Oxygen concentration monitoring method, oxygen concentration monitoring device, and welding system

Abstract

Monitor the oxygen concentration in the welding atmosphere. [Solution] The oxygen concentration monitoring method is a method for monitoring the oxygen concentration in the welding atmosphere when welding an object to be welded. The oxygen concentration monitoring method includes step S10 of measuring the color of the welding light generated when welding an object to be welded, and step S20 of determining whether the oxygen concentration in the welding atmosphere is below a predetermined concentration based on the measured color of the welding light.

Description

Technical Field

[0004] , , , ,

[0005] , , , , ,

[0003] , , , ,

[0001] The present invention relates to an oxygen concentration monitoring method, an oxygen concentration monitoring device, and a welding system.

Background Art

[0006] [Figure 1] Figure 1 is a schematic diagram of the welding system. [Figure 2] Figure 2 is a block diagram of the welding system. [Figure 3] Figure 3 is a flowchart showing an example of an oxygen concentration monitoring method. [Figure 4] Figure 4 is a graph showing the measurement results of the RGB values ​​of welding light in a test example. [Figure 5] Figure 5 is a graph showing the calculation results of the R / B value in the example test. [Figure 6] Figure 6 is a block diagram of a welding system according to another embodiment. [Figure 7] Figure 7 is a flowchart showing an example of an oxygen concentration monitoring method according to another embodiment. [Modes for carrying out the invention]

[0007] Hereinafter, an embodiment of the technology disclosed herein will be described with reference to the drawings. Naturally, the embodiment described herein is not intended to particularly limit the present invention. Furthermore, components and parts that perform the same function will be appropriately denoted by the same reference numerals, and redundant descriptions will be omitted as appropriate.

[0008] Figure 1 is a schematic diagram of welding system 1. Welding system 1 is used when welding a workpiece 5. Welding system 1 may be used, for example, when manufacturing a lithium-ion battery with a case and a lid. In this case, the workpiece 5 may be a case and a lid. However, the type of workpiece 5 is not limited to this. The material of the workpiece 5 is not particularly limited. The workpiece 5 may be, for example, aluminum, an aluminum alloy, or a steel material. As shown in Figure 1, welding system 1 includes a welding apparatus 10.

[0009] The welding apparatus 10 welds the object to be welded 5. The type of welding apparatus 10 is not particularly limited. Various conventionally known welding apparatuses can be used as the welding apparatus 10. In the configuration shown in Figure 1, the welding apparatus 10 is a so-called laser welding apparatus that performs welding by irradiating the object to be welded 5 with laser light L. Although not shown in the figure, the welding apparatus 10 is equipped with a laser oscillator and a scanner head, etc.

[0010] Incidentally, if oxygen gas is present in the welding atmosphere when the workpiece 5 is being welded by the welding apparatus 10, the quality of the weld may deteriorate. Therefore, the inventors want to be able to weld the workpiece 5 while monitoring the oxygen concentration in the welding atmosphere. As a result of diligent research, the inventors have found that there is a correlation between the color of the welding light and the oxygen concentration in the welding atmosphere. According to the inventors' findings, when the welding light is red, the oxygen concentration is relatively low, and when the welding light is blue, the oxygen concentration is relatively high.

[0011] Here, welding light refers to the light generated when welding the object to be welded 5. Welding light includes, for example, laser reflected light, thermal radiation light, and plasma light. Laser reflected light is the light generated when the laser light L emitted from the welding apparatus 10 is reflected by the object to be welded 5. Thermal radiation light is the light emitted from the object to be welded 5 by thermal radiation. Plasma light is the light emitted from the plasma generated when the object to be welded 5 is welded.

[0012] The welding system 1 comprises an oxygen concentration monitoring device 20, a jetting mechanism 50, and a control device 60. The oxygen concentration monitoring device 20 monitors the oxygen concentration in the welding atmosphere when welding the workpiece 5 based on the color of the welding light. The oxygen concentration monitoring device 20 comprises an imaging device 25 and an image processing device 30.

