JPH0957315A - Method and device for estimating width of hot rolled steel strip - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and device for estimating the width of a hot rolled steel strip for drastically reducing the number of generation of cold rolled steel strips having abnormal width without reduction of width yield. SOLUTION: A device 7 for estimating the width of the hot rolled steel strip is provided with a thickness detecting means 8, tension detecting means 9, temp. detecting means 10, width detecting means 11, tension setting means 16, tension deviation detecting means 13 for detecting the tension deviation between a detected tension and a set tension, discriminating means 14 for discriminating between the dimension of the tension deviation and that of predetermined value, carbon equivalent calculating means 15 and weight detecting means 17, and samples the output of each detecting means. And, by determining the entire length of the hot rolled steel strip 1, virtually dividing the hot rolled steel strip 1 equally in the longitudinal direction, processing the sampled data in every virtually equally divided region and substituting the processed data for a preset multiplex regression equation, the width of the hot rolled steel strip at the normal temp. in every divided region is determined. The device is constituted including arithmetic means 12 for calculating the min. width of the hot rolled steel strip from determined widths.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for predicting the strip width of a hot rolled steel strip, which predicts the strip width of a hot rolled steel strip rolled by a hot rolling mill at room temperature from the detected value during hot rolling. Regarding

[0002]

2. Description of the Related Art Conventionally, cold-rolled steel strip is made of low carbon steel or the like, and is subjected to hot rolling, pickling, cold rolling, surface cleaning, annealing, and temper rolling in this order. Is being manufactured. The quality of cold-rolled steel strips has been controlled by many control items, including dimensions, shapes, surface defects, and
Consistent quality control is performed for each item such as material. For example, regarding the strip width of the cold-rolled steel strip, which is one of the most basic quality control items, quality control is consistently performed from the strip width of the hot-rolled steel strip that is the material.

The quality design of the strip width of the hot rolled steel strip is performed as follows. (1) The quality design value of the plate width, which is the minimum necessary plate width in the next process (hereinafter sometimes abbreviated as the quality design value), is the manufacturing specification plate width of the next process, and the necessary margin allowance (for example, 5 mm). It is set by adding. (2) A board width quality design target value at room temperature (hereinafter sometimes abbreviated as a board width target value) is set by adding a board width variation (for example, 5 mm) to the quality design value.

As described above, the quality design of the strip width of the hot-rolled steel strip is performed based on the rational quality design standard so that the strip width of the cold-rolled steel strip satisfies the product standard. The variation in the strip width is mainly due to the variation in the width shrinkage amount after the strip width measurement in the hot rolling step, and the width shrinkage amount is the width shrinkage amount due to the rolling tension on the downstream side of the strip width detection position. And the amount of shrinkage due to cooling. Conventionally,
The variation in the width reduction amount has been estimated based on experience and intuition.

[0005]

As described above, the conventional quality design of the strip width of the hot-rolled steel strip is performed based on the rational quality design standard. However, since the variation in the width reduction amount is estimated by experience and intuition,
There may be a large difference from the actual value of the variation in the width shrinkage amount, which may cause the following problems.

FIG. 7 is a characteristic diagram showing the relationship between the strip width distribution of the hot rolled steel strip at room temperature and the strip width target value, quality design value, and production specification strip width. The horizontal axis of FIG. 7 indicates the plate width based on the production specification plate width, and the vertical axis indicates the appearance frequency of the plate width. When the estimated value of the variation of the width shrinkage amount based on the experience and intuition is smaller than the actual variation of the width shrinkage amount, the strip width distribution K1 of the hot-rolled steel strip at room temperature is centered on the strip width target value W2. A normal distribution is shown, and an area A1 below the quality design value W1 exists on the side of the normal distribution where the plate width is small. The strip width of the hot-rolled steel strip included in this region A1 is smaller than the minimum required strip width, and the necessary margin allowance (W1-WO) for the strip width in the next process cannot be secured. Is an abnormal width steel strip. Further, the abnormal width hot rolled steel strip is caused mainly by the variation in the amount of width shrinkage after the sheet width is detected in the hot rolling step as described above, so it is difficult to find it in the hot rolling step, and the next step It can often be found for the first time by measuring the plate width at. Therefore, when the next step after the hot rolling step is the pickling cold rolling step in which the pickling step and the cold rolling step are continuous, it is possible that the abnormal width steel strip is first obtained after the completion of the cold rolling. As discovered, the cold rolled steel strip will not be able to be delivered as the original product. Further, when it is discovered that the steel strip has an abnormal width after the completion of the cold rolling, the issuance of the remanufacturing instruction is delayed, which may cause a delay in delivery to the customer. On the other hand, if the next step of the hot rolling step is a single pickling step, it can be found that the steel strip has an abnormal width by measuring the strip width in the pickling step. Manufacturing directives can be changed.

In the occurrence of the abnormal width steel strip, the strip width target value is changed to increase the difference between the strip width target value and the quality design value, that is, the estimated value of the variation in the width shrinkage amount based on the experience and intuition. It will be solved. In this case, the strip width distribution K2 of the hot-rolled steel strip at room temperature shows a normal distribution centered on the strip width target value W3 larger than the strip width target value W2.
The entire area of the normal distribution becomes larger than the quality design value W1.
However, in this case, since the plate width is increased as a whole, the trimming allowance at the time of passing the plate in the next process is increased, and therefore, as shown in FIG.
There is a problem that the width yield is reduced by an amount corresponding to the area A2 indicated by the diagonal line.

As described above, in the prior art, since the variation in the amount of width reduction of the hot rolled steel strip is estimated by experience and intuition, the strip width of the hot rolled steel strip at room temperature can be accurately predicted before the next step of threading. However, the problems such as the occurrence of abnormally cold-rolled steel strips with a reduced width and the reduction in width yield remain unsolved.

It is an object of the present invention to solve the above problems and to significantly reduce the number of abnormally width cold-rolled steel strips produced without lowering the width yield, and a method and apparatus for predicting the strip width of hot-rolled steel strips. Is to provide.

