JP4762851B2 - Cross-sectional shape detection method and apparatus - Google Patents

Description

  The present invention relates to a method and an apparatus for detecting a cross-sectional shape of an inorganic material member such as a metal member, a wooden member, an organic member such as a plastic member, and a ceramic member, and more particularly, a cross-sectional shape in a curved portion of a member which is an inspection object. The present invention relates to a cross-sectional shape detection method and apparatus suitable for detection.

  For example, a UO steel pipe of a steel product, which is one of members having a curved portion as an object to be inspected, is manufactured by the steps (a) to (h) shown in FIG. That is, (a) after cutting the width direction end of the thick steel plate that becomes the groove in the subsequent welding process, (b) bend the end with a C press, and (c) be U-shaped with a U press After bending, (d) the end portions are butted together by an O press, and then (e) tack welding, (f) inner surface welding, (g) outer surface welding, and (h) finally the inner pipe The steel pipe is formed into a perfect circular pipe through a pipe expanding process.

  In a cross-sectional profile of UO steel pipe or the like (referring to the outer shape inside or outside the cross-section), the rise of the weld is called a weld bead (hereinafter referred to as “bead”) or extra build. Also, in the cross-sectional profile, the roundness around the toe, which is the boundary between the weld and the base metal, greatly affects the strength of the weld along with the width and height of the bead and how the bead swells around the bead. Therefore, it is important for quality control of the UO steel pipe to quantitatively grasp the roundness in each process of the production line. The deviation of the cross-sectional profile from the true arc around the toe is called peaking, and is conventionally measured using a dedicated jig using a dial gauge as disclosed in Patent Documents 1 and 2. Yes. In a process in which an automatic measuring device cannot be attached due to space or budget constraints, peaking measurement is often performed manually using the dedicated jig.

  An example of a peaking measurement method using a dedicated jig is shown in FIG. Using a jig that has two legs with a constant width W and a dial gauge needle in the middle (hereinafter also referred to as “peaking measurement jig”), the tip of the dial gauge needle is connected to the toe point P. The depth d = CP from the middle point C of the line segment AB to the toe point P is set to the depth of the line segment AB connecting the tip ends A and B of the peak of the peaking measuring jig. measure. d is a quantity having a sign, and d> 0 when the measurement point P is below the line segment AB.

FIG. 12 shows the positional relationship between the cross-sectional profile and a virtual perfect circle having a diameter D passing through the contact points A and B of the leg of the measurement jig on the profile. When the entire cross-sectional profile of the steel pipe is a perfect circle corresponding to the inner diameter D, the depth d ₀ = CP ₀ of the intermediate point P ₀ on the true arc AB with reference to the midpoint C of the chord AB of length W is It is given as equation (1).

Finally, the depth deviation Δd = d−d ₀ from the perfect circle is obtained as peaking which is one of the representative values representing the cross-sectional shape.

  Peaking measurement using the above-mentioned peaking measurement jig is a method that can be performed relatively easily and accurately in a short time without being a skilled worker. However, it still takes time and labor due to manual labor, so the measurement location is limited and intermittent measurement is performed, and it is easy to overlook subtle changes in the shape of the tube axis, and the measurement results vary due to the unevenness of the toe and base parts. There was a problem that.

  On the other hand, various methods and apparatuses for automatically detecting a cross-sectional shape have been proposed. Among them, an optical method is useful in that measurement can be performed without contact. As an optical method for measuring a cross-sectional profile of a steel pipe from a pixel position of a captured light source image in the image data by capturing a light source image irradiated on the inspected part, an optical cutting method or a patent There is a method using an automatic measuring device using a laser distance meter as disclosed in Document 3.

Japanese Utility Model Publication No. 6-49918 JP 59-112209 A JP 2001-201327 A JP 7-40049 A

  In an actual production line, it is not possible to use a large-scale automatic measuring device, and there are a mixture of processes for manually measuring peaking using the above-mentioned peaking measuring jig and processes for using an automatic measuring apparatus using an optical method. May have.

  When measuring and managing how the peaking value has changed between processes on a production line where both the manual measurement process and the automatic measurement process coexist, the peaking is manually measured with a dedicated jig. In consideration of the comparison with the peaking measurement value in the process, an optical measurement method for peaking that is at least as accurate as the measurement method using the dedicated jig and is consistent is required.

