JPH07290126A - Device for estimating friction coefficient - Google Patents

Abstract

PURPOSE:To provide the device for accurately estimating the friction coefficient of a rolling stock during the acceleration and deceleration of the rolling stock. CONSTITUTION:An estimation on the friction coefficient of a rolling stock is made from rolling state data of the rolling stock by a first arithmetic unit 11. Storing is made by a storage device 16 for a plurality of rolling state data at the point of time prior to that of the estimation, as well as for the rolling speed and the friction coefficient corresponding to the data. A statistical processing is carried out by a third arithmetic unit 17 on the stored rolling speed and the friction coefficient; a relation is obtained between the friction coefficient and the rolling speed; and the friction coefficient is determined by such relation and the measured rolling speed. Either the friction coefficient thus determined or the estimated friction coefficient is selected by a switching device 12 and outputted to a thickness control device. Storing is made by a storing device 16 for the friction coefficient thus selected as well as the rolling state data and rolling speed corresponding to the coefficient; and, in the processing at the point of time after that of the estimation, a statistical processing is carried out by the third arithmetic unit 17 on the selected friction coefficient including the rolling speed corresponding to such coefficient.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is necessary for a plate thickness control device or the like for controlling the plate thickness of a material to be rolled in consideration of a friction coefficient between the material to be rolled and a rolling roll of a rolling mill. The present invention relates to a friction coefficient estimation device that estimates a friction coefficient, and more particularly, to improvement in estimation accuracy of a friction coefficient.

[0002]

2. Description of the Related Art A method of rolling a material to be rolled (metal material) by a rolling mill includes a method of pressurizing the material to be rolled by a rolling mill,
There are a method of controlling the tension at the time of rolling the material to be rolled by a rolling mill and a method of using both methods together. The rolling control device that controls the rolling mill controls the rolling mill in consideration of the friction coefficient of the material to be rolled to control the plate thickness, the advanced rate, and the like of the material to be rolled.

One of the most important problems in terms of quality when rolling a material to be rolled by a rolling mill is to control the plate thickness of the material to be rolled at the time of accelerating or decelerating at the start or end of rolling. There is a deterioration in accuracy.

As a countermeasure against this, Japanese Examined Patent Publication No. Sho 54-22782
There is a technique described in the publication. In this technique, as a measure against the deterioration of the plate thickness control accuracy during acceleration / deceleration, two work rolls that focus on the change in the rolling speed and press the material to be rolled from above and below according to the change. A technique for adjusting the clearance (work roll opening) is disclosed.

Further, as another measure, Japanese Patent Laid-Open No. 61-22
There is a technique described in Japanese Patent No. 2625. This applies the law of constant mass velocity to improve the plate thickness control accuracy.

[0006]

[Problems to be Solved by the Invention] Japanese Patent Publication No. 54-2278
The technique described in Japanese Patent Publication No. 2 is very effective when the predicted clearance correction amount is appropriate, but depending on the rolling condition,
The effect was not fully obtained.

It is considered that this is because the friction coefficient fluctuates extremely. That is, the friction coefficient between the material to be rolled and the rolling roll of the rolling mill changes according to the change in speed, and the rolling load applied to the material to be rolled changes with the change in the friction coefficient. it is conceivable that.

It is known that the amount of change in the friction coefficient due to the change in speed during the rolling process is as large as the amount of change in the friction coefficient due to the change in the amount and composition of the lubricant used during rolling. .

Therefore, it is necessary to estimate the fluctuating friction coefficient as accurately as possible during acceleration / deceleration and to control the plate thickness in consideration of the estimated friction coefficient in order to improve the plate thickness control accuracy. It is believed that there is.

On the other hand, according to the technique disclosed in Japanese Patent Laid-Open No. 61-222,625, the control considering the friction coefficient is not performed, but the plate thickness control can be performed with high accuracy. However, in this technology, it is necessary to detect the speed of the material to be rolled rolled at high speed with high accuracy in a non-contact manner. Therefore, a non-contact speed detecting device is required, which further improves the economical aspect. There is a need.

An object of the present invention is to provide a friction coefficient estimating device which can accurately estimate the friction coefficient during acceleration / deceleration of a material to be rolled at a lower cost.

