JPS5911844B2 - Long material bend measuring device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is capable of performing curves on a plane of a long material, such as left and right bends of a section steel, on a path along which the long material is being transported in the longitudinal direction, with improved accuracy. Concerning how to measure.

For example, in the process of manufacturing section steel, the section steel is hot-rolled and becomes bent during the subsequent cooling process, so it is straightened through a bend straightening machine after cooling.

In order to automate this straightening work, the curvature on a plane, that is, the left and right curvature, is measured, and based on that information, the pitch of the straightening roller of the straightening machine and the position of the straightening roller in the vertical direction and the axial direction of the straightening roller are set. This is very important. Furthermore, even after correction, it is necessary to measure the left and right curvature in order to verify the curvature shape. This curve measurement is
Due to the nature of the work, this is done while the long material is being transported.

What you should be careful about here is that the long material may move laterally or skew during transport due to unevenness, backlash, or unevenness on the surface of the roller for transporting the long material, or due to the curve D of the long material itself. It is important to remove the influence of this meandering when measuring the meandering. In addition, especially in the case of product verification,
Measuring local bending is insufficient; it is necessary to understand the shape of the bending force over the entire length of the long material (full-length bending).

This is because, for example, as shown in Figures 12 and 13 below, acceptance or rejection is often determined by the magnitude of the top surface curvature, and this top surface curvature is the amount obtained from the full-length bending force. In addition, the curves that actually occur in long materials include not only simple arcuate curves, but also complex curves such as S-shape and W-shape, and the certification standards are set for each. However, regardless of the curved shape, the amount of top surface curve cannot be obtained unless the total length curve is known. Conventionally, methods for measuring bending of long materials have been disclosed, for example, in Japanese Utility Model Application Publication No. 48-23049 and U.S. Pat. No. 34950.
No. 90, but all of these proposals propose devices for measuring local bends, and do not measure the entire length bend.

In particular, in the latter case, when measuring the camber of the steel material, three detection devices are installed in the transport direction of the steel material to detect the camber of the steel material.
Since the side edges of the steel material are not detected at the same time, the bending measurement value is affected by the meandering of the steel due to the detection time lag, and the measurement accuracy is poor in measuring bending in actual operation transfer lines where meandering occurs frequently. to degrade. The present invention provides a method for measuring the curve of a long material that solves the problems of the conventional method described above. Short-pitch curve V1 that does not require detection To provide a method for measuring the bending of a long material, which eliminates detection errors caused by vibrations of an optical dimension detection device or vibrations of the long material during transportation, and accurately measures the entire length of the material. .

Its feature is that it uses three optical dimension detection devices installed at intervals in the long material transport direction to detect the deviation of the side edge of the long material from the measurement reference line set on the transport line plane. At the same time, a data sampling interval is set by measuring the transport length of the long material, and the deviation signal from the optical dimension detecting device and the installation interval of the optical dimension detecting device are detected for each data sampling interval. The unit length curve b is calculated from , and then the unit length curve b is moving averaged for each arbitrarily determined span, and this moving averaged unit length curve is calculated over the entire length of the long material to calculate the total length curve calculation section. A method for measuring the bending of a long piece of material by sequentially connecting it at intervals to obtain the total bending. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The method for measuring bending of a long material according to the present invention will be explained in detail below based on an embodiment thereof with reference to the drawings.

FIG. 1 is a layout diagram of a measuring device for measuring left-right bending of a section steel according to an embodiment of the present invention.

In the drawing, 3, 4, and 5 are three optical dimension detection devices installed at equal intervals in the transfer direction above the long material transfer line, and in this embodiment, they are photoelectric digital deviation detectors. This detects the deviation of one side edge of the long material 2 from the measurement reference line c defined on the transport plane. The optical dimension detection devices 3, 4, and 5 are not limited to digital systems, but may be analog systems. Reference numeral 1 denotes a line surface for transporting the long material 2 in the direction of the arrow.
Proceed upward in the direction of arrow X.

