JP5245817B2 - Steel plate shape measuring method and shape measuring device - Google Patents

Description

  The present invention relates to a steel plate shape measuring method and a shape measuring device, and is particularly suitable for measuring shapes such as warpage of a steel plate.

As an apparatus for automatically measuring the shape of a steel plate, for example, as described in Patent Document 1 below, a measuring device composed of a plurality of optical distance meters is installed on a steel sheet conveyance line, and the steel plate passes through this measuring device. The distance from the light reflection state to the steel plate surface, that is, the height of the steel plate surface, is detected, and the shape of the steel plate surface is measured continuously with this height.
JP-A-5-237546

  In the manufacture of a steel sheet, generally, a plurality of rolls called cold levelers and hot levelers are arranged one above the other, and the steel sheet is conveyed between these rolls, thereby correcting shape defects such as warpage generated during the manufacture. However, in the case of a steel plate with a thickness of 4 mm or more, which is generally called a thick material, the bending moment required to correct the shape is proportional to the cube of the plate thickness. Absent. Therefore, when a shape defect occurs in a thick material, the steel plate is removed from the line and the shape correction is performed off-line.

Since the steel sheet cannot be stably conveyed offline, the shape measuring apparatus as in Patent Document 1 cannot be applied as it is. In addition, even if it can be applied, the measuring device consisting of a plurality of optical rangefinders is complicated in configuration and requires a gate-type frame to be installed above the passing steel plate. It is disadvantageous.
The present invention has been made paying attention to the problems as described above, and has a simple configuration and a shape measurement method and shape measurement of a steel plate capable of easily and accurately measuring a shape even with a stationary steel plate. The object is to provide an apparatus.

In order to solve the above problem, the shape measuring method of the steel sheet of the present invention, the laser light source is polarized direction, by scanning the laser beam polarized direction, measures the predetermined detection point on stationary steel, measured a method of measuring the detection point group data the shape of the steel sheets, and characterized by performing the arithmetic processing to analyze a regression curved surface from the decimation process and the detection point group data to uniform the density points of the detection point group To do.

The laser shape measuring apparatus of the steel sheet of the present invention, the laser light source is polarized direction, and the laser beam irradiating means for scanning polarization direction has been the laser beam on the steel sheet stationary, which is polarized towards scanned from the laser beam irradiation means Data extraction means for extracting data of a predetermined detection point group on the steel plate stationary by light, thinning processing means for equalizing the point density of the detection point group on the extracted steel plate, and on the thinned steel plate An arithmetic processing means for analyzing a regression surface from the detected point group data and a shape measurement means for measuring the shape of the steel plate from the analyzed regression surface are provided.

The steel plate shape measuring apparatus of the present invention is characterized in that the thinning processing unit sets a thinning amount based on distance information between the laser beam irradiation unit and the steel plate.
Further, the shape measuring apparatus for a steel sheet according to the present invention comprises a correcting means for correcting the coordinates and angle of a predetermined detection group on the steel sheet extracted by the data extracting means, and a predetermined value on the steel sheet corrected by the correcting means. An area selection means for removing the corner of the steel plate from the detection group is provided.

And Thus, according to the steel plate shape measurement method of the present invention, the laser light source is polarized direction, by scanning the polarized direction laser beam, a predetermined detection point on stationary steel was measured and the measured when measuring the shape of the steel plate from the detection point group data, by performing computation processing for analyzing a regression curved surface from the decimation process and the detection point group data to uniform the density points of the detection point group, the shape of the stationary steel Can be measured easily and accurately, and the configuration is simplified because only one laser light source is required.

Further, according to the shape measuring apparatus of the steel sheet of the present invention, the laser light source is polarized direction, by scanning the polarized direction laser beam on the steel sheet stationary and the data of the predetermined detection point group on the stationary steel Extraction is performed, thinning processing is performed to equalize the point density of the detected detection point cloud on the steel plate, the regression surface is analyzed from the detection point cloud data on the thinned steel plate, and the steel plate is analyzed from the analyzed regression curved surface. By measuring the shape, the shape of a stationary steel plate can be measured easily and accurately, and the configuration is simplified because only one laser light source is required.

Further, according to the steel plate shape measuring apparatus of the present invention, the point density of the detection point group on the steel plate can be easily and accurately set by setting the thinning amount based on the distance information between the laser beam irradiation means and the steel plate. It can be made uniform.
Further, according to the steel plate shape measuring apparatus of the present invention, the coordinates and angle of the predetermined detection group on the extracted steel plate are corrected, and the corner of the steel plate is removed from the predetermined detection group on the corrected steel plate. Thus, the shape of the steel sheet can be measured easily and accurately.

