Method and model for predicting shape of welded portion

Abstract

To provide a method and a model for predicting a shape of a welded portion, which can select arc-welding conditions efficiently.SOLUTION: A method for predicting a shape of a welded portion in an object to be arc-welded (a welded object) 4, includes a step in which an arc pressure distribution calculated from brightness of reflection light of arc light on a surface of the welded object 4 is used.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a method and model for predicting the shape of a weld in arc welding. [Background technology]

[0002] The quality of arc welding is evaluated by the weld shape obtained as a result of welding. The weld shape is specified by bead width, reinforcement height, penetration depth, etc. Improvements in welding power source control technology have improved welding stability. However, the welding conditions required to obtain the optimal weld shape still rely on the experience of skilled workers. Even when skilled workers select welding conditions, they may need to conduct repeated test welding when welding with unfamiliar welding materials or new materials. This requires additional work space for the test welding, as well as welding workpieces, welding materials, and time. This increases welding costs. Furthermore, when beginners with less experience select welding conditions, the number of repeated welding tests required to obtain the optimal welding conditions is greater than when skilled workers select welding conditions. This further increases welding costs.

[0003] In light of the above-described background, methods for predicting the shape of a weld have been developed to reduce the number of test welding operations, as described in Patent Document 1 or Patent Document 2. The following describes conventional techniques for predicting the shape of a weld.

[0004] The method described in Patent Document 1 precisely calculates the arc heat input using a heat source model equipped with the function of simulating the current and voltage output by the welding power source, and then calculates the shape of the molten pool based on this arc heat input using heat conduction calculations to estimate the weld bead shape.

[0005] In the method described in Patent Document 2, in order to estimate the penetration depth by hybrid welding using laser and arc, a database is prepared in advance in which the total heat input and penetration depth are associated for each welding condition, and the penetration depth is estimated by using a correlation equation between the welding conditions and the penetration depth calculated from the database.

[0006] Another possible method is to set the arc heat input, arc heat input distribution, peak arc pressure as a physical effect of the arc, and arc pressure distribution by combining virtual parameters, and then calculate the molten pool shape based on this arc heat input and arc pressure using heat conduction calculations to estimate the weld shape. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-200691 [Patent Document 2] Patent Publication No. 2021-79430 Summary of the Invention [Problem to be solved by the invention]

[0008] In the methods described in Patent Documents 1 and 2, the weld shape is predicted based on the heat input. To improve the accuracy of the weld shape prediction, it is necessary to more precisely calculate not only the arc heat input but also the range of influence of the arc's physical action on the molten pool, such as the arc pressure distribution, and set this as a boundary condition for the simulation. However, calculating the arc pressure distribution is not easy, and a great deal of time is spent on this calculation. As a result, a great deal of time is spent on selecting arc welding conditions.

[0009] In view of the above, an object of the present disclosure is to provide a method and a model for predicting the shape of a weld that enable efficient selection of arc welding conditions. [Means for solving the problem]

[0010] In order to achieve the above object, the method and model for predicting the shape of a weld according to the present disclosure are as follows.

[0011] (1) A method for predicting the shape of a weld in an arc welding workpiece, comprising: A method for predicting a shape of a weld, comprising: applying an arc pressure distribution calculated from the brightness of the arc light reflected on the surface of the workpiece to be welded.

[0012] (2) The method for predicting a shape of a welded portion according to (1) above, further comprising the step of binarizing an image of a surface of the workpiece to be welded and calculating the arc pressure distribution.

[0013] (3) A model for predicting the shape of a weld in an arc welding workpiece, A prediction model for a weld shape configured to input an arc pressure distribution calculated from the brightness of the arc light reflected on the surface of the workpiece. [Effects of the Invention]

[0014] According to the present disclosure, a method and a model for predicting the shape of a weld are provided that enable efficient selection of arc welding conditions. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 1 is a schematic diagram illustrating an outline of arc welding. [Figure 2] FIG. 1 is a block diagram illustrating a configuration example of a prediction system according to the present disclosure. [Figure 3] 10 is a flowchart illustrating an example of a procedure for a method for predicting a shape of a weld. [Figure 4] 1 is an example of an image of a surface of an object to be welded. [Figure 5] This is a binarized image of the image in FIG. DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, embodiments of a weld shape prediction method and prediction model according to the present disclosure will be described with reference to the drawings. Each drawing is a schematic diagram and may differ from the actual product. Furthermore, the following embodiments exemplify an apparatus or method for embodying the technical idea of ​​the present disclosure, and are not intended to limit the configuration to that described below. In other words, the technical idea of ​​the present disclosure can be modified in various ways within the technical scope described in the claims.

