JPH0587337B2 - - Google Patents

Description

[Detailed description of the invention]

(Industrial Application Field) The present invention relates to a plate thickness control method in which plate thickness is controlled in a tandem rolling mill by rolling control and tension control. (Prior art and its problems) In tandem rolling mills used for cold rolling of steel plates, the causes of variations in the thickness of the exit side of each rolling stand are variations in the thickness of the input side and variations in the material hardness of the rolled plate. . For this reason, there is a need for plate thickness control that can deal with both of these issues, and the following two methods have become mainstream as conventional control methods used for such purposes. Conventional method () A method that combines the gauge meter AGC on the #1 stand on the upstream side and the X-ray monitor AGC on the final stand. Conventional method () A method in which the plate thickness at the exit side of each stand is varied by combining constant tension control between stands and reduction control by opening and closing the roll gap. Among these, the conventional method () has self-retention properties against changes in material hardness and is highly effective in this regard, but since tension control is not performed, the tension balance often collapses, and rolling There is a problem that the stability of Also,
Since the intermediate stand is not provided with a thickness control function, there is a problem that the overall thickness control effect is poor, and it is particularly difficult to eliminate leading and trailing end off-gauge. On the other hand, in the conventional method (), all variations in the thickness of the outlet plate due to variations in the plate thickness on the inlet side and material hardness are controlled by rolling down, so the load variation at each stand becomes excessive, disrupting the load balance, and causing problems in the product. There are problems such as poor shape and squeezing due to high loads. (Objective of the Invention) This invention is intended to overcome the above-mentioned problems in the prior art, and has a high overall plate thickness control effect and rolling stability, can reduce the amount of off-gauge, and The purpose of the present invention is to provide a plate thickness control method that has small load fluctuations and does not cause product shape defects or shrinkage due to high loads. (Means for Achieving the Object) In order to achieve the above-mentioned object, in the present invention, the amount of variation in the entrance side plate thickness and the amount of variation in material hardness of the plate material to be rolled in each rolling stand are determined, and the amount of variation in the thickness at the entrance side is determined. and the material hardness fluctuation amount is distributed to form first and second disturbance amounts, and the exit side plate thickness fluctuation due to the first disturbance amount is suppressed by reduction control,
Variations in the thickness of the exit side plate due to the second disturbance amount are suppressed by tension control using a target value correction method that takes into account the interference of the tension of the plate material to be rolled between the respective rolling stands. (Example) A Control Principle of Example FIG. 1 is a conceptual diagram of a plate thickness control device used in an example of the present invention. In this embodiment, a tandem rolling mill formed by arranging five rolling stands is used to control the thickness of a rolled plate (for example, a steel plate) Z during cold rolling. Therefore, before explaining the specific structure and operation of the apparatus shown in FIG. 1, the control principle in this embodiment will be explained. As described above, variations in the thickness of the exit side of each stand in a tandem rolling mill are caused by variations in the thickness of the inlet side and changes in material hardness. Moreover, each stand is an interference system connected by the tension of the rolled plate material between the stands. Therefore, by incorporating such interference, the i-th stand (hereinafter referred to as
Write “#i stand”. ) at Δh _i is the following
It can be expressed by equation (1). Δh _i = (∂h _i /∂H _i )ΔH _i + (∂h _i /∂Q _i )ΔQ _i + (∂
h _i /∂S _i )ΔS _i + (∂h _i /∂t _f,i-1 )Δt _f,i-1 + (∂h _i
/∂t _f,i )Δt _f,i (i=1,
2,...5)...(1) Here, ΔH _i = # Amount of change in the entrance side plate thickness at the i stand, ΔQ _i = # Amount of material hardness variation at the i stand, ΔS _i = # Amount of rolling reduction control at the i stand, Δt _{f ,i} = forward tension target value correction amount at ∂i stand, (where Δt _f,0 = Δt _f,5 = 0) (∂h _i /∂H _i ), (∂h _i /∂Q _i ), ( ∂h _i /∂S _i ), (
∂h _i /∂t _f,i ) = various influence coefficients. The various variables and constants appearing in this equation (1) can be determined as follows. First, the material hardness variation ΔQ _i is the material hardness variation ΔQ 1 on the entry side of the #1 stand, and the material hardness variation ΔQ ₁ on the entrance side of the #1 stand.
