JP2017225995A - Device and method for monitoring welding of electroseamed steel pipe - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automatically detect a site of a defect caused by biting of foreign matters, such as scale and iron dust, into a weld surface.SOLUTION: A welding monitoring device 1 in a welding process of an electroseamed steel pipe includes: imaging means 12 for imaging a region including a weld seam extending in a plate thickness direction in which metal plates are butt-welded; image processing means 14 for retrieving images photographed by the imaging means and capturing variation in luminance in the weld seam on the basis of the continuously photographed images; and determination means 16 for determining presence/absence of abnormality by comparing the luminance with a set threshold.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a monitoring method and a monitoring device for detecting in real time a defect that occurs when a foreign matter such as a scale or iron powder bites into a welded surface of an electric resistance welded steel pipe.

  Conventionally, an electric resistance steel pipe, which is a steel pipe formed using high frequency resistance welding (high frequency electric resistance welding), is known. As the material of the steel pipe, a hot-rolled steel sheet having a scale grown on a surface layer, often called a black skin material, is used, and the scale of the hot-rolled steel sheet surface layer is peeled off during the forming process or fin pass roll. In particular, the fin pass roll has an effect of creating a new surface by scraping the welded surface, while causing the scale of the hot-rolled steel sheet surface layer and the iron powder to peel off from the steel sheet itself. When scales peeled off in this way or scale powder or iron powder or the like is mixed into the cooling water, a phenomenon may occur in which the scale powder or iron powder or the like bites into the welded portion. In this case, if the size of these foreign substances is large to some extent, there is a possibility that they will not be melted up to the upset and remain solid on the weld surface and become defective without being discharged. Such a defect not only lowers the toughness of the welded part, but also causes cracking during and after processing, so it is strongly desired to detect it during pipe making. Therefore, Patent Document 1 proposes a technique for detecting such a defect. However, according to the method described in Patent Document 1, it is difficult to detect unless a large amount of scale is mixed so as to remarkably reduce the light emission luminance of the bead portion.

  On the other hand, high-frequency electric seam welding is a process of efficiently welding by concentrating current on the welding surface using the proximity effect and skin effect of the steel material edge at the weld. In high-frequency electric seam welding, a large current flows through the edge of the steel material, so a strong electromagnetic field is formed in the periphery. Since this electromagnetic field is maximized at the welding point (point V), if there is a magnetic body around it, it is easy to jump into the welding point.

JP 2011-036892 A

  By the way, it has not been clear until now what kind of biting process these defects occur. Therefore, this time, an experiment was conducted to artificially generate the phenomenon by sprinkling cooling water mixed with scale or iron powder on the fin path or work coil. As a result of periodically photographing the welded part with a camera installed above the welded part and matching it with cracks or defect occurrence sites in the flattening test, `` the foreign material mixed in the cooling water stays on the weld surface upstream “It may be carried from the side and bitten.”

  In addition, the inventors conducted an artificial scale biting experiment while directly photographing the welding surface using a borescope or a high-power lens, and as a result, foreign matter was carried from the upstream side and biting into the welding surface was found. It was observable, and it became clear that the appearance was different depending on the scale size and the biting side (outer surface / center / inner surface). However, all of them tend to converge to the abutting position (weld line) of the weld surface, and it has been found that the brightness of the weld line part is significantly reduced.

  The present invention has been made with such circumstances, and it is possible to automatically detect a site of a defect that occurs when a foreign object such as a scale or iron powder bites into a welding surface.

  The present invention made to solve the above problems employs the following means. First, the first means continuously forms a cylindrical shape by a roll group while transporting a belt-shaped metal plate, and adds upset by a pair of squeeze rolls from the side to the cylindrical metal plate. The welding process of the ERW steel pipe which manufactures the steel pipe by heating and melting by high-frequency contact resistance welding or high-frequency induction resistance welding while performing heat input control to both ends in the circumferential direction of the metal plate converged in a V shape A welding monitoring device, wherein the image capturing means captures an image of an area including a weld line extending in the thickness direction where the metal plate is abutted, and the image captured by the image capturing means is captured continuously. Image processing means capable of capturing a change in luminance on the weld line based on the photographed image, and a determination means for determining presence / absence of abnormality in comparison with the threshold value for which the luminance is set are provided. Characterized in that there is a welding monitoring device of the welding process the seam welded steel pipe.

