JPH067820A - Looper multivariable controller - Google Patents

Abstract

PURPOSE:To absorb tension variation by the vertical movement of a looper by controlling a rolling mill driving main motor and a looper driving motor according to each rotational speed commanding value calculated by using a process model formed by making an interference system of rolled stock tension and looper height a model. CONSTITUTION:Each rotational speed commanding value of the rolling mill driving main motor 11 of a tandem rolling mill and the looper driving motor 7 is calculated. Control gains are calculated by the setting means 15 of a variable for expressing the process model and a weight function for specifying a response to the looper height and tension between stands and robustness and a control gain arithmetic means 16. The rotational speed commanding value of the rolling mill driving main motor and the rotating speed command value of the looper driving motor are calculated by using the control gains selected by a control gain control means 14 to select these control gains properly according to a rolling condition and a rolling state and further, a control and arithmetic means 13. Consequently, the tension variation is absorbed by the vertical movement of the looper and further, the desired response and robustness can be realized.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a looper multi-variable controller for controlling the height of a looper arranged between stands of a tandem rolling mill and the tension between stands of a rolled material.

[0002]

2. Description of the Related Art Plate thickness and plate width are part of the evaluation criteria for final products in hot rolling and cold rolling. Among these, the automatic plate thickness control is performed on the plate thickness, and the automatic plate width control is performed on the plate width. On the other hand, the tension applied to the material being rolled affects the plate thickness and the plate width, so that the tension is also controlled to be maintained at a certain value.

In particular, the rolled material in hot rolling is heated to a high temperature to reduce the deformation resistance of the rolled material, and if the tension is large, the material is likely to break.
If the tension is set low to prevent this breakage, a tensionless state may occur due to disturbance or erroneous setting, and if this state continues, a large loop may occur between the rolling mill stands and cause an accident. Therefore, in the hot rolling mill, a looper device is provided, the tension is controlled by the looper device, and the height of the looper is controlled from the viewpoint of improving the sheet passing property of the material.

In such a rolled material tension and looper height control device, there are interference from the rolled material tension to the looper height and interference from the looper height to the tension. In the conventional tension control, a method of controlling the rolling material tension and looper height by PID control without suppressing the interference and a rolling material tension by adding a decoupling compensator for suppressing these interferences There are a non-interference control method that controls the looper height independently, and a method that considers the interference system between the looper and the rolling material tension as a multivariable system and applies the optimum control theory (Linear Quadratic). It was

[0005]

In the above-mentioned method using PID control, there is no function of suppressing the mutual interference between the tension of the rolled material and the looper, so that the method lacks quick response and stability. Therefore, recently, non-interference control and optimum control are widely used.

Among these, in the decoupling control, as an example for realizing a decoupling device such as a cross controller, a transfer function of the decompensation device for decoupling is realized on a computer. . In general, this transfer function has a high order, and therefore, the adverse effect of the mismatch between the actual plant and the model may be remarkable, or the calculation accuracy on a computer may be problematic. Also, the looper should act as an original function of the looper to suppress tension fluctuations and play a part of tension control, but since the height of the looper is controlled to be constant by decoupling, that function is fully utilized. There wasn't.

On the other hand, in the method based on the optimal control theory, the control gain can be designed so that the rolling mill driving main motor and the looper are used in cooperation as the operating end. In this optimum control, it is difficult to find a causal law between the weighting matrices Q and R in the evaluation function J shown in the following equation (1) and the response of the actual process, and Q which realizes an appropriate response of the entire control system is realized. , R are found by trial and error to determine the control gain.

[0008]

[Equation 1] Here, x is a state quantity of the control target process, u is an operation quantity given to the control target process by the controller, x ^T is a transposition of x, and u ^T is a transposition of u.

Since trial and error are repeated in this manner, much time is required for designing the control system and adjusting the plant. In addition, in the method based on this optimal control theory, since it is necessary to numerically solve the Riccati equation that cannot be solved analytically, the general formula of the optimal control gain including variables may not be obtained.

There is a method of using a gain table instead of obtaining a general expression. However, in this method, a method is generally used in which a control gain that matches the properties of the rolled material and rolling conditions is obtained in advance and a table is created, and that table is referenced when the control gain is used. Therefore, much time and effort are required to determine, maintain, and manage the values of this gain table.