  However, conventionally, such an appropriate method has not been known. That is, the cross-sectional profile obtained by the optical method is actually composed of a plurality of discrete points. In order to find a discrete point corresponding to the position of the leg tip of the above peaking measuring jig, a set of discrete points that best fit the leg width and satisfy the condition that the straight line connecting the discrete points is perpendicular to the needle is used. Although it is necessary to find it exploratively, there is a problem that the cost of calculation time is high in processing the measurement data obtained by the optical method. For this reason, it is difficult to reduce the measurement interval in the tube axis direction, and a subtle shape change in the tube axis direction is often overlooked.

  The present invention was invented to solve the above-described problems of the prior art, and was measured on an inspected object having a curved portion such as a steel pipe by an optical means such as a light cutting method or a laser distance meter. It is a first object of the present invention to make it possible to derive peaking with high accuracy and low calculation time cost from a cross-sectional profile. It is a second object of the present invention to provide an automatic measuring method of peaking that is consistent with the peaking measurement value by the peaking measuring jig.

The cross-sectional shape detection method according to the first aspect of the present invention is the cross-sectional shape detection method for measuring peaking of the measurement point P on the surface of the bending portion using a member having the bending portion as an object to be inspected, along the surface of the bending portion. The bending portion constituted by a plurality of points by an optical method based on a projection pattern irradiated with linear light or a linear ray locus on the surface of the bending portion obtained by irradiating and scanning with point light. A first step of measuring the cross-sectional profile, a second step of setting a measurement point P on the cross-sectional profile, and a horizontal distance W / 2 from the measurement point P by a distance W / 2 that is half of the preset distance W. A third step of designating a first point and a second point on the cross-sectional profile separated from each other, within a predetermined first search range WL centered on the first point, and the second point A predetermined second search range centered at On each cross-sectional profile in R, a distance between two points from a plurality of discrete points constituting the cross-sectional profile or a set of continuous points on the function obtained by approximating the cross-sectional profile is the distance W. A fourth step for searching the closest points to the first and second endpoints respectively, and a line segment connecting the first and second endpoints as a reference line segment, and the first and second endpoints and the measurement A fifth step of calculating a distance from the reference line segment to the measurement point P based on the position of the point P and a peaking of the measurement point P based on a distance from the reference line segment to the measurement point P are calculated. And a sixth step.
According to a second aspect of the present invention, the optical method is a light cutting method.
The cross-sectional shape detection method of the third invention of the present application is characterized in that the optical method is a triangulation method.
A cross-sectional shape detection device of a fourth invention of the present application is a cross-sectional shape detection device that measures peaking of the measurement point P on the curved surface used in the cross-sectional shape detection method of the first or second invention, Projecting means for irradiating linear light along the curved surface or irradiating and scanning with dotted light to obtain a linear ray trajectory on the curved surface, and an image of the linear ray trajectory Imaging means for obtaining a cross-sectional profile measuring means for deriving a cross-sectional profile of the curved portion constituted by a plurality of points based on an image of the ray trajectory by a predetermined image processing, and each of the second to sixth items Signal processing control means for executing the steps.
A cross-sectional shape detection device of a fifth invention of the present application is a cross-sectional shape detection device that measures peaking of the measurement point P on the curved surface used in the cross-sectional shape detection method of the first or third invention, A light projecting unit that irradiates and scans with point light and obtains a linear ray locus on the surface of the curved portion, a position detection unit that measures the position of the linear ray locus, and a plurality of points. It comprises cross-sectional profile measuring means for deriving a cross-sectional profile of the curved portion, and signal processing control means for executing the second to sixth steps.

According to the present invention, in the peaking measurement of a test object such as a welded steel pipe having a curved portion, the distance AB between the contact points A and B between the leg of the peaking measurement jig and the cross-sectional profile is the leg width W of the jig. The line segment CP connecting the midpoint C of the AB and the measurement point P is equal to the constraint condition that the line segment CP is orthogonal to the AB, so that the automatic measurement is compatible with the manual measurement method using the peaking measurement jig. Measurement is possible.
In addition, according to the present invention, it becomes possible to measure the subtle shape change in the tube axis direction, which is often overlooked by intermittent manual measurement, over the entire length, which is reflected in the creation of the cross-sectional shape and quality control in each manufacturing process. As a result, quality assurance over the entire length can be realized. In addition, it is possible to identify the cause of the shape defect by tracking the matching between the data measured by the peaking measurement jig and the change in the shape between the processes.

