Processing equipment and welding system

Abstract

To provide a processing device and a welding system which can suppress the occurrence of a defect in welding.SOLUTION: A processing device according to an embodiment executes determination processing of determining the state of welding by using a first image obtained by imaging at least a portion of a weld pool. The state includes the first state and the second state that is more unstable than the first state. In the determination processing, when a ripple exists in the weld pool, the welding is determined to be in the second state. The processing device does not correct a condition of the welding when it is determined that the welding is in the first state. The processing device corrects the condition when it is determined that the welding is in the second state.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] FIELD OF THE INVENTION Embodiments of the present invention relate to processing devices and welding systems. [Background technology]

[0002] There is a need to develop technology that can suppress the occurrence of welding defects. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2021 / 049610 Summary of the Invention [Problem to be solved by the invention]

[0004] An object of the present invention is to provide a processing device and a welding system that can suppress the occurrence of defects in welding. [Means for solving the problem]

[0005] The processing device according to the embodiment executes a determination process for determining the state of welding using a first image capturing at least a portion of the molten pool. The state includes a first state and a second state that is more unstable than the first state. In the determination process, if ripples are present in the molten pool, the welding is determined to be in the second state. If the processing device determines that the welding is in the first state, the processing device does not correct the welding conditions. If the processing device determines that the welding is in the second state, the processing device corrects the conditions. [Brief explanation of the drawings]

[0006] [Figure 1] FIG. 1 is a schematic diagram showing a welding system according to an embodiment. [Figure 2] FIG. 2 is a schematic diagram showing the welding process. [Figure 3]FIG. 3 is a flowchart showing a processing method performed by the processing device according to the embodiment. [Figure 4] FIG. 4 is a flowchart showing a first determination method performed by the processing device according to the embodiment. [Figure 5] FIG. 5 is an example of an image captured by the imaging device. [Figure 6] FIG. 6 is an example of an image captured by the imaging device. [Figure 7] FIG. 7 shows an example of an image captured by the imaging device. [Figure 8] FIG. 8 is a graph illustrating the average pixel values of each image. [Figure 9] FIG. 9 is a graph illustrating the variation of pixel values relative to the moving average value. [Figure 10] FIG. 10 is a graph illustrating the variation of pixel values relative to the moving average value. [Figure 11] FIG. 11 is a flowchart showing a second determination method performed by the processing device according to the embodiment. [Figure 12] FIG. 12 shows an example of an image captured by the imaging device. [Figure 13] FIG. 13 is an example of an image captured by the imaging device. [Figure 14] Figures 14(a), 14(c), and 14(e) are examples of images of the molten pool, while Figures 14(b), 14(d), and 14(f) are schematic diagrams showing the state of the molten pool in Figures 14(a), 14(c), and 14(e), respectively. [Figure 15] FIG. 15 is a schematic diagram showing another welding system according to an embodiment. [Figure 16] FIG. 16 is a schematic diagram showing a hardware configuration. DETAILED DESCRIPTION OF THE INVENTION

[0007] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The drawings are schematic or conceptual, and the relationship between the thickness and width of each part, the size ratio between parts, etc. are not necessarily the same as those in reality. Even when the same part is shown, the dimensions and ratios may be different depending on the drawing. In the present specification and the drawings, elements similar to those already explained are given the same reference numerals and detailed explanations will be omitted as appropriate.

[0008] FIG. 1 is a schematic diagram showing a welding system according to an embodiment. The welding system 1 includes a processing device 10, a welding device 20, a control device 30, a storage device 35, an imaging device 40, and a lighting device 50.

[0009] The welding device 20 welds two or more members 5 together. The welding device 20 performs, for example, arc welding. Specific examples of arc welding include tungsten inert gas (TIG) welding, metal inert gas (MIG) welding, metal active gas (MAG) welding, and carbon dioxide arc welding. Here, an example will be described in which the welding device 20 performs MAG welding or MIG welding.

[0010] As shown in FIG. 1, the welding apparatus 20 includes a wire supply unit 21, a wire 22, a wire coil 23, a torch 24, a tube 25, a gas supply unit 26, and a gas source 27.