In addition, it is almost impossible to describe all cases in the gain table, and in case of rolling conditions which do not exist in this gain table, the gain must be approximated from a table similar to this rolling condition. The control performance may be degraded.

In the non-interference control and the optimum control, since the controller is designed based on the process model of the strict control object, when the actual process and the model are different, the entire control system becomes unstable. In some cases For this reason, it is common practice to design the controller so that it does not become unstable by reducing the control gain at the expense of some responsiveness. Even in this case, it is not possible to obtain an index indicating how much the actual process and the model differ from each other and the control system becomes unstable.

In recent years, the H∞ control has received attention because it can design a control system having high robustness. What is robustness?
This is the degree to which the entire control system including the controller is stable even if the controlled process changes for some reason or there is a difference between the controlled process and its model. Generally, a robust control system is desirable. The H∞ control has an index of this robustness, and is also characterized in that the controller can be designed in consideration of the robustness.

However, even when the H∞ control is applied to the looper control system, the tension and the height of the looper are generally made non-interfering. Therefore, as in the case of the non-interference control, the tension is changed by using the looper. It was difficult to design the controller to reduce the power consumption.

The present invention has been made to solve the above problems, and an interference system of the height of the looper arranged between the stands of the tandem rolling mill and the tension between the stands of the rolled material is used as a multivariable system. When controlling, it is an object to obtain a looper multi-variable control device that controls the tension of the looper and the rolled material so that the tension fluctuation is absorbed by the vertical movement of the looper.

[0016]

According to the present invention, each of a rolling mill driving main motor and a looper driving motor of a tandem rolling mill is calculated by using a process model in which an interference system of rolling material tension and looper height is modeled. A looper multivariable control device for controlling according to a rotation speed command value, comprising: a tension command value of a rolled material, a command value of a looper height, a variable expressing a process model, and a response and robustness to a looper height and a tension between stands. The setting means for setting each weighting function that specifies the property, and each set variable and weighting function are substituted into the control gain calculation formula that presupposes that the rolling material tension and the looper height are made non-interfering with each other. Control gain calculating means for obtaining the control gain as a numerical value, the calculated control gain is held, and the control gain is adjusted according to the rolling condition and rolling condition. Based on the control gain management means that selects an appropriate control gain, the selected control gain, and the set rolling material tension command value and looper height command value, the rolling material tension follows the tension command value. And a control calculation means for calculating the rotation speed command value of the rolling mill drive main motor and the rotation speed command value of the looper drive motor such that the looper height follows the looper height command value. is there.

Here, in the process model, a new looper is added by adding a value obtained by weighting the difference between the tension command value of the rolled material and the measured value of the rolled material tension to the set command value of the looper height. It is preferable to have a function of setting the height command value.

[0018]

According to the present invention, in calculating the rotational speed command values of the rolling mill drive main motor and the looper drive motor, the variables representing the process model, the response to the looper height and the tension between stands, and the robustness are designated. The control gains are calculated using the weighting function, appropriate control gains are selected according to the rolling conditions and rolling conditions, and the rotation speed command for the rolling mill drive traction motor is further selected using the selected control gains. Since the value and the rotation speed command value of the looper drive motor are calculated, the tension fluctuation can be absorbed by the vertical movement of the looper, and a desired response and robustness can be realized.

Further, in the set command value of the looper height,
Using a process model with the function of adding a weighted value to the difference between the rolled material tension command value and the measured value of the rolled material tension to obtain a new looper height command value, the weight is set by the setting means. By setting, it is possible to reliably absorb the tension fluctuation by the vertical movement of the looper.

[0020]

The present invention will be described in detail below with reference to the embodiments shown in the drawings. FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention. In the figure, the rolled material 1 is rolled in the order of the i-th stand rolling mill 2 and the (i + 1) th stand rolling mill 3. Here, when the total number of stands of the tandem rolling mill is n, n = 5 to 7 is general. The devices such as the looper shown below are installed between the stands, but they can be easily expanded to other stands by considering the state between the i stand and the i + 1 stand. Therefore, only the two stands will be considered here. . Note that i is in the range of 1 ≦ i ≦ n−1.