Preferred embodiments of the present invention will be described below with reference to the accompanying drawings.
<First section profile derivation method>
FIG. 1 is a schematic diagram showing a configuration of an apparatus used in the first method for deriving a cross-sectional profile and an example of a welded steel pipe as an object to be inspected. Reference numeral 1 denotes a welded steel pipe (only a part of which is shown). Reference numeral 2 denotes light projecting means for creating a linear light projecting pattern on the inspection portion of the object to be inspected. Reference numeral 5 denotes an image pickup means for copying the projection pattern. Reference numeral 6 denotes an image processing device that performs image processing on captured image data to derive a cross-sectional profile. Reference numeral 7 denotes a signal processing control device for executing processing for calculating peaking based on cross-sectional profile data.

  FIG. 1 shows a state where the light projecting means 2 and the imaging means 5 are placed inside the welded steel pipe 1 to detect the cross-sectional shape of the inner surface. In this section profile derivation method, for convenience of calculation, the pipe axis direction of the welded steel pipe 1 of the object to be inspected is taken as the z axis, and the section perpendicular to the pipe axis direction is taken as the xy plane. The projection center direction coincides with the plane formed by the light projection pattern extending from the light projecting means 2, and the emission center direction of the light projection pattern is taken as the y axis. In addition, it is preferable in terms of profile measurement to arrange the beads so that they are close to the center of the projection pattern. The position of the origin O of the xyz coordinate is the intersection of the optical axis direction of the photographing means 5 and the xy plane as will be described later.

  FIG. 2 is a view of the optical arrangement in the first cross-sectional profile derivation method as seen from the x-axis direction. 5a is the imaging surface of the imaging means 5, Op is the lens center (front principal point position) of the imaging lens of the imaging means (hereinafter simply referred to as “lens”), and O is the object (object) side as described above. The origin of the coordinate axis. If the horizontal axis of the imaging surface 5a is the X axis and the axis perpendicular to the X axis in the imaging surface 5a is the Y axis, Oc is the center of the imaging surface 5a and the coordinate origin of the imaging surface 5a (image side) XY plane. . An extension line of a straight line connecting the image side coordinate origin Oc and the lens center Op passes through the object side coordinate origin O, and the object side x axis and the image side X axis are parallel to each other. Further, the optical axis direction OOc of the image pickup means (image pickup system) is arranged so as to form an angle θ with the y axis that is the optical axis direction of the light projecting means 1.

  Here, as the light projecting means 1, for example, a linear light source that converges light emitted from a laser or lamp into a linear light projection pattern at an irradiation position on the object to be inspected using a cylindrical lens, a prism, or the like. Use or scan a point beam that converges in a dot shape at the irradiation position across the bead of the welded steel pipe 1 with a rotation-controlled polygon mirror in the x-axis direction orthogonal to the tube axis direction, A scanning point light source that obtains a light projection pattern of a linear locus is used. When a linear light source is used as the light projecting means 1, reference numeral 4 in FIG. 1 shows a fan-shaped linear beam. When the curved portion of the welded steel pipe 1 that is an object to be inspected is irradiated, A cut image 3 is formed. When a scanning point light source is used as the light projecting means, 4 in FIG. 1 indicates a locus of a point beam scanned in the width direction, and 3 indicates a locus of a point image at an irradiation position on the welded steel pipe 1.

  As the image pickup means 5, for example, a CMOS camera or a CCD camera which is an area sensor is used to pick up a light cut image or a point image locus of a linear locus on the object to be inspected. In the image processing device 6 of FIG. 1, image data of a light section image or point image trajectory imaged from the image pickup means 5 is taken into the image processing device 6, and a light section image or point image (by image processing based on the image data). The pixel position where the image is captured (relative position information of each pixel in the image) is obtained. Then, the correlation between the position of the light projection pattern on the inspection object and the position of the image of the light projection pattern on the image data, which is determined by the optical arrangement of the light projection means 2, the imaging means 5, and the inspection object in FIG. A cross-sectional profile, which is a set of discrete measurement points at intervals in which the contour shape of the surface of the steel pipe including the bead is set in advance, is obtained from the constants indicating the above and the pixel position where the image appears. If the measurement interval is narrow, more accurate profile measurement can be performed, but the measurement time becomes longer and the amount of data increases. Therefore, when high-speed measurement is required, the measurement interval is set so as to satisfy the required measurement accuracy.