[0011] The wire supply unit 21 has a wire coil 23 around which a wire 22 (consumable electrode) is wound. The wire supply unit 21 supplies the wire 22 of the wire coil 23 to a torch 24 via a tube 25.

[0012] The gas supply unit 26 is connected to the gas source 27 and supplies the shielding gas stored in the gas source 27 to the torch 24. In the illustrated example, the shielding gas is supplied to the torch 24 via the wire supply unit 21 and the tube 25. An inert gas (e.g., argon) can be used as the shielding gas. When MAG welding is performed, the shielding gas may include an active gas such as carbon dioxide gas in addition to the inert gas.

[0013] A wire 22 and a shielding gas are supplied to the torch 24. At the tip of the torch 24, the wire 22 protrudes toward the member 5. The torch 24 also ejects the supplied shielding gas around the wire 22.

[0014] The control device 30 controls each element of the welding device 20. Specifically, the control device 30 controls the wire supply speed by the wire supply unit 21, the flow rate of the shielding gas by the gas supply unit 26, the potential at the tip of the wire 22, the potential of the member 5, the current flowing through the tip of the wire 22, the current waveform, etc. For example, the control device 30 controls the respective potentials of the member 5 and the wire 22 so that the wire 22 is a positive electrode (+) and the member 5 is a negative electrode (-).

[0015] Memory device 35 stores data necessary for welding. For example, memory device 35 stores welding conditions such as the wire feed speed, the flow rate of the shielding gas, the voltage and current values during welding, and the current waveform. Control device 30 causes welding device 20 to perform welding in accordance with the preset welding conditions.

[0016] The imaging device 40 captures an image (still image) of the molten pool during welding. The imaging device 40 may also capture a moving image. In this case, a still image is extracted from the moving image. The imaging device 40 is, for example, a camera including a CMOS image sensor or a CCD image sensor.

[0017] The illuminator 50 irradiates the molten pool during welding with light. The illuminator 50 includes a light-emitting diode or a lamp. The illuminator 50 may also emit laser light. The wavelength range of the light emitted from the illuminator 50 is arbitrary. As an example, the illuminator 50 emits infrared light. Note that the illuminator 50 may be omitted if an image usable for determining the welding condition, as described below, can be obtained without illuminating the molten pool with the illuminator 50.

[0018] An optical filter 45 may be provided on the optical path between the molten pool and the imaging device 40. The optical filter 45 selectively transmits only light in a predetermined wavelength range. The optical filter 45 may also be located on the optical path between the lighting 50 and the molten pool. For example, by providing an optical filter 45 that selectively transmits only infrared light, it is possible to prevent overexposure of the image due to arc light emission. A bandpass filter may be used as the optical filter 45.

[0019] The processing device 10 determines the state of stability of the welding based on the image acquired by the imaging device 40. For example, the processing device 10 determines whether the welding is in a first state or a second state. The second state indicates that the welding is more unstable than the first state. The state of the welding may be determined more precisely as one of three or more states including the first state and the second state.

[0020] When the welding is determined to be in the second state, the processing device 10 corrects the preset welding conditions. The processing device 10 transmits the corrected welding conditions to the control device 30. The control device 30 welds the members 5 in accordance with the corrected welding conditions. For example, the welding conditions include the wire feed speed, the shielding gas flow rate, the voltage value, the current value, and the movement speed of the torch 24. The correction adjusts one or more of these conditions.

[0021] FIG. 2 is a schematic diagram showing the welding process. As shown in FIG. 2, during welding, an arc A is generated between the tip of the wire 22 and the member 5. A shielding gas G is sprayed around the arc A. The heat of the arc A causes the wire 22 to melt and drip onto the member 5. The heat of the arc A also melts a portion of the member 5. A molten pool WP is formed by the molten member 5 and the wire 22. When the molten pool WP solidifies, a solidified layer 6 is formed on the surface of the member 5. The multiple members 5 are joined by the solidified layer 6 to form a joined body.

