JP2007054879A - Hybrid welding equipment, and image processing method and image processing program of hybrid welding equipment - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve a problem that image processing in hybrid welding equipment is not sufficiently conducted so that highly precise control of welding is hard to perform. <P>SOLUTION: In the hybrid welding equipment, welding work is conducted in a combination of a 1st welding and a 2nd welding. The hybrid welding equipment is provided with: an image capturing means 3, 41 for capturing the image of a material to be welded; an image processing means 42 for integrating and smoothing a plurality of images captured by the image processing means; and a welding controlling means for controlling the 1st welding and the 2nd welding according to the image processed by the image processing means. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a hybrid welding apparatus, and an image processing method and an image processing program for the hybrid welding apparatus, and in particular, a hybrid welding apparatus that combines laser welding and arc welding, and an image processing method and an image processing for the hybrid welding apparatus. Regarding the program.

  In recent years, for example, by combining laser welding and arc welding, hybrid welding that incorporates the advantages of both laser welding depth and high-speed penetration due to high energy, and sufficient amount of welding by arc welding (filler wire supplied) Has been researched and developed.

  Here, hybrid welding is mainly a combination of laser welding and arc welding, but is not limited to this, and the present invention is a hybrid that combines other types of welding such as electron beam welding and induction heating welding. It can also be applied to welding. In addition to hybrid welding according to the present invention, in addition to hybrid welding according to the present invention, after performing hybrid welding combining laser welding and arc welding in actual thick plate (eg, pipe) welding, etc. Other types of welding may be performed.

  By the way, in the past, in hybrid welding combining laser welding and MIG (Metal Inert Gas) welding, in order to prevent the weld bead from becoming unstable during high-speed welding, the melt formed on the workpiece during welding An imaging means for imaging the pond, a distance calculating means for obtaining a distance from the laser light irradiation point to the contour of the molten pool based on an image taken by the imaging means, and an irradiation point based on the distance obtained by the distance calculating means And a welding means (filler wire) has been proposed (see Patent Document 1, for example).

  Conventionally, there has been proposed an automatic welding amount control method for arc welding in which the groove width is measured with high accuracy and the required welding amount is appropriately controlled according to the variation of the groove width. That is, the inventors of the present invention measure the reference weld pool width from the first layer to the last two layers in the reference welding, and the ratio between the initial layer smoothed weld pool width and the reference weld pool width data is set in advance. The second layer target welding speed is calculated based on the second layer reference welding speed, the second layer welding is performed at this speed, and the weld pool width is measured. Similarly, from the next third layer to the last layer one layer before The final layer welding is performed at the target welding speed calculated based on the ratio of the smoothed weld pool width data before the final layer and the reference weld pool width data and the preset final layer reference welding speed. An automatic welding amount control method for narrow gap multi-layer arc welding has been proposed (see, for example, Patent Document 2).

  Furthermore, conventionally, a TIG (Tungsten) that can stably separate and detect welding electrodes, weld pools, groove walls, and welding wires having different luminance levels from a captured image, and can perform stable control based on image measurement values. Inert Gas) An automatic control method for arc welding has also been proposed. That is, in the luminance image obtained by imaging the region including the welding electrode, the molten pool, the groove wall, and the welding wire by the present inventors, the luminance difference between the welding electrode and the molten pool, the luminance difference between the molten pool and the groove wall. Based on the measured values of the welding electrode, groove wall, molten pool, and welding wire brightness images captured at exposure times such that the brightness differences between the welding wire and the weld pool are relatively large, respectively. An automatic control method of TIG arc welding that performs center line scanning control, swing width control, and welding wire insertion position control has also been proposed (see, for example, Patent Document 3).

JP 2003-181664 A Japanese Patent Laying-Open No. 2005-081418 Japanese Patent Laying-Open No. 2005-081419

  As described above, conventionally, in hybrid welding in which laser welding and MIG welding are combined, the molten pool formed on the workpiece is imaged by the imaging means, and the image is processed by the distance calculating means, and the contour of the molten pool is obtained. And a laser beam irradiation point, and a distance from the laser beam irradiation point to the outline of the molten pool is obtained, and the laser beam irradiation point is determined based on the distance obtained by the distance calculation unit by the moving unit. There has been proposed one that changes the relative position of the supply nozzle of the filler wire.

  In other words, for the welding speed determined by the plate thickness of the workpiece, the distance between the laser beam irradiation point and the supply nozzle (welding wire) is optimally controlled so that the irradiation point is positioned in the molten pool. There has been proposed a hybrid welding technique in which a stable weld bead is always formed regardless of the speed, and a high-quality weld is obtained.

  However, for example, in a hybrid welding apparatus that combines laser welding and arc welding, an apparatus that performs image processing of a melted part and appropriately controls each of laser welding and arc welding has not been put to practical use.

  By the way, for example, the major difference between arc welding and hybrid welding combining laser welding and arc welding (in the case of the first layer of narrow groove multi-layer welding) is the penetration depth and welding speed. In particular, the welding speed is around 20 cm / min in arc welding, whereas it can be around 2 m / min in hybrid welding. Therefore, the control speed in hybrid welding needs to be considerably higher than that in arc welding.

  FIG. 1 is a diagram for explaining an example of a molten pool image capturing timing in an image processing method of a hybrid welding apparatus as a related technique. In FIG. 1, reference numeral 10 denotes a laser irradiation point (laser processing point emission), 20 denotes a welding wire (filler wire), and 100 denotes a molten pool.