1) By sequentially applying the recurrence formula: ΔQ _i =ε _i ΔQ _i-1 (2) using the hardening coefficient ε _i (i = 2 to 5) between the stand and the #i stand. You can ask for it. Here, ΔQ _i can also be obtained by actual measurement, and ΔP ₁ = [M ₁ ∂P ₁ /∂H ₁ /(M ₁ +Q ₁ )]ΔH ₁ −[MQ ₁ /(M ₁ + Q ₁ )] ΔS ₁ + [M ₁ ∂P ₁ / ∂Q ₁ / (M ₁ + Q ₁ )] ΔQ ₁ + [M∂P ₁ / ∂t _f,1 / (M ₁ + Q ₁ )] Δt _{f ,1} ...(3) (where ΔP ₁ and M ₁ are the load fluctuation of #1 stand and Mill constant, respectively) to obtain the following equation (4): ΔQ ₁ = [(M ₁ + Q ₁ ) /M ₁ ∂P ₁ /∂Q ₁ ]ΔP ₁ + [Q ₁ /∂P ₁ /∂Q ₁ ]ΔS ₁ − [∂P ₁ /∂t _f,1 /∂P ₁ /
It can also be calculated by calculation based on ∂Q ₁ ]Δt _f,1 −[∂P ₁ /∂H ₁ /∂P ₁ /∂Q ₁ ]ΔH ₁ (4). That is, in the latter case, ΔH ₁ is obtained by installing an X-ray thickness gauge on the entrance side of #1 stand and tracking the amount of variation in the entrance side plate thickness obtained thereby, and further calculating ΔP ₁ , ΔS ₁ , Δt _{f, 1} in real time by actual measurement, various influence coefficients: (∂P ₁ /
By giving ∂Q ₁ ),... in advance,
ΔQ ₁ can be obtained from equation (4). In addition, in equation (1), the amount of variation in the outlet plate thickness due to the variation in the inlet plate thickness: (∂h _i /∂H _i )ΔH _i …(5) is calculated based on the gauge meter formula, for example, [Q _i / (M _i +Q _i )]ΔH _i …(6) or using a mass flow gauge, V _i-1 (1+f _i-1 )/V _i (1+f _i )H _i −h _i0 …(7) However, it can be determined as follows: V _i = rolling speed of the ∂i stand (mill motor speed), f _i = advance rate of the ∂i stand, h _i0 = target exit plate thickness of the ∂i stand. In this way, each variable and constant in equation (1) can be found. Here, the disturbance vector whose components are ΔH _i and ΔQ _i : ΔH→=(ΔH ₁ , ΔH ₂ ,..., ΔH ₅ ) ^T …(8) ΔQ→=(ΔQ ₁ , ΔQ ₂ ,…, ΔQ ₅ ) ^T …(9) is added, to which the control vector: Δu→ _S = (ΔS ₁ , ΔS ₂ ,…, Consider the case of suppression using ΔS ₅ ) ^T … (10) Δu→ _t = (Δt _f,1 , Δt _f,2 , ..., Δt _f,4 ) ^T … (11). However, in these, "T" represents transposition. Then, the plate thickness variation vector of the plate thickness at the exit side of each stand of the tandem rolling mill: Δh→=(Δh ₁ , Δh ₂ , ..., Δh ₅ ) ^T ...(12) is (1), (8) to (11) According to the formula, it can be written as Δh→= A _H ΔH→+A _Q ΔQ→+A _S Δu→ _S +A _t Δu→ _t (13). Here, A _H = diag[(∂h ₁ /∂H ₁ ), (∂h ₂ /∂H ₂ ),…, (∂
h ₅ /∂H ₅ )] …(14) A _Q = diag[(∂h ₁ /∂Q ₁ ), (∂h ₂ /∂Q ₂ ),…, (∂
h ₅ /∂Q ₅ )] …(15) A _S = diag[(∂h ₁ /∂S ₁ ), (∂h ₂ /∂S ₂ ),…, (∂
h ₅ /∂S ₅ )]…(16)

[Table] …(17)
The symbol "diag" represents a matrix in which each component is arranged diagonally and O is the off-diagonal element. Then, the control vectors Δu→ _S ^* and Δu→ _t ^* are determined such that Δh→ in the above equation (13) is O. In this example, Δh→ is expressed as Δh→ _H = A _H ΔH →+A _S Δu→ _S …(18) Δh→ _Q = A _Q ΔQ→+A _t Δu→ _t …(19) By the two parts Δh→ _H and Δh→ _Q , the following ( 20) Think about it separately as in Eq. Δh→=Δh→ _H +Δh→ _Q …(20) As can be seen from equations (18) and (19), Δh→ _H is the amount of change in the thickness of the outlet side A _H ΔH→ due to the change in the thickness of the inlet side, and the control by the rolling down control. It is composed of the quantity A _S Δu→ _S ,
On the other hand, Δh→ _Q is composed of a variation A _Q ΔQ→ in the outlet side plate thickness due to a variation in material hardness, and a controlled amount A _t Δu→ _t due to tension control using the target value correction method. and,
In this embodiment, control is performed such that Δh→ _H and ΔH→ _Q are individually set to O→. In other words, the disturbance that causes the change in the thickness of the outlet side is distributed into a first amount of disturbance formed by the change in the thickness of the inlet side ΔH→ and a second amount of disturbance formed by the change in material hardness ΔQ→ Therefore, the former plate thickness variation on the exit side is suppressed by rolling down control, and the latter plate thickness variation on the outlet side is suppressed individually by tension control based on the target value correction method. In addition,