[0013] The imaging device 25 captures an image including welding light. The imaging device 25 may capture still images or moving images. In this embodiment, the imaging device 25 captures still images. The placement of the imaging device 25 is not particularly limited, as long as it is in a position where welding light can be captured. For example, the imaging device 25 may be placed above the object to be welded 5.

[0014] The image processing device 30 is communicatively connected to the imaging device 25. The image processing device 30 may be connected to the imaging device 25 by a wired connection or by a wireless connection. The image processing device 30 determines whether the oxygen concentration in the welding atmosphere is below a predetermined concentration by measuring the color of the welding light from the image captured by the imaging device 25. The image processing device 30 may be composed of a computer, for example, which includes a communication interface, storage, memory, and a processor. The communication interface is an interface for sending and receiving data with devices such as the imaging device 25 and the control device 60. The storage stores programs and data necessary for the processor to perform various processing. The memory operates as the processor's work area.

[0015] Figure 2 is a block diagram of welding system 1. The image processing device 30 includes an image acquisition unit 31, a region selection unit 32, an RGB measurement unit 33, an R / B calculation unit 34, and a determination unit 35. The image acquisition unit 31 performs image acquisition processing. The region selection unit 32 performs region selection processing. The RGB measurement unit 33 performs RGB measurement processing. The R / B calculation unit 34 performs R / B calculation processing. The determination unit 35 performs determination processing. Details of each of these processes will be described later.

[0016] As shown in Figure 1, the injection mechanism 50 blows inert gas toward the workpiece 5 to be welded. In the configuration shown in Figure 1, a cylinder 55 containing inert gas is connected to the injection mechanism 50. In this embodiment, nitrogen gas is used as the inert gas. However, argon gas or helium gas may be used instead of nitrogen gas as the inert gas. In the configuration shown in Figure 1, the injection mechanism 50 includes an injection nozzle 51, a flow meter 52, and a flow control valve 53. The inert gas contained in the cylinder 55 is injected from the injection nozzle 51 and blown toward the workpiece 5 to be welded. The flow meter 52 measures the flow rate of the inert gas blown toward the workpiece 5 to be welded. The flow control valve 53 is a valve for adjusting the flow rate of the inert gas blown toward the workpiece 5 to be welded. Conventional known components can be used without particular limitation as the injection nozzle 51, flow meter 52, and flow control valve 53.

[0017] The control device 60 is communicatively connected to the image processing device 30, the flow meter 52, and the flow control valve 53. The control device 60 may be connected to these devices by wire or wirelessly. Based on the result of the image processing device 30's determination of the oxygen concentration in the welding atmosphere, the control device 60 controls the flow rate of inert gas sprayed onto the workpiece 5 by the injection mechanism 50. The control device 60 may be composed of a computer, for example, comprising a communication interface, storage, memory, and a processor. The communication interface is an interface for sending and receiving data with devices such as the image processing device 30, the flow meter 52, and the flow control valve 53. The storage stores programs and data necessary for the processor to perform various processes. The memory operates as the processor's work area. The image processing device 30 and the control device 60 may be a single computer or separate computers.

[0018] As shown in FIG. 2, the control device 60 includes a determination acquisition unit 61, a flow rate acquisition unit 62, and a flow rate control unit 63. The determination acquisition unit 61 acquires the result of the determination process executed by the image processing device 30. The flow rate acquisition unit 62 acquires the measurement result of the flow meter 52. The flow rate control unit 63 controls the operation of the flow rate control valve 53 to control the flow rate of the inert gas sprayed onto the welding object 5.

[0019] Next, the oxygen concentration monitoring method executed by the welding system 1 will be described. FIG. 3 is a flowchart showing an example of the oxygen concentration monitoring method. The oxygen concentration monitoring method is a method for monitoring the oxygen concentration in the welding atmosphere when welding the welding object. The oxygen concentration monitoring method is executed while welding the welding object 5. As shown in FIG. 3, the oxygen concentration monitoring method includes a step S10 of measuring the color of the welding light, a step S20 of determining whether the oxygen concentration is below a predetermined concentration, and a step S30 of increasing the flow rate of the inert gas.