6, 7, 8 are illumination devices, and optical dimension detection devices 3,
4 and 5, respectively. In this embodiment, the emitting light is emitted from below the line surface 1, and the elongated material 2 on the line surface 1 is illuminated.
The light is received by the optical dimension detection devices 3, 4, and 5 as a dark part and the other part as a bright part.

In this embodiment, the optical dimension detecting devices 3, 4, and 5 are composed of a light receiving lens and a photoelectric element array, and therefore, the distances of the bright and dark regions correspond to the distances from the side edges of the field of view of the optical dimension detecting device. The number of pulses is output from the optical dimension detection device. That is, on the illumination devices 6, 7, and 8, the number of pulses corresponding to the distance from the side edge of the field of view to the end in the width direction of the elongated material is output from the optical dimension detection devices 3 to 5. Reference numerals 9, 10, and 11 are optical dimension detecting devices, which are video analyzers in this embodiment, and are paired with the optical dimension detecting devices 3, 4, and 5.As shown in FIG. 15 outputs the distance from the top surface to the bottom surface of the long material 2 as an analog quantity, for example, voltage height, inputs this to the input/output device 18, converts it to a digital quantity with an A-D converter, and outputs it as an analog quantity, for example, the voltage height. Enter it in

At that time, although not shown, the video analyzer 9,
Illumination is applied from the side of 10 and 11 to the long material 2 side, and the portion of the long material 2 is made into a bright part. The reference plane in the vertical direction is the conveyance rollers 14, 1.
5, and this reference surface is the top surface of the video analyzers 9 and 10.
. This is input to the arithmetic processing unit 19 and the vertical position of the long material 2 is grasped. Reference numerals 12 and 13 indicate pulse transmitters, which include a light emitting/receiving mechanism and a pulse transmitting mechanism that outputs pulses corresponding to the number of times of light reception.
, 15 side end faces, and receives the reflected light from the reflective tapes 21, 22. Upon receiving this light, it emits a number of pulses equal to the number of times the light is received.

These constitute the detection command device. The optical dimension detecting devices 3, 4, and 5 are activated based on a signal indicating the length of the long material to be transferred from the detection command device. A control device 16 switches the start signals of the optical dimension detecting devices 3, 4, and 5, and converts the output pulses from the optical dimension detecting devices 3, 4, and 5 into 4-bit BCD (Binary COded Decimal).

18 is an input/output device, which receives signals from the control device 16;
The outputs from the video analyzers 9, 10, 11 and the signals from the operation panel 17 are delivered to the arithmetic processing device 19, and the output (information) from the arithmetic processing device 19 is transferred to the operation panel 17.
and transmits signals to the control device 16.

The arithmetic processing unit 19 receives a signal from the input/output device 18,
Various calculation processes, i.e., calculation of the distance from the reference line at the three points of the optical dimension detection devices 3, 4, and 5 of the long material 2) (including correction calculations based on the vertical deviation of the long material) )
19-1, Calculation of the curve V per unit length of the long material 2 (hereinafter referred to as unit long bend) 19-2, Move for each arbitrarily determined unit long bend b of the long material 2 Average (weighted average in this embodiment), smoothing calculation 19-3 of the detected deviation value, total length bending of the long material 2, and large bend D calculation 19-4 are performed.

Reference numeral 20 denotes a printing device that outputs music information from the arithmetic processing device 19.

Reference numeral 23 denotes a display device that displays the total length bend and the amount of large bend b.

Next, the operation of the method for measuring curve b of a long material according to the present invention will be explained.