[0010]

DISCLOSURE OF THE INVENTION The present invention provides a hot rolled steel strip having a thickness and a hot rolled steel strip between a hot rolling mill and a winder for winding the hot rolled steel strip rolled by the hot rolling mill. The steps of detecting the width, temperature and tension over the entire length, the step of obtaining the carbon equivalent of the hot rolled steel strip, the step of detecting the tension deviation between the detected tension and the preset tension, and the tension deviation being less than the preset value When, is, by substituting the detected value of the plate thickness, plate width and temperature, and the carbon equivalent to the first multiple regression equation in which the plate thickness of the hot rolled steel strip, the plate width, the temperature and the carbon equivalent are variables. , The step of obtaining the strip width of the hot rolled steel strip at room temperature, and when the tension deviation is a predetermined value or more,
Substitute the detected value of the plate thickness, plate width, temperature and tension deviation and the carbon equivalent into a second multiple regression equation that has the plate thickness, plate width, temperature, tension deviation and carbon equivalent of the hot rolled steel strip as variables. Then, the method includes a step of obtaining a sheet width of the hot rolled steel strip at room temperature, and a step of calculating a minimum sheet width of the hot rolled steel strip from the obtained sheet width. Is a method of predicting the plate width. According to the present invention, the first multiple regression equation having a small number of variables and a small calculation load is used when the tension deviation, which occupies most in terms of occurrence frequency, is less than a predetermined value, and the prediction accuracy is high. Second, which has a higher computational load than the multiple regression equation
Since the multiple regression equation is used when the tension deviation, which is small in terms of occurrence frequency, is equal to or greater than a predetermined value, the calculation load is reduced as a whole. When the tension deviation is less than a predetermined value, the first multiple regression equation that does not include the tension deviation as a variable is used, so that the effect of not including the tension deviation as a variable is minimized, and the hot-rolled steel strip as a whole is It is possible to accurately predict the minimum board width of.

Further, according to the present invention, the plate thickness, plate width and temperature of the hot rolled steel strip are changed between the hot rolling mill and the winder for winding the hot rolled steel strip rolled by the hot rolling mill. The step of detecting over, and the step of obtaining the carbon equivalent of the hot rolled steel strip,
The step of obtaining the total length of the hot-rolled steel strip, and the hot-rolled steel strip is virtually equally divided in the longitudinal direction, and the minimum of each detected value is obtained from the detected values of the plate thickness, width and temperature in each virtual equally divided region. Value, average value and maximum value respectively, and selecting any one of these for each detected value, and the selected value and carbon equivalent are the thickness and width of the hot rolled steel strip. ,
Substituting in a predetermined multiple regression equation with temperature and carbon equivalent as variables to obtain the strip width of the hot-rolled steel strip at room temperature for each virtual equal division area, and the hot-rolled steel strip from the obtained strip width And a step of calculating a minimum strip width of the strip, which is a strip width prediction method for a hot-rolled steel strip. According to the present invention, by virtually dividing the hot-rolled steel strip into equal parts, it is possible to predict the plate width of the entire length of the hot-rolled steel strip at room temperature by the number of calculations corresponding to the number of divisions. The calculation time can be significantly reduced, and the strip width of the hot-rolled steel strip can be predicted very efficiently.

Further, the present invention provides a method of rolling the hot rolled steel strip between the hot rolling mill and the winder for winding the hot rolled steel strip rolled by the hot rolling mill. Over the entire length, a step of obtaining the carbon equivalent of the hot rolled steel strip, a step of detecting a tension deviation between the detected tension and a preset tension, a step of obtaining the total length of the hot rolled steel strip, The steel strip is virtually equally divided in the longitudinal direction, and the minimum value, the average value, and the maximum value of each detected value from the detected values of the plate thickness, the plate width, the temperature, and the tension deviation in each virtual equally divided region, respectively. The step of obtaining and selecting any one of these for each detected value, and the value and carbon equivalent selected for each of the detected values are the plate thickness, plate width, temperature and tension of the hot rolled steel strip. Substituting in a multiple regression equation that uses deviation and carbon equivalent as variables A heat treatment characterized by including a step of obtaining a strip width of a hot-rolled steel strip at room temperature for each virtual equal division region, and a step of calculating a minimum strip width of the hot-rolled steel strip from the obtained strip width. This is a method for predicting the strip width of a steel strip. According to the present invention, by virtually splitting the hot-rolled steel strip in the longitudinal direction, it is possible to predict the plate width of the total length of the hot-rolled steel strip at room temperature by the number of calculations corresponding to the number of splits. The plate width calculation time can be significantly shortened, and the plate width of the hot-rolled steel strip can be predicted very efficiently. In addition, since the strip width of the hot rolled steel strip at room temperature is predicted by the multiple regression equation having good strip width prediction accuracy including the tension deviation as a variable, the strip width of the hot rolled steel strip can be accurately predicted.

Further, according to the present invention, the plate thickness, the plate width, the temperature and the tension of the hot rolled steel strip are provided between the hot rolling mill and the winder for winding the hot rolled steel strip rolled by the hot rolling mill. Over the entire length, a step of obtaining the carbon equivalent of the hot rolled steel strip, a step of detecting a tension deviation between the detected tension and a preset tension, a step of obtaining the total length of the hot rolled steel strip, The steel strip is virtually equally divided in the longitudinal direction, and the minimum value, the average value, and the maximum value of each detected value from the detected values of the plate thickness, the plate width, the temperature, and the tension deviation in each virtual equally divided region, respectively. The step of obtaining and selecting any one of these for each detected value, and the value selected for each detected value and the carbon equivalent only when the selected value of the tension deviation is a predetermined value or more And the hot rolled steel strip thickness, width, temperature, tension deviation and Substituting into a predetermined multiple regression equation using the carbon equivalent as a variable, the step of obtaining the strip width of the hot-rolled steel strip at room temperature, and the minimum strip width of the hot-rolled steel strip is calculated from the obtained strip width. And a step for predicting the strip width of a hot rolled steel strip. According to the present invention, by virtually dividing the hot-rolled steel strip into equal parts, it is possible to predict the plate width of the entire length of the hot-rolled steel strip at room temperature by the number of calculations corresponding to the number of divisions. The calculation time can be significantly reduced, and the strip width of the hot-rolled steel strip can be predicted very efficiently. Also,
Only when the tension deviation, which is a condition where a large amount of width shrinkage is likely to occur, is equal to or more than a predetermined value, the strip width of the hot-rolled steel strip at room temperature is used by using the multiple regression equation with good strip width prediction accuracy including the tension deviation as a variable. Therefore, it is possible to significantly reduce the strip width calculation time and to accurately predict the minimum strip width of the hot-rolled steel strip.