[0022] FIG. 3 is a flowchart showing a processing method performed by the processing device according to the embodiment. In the processing method PM, after welding starts, the processing device 10 receives an image of the molten pool WP (step St1). The image is acquired by the imaging device 40. The processing device 10 executes a determination process (step St2). In the determination process, the state of the welding is determined using the image. The processing device 10 determines whether the welding is determined to be in the second state in step St2 (step St3). If the welding is determined to be in the second state in step St2, the processing device 10 corrects the welding conditions (step St4). If the welding is determined to be in the first state, the processing device 10 does not correct the welding conditions. The processing device 10 determines whether the welding has ended (step St5). If the welding has not ended, step St1 is executed again. As a result, the state of the welding is repeatedly determined during welding. The welding conditions are corrected appropriately depending on the state determination result.

[0023] Hereinafter, a method for determining the welding condition in step St2 will be described with reference to a specific example. Either of the following two determination methods can be used to determine the welding condition. In the specific example, a solid wire is used as the wire 22. A partially plated SS400 is used as the member 5. The wire feed speed is 7 m / min. The arc length correction is set to either -3, 0, or 3. The arc length correction is a correction parameter related to the arc length, and relates to the voltage value, current value, and current waveform.

[0024] (First determination method) Fig. 4 is a flowchart showing a first determination method performed by the processing device according to the embodiment. Figs. 5 to 7 are examples of images captured by the imaging device. In the first determination method, the welding condition is determined based on pixel values of the image. First, the processing device 10 cuts out a part of the image acquired by the imaging device 40 (step St211). Specifically, the processing device 10 cuts out a part of the molten pool that is located behind the heat source.

[0025] FIG. 5 shows image IMG1 (an example of a second image) acquired by the imaging device 40. Image IMG1 shows the wire 22, arc A (heat source), molten pool WP, and solidified layer 6. The wire 22 and arc A are moving in direction D1. That is, direction D1 is "forward" as viewed from the position of arc A. An image of portion P1, which is located "rearward" from the position of arc A and opposite direction D1, is cut out. FIG. 6 shows image IMG2 (an example of a first image) of the cut-out portion P1.

[0026] The position to be cut out is set in advance, or the position to be cut out may be set automatically based on the recognition results of the wire 22, the arc A, the molten pool WP, etc. in the image.

[0027] The processing device 10 performs image processing on the image IMG2 (step St212). For example, the image processing may involve edge enhancement (dispersion filtering) or edge detection. Fig. 7 shows an image IMG3 obtained by applying edge enhancement to the image IMG2.

[0028] The processing device 10 calculates a first value from a plurality of pixel values in the processed image (step St213). For example, when a grayscale image or an RGB image is acquired by the imaging device 40, the pixel value is luminance. When an HSV image is acquired by the imaging device 40, the pixel value is hue, saturation, or brightness. For example, the first value is calculated by the average, sum, or weighted average of all pixel values included in the image. The first value may also be calculated by the average, sum, or weighted average of some pixel values included in the image.

[0029] The processing device 10 calculates the moving average value and the variance of the first value (step St214). In the processing method PM, images are repeatedly acquired during welding. The first value is calculated for each image. Therefore, the first value for each image is calculated for each time period. The processing device 10 calculates the moving average value and the variance thereof with respect to time.

[0030] As an example, the imaging device 40 captures video of the molten pool at 500 fps. That is, the imaging device 40 captures 500 images per second. The processing device 10 calculates a first value for each image. The processing device 10 calculates a moving average value and a variance over a 0.2-second interval (100 images). The variance may be standard deviation, variance, or mean square error. The variance may be used as an evaluation value for determining the state of the weld.

[0031] Fig. 8 is a graph illustrating an example of the variation of pixel values of each image with respect to the moving average value, and Figs. 9 and 10 are graphs illustrating an example of the variation of pixel values with respect to the moving average value. In Fig. 8, the horizontal axis represents time T (seconds). The vertical axis represents the average value Ave of pixel values in each image. In Figs. 9 and 10, the horizontal axis represents time T (seconds). The vertical axis represents variation V. Each image used to generate the graphs in Figs. 8 to 10 is an RGB image to which edge enhancement has been applied. The pixel values in each image range from 0 to 255.