When the looper 4 is provided between the i-th stand and the (i + 1) th stand, the tension of the rolled material received by the looper roll is detected by the tension detecting device 5, and the angle of the looper arm converted to the height of the looper is determined. It is detected by the looper height detection device 6. A rotation speed detection device 8 is attached to a looper drive electric motor (hereinafter referred to as a looper electric motor) 7 that drives the looper, and the looper electric motor speed control is performed so as to reduce the deviation between the looper electric motor rotation speed and the rotation speed command value. It is controlled by the device 9.

On the other hand, the main machine speed control device 12 for controlling the speed of the rolling mill drive main electric motor (hereinafter, rolling mill drive main electric motor) 11 of the i-th stand is the main machine speed detected by the main machine speed detecting device 10. The current of the i-th stand main engine 11 is controlled so as to reduce the deviation between the detected value and the speed command value.

The speed command value of the looper electric motor 7 and the rotation speed command value of the i-th stand main unit 11 are controlled and calculated by the control calculation means.
Calculated by 13. In this case, the tension command value of the rolled material, the command value of the looper height, the variable expressing the model of the process, the tension between the stands and the response to the looper height, and the weight function for designating the robustness are respectively set by the setting means 15. The control gain calculation means 16 substitutes each variable and the weighting function into a predetermined control gain calculation formula to obtain the control gain as a numerical value. The control gain managing means 14 holds the calculated control gain, and selects an appropriate control gain in accordance with the rolling condition and the rolling state to control calculating means.
Pass to 13.

FIG. 2 shows a setting means 15, a control gain calculating means 16 and a control gain managing means in the control system shown in FIG.
It is a block diagram corresponding to what removed 14.

In FIG. 2, blocks 17 to 28 are processes to be controlled and correspond to elements denoted by reference numerals 1 to 12 in FIG. 1, respectively. Of these, the block 17 is the main engine speed control system, and includes the main engine speed detecting device 10 and the i-th engine in FIG.
A speed control system including a stand main engine 11 and a main engine speed control device 12 is represented as one block. Block 18 is the influence coefficient from the main machine speed to the rolling material speed, block 19 is the tension generation gain and integrator in the tension generation process, block 20 is the feedback gain in the tension generation process, and the tension generation mechanism is modeled in blocks 19 and 20. It has become.

A block 21 is an influence coefficient from the looper motor rotation speed to the rolled material speed, and a block 22 is an influence coefficient from the rolled material tension to the looper motor torque. block
Reference numeral 23 is a feedback gain from the looper height to the looper motor torque, block 24 is a looper speed controller, which corresponds to the looper motor speed control device 9 in FIG. 1, block 25 is a looper motor torque coefficient, and block 26 is a looper motor. , The transfer function from the torque to the rotation speed, the block 27 is the looper damping coefficient, and the block 28 is the transfer function from the rotation speed of the looper motor to the looper height.
Blocks 24, 25, 26 and 27 in FIG. 2 correspond to the looper speed control system including the looper motor 7, the rotation speed detection device 8 and the control device 9 in FIG.
28 corresponds to the looper 4 in FIG.

On the other hand, blocks 29 to 32 in FIG. 2 are shown in FIG.
2 is a controller corresponding to the control calculation means 13, and each of them is a controller represented by a high-order transfer function. FIG. 2 shows the configuration of the controller when designed under certain design conditions. It goes without saying that another configuration may be used if other design conditions are used.

Further, in order to control both the looper height and the rolled material tension in the looper height control system, the value obtained by multiplying the detected tension by the weight shown in the block 34 is added to the looper height detected value. Is used as a feedback value of the looper height, and a value obtained by multiplying the tension command value by the weight shown in block 33 is added to the looper height command value to obtain a new looper height command value. Block 17 in Figure 2
When the controlled object process model up to ~ 28 is written as a state equation, the following equations (2) and (3) are obtained.

[0029]

[Equation 2]

[0030]

[Equation 3] Here, “Δ” added before each symbol represents a slight change in the symbol. Further, “•” added above each symbol represents a time derivative. So for example Is expressed, where t represents time.

Further, T represents transposition, and the state vector x = [Δt _f Δω _L Δθ ΔV _r
Δx _H ] ^T output vector y = [Δt _f Δθ] ^T input vector u = [ΔV _RREF Δω _LREF ] ^T

[0032] _{· x = Ax + Bu y =} Cx ... (4) However, A, B, C is (2), it represents the following matrix in equation (3).