In the present invention, the step of obtaining the total length of the hot-rolled steel strip is performed, and the hot-rolled steel strip is virtually equally divided in the longitudinal direction, and the plate thickness, width, temperature and The step of obtaining the minimum value, the average value and the maximum value of the respective detected values from the detected value of the tension deviation and selecting any one of these for each detected value, and the selected value of the tension deviation are previously set. When the value is less than the specified value, the selected values of the plate thickness, plate width, and temperature and the carbon equivalent are substituted into the first multiple regression equation, and the plate width of the hot rolled steel strip at room temperature is calculated for each virtual equal division region. And when the selected value of the tension deviation is greater than or equal to a predetermined value, the selected values of plate thickness, plate width, temperature and tension deviation and carbon equivalent are substituted into the second multiple regression equation. And the step of obtaining the strip width of the hot-rolled steel strip at room temperature for each virtual equal division area. Characterized in that it contains. According to the present invention, by virtually splitting the hot-rolled steel strip in the longitudinal direction, it is possible to predict the plate width of the total length of the hot-rolled steel strip at room temperature by the number of calculations corresponding to the number of splits. The plate width calculation time can be significantly shortened, and the plate width of the hot-rolled steel strip can be predicted very efficiently. When the selected value of the tension deviation is less than a predetermined value, the first multiple regression equation with a small calculation load and not including the tension deviation as a variable is used, and the selected value of the tension deviation is a predetermined value. In the above case, the calculation load is larger than that of the first multiple regression formula, and the second multiple regression formula having a good prediction accuracy including the tension deviation as a variable is used. It is possible to accurately predict the strip width of the steel strip.

Further, the present invention is arranged between a hot rolling mill and a winder for winding the hot rolled steel strip rolled by the hot rolling mill,
A sheet thickness detecting means for detecting the sheet thickness of the hot rolled steel strip, a sheet width detecting means arranged between the hot rolling mill and the winder for detecting the sheet width of the hot rolled steel strip, and hot rolling. Temperature detecting means arranged between the rolling mill and the winder to detect the temperature of the hot rolled steel strip, and arranged between the hot rolling mill and the winder to detect the tension of the hot rolled steel strip. The tension detecting means, the tension setting means for setting the tension of the hot rolled steel strip to a predetermined value, the carbon equivalent calculating means for calculating the carbon equivalent, and the tension detected by the tension detecting means and the tension setting means are set. A tension deviation detecting means for detecting a tension deviation from a predetermined set tension, a judging means for judging the magnitude of the tension deviation and a predetermined value in response to an output of the tension deviation detecting means, and an output of each of the detecting means. Is sampled, the total length of the hot-rolled steel strip is obtained, and the hot-rolled steel strip is virtually divided into equal parts in the longitudinal direction. Then, the minimum value, the average value and the maximum value of the detection values of the respective detection means in each virtual equal division area are respectively obtained, and one of these is selected for each detection value to select the tension deviation. When the determined value is less than the predetermined value by the determination means, the thickness of the hot rolled steel strip is determined based on the outputs of the thickness detection means, the width detection means, the temperature detection means and the carbon equivalent calculation means. , The strip width, temperature and carbon equivalent are used as variables to obtain the strip width of the hot rolled steel strip at room temperature. The strip width of the hot rolled steel strip at room temperature is obtained using the first multiple regression equation, and the selected value of the tension deviation is obtained. Is determined to be equal to or greater than a predetermined value by the determination means, the plate thickness detection means, the plate width detection means, the temperature detection means,
Second multiple regression based on the outputs of the tension deviation detection means and the carbon equivalent calculation means, and using the thickness, width, temperature, tension deviation and carbon equivalent of the hot rolled steel strip as variables, the strip width of the hot rolled steel strip at room temperature Using a formula to obtain the strip width of the hot-rolled steel strip at room temperature, and a calculating means for calculating the minimum strip width of the hot-rolled steel strip from the obtained strip width, It is a plate width prediction device. According to the invention, the apparatus for predicting the strip width of a hot-rolled steel strip virtually divides the hot-rolled steel strip in the longitudinal direction, and obtains the strip width of the hot-rolled steel strip at room temperature for each virtual equal division region. Therefore, the strip width of the entire area of the hot-rolled steel strip can be obtained by the number of calculations corresponding to the number of divisions. Therefore, the calculation time of the strip width of the hot-rolled steel strip is significantly shortened, and the minimum strip width of the hot-rolled steel strip at room temperature can be predicted very efficiently. Further, when the tension deviation is less than a predetermined value, that is, when the influence of the tension deviation on the plate width is small, the first multiple regression equation in which the calculation load is small and the tension deviation is not included as a variable is used.
When the tension deviation is equal to or greater than a predetermined value, that is, when the influence of the tension deviation on the strip width is great, the calculation load is the first.
A second multiple regression equation that is larger than the multiple regression equation and includes the tension deviation as a variable and has good prediction accuracy is used. As a result, the first multiple regression equation and the second multiple regression equation can compensate for each other's drawbacks, reducing the overall calculation load and accurately and efficiently predicting the minimum strip width of the hot-rolled steel strip. can do.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a system diagram showing a simplified structure of a strip width predicting apparatus for a hot rolled steel strip according to a first embodiment of the present invention. FIG. 1 also shows the outlet side configuration of the hot rolling equipment. Hot rolled steel strip 1 made from low carbon steel
Is hot-rolled to a predetermined plate thickness in a plurality of hot rolling mills 3, water-cooled to a predetermined temperature in a cooling device 4, and wound by a winder 6 via a deflector roll 5. The strip width predicting device 7 for the hot rolled steel strip has a strip thickness detecting means 8
, Tension detecting means 9, temperature detecting means 10, plate width detecting means 11, calculating means 12, tension deviation detecting means 13
The determination unit 14, the carbon equivalent calculation unit 15, the tension setting unit 16, and the weight detection unit 17 are included.