First, the optical dimension detection devices 3, 4, 5 are installed in the arithmetic processing device 19.
and optical dimension detection device (video analyzer) 9,10
, 11 scale factors and the data sampling interval ΔX1, which is controlled by the detection command device (full length song), the calculated interval interval Δk, and the measurement interval t of the optical dimension detection devices 3, 4, and 5.
is memorized through the operation panel 17, and a reference line (straight line) is drawn in the transport direction of the long material 2 using an enameled wire or steel wire so as to be in contact with the rolls 14, 15 on the line surface 1 by the optical dimension detection devices 3, 4.
, 5, and calculate the plane position of this reference line perpendicular to the long material conveying direction as P, , P2, P3 using the number of pulses of the photoelectric element array of the optical dimension detecting devices 3, 4, 5. It is stored in the device 19.

This reference line does not necessarily have to be stretched parallel to the direction in which the long material 2 is conveyed. Here, the optical dimension detection device installation interval t1, total length curve calculation interval interval Δk1, and data sampling interval Δx will be explained.

First, regarding the optical dimension detection device installation interval t, the smaller t is, the more curves that change in a shorter period can be detected, which is preferable in the sense that the error in detecting curves itself can be reduced, but on the other hand, it is unnecessary. The value of the optical dimension detection device installation interval t is determined from two aspects: the detection period of the curve and the removal of disturbance, and is generally determined by the shortest curve of the curve to be detected. 1/ of the period
2 to 1/5 is desirable. For example, in the case of the above-mentioned steel section, if you want to effectively detect the W-shaped curve of a section steel with a total length of 15 [m], the curve has two periods in its entire length, so the length of one period is 7.5 m. [With MX, optical dimension detection device interval t is 1.5 to 4 [77L? It is very appropriate. Since this t cannot be easily changed, it is necessary to select it in consideration of all the conditions of the long material passing through the line. Next, regarding the interval Δk for calculating the full-length bend b, as described later, the full-length bend is obtained as coordinates for the longitudinal position over the entire length of the long material, but it is not necessary to obtain these coordinates continuously in the longitudinal direction, but intermittently at appropriate intervals. This interval is set as the total length bend calculation section interval Δk. The appropriate value of Δk is, for example, about 1/10 to 1/30 of the total length of the section steel in order to automate the straightening of left and right bends in the section steel mentioned above.
For testing purposes, it is a little more dense than this, but it is appropriate. Therefore, when the length of the section steel is 15CTn], a suitable value of Δk is 0.5 to 1.5 [m]. Next, regarding the data sampling interval ΔX, one of the requirements of the present invention is
It is related to the moving average.

In other words, in order to make the moving average effective, it is necessary to sample data more often than the frequency determined by the total length song b calculation interval interval Δk, for example Δk=0.5T! L
In this case, set Δx = 0.1, etc., sample 5 pieces of data during one section of 0.5 m to find the coordinates of the total length curve, smooth them using a moving average that will be explained in detail later, and calculate the result. As a result, the accuracy of determining the total length bending is increased. Therefore, the calculation of the unit long bend b is performed at every data sampling interval Δx, but the moving average of the unit long bend can be calculated at every total length bend calculation interval Δk. When the tip is moved in the direction of arrow
. measures the deviations of these visual fields from the side edges, and outputs them from the control device 16 to the arithmetic processing device 19 through the input/output device 18.

The calculation of the deviation of the side edge of the elongated material 2 from the reference line is performed by the arithmetic processing unit 19 as follows. FIG. 2 is a sectional view perpendicular to the direction in which the elongated material 2 is transported, showing the relationship between any one of the optical dimension detection devices 3, 4, and 5 and the elongated material 2.

In the figure, the AB plane is the upper surface (level) of the conveyance rollers 14 and 15. Point A and point B are optical dimension detection devices 3, 4,
These are both ends of the visual field of 5. Point C is the point where the reference straight line common to the three optical dimension detection devices 3, 4, and 5 (which does not need to be parallel to the transport direction of the long material 2 as described above) intersects A-B, Point D is the point where the side edge of the long material 2 intersects A-B. The optical dimension detection devices 3, 4, and 5 measure the length of A-D (
The number of pulses P corresponding to sy) is output. Full view RV
Let the number of pulses corresponding to c be R, and measure the length s of A-C and the number of pulses Q corresponding to it using an oven C. At this time, y is determined by the following formula. In the transport direction of the long material, the optical dimension detection devices 3, 4, and 5 are numbered /F6l, 2, and 3, and the deviations in the respective optical dimension detection devices 3, 4, and 5 are designated as Yl, y2, and y3. .