[0032] In the example of Figure 8, an arc is struck at about 0.3 seconds, and welding begins. Between 0.3 and 0.8 seconds, the average value Ave increases significantly and then decreases. After that, the average value Ave generally fluctuates within the range of 0 to 60. Welding ends at 6 seconds.

[0033] Figure 9 shows the variation (standard deviation) relative to the moving average value at each time T. The variation fluctuates significantly temporarily from immediately after the start of welding until around 0.6 seconds. After that, the variation fluctuates generally within the range of 5 to 30.

[0034] The processing device 10 compares the variation with a preset judgment condition. As a specific example, the judgment condition includes a threshold value. The processing device 10 determines whether the variation is less than the preset threshold value (step St215). If the variation is less than the threshold value, the processing device 10 judges that the welding is in a first state (step St216). If the variation is equal to or greater than the threshold value, the processing device 10 judges that the welding is in a second state (step St217). This completes step St2 in the flowchart shown in FIG. 3.

[0035] FIG. 10 shows the same graph as FIG. 9. As an example, as shown in FIG. 10, the threshold value TH is set at V=17. As a result, the welding is determined to be in the second state S2 around 0.7 seconds, 1.2 seconds, 1.9 seconds, 2.8 seconds, 3.6 seconds, 4.2 seconds, and 5.0 to 5.8 seconds. The welding is determined to be in the first state during other periods. If the welding is determined to be in the second state, the welding conditions are corrected. If the welding state is improved by correcting the welding conditions, the variation V can be reduced.

[0036] (Second Judgment Method) In the second determination method, a model is used to determine the welding condition. The model outputs a classification of the welding condition in response to an input image of the weld pool. The model preferably includes a neural network that has undergone supervised learning.

[0037] FIG. 11 is a flowchart showing a second determination method performed by the processing device according to the embodiment. First, the processing device 10 cuts out a part of the image acquired by the imaging device 40 (step St221). Specifically, the processing device 10 cuts out the part of the molten pool that is located directly below the heat source.

[0038] 12 and 13 are examples of images captured by the imaging device. Image IMG11 (an example of a second image) shown in FIG. 12, like image IMG1, shows the wire 22, arc A (heat source), molten pool WP, and solidified layer 6. The wire 22 and arc A are moving in direction D1. An image of portion P2 directly below arc A is cut out. FIG. 13 shows image IMG12 (an example of a first image) of the cut-out portion P2.

[0039] The processing device 10 inputs the cut-out image into the model (step St222). The model classifies the welding state. The processing device 10 acquires the classification result from the model (step St223). The processing device 10 determines the welding state based on the classification result (step St224). For example, the classification result from the model indicates the probability of each of the first state and the second state. If the probability of the first state is higher than the probability of the second state, the processing device 10 determines that welding is in the first state. If the probability of the second state is higher than the probability of the first state, the processing device 10 determines that welding is in the second state. This completes step St2 in the flowchart shown in FIG. 3.

[0040] The advantages of the embodiment will be described. During welding, defects such as spatter, pits, and undercuts can occur. It is desirable to minimize the number and severity of these defects. Previously, attempts have been made to adjust welding conditions using information obtained during welding to prevent defects from occurring. Information that can be used to determine whether adjustments are necessary has also been studied. As a result of their investigation, the inventors discovered that waves in the molten pool become larger, causing ripples to form in the molten pool, especially immediately before pits occur. Based on this knowledge, the inventors have invented a method for determining whether welding conditions need to be adjusted based on the occurrence of ripples.

