming that the gauge meter type plate thickness control as described above is applied, that is, ΔS+ΔP/M→0, Δh=
ΔRe, and the roll eccentricity ΔRe directly appears as a variation in the thickness of the hot rolled steel sheet. Incidentally, the influence of the eccentricity of the rolling rolls on the plate thickness in the case where plate thickness control is not performed is as follows. Generally, in plate thickness control using a gauge meter method, the load fluctuation amount ΔP is calculated by converting the rolling load P into the outlet plate thickness h,
It is given by the following equation (2) as a function with the entrance plate thickness H and deformation resistance K as variables.

ΔP＝∂P／∂hΔh＋∂P／∂HΔH＋∂P／∂KΔK……(2
) From equations (1) and (2) above, Δh can be rewritten as equation (3).

Δh=1/1-1/M ∂P/∂h・(ΔS+ΔRe)+Δhd
...(3) However, Δhd=1/1-1/M ∂P/∂h(1/M ∂P/∂HΔH
+1/M ∂P/∂K ΔK) ... (3') Δhd is the amount of plate thickness variation that occurs due to the inlet thickness change ΔH and the deformation resistance change ΔK when plate thickness control is not performed, i.e. It is a disturbance. As is clear from equation (3), the coefficient 1/1-1/M ∂P/∂h of the roll eccentricity ΔRe is 1 since ∂P/∂h<0.
Therefore, when looking only at the influence of eccentricity of the rolling rolls on the plate thickness, the effect of the eccentricity of the rolling rolls on the plate thickness is reduced when the plate thickness control using the gauge meter method is not applied than when the plate thickness control is applied. Therefore, it can be seen that the influence of roll eccentricity on the plate thickness is enhanced when the gauge meter method is used to control the plate thickness. The present invention has been made based on this knowledge, and its purpose is to obtain data related to plate thickness, such as the amount of change in inlet thickness from the upstream stand, and based on this data, to obtain data related to plate thickness from the upstream stand. The amount of change in plate thickness that should occur in the process is calculated, the amount of change in load target to eliminate this is calculated, and the amount of change in load, including the amount of change in rolling load due to roll eccentricity detected at the downstream stand, is set as the load target. By controlling the load to match the amount of change, continuous rolling eliminates plate thickness variations caused by eccentricity of the rolling rolls, and at the same time eliminates plate thickness variations caused by entrance thickness changes and deformation resistance changes. The purpose of the present invention is to provide a method for controlling the thickness of a machine.

The plate thickness control method for a continuous rolling mill according to the present invention involves determining the inlet thickness change amount, outlet thickness change amount, and load variation amount in the upstream stand of arbitrary two stands in the continuous rolling mill, and based on these, the downstream side Predicting the amount of plate thickness variation that should occur on the exit side of the stand, and based on this plate thickness variation, calculating the load target change amount that should be set on the downstream stand necessary to eliminate this, The present invention is characterized in that the rolling reduction amount of the downstream stand is adjusted so that the amount of load variation including the amount of variation in rolling load due to roll eccentricity detected by the downstream stand coincides with the load target change amount.

Hereinafter, first, the principle of the plate thickness control method for a continuous rolling mill according to the present invention (hereinafter referred to as the method of the present invention) will be explained. Since the plate thickness variation period caused by roll eccentricity is usually extremely small, it is important to detect this as a plate thickness variation using a thickness gauge and adjust the roll gap to eliminate this from the viewpoint of the response characteristics of the control system. It's extremely difficult from that point of view. Therefore, instead of controlling the plate thickness by setting ΔS + ΔP/M → 0 as in the conventional gauge meter method, we first tried to keep the rolling load constant in one stand, in other words, the load fluctuation amount ΔP → 0. Configure the control system. With this, actual roll gap changes due to roll eccentricity can be indirectly detected by rolling load.
By adjusting the rolling reduction variation ΔS so as to keep the roll gap constant, it is possible to eliminate plate thickness fluctuations caused by eccentricity of the rolls in one stand.

However, if there is a disturbance such as a change in the inlet thickness or a change in deformation resistance due to skid marks,
Changes in plate thickness caused by these will appear as they are in the stand. Therefore, in response to these disturbances, the above-mentioned disturbances are predicted in a stand upstream from the stand concerned, the load target change amount ΔPr necessary to eliminate this is calculated, and the rolling load of the one stand is adjusted to the target value. Control is performed so that they match, in other words, ΔP→ΔPr.