The long material 2 is transported in the direction of the arrow X in FIG.
The optical dimension detecting devices 3, 4, and 5 measure the deviation on a plane perpendicular to the transport direction at intervals of [m]. corner (required sampling interval, for example 0.
A method is used in which a detection command device is attached that emits a pulse every time it rotates (every 1 m), and the timing is obtained from the pulse signal and sampling is performed. The symbol representing the sampling time point is n, and the transport distance XCm after the tip of the long material 2 passes through the optical dimension detection device 4 has the relationship x=nΔd. Wash 1, 2 and 3 calculated simultaneously at a certain point n
The deviation of the long material 2 in the optical dimension detection device is yl! 1,
It is expressed as 脣, y go. The unit length curve of the long material 2 is calculated by the arithmetic processing device 19 as follows.

In this example, the unit long curve v at a certain point is the deviation of that point from a straight line when two points in front and back that are 0.5 [m] away from that point are connected with a straight line, as shown in Figure 3. It means q.

The installation interval of optical dimension detection devices 3, 4, and 5 is 0.5 [m]
In general, it can be any Tcm]. The deviation detected by the optical dimension detector 1, 2, 3 at a certain time point n is determined by
6, y product, y group, these are arbitrary reference straight lines RL
It has become a deviation from the At this time, the unit length curve V)Yn is calculated by the following formula. The coefficient in equation (2) above is 1/(
2t)2. As shown in FIG. 5, the long material 2 is bent with a constant curvature over TCm] before and after the point under consideration.

In other words, if it forms an arc of a circle with center 01 and radius R, then νl▼〜＾▼−′ Here, from the above two equations (3) and (4) while assuming that R〉Y, Y′, Eliminate R to obtain ~f-Vl.

In other words, from equation (3), t′ Because this is R〉Y From equation (4) Because this is R〉Y get.

From FIG. 4, Y is clear.

Obtain accordingly.

As described above, the unit long curve b is formed between the tip of the long material 2 passing the A3 optical dimension detection device point and the tail end passing the waterfall 1 optical dimension detection device point, that is, the head Tcm],
It is calculated at every sampling interval ΔXCm] over the entire length of the long material excluding the tail Tcm].

When the long material 2 has a deviation in the vertical direction, as shown in FIG. This is accompanied by an error.

FIG. 14 corresponds to FIG. 2, and the elongated material 2 is in a position where it is raised by a distance Z above the upper surface level of the conveying rollers 14, 15. As in Fig. 2, points A and B are optical dimension detection devices 3 and 4.
, 5, and the point 1C is the position of the reference line.
F is the position of the side edge of the long material 2, point D is the intersection of the perpendicular line drawn from point F to the upper surface level of the conveying rollers 14 and 15 with line A-B, and point E is the optical dimension detection device 3, 4, 5. This is the intersection of the line A-B and the straight line connecting the position φ and point F.

Point H is from position φ of optical dimension detection devices 3, 4, and 5 to A-B.
This is the intersection of the perpendicular line drawn to the line with the A-B line.
As described above, in this state, when the deviation of the long material 2 is calculated using equation (1), the length y of the CE is obtained as the deviation.

However, the true deviation is the CD length y. Unless correction is made for the vertical deviation of the long material, a measurement error of y'-y will occur. Once the vertical position Z of the long material 2 is detected, the true horizontal deviation y is calculated using the following equation.