  First, when the arc welding torch (welding wire 20) is swinging, image measurement at the swing end is required. That is, the rocking width of the welding torch can be calculated by a pair of left and right image measurements. For example, when the rocking frequency is 30 Hz, the distance from the left end to the right end is 16.7 ms. In this case, an interval of 33.3 ms or more is necessary. Therefore, the measurement interval MP of the left and right ends of the welding torch is delayed by one cycle to 50 ms, and the measurement cycle is 20 times / second. Furthermore, in order to avoid the influence of arc fluctuations, for example, an integrated smoothing process of at least about 3 frames is required, and as a result, the measurement cycle is 6.7 times / second.

  Furthermore, in order to use the measured values obtained in this way for controlling welding, it is necessary to eliminate the influence of false detection. For example, the median value processing is performed using the measured values of 5 points (5 times). Etc., and as a result, the control cycle is about 1 second. Here, the welding length of arc welding is, for example, about 3 mm per second, but the welding length of hybrid welding is, for example, about 30 mm per second. Therefore, in a control cycle of about 1 second, hybrid welding is performed. It is difficult to perform control.

  Further, as shown in FIG. 1, for example, three images L1, L2, and L3 when the welding torch is moving to the left end are integrated and smoothed to obtain a left end smooth image LA0. The three images R1, R2 and R3 when moving to the right end are integrated and smoothed to obtain a right end smoothed image RA0.

  At this time, regarding the right end smoothed image RA0, for example, when the three images R1, R2, and R3 used for the integration smoothing process are normal (appropriate exposure), the right end smoothed image RA0 is also normal. However, regarding the left-end smoothed image LA0, for example, when only the image L2 is normal and the images L1 and L3 are over-exposed (overexposure), the wire end portion becomes unclear in the left-end smoothed image LA0, and welding control is performed. It becomes difficult to carry out with high accuracy.

  An object of the present invention is to provide a hybrid welding apparatus capable of performing high-precision welding control, and an image processing method and an image processing program for the hybrid welding apparatus.

  According to the first aspect of the present invention, there is provided a hybrid welding apparatus that combines the first welding and the second welding, the image capturing unit capturing an image of the material to be welded, and the image capturing unit. Image processing means for integrating and smoothing a plurality of images, and welding control means for controlling the first welding and the second welding in accordance with the images processed by the image processing means. A hybrid welding apparatus is provided.

  According to the second aspect of the present invention, there is provided an image processing method for a hybrid welding apparatus that performs a combination of the first welding and the second welding. The image processing method captures a plurality of images of a material to be welded. An image processing method for a hybrid welding apparatus is provided, in which an image used for control of the first welding and the second welding is generated by performing an integration smoothing process on the first image and the second welding.

  According to the third aspect of the present invention, there is provided an image processing program for a hybrid welding apparatus that performs a combination of the first welding and the second welding, and causes a computer to capture a plurality of images of a material to be welded. An image processing program for a hybrid welding apparatus is provided, wherein an image used for controlling the first welding and the second welding is generated by subjecting a plurality of captured images to integration smoothing processing.

  According to a fourth aspect of the present invention, there is provided a medium on which a program to be executed by a computer is recorded. The program causes the computer to capture a plurality of images of a material to be welded, and the plurality of captured images. Is provided for a medium on which an image processing program of a hybrid welding apparatus is recorded, wherein an image used for controlling the first welding and the second welding is generated.

  ADVANTAGE OF THE INVENTION According to this invention, the hybrid welding apparatus which can perform highly accurate welding control, the image processing method and image processing program of a hybrid welding apparatus can be provided.

  Embodiments of a hybrid welding apparatus, an image processing method of the hybrid welding apparatus, and an image processing program according to the present invention will be described below in detail with reference to the accompanying drawings.

  FIG. 2 is a view showing an example of an image of a CMOS camera in the hybrid welding apparatus according to the present invention. In FIG. 2, reference numeral 10 indicates a laser irradiation point (laser processing point emission), 20 indicates a welding wire (filler wire), and 100 indicates a molten pool.

  Here, as will be described in detail later, the CMOS camera provides, for example, an image of a laser welded portion (molten pool) formed by irradiating a workpiece with laser light on the same axis with respect to the irradiated laser light. It has been.

  As shown in FIG. 2, the detection items in the image captured by the CMOS camera of the hybrid welding apparatus according to the present invention include laser processing point position (X coordinate) LX, laser processing point position (Y coordinate) LY, melting Pond end positions (X coordinate) PL, PR and weld pool center PXC (= (PL + PR) / 2), welding wire entry position (Y coordinate) WY, and welding wire position (X coordinate) WXL, WXR and oscillation center WXC (= (WXR + WXL) / 2).

  FIG. 3 is a view schematically showing an embodiment of the hybrid welding apparatus according to the present invention. In FIG. 3, reference numeral 1 is a laser head, 11 is a laser beam, 2 is a welding torch, 20 is a welding wire, and 3 is a CMOS camera. In addition, the welding direction in FIG. 3 shows the relative moving direction of the hybrid welding apparatus provided with the laser welded part (laser head 1) and the arc welded part (welding torch 2) and the workpiece (base material). For example, when the material to be welded is fixed, the hybrid welding apparatus moves from the left back to the right front in FIG.