This makes it possible to simultaneously eliminate plate thickness variations due to roll eccentricity and plate thickness variations due to disturbance.

Therefore, first, the plate thickness fluctuations caused by disturbances such as changes in the thickness at the entrance to the stand and changes in deformation resistance are determined.
In order to find the plate thickness variation due to such disturbances, we focus on any two stands in a continuous rolling mill, for example, two adjacent stands, and calculate the rolling conditions based on the upstream stand (hereinafter referred to as the first stand). The plate thickness fluctuation amount Δhd due to disturbance that would occur if plate thickness control is not performed at the downstream stand (hereinafter referred to as the second stand) is predicted.
All mill stiffness coefficients, etc. related to the stand shall be indicated with the subscript " ₁ ", and all mill stiffness coefficients, etc. related to the second stand shall be indicated with the subscript " ₂ ".

The amount of plate thickness variation Δhd ₂ due to the disturbance predicted on the exit side of the second stand is given by formula (4) by rewriting general formula (3') into the formula for the second stand.

Δhd ₂ = 1/1-1/M ₂ (∂P/∂h) ₂・[1/M ₂ (∂
P/∂H) ₂ ΔH ₂ +1/M ₂ (∂P/∂K) ₂ ΔK ₂ ]...(4) By the way, ∂P/∂K ₂ ΔK ₂ in equation (4) can be rewritten as follows. That is, if we transfer equation (2) to the equation for the first stand, we get equation (5), where (∂P/∂K) ₁ ΔK ₁ =ΔP ₁ −(∂P/∂K) ₁ Δh ₁ −(∂
P/∂K) ₁ ΔH ₁ ...(5) Since the rate of change in deformation resistance is constant across each stand, equation (6) holds true, and ΔK ₁ /K ₁ = ΔK ₂ /K ₂ ... (6) Also, general formula (7) holds true.

P=(∂P/∂K)K...(7) Therefore, from equations (6) and (7), the above (∂P/∂K) ₂ ΔK ₂
can be expressed as follows.

(∂P/∂K) ₂ ΔK ₂ = (∂P/∂K) ₂ K ₂・ΔK ₂ /K ₂ =P ₂ Δ
K ₂ /K ₂ =P ₂ ΔK ₁ /K ₁ =P ₂ /P ₁・P ₁ /K ₁・ΔK ₁ =P ₂ /P ₁ (
∂P/∂K) ₁ ΔK ₁ ...(8) As a result, the above equation (4) can be expressed as the following equation (9).

Δhd ₂ = 1/M ₂ (∂P/∂H) ₂ ΔH ₂ +1/M ₂ P ₂ /P
₁ [ΔP ₁ −(∂P/∂h) ₁ Δh ₁ −(∂P/∂H) ₁ ΔH ₁ ]/
1-1/M ₂ (∂P/∂h) ₂ ...(9) Mill stiffness coefficient M ₂ in formula (9), rolling load P ₁ , P ₂ ,
Partial differential coefficient (∂P/∂h) ₁ , (∂P/∂h) ₂ , (∂P/
∂H) ₁ and (∂P/∂H) ₂ are values that are known or predetermined based on the rolling schedule, and the load fluctuation amount ΔP ₁ in the first stand is detected by a load cell, and Since the inlet thickness change amount ΔH ₂ in the second stand is the outlet thickness change amount Δh ₁ in the first stand, the outlet thickness change amount Δh ₁ can be measured directly with a thickness gauge or by a gauge meter method, that is, Δh ₁ =ΔS Calculated by ₁ + ΔP ₁ /M ₁ . The amount of change in inlet thickness ΔH ₁ at the first stand is measured by a thickness meter, or calculated using the same gauge meter formula as described above as the amount of change in outlet thickness of the stand adjacent to the upstream side. do. The gauge meter formula used here is Δh ₁ = ΔS ₁ + ΔP ₁ /M ₁
Unlike equation (1), it does not include ΔRe, but
This is true if the rolling devices in the first stand and stands further upstream of the first stand use an electric rolling method, or if rolling in these stands does not use gauge meter-based thickness control. This is because it can be replaced with a gauge meter type that does not include ΔRe.

That is, if the plate thickness is not controlled by the gauge meter method, the error between when ΔRe is taken into account and when ΔRe is not taken into account remains at about 1/2 to 1/7 of the roll eccentricity in the stand. In addition, when using the electric rolling down method, the response is at most about 0.5 Hz, as is well known, and cannot follow roll eccentricity of 2 to 5 Hz. This is because the result will be the same as when no plate thickness control is applied.