[0012]

In order to solve the above problems, the present invention provides a rolling means for rolling a material to be rolled and a rolling method for controlling the rolling means in consideration of the friction coefficient of the material to be rolled. A friction coefficient estimating device which is used in a rolling mill having a control device, estimates a friction coefficient, and outputs the estimated friction coefficient to the rolling control device. Means, estimating means for estimating the friction coefficient from the rolling state information about the state of the material to be rolled at the time of rolling, a plurality of the rolling state information at a time point before the estimation time point by the estimating means, and at the previous time point A storage unit that stores a rolling speed and a coefficient of friction corresponding to each of a plurality of rolling state information, and statistical processing is performed on the stored rolling speed and the coefficient of friction, whereby the friction Obtained relation between the number and the rolling speed,
Statistical processing means for obtaining a friction coefficient from the obtained relationship and the measured rolling speed, a friction coefficient obtained by the statistical processing means, and one of the friction coefficients estimated by the estimating means is selected. , And a selection means for outputting to the rolling control device, wherein the storage means also accumulates the selected friction coefficient, rolling state information and rolling speed corresponding to the selected friction coefficient, and the statistics. The processing means is to perform the statistical processing including the selected friction coefficient and the rolling speed corresponding to the selected friction coefficient in the statistical processing after the estimation time by the estimation means. It is a thing.

[0013]

Since the present invention is constructed as described above,
The storage means can sequentially store the selected friction coefficient, rolling state information and rolling speed corresponding to the selected friction coefficient. Then, the statistical processing means performs statistical processing on the rolling speed and friction coefficient that are sequentially accumulated, determine the relationship between the friction coefficient and the rolling speed, from the relationship and the measured rolling speed,
Obtain the latest friction coefficient. Next, since there is a case where the estimated friction coefficient and the statistically obtained friction coefficient do not match, selection is made by the selecting means and the friction coefficient to be sent to the plate thickness control device is determined.

As a result, it is possible to accurately estimate an appropriate friction coefficient according to the rolling speed even during acceleration / deceleration. Since the friction coefficient can be estimated with high accuracy, automatic plate thickness control devices, etc.
In the control device using the friction coefficient as its control parameter, there is an effect that the control accuracy can be stabilized and enhanced. Here, stable means that the plate thickness does not suddenly change, and sophistication means that the amplitude of the plate thickness control accuracy when it is stable is small.

Since the non-contact speed detecting device is unnecessary, the cost can be further reduced.

[0016]

EXAMPLE An example of the present invention will be described below.

The outline of the method of determining the friction coefficient in this embodiment will be described first.

1) First, the actual rolling signal (rolling state information)
Are sequentially collected by measurement or the like. The actual rolling signals include rolling load P, rolling mill entrance side plate thickness H, rolling mill exit side plate thickness h, work roll radius R, rolling mill entrance side tension tb, and rolling mill exit side tension t.
f, there is a deformation resistance km. Of these, rolling load P, rolling mill entrance side plate thickness H, rolling mill exit side plate thickness h, work roll radius R,
For the rolling mill entrance side tension tb and the rolling mill exit side tension tf,
Measured during rolling. For the deformation resistance km, a known functional relationship established between the deformation resistance and the tensile strength is used,
The tensile strength is obtained by a sample test for each lot, and the deformation resistance km is obtained from a known functional relationship.

Next, the friction coefficient μ is obtained from the following equation that holds for the friction coefficient μ.

[0020]

## EQU00001 ## P = f (H, h, R, tf, tb, km,
μ). ． ． (1) μ: Friction coefficient The function f is specifically as follows.

[0021]

[Equation 2]

B: Rolling mill outlet side plate width where:

[0023]

[Equation 3]

C: constant

[0025]

[Equation 4]

[0026]

[Equation 5]

[0027]

[Equation 6]

[0028]

Instead of these equations, other equations may be used depending on the material of the material to be rolled and the rolling method.

Therefore, if the actual rolling signal (rolling amounts) excluding the friction coefficient is determined as the input signal, the friction coefficient can be theoretically calculated (estimated) from the equation (1).

