JP2006043741A - Method and device for evaluating laser welding quality - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and a device for evaluating the laser welding quality capable of correctly grasping the welding situation related to various welding defects. <P>SOLUTION: A CCD camera 8 to receive the reflected laser beams B to be reflected by a laser beam irradiation area via an optical filter 7 is mounted on a laser torch 1 to irradiate laser beams A on a work W to be welded, and the CCD camera 8 picks up images of a molten part 5 including a key hole 4 and its periphery. The obtained images are transmitted to an image analysis means 11 in a signal processing device 9 to allow the image analysis means 11 to operate the length/width ratio of the molten part 5 to be grasped from the images, and the result of operation is transmitted to a quality determination means 12, and defective shrinkage is determined if the length/width ratio is larger than the threshold stored in advance. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a technique for grasping and evaluating a laser welding situation in real time, and in particular, a laser welding quality evaluation method for grasping and evaluating a laser welding situation using reflected light of a laser beam reflected from a laser irradiation region, and Relates to the device.

Conventionally, as a laser welding quality evaluation method and apparatus for grasping and evaluating a laser welding situation using reflected light of a laser beam, there is one described in Patent Document 1. With this, the reflected light of the laser beam reflected from the laser irradiation area is collected by a condensing lens and sent to a CCD camera (imaging device). Based on the luminance level of each pixel from the CCD camera, there is a weld defect Is determined.
JP 2000-42769 A

  By the way, for example, in so-called laser lap welding, in which two plate materials (steel plates, etc.) are overlapped and laser-welded, the welding situation varies greatly depending on the size of the gaps existing between the plate materials, and if the gap becomes larger, it is closed. , Welding defects such as burn-off occur. However, as described in Patent Document 1, in the method of determining the presence or absence of welding defects from the brightness level of the reflected light of the laser, it is possible to accurately grasp the welding situation in which these shrinkage, burnout, etc. occur. There was a problem that it was difficult.

  The present invention has been made in view of the above-described problems, and the object of the present invention is to make it possible to accurately grasp the welding situation that leads to various welding defects, and thereby greatly contribute to the improvement of reliability. The object is to provide a quality evaluation method and apparatus.

  The present invention has been made in order to solve the above-described problem, and the laser welding quality evaluation method according to the present invention images the melted part and its periphery by the reflected light of the laser light reflected from the laser irradiation area, It is characterized in that the welding quality is evaluated based on the shape characteristic of the melted portion grasped from the captured image. In this case, the shape characteristic of the fusion part may be grasped from the luminance integration profile for the image.

  Further, the laser welding quality evaluation apparatus according to the present invention includes an imaging unit that captures an image of the melted part and its periphery by reflected light of laser light reflected from the laser irradiation area, and a shape of the melted part from the image captured by the imaging unit. And a signal processing means for determining a welding quality based on the shape characteristic. In this case, a CCD camera can be used as the imaging means.

  According to the laser welding quality evaluation method and apparatus according to the present invention, since the welding quality is evaluated based on the shape characteristics of the melted part grasped from the captured image, it includes shrinkage, burnout, non-penetrating welding, and the like. Therefore, it is possible to accurately grasp various welding defects, and the reliability of the evaluation of the welding quality is remarkably improved.

The best mode for carrying out the present invention will be described below with reference to the accompanying drawings.
FIG. 1 shows one embodiment of a laser welding quality evaluation apparatus according to the present invention. In the figure, reference numeral 1 denotes a laser torch. In the laser torch 1, an optical system for irradiating a workpiece W with a laser beam A sent from a laser oscillator 2 through an optical fiber 3 is incorporated. The workpiece W is composed of two steel plates W1 and W2 that are superposed on each other. During laser welding, the laser beam A emitted from the laser torch 1 has a predetermined size on the upper steel plate W1. In this state, the laser torch 1 is moved in the welding direction F with respect to the workpiece W. Of course, the position of the laser torch 1 may be fixed and the workpiece W may be moved in the welding direction F.

  In laser welding, as shown in FIG. 1, a keyhole 4 is formed in the irradiation region of the laser beam A on the workpiece W, and a molten portion (melt pool) of metal around the keyhole 4. 5 is formed. The molten part 5 is formed so as to extend to the rear side of the welding direction F, and the molten metal is sequentially solidified from the rear end side of the molten part 5 in accordance with the movement of the laser torch 1 (relative movement with the workpiece W). In FIG. 1, reference numeral 6 denotes a weld bead that is a solidification mark of the molten metal, and two steel plates W <b> 1 and W <b> 2 as the workpieces W are lap welded to each other via the weld bead 6. . As the welding laser beam A, a high output laser beam such as a YAG laser beam or a carbon dioxide laser beam is used.