[0020] In step S10 of measuring the color of the welding light, the color of the welding light generated when welding the welding object 5 is measured. In this embodiment, step S10 of measuring the color of the welding light includes an imaging step S11, an image acquisition step S12, a region selection step S13, an RGB measurement step S14, and an R / B calculation step S15. The imaging step S11 is executed by the imaging device 25. The image acquisition step S12, the region selection step S13, the RGB measurement step S14, and the R / B calculation step S15 are executed by the image processing device 30.

[0021] In the imaging step S11, an image including the welding light is captured by the imaging device 25. The timing at which the imaging step S11 is executed is not particularly limited. The imaging step S11 can be executed at any timing during the welding of the welding object 5. The timing for executing the imaging step S11 may be predetermined. The imaging step S11 may be executed, for example, every second during the period from the start of welding to the end of welding. In the imaging step S11, a color image is captured.

[0022] In the image acquisition step S12, an image acquisition process is executed. The image acquisition process is a process of acquiring, from the imaging device 25, the image captured by the imaging device 25 in the imaging step S11. The image acquisition process is executed, for example, when the image captured in the imaging step S11 is transmitted from the imaging device 25 to the image processing device 30 by wireless communication.

[0023] In the region selection step S13, a region selection process is executed. The region selection process is a process of selecting pixels in the image captured in the imaging step S11 that show the welding light based on a predetermined condition. Here, the predetermined condition may be, for example, the elapsed time from the start of welding. When the welding trajectory by the welding device 10 is predetermined, based on the elapsed time from the start of welding, it is possible to predict which position of the welding object 5 is being welded. Also, the welding light is radiated from the position where welding is being performed and its vicinity. Therefore, by grasping in advance the relationship between the start time from the start of welding and the pixels where the welding light appears, the pixels where the welding light appears can be appropriately selected from the image captured in the imaging step S11. In this embodiment, in the region selection step S13, two or more pixels are selected. However, the number of pixels selected in the region selection step S13 may be one.

[0024] In the RGB measurement step S14, the RGB measurement process is performed. The RGB measurement process measures the RGB values ​​of the pixels selected in the region selection step S13. As described above, in this embodiment, two or more pixels are selected in the region selection step S13. Therefore, in this embodiment, in the RGB measurement step S14, the average value of the RGB values ​​of each selected pixel is calculated. In this embodiment, this calculated value is used as the measured value of the RGB value of the welding light.

[0025] Here, the RGB value is a value expressed by a combination of R, G, and B values. Any color can be represented by the combination of R, G, and B values. The R value is the value of the red component, the G value is the value of the green component, and the B value is the value of the blue component. The R, G, and B values ​​can each be expressed in 256 steps from 0 to 255, for example. For example, if the R value of the welding light measured in the RGB measurement step S14 is sufficiently larger than the G and B values, the color of the welding light is reddish.

[0026] In the R / B calculation step S15, the R / B calculation process is performed. The R / B calculation process calculates the R / B value from the RGB values ​​of the welding light measured in the RGB measurement step S14. The R / B value is the ratio of the R value to the B value, and is calculated by dividing the R value by the B value.

[0027] Step S20, which determines whether the oxygen concentration is below a predetermined concentration, is performed by the image processing device 30. In step S20, a determination process is performed. The determination process determines whether the oxygen concentration in the welding atmosphere is below a predetermined concentration based on the color of the welding light measured in step S10. That is, in the determination process, it is determined whether the oxygen concentration in the welding atmosphere is below a predetermined concentration based on the RGB values ​​measured in step S10. In this embodiment, it is determined whether the oxygen concentration in the welding atmosphere is below a predetermined concentration by determining whether the R / B value calculated in the R / B calculation step S15 is above a predetermined threshold. According to the inventor's knowledge, for example, by setting the predetermined threshold to "1.5", it is possible to determine whether the oxygen concentration in the welding atmosphere is 3% or less. For example, if the calculated R / B value is 2.0, in step S20 it can be determined that the oxygen concentration in the welding atmosphere is 3% or less. If the calculated R / B value is 0.5, step S20 can determine that the oxygen concentration in the welding atmosphere is greater than 3%. However, the threshold set here is not limited to "1.5" and can be set appropriately within the range of 0.3 to 2.0, for example. The predetermined concentration can be changed as appropriate according to the set threshold.