As described above, the friction coefficient changes with the change of the rolling speed, so that the rolling quantities are collected at minute time intervals such as 100 milliseconds, and the rolling coefficient changes momentarily. Estimate the friction coefficient corresponding to the speed. The relationship between the rolling speed and the coefficient of friction can be expressed as shown in FIG. 2 when the rolling mill exit side plate thickness h, the work roll radius R, and the rolling mill exit side tensions tf and tb are expressed as parameters (these are the same). When the speed is low, the coefficient of friction is large, and as the speed is higher, the coefficient of friction is smaller. In FIG. 2, the horizontal axis V indicates the rolling speed, and the horizontal axis μ indicates the friction coefficient. Even if the above parameters are the same, the calculated friction coefficient has variations as shown in FIG. 2 even for the same rolling speed due to measurement errors and the like. Therefore,
The friction coefficient is statistically processed in the entire speed range, for example, using a regression analysis method. As the regression analysis, for example, regression analysis is performed using an exponential function.

2) The friction coefficient corresponding to the rolling speed obtained in the above item 1) is replaced with the friction coefficient used at that time to obtain a more optimal rolling state. In this replacement, of course, no match can be expected with the friction coefficient up to that point, so smoothing is performed in time series when replacing the old numerical value with the new numerical value. That is, instead of suddenly changing from the old number to the new number, change gradually. As a method of changing, the following expression 7 is used. In this way, the friction coefficient registered in the plate thickness control device can be smoothly replaced with a new friction coefficient. In most rolling, the process up to this stage enables stable plate thickness operation.

3) As a rare case, even if the parameters in FIG. 2 are the same, that is, similar rolling conditions are satisfied, the estimated friction coefficient does not match the previously estimated friction coefficient. To do. This is often
This is because the deformation resistance km used in Equation 1 varies. For example, since the deformation resistance km depends on the material, the deformation resistance km may change due to a slight change of the material even in the same lot.

Deformation resistance km used in Equation 1
Is obtained by performing a tensile test using a sample cut out from the lot, so it may deviate from the actual value. Therefore, regarding the friction coefficient, it is necessary to take this effect into consideration, and specifically, a regression analysis of a plurality of rolling speed vs. friction coefficient graphs is performed to evaluate as an average friction coefficient. The replacement procedure with the existing friction coefficient in this case is the same as that described above. It should be noted that prior to the replacement, the numerical data is displayed, and the execution / cancellation of the replacement can be selected by the judgment of the engineer.

Next, the configuration and operation of this embodiment will be described.

FIG. 1 shows the structure of the rolling apparatus according to this embodiment. A metal rolling machine 1 to be controlled is a four-high rolling machine consisting of a pair of work rolls 1a and a pair of reinforcing rolls 1b. Is.

The present rolling mill comprises a rolling mill 1 and a work roll 1
a, a reinforcing roll 1b, a speed generator 1c, a rolling load meter 1d, a left reel 2a, a right reel 2b, a left thickness meter 3a, a right thickness meter 3b, a left tensiometer 4a, and a right tension Total 4
b, the deformation resistance setting device 10, the first computing device 11, the switching device 12, the initial value setting device 13, and the second computing device 14
And a storage device 16, a third computing unit 17, and an automatic plate thickness control device 18.

Each time the rolled material 5 passes between the work rolls 1a from the left reel 2a to the right reel 2b or from the right reel 2b to the left reel 2a (one pass is counted as one pass). It is thinned by the reduction force by the machine or the tension between the rolling mill and the reel. A predetermined thickness is obtained by multiple passes. Left reel 2a to right reel 2b
If sent to, the left side will be the entrance side. In the opposite case,
The right side is the entrance side.

FIG. 3 shows the state of acceleration / deceleration during rolling.
This is because the rolled material wound on the reel has a length of 2000m.
Is the case. The horizontal axis represents time and the vertical axis represents speed. There are various lengths of the material to be rolled, but it is about 1000 to 3000 m. In the case of FIG. 3, the time required for one pass is 215 seconds, the acceleration period is 20 seconds, and the deceleration period is 10 seconds. The length of the rolled material that passes through the rolls during the acceleration period is 100 m,
The length of the rolled material passing through the rolls during the deceleration period is 50 m.