Other methods of forming the first and second disturbance amounts will be described later. First, in order to change Δh→ _H in equation (18) to O→, A _H ΔH→+A _S Δu→ _S ^* =O→ (21). Therefore, the inverse matrix of A _S is A _S ^-1
Then, Δu→ _S ^* can be found as follows: Δu→ _S ^* =−A _S ^-1 A _H ΔH→ (22). As can be seen from equation (8), ΔH→ is a five-component vector, and A _S ^-1 A _H is (5×
5), Δu→ _S ^* is completely defined by this equation (22). On the other hand, to set Δh→ _Q in equation (18) to O→, A _Q ΔQ→+A _t Δu→ _t ^* =O→ (23). However, in this case, as seen from equation (17), A _t is a (5 × 4) matrix, so
There is no inverse matrix of A _t , and Δ
It is not possible to determine u→ _t ^* . Therefore, for example, Δu→ t _* that minimizes J(u→ _t )=Δh→ _Q ^T WΔh→ _Q (24 ⁾ is determined by the weighted least squares method. However, W is a weight matrix and is defined as W=diag[W ₁ , W ₂ , ..., W ₅ ] ...(25), and the weight component W _i (i=1, 2, ..., 5)
For example, when giving top priority to #5 stand exit side plate thickness control, W ₁ < W ₂ <...<W ₅ (26) and exit side plate thickness control of #1 and #5 stands. When giving priority to the same degree, W ₂ , W ₃ , W ₄ <W ₁ =W ₅ (27) are constants determined respectively. Specifically, when Δu→ _t ^* is determined by ∂J(u→ _t )/∂u→ _t = O …(28), Δu→ _t ^* = −(A _t ^T WA _t ) ⁻¹ A _t ^T WA _Q ΔQ→ …(29). In order to avoid interference between actuators to be operated, tension control based on Δu→ _t ^* is realized by speed control of the mill motor of each stand. Note that the coefficient matrix appearing in equations (22) and (29): A _S ^-1 A _H …(30) (A _t ^T WA _t ) ^-1 A _t ^T WA _Q …(31) is Find it by calculation. B. Specific Structure and Operation of the Embodiment The specific structure and operation of the embodiment shown in FIG. 1, which performs plate thickness control in accordance with such a control principle, will be explained with reference to the same figure. This embodiment has an arrangement of five rolling stands as described above, and each stand is equipped with a rolling roll 1, a hydraulic rolling device 2, a mill motor 3, a load cell 4, and the like. Tension meters T ₁₂ , T ₂₃ , . . . , T ₄₅ are provided between each stand. In order to perform the above control in this device, first, the thickness H 1 measured by the X-ray thickness meter 5 installed on the entrance side of the # ₁ stand is compared with the target thickness H ₀ set in advance. By taking the difference, the plate thickness deviation: ΔH ₁ = H ₁ − _{H 0} ...(32) is determined (block 11 in Figure 1), and the deviation is calculated according to the speed of the rolled plate material Z, which is actually measured or calculated from the roll speed. Tracking. On the other hand, the material hardness change ΔQ ₁ in the #1 stand is determined by taking the difference between the reference hardness Q ₀ and the hardness Q ₁ on the entrance side of the #1 stand obtained by actual measurement (block 21). However, as shown by the broken line around the same block 21, this ΔQ ₁ is the measured value of the load deviation ΔP 1 in the #1 stand, the value obtained by tracking ΔH ₁ , and the value obtained by tracking the load deviation ΔP ₁ in the #1 stand and the #2 stand. It may also be determined by calculating equation (4) using the tension deviation Δt _f,1 , etc. This ΔQ ₁ is also tracked according to the speed of the plate material Z to be rolled. Next, when the relevant part of the plate material Z to be rolled comes directly under the #1 stand, use equation (1) to calculate Δh ₁ using, for example, a mass flow gauge, and set this as ΔH ₂ (block 12). . However, ΔS ₁ in equation (1),