[0033]

[Equation 4] The meaning of each variable in the above equation of state is as follows.

G _r : Gear ratio between looper and looper motor J: Looper motor inertia factor K ₁₀ : Tension feedback coefficient E: Young's modulus of rolled material L: Distance between stands t _f : Forward tension V _r : Speed of main machine Z: Looper damping coefficient α: Influence coefficient from main machine speed to rolling material speed θ: Looper height (expressed in angle) φ: Looper motor torque coefficient ω _L : Looper rotation speed T _v : Main machine speed control system time constant F ₁ : looper (weight of the rolled material, the load torque by the looper own weight) gain from the height to the looper drive torque F _2: looper influence coefficient from the rotational speed to the strip speed F _3: influence coefficient from the tension to the looper motor torque x _H: looper speed controller internal variables K, T _21: looper speed controller control constants subscript REF: command value of that symbol, where the looper in the looper height control system In order to control both the To rolling material tension is, to change the above-mentioned equation (3) as follows.

[0035]

[Equation 5] Note that the control amount Δθ in equation (3) is Δy in equation (5).
_2, and this Δy ₂ is changed by the following equation.

Δy ₂ = C ₁ · Δt _f + Δθ (6) If the weight C ₁ is increased, the controllability of the rolled material tension t _f is increased, and the rolled material tension itself is well controlled, but the looper height is increased. The fluctuation of θ becomes large. Further, if the weight C ₁ is reduced, the specific gravity of the rolled material tension t _f is reduced and the looper height θ is increased.
Is controlled to be constant. If the weight C _{1 is set} to 0, it becomes the same as the model shown in the equation (3).

Blocks 33 and 34 in FIG. 2 are added by introducing C ₁ in the equation (5). The equivalent conversion of FIG. 2 can be drawn as shown in FIG. From FIG. 3, C ₁ is the tension command value Δt _fREF of the rolled material and the actual tension Δt.
When a deviation occurs between _f and the looper height command value, the target looper height value is changed from C ₁ · (Δt _fREF −Δ
It can be understood that it has the meaning of trying to absorb the tension fluctuation by changing only t _f ).

The method of determining the transfer function of blocks 29 to 32 in FIG. 2 corresponding to the control calculation means 13 in FIG. 1 is as follows.

Basically, it is determined by using H∞ control. Incidentally, FIG. 4 is a block diagram of the H ∞ control when C ₁ is not introduced, and is an example obtained when the controller is designed by the H ∞ control using the equations (2) and (3). The outline of the H ∞ control will be described with reference to FIG. 4 as an example.

In FIG. 4, weighting functions W ₁₁ , W ₁₂ , W ₂₁ and W ₂₂ are added for designing the controller, which are represented by blocks 39, 40, 41 and 42, respectively. This weight function is generally given by a high-order transfer function. FIG. 5 shows the relationship between the frequency and the weighting function, the sensitivity function and the complementary sensitivity function. In FIG. Is set.

In the H∞ control, the following mixed sensitivity problem is generally set. What is this mixed sensitivity problem?

[0042]

[Equation 6] A dude ‖ - ‖ _H∞ is ‖E (jω) || _H∞ <1 and E as representing the H∞ norm (j [omega]) is a problem which issues Heading a transfer function such that the internal stability. In order to derive this transfer function, a design CAD system for gain calculation is necessary, and the transfer function is obtained by numerical expression. Therefore, the designer can obtain the transfer function by setting the weighting functions such as W ₁₁ and W ₁₂ and using the design CAD system for gain calculation.

By designating these four weighting functions, the desired quick response and robustness of the control system can be obtained.
The mechanism is as follows. FIG. 5 shows W ₁₁ , W ₁₂ , sensitivity functions G _STC , G _SHC, and complementary sensitivity functions G _TTC , G _THC among the four weighting functions. Here, the subscript TC indicates tension control, and HC indicates looper height control. Generally, the sensitivity function mainly relates to the quick response of the control system, and the complementary sensitivity function mainly relates to the robustness of the control system. Therefore, the sensitivity function G _STC is the tension control speed response, the sensitivity function G _SHC is the looper height control response, the complementary sensitivity function G _TTC is the tension control robustness, and G _THC is the looper height control robustness. It serves as an indicator of sex.