[0041] Figures 14(a), 14(c), and 14(e) are examples of images of the molten pool, while Figures 14(b), 14(d), and 14(f) are schematic diagrams showing the state of the molten pool in Figures 14(a), 14(c), and 14(e), respectively. In image IMG21 of Figure 14(a), small waves W caused by dripping of molten wire 22 spread gently concentrically in the weld pool WP, as shown in Figure 14(b). There are virtually no ripples in the weld pool WP. In image IMG22 of Figure 14(c), numerous irregularities are present in the weld pool WP. Compared to image IMG21, the surface of the weld pool WP is wavy, with numerous ripples R occurring as shown in Figure 14(d). In image IMG23 of Figure 14(e), similar to image IMG22, numerous irregularities are present in the weld pool WP. In addition, in image IMG23, larger ripples R are present on the surface of the weld pool WP due to bubbles B, as shown in Figure 14(f).

[0042] In areas where there are no ripples, the direction of reflected light is almost uniform. Therefore, there is little change in pixel value in areas where there are no ripples. On the other hand, the direction of reflected light changes near the ripples. In the image, there is a greater change in pixel value near the ripples than in areas where there are no ripples. As a result, for example, the average pixel value of image IMG22 is greater than the average pixel value of image IMG21. By calculating a moving average between multiple images and calculating the variation relative to this moving average, images in which pixel values change can be identified. In other words, images in which ripples are present in the molten pool WP can be identified. As a result, if ripples are present in the molten pool WP in the image, it is determined that the welding is in the second state.

[0043] In addition to pixel values, the welding condition can be determined by classifying the weld pool condition using a model. When using a model, images without ripples, as shown in Figure 14(a), and images with ripples, as shown in Figures 14(c) and 14(e), are prepared in advance during learning. These images and the state classification results (training data) are used to train the model. As a result, images in which ripples are present in the weld pool WP are classified as welding in the second state.

[0044] By correcting the welding conditions when ripples are present, the occurrence of defects (particularly pits) can be suppressed, and the degree and number of defects can be reduced. For example, according to the embodiment, the occurrence of poor welding can be suppressed, and the quality of the joined body can be improved.

[0045] Preferably, the welding conditions to be corrected include the voltage value, the current value, the current waveform, or the shielding gas flow rate. The inventors have confirmed through their investigations that these conditions are correlated with the occurrence of ripples. Specifically, when the welding is determined to be in the second state, the processing device 10 performs one or more of the following actions: increasing the voltage value, increasing the current value, and increasing the shielding gas flow rate.

[0046] The voltage value, current value, and shielding gas are thought to be related to the welding condition for the following reasons. When the flow rate of shielding gas is low, the molten pool is more likely to be exposed to air. As a result, components in the air are more likely to enter the molten pool. When components in the air enter the molten pool, some of the components are released as gas within the molten pool as it cools, leading to the formation of pits. Increasing the shielding gas flow rate can reduce the exposure of the molten pool to air. When the current or voltage is increased, the amount of heat input to the molten pool increases. This keeps the molten pool in a molten state for longer, promoting the release of gas and suppressing the formation of pits.

[0047] After the welding conditions are corrected, welding may be continued under the corrected welding conditions, or, when the welding is subsequently determined to be in the first state, the welding conditions may be changed from the corrected values to preset standard values.

[0048] Preferably, the processing device 10 does not perform image-based condition assessment from the start of welding until the first time has elapsed. Alternatively, the processing device 10 ignores assessment results obtained from the start of welding until the first time has elapsed. In either case, the welding conditions are not corrected regardless of the welding condition. For example, as shown in FIG. 9, the variation V is large until 0.8 seconds due to fluctuations in the size of the molten pool immediately after the start of welding. The period from 0 seconds to 0.8 seconds is an example of the first time. If the welding condition is assessed based on images from 0 seconds to 0.8 seconds, the welding conditions may be unnecessarily corrected. This may also result in reduced welding stability. By not correcting the welding conditions based on the assessment results during the first time, stability immediately after the start of welding can be improved.

[0049] The first time period may be set in advance by a user. Alternatively, the first time period may be set based on data obtained from an image. For example, the end point of the first time period is set to the timing when the variation V first falls below the threshold value TH after the variation V first exceeds the threshold value TH. When a model is used, the end point of the first time period is set to the timing when the welding state is first classified as the first state after the welding state is first classified as the second state.