The plate thickness detecting means 8 is arranged between the final stand of the plurality of hot rolling mills 3 and the winding machine 6 for winding the hot-rolled steel strip 1 rolled by the hot rolling mills 3, The plate thickness of the steel strip 1 is detected by X-ray. The tension detecting means 9 is arranged between the plate thickness detecting means 8 and the winder 6 and measures the load applied to the tension detecting roll to detect the tension. The temperature detecting means 10 is realized by a radiation thermometer or the like and is arranged between the cooling device 4 and the winder 6 to detect the temperature of the hot-rolled steel strip 1. The strip width detecting means 11 is realized by, for example, a photoelectric width meter, is arranged close to the downstream side of the temperature detecting means 10, and detects the strip width of the hot-rolled steel strip 1. The tension deviation detection means 13 is realized by a subtraction circuit or the like, and detects a tension deviation between the detected tension of the tension detection means 9 and the preset tension set by the tension setting means 16. The determination means 14 is realized by a comparison circuit or the like, and in response to the output of the tension deviation detection means 13, determines the magnitude of the tension deviation and the magnitude of the predetermined value. The tension setting means 16 inputs a predetermined tension value from a keyboard or the like to set the tension of the hot rolled steel strip. The weight detecting means 17 is realized by a weighing machine or the like, and detects the weight of the hot-rolled steel strip 1 wound by the winder 6 or the weight of a flat steel strip called a slab before hot rolling.

The carbon equivalent calculating means 15 is realized by a personal computer or the like, and the carbon equivalent of the hot-rolled steel strip 1 (hereinafter sometimes abbreviated as carbon equivalent).
Is calculated based on a predetermined calculation formula. Carbon equivalent is an index value representing the strength and deformability of the hot rolled steel strip,
The calculation formula is experimentally determined based on the chemical composition of the hot rolled steel strip. The formula for calculating the carbon equivalent is, for example, Ceq for the carbon equivalent, C (%) for the carbon content of the hot rolled steel strip, Mn (%) for the manganese content, Si (%) for the silicon content, and Ni for the nickel content. (%), Chromium content is Cr
(1), where molybdenum content is Mo (%) and vanadium content is V (%).

[0019]

[Equation 1]

The calculating means 12 is realized by a computer or the like, samples the output of each detecting means, determines the total length of the hot-rolled steel strip 1, and virtually divides the hot-rolled steel strip 1 in the longitudinal direction. (For example, 12 divisions), the data sampled for each virtual equal division area (hereinafter, may be abbreviated as a division area) is processed, and the processed data is substituted into a preset multiple regression equation described below, The strip width of the hot-rolled steel strip at room temperature is calculated for each divided region, and the minimum strip width of the hot-rolled steel strip is calculated from the obtained strip width.

FIG. 2 is a schematic plan view showing a hot rolled steel strip developed in the longitudinal direction. The strip width WA of the hot-rolled steel strip 1 fluctuates in the longitudinal direction due to variations in the rolling tension, and the narrow neck portions N1 and N2 are formed so as to face both end portions of the strip width of the hot-rolled steel strip 1. Has been done. At the time of hot rolling, the strip width center line C1 of the hot rolled steel strip 1 is adjusted so as to coincide with the center line of the hot rolling mill 3, so that the neck portions N1 and N2 have the center line C.
It is formed almost symmetrically with respect to 1. As described above, the hot rolled steel strip 1 is virtually divided into 12 equal parts (m1 to m12), and the plate width of the hot rolled steel strip 1 is predicted for each divided region. The total length L of the hot-rolled steel strip 1 is, for example, 600 m, and the length u of the divided region is, for example, 50 m.

FIG. 3 is a block diagram showing the electrical construction of the apparatus for predicting the width of a hot rolled steel strip. The calculating means 12 which is a processing circuit samples the output signals of the plate thickness detecting means 8, the plate width detecting means 11, the temperature detecting means 10 and the tension detecting means 9. The tension deviation detecting means 13 detects a tension deviation between the sampled detected tension and the tension set by the tension setting means 16, and sends a tension deviation detection signal to the computing means 12. The carbon equivalent calculating means 15 calculates the carbon equivalent and sends a signal representing the carbon equivalent to the calculating means 12, and the weight detecting means 1
7 sends a weight detection signal to the calculation means 12. Discrimination means 14
Determines the magnitude of the tension deviation and a predetermined value, and sends a tension deviation determination signal to the calculation means 12. Computing means 1
2, the total length L of the hot-rolled steel strip 1 is obtained based on the above signals, the hot-rolled steel strip 1 is virtually divided into 12 equal parts in the longitudinal direction, and the hot-rolled steel strip 1 at room temperature is divided for each divided region. The sheet width of the hot-rolled steel strip 1 is calculated from the obtained sheet width, and the sheet width is displayed on the display device 18 and printed by the printing device 19.

FIG. 4 is a flow chart for explaining the plate width predicting method according to the first embodiment of the present invention. At step a1, sampling of each of the detected values is performed. That is, the detection signals of the plate thickness detecting means 8, the plate width detecting means 11, the temperature detecting means 10, and the tension detecting means 9 are sent to the computing means 12 and sampled as the data for predicting the sheet width of the hot rolled steel strip at room temperature. At step a2, the carbon equivalent is calculated. The calculation of the carbon equivalent is performed by the carbon equivalent calculating means 15 by substituting the chemical components of the hot-rolled steel strip 1 into the equation (1). The chemical composition of the hot-rolled steel strip 1 is analyzed in the steelmaking process which is the upper process, and for example, C: 0.04%, Mn: 0.25%, Si: 0.02.
%, Ni: 0.02%, Cr: 0.02%, Mo: 0.
01% and V: 0.01%. In this case, the carbon equivalent Ceq is 0.09. In step a3, the tension deviation detecting means 13 detects the tension deviation. The tension deviation is detected by detecting the deviation between the sampled tension detection value and the tension setting value set by the tension setting means 16. The detected tension deviation signal is sent to the calculating means 12. At step a4, the total length L of the hot-rolled steel strip 1 is calculated. The calculation of the total length L is performed using the detection values of the plate thickness detecting means 8, the plate width detecting means 11 and the weight detecting means 17.

At step a5, data processing is performed. The data processing is performed by the calculation means 12 as follows. (A) The total length L of the hot-rolled steel strip 1 is virtually divided into 12 equal parts. (B) For each divided area, the minimum value, the average value and the maximum value of each detected value are calculated from the detected values of the plate thickness, the plate width, the temperature and the tension deviation in each area. (C) From the calculated values obtained in (b) above,
For each divided area, for example, select the average value of the plate thickness, the minimum value of the plate width, the maximum value of the temperature, the maximum value of the tension deviation,
It is saved as representative data of each divided area.