This makes it possible to predict plate thickness variations on the exit side of the second stand that occur due to changes in inlet thickness and changes in deformation resistance.

Next, the load target change amount ΔPr ₂ to be set at the second stand is calculated using the plate thickness variation amount Δhd ₂ due to the disturbance predicted in the above process, and the load variation amount ΔP 2 detected at the second stand is used as the load change amount ΔPr ₂ . So-called load control is performed to adjust the reduction amount to match the date mark change amount ΔPr ₂ . First, the block diagram of the load control system shown in FIG. 2 will be explained below.

Now, the load target change amount ΔPr ₂ and the load variation amount ΔP ₂
It is assumed that the servo valve operates and the pressure oil flow rate q changes according to the deviation signal ε. Here, if the proportional gain between the deviation signal ε and the pressure oil flow rate q is Kv, and the time constant of the servo valve is Tv, the relationship between the pressure oil flow rate q and the deviation signal ε is as shown in equations (10) and (11). be.

q=-Kv/1+rTvε ...(10) ε=ΔPr ₂ −ΔP ₂ ...(11) However, r is the Laplace operator.In addition, when the pressure oil flow rate is integrated over time, The change, that is, the reduction change amount ΔS ₂ .

ΔS ₂ = 1/r 1/Aq ...(12) However, A: Hydraulic cylinder cross-sectional area Furthermore, regarding the load fluctuation amount ΔP ₂ , the following equation (13) for the second stand is obtained using equations (1) and (3). can be guided.

ΔP ₂ = −M ₂ Q ₂ /M ₂ +Q ₂ (ΔS ₂ +ΔRe ₂ )+ΔPd ₂ ...(13) However, Q: Plasticity coefficient Next, the dynamic characteristics of the load fluctuation amount ΔP ₂ with respect to the load target change amount ΔPr ₂ are Find the transfer function Gp(r) shown.

First, by eliminating ε and q from equations (10), (11), and (12), we get equation (14).

ΔS ₂ = -1/r 1/A Kv/1+rTv (ΔPr ₂ −ΔP ₂ ) ...(14) If the reduction change amount ΔS ₂ is eliminated from equations (13) and (14), it becomes equation (15). .

(1+M ₂ Q ₂ /M ₂ +Q ₂ 1/r 1/A Kv/1+rTv)ΔP ₂
=M ₂ Q ₂ /M ₂ +Q ₂ 1/r 1/A Kv/1+rTvΔPr ₂ +ΔP
d ₂ −M ₂ Q ₂ /M ₂ +Q ₂ ΔRe ₂ ...(15) Therefore, if we set T=A/Kv M ₂ +Q ₂ /M ₂ Q ₂ , equation (15) can be written as equation (16). It can be changed.

(1+1/rT 1/1+rTv)ΔP ₂ =1/rT 1
/1+rTvΔPr ₂ +ΔPd ₂ −M ₂ Q ₂ /M ₂ +Q ₂ ΔRe ₂ ……(16)
Furthermore, by setting GP(r)=1/1+rT(1+rTv), equation (16) can be rewritten as equation (17).

ΔP ₂ = Gp(r)ΔPr ₂ +(1−Gp(r))・
(ΔPd ₂ −M ₂ Q ₂ /M ₂ +Q ₂ ΔRe ₂ ) (17) Gp(r) is a transfer function indicating the dynamic characteristics of the load variation amount ΔP ₂ with respect to the load target change amount ΔPr ₂ .

The above is an analysis of the dynamic characteristics of the rolling load when the load control system as shown in FIG. 2 is configured, and the plate thickness deviation Δh ₂ of the second stand at this time is determined below.

First, if we rewrite equation (13), we get ΔS ₂ +ΔRe ₂ =−M ₂ +Q ₂ /M ₂ Q ₂ (ΔP ₂ −ΔPd ₂ ) ...(18), so if we substitute this into equation (1), we get (19 ), Δh ₂ = −M ₂ +Q ₂ /M ₂ Q ₂ (ΔP ₂ −ΔPd ₂ ) + ΔP ₂ /M
₂ = (1/M ₂ + 1/Q ₂ )ΔPd ₂ −1/Q ₂ ΔP ₂ (19) Substituting equation (17) into this gives equation (20).

Δh ₂ = -1/Q ₂ Gp(r) ΔPr ₂ +(1/M ₂ +1/Q ₂ Gp(
r)) ΔPd ₂ + (1-Gp(r)) M ₂ /M ₂ +Q ₂ ΔRe ₂ ……(
20) Equation (20) represents the plate thickness deviation when the load control system is configured as shown in Figure 2.