  As a first means, the image processing means has a setting area covering at least the plate thickness of the weld line extracted from the photographed image in a rectangular shape, and calculates the maximum luminance in the section divided into the setting area. It is preferable to include a means for calculating the average luminance for each image based on the luminance distribution as the representative value of each area.

  In the first means, the determination means for determining the presence / absence of abnormality includes a threshold value for determining the presence / absence of abnormality determined by using average brightness at normal time and abnormal time obtained in advance by a preliminary test for generating abnormality. And means for comparing the threshold with an average luminance for each image.

  In the first means, the image processing means calculates a normal average brightness from a plurality of images at the start when it is confirmed that no abnormality has occurred, sets a threshold, and sets the threshold as an average brightness for each image. It is preferable to have a configuration including means for comparing.

  In the first means, it is preferable that the determination means for determining the presence / absence of abnormality includes a means for determining that an abnormality has occurred when an average luminance calculated for each image is equal to or less than a set threshold value.

  In the first means, it is preferable to include a means for performing marking and / or position information transmission to a host computer at a position where the weld surface brightness is lowered.

  Further, the second means is to continuously form a cylindrical shape by a group of rolls while transporting the belt-shaped metal plate, and add upset by a pair of squeeze rolls from the side to the cylindrical metal plate. The welding process of the ERW steel pipe which manufactures the steel pipe by heat-melting and butting by high-frequency contact resistance welding and high-frequency induction resistance welding while controlling the heat input to both ends in the circumferential direction of the metal plate converging in a V shape A method of monitoring the welding, wherein an image of a region including a weld line extending in the plate thickness direction where the metal plate is abutted is captured, the image captured by the image capturing means is captured, and the continuously captured image is captured. It is the welding monitoring method of the welding process of an ERW steel pipe characterized by catching the luminance change of a welding line based on this, and distinguishing presence or absence of abnormality compared with the threshold which set the above-mentioned luminance.

  In the second means, when capturing the luminance change of the weld line based on the images photographed in time series, a setting area covering at least the plate thickness of the weld line is extracted in a rectangular shape from the photographed image. It is preferable to calculate the average luminance for each image based on the luminance distribution in which the maximum luminance in the section divided into the setting areas is the representative value of each area.

  In the second means, when determining the presence / absence of the abnormality, a threshold value determined using the average brightness at the normal time and the abnormal time determined in advance by a preliminary test for generating the abnormality is determined, and the threshold value Is preferably compared with the average luminance for each image.

  Secondly, in the means, when capturing the luminance change of the weld line based on the images taken in time series, the average luminance at the normal time is calculated from a plurality of images at the start when it is confirmed that there is no abnormality, and normal Preferably, the lower limit of the state is set as a threshold value, and the threshold value is compared with the average luminance for each image.

  In the second means, when the presence / absence of the abnormality is determined, it is preferable to determine that the abnormality has occurred when the average luminance calculated for each image is equal to or less than a set threshold value.

  In the second means, it is preferable to perform marking and / or position information transmission to a host computer for the position where the brightness of the weld surface is lowered.

  When the first means and the second means are used, it is possible to automatically detect a site of a defect that occurs when a foreign matter such as a scale or iron powder bites into the welding surface.

It is the perspective view which showed the state which is monitoring the high frequency electric-welding welding part. It is the figure which showed the positional relationship of an imaging | photography process means and a welding location. It is the figure which showed the algorithm of the image processing in embodiment. It is the figure which showed the relationship between an approximate line and a setting area. It is the figure which showed notionally the relationship between a setting area and an area. It is the image which image | photographed the healthy state which the foreign material has not bitten. It is the image which image | photographed the abnormal state which the foreign material bite.

  Hereinafter, embodiments of the invention will be described. 1 and 2 show a state in which a high-frequency electro-welded weld is being monitored. FIG. 1 is a bird's eye view and FIG. 2 is a side view. FIGS. 1 and 2 show (a) and (b), respectively, where (a) is monitored from above by the image photographing means 12a using a probe, and (b) is a normal image photographing means 12b (for example, a long focal point). The case of monitoring from an oblique upstream using a CCD camera equipped with a lens) is shown. In order to perform high-frequency electric seam welding, prior to welding, the belt-shaped metal plate 22 is continuously formed into a substantially cylindrical shape by a roll group (not shown) while being conveyed. While applying an upset to the substantially cylindrical metal plate 22 by a pair of squeeze rolls 23 from the side, performing heat input control to both ends in the circumferential direction of the metal plate 22 converging in a V shape. Then, the steel pipe is manufactured by heating and melting by high-frequency contact resistance welding or high-frequency induction resistance welding. This method of manufacturing a steel pipe is a general method for manufacturing a steel pipe using electric resistance welding.