The sensitivity function and the complementary sensitivity function are the responses of the closed loop system after the weighting function is set and the transfer function is calculated. The sensitivity function G _STC and the complementary sensitivity function G _TTC related to tension control are the weighting function W. ₁₁ , determined by W ₂₁ ,
The sensitivity function G _SHC and the complementary sensitivity function G _THC related to the looper height control are determined by the weighting functions W ₁₂ and W ₂₂ .

The index of the quick response is the frequency in the vicinity of where the sensitivity function cuts the 0 db line, and in FIG. 5, the response of the tension control is approximately 7 rad / s in terms of the crossing angular frequency. The robustness index is the gain difference between the complementary sensitivity function and the reciprocal of the weighting function, and the robustness index of the looper height control system in FIG. 5 is about 20 db which is the difference between W ₂₂ ^-1 and G _THC .
This is because the error between the actual process and the model is about 20db (= 10
It means that stability will be maintained even if there is.

Designing the robustness means that the control system is stable even if the process to be controlled changes over a wide range, and a kind of controller gain can cope with a wide range of rolling conditions. Therefore, it is not necessary to have many control gains depending on the rolling condition.

For the process model of FIG. 2 or FIG. 3 according to this embodiment, the transfer function can be determined in exactly the same manner as in the case of FIG. H∞ for conventional process models
In the case of the control system of FIG. 4 to which the control is applied, the tension and the looper height are evaluated separately to design the response and robustness of each. For this reason, the tension and the height of the looper are made almost non-interfering with each other, and it is difficult to design the tension fluctuations to be absorbed by the vertical movement of the looper.

On the other hand, the control system shown in FIGS. 2 and 3 according to this embodiment evaluates the tension and the looper height + weighted tension. Therefore, it is possible to design the looper to suppress the tension fluctuation. As a result, when the transfer function of FIG. 2 and the transfer function of FIG. 4 are compared, the structure of the block 30 is different, and even if the block 30 is designed with the same sensitivity function and complementary sensitivity function, FIG. 2 and FIG. Block 29-32
The values of the coefficients K ₁₁ , a ₁₁₁ , b _110, etc., which form the above are also different.

In this embodiment, if the control gain calculated by the setting means 15 under the same or similar conditions and conditions as the rolling condition and the rolling condition is recorded in the control gain managing means 14, the control gain is obtained. Control calculation means
Pass to 13.

When there is no control gain with the same or similar rolling conditions and rolling conditions, the above equations (2) and (5) are used.
The variables T _v , E, L, etc. in the formula and the weight C ₁ are variables representing the process model, and specific values are set according to the rolling conditions and rolling conditions, and W ₁₁ , W ₁₂ , W ₂₁ , W ₂₂ are The weighting functions are set by the setting means 15 as a weighting function for designating the response of the rolled material tension and the looper height and the robustness, and are passed to the control gain calculating means 16.

When setting the various parameters by the setting means 15, it is possible to automatically determine whether or not the control gain management means 14 has control gains having the same or similar rolling conditions and rolling conditions. Alternatively, a person may judge. Further, a so-called table pickup method, in which control gains corresponding to all presumed rolling conditions and rolling conditions are calculated and stored and the control gains are selected at the time of operation, can also be implemented.

FIG. 6 shows a simulation result of the control system according to this embodiment. This is a simulation of a 7-stand hot rolling mill. 6 (a), a, b, ..., F respectively represent the tension between the first and second stands, the tension between the second and third stands, ..., The tension between the sixth and seventh stands, and g in FIG. 6 (b). , H, ..., L are the looper height between the 1st and 2nd stands, the looper height between the 2nd and 3rd stands, ..., 6th-
Shows the height of the looper between 7 stands. The target process in this simulation is the block shown in Fig. 2.
Not a simplified model as shown in 17-28, but a model that describes the skid marks added during hot rolling, disturbances such as roll eccentricity, main machine control system, looper controller, and automatic plate thickness control system in detail. It is used to simulate the actual rolling with high accuracy. Also, in this simulation, accelerated rolling is performed from 27 seconds to 53 seconds, and it is premised that the rolling material tension and looper height are made non-interfering under the same conditions where the period of disturbance is shortened in the latter half of the simulation. FIG. 7 shows the result of simulation with the designed control system. 7 (a), m, n, ..., r represent the tension between the 1st and 2nd stands, the tension between the 2nd and 3rd stands, ..., and the tension between the 6th and 7th stands, respectively, and s in FIG. 7 (b). , T, ..., x
Represents the looper height between the first and second stands, the looper height between the second and third stands, ..., The looper height between the sixth and seventh stands, respectively. Here, since the looper height is almost fixed, it does not seem that the tension fluctuation is suppressed by the vertical movement of the looper.

