JP5858265B1 - Reduction control device and reduction control method - Google Patents

Abstract

Disclosed is a roll-down control device and a control method capable of identifying a dominant phase characteristic in a control band based on a step response test of a roll-down control system and estimating a control delay time with high accuracy from the phase characteristic. SOLUTION: Using step response test means 410 for performing a step response test of a rolling control system using a fluctuation frequency band of an entry side plate thickness of a material to be rolled 200 as a control band, and the test result and a measured output dead time. , Parameter calculating means 420 for obtaining the damping coefficient and natural frequency of the reduction control system in the control band, and a parameter for obtaining the parameter of the optimum approximate solution for the phase characteristics of the reduction control system using the damping coefficient, natural frequency and output dead time Computation means 430 and control delay time computation means 440 for obtaining the control delay time of the reduction control system using the parameters of the optimum approximate solution and the output dead time are provided, and the reduction control system is operated based on the control delay time. . [Selection] Figure 6

Description

  The present invention relates to a reduction control device and a reduction control method for controlling the thickness of a material to be rolled in a rolling mill, and more particularly to a feed forward automatic thickness control (FFAGC) technology.

The FFAGC controls the outlet side thickness of the material to be rolled to a target value by the roll control device controlling the roll opening degree according to the detected value of the inlet side thickness of the material to be rolled. Here, a localization control system that controls the oil column position of a hydraulic cylinder for driving a rolling roll to a predetermined value is widely used in the reduction control device.
If there is no response delay or dead time in this localization control system, it is possible to match the delivery side plate thickness of the material to be rolled to the target value, but generally there is a control delay time including dead time in the localization control system. To do. Therefore, in order to realize accurate sheet thickness control by FFAGC, it is necessary to accurately estimate the control delay time and reflect it in actual operation.

  In view of the above points, for example, Patent Document 1 inputs time series data of a reduction position feedback value having a time delay into a neural network learned using time series data of a reduction position command value. A plate thickness control device (first prior art) is described in which a delay time of a reduction position feedback value with respect to a reduction position command value is calculated based on the similarity between output time-series data.

  Further, in Patent Document 2, for the centralized controller that simultaneously performs the rolling position control, looper torque control, and mill speed control, the exit side plate thickness of the rolling mill, the looper angle, the deviation of the tension between the stands, the set value, and the like are variables. The state predictor is configured by the state equation, the control system based on the delay time and the control operation amount is obtained by using the state predictive control theory by obtaining the delay time of the operation path of the control system and the signal of the detection path. Describes a rolling mill control method (second prior art) for obtaining a control operation amount that compensates for the delay time by estimating the internal state of the rolling mill.

Japanese Patent No. 3266682 (paragraphs [0003] to [0008], FIG. 1 etc.) Japanese Patent Laid-Open No. 11-90518 (paragraphs [0012] to [0040], FIG. 2, FIG. 3, etc.)

However, in the first conventional technique, a large amount of calculation processing and time are required for learning a neural network and acquiring a large number of similarity data used for calculating a delay time.
In addition, since the second prior art is intended for a centralized controller, a state predictor must be configured using a state equation having many types of variables, and a high-speed and expensive arithmetic processing device is required. There was a problem of becoming.

  Therefore, the problem to be solved by the present invention is to identify the dominant phase characteristic in the control band as the fluctuation frequency band of the inlet side plate thickness based on the step response test of the reduction control system that can be easily performed, and from the phase characteristic An object of the present invention is to provide a roll-down control device and a roll-down control method in which the control delay time is estimated and reflected in the localization control.

In order to solve the above-described problem, the reduction control device according to claim 1 is configured to target the outlet side thickness of the material to be rolled by the reduction control system adjusting the rolling roll opening according to the inlet side thickness of the material to be rolled. A rolling control device that matches a value, and includes a secondary vibration system in which a transfer function representing a dynamic characteristic from a rolling roll opening command to a rolling roll opening in the rolling control system has an output dead time. In
A test means for performing a step response test of the rolling-down control system using the fluctuation frequency band of the inlet side plate thickness as a control band;
By using the result and the measured the output dead time of the step response test, the first parameter calculating means for calculating the damping coefficient and natural frequency of the pressure control system in the control band,
Second parameter calculating means for obtaining a parameter of an optimal approximate solution for the phase characteristic of the reduction control system using the damping coefficient and natural frequency and the output dead time;
Using the parameter of the optimum approximate solution and the output dead time, control delay time calculating means for obtaining a control delay time of the reduction control system,
The reduction control system is operated based on the control delay time.

