Welding condition determination device and welding condition determination method

Abstract

To provide a welded state determination device and a welded state determination method which are advantageous for simplifying a device determining welded state.SOLUTION: A welded state determination device 1 comprises: a photographing device 10 that has a wide-angle lens 10a and photographs, through the wide-angle lens 10a, a rear surface of a weld zone 92 including the whole of a weld line WL set on a rear side of the weld zone 92, when a first member 90 is welded to a second member 91; and a control calculation portion 11 that determines whether a state of the weld zone 92 is in a good state or not, on the basis of a photographed image photographed by the photographing device 10. The photographing device 10 is arranged in such a posture that a direction in which the rear side of the weld zone 92 is photographed inclines with respect to an extension direction of the weld line WL.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a welding condition determination device and a welding condition determination method. [Background technology]

[0002] For example, a welding defect that can occur in full-penetration welding using laser welding or laser-arc hybrid welding is a back-side swell. Conventionally, there has been a technique for directly observing the back side of a weld to more accurately grasp the state of the back-side swell formed on the back side of a weld as an indicator for suppressing the occurrence of a back-side swell. Patent Document 1 discloses a technique for a method for monitoring whether the back-side swell is formed properly by measuring plasma light generated on the side opposite to the front side of the weld where the laser is irradiated, i.e., on the back side, using a sensor from the back side of the weld. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 10-76383 Summary of the Invention [Problem to be solved by the invention]

[0004] In the method disclosed in Patent Document 1, in order for the sensor to constantly observe the back side of the weld where the reverse side is formed, a mechanism is required to move either the observation device or the holding mechanism that holds the welding object. Furthermore, since full penetration welding using laser welding or the like generates spatter or fumes, the sensor needs a mechanism to deal with spatter, etc.

[0005] Therefore, an object of the present disclosure is to provide a welding condition determination device and a welding condition determination method that are advantageous in simplifying the device that determines the welding condition. [Means for solving the problem]

[0006] A welding condition determination device according to one aspect of the present disclosure includes an imaging device having a wide-angle lens that, when two components or parts are welded, captures an image of the back surface of the weld, including the entire weld line set on the back side of the weld, through the wide-angle lens; and a control and calculation unit that determines whether the condition of the weld is good or bad based on the image captured by the imaging device, wherein the imaging device is positioned in a position such that the direction in which the image of the back surface of the weld is captured is inclined relative to the extension direction of the weld line.

[0007] In addition, a welding condition determination device according to another aspect of the present disclosure includes an imaging device having a wide-angle lens that photographs the back side of the weld through the wide-angle lens when two components or parts are welded, and a control and calculation unit that determines the quality of the weld based on the image captured by the imaging device, and the imaging device is positioned so that the entire weld line set on the back side of the weld is within the angle of view.

[0008] In each of the above-described welding condition determination devices, the control and calculation unit may determine the quality of the backside ripple formed along the weld line based on the captured image. Here, the control and calculation unit may measure the amount of spatter generated from the captured image and determine the quality of the backside ripple based on whether the amount of spatter generated is within a predetermined allowable amount. The control and calculation unit may measure the amount of deviation of a high-brightness area representing the area where the backside ripple is generated from the captured image and determine the quality of the backside ripple based on whether the deviation is within a predetermined allowable range. The control and calculation unit may calculate the amount of variation in the brightness of the backside of the weld from the captured image and determine the quality of the backside ripple based on whether the variation is within a predetermined allowable range. The control and calculation unit may acquire waveform data by applying artificial intelligence to data accumulated in advance from a construction test that combines the acquisition of observed brightness waveforms and the determination of the quality of the backside ripple, and calculate the amount of variation by referring to the waveform data. The two components or parts are joined together by partial penetration welding, and the control and calculation unit calculates the brightness distribution on the back side of the weld based on the captured image, and may determine whether the partial penetration welding is successful or not based on whether there is an area in the brightness distribution that exceeds a predetermined threshold.

[0009] Furthermore, a welding condition determination method according to one aspect of the present disclosure includes an imaging process of photographing the back surface of the weld, including the entire weld line set on the back surface of the weld when two members or parts are welded, using a wide-angle lens from a direction inclined with respect to the extension direction of the weld line, and a process of determining whether the condition of the weld is good or bad based on the image captured in the imaging process. [Effects of the Invention]

[0010] According to the present disclosure, it is possible to provide a welding condition determination device and a welding condition determination method that are advantageous in simplifying a device that determines a welding condition. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a schematic configuration of a welding condition determination device according to a first embodiment. [Figure 2] FIG. 2 is a perspective view showing a state in which the photographing device photographs a back wave in the first embodiment. [Figure 3A] FIG. 10 is a side view of the welding object and the imaging device as viewed along the X direction. [Figure 3B] FIG. 2 is a plan view of the welding object and the imaging device as viewed along the Z direction. [Figure 4] 4 is a flowchart showing a welding condition determination process in the first embodiment. [Figure 5A] FIG. 10 is a diagram showing a first photographed image taken when the state of the back waves is good. [Figure 5B] FIG. 10 is a diagram showing a second photographed image taken when the state of the back waves is poor. [Figure 6A] FIG. 10 is a diagram showing a first processed image obtained by binarizing the first captured image. [Figure 6B] FIG. 10 is a diagram showing a second processed image obtained by binarizing the second captured image. [Figure 7] 10 is a graph showing the amount of spatter generated per image versus the imaging time. [Figure 8] 10 is a flowchart showing a welding condition determination process in a second embodiment. [Figure 9A]FIG. 10 is a diagram showing a third photographed image taken when the state of the back waves was good. [Figure 9B] FIG. 10 shows a third processed image to which ellipse fitting is applied. [Figure 10A] FIG. 10 is a diagram showing a fourth photographed image taken when the state of the back waves is poor. [Figure 10B] FIG. 10 shows a fourth processed image to which ellipse fitting is applied. [Figure 11] 10 is a graph showing the deviation between the reference line and the center of the high brightness area with respect to the shooting time. [Figure 12] This is a diagram comparing an actual photo of a back wave with the results of a judgment of whether the back wave condition is good or bad. [Figure 13] FIG. 10 is a perspective view showing a state in which the photographing device photographs a back swell occurring at a butt joint. [Figure 14] 10 is a flowchart showing a welding condition determination process in a third embodiment. [Figure 15] 10 is a flowchart showing a welding condition determination process in a fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, several exemplary embodiments will be described with reference to the drawings. Hereinafter, the dimensions, materials, and other specific numerical values ​​shown in each embodiment are merely examples and do not limit the present disclosure unless otherwise specified. Furthermore, elements having substantially the same functions and configurations are assigned the same reference numerals to avoid redundant explanation, and elements not directly related to the present disclosure are not shown.