  As shown in FIG. 3, the CMOS camera 3 is arranged coaxially with the laser beam 11 irradiated from the laser head 1 to the workpiece. In addition, the CMOS camera 3 is provided with an interference filter (transmission main wavelength 955 nm) in order to better capture the molten pool radiation light and the laser processing point emission by eliminating the influence of the arc light. The wire 20 and the laser irradiation point (laser processing point) 10 can be imaged at high speed (for example, 2000 frames per second).

  FIG. 4 is a block diagram schematically showing one embodiment of the hybrid welding apparatus according to the present invention. In FIG. 4, reference numeral 4 denotes a control computer (control PC), 42 denotes an image processing board, and 43 denotes a serial interface (serial I / F).

  As shown in FIG. 4, the video from the CMOS camera 3 is input to the image processing board 42. The asynchronous shutter trigger is output from the laser head left-right axis servo controller (swing controller) and input to the image processing board 42, and the shutter trigger for the CMOS camera 3 is output from the image processing board 42. Then, the image data processed by the image processing board 42 is output to the control device via the serial interface 43 (for example, RS-232C).

  FIG. 5 is a view for explaining laser welding line center scanning control in one embodiment of the hybrid welding apparatus according to the present invention.

  As shown in FIG. 5, the laser welding line center copying is performed by the center LX of the laser processing point 10 in the welded portion image captured by the CMOS camera 3 and the center PXC of the welding line (that is, the groove center = the molten pool 100). The processing point position is controlled so that the (center) matches. In the hybrid welding apparatus of this embodiment, since the CMOS camera 3 provided coaxially with the laser beam 11 is interlocked with the laser head 1 (see FIG. 3), the laser processing point 10 is stationary and melted. The pond 100 (groove) appears to move. Then, a difference δX between the center PXC of the molten pool 100 and the center LX of the laser processing point 10 is obtained, and the position of the laser head 1 is controlled so that this becomes zero.

δX = LX−PXC = LX− (PR−PL) / 2
Here, assuming that the measurement cycle time is Δt and the control delay time is td, in this embodiment, Δt> td. However, since the target work (material to be welded) is a straight line, it is sufficient for the copying control.

  A value obtained by multiplying the center position deviation amount δX by the horizontal pixel resolution (mm / pixel) is transmitted as the position control amount of the laser processing point. Note that the horizontal resolution is, for example, 0.015 mm / pixel, and when a variation of about 10 pixels is estimated, the accuracy is ± 0.15 mm.

  FIG. 6 is a view for explaining arc welding line center scanning control in one embodiment of the hybrid welding apparatus according to the present invention.

  First, the arc welding line center scanning control is independent of the control of the laser head 1 using the image of the molten pool 100 (welding wire 20) imaged by the CMOS camera 3.

  As shown in FIG. 6, the arc welding line center scanning control is performed by changing the welding wire intrusion position WXL from the swing left end image (see FIG. 6A) and the WXR from the right end image (see FIG. 6B). Measure and calculate the swing center position WXC of the welding wire 20 (arc welding torch) from these.

WXC = (WXL + WXR) / 2
At the same time, the center PXC of the weld pool 100 (average value of the center of the weld pool at the left and right ends of the swing) is also calculated, and the difference value between the swing center position WXC and the weld pool center PXC, that is, the arc welding line center displacement δX. Ask for.
δX = PXC-WXC

  FIG. 7 is a view for explaining laser / arc distance control in one embodiment of the hybrid welding apparatus according to the present invention.

  As shown in FIG. 7, the laser-arc distance control is performed by using an arc welding torch (welding wire 20) in order to keep the distance between laser welding and arc welding constant using an image captured by the CMOS camera 3. ) Position control is required. The wire entry position WYL is measured from the image captured at the left end of the swing (see FIG. 7A), and the laser-arc distance δLA-L at the left end of the swing is calculated (before laser welding).

δLA-L = LY-WYL
Next, the wire entry position WYR is measured from the image captured at the right end of the swing (see FIG. 7B), and the laser-arc distance δLA-R at the left end of the swing is calculated.

δLA-R = LY-WYR
Then, the average δLA of the distance between the laser and the arc calculated at the left end of the swing and at the right end becomes a desired value.

δLA = (δLA-L + δLA-R) / 2
Here, when the reference laser-arc distance is LAstd, and the direction in which the welding torch advances (the direction approaching the laser processing point) is +, the control amount ΔL is
ΔL = δLA−LAstd
It becomes. If ΔL is negative (−), the welding torch 2 (welding wire 20) is retracted from the laser processing point 10.

  FIG. 8 is a view for explaining the swing width control in one embodiment of the hybrid welding apparatus according to the present invention. FIG. 8 (a) shows the reference molten pool width and swing width, and FIG. Indicates the current weld pool width and rocking width.

The rocking width control is performed so as to have a rocking width corresponding to the fluctuation of the groove width (root gap), and the weld pool width PWstd (= PRstd−PLstd: FIG. 8 ( a) and a difference value between the welding wire swinging width EWstd (= WXRstd−WXLstd: see FIG. 8B) is recorded. When the actual weld pool width is PWcur (= PRcur-PLcur) and the actual oscillation width is EWcur,
ΔW = (PWcur−EWcur) − (PWstd−EWstd)
ΔW is transmitted to the control device as the fluctuation range increase / decrease value.