At step a6, n indicating the division number is n
= 1 is set, and the sheet width prediction of the hot rolled steel strip in the first divided region m1 is started. In step a7, it is determined whether or not the maximum value of the tension deviation in the first divided area m1 is a predetermined value, for example, less than 10 tons. If this determination is affirmative, the process proceeds to step a8. Step a8
Then, the plate width of the hot-rolled steel strip 1 at room temperature is calculated by using a preset first multiple regression equation described later. The first multiple regression equation is a multiple regression equation for estimating the strip width of the hot rolled steel strip at room temperature by using the strip thickness, strip width and temperature of the hot rolled steel strip 1 during hot rolling and the carbon equivalent as variables. Since the deviation is not included in the variable, it is particularly preferably used when the maximum value of the tension deviation is small, that is, when the influence of the tension deviation on the plate thickness is small. The first multiple regression equation is WA for the strip width of the hot-rolled steel strip at room temperature, WB for the strip width of the hot-rolled steel strip during hot rolling, TA for the strip thickness, TC for the temperature, and carbon equivalent of the hot-rolled strip. Is expressed as Ceq, it is expressed by equation (2).

WA = WB- {aTA + (bTC-c) WB-dCeq} (2) However, the coefficients a, b, c and d are constants.

The calculation of the plate width in step a8 is performed as follows.
It is performed by substituting the representative data in the first divided region m1, that is, the average value of the plate thickness, the minimum value of the plate width, the maximum value of the temperature, and the carbon equivalent into the formula (2). The part of the right side of the equation (2) excluding the first term represents the amount of width shrinkage, and the coefficients a to d are determined in advance from the past rolling record of the hot rolling equipment.

If the determination in step a7 is negative, the process proceeds to step a9. In step a9, the plate width of the hot-rolled steel strip at room temperature is calculated by using the preset second multiple regression equation described later. The second multiple regression equation is a multiple regression equation for estimating the sheet width of the hot rolled steel strip at room temperature with the thickness, strip width, temperature and tension deviation of the hot rolled steel strip during hot rolling and carbon equivalent as variables. Since the tension deviation is included in the variable, it is particularly preferably used when the maximum value of the tension deviation is large, that is, when the influence of the tension deviation on the plate width is large. The second multiple regression equation is represented by the equation (3) when the tension deviation of the hot-rolled steel strip during hot rolling is PA and the other symbols are the same as the equation (2).

WA = WB- {eTA + (fTC-g) WB-hCeq + iPA} (3) However, the coefficients e, f, g, h, and i are constants.

The calculation of the plate width in step a9 is as follows.
It is performed by substituting the representative data in the first divided region m1, that is, the average value of the plate thickness, the minimum value of the plate width, the maximum value of the temperature and the maximum value of the tension deviation, and the carbon equivalent in the expression (3). The portion of the right side of the equation (3) excluding the first term represents the amount of width shrinkage, and the coefficients e to i are determined in advance from the past rolling record of the hot rolling equipment.

At step a10, it is judged whether or not n indicating the division number is n = 12. If this determination is negative, the process proceeds to step a11, n is incremented by +1 and the process returns to step a7 again. In step a7, similar to the case of the first divided region m1, the plate width prediction of the hot-rolled steel strip 1 in the second divided region m2 is started.
The processing loop from step a7 to step a11 is repeated until the determination in step a10 becomes affirmative, that is, until the plate width prediction of the twelfth divided region m12 is completed. If the determination in step a10 is affirmative, the process proceeds to step a12. At step a12, the minimum plate width is calculated from the plate widths obtained for each of the divided areas.

At step a13, it is judged whether or not the minimum width is equal to or more than the quality design value W1. If this determination is affirmative, the process proceeds to step a14. In step a14, it is predicted that the strip width of the hot-rolled steel strip 1 at room temperature will be larger than the minimum required strip width over the entire region, so that the next step is promoted according to the initially set manufacturing instruction. If this determination is negative, the process proceeds to step a15. Step a15
Since it is predicted that the strip width of the hot-rolled steel strip 1 at room temperature may be smaller than the minimum required strip width, the initially set manufacturing directive was changed, and the next process was promoted in accordance with the changed manufacturing directive. To be done.

Next, the prediction accuracy of the first multiple regression equation and the second multiple regression equation (hereinafter sometimes collectively referred to as multiple regression equation) used in the first embodiment of the present invention will be described. . Since the multiple regression equation of the first embodiment of the present invention is a multiple regression equation for predicting the strip width of the hot-rolled steel strip 1 at room temperature, the evaluation of the prediction accuracy of the multiple regression equation is essentially the hot rolling at room temperature. The strip width of the steel strip 1 should be used as a prediction target. However, since the board width is predicted through the width shrinkage amount as shown in the first multiple regression equation (2) and the second multiple regression equation (3), the width shrinkage amount is predicted. By also setting, it is possible to evaluate the prediction accuracy of the multiple regression equation.

FIG. 5 is a characteristic diagram showing the prediction accuracy of the multiple regression equation with the width reduction amount as the prediction target. The horizontal axis of FIG. 5 is the actual width reduction amount, and the vertical axis is the calculation width reduction amount. Straight line L
1 is 45 ° where the actual width reduction amount and the calculated width reduction amount match
It is a line, and the prediction accuracy of the multiple regression equation becomes higher as it gets closer to the straight line L1, and the prediction accuracy of the multiple regression equation becomes lower as it goes away from the straight line L1. The straight line L2 is a straight line relating to the first embodiment of the present invention, and shows the prediction accuracy of the first multiple regression equation when the tension deviation is less than 10 tons. The straight line L3 is a straight line relating to the first embodiment of the present invention, and shows the prediction accuracy of the second multiple regression equation when the tension deviation is 10 tons or more. A straight line L4 shows the prediction accuracy of the first multiple regression equation when the tension deviation is not limited, and a straight line L5 shows the prediction accuracy of the second multiple regression equation when the tension deviation is not limited. . From FIG. 5, each straight line is located in the order of straight lines L5, L3, L2, and L4 in the order close to the straight line L1, and among these straight lines L5, L3,
Both L2 are very close to the straight line L1, and the straight line L
It can be seen that only 4 is slightly separated from the straight line L1. That is, from FIG. 5, the prediction accuracy of the second multiple regression equation that includes the tension deviation as a variable is high under any condition, and the prediction accuracy of the first multiple regression equation that does not include the tension deviation as a variable is only when the tension deviation is small. It turns out to be expensive. Therefore, the prediction precisions of the first multiple regression equation and the second multiple regression equation used in the first embodiment of the present invention are both very good.