[0013] (First embodiment) Fig. 1 is a block diagram showing a schematic configuration of a welding condition determination device 1 according to a first embodiment. Fig. 2 is a perspective view showing a state in which an image capturing device 10 included in the welding condition determination device 1 captures an image of a backside bead 93 during welding. The welding condition determination device 1 determines whether the state of the backside bead 93 is good or bad based on an image obtained by capturing an image of the backside bead 93 while a welding machine 80 is welding two members or parts together.

[0014] The welding that generates the back seam 93 that is the subject of the quality judgment by the welding condition judgment device 1 is, for example, full-penetration welding employing laser welding or laser-arc hybrid welding. In this embodiment, the welding condition judgment device 1 judges the quality of the back seam 93 of a weld 92 performed by, for example, a welding machine 80 employing laser welding. The welding machine 80 includes a laser torch 81. The laser torch 81 introduces a laser emitted from a laser oscillator (not shown) and irradiates the welding position with laser light Ls via a focusing optical system. The welding machine 80 performs butt welding by moving the irradiation position of the laser light Ls along the welding position.

[0015] Furthermore, the objects to be welded in this embodiment are a first member 90 and a second member 91, each of which is a metal plate. As an example, the welding machine 80 performs full-penetration welding of the end face of the second member 91 in a state where they are perpendicular to the main plane of the first member 90, to produce a T-joint. At this time, a laser beam Ls is irradiated onto the joint between the first member 90 and the second member 91, forming a weld 92. At the same time, a back rib 93 is generated on the back side of the weld 92, which is on the opposite side of the second member 91 from the side irradiated with the laser beam Ls.

[0016] Hereinafter, directions in the XYZ coordinate system are set as shown in FIG. 2, assuming that a welding machine 80 welds a first member 90 and a second member 91 to create a T-joint. The XY plane is defined as a plane parallel to the main plane of the first member 90 to which the end face of the second member 91 is welded. The Y direction is along the extension direction of the second member 91. In other words, the welding direction in this embodiment is along the Y direction. The X direction is perpendicular to the Y direction. Furthermore, the Z direction is perpendicular to the XY plane. In other words, the joining direction of the second member 91 to the main plane of the first member 90 is along the Z direction.

[0017] 2, a weld line WL is set as a reference for the position where a back rib 93 occurs during welding. At the joint where the main plane of the first member 90 and the end face of the second member 91 face each other, two imaginary intersection lines between the first member 90 and the second member 91 are generated, each extending along the Y direction, which is the welding direction. The weld line WL is one imaginary intersection line on the side irradiated with the laser light Ls, and the other imaginary intersection line that occurs on the opposite side, i.e., the back side, across the second member 91. In other words, the back rib 93 occurs approximately on the weld line WL in line with the formation of the weld 92.

[0018] The welding condition determination device 1 includes an imaging device 10, a control and calculation unit 11, a storage unit 12, an input unit 13, and an output unit 14.

[0019] The photographing device 10 photographs at least a portion of the welding object that includes the entire weld seam WL. The photographing device 10 is, for example, a CMOS camera. A wide-angle lens 10a is used for photographing with the photographing device 10. Hereinafter, the image photographed by the photographing device 10 will be simply referred to as a "photographed image."

[0020] 3A and 3B are diagrams illustrating the installation position of the imaging device 10 relative to the welding object and the imaging range P of the imaging device 10. FIG. 3A is a side view of the welding object and the imaging device 10 viewed along the X direction so as to visually confirm the side on which a back rib 93 occurs on the weld line WL. FIG. 3B is a plan view of the welding object and the imaging device 10 viewed along the Z direction so as to visually confirm the second member 91. In FIGS. 3A and 3B, the imaging range P is represented as the area sandwiched between two dashed dotted lines. In the example shown in FIGS. 3A and 3B, the weld line WL is set over the entire length L in the Y direction of each of the first member 90 and the second member 91, and a back rib 93 is shown to occur along the entire weld line WL.

[0021] The camera 10 is positioned so that the direction in which it captures the weld line WL, the back ribs 93, is inclined relative to the extension direction of the weld line WL set on the welding object. In the Y direction, which is the extension direction of the weld line WL, the camera 10 is positioned outside the range in which the weld line WL is set. For example, as shown in FIGS. 3A and 3B , if the weld line WL is set to a length L equal to the Y-direction dimensions of the first member 90 and the second member 91, at least a portion of the camera 10 is not located within the Y-direction dimension of the welding object defined by the length L. The camera 10 is positioned so that the entire portion where the back ribs 93 are generated, i.e., the entire weld line WL, is simultaneously captured within the angle of view. In addition to being positioned based on at least one of the two placement conditions, the camera 10 can include the entire weld line WL within the shooting range P by using a wide-angle lens 10a. In other words, the position and orientation of the camera 10 relative to the welding object do not change during welding.

[0022] Control and calculation unit 11 is, for example, a CPU (Central Processing Unit), and executes a program related to a welding condition determination step as a welding condition determination method according to this embodiment. Control and calculation unit 11 is electrically connected to imaging device 10, memory unit 12, input unit 13, and output unit 14. In addition, when the determination result obtained by executing the welding condition determination step is to be reflected in real time in welding machine 80 during welding, control and calculation unit 11 may be electrically connected to welding machine 80.