Further, in the reduction control method according to claim 2, the reduction control system adjusts the rolling roll opening according to the entry side plate thickness of the material to be rolled, so that the output side plate thickness of the material to be rolled matches the target value. In the rolling control method , the transfer function representing the dynamic characteristics from the rolling roll opening command to the rolling roll opening in the rolling control system includes a secondary vibration system having an output dead time .
A first step of performing a step response test of the rolling-down control system using the fluctuation frequency band of the entry side plate thickness as a control band;
By using the result and the measured the output dead time of the step response test, a second step of obtaining the attenuation factor and natural frequency of the pressure control system in the control band,
A third step of obtaining a parameter of an optimum approximate solution for the phase characteristic of the reduction control system using the damping coefficient and natural frequency and the output dead time;
Using a parameter of the optimal approximate solution and the output dead time to obtain a control delay time of the reduction control system,
The reduction control system is operated based on the control delay time.

According to the present invention, parameters necessary for FFAGC are automatically determined by arithmetic processing means such as a PLC (programmable logic controller) based on the step response of the rolling control system, and the entry side thickness of the material to be rolled is determined. By estimating the control delay time adapted to the fluctuating frequency band and reflecting it in the localization control, it is possible to control the plate thickness while suppressing the variation in the outlet side plate thickness.
In addition, since a procedure for acquiring and learning a large number of data is unnecessary, the burden of the arithmetic processing is small, and it can be realized by a general-purpose arithmetic processing device.

It is a block diagram of the reduction control system in the embodiment of the present invention. It is the block diagram which deform | transformed FIG. It is a wave form diagram which shows the time response of a secondary vibration system. FIG. 20 is an explanatory diagram for deriving Formula 15. It is a wave form diagram which shows the time response which considered the output dead time. It is a schematic block diagram of the rolling-down control apparatus which concerns on embodiment of this invention. It is a figure which shows the entrance side deviation (FIG. 7 (a)) and exit side deviation (FIG. 7 (b), (c)) of board thickness for demonstrating the effect by embodiment of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
First, FIG. 1 is a block diagram of a reduction control system (localization control system) in this embodiment. In FIG. 1, R is a roll opening command of a rolling roll, Y is a roll opening, 10 is an adjustment term, 20 is a servo valve, 30 is a hydraulic cylinder that adjusts the roll opening when oil is supplied from the servo valve 20, Reference numeral 40 denotes a dead time element caused by an operation delay of the servo valve 20 or the hydraulic cylinder 30 or the like.

Here, when the dynamic characteristic from the roll opening degree command R to the roll opening degree Y is expressed by the transfer function G (s), the following formula 1 is obtained.

In Equation 1, the coefficient K of the adjustment term 10 is
K = K ₀ × K _s × K _c
(K ₀ : adjustment coefficient, K _s : coefficient for converting the opening degree of the servo valve 20 into oil flow rate, K _c : coefficient for converting the oil flow rate into piston movement amount of the hydraulic cylinder 30) , T _s is a time constant of the servo valve 20, T _{d is a} dead time, and s is a Laplace operator.

The numerator and denominator _Td in Equation 1 can be considered as output dead time and dead time in internal delay, respectively.
The transfer function G (s) of Expression 1 is transformed as Expression 2 using the damping coefficient ζ (0 <ζ <1) and the natural frequency ω as parameters.

Equation 2 represents Equation 1 as a series combination of a secondary vibration system having an output dead time _Td and multiplicative uncertainty.
The above formula 2 is represented by a block diagram as shown in FIG. 2. 40 is the dead time element, 50 is the secondary vibration system, and 60 is an internal delay element corresponding to Δ in formula 2.

  The internal delay element Δ in FIG. 2 has a remarkable phase lag in a high frequency band than the control band of the reduction control system (the fluctuation frequency band of the entry side plate thickness of the rolling mill). On the other hand, among the dominant characteristics in the control band indicated by reference numeral 70, the parameters constituting the secondary vibration system 50 can be identified by a step response test with respect to the reduction control system. Therefore, here, a procedure for identifying the parameters (damping coefficient ζ and natural frequency ω) constituting the secondary vibration system 50 by excluding the internal delay element Δ and considering the step response of the reduction control system will be described.

数式３のステップ入力が与えられた二次振動系の応答Ｙ（ｓ）は、数式４となる。

Now, the transfer function U (s) of the step input having the magnitude R is expressed by Equation 3.

The response Y (s) of the secondary vibration system given the step input of Equation 3 is expressed by Equation 4.

Further, the time response y (t) is expressed by Equation 5.