Hereinafter, a steel plate shape measuring apparatus according to an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a schematic overall view of a shape correcting device using the steel plate shape measuring device of the present embodiment. Reference numeral 1 in the drawing denotes a press machine that corrects the shape of the steel sheet S. An entrance bed 3 is disposed on the entry side of the press machine 1, and an exit bed 4 is disposed on the exit side of the press machine 1. . Each of the beds 3 and 4 is provided with a large number of rollers for transporting the steel sheet S, and the transport state of the steel sheet S can be controlled by controlling the rotation state of the rollers. In the case of the press machine 1 of the present embodiment, the steel plate S is pressurized from above with the pressurizing ram 2, and a bending moment is mainly applied to the steel plate S to correct the shape of the steel plate. The shape of the steel plate S is measured by a steel plate shape measuring device described later. The parameters for correcting the shape of the steel sheet include, for example, the curvature obtained from the shape of the steel sheet S, the pressure applied by the pressure ram 2, the position and interval of the laying bar called shim, the position of the steel sheet S, that is, the position of the steel sheet S by the beds 3 and 4. Examples include a transport state. In the steel plate shape correction by the press machine 1 of the present embodiment, two shims are laid under the steel plate S, and the steel plate S in the portion between the shims is pressed with the pressure ram 2. The bending moment due to the pressure ram 2 is generated only in the steel plate S in the portion between the shims. The above-described various parameters are adjusted in consideration of the deformation amount of the steel sheet S due to this bending moment and the return amount when pressure is released, the so-called springback amount.

  On the side of the entrance bed 3, a shape measuring device 5 and a control device 6 were installed. Among these, the shape measuring device 5 is a laser distance meter that detects the distance to the detection point with laser light. As a method for detecting a distance with a laser rangefinder, there are a known phase difference method and a Time of Flight method. The phase difference method is a method of irradiating an object with intensity-modulated laser light, receiving the reflected light with a light receiving sensor, and calculating the distance from the phase difference between the oscillated laser light and the received reflected light. The Time of Flight method is a method for calculating the distance from the time difference between the incident wave and the reflected wave of the laser light and the speed of the light. By measuring the distance to each detection point of the steel sheet S with a laser distance meter, the shape of the steel sheet S can be obtained as described later.

As shown in FIG. 2, the laser distance meter of the present embodiment has a laser light source 11 mounted on a turntable 12, and a known galvanometer mirror 13 is disposed at the laser emission port of the laser light source 11. The rotation center of the turntable 12 coincides with the laser light from the laser emission port of the laser light 11, and the rotation axis of the galvano mirror 13 is orthogonal to the rotation axis of the turntable 12. In the present embodiment, by rotating the galvanometer mirror 13, the laser light from the laser light source 11, and the polarization direction primarily longitudinal direction of the steel sheet S, i.e. in the transport direction of the steel sheet S in FIG. 1, to rotate the turntable 12 it allows the laser light to be polarized towards the galvanometer mirror 13 scans primarily the width direction of the steel sheet S, i.e. in the transport direction and orthogonal direction of the steel sheet S in FIG.

  The control device 6 is constructed with a computer system such as a host computer, and controls the operating state of the press machine 1 and the beds 3 and 4 or measures the shape of the steel sheet S by performing, for example, arithmetic processing described later. . Of course, according to the measured shape of the steel sheet S, for example, the press machine 1 and the beds 3 and 4 can be automatically controlled so that the shape of the steel sheet S is corrected.

Next, calculation processing for measuring the shape of the steel sheet S performed by the control device 6 will be described with reference to the flowchart of FIG. This calculation process is performed at the same time for example the shape measurement start command of the steel sheet and, first in step S1, and the object of the predetermined region containing the steel sheet S, so as to be biased toward-scanning a laser beam to the object, the detection point So-called scanning is performed to read distance information.
Next, the process proceeds to step S2, and the data of the portion of the steel sheet S whose shape is measured is extracted from the distance data of the scanned object (predetermined region). For example, when the steel sheet S is in a state as shown in FIG. 1, this data extraction can be performed by determining that the distance data continuously changing over a long range in the transport direction is the data of the steel sheet S.

Next, the process proceeds to step S3, and the extracted distance data of the steel sheet is converted into a coordinate system in which, for example, the longitudinal direction of the steel sheet is the x axis, the width direction is the y axis, and the height direction is the z axis, with angle conversion. to correct.
Next, the process proceeds to step S4, where region selection is performed to remove data of unstable corner portions (edge portions in the figure) of the extracted and corrected steel plate data.
Next, the process proceeds to step S5, where detection point thinning is performed so that the density of data detection points is uniform. However, if the total number of data detection points is small or the density of data detection points is uniform, step S5 may be omitted.