  A method for obtaining a constant indicating the correlation between the position of the projection pattern on the object to be inspected and the position of the image corresponding to the projection pattern on the image data will be described in detail below. From the imaging formula of the lens in geometric optics, if the distance from the lens center Op to the object side origin O in FIG. 2 is L and the focal length of the lens is f, the object side origin O is for the image captured on the imaging surface 5a. The magnification M of the object is expressed by M = (L−f) / f. However, at the point Q at the height of y from the object-side origin O in the y-axis direction, the magnification M ′ is M ′ = M · (Lf−y · cos θ) because it is close to the imaging surface side by y · cos θ. / (L-f).

  Considering the magnification on the plane perpendicular to the imaging plane passing through the point of height y for the point Q (x, y) on the xy plane and its corresponding image (X, Y) on the imaging plane XY, The following relational expression holds.

  If this is rearranged, a coordinate conversion formula from the image side (imaging surface side) to the object side (inspected object side) is obtained as in the following formulas (4) and (5).

  In addition to calculating peaking, the signal processing control device 7 also controls imaging conditions such as the irradiation time of the light projecting means 1 and the exposure time of the imaging means 5.

<Second section profile derivation method>
FIG. 3 is a schematic diagram illustrating an apparatus configuration in the second section profile derivation method in the section shape detection method and apparatus of the present invention, taking a welded steel pipe as an example to be inspected. In FIG. 3, reference numeral 5 'denotes a distance measuring means having a measurement principle similar to that of a laser distance meter in which a scanning point light source (for example, a laser light source, not shown) and a PSD (position detector, not shown) are integrated. The distance from the distance measuring means itself to the surface of the welded steel pipe 1 to be measured can be directly measured by the triangulation method. The xy coordinate system is the same as the first section profile derivation method. For example, when the welded steel pipe is placed horizontally with the bead down, the distance measuring means 5 ′ is installed vertically above the measurement location of the bead. The trajectory of the laser beam emitted from the scanning point light source in the distance measuring means 5 ′ scanned in the x-axis direction orthogonal to the tube axis direction is indicated by 4, and 3 is the scan of the irradiation point of the laser beam on the welded steel tube 1. Show the trajectory. Since the distance to the surface of the welded steel pipe 1 has already been calculated by the distance measuring means 5 ′, the image pickup means is unnecessary, and the signal processing control device 7 ′ does not need image processing for calculating the cross-sectional profile. That is, in the case of this method for deriving a cross-sectional profile, the signal processing control device 7 ′ takes in the distance data from the distance measuring means 5 ′, calculates the cross-sectional profile based on the distance data, calculates the cross-sectional shape described later, and The distance unit 5 'is controlled. As described above, a cross-sectional profile, which is a set of discrete measurement points at intervals in which the contour shape of the steel pipe surface including the beads is set in advance, is obtained.

  Based on the cross-sectional profiles derived by the first and second methods described above, a toe point that is a boundary between a steel pipe base material and a bead that is a measurement point is detected. Since the slope of the tangent line of the cross-sectional profile changes moderately at the toe, but the cross-sectional profile is differentiated, it has a peak at the toe. Actually, since the cross-sectional profile consists of discrete points and includes noise, as described in Patent Document 4, the method uses a combination of smoothing and differential operation, or smoothing adaptive differentiation such as the Savitzky-Golay method. Detect a toe point.

  Below, the peaking calculation procedure and calculation method at the toe point or the designated measurement point automatically detected from the cross-sectional profile of the welded steel pipe 1 will be described.

<First Peaking Detection Procedure>
FIG. 4 shows the relationship between the cross-sectional profile of the welded steel pipe 1 obtained by one of the optical methods described above and the discrete point sequence on the cross-sectional profile that is a set of actual measurement points. Ac and Bc indicate points on the cross-sectional profile at a distance of W / 2 in the horizontal direction from the measurement point P, respectively. As a first peaking detection procedure, first, an end point having a distance between two points close to W is searched from a discrete point sequence constituting a cross-sectional profile within the search ranges WL and WR centered on Ac and Bc, respectively. A method for obtaining the distance from the reference line segment connecting the end points to the measurement point P will be described.

_{Let the} cross-sectional profile coordinate point sequences in the search ranges WL and WR be [A ′ _n ] and [B ′ _n ], respectively. For the point sequence [A ' _n ], A' ₁ is the left neighbor of A ' ₀ , A' ₂ is the left neighbor of A ' ₁ , ... A' _n is the left neighbor of A ' _n-1 in terms of, 'for the _[n, B point sequence B]' ₁ is B _'0 right next to the, B' ₂ is B _'1 to the right, ···, B' _n is B _'n-1 In this way, the indexing is performed in the ascending order from the measurement point P to the far side. The equidistant points from the measurement point P found in the k-th search are represented by A _k and B _k .