  In the present embodiment, at the rear end portion of the laser torch 1, a CCD camera that receives the reflected light B of the laser light reflected from the melting portion 5 and its periphery through the optical filter 7 coaxially with the welding laser light A ( An imaging means) 8 is attached. The optical filter 7 is an interference filter whose transmission band is the wavelength component of the laser light. The CCD camera 8 images the melted part 5 and its surroundings with the reflected light B of the laser light that has passed through the optical filter 7, An image (visualized image) of the fusion part 5 and its periphery is obtained. In addition, a signal processing device (signal processing means) 9 is separately provided, and an image captured by the CCD camera 8 is sent to the signal processing device 9 through a signal line 10. . The signal processing device 9 analyzes an image picked up by the CCD camera 8 by a method described later, and determines the shape characteristic of the melted part and welding based on the analysis result of the image analyzing means 11. And a quality judgment means 12 for judging quality quality.

  Here, the image of the melted portion 5 obtained from the reflected light B of the laser beam has a high-luminance region 21 with the above-described keyhole 4 as a dark field 20 as shown in FIG. Appears in a donut shape. And the shape of this high-intensity area | region 21 changes according to a welding condition, for example, is divided | segmented before and after the welding direction F, as shown to the same figure (b) and (c). In this case, the arcuate front high brightness portion 21a located on the front side in the welding direction is on the front wall portion of the keyhole 4, and the rear high brightness portion 21b located on the rear side in the welding direction is the rear wall of the keyhole 4. Each of which corresponds to a portion (usually called a weld pool tip), and the splitting of the rear high-luminance portion 21b moves backward in the welding direction F with respect to the front high-luminance portion 21a. It appears as a phenomenon, that is, as a phenomenon that the tip of the molten pool recedes. In FIG. 2, the image is represented as a schematic diagram for the sake of clarity. Hereinafter, all the images are represented as schematic diagrams.

  As a result of various investigations on the relationship between the split state of the high-luminance region 21 and the actual welding quality, the present inventors have found that as the amount of shrinkage increases, the front high-luminance portion as shown in FIG. In a situation in which the rear high-luminance part 21b recedes to the rear side in the welding direction F with respect to 21a and is nearly burned out, as shown in FIG. It has been found that the part 21b is largely retracted to the rear side in the welding direction F.

  Therefore, the inventors normalize the total length L of the melted portion grasped from the high luminance region (image) 21 with the width (bead width) W of the melted portion as shown in FIG. Various investigations were made on the relationship between (L / W) and welding quality. FIG. 4 shows the result of laser lap welding of two steel plates W1 and W2 having a thickness of 0.7 mm. As the L / W ratio increases, the actual shrinkage (actual shrinkage) is shown. Amount) was found to increase linearly.

  By the way, about the image of the fusion part shown in the above-mentioned FIG. As shown in FIGS. 5 (a), (b), and (c) correspond to (a), (b), and (c) in FIG. 2, respectively, and the distribution of each luminance integrated value (hereinafter referred to as luminance integration). In the profile, peaks P1 and P2 corresponding to the front high luminance part 21a and the rear high luminance part 21b appear. In the present embodiment, the distance l between the rising point of the peak P1 corresponding to the front high luminance part 21a and the position of the peak P2 corresponding to the rear high luminance part 21b is the total length L (FIG. 3) of the melting part 5. The length L is calculated by the image analysis means 11 in the signal processing device 9. 2A to 2C, the luminance values are integrated in the horizontal axis direction, and the distribution of the luminance integrated values in the vertical axis direction (direction orthogonal to the welding direction) is obtained. When viewed, a trapezoidal luminance integration profile (not shown) is obtained. In the present embodiment, the distance connecting rising points generated above and below the trapezoidal luminance integration profile is defined as the full width W (FIG. 3) of the melting portion 5, and this width W is also calculated by the image analysis means 11. ing.

  An analysis area is set for the image obtained by the CCD camera 8 in order to improve the accuracy and speed of image analysis by the image analysis means 11. As shown in FIG. 3, this analysis area is set to a predetermined value on the left side of the rising point of the peak P1 (corresponding to the front high luminance portion 21a) shown in the luminance integration profile in the horizontal axis direction (FIG. 5). The position where the number of pixels is offset is the analysis start position S1 in the horizontal axis direction, the position where the predetermined number of pixels is offset from the analysis start position S1 to the rear side in the welding direction is the analysis end position S2, and the vertical axis direction table The position obtained by offsetting the predetermined number of pixels upward from the upper rising point of the shape luminance integration profile is the upper analysis position S3, and the predetermined number of pixels is offset downward from the lower rising point of the same trapezoidal luminance integration profile. This position was designated as a lower analysis position S4, which was an area surrounded by these positions S1 to S4. In this case, the analysis end position S2 is set to be sufficiently large so that the tip portion can enter even when the tip portion of the molten pool is largely retracted (FIG. 2 (c)).