[0028] As shown in Figure 3, in step S20, if it is determined that the R / B value is smaller than a predetermined threshold, step S30 is performed to increase the flow rate of the inert gas. That is, in step S20, if it is determined that the oxygen concentration in the welding atmosphere is greater than a predetermined concentration, step S30 is performed to increase the flow rate of the inert gas. In step S30, which increases the flow rate of the inert gas, the flow rate of the inert gas injected by the injection mechanism 50 is increased. In this embodiment, the control device 60 acquires the result of the determination process performed by the image processing device 30, and the control device 60 executes control to increase the flow rate of the inert gas. In this embodiment, the control device 60 increases the flow rate of the inert gas by controlling the flow control valve 53. Here, the amount of increase in the flow rate of the inert gas is set appropriately according to the welding conditions of the object to be welded 5, etc.

[0029] The following describes test examples demonstrating the technologies disclosed herein. However, this is not intended to limit the technologies disclosed herein to the following test examples.

[0030] In the test example, two aluminum plates were prepared as the objects to be welded. In the test example, the two aluminum plates were laser welded. The laser welding conditions were as follows: laser wavelength of 1070 nm, laser beam diameter of 0.6 mm, laser output of 2000 W, and welding speed of 200 mm / s. In the test example, while measuring the oxygen concentration in the welding atmosphere with an oxygen concentration sensor, the welding light generated during laser welding was captured with an imaging device, and the RGB values ​​of the welding light were measured. In the test example, the RGB values ​​of the welding light were measured when the oxygen concentration in the welding atmosphere was 1.0%, 3.0%, 5.0%, 10.0%, and 20.7%.

[0031] Figure 4 is a graph showing the measurement results of the RGB values ​​of welding light in a test example. The horizontal axis of the graph in Figure 4 represents the oxygen concentration in the welding atmosphere. The vertical axis of the graph in Figure 4 represents the intensity of the R, G, and B values. The R, G, and B values ​​are each expressed in 256 steps from 0 to 255. As can be seen from Figure 4, as the oxygen concentration increases, the R value decreases and the B value increases. Therefore, when the oxygen concentration is relatively high, the welding light color becomes bluish. When the oxygen concentration is relatively low, the welding light color becomes reddish.

[0032] Figure 5 is a graph showing the calculation results of the R / B value in the test example. The graph in Figure 5 is calculated by dividing the R value at each oxygen concentration shown in the graph in Figure 4 by the B value at each oxygen concentration shown in the graph in Figure 4. The horizontal axis of the graph in Figure 5 represents the oxygen concentration in the welding atmosphere. The vertical axis of the graph in Figure 5 represents the R / B value. As can be seen from Figure 5, there was a tendency for the R / B value to decrease as the oxygen concentration increased. Therefore, it was found that there is a correlation between the R / B value and the oxygen concentration in the welding atmosphere, and that the oxygen concentration in the welding atmosphere can be monitored using the R / B value.

[0033] As shown in Figure 3, the oxygen concentration monitoring method in the above embodiment includes a step S10 of measuring the color of the welding light and a step S20 of determining whether the oxygen concentration is below a predetermined concentration. According to the inventors' findings, there is a correlation between the color of the welding light and the oxygen concentration in the welding atmosphere. Therefore, this method makes it easy to monitor the oxygen concentration in the welding atmosphere while welding the object to be welded 5.