As described above, in the first embodiment of the present invention, the total length of the hot-rolled steel strip 1 is virtually divided into 12 equal parts, and representative data is selected from the data sampled for each virtual equal division area. Then, the representative data is substituted into a preset multiple regression equation, and the strip width of the hot-rolled steel strip 1 at room temperature is calculated for each virtual equal division region. Therefore, it is possible to predict the strip width of the entire area of the hot-rolled steel strip 1 by a total of 12 computations, and the computation time is extremely large compared to the case where the strip width of the entire length is calculated without virtually dividing it equally. Can be shortened to Further, in the first embodiment of the present invention, the first multiple regression equation having a small number of variables and a small calculation load is used when the tension deviation for which good prediction accuracy is desired is less than 10 tons, and the prediction accuracy is high. However, the second multiple regression equation, which has a larger calculation load than the first multiple regression equation, is used in the first multiple regression equation when the tension deviation, which cannot be expected to have high prediction accuracy, is 10 tons or more. In this way, the first multiple regression equation and the second multiple regression equation are applied in a well-balanced manner so as to make up for each other's drawbacks, so the overall calculation load is further reduced, and the minimum strip width of the hot-rolled steel strip is accurately determined. And can be predicted efficiently.

FIG. 6 is a transition chart showing the number of occurrence of abnormal width steel strips by month. The transition line S1 is a transition line showing the number of occurrence of abnormal width steel strips for each month in the period to which the strip width prediction method for a hot rolled steel strip according to the first embodiment of the present invention is applied.
2 is a transition line showing the number of abnormally wide steel strips generated by the conventional method for each month. It can be seen from FIG. 6 that the number of abnormal width steel strips is remarkably reduced by applying the strip width prediction method for the hot-rolled steel strip according to the first embodiment of the present invention. This means that when the sheet width prediction predicts that the sheet width of the hot-rolled steel strip at room temperature becomes smaller than the minimum required sheet width, an appropriate sheet width target based on the sheet width predicted before proceeding to the next step. This is because the manufacturing instruction is changed to a value and the next process is promoted according to the changed manufacturing instruction of the plate width. As described above, by applying the first embodiment of the present invention, the manufacturing instruction of the strip width is changed in a state where the excessive change of the strip width target value is avoided, so that the width yield of the cold-rolled steel strip is reduced. Without this, the number of abnormally width cold-rolled steel strips can be significantly reduced. Therefore, the manufacturing yield of the cold rolled steel strip is significantly improved.

As a second embodiment of the present invention, the total length of the hot-rolled steel strip 1 is not virtually equally divided, and the tension deviation is 10%.
When it is less than tonnes, the first multiple regression equation is used, and when the tension deviation is 10 tons or more, the second multiple regression equation is used to obtain the strip width of the hot-rolled steel strip 1 over the entire length, and the strip width of the obtained strip width is calculated. The minimum strip width of the hot-rolled steel strip 1 may be calculated from the inside. The configuration of the strip width predicting device for a hot rolled steel strip in the second embodiment of the invention is the same as that of the first embodiment of the invention except that the weight detecting means 17 is not included. The plate width prediction method for the hot-rolled steel strip 1 in the second embodiment is the same as that of the first embodiment of the invention except that the entire length of the hot-rolled steel strip 1 is not virtually equally divided. Therefore, in the second embodiment of the invention, steps a4 to a6 and steps a10 to a11 in the flowchart of FIG. 4 are omitted. Thus, in the second embodiment of the invention,
Hot rolled steel strip 1 without virtually dividing the entire length of hot rolled steel strip 1
Since the strip width is calculated over the entire length, the minimum strip width of the hot-rolled steel strip 1 can be predicted more accurately than in the first embodiment of the invention. Further, the first multiple regression equation having a small number of variables and a small calculation load is used when the tension deviation, which accounts for most of the occurrence frequency, is small. Since the large second multiple regression equation is used when the tension deviation is small in terms of frequency of occurrence and the tension deviation is large, the calculation load is reduced as a whole.

As a third embodiment of the present invention, the entire length of the hot-rolled steel strip 1 is virtually equally divided, and the second multiple regression equation is used for each virtual equally divided region regardless of the magnitude of the tension deviation. Alternatively, the strip width of the hot-rolled steel strip 1 may be obtained, and the minimum strip width of the hot-rolled steel strip 1 may be calculated from the obtained strip widths. The configuration of the strip width predicting apparatus for the hot-rolled steel strip 1 according to the third embodiment of the invention is the same as that of the first embodiment of the invention except that the determining means 14 is not included. The strip width prediction method for the hot-rolled steel strip 1 according to the third aspect of the invention is different from that of the invention except that the determination is not performed based on the magnitude of the tension deviation and the strip width is not calculated by the first multiple regression equation. This is the same as the first embodiment. Therefore, in the third embodiment of the invention, steps a7 to a7 in the flowchart of FIG.
a8 is omitted. As described above, in the third embodiment of the invention, the entire length of the hot-rolled steel strip 1 is virtually divided into equal parts, and
Since the strip width of the hot-rolled steel strip 1 is obtained only by using the second multiple regression equation having a better prediction accuracy than the multiple regression equation, the strip width calculation time is extremely longer than that in the case where virtually no equal division is performed. It can be significantly shortened. Further, since the prediction accuracy of the second multiple regression equation in the third embodiment of the present invention is extremely high as represented by the straight line L5 in FIG. 5, the third embodiment of the present invention is The minimum strip width of the hot-rolled steel strip 1 can be predicted more accurately and efficiently than in the first embodiment of the invention.