This equation (3) is expressed as (Z/h){(r/R,
). A correction term P−q} is added. H
, q, r, and s are measured in advance. In this example,
As shown in FIG. 1, the vertical deviation amount Z of the long material 2 is detected by optical dimension detection devices (video analyzers) 9, 10, and 11, and this is input to the input/output device 18 as an analog value. It is converted into a digital quantity by the AD converter in the output device 18 and inputted to the arithmetic processing unit 19, and the horizontal displacement amount y of the elongated material 2 is calculated by equation (8).

As already mentioned, the unit long piece VYn is obtained from y ancient, YA, and yλ obtained in this way.

The unit length bend Yn has the following errors: (1) short pitch bends that the long material 2 has and does not need to be detected; (Ii) scales on the side edges of the long material 2. , errors due to adhesion of water, etc., and unevenness of the side edges themselves regardless of bending; (11i) optical dimension detection device 3;
The vibration ranges 4 and 5 include errors caused by vibrations of the long material during conveyance.

Therefore, in the long material bend measuring device according to the present invention, the unit long bend BYn measured and calculated at each data sampling interval Δx is smoothed by moving averaged for each arbitrarily determined span, and the influence of the error is smoothed. I'm trying to reduce that. In other words, a digital filter is used. The above-mentioned smoothing can be done by a simple average filter, or a weighted average filter, where m is the moving average span.

In this embodiment, a weighted average filter type 01 is used.
To explain this smoothing effect, that is, the effect of the filter, first, the frequency characteristics of the filter are shown in FIG. This is when a weighted average filter is used with the moving average span m = 10. It shows the gain of the filter when changing p of the sinusoidal bending force expressed by the pitch of the filter.

In this figure, the horizontal axis is the ratio of P to the sampling interval ΔX, and the vertical axis is the gain. As is clear from this figure, songs with a pitch of P/ΔX of 10 or less are P
It can be seen that the attenuation is almost zero except when /Δx=1. To further explain this effect in relation to the error αi) above, let us assume that the long material 2 has a sinusoidal bend as shown by formula I. A = for simplicity
1 [MO.

(2) What is the sampling using formula? It can be seen that the unit long curve BYn at point n becomes, and the unit long curve D also has a sine wave shape. The amplitude of this sine wave varies depending on the distance t between the optical dimension detecting devices 3, 4, and 5 and the pitch P of the music D, as can be seen from the equation (self).

As an example, the amplitude in the case of t=3 [m] and Δx=0.1 [m] is shown in FIG. 9, with the horizontal axis being the bending pitch P. The same figure also shows the frequency characteristics of the amplitude when a weighted average filter is added when the value is m-10. As is clear from this figure, if we want to detect a bend with a pitch of 3 [m] or more due to the requirements for using a bend straightening machine, on the other hand, looking at the amplitude characteristics before weighted averaging, the pitch is 2 [m]. Although the front and rear bends are detected as bends, this is unnecessary and also causes an error. On the other hand, looking at the amplitude characteristics smoothed by weighted averaging, it can be seen that the unnecessary pitch waves are attenuated to less than half. The characteristics of the weighted average can be adjusted by the sampling interval Δx and the moving average span m. A simple average (arithmetic average) filter can also be used as a digital filter, and although its characteristics are lower than that of a weighted average filter, it is useful because it is easier to operate. Now, to explain the relationship between unit length bending and curvature of the long material 2, it is 0.5 [m] before and after the point we are currently considering.
], assuming that the curvature of the long material is constant and that part of the long material forms an arc of a circle with center φ and radius R as shown in FIG. 7, then in ΔφCD, It will be explained that when the long material 2 meanderes on the transfer line, there is no effect on the horizontal displacement measurement.

As shown in FIG. 6, the long material 2 is moved from the initial position I- to
At a certain point n, translate by Un and rotate at an angle θn (/F6.l, 3 Vn in the optical dimension detection device)
Suppose that the position and direction are , which corresponds to , and . If the deviation amounts when there is no meandering are y old, y2, and y, then the optical dimension detection device when there is meandering. V). 1,
The measured values in 2 and 3 are respectively.