  In the image processing method of the hybrid welding apparatus of the present invention, for example, an image of the weld pool 100 is taken (imaged) using the high-speed CMOS camera 3, and the oscillation center and oscillation width of the arc welding torch 2 are measured. In order to do this, a weld pool image (welding wire image) at the swing end is required. Furthermore, in order to integrate and smooth a plurality of frames in the vicinity of the swing end imaged by the high-speed CMOS camera 3, for example, a trigger for notifying the start of integration is used to swing the wire from the swing axis (left and right axis) servo controller. Output at a position of 1.5 ms before the end.

  FIG. 9 is a view for explaining an example of the image integration smoothing process in the image processing method of the hybrid welding apparatus according to the present invention.

  As shown in FIG. 9, for example, the original images I−1 to In captured by the CMOS camera 3 are images of the welding wire 20 due to the influence of arc light 200, laser plume light, or the like depending on the imaging timing of the camera. In some cases, the image becomes unclear or the brightness of the entire image decreases due to an arc short circuit, making it difficult to distinguish the welding wire 20. Therefore, in the image processing method of the hybrid welding apparatus of the present invention, luminance is integrated (for example, three images are integrated) by the adder 401, and further, bit shift is performed by the bit shift circuit 402 for smoothing processing. Then, an image IA obtained by integrating and smoothing the three images is generated.

  As is clear from FIG. 9, in the image IA that has been subjected to the integrated smoothing process, the weld pool 100, the welding wire 20, and the like can be clearly recognized, and by using the image IA that has been subjected to the integrated smoothing process, Each control of the laser welding and arc welding described above can be performed accurately.

  10A and 10B are diagrams for explaining an example of the change in the arc light amount in the image processing method of the hybrid welding apparatus according to the present invention. FIG. 10A is an image when the arc light amount is appropriate, and FIG. FIG. 10C shows an image when the light intensity is excessive, and FIG. 10C shows an image when the arc light intensity is excessive.

  First, as shown in FIG. 10A, the image when the arc light amount is appropriate is that when the arc light amount and the molten pool radiation light are appropriate, and the light emission is uniformly spread in the groove, and the laser Processing point light emission (laser irradiation point) 10, welding wire 20, and molten pool radiation (molten pool 100) can be clearly identified. On the other hand, as shown in FIG. 10B, in the image when the arc light amount is excessive, the arc light 200 may be biased or the end of the welding wire 20 may be blocked. As shown in c), in the image when the arc light quantity is too small, the light emission is too weak and the welding wire 20 and the groove wall become unclear.

  Therefore, in the image processing method (hybrid welding apparatus) of the hybrid welding apparatus according to the present invention, an image used for control of laser welding and arc welding is generated by integrating and smoothing a plurality of images.

  FIG. 11 is a diagram for explaining an example of the molten pool image capturing timing in the image processing method of the hybrid welding apparatus according to the present invention.

  As described above, the image processing method of the hybrid welding apparatus according to the present invention generates an image used for control of laser welding and arc welding by integrating and smoothing a plurality of images.

  However, for example, when many of the images to be integrated are unclear due to the magnitude of the arc light amount (for example, 2 of the 3 screens to be integrated are dark and the wire image is unclear), the integration smoothing process is performed. Even if it goes, the effect of improving the image quality is not obtained. Therefore, as shown in FIG. 11, for example, seven screens L11 to L17 are captured every 0.5 ms in ± 1.5 ms before and after the rocking end of the welding torch 2 (welding wire 20), and the appropriate brightness is obtained. Only the three screens L12, L14, and L17 are subjected to integration smoothing processing to generate a left-end smoothed image LA. Similarly, the right-end smoothed image RA is also generated by performing integration smoothing processing only on a screen with appropriate brightness.

  In FIG. 11, screens L11 and L13 are those when the arc light quantity is too low (arc short circuit: NG1), and screens L15 and L16 are those when the arc light quantity is excessive (NG2). , L13, L15, and L16 are not used for the integration smoothing process.

  Here, if the arc fluctuation is sufficiently fast, a camera capable of high-speed capture (such as a CMOS camera) is used, and images are captured at intervals of about 0.5 ms in the vicinity of the swing end (± 1.5 ms). If the integrated smoothing process is performed by taking in, for example, when the arc welding torch 2 is swinging at 30 Hz, the measurement cycle is 30 times / second (a pair of left and right). Here, the measurement cycle is determined by the oscillation cycle. When the oscillation width is 2 mm and the oscillation amplitude is 30 Hz, the position of the welding wire 20 at ± 1.5 ms from the left and right oscillation ends is about ± 0.04 mm. When the diameter is 1.2 mm, the deviation is only ± 3.3%, and it can be seen that the integration smoothing process is sufficient.

  In order to use the measured value for control, smoothing processing such as median value processing is required from the measured value of 5 to 10 (or 2 to 5) points in order to eliminate the influence of false detection. The cycle is about 0.17 to 0.33 (or 0.07 to 0.17) seconds. Further, by excluding an image unsuitable for the integration smoothing process (the arc light amount is excessively small or excessive) by the magnitude of the luminance integrated value, a good smoothed image can be obtained. Furthermore, since the image quality of the smoothed image is improved, the number of median processing points for reducing the influence of erroneous detection can be reduced, and the control cycle can be shortened.