As a fourth embodiment of the present invention, the entire length of the hot-rolled steel strip 1 is virtually divided into equal parts, and only when the magnitude of the tension deviation is 10 tons or more in each of the virtual equal-divided regions, the second part is formed. The strip width of the hot-rolled steel strip 1 may be obtained using a multiple regression equation, and the minimum strip width of the hot-rolled steel strip 1 may be calculated from the obtained strip widths. Hot Rolled Steel Strip 1 in Fourth Embodiment of Invention
The configuration of the strip width predicting device is the same as that of the first embodiment of the invention, and the hot-rolled steel strip 1 according to the fourth embodiment of the invention is used.
The plate width predicting method is the same as the first embodiment of the invention except that the plate width is not calculated at all when the magnitude of the tension deviation is less than 10 tons. Therefore, in the fourth embodiment of the invention, step a8 in the flowchart of FIG. 4 is omitted, and if the determination in step a7 is affirmative, the process proceeds to step a10. As described above, in the fourth embodiment of the invention, the entire length of the hot-rolled steel strip 1 is virtually divided into equal parts, and the influence of the tension deviation on the strip width is large, and the magnitude of the tension deviation easily causes width contraction. Since the strip width of the hot-rolled steel strip 1 is obtained by using the second multiple regression equation only when is 10 tons or more, the strip width calculation time can be further shortened as compared with the third embodiment of the invention. It is possible. Further, since the prediction accuracy of the second multiple regression equation in the fourth embodiment of the present invention is extremely high as represented by the straight line L3 in FIG. 5, the fourth embodiment of the present invention is It is possible to predict the minimum strip width of the hot-rolled steel strip 1 extremely accurately as in the third embodiment of the invention and more efficiently than in the third embodiment of the invention.

As a fifth embodiment of the present invention, the entire length of the hot-rolled steel strip 1 is virtually equally divided, and only the first multiple regression equation is used which does not include the tension deviation for each virtual equally divided region. Alternatively, the strip width of the hot-rolled steel strip 1 may be obtained, and the minimum strip width of the hot-rolled steel strip 1 may be calculated from the obtained strip widths. The strip width predicting apparatus for the hot-rolled steel strip 1 according to the fifth embodiment of the present invention is configured by a tension detecting means 9, a tension deviation detecting means 13,
The method is the same as that of the first embodiment of the invention except that the determination means 14 and the tension setting means 16 are not included, and the strip width prediction method for the hot-rolled steel strip 1 according to the fifth embodiment of the invention is It is the same as the first embodiment of the present invention except that the determination is not performed according to the magnitude of the deviation and the plate width is not calculated by the second multiple regression equation. Therefore, in the fifth embodiment of the invention, steps a7 and a9 in the flowchart of FIG. 4 are omitted. As described above, in the fifth embodiment of the invention, the entire length of the hot-rolled steel strip 1 is virtually divided into equal parts, and only the first multiple regression equation, which has a smaller calculation load than the second multiple regression equation, is used. Since the strip width of the steel strip 1 is calculated, it is possible to significantly reduce the strip width calculation time as compared with the case where the strip width is not virtually evenly divided. The board width can be calculated in a shorter time than the form. The fifth embodiment of the present invention
The prediction accuracy of the first multiple regression equation in the form of FIG.
Although the accuracy is slightly low as represented by the straight line L4, the strip width of the hot-rolled steel strip 1 at room temperature can be predicted more accurately than the method based on experience and intuition in the conventional technique.

As described above, in the second to fifth embodiments of the present invention, it is possible to accurately and efficiently predict the strip width of the hot-rolled steel strip 1 at room temperature. When the occurrence of the steel strip is predicted, the manufacturing instruction can be changed appropriately as in the first embodiment of the invention, and the number of the abnormal width steel strips can be significantly reduced. The sheet width prediction of the hot rolled steel strip at room temperature in the first to fifth embodiments of the invention may be performed online during hot rolling, or offline after hot rolling is completed. Good.

[0042]

As described above, according to the present invention, the minimum strip width of the hot-rolled steel strip at room temperature can be predicted extremely accurately and efficiently. When it is predicted that the width will be smaller than the minimum required board width in the process, the board width manufacturing command can be changed to an appropriate board width target value based on the predicted board width. As a result, the manufacturing instruction of the strip width is changed while avoiding the excessive change of the strip width target value.Therefore, the width yield of cold-rolled steel strip does not decrease, and the number of abnormal width cold-rolled steel strips is generated. Can be significantly reduced. Therefore, the manufacturing yield of the cold rolled steel strip is significantly improved.

[Brief description of drawings]

FIG. 1 is a system diagram showing a simplified configuration of a strip width prediction device for a hot rolled steel strip according to a first embodiment of the present invention.

FIG. 2 is a schematic plan view showing a hot-rolled steel strip developed in the longitudinal direction.

FIG. 3 is a block diagram showing an electrical configuration of a strip width prediction device for a hot rolled steel strip.

FIG. 4 is a flowchart for explaining a plate width prediction method according to the first embodiment of this invention.

FIG. 5 is a characteristic diagram showing the prediction accuracy of the multiple regression equation with the width reduction amount as a prediction target.

FIG. 6 is a transition chart showing the number of occurrence of abnormal width steel strips by month.

FIG. 7 is a characteristic diagram showing a relationship between a strip width distribution of a hot rolled steel strip at room temperature and a strip width target value, a quality design value, and a production specification strip width.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Hot-rolled steel strip 3 Hot rolling mill 6 Winding machine 7 Sheet width prediction device for hot-rolled steel strip 8 Sheet thickness detection means 9 Tension detection means 10 Temperature detection means 11 Sheet width detection means 12 Calculation means 13 Tension deviation detection means 14 determination means 15 carbon equivalent calculation means

Claims (6)