[0023] The storage unit 12 is, for example, a semiconductor memory or a magnetic storage device such as an HDD (hard disk drive), and stores (preserves) programs, databases, etc. The control and calculation unit 11 can read out programs or data stored in the storage unit 12 as needed. The control and calculation unit 11 can also write new data to the storage unit 12 as needed. For example, the storage unit 12 stores captured images.

[0024] Furthermore, memory unit 12 has a program storage unit 15. Program storage unit 15 stores at least a control program related to a series of controls in the welding condition determination process. Furthermore, when the welding condition determination process according to a third embodiment described below is employed, program storage unit 15 may store a machine learning program related to artificial intelligence (AI). In this case, memory unit 12 may have a database storage unit 16 that stores a database to be referenced when the machine learning program is executed.

[0025] The input unit 13 is, for example, an operation input device such as a keyboard or a touch panel, a receiving device that receives information from an external device, or an input port that reads information from a portable memory device such as a memory card. An operator can directly input various information from the input unit 13 as an operation input device. Alternatively, the control and calculation unit 11 can receive various information from an external server or the like via a communication network such as a LAN or the Internet from the input unit 13 as a receiving device. Furthermore, the control and calculation unit 11 can transfer various information from a memory device via the input unit 13 as an input port. Information that can be input by the input unit 13 is, for example, the allowable amount of spatter generation that is referenced in the step of determining whether the amount of spatter generation is an allowable amount (S103; see FIG. 4). Details of the amount of spatter generation, etc. will be described later in conjunction with the explanation of the welding condition determination step.

[0026] The output unit 14 is a display unit such as a monitor, a transmitting device that transmits information to the outside, or an output port that writes information to a memory device. The control and calculation unit 11 can display the judgment result obtained in the welding condition judgment process on the output unit 14 as a display unit, so that an operator or the like can directly recognize the judgment result. Alternatively, the control and calculation unit 11 can transmit the judgment result from the output unit 14 as a transmitting device to an external server or the like via a communication network. Furthermore, the control and calculation unit 11 can transfer the judgment result from the output unit 14 as an output port to a memory device.

[0027] Next, the operation of the welding condition determination device 1 and the welding condition determination method according to this embodiment will be described.

[0028] Generally, when poor back beading occurs during full penetration welding using laser welding or the like, the entire plate thickness is not melted at the joint, and the desired joint strength may not be obtained. Therefore, the welding condition judgment device 1 determines whether the state of the back beading 93 during welding is good or bad, as an index for suppressing the occurrence of poor back beading, by performing a welding condition judgment described in detail below.

[0029] 4 is a flowchart showing a welding condition determination process as a welding condition determination method according to the first embodiment. The welding condition determination process includes a series of steps according to the flowchart shown in FIG. 4 and is executed by control calculation unit 11 as a control program of welding condition determination device 1.

[0030] The control and calculation unit 11 starts executing the welding condition determination process in conjunction with the start of welding by the welding machine 80. First, the control and calculation unit 11 executes the back wave photographing process S100. In the back wave photographing process S100, the control and calculation unit 11 causes the photographing device 10 to photograph the welding object. Here, the photographing range P of the photographing device 10 includes at least the entire weld line WL. Therefore, even if the position or posture of the photographing device 10 is not changed in accordance with the progress of the welding process, the photographed image will capture the back wave 93 that is occurring at the time of photographing. Then, the control and calculation unit 11 stores the photographed image in the memory unit 12.

[0031] Next, the control and calculation unit 11 executes an image processing step S101. In the image processing step S101, the control and calculation unit 11 first acquires a captured image from the storage unit 12.

[0032] 5A and 5B are diagrams showing examples of captured images processed in the image processing step S101. FIG. 5A is a diagram showing a first captured image IM1 captured when the condition of the backside wave 93 is good. FIG. 5B is a diagram showing a second captured image IM2 captured when the condition of the backside wave 93 is poor. The captured images may be grayscale images as shown in FIGS. 5A and 5B, or may be color images. The captured images shown in FIGS. 5A and 5B include portions of the first member 90 and the second member 91, and the areas located on the back side of the welded portion 92 that was welded at the time of capture, i.e., the areas where the backside wave 93 occurs, are shown with high brightness.

[0033] Then, the control and calculation unit 11 extracts high-luminance areas from the acquired photographed image, for example, by binarizing an image of a specific color component using a preset brightness threshold value. Hereinafter, the image binarized based on the photographed image will be simply referred to as the "processed image."

[0034] 6A and 6B are diagrams showing examples of processed images. FIG. 6A is a diagram showing a first processed image IM3 obtained by binarizing the first captured image IM1 shown in FIG. 5A. FIG. 6B is a diagram showing a second processed image IM4 obtained by binarizing the second captured image IM2 shown in FIG. 5B. First, referring to the first processed image IM3 shown in FIG. 6A, the area where the reverse side wave 93 occurs is extracted as a first high-luminance area 94, and also a second high-luminance area 95 different from the first high-luminance area 94 is extracted. On the other hand, referring to the second processed image IM4 shown in FIG. 6B, the area where the reverse side wave 93 occurs is extracted as a first high-luminance area 94, but the second high-luminance area 95 is not extracted.

[0035] Next, the control and calculation unit 11 executes a spatter generation amount measurement step S102. In the spatter generation amount measurement step S102, the control and calculation unit 11 determines that the amount of second high-brightness regions 95 extracted from the processed image in the image processing step S101 corresponds to the amount of spatter generated during welding, and measures the amount of spatter generated [pieces].

[0036] Next, the control and calculation unit 11 determines whether the amount of spatter generated determined in the spatter generation amount measurement step S102 is an allowable amount (S103). The allowable amount of spatter generated is determined in advance, for example, for each plate thickness of the welding target or welding conditions, taking into account the relationship when the state of the back-side bead 93 is good. Here, if the control and calculation unit 11 determines that the amount of spatter generated is an allowable amount (YES), it determines that the state of the back-side bead 93 is "good" (S104). On the other hand, if the control and calculation unit 11 determines that the amount of spatter generated is not an allowable amount (NO), it determines that the state of the back-side bead 93 is "poor" (S105).