Calculating the unit length bending radius from this measured value using equation (2) gives ~bl/no.
1, and the amount due to meandering is completely erased.

Therefore, even if the long material 2 is transported in a meandering state,
songs) can be measured without being affected by this. As described above, the tail end of the long material 2 is connected to the waterfall 1 and the optical dimension detection device 3.
The unit long curve YCx] is calculated at every sampling interval ΔXCm until the end of passing through.

The unit long curve v smoothed value Y is determined by moving average for each interval ΔXcTn]. However, Y does not necessarily need to be determined at every sampling point of every ΔXCm, and may be determined every ΔKCm for calculating the total length curve. Here, let the unit length curve smoothing value when x = t be Y, and hereafter the unit length curve smoothing values obtained for each ΔKCm] are Y2, Y3,...
・Suppose. Among Yl, Y2, . . ., those near both ends for which a smoothed value cannot be obtained are substituted with unit length bending. After the tail end of the long material 2 has passed through the cleaning 1 optical dimension detection device 3, using the unit length bending smoothing values Yl, Y2, Y3, . . . stored in the arithmetic processing device 19, The total length bend is calculated, and the method will be explained below. first,
The curve amount Yi of 2Δk [m] is calculated from the unit length curve smoothed values Yl, Y2, Y3, . . . (hereinafter generally referred to as Yi).

If we note that the unit length bend is 1 [m] (equal), then by the same way of thinking as the coefficient in equation (2) is 1/(2t)2, -
−F
The relationship aA~ is derived.

As shown in Figure 11, the total length of the song is 2∆KCm]. It is found by overlapping the triangles formed by i.

? A triangle formed by is a straight line connecting two points separated by a distance of 2ΔKC77L (when drawn in the diagram, the one on the left side will be called the left end point, and the one on the right side will be called the right end point) and the third point.
The distance between the point (this will be called the center point) is Y'1
What is the superposition of one triangle that is a triangle formed by three points such as ? A method is used in which the left end point and center point of the triangle created by 7 and +1 are superimposed on the center point and right end point of the triangle created by i, respectively. At this time, as shown in Figure 11, it is necessary to know the distance Ai between the right end point and the straight line passing through the left end point and the center point of the triangle, but it is obvious. 〇 Find the total length curve by superimposing the above triangles. Sometimes, as shown in Figure 15, TCm from both ends of the long material]
If the coordinates of the two points are 0, then the total length curve b coordinate F(0)
, F(1), . . . , F(K) are obtained by the following iterative calculation.

Here, K is the total length of the long material (LCm) (amount cut off to be a multiple of Δk), then K = (L-2t)
It is clear from FIG. 15 that the amount is /Δk+2. First, for F(0) and F(1) near the tip, F(
2)? For F(3),...,F(K-2), we use the quantity G(1) for convenience of calculation, and find it by repeating the following two equations, Furthermore, regarding F (dai), 1 gate-'l▼Tomoetenj: lie no-. -~ is.

(]8> The initial value of G in the formula is And it is sufficient.

With the above, from the position t-ΔKCm] of the long material to t+(K-
1) Each coordinate up to Δk was determined.

Furthermore, the coordinates of the tip and tail of the long material are determined as follows. Δk for the tip and then Δk for the tail.

Here, the coordinates of the leading edge and the tailing edge are F(1-ite), F(K
-1+4v). The coordinates over the entire length of the long material 2 are determined by the above calculation.

This series of coordinates is called a total length curve. In order to obtain the large curve, the coordinates of the total length curve obtained above may be transformed so that the coordinates become zero at both ends of the long material 2, and the maximum value thereof may be obtained. Omagari b is shown in FIGS. 12 and 13. The bending information obtained as described above is outputted to the printing device 20 based on the signal from the operation panel 17, and the total length bending and large bending amount are displayed on the display device 23 through the input/output device 18.

Based on this information, it is also possible to automatically set, for example, a roller straightening machine for section steel.