[0050] For welding device 20, torch 24 may be hand-held or machine-held. Preferably, torch 24 is machine-held.

[0051] FIG. 15 is a schematic diagram showing another welding system according to an embodiment. In the welding system 1a shown in Fig. 15, the welding device 20 further includes a moving device 28. In the example shown in Fig. 15, one device is provided as the wire supply unit 21 and the gas supply unit 26. The imaging device 40 and the lighting 50 are omitted.

[0052] The moving device 28 includes an actuator (not shown). The moving device 28 is controlled by the control device 30 and moves the torch 24. Welding is performed while the torch 24 moves relative to the member 5. The moving device 28 may be an XY Cartesian robot, a horizontal articulated robot, or a vertical articulated robot.

[0053] Imaging device 40 and lighting 50 (not shown) also move in synchronization with torch 24. In welding system 1a, as in welding system 1, the welding state is determined using images taken during welding, and welding conditions are corrected as appropriate. This reduces the degree and number of defects.

[0054] In the above, an example in which arc welding is performed has been described. The embodiment can also be applied to laser welding. In laser welding, a molten pool is formed in the portion of the member 5 irradiated with a laser (heat source). The welding conditions can be corrected depending on whether ripples are present in the molten pool, thereby improving the stability of the welding.

[0055] In addition to the determination based on the pixel values of the image, the welding condition may also be determined using the intensity distribution of reflected laser light. For example, the lighting device 50 irradiates the molten pool WP with laser light. The laser light is reflected by the molten pool. A light receiving device that receives the reflected light is provided instead of the imaging device 40. The light receiving device generates an intensity distribution of the reflected light and transmits it to the processing device 10. The processing device 10 determines the welding condition from the periodicity of the intensity distribution.

[0056] Alternatively, the welding condition may be determined using the intensity distribution of infrared light from the molten pool. For example, a thermal camera is provided as the imaging device 40. The thermal camera detects infrared light emitted from the molten pool. The thermal camera generates an intensity distribution of the infrared light and transmits it to the processing device 10. The processing device 10 determines the welding condition from the periodicity of the intensity distribution.

[0057] FIG. 16 is a schematic diagram showing a hardware configuration. Each of the processing device 10 and the control device 30 is realized by a general-purpose or dedicated computer. The functions of each of the processing device 10 and the control device 30 may be realized by cooperation of multiple computers. Each of the processing device 10 and the control device 30 includes, for example, the hardware configuration shown in FIG. 16.

[0058] The computer 90 shown in FIG. 16 includes a CPU 91 , a ROM 92 , a RAM 93 , a storage device 94 , an input interface 95 , an output interface 96 , and a communication interface 97 .

[0059] The ROM 92 stores programs that control the operation of the computer. The ROM 92 stores programs necessary for the computer to execute each of the above-mentioned processes. The RAM 93 functions as a storage area in which the programs stored in the ROM 92 are expanded.

[0060] The CPU 91 includes a processing circuit. The CPU 91 uses a RAM 93 as a work memory and executes a program stored in at least one of a ROM 92 and a storage device 94. During program execution, the CPU 91 controls each component via a system bus 98 and executes various processes.

[0061] The storage device 94 stores data necessary for executing the program and data obtained by executing the program. The storage device 94 functions as the storage device 35.

[0062] The input interface (I / F) 95 connects the computer 90 and the input device 95a. The input I / F 95 is, for example, a serial bus interface such as USB. The CPU 91 can read various data from the input device 95a via the input I / F 95.

[0063] The output interface (I / F) 96 connects the computer 90 and the display device 96a. The output I / F 96 is, for example, a video output interface such as a Digital Visual Interface (DVI) or a High-Definition Multimedia Interface (HDMI (registered trademark)). The CPU 91 can transmit data to the display device 96a via the output I / F 96 and display an image on the display device 96a.