  The image analysis means 11 in the signal processing device 9 has a function of setting the analysis area S in the image captured by the CCD camera 8 and the length L and width W of the melting part grasped from the image in the analysis area S. And a function of calculating the L / W ratio based on these L and W. On the other hand, the pass / fail judgment means 12 stores in advance a threshold Ra of the L / W ratio set in the correlation diagram shown in FIG. FIG. 4 shows a result of laser lap welding of two steel plates W1 and W2 having a plate thickness of 0.7 mm as described above. For example, when the amount of shrinkage reaches 50% of the plate thickness, For this purpose, the threshold value Ra may be set to about 1.75. The pass / fail judgment means 12 compares the L / W ratio obtained by the image analysis means 11 with a threshold value Ra stored in advance, and when the L / W ratio exceeds the threshold value Ra, welding failure (retraction) It has a function to determine that it is defective.

  In the laser welding quality evaluation apparatus configured as described above, imaging at a high speed (for example, 30 frames per second) by the CCD camera 8 is started simultaneously with the start of welding, whereby image analysis means in the signal processing apparatus 9 is started. 11 is sent at high speed an image of each frame picked up by the CCD camera 8. The image analysis means 11 sets the analysis area S in the image sent from the CCD camera 8, analyzes the image in the analysis area S, and calculates the ratio between the total length L and the total width W of the high luminance region 21. (L / W ratio), that is, the length / width ratio of the welded portion grasped from the image is obtained, and this data is sent to the quality determination means 12. Then, the pass / fail determination means 12 compares this L / W ratio with a previously stored threshold value Ra (FIG. 4), and determines that it is poor when the L / W ratio exceeds the threshold value Ra. If necessary, this determination result may be displayed on the display device together with the image captured by the CCD camera 8.

  By the way, in laser lap welding, if the gap between the two steel plates W1 and W2 that are the workpieces W is increased, the phenomenon of burn-out occurs, but this phenomenon of burn-out occurs during welding. The case where it occurs and the case where it occurs immediately after the start of welding are different. If burnout occurs during welding, the molten pool tip melts back after abnormally retreating, and welding starts again from that point, so the image immediately after the burnout is close to normal welding ( FIG. 2 (a)). In this case, the full length L of the melted part 5 grasped from the image is once expanded and then suddenly returns to the value of normal welding. Therefore, the movement of the full length L is analyzed by the image analysis means 11. By grasping by, it is possible to grasp the burn-out during welding.

  On the other hand, when burn-off has occurred immediately after the start of welding, the reflected light from the tip of the molten pool (the rear wall portion of the keyhole) cannot be obtained, so the image obtained with the CCD camera 8 is shown in FIG. As shown in (a), at first glance, it is similar to the image at normal welding (FIG. 2 (a)). In this case, it is difficult to clearly distinguish the luminance integration profile in the horizontal axis direction obtained as described above from the luminance integration profile during normal welding (FIG. 5A) as shown in FIG. 6B. It is. However, when the image shown in FIG. 6A is observed in detail, a ring-shaped high luminance portion 21c is observed behind the front high luminance portion 21a (the portion of the keyhole 4). Therefore, when the minimum value filter processing is performed on the image and the luminance integration profile in the horizontal axis direction is obtained for the image after this processing, as shown in FIG. 6C, the three peaks Pa, Pb, and Pc are clear. Appears on Among these, the left peak Pa is due to the reflected light from the front wall portion of the keyhole 4, and corresponds to the peak P1 (FIG. 5) corresponding to the above-described front high brightness portion 21a having an arcuate shape. The two peaks Pb and Pc on the right side are due to the reflected light from both side portions of the ring-shaped high luminance portion 21c. Therefore, it is possible to grasp the burn-out immediately after the start of welding by capturing the characteristics of the luminance integration profile by image analysis by the image analysis unit 11. In this case, it may be characterized by the number of peaks or by the distance between each peak.