In this embodiment, the distance in the conveying direction of the long material 2 is detected by the rotation angle of the conveyance rollers 14 and 15, but the present invention is not limited to this. Of course, it is also possible to determine the transfer distance of 2 and output the detection command signal based on this.

Moreover, it goes without saying that the sampling interval Δx can be set arbitrarily. The method for measuring bending of a long material according to the present invention is constructed and operated as described above.
The amount of bending of a long material can be detected automatically and digitally with high accuracy, and there is no need for personnel to directly measure left and right bends, which has the effect of completely eliminating dangerous work. Indirectly, for example, automatic control of a bend straightening machine for shaped steel is enabled, which improves straightening accuracy, thereby reducing re-straightening work, resulting in improvements in yield, production efficiency, and quality. In addition, in the method for measuring the bending of long materials according to the present invention, smoothing by a moving average (digital filter) is performed when calculating the unit long bending, so that errors occur as shown in Fig. 10. It has the advantage of being extremely small.

In this figure, the shaded graph is for the case without a filter.
=36, mean value is -0.00mm 1 standard deviation is 0.05
m! , The graph with diagonal lines is for filter adjustment and n=
36, average value is 0.017nT111 standard deviation is 0.0
It is 3mm. Furthermore, since the plane deviation is corrected for the vertical deviation of the elongated material, there is an advantage that the error caused by this plane of the unit elongated curve becomes extremely small as shown in FIG. 16. In this figure, the hatched graph shows the case without vertical correction, n=34, the average value is 0.05 mm, 1 standard deviation is 0.18, and the hatched graph shows the case with vertical correction, n=34, average The value is -0.03 mm and the standard deviation is 0.08.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an embodiment of the method for measuring the curve b of a long material according to the present invention, and FIG. 3 is a diagram explaining the unit length bending of a long material, FIG. 4 is an explanatory diagram illustrating the deviation amount of the long material in the reference line 7 and the optical dimension detection device, and FIG. 5 is a diagram explaining the unit length bending of a long material. An explanatory diagram of the calculation formula for the unit length bending length of the length material,
Fig. 6 is an explanatory diagram of the influence of meandering of a long material on deviation measurement, Fig. 7 is an explanatory diagram of the relationship between the unit length curve and curvature of a long material, and Fig. 8 is an explanatory diagram of the unit length of a long material. A diagram showing the weighted average dynamic characteristics when taking a moving average of bends. Figure 9 is a diagram showing the removal of measured values of short pitch bends v by weighted averaging of unit long bends of long materials. FIG. 11 is a graph showing the error reduction due to the weighted average of unit length bends of the material; FIG. Figure 13 and Figure 13 are diagrams explaining the amount of large bending, Figure 14 is a diagram explaining the error in horizontal displacement measurement due to vertical displacement of a long material, and Figure 15 is a diagram explaining the total length bending force from a unit length bending. Explanatory diagram when calculating, 1st
FIG. 6 is a diagram showing the measurement error distribution after removing and correcting the horizontal deviation measurement error due to the vertical deviation of the long material. In the drawing, 2 is the long material, 1 is the transfer route, C is the measurement reference line, and 3
~5 is an optical dimension detection device, 12 and 13 are detection command devices,
19 is an arithmetic processing unit, 20 is a printing device, 23 is a display device, and 9 to 11 are optical dimension detection devices.

Claims (1)

[Claims] 1 Three optical dimension detection devices installed at intervals in the long material transport direction simultaneously detect the deviation of the side edges of the long material from the measurement reference line set on the transport line surface, and A data sampling interval is determined by measuring the moving length of the length of the length every time, and a unit length bending is calculated from the deviation signal from the optical dimension detecting device and the installation interval of the optical dimension detecting device for each data sampling interval. Then, the unit length bend is moving averaged for each arbitrarily determined span, and the moving averaged unit length bend is sequentially connected at each full length bend calculation interval interval over the entire length of the long material to obtain the full length bend. A method for measuring the bending of long materials.