[0034] According to the embodiment described above, step S10 for measuring the color of welding light includes an imaging step S11, a region selection step S13, and an RGB measurement step S14. In the imaging step S11, an image including welding light is captured. In the region selection step S13, a region in the captured image that shows welding light is selected based on predetermined conditions. In the RGB measurement step S14, the RGB values ​​of the selected region are measured. This method allows for monitoring of oxygen concentration by capturing an image of the workpiece 5 being welded. Therefore, oxygen concentration can be easily monitored.

[0035] According to the embodiment described above, in the region selection step S13, two or more pixels in which welding light is captured are selected based on predetermined conditions. In the RGB measurement step S14, the average value of the RGB values ​​in each selected pixel is calculated. This calculates the average value of the RGB values ​​in multiple pixels of the captured image, thereby improving the measurement accuracy of the RGB values ​​in step S14.

[0036] According to the embodiment described above, step S10 for measuring the color of the welding light includes an R / B calculation step S15. In the R / B calculation step S15, an R / B value, which is the ratio of the R value to the B value, is calculated from the RGB values ​​of the measured welding light. In step S20 for determining whether the oxygen concentration is below a predetermined concentration, it is determined whether the oxygen concentration in the welding atmosphere is below a predetermined concentration by determining whether the calculated R / B value is above a predetermined threshold. This reduces noise caused by, for example, external light. As a result, the accuracy of oxygen concentration monitoring is improved. Here, external light includes, for example, light irradiated from lighting installed in the space where the object to be welded 5 is being welded.

[0037] According to the above embodiment, the oxygen concentration monitoring method includes step S30, which increases the flow rate of inert gas when it is determined that the oxygen concentration in the welding atmosphere is greater than a predetermined concentration. This allows the flow rate of inert gas blown onto the workpiece 5 to be welded to be controlled based on the oxygen concentration monitoring result. Therefore, the oxygen concentration in the welding atmosphere when welding the workpiece 5 can be appropriately controlled.

[0038] The above describes one embodiment of the proposed technology. However, the above-described embodiment is merely an example, and the technology can be implemented in other ways.

[0039] Figure 6 is a block diagram of a welding system 1 according to another embodiment. In the embodiment shown in Figure 6, a machine learning model 38 determines whether the oxygen concentration in the welding atmosphere is below a predetermined concentration. In the embodiment shown in Figure 6, the image processing device 30 includes an image acquisition unit 31, an RGB measurement unit 33, a determination unit 35, and a storage unit 36. Similar to the embodiment described above, the image acquisition unit 31 performs image acquisition processing. The RGB measurement unit 33 performs RGB measurement processing. The determination unit 35 performs determination processing.

[0040] The memory unit 36 ​​stores a machine learning model 38. The machine learning model 38 is a pre-trained model that has been trained using machine learning. Various machine learning algorithms can be used for the machine learning model 38; for example, a neural network algorithm is used. In the configuration shown in Figure 6, the machine learning model 38 has been pre-trained using the distribution of RGB values ​​in an image containing welding light and the oxygen concentration in the welding atmosphere as training data. The machine learning model 38 may be one generated within the image processing device 30, or one generated by an external computer.

[0041] Figure 7 is a flowchart showing an example of an oxygen concentration monitoring method according to another embodiment. The oxygen concentration monitoring method includes a step S10 for measuring the color of welding light, a step S20 for determining whether the oxygen concentration is below a predetermined concentration, and a step S30 for increasing the flow rate of inert gas. In the embodiment shown in Figure 7, the step S10 for measuring the color of welding light includes an imaging step S11, an image acquisition step S12, and an RGB measurement step S14.

[0042] In the configuration shown in Figure 7, the imaging step S11 and the image acquisition step S12 are performed in the same manner as in the configuration shown in Figure 3. In the configuration shown in Figure 7, the RGB measurement step S14 measures the distribution of RGB values ​​in the image captured in the imaging step S11. In the RGB measurement step S14, the distribution of RGB values ​​may be measured over the entire range of the captured image, or the distribution of RGB values ​​may be measured in a specific region of the captured image. Here, the specific region includes welding light. The method for selecting the specific region is not particularly limited. As a method for selecting the specific region, for example, the region selection process performed in the region selection step S13 in the configuration shown in Figure 3 may be used.