By the way, in the present invention, the plate thickness fluctuation amount Δhd ₂ due to disturbance in the stand is predicted, and the command value of the load control system (i.e., the load target change amount) ΔPr ₂ is calculated as (21)
Determine by formula.

∆Pr ₂ = (M ₂ + Q ₂ ) ∆hd ₂ = (1 + Q ₂ /M ₂ ) ∆Pd ₂ ... (21) In this case, the plate thickness deviation ∆h ₂ of the stand concerned can be calculated by converting equation (21) to equation (20). By substituting into , it can be seen that it is given by the following equation (22).

Δh ₂ = (1-Gp(r)) ΔPd ₂ /M ₂ + (1-Gp(r)) M
₂ /M ₂ +Q ₂ ΔRe ₂ = (1-Gp(r)) (Δhd ₂ +M ₂ /M ₂ +Q
₂ ΔRe ₂ ) ...(22) Equation (22) represents the plate thickness variation due to disturbance Δhd ₂ and the plate thickness deviation due to the roll eccentricity ΔRe ₂ of the stand when the present invention is implemented. be.

Next, the frequency characteristics (dynamic characteristics) of the transfer function Gp(r)
Let's take a look.

The above-mentioned transfer function Gp(r) can be expressed as in equation (23).

Gp(r)=1/1+rT+r ² TTv=1/TTv
/r ² +r1/Tv+1/TTv...(23) Now when we define ω _o and ζ as follows Equation (23) becomes like equation (24).

Gp(r)=ω _o ² /r ² +2ζω _o r+ω _o ² (24) This is known as a “second-order lag system,” where ω _o is the natural frequency and ζ is the damping coefficient.

Incidentally, the frequency response of the second-order lag system is expressed as in equation (25) as is well known.

Gp (jω) = ω _o ² / (ω _o ² − ω ² ) + j2ζω _o ω……(25
) However, j: imaginary unit ω: angular frequency (rad/sec) And when the angular frequency is ω < ω _o , in other words, when the frequency is f < _fo (here, _fo = ω _o /2π) ( 26)
The formula holds true.

Gp(jω)1...(26) Using this property to investigate the frequency response of the plate thickness deviation _Δh2 , we get the following. In other words, since equation (27) holds true from equation (22), Δh ₂ (jω) = (1−Gp(jω)) (Δhd ₂ (jω) + M ₂
/M ₂ +Q ₂ ΔRe ₂ (jω)) (27) When ω<ω _o , in other words, f< _fo , equation (28) holds true from equation (26).

Δh ₂ (jω)0 ... (28) By the way, as is well known, the natural frequency ω _o of the hydraulic pressure reduction system is ω _o =60 rad/sec or more, in other words, f _o = 10 Hz.
That's all.

On the other hand, the frequency component of the plate thickness variation Δhd ₂ due to disturbance is f < 0.5 Hz, and the frequency component of the roll eccentricity ΔRe ₂ is f < 5 Hz, so in any case, f < _fo holds true (28 ), that is, Δh ₂ 0 holds true.

As explained above, instead of controlling so that ΔS + ΔP/M → 0 as in the conventional plate thickness control system, the target load change amount ΔPr ₂ is determined based on the measured data of the upstream stand as in the present invention. , ΔP→ΔPr _{2 ,} Δh ₂ =0 can be achieved regardless of the plate thickness fluctuation amount Δhd ₂ due to disturbance and the roll eccentricity ΔRe ₂ .

DESCRIPTION OF THE PREFERRED EMBODIMENTS An apparatus for carrying out the method of the present invention will be specifically described below with reference to the drawings. FIG. 1 is a schematic diagram showing an apparatus and its control system for carrying out the method of the present invention, and FIG.
The figure is also a block diagram of the load control system in the second stand. 1 in the figure is the rolled material, ST ₁ , ST ₂
shows a stand located on the upstream side, ie, the first stand, and a stand located on the downstream side, ie, the second stand, among adjacent stands in the moving direction of the rolled material 1 in the continuous rolling mill. 2 is a calculation device which calculates the amount of load fluctuation ΔP ₁ detected by the load cell 21 attached to the first stand, and the amount of reduction change detected by the electric reduction device 22. In addition to ΔS ₁ , the inlet thickness variation ΔH ₁ detected by the thickness gauge 23 placed on the entrance side of the first stand is read, and the mill rigidity coefficient M ₁ of the first stand is read from the rolling schedule calculation device 3. Other partial differential coefficients (∂P/∂h) ₁ , etc. are read, and based on these, the amount of plate thickness variation Δhd ₂ caused by disturbance at the second stand is calculated in accordance with equation (9), and this predicted value is the plate thickness variation. The quantity Δhd ₂ is output to register 4 and stored therein.