  FIG. 5 is a flowchart of a procedure for finding the end points A and B of the reference line segment from the discrete cross-sectional profile point sequence measured in the search ranges WL and WR, and calculating the peaking depth d = CP at the measurement point P. Show.

  In step S101, first, the toe point that is the measurement point P is detected and designated by the method described above.

  In step S102, the search ranges WL and WR for the end points A and B whose coordinates are unknown are set around the two points Ac and Bc that are at a distance of W / 2 in the horizontal direction from the measurement point P, respectively.

In step S103, selecting a point A _k in the search range WL. The initial point A ₀ is a point A ′ ₀ closest to the measurement point P within the search range WL. 'In part point sequence of _{n, for example, the sequence of points {A sequence of points {A _k} is the sequence of points A}' when extracting the _n} from the point sequence {A _k} at intervals [Delta] n is a multiple of n of [Delta] n, That is, n = mΔn (m = 0, 1,...) And A _k is A ′ _n .

In step S104, the distance from the measurement point P to A _k is set to R _k = PA _k, and a discrete point B _k approximately PA _k = PB _k = R _{k is} obtained. 'Carries out the search from the _{n in ascending order of the index, from the measurement point P B point sequence in the right search range WR B}' distance PB _'n is the first PB' _n ≧ R _k points satisfy the B _'n to _{n Let} B _k . Search for B _k starts from the search end point B _k-1 corresponding to the previous A _k-1.

At step S105, step if the end points A _k, from the coordinates of B _k of the line segment A _k B _k the length W _k calculated, Ashihaba W or more peaking measuring jig as a target, i.e. becomes W _k ≧ W It is determined to exit the loop from S103 to S104. As described above, two end points near the distance W are derived by the iterative calculation process.

In step S106, the end points A _k and B _k finally obtained are regarded as the end points A and B of the reference line segment and the length W _k of the line segment A _k B _k as W, and R = R _k The distance CP = d between the line segment having the length W and the measurement point P is calculated by the equation (6).

  The sign is + when the measurement point P is below the line segment AB and-when it is above the line segment AB. In the embodiment of the method for detecting a cross-sectional shape of the present invention, an example of measuring the peaking of the inner surface of the welded steel pipe has been given, but the peaking measurement on the outer surface can also be performed by attaching a sign to the distance from the reference line segment to the measurement point P. Applicable.

In step S107, the peaking value Δd = d−d ₀ is calculated by subtracting the depth d ₀ in the case of a perfect circle as a final reference from the equation (1). When the distance between the discrete points is dense, according to the present method, a measurement value with sufficient accuracy can be obtained only by searching for the points of the reference line segment. Further, assuming that the number of discrete points in the search ranges WL and WR is NL and NR, respectively, a combination of a maximum of NL × NR points can be considered. However, according to this method, the search is completed on the order of several times NL. .

<Second Peaking Detection Procedure>
As a second peaking detection procedure, a distance between two points is set to W among continuous points on a function obtained by approximating a cross-sectional profile based on a discrete point sequence to a cross-sectional profile within the search ranges WL and WR. A method for estimating a near end point and obtaining a distance from a reference line segment connecting the end points to the measurement point P will be described.

FIG. 6 shows the relationship between the cross-sectional profile of the welded steel pipe 1 obtained by any one of the above optical methods and the end points A and B of the line segment having a known length W and unknown coordinates. Here, the length W is the same as the leg width of the peaking measurement jig, A and B correspond to the points where the legs contact the cross-sectional profile, C is the midpoint of the line segment AB, and the measurement point P is the dial gauge. It matches the needle tip of. The length of the line segment CP corresponds to the peaking depth to be obtained. Ac and Bc indicate points on the cross-sectional profile at a distance of W / 2 in the horizontal direction from the measurement point P, respectively. L1 and L2 indicate curves approximating a cross-sectional profile within a range having the widths WL and WR designated with Ac and Bc as the centers. A _k and B _k indicate the intersections of a circle with a radius R _k centered on the measurement point P and the approximate curve. A ₀ is a right boundary point close to the measurement point P side on the approximate curve in the range WL.

  With reference to FIG. 7, the calculation algorithm of the peaking depth d = CP at the measurement point P will be described. In step S201, first, the toe point that is the measurement point P is detected and designated by the method described above.