  In laser lap welding, a so-called non-penetrating welding phenomenon may occur in which the laser beam A does not penetrate the workpiece W. In such non-penetrating welding, the reflected light from the center of the laser irradiation area becomes extremely strong, so that the image obtained by the CCD camera 8 has the above-mentioned bow shape as shown in the upper part of FIG. A circular high luminance part 21d is present behind the front high luminance part 21a. In this case, as shown in the lower part of FIG. 7B, the luminance integration profile in the horizontal axis direction for the image appears widely on the right side with respect to the left side with the valley portion as the center. On the other hand, at the time of normal welding (through welding), the high luminance region 21 appears in a donut shape as shown in the upper part of FIG. 7A, and the luminance integration profile is shown in FIG. As shown in the lower part, the left side of the valley appears widely with respect to the right side. The left and right widths E1 and E2 centering on the valley correspond to energy, and the comparison between (a) and (b) in FIG. 7 shows that the right and left energy ratio E1 / E2 is relatively large during normal welding. It appears that the left / right energy ratio E1 / E2 appears relatively small during non-penetrating welding. Accordingly, by capturing this energy ratio E1 / E2 ratio by image analysis by the image analysis means 11, it is possible to determine that non-penetrating welding has occurred when the E1 / E2 ratio is equal to or less than a predetermined value. .

  In the above embodiment, the reflected light B of the welding laser beam A is taken into the CCD camera 8 coaxially with the laser beam A. However, in the present invention, the CCD camera 8 is placed on the side of the laser torch 1. It may be arranged so that the reflected light B is taken into the CCD camera 8 non-coaxially with the welding laser light A. However, in this case, since the obtained image is different from the case where the reflected light is captured coaxially, it is necessary to grasp in advance the relationship between the image obtained by the CCD camera and the welding quality.

  Moreover, in the said embodiment, although the example applied to the lap welding of the two steel plates W1 and W2 was shown, this invention can be applied to butt welding of two members, fillet welding, etc. besides this Of course.

It is a mimetic diagram showing one embodiment of a laser welding quality evaluation device concerning the present invention. It is a schematic diagram which shows the image obtained with this laser welding quality evaluation apparatus according to a welding condition. It is a schematic diagram which shows the setting point of the full length and full width with respect to an image, and the setting point of an analysis area | region. It is a graph which shows the correlation with the length / width ratio of the fusion | melting part grasped | ascertained from an image, and measurement shrinkage. 3 is a graph showing a luminance integration profile in the horizontal axis direction for each image shown in FIG. 2. It is a figure which shows the image in case burn-out has generate | occur | produced immediately after the start of welding, the luminance integration profile about the image, and the Kido integration profile after signal processing. It is a figure which contrasts the image and brightness | luminance integration profile at the time of non-penetrating welding with the image and brightness | luminance integration profile at the time of normal welding.

Explanation of symbols

1 Laser torch 4 Keyhole 5 Weld pool (melting part)
7 Optical filter 8 CCD camera (imaging means)
9 Signal processing equipment (signal processing means)
DESCRIPTION OF SYMBOLS 11 Image analysis means 12 Pass / fail judgment means 21 High-intensity part in image A Laser beam for welding B Reflected light of laser light W

Claims (8)

  The fusion part and its periphery are imaged by the reflected light of the laser beam reflected from the laser irradiation area, and the welding quality is evaluated based on the shape characteristic of the fusion part grasped from the taken image. Laser welding quality evaluation method.   2. The laser welding quality evaluation method according to claim 1, wherein a shape characteristic of the melted portion is grasped from a luminance integrated profile for the image.   The length / width ratio of the melted portion is obtained from the luminance integration profile in the welding direction and the luminance integration profile in the direction orthogonal to the welding direction, and if the length / width ratio is greater than a predetermined value, it is determined that the shrinkage is defective. The laser welding quality evaluation method according to claim 2, wherein:   3. The length of the melted portion is obtained from a luminance integrated profile in the welding direction, and when the length rapidly decreases after abnormally expanding, it is determined that burn-out has occurred during welding. The laser welding quality evaluation method described in 1.   3. The laser according to claim 2, wherein when the three peaks centering on the keyhole appear in the luminance integration profile in the welding direction, it is determined that burn-out has occurred immediately after the start of welding. Welding quality evaluation method.   3. The laser according to claim 2, wherein an energy ratio before and after the valley portion of the luminance integration profile in the welding direction is obtained, and non-penetrating welding is determined when the energy ratio is smaller than a predetermined value. Welding quality evaluation method.   An imaging unit that captures an image of the melted part and its periphery by reflected light of laser light reflected from the laser irradiation area, and obtains a shape characteristic of the melted part from an image captured by the image capturing unit, and the shape characteristic A laser welding quality evaluation apparatus comprising signal processing means for determining welding quality based on the signal processing means. 8. The laser welding quality evaluation apparatus according to claim 7, wherein the imaging means is a CCD camera.

Images