  Next, the process proceeds to step S6, where the shape of the steel sheet surface is regarded as a certain curved surface, and regression surface analysis is calculated so that the data of the detection points satisfy the curved surface. However, when handling the data of the detection points as they are, for example, when the calculation process such as the function process is not performed on the shape of the steel sheet surface in the subsequent process, step S6 may be omitted. However, since the detection point data is generally an aggregate of discrete points that are discontinuous and unequally spaced, it is preferable to perform arithmetic processing on a regression surface that interpolates the detection points in this processing because subsequent handling becomes easier.

Next, the process proceeds to step S7, where the regression surface obtained in step S6 is identified so that it can be evaluated as an actual steel sheet surface (virtual is made coincident with reality), that is, after the shape is measured, the process returns to the main program.
Next, the operation of the arithmetic processing in FIG. 3 will be described in detail using the detected data. FIG. 4a shows the distance data of the predetermined area (object) obtained in step S1 of the calculation process of FIG. 3 in terms of brightness. Among these, the distance data of the steel sheet is continuously visible in white in the middle in the left-right direction in the figure. As can be seen from the figure, this distance data is farther upward (or left and right) in the figure. Therefore, in step S2 of the calculation process of FIG. 3, the continuously changing distance data is extracted as the distance data of the steel plate.

Then, for example, as shown in FIG. 5, the longitudinal direction the laser light of the polarization direction of the steel plate, i.e. a rotation angle of the galvanometer mirror theta, scanning of the laser beam in the width direction of the steel sheet, i.e. the rotation angle of the turntable When the distance to the detection point is r, the vertical direction in FIG. 4a is the x coordinate, the horizontal direction is the y coordinate, and the paper direction is the z coordinate, the x coordinate X, y coordinate Y, and z coordinate Z of the detection point are , Respectively, is represented by the following formula (1). The detected point data subjected to coordinate conversion is shown in FIG.

  In step S3 of the arithmetic processing in FIG. 3, the detection point data thus coordinate-transformed is further defined as the longitudinal direction of the steel plate as x coordinate, the width direction of the steel plate as y coordinate, and the height direction of the steel plate as z coordinate. Convert to a coordinate system. For example, when the coordinate system after conversion is rotated by the rotation angle α with respect to the coordinate axis x-axis before conversion, the x-coordinate X ′, y-coordinate Y ′, and z-coordinate Z ′ after conversion are the following two formulas, respectively. Appears.

  Similarly, for example, when the coordinate system after conversion is rotated by the rotation angle α with respect to the coordinate axis y-axis before conversion, the x-coordinate X ′, y-coordinate Y ′, and z-coordinate Z ′ after conversion are respectively It is expressed by the following three formulas.

  Similarly, for example, when the coordinate system after conversion is rotated by the rotation angle α with respect to the coordinate axis z-axis before conversion, the x-coordinate X ′, y-coordinate Y ′, and z-coordinate Z ′ after conversion are respectively It is expressed by the following four formulas.

In addition, for example, when the coordinate systems after conversion with respect to the x-axis, y-axis, and z-axis before conversion are respectively translated by L _X , L _Y , and L _Z , the x-coordinate X ′ after conversion, The y coordinate Y ′ and the z coordinate Z ′ are expressed by the following five equations, respectively.

  The coordinate data of the x-axis and the y-axis converted using these conversion formulas are shown in FIG. 4c. Further, coordinate data taking the z-axis into account is shown in FIG. 6 as three-dimensional data. For example, as clearly shown in FIG. 4c, the left direction of the figure, that is, the side closer to the shape measuring apparatus has a higher point density of detection points, so-called dense, whereas the right direction of the figure, that is, the side far from the shape measuring apparatus is detected. The dot density is low, so-called rough. For example, when the distance data to the steel plate is detected every time the rotation angle θ of the galvanometer mirror rotates by a certain angle, as shown in FIG. 7, the point density of the detection points (unit length in the figure) is near the laser distance meter. The larger the number of points per unit) and the farther from the laser distance meter, the smaller the point density of the detection points. If the regression surface analysis is performed using all of this data, the number of data is too large and the calculation load and memory become large, and the regression surface may not be analyzed correctly. That is, the data obtained by the laser distance meter does not necessarily represent the detection point accurately, and the shape such as warpage occurring in the steel plate is not always uniform in the longitudinal direction, This is because there is a possibility that coarse detection point data may be disregarded with emphasis on dense detection point data.