  In the above, the images (screens) used for the integrated smoothing process are images before and after the rocking end of the welding wire 20, but are not limited to images of ± 1.5 ms before and after the rocking end. The number of images used for the integration smoothing process is not limited to three.

  FIG. 12 is a diagram for explaining an example of appropriate image determination based on an integrated image luminance value in the image processing method of the hybrid welding apparatus according to the present invention.

As shown in FIG. 12, in order to determine an appropriate brightness, the luminance integrated value SUM-P in the set rectangular range in the image is set to threshold values Th1 and Th2 that are set in advance.
Th1 <SUM-P <Th2
It is necessary to determine whether or not

  That is, the luminance integrated value in the rectangular area (Xs, Ys) to (Xe, Ye) is calculated, and if the value is within the range of preset threshold values Th1 to Th2, an effective image is obtained, and the smoothed data Used for generation. In addition, when the luminance integrated value in the rectangular areas (Xs, Ys) to (Xe, Ye) is less than the threshold value Th1 or exceeds the threshold value Th2, it is determined as an invalid image.

  That is, out of a plurality of images taken by the CMOS camera 3, the image includes the images of both ends of the weld pool 100 and the welding wire 20, and does not include the image of the laser irradiation point (laser processing light emission point) 10. The luminance integrated value SUM-P in the area is calculated. Then, an image in which the calculated luminance integrated value SUM-P is within a predetermined range (Th1 <SUM-P <Th2) is used for the integration smoothing process.

  Then, based on an image generated by integrating and smoothing an image in which the luminance integrated value P in the rectangular region is within a predetermined range, the center scanning control of the laser welding line in laser welding, the laser of laser welding Control of the distance between the irradiation point and the welding wire tip of arc welding, arc welding line center scanning control of arc welding, swing width control of the welding wire, and the like are performed.

  FIG. 13 is a diagram for explaining an example of a molten pool end detection process in the image processing method of the hybrid welding apparatus according to the present invention.

  As shown in FIG. 13, in order to obtain the center of the molten pool 100, the left and right ends of the molten pool 100 are set so that the groove crosses the measurement line, and the luminance change on the line is examined. Detect by. Here, three measurement lines DL <b> 1 to DL <b> 3 are set, and the median value among the three X coordinate points detected by the lines DL <b> 1 to DL <b> 3 is the end of the molten pool 100.

  In the case of the left end point, primary differentiation and secondary differentiation are performed on the luminance data on the measurement lines DL1 to DL3 from the outside of the weld pool 100 toward the center. A point at which the primary differential value exceeds the preset threshold value Th10 for the first time and the secondary differential value crosses zero (+ → −) immediately after that is detected. And However, it is necessary that the first-order differential value at the zero cross point is equal to or greater than the threshold value Th10. If the search point exceeds the set range, a detection error is assumed. In the case of the right end point, the process is similarly performed from the outside of the molten pool 100 toward the center.

  Since the laser processing point (laser irradiation point) is seen on the same axis by the CMOS camera 3, its position is fixed in the image, and its coordinates can be known in advance before welding.

  Thus, the molten pool edge part and laser processing point in the image of the coaxial camera (CMOS camera 3) required for laser welding line center copying can be detected.

  FIG. 14 is a diagram for explaining an example of a welding wire detection process in the image processing method of the hybrid welding apparatus according to the present invention.

  Detection of the position of the welding wire 20 is necessary for arc welding line center scanning control and oscillation width scanning control. First, a measurement line for detecting the welding wire entry position is set.

  As shown in FIG. 14, the position of the welding wire 20 is detected by using the measurement lines DL1 to DL3 used to detect the left and right ends of the weld pool 100 described with reference to FIG. The position in the X direction (perpendicular to the traveling direction) is detected.

  The luminance data of the measurement line DL1 is subjected to moving average processing to obtain the minimum value in the inner range of the left and right ends of the weld pool 100, and this is set as the X coordinate candidate of the welding wire 20. Further, the same processing is performed on the remaining two measurement lines DL2 and DL3, and the X coordinate value obtained that has the median value is defined as the X coordinate WXL (at the left end of the swing) of the welding wire 20. Further, the luminance value at that time is P (WXL). Note that the X coordinate of the welding wire 20 at the right end of the swing is WXR, and the luminance value at that time is P (WXR)).

  This X coordinate is set as the wire penetration position measurement line DL4, the luminance data is examined from the lower side of the image toward the laser machining point, and the point where the value exceeds the set threshold value (= P (WXL) * α). The wire entry position candidate WYL (or WYR) is used. Furthermore, the wire entry position WY is an average value of WYL and WYR [(WYL + WYR) / 2]. The swinging width EW is EW = WYR−WYL.

  15 and 16 are flowcharts for explaining an example of image processing of the hybrid welding apparatus according to the present invention. Note that the flowcharts of FIGS. 15 and 16 use the five-point (five times: M = 5) measurement values to remove the influence of false detection from the measurement values and use it for welding control. An example is shown in which a smoothing process such as the above is performed and a screen is captured every 0.5 ms within ± 1.5 ms before and after the rocking end of the welding torch 2 (welding wire 20).