For Δt _f,1, etc., actual measured values or calculated values are used, and for ΔQ ₁ , the value obtained in block 21 is used. Regarding the material hardness variation ΔQ ₁ , ΔQ ₂ is also determined by multiplying ΔQ ₁ by the hardness coefficient ε ₁ according to equation (2) (block 22). Similarly, as shown in blocks 10 and 20 in FIG. ₁ and in _FIG . It is performed up to the buttocks. In this way, in the state where the disturbance vectors ΔH→ and ΔQ→ are found, ΔH→ has A _S ^-1 A _H , and ΔQ→ has (A _t ^T WA _t ) ^-1 A _t ^T WA _Q is respectively multiplied to calculate the reduction control vector Δu→ _S ^* and the tension target value correction vector Δu→ _t ^* (blocks 13 and 23). Then, Δu→ _S ^* is directly output to the hydraulic pressure lowering device 2 via the hydraulic control system 14. On the other hand, the tension deviation value Δu→ _t measured by the tension meters T ₁₂ , ... is subtracted from Δu→ _t ^* output from the block 23, and a new target deviation (Δu→ _t ^* −Δu → _t ). Then, in the tension control system 24, the set tension value is changed according to this new target deviation, and the speed of the mill motor 3 is changed accordingly. Therefore, here, the "constant value (=target value)" in the tension "constant" control is corrected. Since control is performed that takes into account the interference of the tension between each stand, the tension between each stand is changed to an appropriate value as a whole. In this way, variations in the thickness of the outlet side due to variations in the plate thickness on the inlet side can be controlled by reduction control, and variations in plate thickness on the outlet side due to changes in material hardness can be controlled using a tension target value correction method that incorporates the interference of tension between each stand. Each is suppressed independently by tension control. C Data Example FIGS. 3 and 4 show data examples comparing the effects of the above embodiment and the conventional method, respectively. Of these, FIG. 3 shows the variation in the exit side plate thickness (a in the figure) and the variation in tension (b in the figure) for the conventional method () already explained in the "Prior Art" section and the above embodiment. This shows changes over time. As can be seen from FIG. 3, the above embodiment has the following advantages over the conventional method (). In the above-mentioned embodiment, variations in the outlet side plate thickness due to variations in material hardness are suppressed with higher precision than in the conventional method (FIG. 1A). Since tension control is performed, tension fluctuations are small, and the tension is stably controlled around the set value (FIG. 3(b)). Additionally, in connection with these, there is also the advantage that the amount of off-gauge at the leading and trailing ends can be reduced. On the other hand, FIG. 4 shows load fluctuations in comparison with the conventional method (2009), and the advantages of this example in this comparison are as follows. Load fluctuations are small and stable because the reduction control only responds to changes in plate thickness on the entry side. Therefore, the rolled plate material Z does not suffer from poor shape or shrinkage due to high loads. In this way, the above embodiments have many advantages compared to each of the conventional methods. D. Modification By the way, this invention is not limited to the above embodiment, and the following generalization is also possible.
That is, in the above embodiment, the first disturbance amount was formed by the entrance side plate thickness variation itself, and the disturbance amount was formed by the material hardness variation itself, and these were suppressed by the reduction control and tension control, respectively. When forming the first and second disturbance amounts, appropriate weights α _i and β _i are assigned to the entrance side plate thickness variation and material hardness variation, respectively, and the two disturbance quantities are distributed. The amount of disturbance may be made to correspond to the pressure reduction control and the tension control, respectively. In other words, the variation amount Δh _si in the first disturbance amount corresponding to the reduction control and the variation amount Δh _ti in the second disturbance amount corresponding to the tension control are expressed as follows:
Δh _si = α _i (∂h _i /∂H _i )ΔH _i +β _i (∂h _i /∂Q _i )ΔQ, respectively.