Here, FIG. 3 is a waveform diagram showing the time response y (t) of the secondary vibration system, in which R is a step response target value (corresponding to the roll opening command R), and R _m is a step response. The maximum value, t _m, is the maximum overshoot time. As will be described later, R _m / R is referred to as an overshoot rate R _r .
From FIG. 3, the following formulas 6 and 7 are obtained.

When t = t _m , the time response y (t) is expressed by Equation 8. Further, when the overshoot rate R _r is obtained by dividing Equation 8 by R, Equation 9 is obtained.

Next, the time response y (t) is differentiated to obtain Equation 10.

Further, Expression 11 and Expression 12 are obtained from Expression 10.

Here, the mathematical formula 14 is derived based on the mathematical formula 12 and the mathematical formula 13 written in the mathematical formula 5, and as apparent from FIG. 4, tan ψ = tan (π + ψ), so that the mathematical formula 15 is obtained.

From Equation 9 and Equation 14 described above, the overshoot rate R _r becomes Equation 16, and Equation 17 and Equation 18 are obtained from Equation 16.

Accordingly, the damping coefficient ζ and the natural frequency ω are expressed by Expression 19 and Expression 20, respectively.

Here, the output dead time _Td shown in FIG. 5 can be measured, and when Equation 20 is transformed in consideration of this output dead time _Td , Equation 21 is obtained.

数式２２により表される位相特性ψ（ｆ）を、数式２３のように一次関数にて表す。

Next, the phase characteristic ψ (f) of the reduction control system is expressed by Equation 22 using the damping coefficient ζ and the natural frequency ω identified by the above step response and the measurable output dead time T _d. It is expressed in

The phase characteristic ψ (f) expressed by Expression 22 is expressed by a linear function as Expression 23.

Here, the parameter α in Expression 23 minimizes the evaluation expression shown in Expression 24, and is a constant that satisfies Expression 25. In Equation 24, f _cmin and f _cmax are the minimum frequency and the maximum frequency of the assumed control band.

Therefore, the parameter α is expressed by Equation 26.

ただし、正数ｎは設計パラメータであり、更に数式２８を条件とする。

Formula 26 can be easily solved in a discrete form such as Formula 27 by using PLC, for example.

However, the positive number n is a design parameter and is further conditional on Equation 28.

The control delay time T _cd to be set in the control device can be expressed as Equation 29 using the phase characteristic ψ (f) and the measured output dead time T _d .

数式３０によって表される制御遅れ時間Ｔ_ｃｄは定数となるため、この実施形態によれば、圧延機の入側板厚の変動周波数に応じて制御遅れ時間Ｔ_ｃｄを逐次計算する必要がなくなるという利点がある。 Further, when Expression 23, which is an approximate expression for phase delay, is substituted into Expression 29, Expression 30 is obtained.

Since the control delay time T _cd represented by Expression 30 is a constant, according to this embodiment, it is not necessary to sequentially calculate the control delay time T _cd according to the fluctuation frequency of the entry side plate thickness of the rolling mill. There is.

As described above, an appropriate control delay time T _cd for performing highly accurate reduction control in FFAGC can be calculated by the following procedures 1 to 3.
[Procedure 1]: Using the result of the step response test and the output dead time _Td , the damping coefficient ζ and the natural frequency ω, which are parameters of the rolling-down control system, are obtained by Equations 19 and 21.
[Procedure 2]: Using the ζ and ω obtained in [Procedure 1], the parameter α of the optimum approximate solution ψ _op (f) for the phase characteristic ψ (f) of the reduction control system is obtained by Equation 27.
[Procedure 3]: The control delay time T _cd is calculated by Equation 30 using the parameter α obtained in [Procedure 2] and the output dead time T _d .

The above steps 1 to 3 can be easily implemented as a PLC program, and the calculation of the control delay time required for operation can be automated.
In particular, the control delay time can be accurately and easily identified as compared with the method of reading and determining the analog chart output of the reduction control system.

FIG. 6 is a schematic configuration diagram of the rolling control device according to the embodiment of the present invention.
In FIG. 6, 110 is a rolling control unit of the rolling mill, 120 is a rolling mechanism provided with a servo valve, a hydraulic cylinder, etc., 130 and 140 are plate thickness sensors on the entry side and the exit side, 200 is a material to be rolled, and 300 is a material to be rolled. It is a rolling roller. Here, in the case of configuring a control system only of FFAGC, a feedback signal from the plate thickness sensor 140 on the output side is not indispensable.