  First, when image processing of the hybrid welding apparatus is started, in step ST1, the number of times of measurement M is set to 1, and the process proceeds to step ST2 to determine whether or not the welding torch 2 is 1.5 ms before the swing end. If it is determined in step ST2 that it is 1.5 ms before the swing end, the process further proceeds to step ST3 to determine whether it is the swing right end side, and is not 1.5 ms before the swing end. If it is determined, the process returns to step ST2, and the process is repeated until it is determined that it is 1.5 ms before the swing end.

  If it is determined in step ST3 that it is on the right side of the swing, the process proceeds to step ST4, and one screen is captured. Further, the process proceeds to step ST5. To go to step ST10.

  In step ST5, as described with reference to FIG. 12, it is determined whether or not the integrated luminance value of the screen captured in step ST4 is within the set range, and if it is determined that the integrated luminance value is within the set range. Then, the process proceeds to step ST6, image integration / smoothing processing is performed, and the process proceeds to step ST7. Similarly, in step ST10, it is determined whether or not the luminance integrated value of the screen captured in step ST9 is within the setting range. If it is determined that the luminance integrated value is within the setting range, the process proceeds to step ST11, Integration / smoothing is performed, and the process proceeds to step ST12.

  In step ST7, it is determined whether or not the swing end is 1.5 ms later. If it is determined that the swing end is 1.5 ms later, the process proceeds to step ST14, and it is determined that the swing end is not 1.5 ms later. Then, the process proceeds to step ST8, is delayed by 0.5 ms, returns to step ST4, and the same process is repeated until it is determined in step ST7 that the swing end is 1.5 ms later. Similarly, in step ST12, it is determined whether or not the swing end is 1.5 ms later. If it is determined that the swing end is 1.5 ms later, the process proceeds to step ST14. If it is determined that there is not, the process proceeds to step ST13, and after a delay of 0.5 ms, the process returns to step ST9, and the same processing is repeated until it is determined in step ST12 that the oscillation end is 1.5 ms later.

  In step ST14, it is determined whether there are M pairs on the left and right of the swing end. If it is determined that there are M pairs, the process proceeds to step ST15, and it is determined that there are no M pairs. Then, the process returns to step ST2, and the same processing is repeated until it is determined that there are M pairs.

  In step ST15, as described above, the center of the Mth molten pool, the molten pool width, the swinging width, the distance between the wire entry position and the laser processing point, and the like are calculated, and the process proceeds to step ST16. In step ST16, it is determined whether M is 5 or more (M ≧ 5). If it is determined that M is not 5 or more (M = 1, 2, 3, 4), the process proceeds to step ST17. The process returns to step ST2 with M = M + 1, and if it is determined that M is 5 or more (M = 5), the process proceeds to step ST18.

  In step ST18, smoothing processing of the calculated values such as the molten pool center, the molten pool width, the swinging width, the distance between the wire entry position and the laser processing point (average of five calculated values: integrated smoothing processing) is performed. Go to step ST19. In step ST19, each control amount is calculated and transmitted according to the image obtained by integrating and smoothing a plurality of images in step ST18, and the process proceeds to step ST20 to determine whether or not to end the process, and to determine that the process is to end. Until the above-described step ST1 is repeated.

  The control amounts in step ST19 are, for example, laser beam welding line center scanning control in laser welding, distance control between the laser irradiation point and the arc welding welding wire tip, and arc welding arc welding line center scanning control. And a control amount used for controlling the swinging width of the welding wire.

  Here, the number of times of measurement (M) performed to eliminate the influence of false detection from the measured value is not limited to five, and the screen and the timing of capturing are set to 0. Of course, it is not limited to fetching every 5 ms.

FIG. 17 is a view schematically showing another embodiment of the hybrid welding apparatus according to the present invention.
As is clear from a comparison between FIG. 17 and FIG. 3 described above, the hybrid welding apparatus of the present embodiment corresponds to a hybrid welding apparatus shown in FIG.

  The CCD camera 5 is arranged obliquely in front of the welding direction (for example, an angle of about 45 °), and is provided with an interference filter (transmission main wavelength is 955 nm) similar to that of the CMOS camera 3, and measures the laser processing point position.

  FIG. 18 is a block diagram schematically showing another embodiment of the hybrid welding apparatus according to the present invention. In FIG. 18, reference numeral 4 is a control computer (control PC), 42 is an image processing board, and 43 is a serial interface (serial I / F).

  As shown in FIG. 18, the video of the CMOS camera 3 is input to CH1 of the image processing board 42, and the video of the CCD camera 5 is input to CH0 of the image processing board 42. The asynchronous shutter trigger is output from the laser head left / right axis servo controller and input to the image processing board 42. A shutter trigger for the CMOS camera 3 is output from the image processing board 42.

  FIG. 19 is a view showing an example of an image of a CCD camera in the hybrid welding apparatus according to the present invention.

  As shown in FIG. 19, detection items in an image captured by the CCD camera 5 of the hybrid welding apparatus according to the present invention include a laser processing point position (X coordinate) lx and a laser processing point position (Y coordinate). ly.

  FIG. 20 is a view for explaining laser head height tracking control in another embodiment of the hybrid welding apparatus according to the present invention.

  As shown in FIG. 20, the laser head height scanning control is performed by measuring the laser processing point center coordinate ly image in the welded portion image captured by the CCD camera 5 and using the difference value δly from the reference coordinate value lystd to determine the laser head. A height fluctuation value Δd is calculated.