[0043] In step S20, which determines whether the oxygen concentration is below a predetermined concentration, a determination process is performed. In the configuration shown in Figure 7, in step S20, the distribution of measured RGB values ​​is input to the machine learning model 38 to determine whether the oxygen concentration in the welding atmosphere is below a predetermined concentration. That is, in the configuration shown in Figure 7, in step S20, the machine learning model 38 is used to determine whether the oxygen concentration in the welding atmosphere is below a predetermined concentration from the distribution of RGB values ​​of the measured image.

[0044] The configuration of the injection mechanism 50 is not limited to the form shown in Figure 1. For example, the injection mechanism 50 may include a mass flow controller instead of the flow meter 52 and flow control valve 53. Also, the welding system 1 does not have to include the injection mechanism 50. In this case, step S30, which increases the flow rate of the inert gas, is not performed.

[0045] In the above-described embodiment, the color of the welding light was measured by imaging the welding light. However, the method for measuring the color of the welding light is not limited to this. For example, the color of the welding light may be measured by a color sensor equipped with a photodiode or the like.

[0046] In the configuration shown in Figure 3, in step S20, it was determined whether the oxygen concentration was at a predetermined concentration using the R / B value, but the method for determining the oxygen concentration is not limited to this. For example, it may be determined whether the oxygen concentration is at a predetermined concentration by determining whether the R value is above a predetermined threshold. Here, the threshold for the R value can be appropriately set, for example, within the range of 20 to 150. Alternatively, it may be determined whether the oxygen concentration is at a predetermined concentration by determining whether the B value is below a predetermined threshold. Here, the threshold for the B value can be appropriately set, for example, within the range of 0 to 150.

[0047] The technologies disclosed herein have been described in detail above. Unless otherwise specified, the embodiments and other details mentioned herein do not limit the present invention. Furthermore, the technologies disclosed herein can be modified in various ways, and each component and each process mentioned herein may be omitted or combined as appropriate, unless no particular problems arise. This specification also includes the disclosures described in the following sections.

[0048] Section 1: An oxygen concentration monitoring method for monitoring the oxygen concentration in the welding atmosphere when welding an object to be welded, A step of measuring the color of the welding light generated when welding the aforementioned object to be welded, A step of determining whether the oxygen concentration in the welding atmosphere is below a predetermined concentration based on the measured color of the welding light. A method for monitoring oxygen concentration, including the following.

[0049] Section 2: The step of measuring the color of the welding light is, The steps include capturing an image including the welding light, A step of selecting pixels from the captured image that show the welding light based on predetermined conditions, The steps include measuring the RGB values ​​of the selected pixels and The oxygen concentration monitoring method described in item 1, including the method described in item 1.

[0050] Section 3: In the step of selecting the aforementioned pixels, Based on the predetermined conditions mentioned above, two or more pixels are selected. In the step of measuring the RGB values, The oxygen concentration monitoring method according to item 2, wherein a value is calculated by averaging the RGB values ​​in each selected pixel.

[0051] Section 4: In the step of measuring the color of the welding light, The RGB values ​​of the welding light are measured, From the measured RGB values ​​of the welding light, the R / B value, which is the ratio of the R value to the B value, is calculated. In the step of determining whether the oxygen concentration is below the predetermined concentration, An oxygen concentration monitoring method according to any one of items 1 to 3, wherein the calculated R / B value is greater than or equal to a predetermined threshold, thereby determining whether the oxygen concentration in the welding atmosphere is less than or equal to the predetermined concentration.

[0052] Section 5: The step of measuring the color of the welding light is, The steps include capturing an image including the welding light, The steps include measuring the distribution of RGB values ​​in the captured image and Includes, In the step of determining whether or not the concentration is below the predetermined level, The oxygen concentration monitoring method according to item 1, wherein a machine learning model, which has been pre-trained using the distribution of RGB values ​​of the aforementioned image and the oxygen concentration in the welding atmosphere as training data, is used to determine whether or not the oxygen concentration in the welding atmosphere is below a predetermined concentration based on the measured distribution of RGB values ​​of the aforementioned image.