  In step S202, the search ranges WL and WR for the end points A and B whose coordinates are unknown are set with the two points Ac and Bc on the cross-sectional profile at a distance of W / 2 in the horizontal direction from the measurement point P, respectively. To do.

  In step S203, the cross-sectional profiles in the ranges WL and WR are approximated by the left and right curves L1 and L2, respectively, by a function approximation method such as a least square method.

In step S204, the central sets the radius R _k of a circle in the measurement point P. The initial value R ₀ is, for example, the distance from the right boundary point A _{0 of the} WL near the measurement point P to the measurement point P among the points on the approximate curve L 1. R _k in the k-th setting is increased by a step size ΔR so that R _k = R ₀ + k · ΔR. The step size ΔR is specified in consideration of calculation costs such as accuracy and processing time.

In step S205, end points A _k and B _k on approximate curves L1 and L2 such that PA _k = PB _k = R _{k are} approximately _obtained . That is, to obtain the intersection between the circle of the approximate curve L1, L2 and radius R _k respectively, and end point A _k, B _k.

At step S206, When the end point A _k, from the coordinates of B _k of the line segment A _k B _k the length W _k calculated, Ashihaba W or more peaking measuring jig as a target, i.e. becomes W _k ≧ W, A determination is made to exit the loop from step S204 to step S205. As described above, two end points near the distance W are derived by the iterative calculation process.

In step S207, the end points A _k and B _k finally obtained are regarded as the end points A and B of the reference line segment and the length W _k of the line segment A _k B _k as W, and R = R _k . The distance CP = d between the line segment of the length W and the measurement point P is calculated by (6).

In step S208, the peaking value Δd = d−d ₀ is calculated by subtracting the depth d ₀ in the case of a perfect circle as the final reference from the equation (1).

  A function such as a parabola or a circle is used when the ranges WL and WR including the end points A and B of the line segment having the length W are wide on the cross-sectional profile, but may be approximated by a straight line when narrow.

<Third Peaking Detection Procedure>
When the discrete point sequence on the cross-sectional profile within the search ranges WL and WR is approximated by a straight line in the second peaking detection procedure, the depth CP connecting the measurement point P and the midpoint C of the line segment AB is directly calculated. A calculation formula can be derived.

  First, a calculation formula used in the third peaking detection procedure is derived using FIG. The straight lines L1 and L2 are straight lines (that is, tangent lines) obtained by approximating the peripheral portions of the end points A and B in FIG.

  Point E is a perpendicular foot drawn from the middle point C of the line segment AB to the straight line L1, point D is a perpendicular foot drawn from the measurement point P to the straight line CE, and point F is a perpendicular line drawn from the measurement point P to the straight line L1. Feet, point H is a perpendicular foot drawn from the middle point C to the straight line L2, point G is a perpendicular foot drawn from the measurement point P to the straight line CH, and point I is a perpendicular foot drawn from the measurement point P to the straight line L2. is there.

  Point R is the intersection of the straight line L1 and the extension of the straight line CP, point S is the intersection of the straight line L2 and the extension of the straight line CP, M1 is the bisector of the angle formed by the two straight lines L1 and L22, and M2 is the measurement point A straight line passing through P and parallel to the bisector M1.

U ₁ is the distance from the measurement point P to the straight line L1 (line segment PF), and U ₂ is the distance from the measurement point P to the straight line L2 (line segment PI). α is a half angle of the angle formed by the two straight lines L1 and L2, and δ is an angle formed by the line segment CP with respect to the bisector direction M2. In the following description, for example, the length of the line segment PF is simply referred to as PF.

  The straight line CR is inclined by δ with respect to the straight line M2, and since ∠ACR = ∠AEC = 90 ° and DP // L1, ∠CRA = ∠ACE = ∠CPD = α−δ. Similarly, since ∠BCS = ∠BHC = 90 ° and GP // L2, ∠CSB = ∠BCH = ∠CPG = α + δ.

From AC = BC = W / 2, CE = W / 2 · cos (α−δ), CD = d · sin (α−δ), U ₁ = PF = DE = CE−CD, the following expression (7) holds. . Similarly, the following equation (8) is established from CH = W / 2 · cos (α + δ), CG = d · sin (α + δ), and U ₂ = PI = GH = CH−CG.

  When the relational expressions (7) and (8) are transformed, the following expressions (9) and (10) are derived.