[0037] In particular, if the state of the back wave 93 is determined to be "poor" (S105), the control and calculation unit 11 may subsequently determine whether the amount of spatter generated measured in the spatter generation amount measurement step S102 is greater than the allowable amount (S106). Here, if the control and calculation unit 11 determines that the amount of spatter generated is greater than the allowable amount (YES), it may determine that the penetration amount at the time of photographing is excessive (S107). In particular, if the welded portion 92 has excessive penetration, it may also be determined that burn-through has occurred. On the other hand, if the control and calculation unit 11 determines that the amount of spatter generated is not greater than the allowable amount, i.e., is less than the allowable amount (NO), it may determine that the penetration amount at the time of photographing is insufficient (S108). In particular, if the welded portion 92 has insufficient penetration, it may also be determined that the welding output has decreased.

[0038] After each of the determinations in S104, S107, and S108, the control and calculation unit 11 next determines whether it is necessary to subsequently determine the quality of the state of the backside wave 93 based on another captured image (S109). The other captured image refers to an image obtained by capturing an image captured a certain time after the image that was used as the basis for determining the quality of the state of the backside wave 93 up to this point in the backside wave capturing step S100 during a series of welding operations. Here, if the control and calculation unit 11 determines that it is necessary to determine the quality of the state of the backside wave 93 based on another captured image (YES), the process returns to the image processing step S101. On the other hand, if the control and calculation unit 11 determines that it is not necessary to determine the quality of the state of the backside wave 93 based on another captured image (NO), the process ends the welding condition determination step. An example of a case where it is not necessary to determine the quality of the state of the backside wave 93 based on another captured image is when welding has been completed over the entire weld line WL.

[0039] Next, the effects of the welding condition determination device 1 and the welding condition determination method will be described.

[0040] The welding condition determination device 1 and the welding condition determination method according to this embodiment can be applied when two members or parts are welded together. In the above example of this embodiment, two members, a first member 90 and a second member 91, are to be welded together.

[0041] In this case, welding condition determination device 1 is equipped with photographing device 10 having wide-angle lens 10a that photographs the back surface of welded portion 92, including the entire weld line WL set on the back side of welded portion 92, via wide-angle lens 10a. Welding condition determination device 1 also includes control and calculation unit 11 that determines the quality of the condition of welded portion 92 based on the image photographed by photographing device 10. Photography device 10 is positioned in a position such that the direction in which the photograph is taken of the back surface of welded portion 92 is inclined relative to the extension direction of weld line WL.

[0042] Welding condition determination device 1 also includes photographing device 10 having wide-angle lens 10a that photographs the back surface of welded portion 92 through wide-angle lens 10a, and control and calculation unit 11 that determines the quality of the condition of welded portion 92 based on the image photographed by photographing device 10. Photography device 10 is placed in a position where the entire weld line WL set on the back surface of welded portion 92 fits within the angle of view.

[0043] The welding condition determination method further includes a photographing step of photographing the rear surface of welded portion 92, including the entire weld line WL set on the rear surface of welded portion 92, via wide-angle lens 10a from a direction inclined with respect to the extension direction of weld line WL. The welding condition determination method also includes a step of determining whether the condition of welded portion 92 is good or bad based on the image photographed in the photographing step.

[0044] Here, the image captured by the photographing device 10 corresponds to the first photographed image IM1 or the second photographed image IM2 in the above example of this embodiment. In the above example of this embodiment, the photographing step corresponds to the back wave photographing step S100 shown in the flowchart of Fig. 4. Furthermore, in the above example of this embodiment, the step of determining whether the state of the welded portion 92 is good or bad corresponds to S103 to S105 shown in the flowchart of Fig. 4.

[0045] First, when determining whether the condition of the weld 92 is good or bad, a photographed image obtained by photographing the back surface of the weld 92 is referenced. Therefore, for example, when two members or parts are joined by full-penetration welding such as laser welding, a photographed image of the back surface 93 generated during welding is referenced. Therefore, according to this embodiment, the condition of the back surface 93 can be grasped more accurately than when the back surface 93 is indirectly observed.

[0046] The captured image referred to here was captured through a wide-angle lens 10a by the camera 10 positioned such that the direction in which the camera captures the back surface of the weld 92 is inclined relative to the extension direction of the weld seam WL. Therefore, as illustrated in FIGS. 3A and 3B , the camera 10 is positioned farther away from the back surface of the weld 92, where the back surface ripple 93 occurs, and is therefore less susceptible to spatter or fumes generated during full-penetration welding, such as laser welding. This embodiment is therefore advantageous in simplifying countermeasures against spatter or fumes in the camera 10. Furthermore, as illustrated in FIGS. 3A and 3B , the camera 10 can be positioned so that the entire area in which the back surface ripple 93 may occur is included in the image capture range P, thereby enabling the capture of images including the entire back surface ripple 93 at all times. Therefore, this embodiment eliminates the need to move either the camera 10 or the components to be welded as the welding operation progresses, thereby simplifying the configuration of the welding condition determination device 1 or the holding mechanism for the welded object, etc.

[0047] Furthermore, the camera 10 may be specified to be positioned so that the entire weld line WL set on the back side of the welded portion 92 fits within the angle of view. This specification also simplifies the configuration of the welding condition determination device 1 or the welding machine 80, since the entire area where back seams 93 may occur is included in the imaging range P. Furthermore, by capturing images through the wide-angle lens 10a, the camera 10 can be positioned in a position where the direction in which the image is captured of the back side of the welded portion 92 is inclined with respect to the extension direction of the weld line WL. Therefore, this specification also allows the camera 10 to be positioned further away from the back side of the welded portion 92, which is advantageous for simplifying measures against spatter or fumes in the camera 10.