  The welding monitoring apparatus 1 according to the present embodiment used in such a method for manufacturing a steel pipe using electric resistance welding is an image that captures an image of a region including a weld line extending in the plate thickness direction in which the metal plate 22 is abutted. Imaging means 12a and 12b are provided. Compared with the image processing means 14 capable of capturing the images photographed by the image photographing means 12a and 12b and capturing the change in the luminance of the weld line based on the photographed images, and a threshold value for which the luminance is set. A discriminating means 16 for discriminating the presence or absence of abnormality is provided. If such a welding monitoring apparatus 1 is used, it becomes possible to automatically detect a site of a defect that occurs when a foreign matter such as a scale or iron powder bites into the welding surface. Moreover, since it is the structure which catches the change of the brightness | luminance in a weld line based on the image of the area | region containing a weld line, it becomes possible to grasp | ascertain the biting of a comparatively small amount of scales.

  If the welding monitoring apparatus 1 of this embodiment is used, it will also become possible to detect the defective part used as removal object at an early stage. It also leads to the realization of the welded part of the steel pipe. The image photographing means 12a and 12b are located upstream and above the V-shaped convergence region 29 where the metal plate 22 converges in a V shape, as can be understood from the matters shown in FIGS. In the case of the image photographing means 12a, the welding line is photographed from the upstream side by the probe, and in the case of the image photographing means 12b, the welding line is photographed from the obliquely upstream side.

  In the present embodiment, an impeder 32 is disposed inside a metal plate 22 bent into a cylindrical shape, and a work coil 34 is disposed around the steel plate. This is high frequency induction resistance welding in which high frequency power is supplied from the high frequency power source 36 to the work coil 34. The form of the electric resistance welding is not limited to such a form, and high-frequency contact resistance welding such as a form in which high-frequency power is supplied to the contact tip brought into contact with the metal plate 22 bent into a cylindrical shape. It is also possible.

  Next, an image processing algorithm will be described with reference to FIG. Here, description will be given of a case where a CCD camera is used as a color camera capable of obtaining a color image. First, a captured image is acquired (S001). Next, a red component is extracted from the photographed image (S002). After extracting the red component from the captured image, an approximate line 52 of the inner edge is obtained as shown in FIG. 4 (S003). The approximate line 52 is obtained by the least square method by searching the high luminance part in the left-right direction from the center position of the lower end of the image to detect the edge position. The intersection of the obtained two approximate lines 52 is detected as the lower end (inner surface side) of the weld line (S004).

  A setting area 6 for detecting luminance is set with reference to this point. The setting area 6 of the present embodiment is set in a rectangular shape, and in the left-right direction, the setting area 6 is set so as not to include a region that is brighter than the welded portion on the left and right sides around the intersection of the approximate lines 52. ing. The vertical direction of the setting area 6 is set so as to cover at least the entire plate thickness. The reason why the setting area 6 is rectangular is that it is assumed that the welding line is not necessarily perpendicular to the image or a straight line.

  Usually, since the weld line has the highest luminance as compared with the peripheral portion, the setting area 6 may have a length of several mm or more on the left and right sides. Further, regarding the length of the setting area 6 in the vertical direction, it is conceivable that foreign matter partially bites in, and therefore it is preferable to set the length to cover the entire thickness of the metal plate 22.

  In the present embodiment, as conceptually shown in FIG. 5, the setting area 6 is divided into a plurality of sections 62, and the luminance distribution in the setting area 6 can be compared with the set of sections 62. In this specification, the area 62 extending in the left-right direction on the image is referred to as a line. In the setting area 6 of the present embodiment, lines extending in the left-right direction are arranged in a line in the vertical direction.