The register 4 reads the number of pulses emitted from the pulse generator 31 attached to the first stand in accordance with the circumferential speed of the work roll, in other words, the moving speed of the material to be rolled 1, and calculates the plate thickness fluctuation amount Δhd due to disturbance. The plate thickness fluctuation amount Δhd ₂ is output to the calculation device 5 in synchronization with the predicted position reaching the second stand. The calculation device 5 calculates the mill stiffness coefficient for the second stand from the rolling schedule calculation device 3.
_M2 , other partial differential coefficients (∂P/∂h) _2, etc. are read, and based on these, the target load change amount is calculated according to equation (21), and this is input to the comparator 6. The load variation amount ΔP 2 detected by the load cell 61 attached to the second stand is input to the comparator 6, and the comparator 6 inputs the load variation amount ΔP ₂ and the target load change amount ΔPr ₂ at the second stand. Compare with ΔP ₂ ,
In order to eliminate the deviation, in other words, to make the load fluctuation amount ΔP ₂ match the target load change amount ΔPr ₂ , a command signal is issued to the servo valve 8 for controlling the pressure oil for the hydraulic pressure lowering device 9, and the hydraulic pressure is lowered. Device 9
The amount of reduction of the second stand is adjusted by , and the load variation amount ΔP ₂ is made to match the target load change amount ΔPr ₂ .
Specifically, as shown in FIG.
The target load change amount ΔPr ₂ and the load variation amount ΔP ₂ detected by the load cell 61 are input into the servo valve 8, and a signal corresponding to the difference between them is sent to the servo valve 8. The servo valve 8 operates to flow an amount of pressurized oil obtained by multiplying the difference signal by a proportional gain Kv. The dynamic characteristics of the servovalve are expressed by the first-order lag of the time constant Tv. As the pressure oil flows into or out of the hydraulic cylinder, the pressure reduction changes by ΔS ₂ .

At this time, the load fluctuation amount ΔP ₂ changes as shown in equation (13).

This load variation amount ΔP ₂ is made to match the target load change amount ΔPr ₂ . The above-mentioned control is performed by determining a set of adjacent stands of a continuous rolling mill and controlling the plate thickness between the first stand located on the upstream side and the second stand located on the downstream side for each set. Although the case where this is carried out is shown, it goes without saying that this may be carried out between any two stands.

As described above, in the method of the present invention, in order to make the amount of load fluctuation, including the amount of fluctuation in rolling load caused by roll eccentricity detected by the downstream stand, match the target load change amount, the downstream stand is adjusted according to the deviation. Since the amount of rolling reduction is adjusted, roll eccentricity can be understood as the amount of variation in rolling load, and there is no need for a device to measure the phase, size, etc. of roll eccentricity, or a storage device for the measured values, and the equipment for the control system can be improved. This has the excellent effect of simplifying both the control process and the control process, and also eliminating plate thickness variations caused by skid marks and the like as well as plate thickness variations caused by roll eccentricity.

[Brief explanation of the drawing]

FIG. 1 is a schematic diagram showing an apparatus for carrying out the method of the present invention and its control system, and FIG. 2 is a block diagram showing a load control system in a downstream stand. DESCRIPTION OF SYMBOLS 1... Material to be rolled, 2... Computing device, 3... Schedule computing device, 4... Register, 5... Computing device, 6... Comparator, 8... Servo valve, 9...
...Hydraulic pressure reduction device, 21...Load cell, 22...
Electric rolling down device, 23...Thickness gauge, 31...Pulse generator.

Claims (1)

[Claims] 1. Find the inlet thickness change, outlet thickness change, and load variation in the upstream stand of any two stands in a continuous rolling mill, and calculate the thickness variation that should occur at the exit side of the downstream stand based on these. Based on this plate thickness variation, the target load change amount that should be set on the downstream stand to eliminate this variation is calculated, and the rolling reduction caused by roll eccentricity detected on the downstream stand is A method for controlling plate thickness in a continuous rolling mill, comprising adjusting a reduction amount of a downstream stand so that a load variation amount including a load variation amount matches the load target change amount.