[0048] As described above, according to the present embodiment, it is possible to provide a welding condition determination device 1 and a welding condition determination method that are advantageous in simplifying the device that determines the welding condition.

[0049] In the welding condition determination device 1 and the like, the control and calculation unit 11 may determine whether the condition of the back seam 93 formed along the weld line WL is good or bad based on the captured image.

[0050] The quality of the state of the back bead 93 is directly related to the quality of full penetration welding by laser welding etc. Therefore, according to the welding condition determination device 1 etc., the quality of the state of the back bead 93 is determined and as a result the quality of the welding condition is determined, which can be advantageous in terms of, for example, simplifying the determination of the quality of the welding condition or improving the accuracy of the determination of the quality of the welding condition.

[0051] Furthermore, in the welding condition determination device 1, etc., the control calculation unit 11 may measure the amount of spatter generated from the captured image and determine whether the condition of the back bead 93 is good or bad based on whether the amount of spatter generated is within a predetermined allowable amount.

[0052] The amount of spatter generated during full penetration welding such as laser welding is related to the amount of penetration, i.e., the state of the weld 92. Therefore, with this welding state determination device 1, it is possible to easily determine whether the state of the back bead 93 is good or bad from the amount of spatter generated. Also, as illustrated above, in this case, by determining whether the amount of spatter generated is more or less than the allowable amount (S106), it is possible to easily determine whether the penetration is excessive or insufficient.

[0053] Here, if the state of backside bead 93 is determined to be poor in an image captured at a certain timing during welding while the welding condition is being determined, control calculation unit 11 can reflect the determination result in welding machine 80 in real time. For example, control calculation unit 11 repeatedly determines whether the state of backside bead 93 is good or bad over time during a series of welding operations. In this case, when control calculation unit 11 determines that the state of backside bead 93 is poor (S105), it may instruct welding machine 80 to stop welding. Furthermore, when control calculation unit 11 determines that the penetration amount is excessive or insufficient (S107 or S108), it may execute feedback control to cause welding machine 80 to change welding conditions, etc., depending on the penetration amount.

[0054] On the other hand, in this embodiment, the control calculation unit 11 does not reflect the judgment result regarding the quality of the condition of the back wave 93 to the welding machine 80 in real time, but rather may output the quality of the condition of the entire welded section 92 as a final result from the output unit 14 after the welding operation is completed.

[0055] FIG. 7 is a graph showing the amount of spatter generated per image (number of spatters) versus the time t at which the back wave 93 was photographed in the back wave photographing step S100. In FIG. 7, the allowable range AR, defined by the allowable amount of spatter generated, is shown as the area surrounded by a two-dot chain line. As shown in FIG. 7, if the amount of spatter generated measured based on the images sequentially photographed during the welding process falls within the allowable range AR, it is considered that the condition of the back wave 93 is good throughout the entire weld 92. Then, by recognizing the final result shown in FIG. 7 from the output unit 14, an operator can determine that the condition of the entire weld 92 is good and that the quality of the T-joint produced by this welding process is also good.

[0056] In contrast to this, if the amount of spatter generated falls outside the allowable range AR, as opposed to the final result shown in Figure 7, it is considered that a portion of the weld 92 has a poor state of back wave 93. In this case, an operator or the like can recognize from the output unit 14 the final result that there is a portion where the state of back wave 93 is poor, and can determine that the quality of the T-joint produced by this welding operation is poor. Furthermore, according to the determination in S106 above, depending on whether the amount of spatter generated is more or less than the allowable amount, the operator or the like can recognize from the output unit 14 the final result that there is excessive or insufficient penetration in the weld 92 of the T-joint produced by this welding operation.

[0057] In the above description, the range of the weld 92, i.e., the range of the weld line WL located on the back side of the weld 92, is set to the entire length L in the Y direction of the first member 90 and the second member 91. In this case, as shown in FIGS. 3A and 3B , part or all of the imaging device 10 does not face either the first member 90 or the second member 91 in the X direction or the Z direction. In contrast, the range of the weld line WL may be set to part of the length L in the Y direction of the first member 90 or the second member 91. In this case, part or all of the imaging device 10 may be disposed so as to face the first member 90 or the second member 91 in the X direction or the Z direction.

[0058] (Second embodiment) In the first embodiment, the control calculation unit 11 determines whether the amount of spatter generated, which is identified based on the photographed image, is an allowable amount, thereby determining whether the state of the back wave 93 is good or bad. In contrast, in the second embodiment, the control calculation unit 11 determines whether the deviation of the high brightness area, which is identified based on the photographed image, is within an allowable range, thereby determining whether the state of the back wave 93 is good or bad.

[0059] Fig. 8 is a flowchart showing a welding condition determination step as a welding condition determination method according to the second embodiment. The welding condition determination step is executed by control calculation unit 11 as a control program for welding condition determination device 1, which includes a series of steps according to the flowchart shown in Fig. 8. Note that the basic configuration of welding condition determination device 1 according to this embodiment and the welding object are the same as those in the first embodiment, and therefore, the same reference numerals will be used for the respective components and the welding object, and detailed description thereof will be omitted.

[0060] The control and calculation unit 11 starts executing the welding condition determination process in conjunction with the start of welding by the welding machine 80. First, the control and calculation unit 11 executes a reverse side wave photographing process S200. The reverse side wave photographing process S200 is the same as the reverse side wave photographing process S100 in FIG. 4 in the first embodiment.

[0061] Next, the control and calculation unit 11 executes an image processing step S201. In the image processing step S201, the control and calculation unit 11 first acquires a captured image from the storage unit 12, as in the first embodiment, and extracts a first high-luminance region 94 as a region where a reverse ripple 93 occurs from a processed image obtained by binarizing the captured image. Then, the control and calculation unit 11 performs ellipse fitting on the extracted first high-luminance region 94, and identifies a center 97 of the obtained ellipse 96 (see FIGS. 9A to 10B). The least squares method may be used for the ellipse fitting here.