As a result of this simulation, in the present embodiment, the variation of the looper height becomes large, but the variation of the rolling material tension can be suppressed to be small as compared with the method of making the rolled material tension and the looper height non-interfering. I can confirm. Therefore, by designing the variation of the looper height within the allowable range by the weight C ₁ or the like, the rolled material tension can be controlled better than in the case of non-interference.

In the above embodiment, a quadruple rolling mill in which a backup roll is arranged outside the work roll, and a looper provided between these rolling mills is driven by an electric motor is used. Is not limited to this, it is clear that the present invention can be applied to other rolling mills provided with an intermediate roll or the like, or to those hydraulically driving a looper arranged between rolling mills. Is.

[0055]

As is apparent from the above description, the present invention calculates the control gain by using the variable expressing the process model, the response to the looper height and the tension between stands, and the weighting function for designating the robustness, These control gains are selected according to the rolling conditions and rolling conditions, and the selected control gains are used to determine the rotation speed command value for the rolling mill drive main motor and the rotation speed command value for the looper drive motor. Since the calculation is performed, the tension fluctuation can be absorbed by the vertical movement of the looper, and further, desired response and robustness can be realized.

Further, the set command value of the looper height is
By using a process model equipped with a function that adds a weighted value to the difference between the rolling material tension command value and the rolling material tension measurement value to create a new looper height command value, tension fluctuations It can be reliably absorbed by moving up and down.

[Brief description of drawings]

FIG. 1 is a block diagram showing the overall configuration of an embodiment of the present invention together with a rolling system.

FIG. 2 is a block diagram showing a detailed configuration of a main part of one embodiment of the present invention.

FIG. 3 is a block diagram showing a detailed configuration of a main part of an embodiment of the present invention.

FIG. 4 is a block diagram showing a detailed configuration of a main part of a conventional multivariable control device, which is shown for comparison with an embodiment of the present invention.

FIG. 5 is an explanatory diagram for explaining quick response and robustness according to an embodiment of the present invention.

FIG. 6 is a diagram showing the relationship between tension and looper height and time for explaining the operation of one embodiment of the present invention.

FIG. 7 is a diagram showing the relationship between tension and looper height and time in a conventional multi-variable control device in order to be symmetrical with the embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Rolled material 2 i-th stand rolling mill 3 i + 1-th stand rolling mill 4 Looper 5 Tension detection device 6 Looper height detection device 7 Looper drive electric motor 8 Rotation speed detection device 9 Control device 10 Main machine speed detection device 11 Rolling machine drive main electric motor 12 Main Engine Speed Control Device 13 Control Calculation Means 14 Control Gain Management Means 15 Setting Means 15 Control Gain Calculation Means

Claims (2)

[Claims] 1. A looper for controlling a rolling mill drive main motor and a looper drive motor of a tandem rolling mill in accordance with respective rotation speed command values calculated using a process model modeling an interference system of rolling material tension and looper height. A multivariable control device, comprising: a rolling material tension command value, a looper height command value, variables expressing the process model, and a weighting function that specifies the response to the looper height and stand tension and robustness. Substituting the setting means to be set respectively, each of the set variables and the weighting function into the control gain calculation formula which is premised on the mutual decoupling of the rolling material tension and the looper height, and the control gain is set to a numerical value. Control gain calculating means, which holds the calculated control gain, and which is suitable for the rolling condition and rolling condition. Based on the control gain management means for selecting a desired control gain, the selected control gain, and the set tension command value and looper height command value of the rolled material, and , A control calculation means for calculating the rotation speed command value of the rolling mill drive main motor and the rotation speed command value of the looper drive motor so that the looper height follows the looper height command value. Looper multi-variable controller. 2. The process model is newly added by adding a weighted value to a difference between a tension command value of the rolled material and a measured value of the rolled material tension to the set command value of the looper height. The looper multi-variable control device according to claim 1, further comprising a function of setting a looper height command value.