In this rolling process, the entrance and exit plate thicknesses H and h of the rolling mill 1 are measured by the radiation thickness gauges 3a and 3b. The radiation thickness gauges 3a and 3b are the rolled material 5
The thickness is measured by the amount of radiation passing through. Similarly, the inlet-side and outlet-side tensions tb and tf are measured by the tensiometers 4a and 4b, and these measurement signals are obtained by the first arithmetic unit 11
Entered in. The tensiometers 4a and 4b measure the tension by pushing the rolled material 5 with the tensiometers 4a and 4b and measuring the repulsive force at that time.

The first calculator 11 further receives the rolling load P from the rolling load meter 1d, the rolling speed at that time from the speed generator 1c, and the deformation resistance stored in the deformation resistance setting unit 10. The value is entered.

After that, the first arithmetic unit 11 receives an interrupt signal from a timer (not shown) or the like, and calculates the friction coefficient μ corresponding to the rolling speed by the equation 1 every moment. Here, the calculated friction coefficient μ is sent to the second computing unit 14 via the switching unit 12.

The second computing unit 14 includes the initial value setting unit 13
A reference friction coefficient value group 14a before starting the estimation from is given. The numerical value stored in the initial value setting device 13 is obtained by statistical processing when the operation is started after the operation is stopped or when the operation is just started and the amount of data is small. This is the case when the numerical values are not stable. Then, the output of the first computing unit 11 and the friction coefficient value group 14a are displayed side by side on a display device (not shown), and both are replaced if necessary by a rolling engineer's instruction 14b.

Which one is selected depends on the judgment of the engineer. The selection may be based on the factual fact that the rolling state is unstable, or may be a value empirically determined to be close to a value considered appropriate by the engineer. If there is no instruction from a technician, it depends on the initial value setting of the apparatus. Therefore, when changing the initial value, the desired value is switched.
In order to prevent the rolling condition from being disturbed, the selection is not performed during rolling, but is performed at the time of switching lots or between passes.

At the time of replacement, smoothing processing is executed in order to suppress a sudden change in the friction coefficient. The smoothing process is, for example, a known exponential smoothing method, that is,

[0046]

## EQU00007 ## y (i) =. Alpha..times.y (i-1) + (1
−α) × x (i).

Here, α represents a smoothing coefficient (0 <α <1), x (i) is an input signal to the smoother (second arithmetic unit 14), and y (i) is an output signal of the smoother. Is. Subscript (i)
Represents the calculation timing, and (i-1) represents the previous calculation timing.

In this way, the smoothed friction coefficient 1
5 is sent to a control device such as the automatic plate thickness control device 18.

By the way, it is usual that a large amount of material to be rolled is produced under the same type of rolling conditions (the parameters in FIG. 2 are the same). At that time, comparison with the friction coefficient calculated in the past is made. The signal 15 is supplied to the storage device 16 in order to be able to do so.

The storage device 16 stores, in addition to the friction coefficient 15 obtained at each calculation timing, the friction coefficient obtained in advance by experiments, the friction coefficient obtained for other rolled materials, and the like. The storage device 16 retrieves and extracts the friction coefficient in the same kind of rolling state (such as the friction coefficient already obtained by calculation), and transmits it to the third computing unit 17. Here, the same kind of rolling condition means that the signals input to the first computing unit 11 are within a certain range, and the roll surface properties (surface roughness) of the work roll 1a and the like, which will be described later, are extremely similar. Means there is. Particularly, in the input signal to the first arithmetic unit 11, the deformation resistance value given from the deformation resistance setting unit 10 is
Since it changes considerably depending on the processing rate of the material 5 to be rolled, it is necessary to consider such as applying the same pass signal of the same rolling schedule. The same reduction schedule means the first pass
For example, rolling from 3.5 mm to 3.0, rolling from 3.0 to 2.7 in the second pass, and 2.7 in the third pass.
It means that the plate thickness is the same for all of the rolling schedules of rolling to 2.5.

The third computing unit 17 performs a regression analysis of these numerical groups. That is, specifically, the calculated friction coefficient in the same rolling state with respect to the rolling speed is the first (first lot)
From the nth to nth lot (nth lot), 1
Regression analysis is performed on the data from the first line to the nth line to obtain one curve by the regression analysis. This is shown in FIG.