If y-res is the image resolution,
Δd = δly * cos θ * yres
= (Lystd−ly) * cos θ * yres
Here, the image resolution yres is 0.03 mm / pixel, and when θ is 45 °, the height detection resolution is 0.04 mm / pixel.

  Thus, by providing the CCD camera 5, it is possible to perform laser head height copying control.

  FIG. 21 is a diagram for explaining an example of horizontal / vertical projection processing in the image processing method of the hybrid welding apparatus according to the present invention.

  In the laser processing point detection processing by the CCD camera 5, the laser processing point position in the image of the CCD camera 5 is used for height measurement. Since laser processing point emission is picked up brighter than arc light in an image that has passed through a 955 nm interference filter having a near infrared wavelength, it can be detected by performing binarization processing. However, since the arc emission may be strong depending on the case, the influence of the arc light can be reduced by performing the integration smoothing processing on a plurality of images as in the coaxial camera (CMOS camera 3).

  As shown in FIG. 21, the position detection uses horizontal direction projection processing (projection). This process is to integrate the luminance values on one line in the horizontal direction. If this is sequentially performed in the vertical direction, the vertical distribution ΣP (i, Y) of the projection values can be obtained. As is clear from FIG. 21, for example, in a portion where the bright part is long in the horizontal direction, the integrated value of luminance (gray value projection data) appears large, and the point where this projection data is maximum is the vertical coordinate of the laser processing point. It is.

  Similarly, the vertical direction projection process is to integrate luminance values on one line in the vertical direction. When this vertical direction projection process is sequentially performed in the vertical direction, a horizontal distribution ΣP (X, j) of projection values is obtained. As can be seen, the point where the projection data is maximum is the horizontal coordinate of the laser processing point.

  As described above, the vertical and horizontal coordinates (X coordinate: lx and Y coordinate: ly) of the laser processing point can be detected.

  FIG. 22 is a diagram for explaining an example of a medium on which a quality determination program for a laser weld according to the present invention is recorded. In FIG. 22, reference numeral 310 denotes a processing device, 320 denotes a program (data) provider, and 330 denotes a portable storage medium.

  The above-described laser weld quality determination method according to the present invention is given as a program (data) for the processing device 310 as shown in FIG. 22 and executed by the processing device 310, for example. The processing device 310 includes an arithmetic processing device main body 311 including a processor, and a processing device side memory (for example, a RAM (Random Access) for storing a result of giving a program (data) to the arithmetic processing device main body 311 or processing. Memory), hard disk) 312 and the like. The program (data) provided to the processing device 310 is loaded and executed on the main memory of the processing device 310.

  The program (data) provider 320 has means (line-destination memory: for example, DASD (Direct Access Storage Device)) 321 for storing the program (data). For example, the program (data) is provided via a line such as the Internet. Is provided to the processing device 310, or is provided to the processing device 310 via a portable storage medium 330 such as a CD-ROM, an optical disk, or a floppy disk. Needless to say, the medium on which the quality determination program for the laser welded portion according to the present invention is recorded includes the processing device side memory 312, the line destination memory 321, the portable storage medium 330, and the like. .

  The present invention is mainly applied to hybrid welding in which laser welding and arc welding are combined, but is not limited thereto, and can be applied to hybrid welding in which various types of welding are combined. Moreover, the hybrid welding to which the present invention is applied can be used in combination with other types of welding that are conventionally known.

It is a figure for demonstrating the example of the molten pool image taking-in timing in the image processing method of the hybrid welding apparatus as related technology. It is a figure which shows an example of the image of the CMOS camera in the hybrid welding apparatus which concerns on this invention. It is a figure showing typically one example of a hybrid welding device concerning the present invention. 1 is a block diagram schematically showing an embodiment of a hybrid welding apparatus according to the present invention. It is a figure for demonstrating laser welding line center scanning control in one Example of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating arc welding line center scanning control in one Example of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating the laser-arc distance control in one Example of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating the rocking | swiveling width control in one Example of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating an example of the image integration smoothing process in the image processing method of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating the example of the change of the arc light quantity in the image processing method of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating the example of the molten pool image taking-in timing in the image processing method of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating the example of the appropriate image determination by the image luminance integrated value in the image processing method of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating an example of the molten pool edge part detection process in the image processing method of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating an example of the welding wire detection process in the image processing method of the hybrid welding apparatus which concerns on this invention. It is a flowchart (the 1) for demonstrating an example of the image processing of the hybrid welding apparatus which concerns on this invention. It is a flowchart (the 2) for demonstrating an example of the image processing of the hybrid welding apparatus which concerns on this invention. It is a figure which shows typically the other Example of the hybrid welding apparatus which concerns on this invention. It is a block diagram showing roughly other examples of a hybrid welding device concerning the present invention. It is a figure which shows an example of the image of the CCD camera in the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating the laser head height copy control in the other Example of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating an example of the horizontal and vertical projection process in the image processing method of the hybrid welding apparatus which concerns on this invention. It is a figure for demonstrating the example of the medium which recorded the quality determination program of the laser welding part which concerns on this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Laser head 2 Welding torch 3 CMOS camera 4 Control computer 5 CCD camera 10 Laser irradiation point (laser processing point)
11 Laser light 20 Welding wire (filler wire)
42 image processing board 43 serial interface 100 molten pool 310 processing device 311 arithmetic processing device main body 312 processing device side memory 320 program (data) provider 330 portable storage medium