_i ...(33) Δh _ti =(1−α _i )(∂h _i /∂H _i )ΔH _i +(1−β _i )(∂h _i /∂Q _i )ΔQ _i …(34). However, the weights α _i and β _i are arbitrarily determined for each stand. In this case, the equation corresponding to equation (13) is Δh→=Δh→ _S +Δh→ _t +A _S Δu→ _S +A _t Δu→ _t …(1
3)′ is obtained. However, Δh→ _S = (Δh _S1 , Δh _S2 ,...Δh _S5 ) ^T ...(35) Δh→ _t = (Δh _t1 , Δh _t2 ,...Δh _t5 ) ^T ...(36). By using the same method as in the above embodiment for this equation (13)', the reduction control amount Δu→ _S ^* and the tension control amount Δu→ _t ^* can be obtained from the above equation (22) and (29), respectively.
Corresponding to the formula, Δu→ _S ^* =−A _S ^-1 Δh→ _S …(22)′ Δu→ _t ^* =−(A _t ^T W′A _t ) ⁻¹ A _t ^T Δh→ _t …(29) ′. However, W' is a weight matrix similar to W. Such a method is effective in the following cases. In other words, some rolled plates have one of the disturbances ΔH _i and ΔQ _i considerably larger than the other. If the above-described embodiments are applied to such a device as is, the load will be concentrated on the reduction control or the tension control using the target value correction method, which may cause load imbalance, poor shape, or plate breakage. Therefore, for such special materials,
By determining the magnitude relationship between the above ΔH _i and ΔQ _i based on the steel type and the measured values in the previous process, and appropriately selecting the above α _i and β _i accordingly, good results can be obtained even for special materials. Highly accurate sheet thickness control can be achieved in the rolling state. For ordinary materials, the configuration of the above embodiment is sufficiently effective, and it is not necessary to make such modifications. Note that it can be easily confirmed that if α _i =1 and β _i =0 in the above equations (33) and (34), they match the above embodiments. Therefore, if α _i and β _i are made variable so that they can be set arbitrarily, the method includes the above embodiments. (Effects of the Invention) As explained above, according to the present invention, the first and second disturbance amounts are formed by distributing the entry side plate thickness variation amount and the material hardness variation amount, and the disturbance amount is determined by these disturbance amounts. The resulting plate thickness fluctuations on the exit side are individually suppressed by reduction control and tension control using a target value correction method that incorporates tension interference, resulting in high plate thickness control effects and rolling safety as a whole. In addition, it is possible to obtain a plate thickness control method that can reduce the amount of off-gauge, has small load fluctuations in each stand, and does not cause product shape defects or shrinkage due to high loads.

[Brief explanation of the drawing]

FIG. 1 is a conceptual diagram of control according to an embodiment of the present invention, FIG. 2 is a diagram showing tracking in the embodiment, and FIGS. 3 and 4 are diagrams illustrating the effects of the embodiment. Z...Rolled plate material, 1...Rolling roll, 2...
Hydraulic lowering device, 3...mill motor, 4...load cell, 5...X-ray thickness gauge, ~...rolling stand, _T12 ~ _T45 ...tension meter.

Claims (1)

[Scope of Claims] 1. A method for controlling plate thickness in a tandem rolling mill having two or more rolling stands arranged, which comprises: determining the amount of variation in the entrance side plate thickness and the amount of variation in material hardness of a rolled plate material in each rolling stand; The input side plate thickness variation amount and the material hardness variation amount are distributed to form first and second disturbance amounts, the exit side plate thickness variation due to the first disturbance amount is suppressed by reduction control, and the output side plate thickness variation due to the first disturbance amount is suppressed by reduction control. 2. A plate in a tandem rolling mill, characterized in that variation in the exit side plate thickness due to the amount of disturbance described in item 2 is suppressed by tension control using a target value correction method that takes into account the interference of the tension of the plate to be rolled between each rolling stand. Thickness control method. 2. A plate thickness control method in a tandem rolling mill according to claim 1, wherein the first disturbance amount is the input side plate thickness variation amount itself, and the second disturbance amount is the material hardness variation amount itself. .