Therefore, as shown in FIG. 7, the thinning process is performed in step S5 of the calculation process of FIG. 3 so that the point density of the detection points in the longitudinal direction of the steel sheet is uniform. The thinning amount is determined according to the distance between the shape measuring device, that is, the laser beam irradiation device and the detection point. As this distance information, a method of obtaining from the position of the steel plate and the size of the steel plate or a distance measured in advance with a laser distance meter may be used. For example, the longitudinal direction of the steel sheet to be processed is equally divided into i = 1 to n sections, the length in the steel sheet longitudinal direction of this section is the unit length Lp, and the number of detection points per unit length Lp is (N ) _I , where Np is the number of detection points per unit length Lp after processing, the interval (dN) _i for performing the thinning processing is expressed by the following six equations.

For example, when the processing is performed based on the unit length Lp in the longitudinal direction of the steel sheet, if i = 1 to n, the starting position (X ₁ ) _i of the detection point range of the steel sheet that performs the thinning process of each section is as follows: Appears in Equation 7.

Further, the end position (X ₂ ) _i of the detection point range of the steel sheet that performs the thinning process in each section is expressed by the following eight equations.

Therefore, the number of detection points (N) _{i for} each section per unit length Lp is expressed by the total number of detection points that satisfy the following formula (9).

For example, if the total number of detected points after the thinning process is Np, the number of detected points (Np) _{i for} each section after the thinning process is expressed by the following equation (10).

Accordingly, the interval (dN) _i for performing the thinning process may be adjusted for each section so as to satisfy the following expression (11).

  An example of the number of detection points per unit length after the thinning process is shown in Table 1 below, a histogram thereof is shown in FIG. 8, and as a result, a three-dimensional map of the thinned detection data is shown in FIG. From Table 1 below, it can be seen that the difference in the number of detected points, which was about 600 points before the thinning-out process, is reduced to a difference of several tens of points.

  In step S6 of the calculation process of FIG. 3, the regression surface analysis is performed on the detection point data in which the point density is uniformed by the thinning, assuming that the surface of the steel plate approximates a certain curved surface. The following 12 equations were used as the regression surface calculation method.

  Further, f in the equation is an interpolated curved surface f, which is an arbitrary curved surface having a constraint condition of passing through a data point shown in the following equation (13).

  Although there are an infinite number of interpolated curved surfaces that satisfy the constraint conditions of the equation 13, the thin plate spline curved surface is an optimal interpolated curved surface in the sense of minimizing the curvature. The functional of the thin plate spline curved surface f is expressed by the following equation (14).

  A thin plate spline curved surface can be obtained by applying a variational principle to the thin plate spline functional. When the variational principle is applied to the thin plate spline functional of the above 14 equations, the following 15 equations are obtained.

  From the equation (15), the differential equation of the thin plate spline curved surface is expressed by the following equation (16).

  The boundary condition of the thin plate spline curved surface is given by the following equation (17).

In general, the analytical solution of the differential equation of the equation 16 cannot be obtained under the boundary condition of the equation 17, but if the interpolation region Ω is infinite, the solution of the differential equation of the equation 16 is The following 18 thin plate spline curved surfaces are obtained with c ₁ , c ₂ , and c ₃ as unknown parameters.

The fourth term on the right side of the thin plate spline curved surface of Equation 18 is the sum of the Green functions. Further, equations 18-1 to 18-3 are constraint conditions.
The unknown parameter of the regression surface was obtained from the evaluation function based on the information amount. The number of parameters that minimizes the information amount criterion is the optimum number of parameters. There are various information amount standards, but the Akaike information criterion is known as the simplest. The evaluation function based on the Akaike information amount is given by the following equation (19).

  By combining the above 18 to 19 equations, it is possible to analyze a regression surface that interpolates detection points. FIG. 10 shows calculation results of regression surface analysis using the detection points of FIG. In the figure, both the x-axis and the y-axis are displayed as a 40-division rectangular mesh.

  Then, in step S7 of the calculation process of FIG. 3, using the analyzed regression surface, identification is performed to evaluate the actual shape of the steel sheet. FIG. 11 superimposes the height distribution in the z-axis direction of the detection points in the range of the width ± 50 mm in the central portion in the width direction of the steel sheet and the height distribution in the z-axis direction of the analyzed regression surface. As is clear from the figure, in this embodiment, an optimal regression surface can be obtained for the distribution in the height direction of the steel sheet. FIG. 12 is a plot of the height distribution divided into 40 in both the x-axis direction and the y-axis direction. From the figure, by analyzing the regression surface, it is possible to extract height information at an arbitrary position from the regression surface regardless of the position of the detection point.