[0062] 9A and 9B are diagrams showing examples of an ellipse 96 and a center 97 of the ellipse 96 obtained based on an image taken when the condition of the back swell 93 was good. Fig. 9A is a diagram showing a third captured image IM5 to which the ellipse 96, center 97, and reference line RL have been added. Fig. 9B is a diagram showing a third processed image IM6 used to obtain the ellipse 96 and center 97 of the ellipse 96 applied to the first high-brightness region 94.

[0063] 10A and 10B are diagrams showing examples of an ellipse 96 and a center 97 of the ellipse 96 obtained based on an image taken when the condition of the back wave 93 was poor. Fig. 10A is a diagram showing a fourth captured image IM7 to which the ellipse 96, the center 97, and a reference line RL have been added. Fig. 10B is a diagram showing a fourth processed image IM8 used to obtain the ellipse 96 and the center 97 of the ellipse 96 applied to the first high-brightness region 94.

[0064] Next, the control and calculation unit 11 executes a high-brightness area displacement measurement step S202. In the high-brightness area displacement measurement step S202, the control and calculation unit 11 first recognizes the center 97 of the ellipse 96 extracted from the processed image in the image processing step S201 as corresponding to the center of the high-brightness area, which is the area where the reverse ripple 93 occurs during welding. The control and calculation unit 11 also determines a reference line RL that serves as a reference for measuring the displacement of the high-brightness area. In this embodiment, the weld line WL described in the first embodiment is used as the reference line RL. Then, the control and calculation unit 11 measures the displacement of the center of the high-brightness area from the reference line RL.

[0065] Next, the control and calculation unit 11 determines whether the deviation of the high-brightness region identified in the high-brightness region deviation measurement step S202 is within an allowable range (S203). The allowable range of the high-brightness region deviation is predetermined, for example, based on the relationship between the plate thickness of the welding object and the welding conditions and the state of the backside bead 93 when it is good. If the control and calculation unit 11 determines that the deviation of the high-brightness region is within the allowable range (YES), it determines that the state of the backside bead 93 is "good" (S204). As can be seen from the third captured image IM5 in FIG. 9A, when the state of the backside bead 93 is good, the center 97 of the ellipse 96 is located close to the reference line RL. On the other hand, if the control and calculation unit 11 determines that the deviation of the high-brightness region is not within the allowable range (NO), it determines that the state of the backside bead 93 is "poor" (S205). As can be seen by referring to the fourth captured image IM7 in Figure 10A, when the condition of the back swell 93 is poor, the center 97 of the ellipse 96 is located farther from the reference line RL than when the condition of the back swell 93 is good.

[0066] After the determinations in S204 and S205, the control and calculation unit 11 next determines whether it is necessary to subsequently determine the quality of the state of the backside seam 93 based on another captured image (S206), similar to S109 in FIG. 4 in the first embodiment. Here, if the control and calculation unit 11 determines that it is necessary to subsequently determine the quality of the state of the backside seam 93 based on another captured image (YES), the process returns to the image processing step S201. On the other hand, if the control and calculation unit 11 determines that it is not necessary to subsequently determine the quality of the state of the backside seam 93 based on another captured image (NO), the process ends.

[0067] In this way, the control calculation unit 11 may measure the amount of deviation of the high-brightness area in the captured image where the back wave 93 is occurring, and determine whether the condition of the back wave 93 is good or bad based on whether the amount of deviation is within a predetermined tolerance range.

[0068] According to the welding condition determination device 1, it is possible to easily determine whether the condition of the back seam 93 is good or bad from the amount of deviation of the high brightness area. In this case, if the amount of deviation of the high brightness area is not within the allowable range, it means that the welding target position is not appropriate, that is, a target position deviation has occurred, and therefore, the occurrence of a target position deviation can be easily detected.

[0069] As in the first embodiment, the control and calculation unit 11 may cause the output unit 14 to output the final result indicating whether the overall state of the welded portion 92 is good or bad after the welding operation is completed.

[0070] FIG. 11 is a graph showing the deviation between the reference line RL and the center of the high-brightness area versus time t at which the back wave 93 was photographed in the back wave photographing step S200. In FIG. 11, the allowable range AR of the high-brightness area deviation is shown as the area surrounded by a two-dot chain line. As shown in FIG. 11, if the deviations of the high-brightness areas measured based on the images photographed sequentially during welding are all within the allowable range AR and are distributed symmetrically with respect to the reference line RL, it is considered that the condition of the back wave 93 is good throughout the entire weld 92. In contrast, if the deviations of the high-brightness areas fall outside the allowable range AR or the distribution of the deviations of the high-brightness areas is biased toward either the positive or negative side of the reference line RL, it is considered that misalignment has occurred in the weld 92, and that there is a portion where the condition of the back wave 93 is poor.

[0071] In the above description of this embodiment, the quality of the state of the back wave 93 is determined from the deviation of a high-brightness area identified based on a captured image. However, instead of an area expressed by brightness, for example, an area expressed by the intensity of each RGB color may be identified, and the quality of the state of the back wave 93 may be determined from the deviation of that RGB area.

[0072] 12 is a diagram showing an example of a comparison between a photograph of a back rib 93 that occurs when a first member 90 and a second member 91 are actually joined by full penetration welding and the results of a judgment of the quality of the back rib 93 in the first and second embodiments. The back rib 93 occurs in a straight line along the boundary between the first member 90 and the second member 91. In addition, in FIG. 12, the portion of the back rib 93 that is visually judged to be in a good condition based on the actual condition of the back rib is marked with "good back rib," while the portion that is judged to be in a poor condition is marked with "poor back rib."

[0073] Here, the first section IV1 corresponds to the back side of the section irradiated with the laser light Ls by the welding machine 80, i.e., the section where the backside ribs 93 occur on the weld line WL. In contrast, the second section IV2 corresponds to the section where the state of the backside ribs 93 is determined to be good based on the amount of spatter generated in the first embodiment. Meanwhile, the third section IV3 corresponds to the section where the state of the backside ribs 93 is determined to be good based on the high-brightness area deviation or the RGB area deviation in the second embodiment.