In FIG. 2, it is known that when n becomes large, the wear of the work roll 1a and the properties of the roll surface change, and as a result, the friction coefficient is affected, so generally n <10 is applied. To do. From this viewpoint, the distinction of the roll surface roughness is also a discriminating factor for whether or not the rolling is of the same type. The roll surface roughness is an amount indicating how much is so-called dull (a rough surface) or bright (a smooth surface).

In FIG. 2, although there are many plot points,
It is known that the variation generally has a certain tendency, that is, an exponential distribution state as shown in FIG.

Therefore, each plot point in FIG. 2 is subjected to regression analysis (for example, exponential regression analysis) to obtain a function yμ representing a friction coefficient.

The regression function yμ, which is the output of the third computing unit 17, is displayed side by side with the output of the first computing unit 11 on a display device (not shown). One of them is selectively adopted by the switching device 12, and becomes a source signal of the actual rolling control friction coefficient 15.

Which one is selected depends on the judgment of the engineer. The selection may be based on the factual fact that the rolling state is unstable, or a value that is empirically determined to be close to a value considered appropriate by the engineer may be selected. If there is no instruction from a technician, it depends on the initial value setting of the apparatus. Therefore, when changing the initial value, the desired value is switched. In order to prevent the rolling condition from being disturbed, the selection is not performed during rolling, but is performed at the time of switching lots or between passes.

In this way, the coefficient of friction is not only calculated by the equation 1 (calculation data), but the similar rolling data is subjected to regression analysis to actually filter (smooth) the calculation data. Therefore, stable numerical data are provided without misusing the protruding singular value, and as a result, stable rolling is possible.

In this example, the constituent elements such as the arithmetic unit are independent devices, but the functions of these plural constituent elements may be realized by one processing device such as a control computer. Good.

Further, regarding the substitution of the friction coefficient value group in the second computing unit 14 and the substitution in the switching unit 12, a case where a rolling engineer gives an instruction is illustrated, but the automatic substitution is performed under a specific substitution condition. It is also possible to let. For example, in the case of the switching device 12, the friction coefficient (output of the third calculator 17) obtained by regression analysis is normally used for all speed regions, but the friction coefficient obtained from the equation 1 (first calculator 11) is used. Output) falls within the range of ± 10% of the friction coefficient obtained by the regression analysis, it is possible to adopt a condition such as substitution.

The friction coefficient obtained by the regression analysis is normally used because there are few mistakes because it is averaged. Further, for example, the reason why the substitution is made in the case of falling within the range of ± 10% is that the friction coefficient according to Formula 1 is considered to be the most suitable value for the rolling state at that time if there is no error in the measurement. If it does not fall within the range of ± 10%, an error in measurement or the like is considered, and therefore replacement is not performed.

As described above, according to the present embodiment, the non-contact speed detecting device as in the prior art is not required, and the friction coefficient is accurately estimated during rolling, and the value is sequentially applied to be appropriate. The plate thickness control amount can be determined. As a result, the accuracy of controlling the plate thickness during acceleration / deceleration is improved.

[0062]

According to the present invention, since the friction coefficient can be estimated with high accuracy in real time, it is possible to stabilize and improve the control accuracy in a control device such as an automatic plate thickness control device which uses the friction coefficient as its control parameter. There is an effect that it can be realized.

[Brief description of drawings]

FIG. 1 is a block diagram of a rolling apparatus according to the present invention.

FIG. 2 is a graph showing an example of actual measurement of rolling speed and friction coefficient in the present invention and a regression analysis result thereof.

FIG. 3 is an explanatory diagram showing changes in rolling speed in time series during rolling.