  In the present embodiment, the maximum luminance is calculated for each of the areas 62, and the cross-sectional luminance of the weld line is obtained. Specifically, the maximum luminance in the area 62 is set as a representative value of each area 62, and the average luminance is calculated for each image based on the luminance distribution that is the distribution of the representative value. Unlike this embodiment, it may be possible to take the brightness of one section in the vertical direction, but the weld line may move slightly left and right due to molding, material shape, strength variations, etc., so the maximum for each line It is preferable to take luminance.

  In the present embodiment, as shown in FIG. 3, after S004, the maximum luminance is calculated for each line whose left-right width is set to 1 mm (S005). Next, the average luminance is calculated by calculating the average value of the maximum luminance obtained for each line (S006). Further, this average luminance is compared with a preset threshold value (S007). As a result, if it is determined that the average luminance is equal to or higher than the set luminance, a normal determination is made (S008). If it is determined that the average brightness is not equal to or higher than the set brightness, a defect is determined (S009). If the determination is made in S008 or S009, the process returns to the beginning, and thereafter the same process is repeated. Note that the threshold value may be determined in advance by confirming the luminance that is a boundary between abnormality and normality through a preliminary test for intentionally generating a defect. Alternatively, it may be confirmed that there is no biting in a plurality of images taken in a few seconds immediately after the start, and an average value thereof is calculated to determine the lower limit value of the normal state as a threshold value.

  On the basis of the detection information obtained in this way, it is possible to have a configuration including means for performing marking and / or transmission of positional information to a host computer at a position where the brightness of the welding surface is reduced. Specifically, it is possible to adopt a configuration in which marking (ink method) is performed immediately after welding, or a configuration in which the detection position is transmitted to the process controller while measuring the length from the top of the coil. By adopting such a configuration, it becomes possible to remove the defective portion in the refining process, and high quality control can be realized.

  Next, examples will be described. First, detection processing was performed on the images shown in FIGS. 6 and 7 using the algorithm shown in FIG. The image shown in FIG. 6 is in a healthy state where no foreign matter is bitten, and the image shown in FIG. 7 is an abnormal time when the foreign matter is bitten. The average brightness was “120 to 150” when the sound was photographed in a healthy state, but was significantly reduced to “35” when the biting of a foreign object was photographed. From the above, it has been confirmed that if a welding monitoring method using the welding monitoring apparatus 1 of the present embodiment is employed, a defect in the ERW steel pipe can be detected.

  Next, in the actual production line, the welding monitoring apparatus 1 was installed, and the state of the welding surface was measured while continuously performing photographing and image processing. The target pipe is an actual pipe of φ100mm × 4mmt. The camera used for imaging was 200 frames / second, and the exposure time was 1/10000 seconds. While photographing the welded portion, the cooling water mixed with the scale was sprinkled around the work coil 34 to generate a simulated defect. A flat test was conducted from the top after pipe making, and the position where the crack occurred was measured. Prior to the start of the test, a notch was made in advance in the coil end (metal plate end), and the frame number reflected in the image was recorded. Since the tube-forming speed of 40 mpm at the test site is constant, it corresponds to about 3.3 mm per frame. Starting from the image where the cutout was taken, the image detected abnormally by the algorithm described above was converted into an actual distance, and matched with the actual image. Table 1 shows the test results for four pipes. The numerical value on the left side of the hyphen in “No.” in Table 1 is the pipe number. The unit of position in Table 1 is “m”.

  Since the results shown in Table 1 were obtained, it was found that the actual fruit confirmation position obtained by tracking and flattening was substantially coincident with the position of the image.

  The present invention is not limited to the above-described embodiments, and it goes without saying that the present invention can be applied without departing from the spirit of the present invention. For example, in an image photographed using a color camera, it is preferable to use a red component having the highest radiance or a steel material for processing. However, if photographing can be performed with a sufficient luminance level, the same processing can be performed with a green or blue component. In addition, when a monochrome camera is used, it is not necessary to extract color components.