Thus, according to the shape measuring apparatus of the steel sheet of the present embodiment, the laser light source is polarized direction scans polarization direction has been the laser beam on the steel sheet stationary and the predetermined detection point group on the stationary steel Extract the data, perform the thinning process to equalize the point density of the detected detection point cloud on the steel plate, analyze the regression surface from the detection point cloud data on the thinned steel plate, and analyze the regression surface By measuring the shape of the steel plate, the shape of the stationary steel plate can be measured easily and accurately, and the configuration is simplified because only one laser light source is required.

Also, by setting the amount of data to be thinned out based on the distance information between the laser beam irradiation device and the steel plate, the point density of the detection point group on the steel plate can be made uniform easily and accurately.
In addition, by correcting the coordinates and angle of the predetermined detection group on the extracted steel plate, and removing the corner of the steel plate from the predetermined detection group on the corrected steel plate, the shape of the steel plate can be measured easily and accurately. can do.

Further, by analyzing the regression surface, it is possible to extract not only the data of the detection point but also the shape data at an arbitrary position on the regression surface.
Further, since the interpolated regression curved surface is analyzed based on a plurality of detection point data, there is an effect that the analysis accuracy of the curved surface is improved as compared with the original detection point data.
Further, the regression surface analysis converts discrete data into a continuous function, which has the effect of improving the data utilization, such as differential processing and processing with analysis software.

  In the embodiment, the shape of a single steel plate is measured. However, for example, a plurality of steel plates can be arranged in the width direction, and the shapes of these steel plates can be measured simultaneously. And in that case, the shape measurement length in the width direction of the steel plate becomes long. For example, in the case of the embodiment, the difference in density of the measurement points in the rotation direction of the turntable becomes large. The detection point thinning process may be performed using the principle.

BRIEF DESCRIPTION OF THE DRAWINGS It is a schematic whole block diagram of the shape correction apparatus which shows one Embodiment of the shape measuring apparatus of the steel plate of this invention. It is a schematic block diagram of the laser irradiation apparatus which comprises the laser distance meter in the shape measuring apparatus of FIG. It is a flowchart which shows the arithmetic processing for the shape measurement of the steel plate performed within the control apparatus of FIG. It is explanatory drawing of the extraction of data by the arithmetic processing of FIG. 3, and coordinate system angle correction. It is explanatory drawing of the distance and rotation angle by the laser rangefinder of FIG. It is a three-dimensional map of data after extraction and coordinate system angle correction. It is explanatory drawing of the number of detection points by the laser distance meter of FIG. It is explanatory drawing of the thinning-out process by the arithmetic processing of FIG. It is a three-dimensional map of the thinned data. It is a three-dimensional map of the regression surface analyzed by the arithmetic processing of FIG. It is explanatory drawing which accumulated the shape of the steel plate width direction center part extracted from the regression surface of FIG. 10, and actual data. It is a three-dimensional map of the steel plate shape taken out from the regression surface of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Press machine 2 Pressurization ram 3 Incoming bed 4 Outgoing bed 5 Shape measuring device 6 Control device 11 Laser light source 12 Turntable 13 Galvano mirror

Claims (4)

A laser light source and polarization direction, by scanning the laser beam polarized direction, a method of the predetermined detection point on stationary steel was measured, to measure the shape of the steel plate from the measured detection point group data, shape measurement method of the steel sheet, wherein the thinning processing and the detection point group data to uniform the density points of the detection point group by carrying out calculation processing for analyzing a regression curved surface. A laser light source and polarization direction, the polarization direction has been the laser beam irradiation means for a laser beam is scanned on the steel plate stationary, a predetermined detection point group on the stationary steel by laser beam polarized direction scanning by the laser beam irradiation means Data extraction means for extracting the data, thinning processing means for equalizing the point density of the detected detection point cloud on the steel plate, and arithmetic processing for analyzing the regression surface from the detection point cloud data on the thinned steel plate A steel plate shape measuring apparatus comprising: means; and shape measuring means for measuring the shape of the steel plate from the analyzed regression surface.   The steel sheet shape measuring apparatus according to claim 2, wherein the thinning processing unit sets a thinning amount based on distance information between the laser beam irradiation unit and the steel plate.   Correction means for correcting the coordinates and angle of a predetermined detection group on the steel plate extracted by the data extraction means, and region selection for removing the corner of the steel plate from the predetermined detection group on the steel plate corrected by the correction means The shape measuring device for a steel sheet according to claim 2, further comprising: means.

Images