Claims (22)

A hybrid welding apparatus that combines the first welding and the second welding,
Image capturing means for capturing an image of a material to be welded;
Image processing means for integrating and smoothing a plurality of images captured by the image capturing means;
A hybrid welding apparatus, comprising: a welding control unit that controls the first welding and the second welding according to an image processed by the image processing unit. The hybrid welding apparatus according to claim 1,
The first welding is laser welding; and
The hybrid welding apparatus, wherein the second welding is arc welding.   The hybrid welding apparatus according to claim 2, wherein the image capturing unit includes a CMOS camera capable of high-speed image capturing.   3. The hybrid welding apparatus according to claim 2, wherein the CMOS camera is provided at a position for taking an image of the welded material coaxially with respect to a laser beam irradiated in the laser welding. Hybrid welding equipment.   3. The hybrid welding apparatus according to claim 2, wherein the image capturing means captures images before and after a rocking end of a welding wire used for the arc welding.   3. The hybrid welding apparatus according to claim 2, wherein the image processing unit calculates a luminance integrated value in a predetermined region in the image among the plurality of images captured by the image capturing unit, and calculates the calculated luminance. A hybrid welding apparatus characterized in that an image having an integrated value within a predetermined range is used for the integrated smoothing process.   7. The hybrid welding apparatus according to claim 6, wherein the predetermined area in the image for calculating the luminance integrated value is a rectangular area that includes images of both ends of the weld pool and the welding wire and does not include an image of the laser irradiation point. A hybrid welding apparatus characterized in that there is.   The hybrid welding apparatus according to claim 7, wherein the welding control means includes a center scanning control of a laser welding line in the laser welding, a control of a distance between a laser irradiation point of the laser welding and a welding wire tip of the arc welding. A hybrid welding apparatus that performs arc welding line center scanning control of the arc welding or swing width control of the welding wire. The hybrid welding apparatus according to claim 4, further comprising:
A hybrid welding apparatus comprising a CCD camera that captures an image of the material to be welded at a predetermined angle with respect to the surface of the material to be welded.   10. The hybrid welding apparatus according to claim 9, wherein the CCD camera measures a laser processing point position by the laser welding. An image processing method for a hybrid welding apparatus that combines first welding and second welding,
Capture multiple images of the material to be welded,
An image processing method for a hybrid welding apparatus, characterized in that an integrated smoothing process is performed on the plurality of captured images to generate an image used for controlling the first welding and the second welding. In the image processing method of the hybrid welding apparatus of Claim 11,
The first welding is laser welding; and
The image processing method for a hybrid welding apparatus, wherein the second welding is arc welding.   13. The hybrid welding apparatus image processing method according to claim 12, wherein the plurality of images of the welded material are captured using a CCD camera capable of high-speed imaging. Processing method.   13. The image processing method for a hybrid welding apparatus according to claim 12, wherein the CCD camera is provided at a position for taking an image of the material to be welded coaxially with respect to a laser beam irradiated in the laser welding. An image processing method for a hybrid welding apparatus.   13. The hybrid welding apparatus image processing method according to claim 12, wherein the plurality of images of the workpiece include images before and after a rocking end of a welding wire used for the arc welding. Image processing method.   The image processing method of the hybrid welding apparatus according to claim 12, wherein an image used for controlling the laser welding and the arc welding is an integrated luminance value in a predetermined region in the image among the plurality of captured images. An image processing method for a hybrid welding apparatus, characterized in that an image in which the calculated luminance integrated value is within a predetermined range is generated by the integrated smoothing process.   17. The image processing method for a hybrid welding apparatus according to claim 16, wherein the predetermined area in the image for calculating the integrated luminance value includes images of both ends of the weld pool and a welding wire, and includes an image of a laser irradiation point. An image processing method for a hybrid welding apparatus, wherein the image processing method is a rectangular region having no shape.   18. The image processing method for a hybrid welding apparatus according to claim 17, wherein the laser welding and the arc welding are controlled by center scanning control of a laser welding line in the laser welding, a laser irradiation point of the laser welding, and the arc welding. An image processing method for a hybrid welding apparatus, comprising: control of a distance to a welding wire tip, arc welding line center scanning control of arc welding, or swing width control of the welding wire. The image processing method for a hybrid welding apparatus according to claim 14, further comprising:
An image processing method for a hybrid welding apparatus, wherein an image of the material to be welded captured by a CCD camera provided at a predetermined angle with respect to the surface of the material to be welded is captured.   20. The image processing method for a hybrid welding apparatus according to claim 19, wherein the image photographed by the CCD camera is used for measurement of a laser processing point position by the laser welding. . An image processing program for a hybrid welding apparatus that combines first welding and second welding,
On the computer,
Capture multiple images of the material to be welded,
An image processing program for a hybrid welding apparatus, wherein an image used for controlling the first welding and the second welding is generated by performing an integration smoothing process on the plurality of captured images. A medium recording a program to be executed by a computer, the program being stored in the computer
Capture multiple images of the material to be welded,
A medium on which an image processing program of a hybrid welding apparatus is recorded, wherein an image used for controlling the first welding and the second welding is generated by subjecting the plurality of captured images to an integration smoothing process. .

Images