Also, 410 is a test means for performing a step response test of the reduction control system, and 420 is a damping coefficient ζ and a natural frequency according to Equations 19 and 21, using the result of the step response test and the measured output dead time _Td. The first parameter calculating means 430 for obtaining ω uses the damping coefficient ζ, the natural frequency ω, and the output dead time T _d, and the optimum approximate solution ψ _op for the phase characteristic of the reduction control system using Equation 22 and Equation 27. Second parameter calculation means 440 for determining the parameter α in (f) is control delay time calculation means 440 for calculating the control delay time T _cd using Equation 30 using the parameter α and the output dead time T _d . Each of these means 410, 420, 430, 440 is realized by hardware such as PLC and a program mounted thereon.

The reduction control unit 110 can reduce the deviation from the target value of the plate thickness due to the control delay by driving the reduction mechanism 120 in accordance with the command in consideration of the control delay time _Tcd .
Here, a specific method for controlling the reduction control unit 110 and the reduction mechanism 120 in consideration of the control delay time T _cd is as follows.
First, assuming that the distance from the entry side thickness sensor 130 to the rolling position by the rolling roller 300 is L and the speed of the material 200 to be rolled is V, the thickness portion measured by the thickness sensor 130 at a certain time T ₀ is time _{T 1} to reach the rolling position _is a _{T 1 = T 0 + L /} V. That is, when the time (L / V) has elapsed after the plate thickness is measured by the plate thickness sensor 130, the roll opening needs to be set to a desired target value.
The control delay time T _cd calculated according to the embodiment of the present invention is the time from when the command is given to the reduction control unit 110 until the actual roll opening reaches the target value (output of the reduction control unit 110). Therefore, in order to make the roll opening reach the target value at the time T ₁ described above, a command may be given to the reduction control unit 110 at the time (T ₁ -T _cd ).

Moreover, FIG. 7 is a figure which shows the board | substrate thickness entrance side deviation (FIG. 7 (a)) and exit side deviation (FIG. 7 (b), (c)) for demonstrating the effect of embodiment of this invention. FIG. 7B shows an outgoing side deviation when an appropriate control delay time is set by applying the present invention, and FIG. 7C shows an outgoing side deviation when the control delay time is not appropriate.
As is clear from these figures, when the control delay time T _cd is not set appropriately, the variation in the inlet side plate thickness cannot be sufficiently removed, but according to the present invention, the outlet side deviation is reduced. The plate thickness can be controlled with high accuracy to the target value.

  The present invention can be used for various FFAGC regardless of cold rolling or hot rolling.

10: Adjustment term 20: Servo valve 30: Hydraulic cylinder 40: Dead time element 50: Secondary vibration system 60: Internal delay element 70: Dominant characteristic in control band 110: Reduction control unit 120: Reduction mechanism 130, 140: Plate Thickness sensor 200: material to be rolled 300: rolling roller 410: step response test means 420: first parameter calculation means 430: second parameter calculation means 440: control delay time calculation means

Claims (2)

A rolling control device that adjusts a rolling roll opening degree according to an entry side plate thickness of a material to be rolled to adjust a delivery side plate thickness of the material to be rolled to a target value, and rolling in the rolling control system In the rolling control device including a secondary vibration system in which the transfer function representing the dynamic characteristics from the roll opening command to the rolling roll opening has an output dead time ,
A test means for performing a step response test of the rolling-down control system using the fluctuation frequency band of the inlet side plate thickness as a control band;
By using the result and the measured the output dead time of the step response test, the first parameter calculating means for calculating the damping coefficient and natural frequency of the pressure control system in the control band,
Second parameter calculating means for obtaining a parameter of an optimal approximate solution for the phase characteristic of the reduction control system using the damping coefficient and natural frequency and the output dead time;
Using the parameter of the optimum approximate solution and the output dead time, control delay time calculating means for obtaining a control delay time of the reduction control system,
A roll-down control device that operates the roll-down control system based on the control delay time. A rolling control method in which a rolling control system adjusts a rolling roll opening degree according to an inlet side plate thickness of a material to be rolled to match an outlet side plate thickness of the material to be rolled with a target value, and rolling in the rolling control system In the rolling control method including a secondary vibration system in which the transfer function representing the dynamic characteristics from the roll opening command to the rolling roll opening has an output dead time ,
A first step of performing a step response test of the rolling-down control system using the fluctuation frequency band of the entry side plate thickness as a control band;
By using the result and the measured the output dead time of the step response test, a second step of obtaining the attenuation factor and natural frequency of the pressure control system in the control band,
A third step of obtaining a parameter of an optimum approximate solution for the phase characteristic of the reduction control system using the damping coefficient and natural frequency and the output dead time;
Using a parameter of the optimal approximate solution and the output dead time to obtain a control delay time of the reduction control system,
A rolling-down control method, wherein the rolling-down control system is operated based on the control delay time.

Images