[0053] Item 6: When welding the aforementioned object to be welded, an inert gas is blown toward the object to be welded. A method for monitoring oxygen concentration according to any one of items 1 to 5, comprising the step of increasing the flow rate of an inert gas if it is determined that the oxygen concentration in the welding atmosphere is greater than the predetermined concentration.

[0054] Section 7: An oxygen concentration monitoring device that monitors the oxygen concentration in the welding atmosphere when welding an object to be welded, An imaging device for capturing welding light generated when the aforementioned object to be welded is being welded, An image processing device that determines whether the oxygen concentration in the welding atmosphere is below a predetermined concentration by measuring the color of the welding light from the image captured by the imaging device. An oxygen concentration monitoring device equipped with the following features.

[0055] Section 8: The image processing device A process of acquiring the image captured by the imaging device from the imaging device, A process of selecting pixels from the captured image in which the welding light is visible, A process for measuring the RGB values ​​of the selected pixels, A process to determine whether the oxygen concentration in the welding atmosphere is below a predetermined concentration based on the measured RGB values. The oxygen concentration monitoring device described in item 7, configured to perform the following:

[0056] Section 9: In the process of selecting the aforementioned pixels, Based on the predetermined conditions mentioned above, two or more pixels are selected. In the step of measuring the RGB values, The oxygen concentration monitoring device according to item 8, wherein a value is calculated by averaging the RGB values ​​in each selected pixel.

[0057] Section 10: The image processing device is configured to further perform a process to calculate the R / B value, which is the ratio of the R value to the B value, from the measured RGB values. In the process of determining whether the oxygen concentration is below the predetermined concentration, An oxygen concentration monitoring device according to item 8 or 9, which determines whether the oxygen concentration in the welding atmosphere is less than or equal to a predetermined concentration by determining whether the calculated R / B value is greater than or equal to a predetermined threshold.

[0058] Section 11: The image processing device stores a machine learning model that has been pre-trained using the distribution of RGB values ​​in the image and the oxygen concentration in the welding atmosphere as training data. The aforementioned image processing device is A process of acquiring an image captured by the aforementioned imaging device from the aforementioned imaging device, A process for measuring the distribution of RGB values ​​in the captured image. The process involves inputting the measured distribution of RGB values ​​into the machine learning model to determine whether or not the oxygen concentration in the welding atmosphere is below a predetermined concentration. The oxygen concentration monitoring device described in item 7, configured to perform the following:

[0059] Section 12: A welding device for welding objects to be welded, An oxygen concentration monitoring device as described in any one of paragraphs 7 through 11 and A welding system equipped with [specific features / features].

[0060] Section 13: A spray mechanism that blows an inert gas toward the object to be welded, A control device that controls the flow rate of inert gas sprayed onto the workpiece by the injection mechanism based on the determination result of the oxygen concentration in the welding atmosphere by the front-end image processing device, and The welding system described in item 12, comprising: [Explanation of symbols]

[0061] 1. Welding System 5. Object to be welded 10 Welding equipment 20. Oxygen concentration monitoring device 25 Imaging device 30 Image Processing Device 38 Machine Learning Models 50 Injection mechanism 60 Control device

Claims

1. An oxygen concentration monitoring method for monitoring the oxygen concentration in the welding atmosphere when welding an object to be welded, A step of measuring the color of the welding light generated when welding the aforementioned object to be welded, A step of determining whether the oxygen concentration in the welding atmosphere is below a predetermined concentration based on the measured color of the welding light. A method for monitoring oxygen concentration, including the following.

2. The step of measuring the color of the welding light is, The steps include capturing an image including the welding light, A step of selecting pixels from the captured image that show the welding light based on predetermined conditions, The steps include measuring the RGB values ​​of the selected pixels and The oxygen concentration monitoring method according to claim 1, including the method described in claim 1.