DESCRIPTION OF SYMBOLS 1 Welding monitoring apparatus 6 Setting area 12a, b Image photographing means 14 Image processing means 16 Discriminating means 22 Metal plate 52 Approximate line 62 Area

Claims (12)

Metal that is continuously formed into a cylindrical shape by a group of rolls while transporting a belt-shaped metal plate, and is converged into a V shape while being upset by a pair of squeeze rolls from the side of the cylindrical metal plate. A welding monitoring device for the welding process of an ERW steel pipe for producing a steel pipe by heat melting by high frequency contact resistance welding or high frequency induction resistance welding while performing heat input control to both ends in the circumferential direction of the plate,
Image photographing means for photographing an image of a region including a weld line extending in a thickness direction in which the metal plate is abutted; and
Image processing means capable of capturing an image captured by the image capturing means and capturing a change in luminance in the weld line based on continuously captured images;
A discriminating means for discriminating the presence or absence of an abnormality by comparing the brightness with a set threshold;
A welding monitoring device for a welding process of an electric resistance welded steel pipe, comprising: In the image processing means, a setting area covering at least the plate thickness of the weld line is extracted in a rectangular shape from the captured image,
2. The electro-sewing device according to claim 1, further comprising means for calculating an average luminance for each image based on a luminance distribution in which the maximum luminance in the area divided into the setting area is a representative value of each area. Welding monitoring device for steel pipe welding process. The discriminating means for discriminating the presence / absence of an abnormality includes a threshold value for determining the presence / absence of an abnormality determined using an average luminance at the time of normality and abnormality obtained in advance by a preliminary test for generating an abnormality,
The welding monitoring apparatus for a welding process of an ERW steel pipe according to claim 1 or 2, further comprising means for comparing the threshold value with an average luminance for each image. The image processing means calculates a normal average luminance and sets a threshold value from a plurality of images at the start when it is confirmed that no abnormality has occurred,
The welding monitoring apparatus for a welding process of an ERW steel pipe according to claim 1 or 2, further comprising means for comparing the threshold value with an average luminance for each image.   The determination means for determining the presence / absence of abnormality comprises means for determining that an abnormality has occurred when an average luminance calculated for each image falls below a set threshold value. The welding monitoring apparatus of the welding process of the ERW steel pipe of 1 description.   The welding of the ERW steel pipe according to any one of claims 1 to 5, further comprising means for performing marking and / or transmission of positional information to a host computer at a position where the brightness of the welding surface is lowered. Process welding monitoring device. Metal that is continuously formed into a cylindrical shape by a group of rolls while transporting a belt-shaped metal plate, and is converged into a V shape while being upset by a pair of squeeze rolls from the side of the cylindrical metal plate. There is a welding monitoring method of the welding process of an ERW steel pipe that manufactures a steel pipe by heat melting and welding by high frequency contact resistance welding and high frequency induction resistance welding while performing heat input control to both ends in the circumferential direction of the plate,
Taking an image of an area including a weld line extending in the thickness direction where the metal plate is abutted, capturing an image taken by the image taking means, and changing the brightness of the weld line based on the continuously taken image And determining whether or not there is an abnormality by comparing the brightness with a set threshold value, and a welding monitoring method for a welding process of an ERW steel pipe. When capturing the brightness change of the weld line based on the image taken in the time series, the setting area that covers at least the plate thickness of the weld line from the photographed image is extracted in a rectangular shape,
8. The welded ERW steel pipe according to claim 7, wherein the average brightness is calculated for each image based on a brightness distribution in which the maximum brightness in the section divided into the setting areas is a representative value of each section. 9. Process welding monitoring method.   When determining the presence / absence of the abnormality, the threshold value determined using the average luminance at the normal time and the abnormal time determined in advance by a preliminary test for generating the abnormality is determined, and the threshold value is determined as the average luminance for each image. The welding monitoring method of the welding process of the ERW steel pipe according to claim 7 or 8, wherein   When capturing changes in the brightness of the weld line based on the images taken in time series, the average brightness at the normal time is calculated from a plurality of images at the start when it is confirmed that no abnormality has occurred, and the lower limit of the normal state is used as a threshold value. The welding monitoring method according to claim 7 or 8, wherein the threshold value is set and compared with an average luminance for each image.   11. The electric resistance welded steel pipe according to claim 7, wherein, when determining whether or not there is an abnormality, it is determined that an abnormality has occurred when an average luminance calculated for each image is equal to or less than a set threshold value. 11. Monitoring method of welding process.   The welding monitoring method for the welding process of an ERW steel pipe according to any one of claims 7 to 11, wherein the position of the weld surface brightness is reduced and / or position information is transmitted to a host computer. .