[Explanation of symbols]

1 ... Rolling mill, 1a ... Work roll, 1b ... Reinforcing roll, 1
c ... Speed generator, 1d ... Rolling load meter, 2a, 2b ... Left and right reels, 3a, 3b ... Left and right thickness gauge, 4a, 4b ... Left and right tension meter, 5 ... Rolled material, 10 ... Deformation resistance setting device, 11 ... First computing unit, 12 ... Switching unit, 12a ... Instruction of rolling engineer, 13 ... Initial value setting unit, 14 ... Second computing unit, 14a
... Standard friction coefficient numerical value group, 14b ... Instruction of rolling engineer, 1
5 ... Smoothed friction coefficient, 16 ... Storage device, 17 ... Third computing unit

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Masaaki Nakajima 52-1 Omika-cho, Hitachi-shi, Ibaraki Hitachi Ltd. Omika factory

Claims (7)

[Claims] 1. A rolling apparatus having a rolling means for rolling a material to be rolled and a rolling control device for controlling the rolling means in consideration of the friction coefficient of the material to be rolled, for estimating the friction coefficient. Then, in the friction coefficient estimation device that outputs the estimated friction coefficient to the rolling control device, from the speed reception means that receives the measured value of the rolling speed of the material to be rolled, and the rolling state information about the state of the material to be rolled at the time of rolling Estimating means for estimating the friction coefficient, a plurality of the rolling state information at a time point earlier than the time point estimated by the estimating means, and a rolling speed and a friction coefficient corresponding to each of the plurality of rolling state information at the previous time point And a rolling means for storing the rolling speed and the friction coefficient stored therein to obtain the relationship between the friction coefficient and the rolling speed. Of the friction coefficient obtained from the coefficient and the measured rolling speed, the friction coefficient obtained by the statistical processing means, and the friction coefficient estimated by the estimating means are selected to perform the rolling control. With a selection means for outputting to the device, the storage means also accumulates the selected friction coefficient, rolling state information and rolling speed corresponding to the selected friction coefficient, the statistical processing means, In the statistical processing at a time point after the estimation time point by the estimating means, the friction coefficient is characterized by performing the statistical processing including the selected friction coefficient and the rolling speed corresponding to the selected friction coefficient. Estimator. 2. The friction coefficient estimating device according to claim 1, wherein the rolling state information includes a rolling load applied to the material to be rolled, a thickness of the material to be rolled, and a deformation resistance of the material to be rolled. , At least included, the rolling load and the thickness of the material to be rolled are actually measured at the time of estimation, and the deformation resistance is stored in memory. . 3. The friction coefficient estimating device according to claim 1 or 2, wherein the estimating means estimates the friction coefficient from a theoretical expression representing a relationship between the rolling state information and the friction coefficient. Characteristic friction coefficient estimation device. 4. The friction coefficient estimating device according to claim 1, 2 or 3, wherein an initial value storing means for storing an initial value of the friction coefficient set before the start of the estimation, and the friction when the estimation is started. An apparatus for estimating a friction coefficient, comprising: an initial value of the coefficient; and a start-time selection means for selecting one of the selected friction coefficients. 5. The friction coefficient estimating device according to claim 4, further comprising a smoothing means for outputting an interpolated value of the two friction coefficients to be selected when the start time selecting means makes a selection. A friction coefficient estimating device, wherein the output of the smoothing means is sent to the rolling control device. 6. A rolling device having a rolling means for rolling a material to be rolled and a rolling control device for controlling the rolling means in consideration of the friction coefficient of the material to be rolled, for estimating the friction coefficient. Then, in the friction coefficient estimation device that outputs the estimated friction coefficient to the rolling control device, the friction coefficient is determined from the speed measuring means that measures the rolling speed of the material to be rolled and the rolling state information related to the state of the material to be rolled during rolling. Estimating means for estimating, the rolling state information at a time point before the estimation time point by the estimating means, rolling state information corresponding to the estimated friction coefficient, and corresponding to each of the plurality of rolling state information A storage means for storing the rolling speed and the friction coefficient, and a statistical process for the stored rolling speed and the friction coefficient are used to obtain the relationship between the friction coefficient and the rolling speed. A friction coefficient estimating device comprising: statistical processing means for obtaining a friction coefficient from the obtained relationship and the measured rolling speed. 7. The friction coefficient estimating device according to claim 1, rolling means for rolling a material to be rolled, and the rolling means in consideration of the coefficient of friction of the material to be rolled. It has a rolling control device for controlling and a speed measuring means for measuring the rolling speed of the material to be rolled, and the rolling control device controls the rolling means on the basis of the estimated friction coefficient. Rolling equipment.