[0074] As shown in FIG. 12, in the example, it has been confirmed that both the second section IV2 and the third section IV3 are actually included in the areas where the state of the back waves is good.

[0075] Furthermore, in the first and second embodiments, the example has been given of a case where the welding application object is a T-joint in which the end face of the second member 91 is welded perpendicular to the main plane of the first member 90. However, the welding application object is not limited to welded joints such as T-joints, and may be, for example, a butt joint in which the end faces of two metal plates are butted together in parallel to each other and welded through penetration.

[0076] FIG. 13 is a perspective view showing a state in which the camera 10 is capturing an image of a back rib 103 during welding when the welding target is a butt joint. In this case, the welding target is a third member 100 and a fourth member 101, each of which is a metal plate. The welding machine 80 performs full penetration welding between the end faces of the third member 100 and the fourth member 101 while they are butted together in the X direction to create a butt joint. At this time, a laser beam Ls is irradiated onto the joint between the third member 100 and the fourth member 101, forming a weld 102. At the same time, a back rib 103 is generated along the weld line WL on the back side of the weld 102, opposite the side irradiated with the laser beam Ls. Even when the welding target is a butt joint, the camera 10 is positioned according to the positioning conditions described in the first embodiment.

[0077] The welding object may be a welded joint other than a T-joint or a butt joint. Furthermore, the specific shape of the weld line WL set at the joint of the welding object may be curved rather than linear.

[0078] (Third embodiment) In the first and second embodiments, the control and calculation unit 11 performs image processing such as binarization on the captured image, and then determines whether the amount of spatter generated is within the allowable range, thereby determining whether the state of the back wave 93 is good or bad. In contrast, in the third embodiment, the control and calculation unit 11 directly obtains the brightness from the captured image, and determines whether the state of the back wave 93 is good or bad from the amount of variation in the brightness.

[0079] Fig. 14 is a flowchart showing a welding condition determination step as a welding condition determination method according to the third embodiment. The welding condition determination step is executed by control calculation unit 11 as a control program for welding condition determination device 1, which includes a series of steps according to the flowchart shown in Fig. 14. Note that the basic configuration of welding condition determination device 1 according to this embodiment and the welding object are the same as those in the first embodiment, and therefore, the same reference numerals will be used for the respective components and the welding object, and detailed description thereof will be omitted.

[0080] Control and calculation unit 11 starts executing the welding condition determination process in conjunction with the start of welding by welding machine 80. First, control and calculation unit 11 executes back surface luminance observation process S300. In back surface luminance observation process S300, control and calculation unit 11 causes photographing device 10 to photograph the welding object. At this time, as in the first embodiment, the photographing range P of photographing device 10 includes the back surface of welded portion 92, which includes at least the entire weld line WL. Then, control and calculation unit 11 observes the luminance (brightness) of the back surface where back waves 93 occur from the photographed image at predetermined intervals over a predetermined time, known as an averaging time.

[0081] Next, the control and calculation unit 11 executes a luminance fluctuation amount calculation step S301. In the luminance fluctuation amount calculation step S301, the control and calculation unit 11 calculates the amount of fluctuation in the back surface luminance using, as an index, the average value, standard deviation, or disturbance intensity expressed as (standard deviation / average value) of the observed values ​​of the back surface luminance observed in the back surface luminance observation step S300.

[0082] Next, the control and calculation unit 11 determines whether the amount of variation in the back surface brightness determined in the brightness variation calculation step S301 is within an allowable range (S302). The allowable range of the amount of variation in the back surface brightness is determined in advance, for example, for each plate thickness of the welding object or each welding condition, taking into account the relationship when the state of the back surface bead 93 is good. If the control and calculation unit 11 determines that the amount of variation in the back surface brightness is within the allowable range (YES), it determines that the state of the back surface bead 93 is "good" (S303). On the other hand, if the control and calculation unit 11 determines that the amount of variation in the back surface brightness is not within the allowable range (NO), it determines that the state of the back surface bead 93 is "poor" (S304). Then, after making the determination in S303 or S204, the control and calculation unit 11 ends the execution of the welding condition determination step.

[0083] In this way, the control calculation unit 11 may calculate the amount of variation in the brightness of the back surface of the welded portion 92 from the captured image, and determine whether the condition of the back wave 93 is good or bad based on whether the amount of variation is within a predetermined tolerance range.

[0084] According to the welding condition determination device 1 and the like, the quality of the condition of the backside wave 93 can be determined simply from the amount of variation in the brightness of the backside of the welded portion 92.

[0085] In the above description of this embodiment, the amount of variation in brightness referred to when determining whether the state of the back wave 93 is good or bad is based on the brightness of the back surface observed from the captured image at predetermined intervals during the averaging time. However, the amount of variation in brightness may be obtained by referring to time history observation data (waveform data) of brightness having a time length similar to the averaging time, instead of the brightness of the back surface observed at predetermined intervals during the averaging time.

[0086] Alternatively, the control calculation unit 11 may acquire waveform data by applying the accumulated data, which has been obtained in advance through a construction test that combines the acquisition of observed brightness waveforms and the determination of the quality of the back wave condition, to artificial intelligence, and then calculate the amount of variation by referring to the waveform data.

[0087] This welding condition determination device 1, etc., can be advantageous, for example, in increasing the speed or accuracy of processing related to welding condition determination when simply determining whether the condition of the backside wave 93 is good or bad based on the amount of variation in brightness on the back side of the welded portion 92.

[0088] (Fourth embodiment) In each of the above embodiments, the quality of the welded state in full penetration welding is determined by determining the quality of the backside rib 93. In contrast, in the fourth embodiment, the quality of the welded state in partial penetration welding in which the backside rib does not appear on the back side of the weld 92 is determined based on the luminance distribution on the back side of the weld 92.