[Claims] 1. The thickness, width, temperature and tension of the hot-rolled steel strip are controlled over the entire length between the hot-rolling mill and the winder for winding the hot-rolled steel strip rolled by the hot-rolling mill. The step of detecting, the step of obtaining the carbon equivalent of the hot rolled steel strip, the step of detecting the tension deviation between the detected tension and the preset tension, and the step of detecting the tension deviation when the tension deviation is less than the preset value. Substituting the detected value of the plate thickness, the plate width and the temperature and the carbon equivalent into a predetermined first multiple regression equation in which the plate thickness, the plate width, the temperature and the carbon equivalent are variables, and A step of obtaining a strip width, and when the tension deviation is equal to or more than a predetermined value, the strip is subjected to a predetermined second multiple regression equation having the thickness of the hot rolled steel strip, the width, the temperature, the tension deviation and the carbon equivalent as variables. If the detected values of thickness, width, temperature and tension deviation are Substituting the carbon equivalent to the step of obtaining the strip width of the hot-rolled steel strip at room temperature, and calculating the minimum strip width of the hot-rolled steel strip from the obtained strip width. Method for predicting strip width of hot rolled steel strip. 2. The plate thickness, plate width and temperature of the hot rolled steel strip are detected over the entire length between the hot rolling mill and the winder for winding the hot rolled steel strip rolled by the hot rolling mill. A step, a step of obtaining the carbon equivalent of the hot-rolled steel strip, a step of obtaining the total length of the hot-rolled steel strip, and a step of virtually dividing the hot-rolled steel strip in the longitudinal direction into the virtual equal division regions. A step of obtaining a minimum value, an average value and a maximum value of the respective detected values from the detected values of the thickness, the plate width and the temperature, and selecting any one of these for each detected value; The carbon equivalent and the plate thickness of the hot rolled steel strip,
Substituting the strip width, temperature and carbon equivalent into a predetermined multiple regression equation with variables to obtain the strip width of the hot-rolled steel strip at room temperature for each virtual equally divided region, and from the strip width obtained And a step of calculating a minimum strip width of the hot-rolled steel strip. 3. The plate thickness, plate width, temperature and tension of the hot rolled steel strip are adjusted over the entire length between the hot rolling mill and the winder for winding the hot rolled steel strip rolled by the hot rolling mill. The step of detecting, the step of obtaining the carbon equivalent of the hot rolled steel strip, the step of detecting the tension deviation between the detected tension and the preset tension, the step of obtaining the total length of the hot rolled steel strip, Virtually equally divided in the longitudinal direction, the plate thickness in each virtual equally divided region, the plate width, the minimum value of each detected value from the detected value of the temperature and tension deviation, the average value and the maximum value, respectively, The step of selecting any one of these for each detected value, and the value and carbon equivalent selected for each of the detected values are the sheet thickness, sheet width, temperature, tension deviation and carbon of the hot rolled steel strip. Substituting into a predetermined multiple regression equation with equivalence as a variable A hot-rolled steel strip characterized by including a step of obtaining a strip width of a hot-rolled steel strip at room temperature for each region, and a step of calculating a minimum strip width of the hot-rolled steel strip from the obtained strip width. Board width prediction method. 4. The plate thickness, plate width, temperature and tension of the hot rolled steel strip are adjusted over the entire length between the hot rolling mill and the winder for winding the hot rolled steel strip rolled by the hot rolling mill. The step of detecting, the step of obtaining the carbon equivalent of the hot rolled steel strip, the step of detecting the tension deviation between the detected tension and the preset tension, the step of obtaining the total length of the hot rolled steel strip, Virtually equally divided in the longitudinal direction, the plate thickness in each virtual equally divided region, the plate width, the minimum value of each detected value from the detected value of the temperature and tension deviation, the average value and the maximum value, respectively, The step of selecting any one of these for each detected value, and the value selected for each detected value and the carbon equivalent only when the selected value of the tension deviation is a predetermined value or more,
Substituting the plate thickness, plate width, temperature, tension deviation and carbon equivalent of the hot-rolled steel strip into a predetermined multiple regression equation to obtain the plate width of the hot-rolled steel strip at room temperature, And a step of calculating a minimum strip width of the hot rolled steel strip from the strip width. 5. The step of obtaining the total length of a hot-rolled steel strip, and the hot-rolled steel strip being virtually divided in the longitudinal direction, and the plate thickness, plate width, temperature and tension deviations in each virtual equally divided region. The step of obtaining the minimum value, the average value and the maximum value of each detected value from the detected values and selecting any one of these for each detected value; and the selected value of the tension deviation is less than a predetermined value. When, the selected values of the plate thickness, plate width, and temperature and the carbon equivalent are substituted into the first multiple regression equation to obtain the plate width of the hot rolled steel strip at room temperature for each virtual equal division region. And when the selected value of the tension deviation is equal to or greater than a predetermined value, the selected values of the plate thickness, the plate width, the temperature, and the tension deviation and the carbon equivalent are substituted into the second multiple regression equation to obtain each value. Determining the strip width of the hot-rolled steel strip at room temperature for each virtual equal division area. The method for predicting the strip width of a hot rolled steel strip according to claim 1. 6. A plate thickness detecting means arranged between a hot rolling mill and a winder for winding the hot rolled steel strip rolled by the hot rolling mill, for detecting the thickness of the hot rolled steel strip. , A strip width detecting means disposed between the hot rolling mill and the winder to detect the strip width of the hot rolled steel strip, and disposed between the hot rolling mill and the winder. A temperature detecting means for detecting the temperature of the strip, a tension detecting means arranged between the hot rolling mill and the winder for detecting the tension of the hot rolled steel strip, and a value for predetermining the tension of the hot rolled steel strip. Tension setting means, the carbon equivalent calculating means for calculating the carbon equivalent, and the tension deviation detecting means for detecting the tension deviation between the detected tension of the tension detecting means and the preset tension set by the tension setting means. And a determination unit that determines the magnitude of the tension deviation and a predetermined value in response to the output of the tension deviation detection unit, Note The output of each detection means is sampled, the total length of the hot-rolled steel strip is determined, the hot-rolled steel strip is virtually equally divided in the longitudinal direction, and the minimum value of the detection values of the respective detection means in each virtual equal division area is obtained. , When an average value and a maximum value are respectively obtained, one of them is selected for each detected value, and the selected value of the tension deviation is determined to be less than a predetermined value by the determination means. Based on the outputs of the plate thickness detecting means, the plate width detecting means, the temperature detecting means and the carbon equivalent calculating means, the plate thickness of the hot rolled steel strip, the plate width, the temperature and the carbon equivalent of the hot rolled steel strip at room temperature are used as variables. When the plate width of the hot-rolled steel strip at room temperature is obtained using the first multiple regression equation for obtaining the plate width, and when the selected value of the tension deviation is determined to be equal to or greater than the predetermined value by the determination means, Thickness detection means, width detection means, temperature detection means Secondly, based on the outputs of the tension deviation detecting means and the carbon equivalent calculating means, the plate width of the hot rolled steel strip, the width, the temperature, the tension deviation and the carbon equivalent are used as variables to obtain the strip width of the hot rolled steel strip at room temperature. A hot-rolled steel strip characterized by including a calculation means for obtaining a strip width of the hot-rolled steel strip at room temperature using a regression equation and calculating a minimum strip width of the hot-rolled steel strip from the obtained strip width. Width prediction device.