  When these equations (9) and (10) are given a straight line that approximates the periphery of the end points A and B of the line segment of length W on the cross-sectional profile, the CP length d and two This is an equation for calculating the angle δ formed by CP with respect to the bisector M1 of the angle formed by the straight lines L1 and L2 as an unknown. If the direction angle δ is eliminated, a quartic equation for CP = d is obtained, but the direction of the line segment CP is almost equal to the direction of the angle bisectors M1 and M2, and δ is a minute amount. ) And (10) From the following relational expressions (11) and (12) obtained by further modifying each, it is sufficient to calculate δ by a first order approximation and calculate the depth d.

That is, first, the initial value d ₀ of the depth CP = d is calculated with δ = 0 in equation (11). Δ = δ ₁ is calculated from Equation (12) using the initial value d ₀ of the depth d. In equation (11), the final depth d is calculated with δ = δ ₁ .

When the measurement point P is above the line segment AB, d <0 must be satisfied. However, in order to correspond to the sign, when the measurement point P is above the straight lines L1 and L2, U ₁ > 0, U ₂ > 0, U ₁ <0, U ₂ <0 in the lower case.

  FIG. 9 shows a flowchart of a procedure for finding the end points A and B of the reference line segment on a straight line that approximates a discrete cross-sectional profile in the search ranges WL and WR, and calculating the peaking depth d = CP at the measurement point P. .

  With reference to FIG. 9, the calculation algorithm of the peaking depth d = CP at the measurement point P will be described. In step S301, first, the toe point that is the measurement point P is detected and designated by the method described above.

  In step S302, the search ranges WL and WR for the end points A and B whose coordinates are unknown are set around the two points Ac and Bc on the cross-sectional profile at a distance W / 2 in the horizontal direction from the measurement point P, respectively. To do.

  In step S303, the cross-sectional profiles in the ranges WL and WR are approximated by straight lines L1 and L2, respectively, by a function approximation method such as a least square method.

In step S304, the length W of the reference line segment, the distances U ₁ and U ₂ from the measurement point P to the straight lines L1 and L2, and the half angle α formed by the straight lines L1 and L2 are known, and the equations (11) and (12 ), The length d of the line segment connecting the midpoint C to the measurement point P and the deviation angle δ from the bisector direction between the angles formed by L1 and L2 are obtained.

In step S305, the peaking value Δd = d−d ₀ is calculated by subtracting the depth d ₀ in the case of a perfect circle as a final reference from the equation (1).

  In realizing the cross-sectional shape detection device of the present invention, a personal computer is used as an image processing device and a signal processing control device, and each cross-sectional profile derivation method and each peaking detection procedure in the present invention are processed in, for example, a personal computer. This can be realized by a computer CPU executing a computer program. Further, in the cross-sectional shape detection apparatus of the present invention, a keyboard and a mouse may be used to set data for each process, and intermediate results and peaking derivation results of each process are displayed on a computer display. Display (not shown).

  A sample having a length of 200 mm cut out from an actual UO steel pipe was placed on an automatic stage, and peaking measurement was performed while moving the sample in the pipe axis direction. FIG. 10 is a graph plotting changes in the longitudinal (tube axis) direction of measured values (solid line) obtained by the method of the present invention and measured values (◯) obtained by applying a dedicated peaking measurement jig. is there. Both agree well with an accuracy of ± 0.2 mm, and the present method is consistent with the method using the peaking measurement jig. Therefore, both measured values can be used for quality control.

  This method is a calculation method using information on multiple discrete points, and there is little variation in measurement due to the distance between adjacent discrete points and the unevenness of the cross section, and the cross-sectional shape of the curved part can be calculated easily and at high speed. Subtle changes in the tube axis direction that cannot be captured by intermittent measurement can be measured over the entire length at fine intervals. In particular, peaking that affects the welding strength can be measured by a method compatible with the peaking measurement jig in each manufacturing process, and comparison with hand-measured values between processes and tracking of shape changes can be performed. In addition, by accumulating data, it is possible to determine a peaking control value that can guarantee the welding strength, and by reflecting it in manufacturing and quality control, it is possible to guarantee the quality over the entire length.