[0064] A communication interface (I / F) 97 connects the computer 90 to a server 97a external to the computer 90. The communication I / F 97 is, for example, a network card such as a LAN card. The CPU 91 can read various data from the server 97a via the communication I / F 97. A camera 99 captures images of the molten pool and stores the images in the server 97a. The camera 99 functions as the imaging device 40.

[0065] The storage device 94 includes one or more selected from a hard disk drive (HDD) and a solid state drive (SSD). The input device 95a includes one or more selected from a mouse, a keyboard, a microphone (voice input), and a touchpad. The display device 96a includes one or more selected from a monitor and a projector. A device having the functions of both the input device 95a and the display device 96a, such as a touch panel, may also be used.

[0066] The various data processing operations described above may be recorded as a program that can be executed by a computer on a magnetic disk (such as a flexible disk or hard disk), an optical disk (such as a CD-ROM, CD-R, CD-RW, DVD-ROM, DVD±R, DVD±RW), a semiconductor memory, or other non-transitory computer-readable storage medium.

[0067] For example, information recorded on a recording medium can be read by a computer (or an embedded system). The recording medium may have any recording format (storage format). For example, a computer reads a program from the recording medium and causes a CPU to execute instructions written in the program based on the program. The computer may acquire (or read) the program via a network.

[0068] According to the embodiments described above, a processing device, a welding system, a processing method, a program, and a storage medium are provided that are capable of suppressing the occurrence of defects in welding.

[0069] Although several embodiments of the present invention have been described above, these embodiments are presented by way of example only and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, modifications, etc. can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as set forth in the claims. Furthermore, the above-described embodiments can be implemented in combination with each other. [Explanation of symbols]

[0070] 1, 1a: welding system, 5: component, 6: solidified layer, 10: processing device, 20: welding device, 21: wire supply unit, 22: wire, 23: wire coil, 24: torch, 25: tube, 26: gas supply unit, 27: gas source, 28: moving device, 30: control device, 35: storage device, 40: imaging device, 45: optical filter, 50: lighting, 90: computer, 91: CPU, 92: ROM, 93: RAM, 94: storage device, 95: input interface, 95a: input device, 96: output interface, 96a: display device, 97: communication interface, 97a: server, 98: system bus, 99: camera, A: arc, B: bubble, D1: direction, G: shielding gas, PM: processing method, R: ripple, S2: second state, TH: Threshold, V: Variation, W: Wave, WP: Weld pool

Claims

1. A determination process for determining the state of welding using a wire using a first image that captures at least a part of a molten pool, the state including a first state and a second state that is more unstable than the first state, and the welding being determined to be in the second state if ripples other than waves caused by dripping of the molten wire are present in the molten pool, the determination process being performed; When the welding is determined to be in the first state, the welding conditions are not corrected, and when the welding is determined to be in the second state, the conditions are corrected. A processing device, The presence of ripples other than waves caused by dripping of the molten wire is determined by calculating an evaluation value based on pixel values of the first image and comparing the evaluation value with a predetermined threshold value.

2. The processing apparatus according to claim 1 , wherein the first image is generated by cutting out a portion of the molten pool located behind the heat source from a second image of the molten pool.

3. The calculation of the evaluation value is calculating a first value that is an average, sum, or weighted average of a plurality of the pixel values in the first image; Calculating a moving average and a variance of the first values for each of the plurality of first images; Including, The processing device according to claim 1 , wherein the variation is used as the evaluation value.

4. 4. The processing device according to claim 1, wherein the evaluation value is calculated from the first image to which edge enhancement or edge detection has been applied.

5. 5. The processing device according to claim 1, wherein the conditions are not corrected until a first time period has elapsed since the start of the welding, regardless of the result of the state determination.

6. The first image shows the molten pool during arc welding, 6. The processing apparatus according to claim 1, wherein the condition to be corrected is one or more selected from a current value, a voltage value, a current waveform, and a shielding gas flow rate.

7. A processing device according to any one of claims 1 to 6; an imaging device that acquires the first image; a control device that causes the welding device to perform the welding under the corrected conditions; A welding system comprising:

Images