3. In the step of selecting the aforementioned pixels, Based on the predetermined conditions mentioned above, two or more pixels are selected. In the step of measuring the RGB values, The oxygen concentration monitoring method according to claim 2, wherein a value is calculated by averaging the RGB values ​​in each of the selected pixels.

4. In the step of measuring the color of the welding light, The RGB values ​​of the welding light are measured, From the measured RGB values ​​of the welding light, the R / B value, which is the ratio of the R value to the B value, is calculated. In the step of determining whether the oxygen concentration is below the predetermined concentration, The oxygen concentration monitoring method according to claim 1, wherein the calculated R / B value is greater than or equal to a predetermined threshold, thereby determining whether the oxygen concentration in the welding atmosphere is less than or equal to a predetermined concentration.

5. The step of measuring the color of the welding light is, The steps include capturing an image including the welding light, The steps include measuring the distribution of RGB values ​​in the captured image and Includes, In the step of determining whether or not the concentration is below the predetermined level, The oxygen concentration monitoring method according to claim 1, wherein a machine learning model, which has been pre-trained using the distribution of RGB values ​​in the aforementioned image and the oxygen concentration in the welding atmosphere as training data, is used to determine from the measured distribution of RGB values ​​in the aforementioned image whether or not the oxygen concentration in the welding atmosphere is below a predetermined concentration.

6. When welding the aforementioned object to be welded, an inert gas is blown toward the object to be welded. The oxygen concentration monitoring method according to claim 1, further comprising the step of increasing the flow rate of an inert gas when it is determined that the oxygen concentration in the welding atmosphere is greater than the predetermined concentration.

7. An oxygen concentration monitoring device that monitors the oxygen concentration in the welding atmosphere when welding an object to be welded, An imaging device for capturing welding light generated when the aforementioned object to be welded is being welded, An image processing device that determines whether the oxygen concentration in the welding atmosphere is below a predetermined concentration by measuring the color of the welding light from the image captured by the imaging device. An oxygen concentration monitoring device equipped with the following features.

8. The image processing device A process of acquiring the image captured by the imaging device from the imaging device, A process of selecting pixels from the captured image in which the welding light is visible, A process for measuring the RGB values ​​of the selected pixels, A process to determine whether the oxygen concentration in the welding atmosphere is below a predetermined concentration based on the measured RGB values. The oxygen concentration monitoring device according to claim 7, configured to perform the following:

9. In the process of selecting the aforementioned pixels, Based on the predetermined conditions mentioned above, two or more pixels are selected. In the step of measuring the RGB values, The oxygen concentration monitoring device according to claim 8, wherein an average value of the RGB values ​​in each selected pixel is calculated.

10. The image processing device is configured to further perform a process to calculate the R / B value, which is the ratio of the R value to the B value, from the measured RGB values. In the process of determining whether the oxygen concentration is below the predetermined concentration, The oxygen concentration monitoring device according to claim 8, which determines whether the oxygen concentration in the welding atmosphere is less than or equal to a predetermined concentration by determining whether the calculated R / B value is greater than or equal to a predetermined threshold.

11. The image processing device stores a machine learning model that has been pre-trained using the distribution of RGB values ​​in the image and the oxygen concentration in the welding atmosphere as training data. The aforementioned image processing device is A process of acquiring an image captured by the aforementioned imaging device from the aforementioned imaging device, A process for measuring the distribution of RGB values ​​in the captured image, The process involves inputting the measured RGB value distribution into the machine learning model to determine whether the oxygen concentration in the welding atmosphere is below a predetermined concentration. The oxygen concentration monitoring device according to claim 7, configured to perform the following:

12. A welding device for welding objects to be welded, An oxygen concentration monitoring device according to any one of claims 7 to 11 and A welding system equipped with [specific features / features].

13. A spray mechanism that blows an inert gas toward the object to be welded, A control device that controls the flow rate of inert gas sprayed onto the workpiece by the injection mechanism based on the determination result of the oxygen concentration in the welding atmosphere by the front-end image processing device, and The welding system according to claim 12, comprising:

Images