In embodiment of this invention, it is the schematic of the apparatus structure of the 1st cross-sectional profile derivation method, and is a figure which shows the whole structure by the light cutting method using a linear light source or a point light source. It is a figure which shows the optical arrangement | positioning by the light cutting method using a linear light source or a point light source, and the coordinate system of an object side and an image side in the 1st cross-sectional profile derivation method. In embodiment of this invention, it is the schematic of the apparatus structure of the 2nd cross-sectional profile derivation method, and is a figure which shows the whole structure at the time of using a ranging means. It is a figure which shows the positional relationship of search range WL, WR of the end points A and B of the reference line segment on a cross-sectional profile, and a discrete point sequence in the 1st peaking detection procedure in the cross-sectional shape detection method of this invention. It is a flowchart which calculates a peaking value in the 1st peaking detection procedure in the cross-sectional shape detection method of this invention. In the 2nd peaking detection procedure in the section shape detection method of the present invention, it is a figure showing the positional relation between search range WL, WR of end points A and B of a reference line segment on a section profile, and an approximate curve. It is a flowchart which calculates a peaking value in the 2nd peaking detection procedure in the cross-sectional shape detection method of this invention. It is explanatory drawing used for derivation | leading-out of a geometric relational expression in the 3rd peaking detection procedure in the cross-sectional shape detection method of this invention. It is a flowchart which calculates a peaking value in the 3rd peaking detection procedure in the cross-sectional shape detection method of this invention. It is the figure which plotted the measured value obtained by applying the cross-sectional shape detection method of this invention to the sample cut out from the actual UO steel pipe, and the measured value in a dedicated jig in the longitudinal direction. It is a figure explaining the positional relationship on the cross-sectional profile of the measuring method of the peaking which uses the dial gauge which is a prior art, and a measurement jig | tool of a UO steel pipe. It is a figure explaining the peaking in the toe part of a UO steel pipe. It is a figure explaining the manufacturing process of a UO steel pipe.

Explanation of symbols

1: Welded steel pipe 2: Projection device (linear laser light source or scanning dot laser light source)
3: Light cut image 4: Trace of linear laser beam or scanning point beam 5: Imaging means (camera)
5 ': Laser distance meter 6: Image processing device 7: Signal processing control device (first)
7 ': Signal processing control device (second)
10: Bead part 11: Toe part 12: Peripheral part outside bead P: Toe part

Claims (5)

In the cross-sectional shape detection method for measuring the peaking of the measurement point P on the surface of the bending portion, using a member having the bending portion as an inspection object,
Based on a light projection pattern irradiated with linear light along the curved surface, or a linear ray trajectory on the curved surface obtained by irradiating and scanning with dotted light, a plurality of points are obtained by an optical method. A first step of measuring a cross-sectional profile of the curved portion configured;
A second step of setting a measurement point P on the cross-sectional profile;
A third step of designating a first point and a second point on the cross-sectional profile that are separated from the measurement point P in the horizontal direction by a distance W / 2 that is half of the preset distance W;
Cross-sectional profiles on respective cross-sectional profiles in a predetermined first search range WL centered on the first point and in a predetermined second search range WR centered on the second point The first and second end points are the points where the distance between the two points is the closest to the distance W from the set of continuous points on the function obtained by approximating the cross-sectional profile or a plurality of discrete points constituting And a fourth step to search as
A line segment connecting the first and second end points is used as a reference line segment, and a distance from the reference line segment to the measurement point P is calculated based on the positions of the first and second end points and the measurement point P. And a fifth step
And a sixth step of calculating peaking of the measurement point P based on a distance from the reference line segment to the measurement point P.   The cross-sectional shape detection method according to claim 1, wherein the optical method is a light cutting method.   The cross-sectional shape detection method according to claim 1, wherein the optical method is a triangulation method. A cross-sectional shape detection apparatus for measuring peaking of the measurement point P on the curved portion surface used in the cross-sectional shape detection method according to claim 1 or 2,
Irradiating linear light along the curved surface, or projecting means for irradiating and scanning with dotted light to obtain a linear ray locus on the curved surface;
Imaging means for obtaining an image of the linear ray trajectory;
Cross-sectional profile measuring means for deriving a cross-sectional profile of the curved portion constituted by a plurality of points based on an image of the ray trajectory by predetermined image processing;
A cross-sectional shape detection apparatus comprising: signal processing control means for executing the second to sixth steps. A cross-sectional shape detection device for measuring peaking of the measurement point P on the curved portion surface used in the cross-sectional shape detection method according to claim 1,
Projecting means for irradiating and scanning point light to obtain a linear ray locus on the surface of the curved portion;
Position detecting means for measuring the position of the linear ray trajectory;
Cross-sectional profile measuring means for deriving a cross-sectional profile of the curved portion constituted by a plurality of points;
A cross-sectional shape detection apparatus comprising: signal processing control means for executing the second to sixth steps.

Images