[0089] Fig. 15 is a flowchart showing a welding condition determination step as a welding condition determination method according to the fourth embodiment. The welding condition determination step is executed by control calculation unit 11 as a control program for welding condition determination device 1, which includes a series of steps according to the flowchart shown in Fig. 15. Note that the basic configuration of welding condition determination device 1 according to this embodiment and the welding object are the same as those in the first embodiment, and therefore, the same reference numerals will be used for the respective components and the welding object, and detailed description thereof will be omitted below.

[0090] Control and calculation unit 11 starts executing the welding condition determination process in conjunction with the start of welding by welding machine 80. First, control and calculation unit 11 executes a back surface luminance observation process S400. In the back surface luminance observation process S400, control and calculation unit 11 causes image capture device 10 to capture an image of the welding object, similar to the third embodiment. Then, control and calculation unit 11 observes the luminance of the back surface where back waves 93 occur from the captured image.

[0091] Next, the control and calculation unit 11 executes a luminance distribution calculation step S401. In the luminance distribution calculation step S401, the control and calculation unit 11 calculates the distribution of the back surface luminance observed in the back surface luminance observation step S400.

[0092] Next, the control and calculation unit 11 determines whether the distribution of backside luminance identified in the luminance distribution calculation step S401 includes a region exceeding a threshold (S402). The distribution of backside luminance is defined, for example, by width, length, or area. The threshold for the distribution of backside luminance is determined in advance, taking into account the relationship between the material of the welding object, the type of joint, the penetration depth, and other processing conditions and the results of good partial penetration welding. If the control and calculation unit 11 determines that the distribution of backside luminance does not include a region exceeding the threshold (NO), it determines that the partial penetration welding is "good" (S403). On the other hand, if the control and calculation unit 11 determines that the distribution of backside luminance includes a region exceeding the threshold (YES), it determines that the partial penetration welding is "poor" (S404). After making the determination in S403 or S404, the control and calculation unit 11 terminates the execution of the welding condition determination step.

[0093] In this way, two members or portions may be joined together by partial penetration welding. The control and calculation unit 11 may calculate the luminance distribution on the back surface of the weld 92 based on the captured image, and determine whether the partial penetration welding is successful based on whether the luminance distribution includes an area exceeding a predetermined threshold.

[0094] This welding condition determination device 1 and the like can be applied to determine the quality of partial penetration welding not only in the full-penetration welding exemplified in the above embodiments, but also in cases where two members or parts are joined to each other by partial penetration welding.

[0095] In the above description of this embodiment, the quality of partial penetration welding is determined from the distribution of backside luminance determined based on the captured image. However, instead of a distribution expressed in luminance, for example, a distribution expressed by the intensity of each RGB color may be determined, and the quality of partial penetration welding may be determined from the distribution of that RGB region.

[0096] In addition, in the above-described embodiments, examples have been given in which welding machine 80 employs a laser welding method or a laser-arc hybrid welding method. However, welding machine 80 may employ not only the above-described welding methods that use laser light, but also various welding methods that are classified as arc welding that does not use laser light, for example.

[0097] Although several embodiments have been described, the embodiments can be modified or varied based on the above disclosure. All components of the above embodiments and all features described in the claims may be individually extracted and combined, unless they contradict each other. [Explanation of symbols]

[0098] 1. Welding condition determination device 10 Imaging equipment 10a wide-angle lens 11 Control and calculation section 90 First member 91 Second member 92 Welded parts 93 Back Wave WL welding line

Claims

1. an imaging device having a wide-angle lens, which, when two members or parts are welded, captures an image of the rear surface of the weld, including the entire weld line set on the rear surface of the weld, through the wide-angle lens; a control and calculation unit that determines whether the state of the weld is good or bad based on the image captured by the imaging device; Equipped with The welding condition determination device is configured such that the photographing direction of the rear surface of the weld is inclined relative to the extension direction of the weld line.

2. an imaging device having a wide-angle lens, which captures an image of the back surface of the welded portion through the wide-angle lens when two members or parts are welded; a control and calculation unit that determines whether the state of the weld is good or bad based on the image captured by the imaging device; Equipped with The welding condition determination device is configured such that the photographing device is positioned so that the entire weld line set on the back side of the welded portion is within the angle of view.

3. The welding condition determination device according to claim 1 or 2, wherein the control and calculation unit determines whether a state of a back bead formed along the weld line is good or bad based on the captured image.

4. 4. The welding condition determination device according to claim 3, wherein the control and calculation unit measures the amount of spatter generated from the captured image and determines whether the state of the back-strip is good or bad based on whether the amount of spatter generated is within a predetermined allowable amount.

5. 4. The welding condition determination device according to claim 3, wherein the control and calculation unit measures a deviation amount of a high-brightness area in the captured image where the back swell occurs, and determines whether the state of the back swell is good or bad based on whether the deviation amount is within a predetermined tolerance range.

6. 4. The welding condition determination device according to claim 3, wherein the control and calculation unit calculates a variation in brightness of the back surface of the weld from the captured image, and determines whether the state of the back wave is good or bad based on whether the variation is within a predetermined allowable range.

7. 7. The welding condition determination device according to claim 6, wherein the control and calculation unit acquires waveform data by applying to an artificial intelligence data accumulated in advance by performing a construction test that combines acquisition of an observed waveform of the luminance and determination of the quality of the back wave state, and calculates the amount of variation by referring to the waveform data.

8. The two members or portions are joined together by partial penetration welding; 3. The welding condition determination device according to claim 1, wherein the control and calculation unit calculates a luminance distribution of the back surface of the weld based on the captured image, and determines whether the partial penetration welding is good or bad based on whether the luminance distribution includes an area exceeding a predetermined threshold value.

9. an imaging process for imaging the rear surface of the welded portion, including the entire weld line set on the rear surface of the welded portion, using a wide-angle lens from a direction inclined with respect to the extension direction of the weld line when two members or parts are welded; a step of determining whether the state of the weld is good or bad based on the image captured in the photographing step; A welding condition determination method including:

Images