[0017] (Outline of the method for predicting weld shape) By executing the weld shape prediction method according to the present disclosure, the shape of a weld obtained by arc welding can be predicted. Arc welding, as shown in FIG. 1 , is a method in which an arc is generated between workpieces 4 and a welding torch 3, and the workpieces 4 are melted and joined by the heat generated by the arc. The portion of the workpieces 4 joined by melting them is also referred to as a weld 5. The weld 5 includes a portion of the workpieces 4 that has solidified after melting. The weld 5 also includes a heat-affected zone as defined in JIS Z 3001. The heat-affected zone is a portion that has not melted due to the heat of welding but has changed in structure, metallurgical properties, mechanical properties, etc. The prediction method according to the present disclosure may predict the shape of a weld 5 that includes at least a portion of the heat-affected zone, or may predict the shape of a weld 5 that does not include the heat-affected zone.

[0018] The shape of the weld 5, i.e., the weld shape, is specified by the bead width, the reinforcement height, the penetration depth, etc. The weld shape can be controlled by the arc welding conditions. Conversely, the arc welding conditions must be selected to obtain a desired weld shape. The weld shape may include the cross-sectional shape of the weld 5. The weld shape may also include other shapes of the weld 5, such as the surface shape or the toe shape of the weld 5.

[0019] In order to select arc welding conditions, the shape of the weld may be estimated by simulating arc welding. The arc welding simulation includes calculation of heat conduction in arc welding. Specifically, the arc welding conditions used in the simulation are adjusted so that the difference between the desired weld shape and the weld shape estimated by the arc welding simulation is small. The parameters of the arc welding conditions when the difference between the desired weld shape and the estimated weld shape is sufficiently small are selected as appropriate arc welding conditions.

[0020] In an arc welding simulation, the shape of a weld is estimated by setting assumed values ​​for virtual parameters representing the arc heat input or arc pressure corresponding to the arc welding conditions and running the simulation, and each parameter is adjusted by repeatedly setting and estimating the assumed value so that the estimated weld shape approaches the actual weld shape. It is desired to reduce the number of attempts to adjust the parameters of the arc welding conditions set in the simulation so that the arc welding conditions can be selected efficiently.

[0021] A weld shape prediction model may be used for the arc welding simulation. The weld shape prediction model is configured to output a prediction result of the weld shape obtained by arc welding applying the arc welding conditions when the arc welding conditions are input. The arc welding conditions for obtaining the desired weld shape are determined by repeating a simulation using the prediction model while changing the arc welding conditions so that the prediction result of the weld shape approaches the desired weld shape.

[0022] The prediction model may be, for example, a model that applies the finite element method (FEM). When the prediction model is a model that applies the finite element method, parameters determined according to the arc welding conditions are set as boundary conditions of the finite element method. The prediction model may also be a model that applies various other methods, such as the finite difference method (FDM).

[0023] The boundary conditions determined according to the arc welding conditions include the arc pressure distribution, which is the distribution of arc pressure that physically acts on each part of the surface of the work-piece 4. Calculating the arc pressure distribution from the arc welding conditions is not easy, although it can be realized using, for example, a different model.

[0024] During arc welding, a weld pool is formed on the surface of the workpiece 4. The arc pressure acting on the weld pool surface deforms the weld pool surface. The deformation of the weld pool surface affects the weld shape. The arc light is reflected in different directions between deformed and non-deformed areas of the weld pool surface. Arc light is light generated by the arc. As a result, a brightness distribution appears on the surface of the weld workpiece 4 when viewed from one direction. Given that the deformation of the weld pool surface is caused by the action of arc pressure, the brightness of the surface of the weld workpiece 4 reflects the arc pressure distribution. Therefore, the arc pressure distribution can be easily calculated from the surface brightness of the weld workpiece 4. When the arc pressure distribution is set as the boundary condition of the weld shape prediction model, the weld shape can be predicted by calculating the energy conservation equation, which takes into account the physical effects on the weld pool.

[0025] Therefore, in the method for predicting a weld shape according to the present disclosure, a measured value of the arc pressure distribution that physically acts on the surface of the work-piece 4 is set as a boundary condition of the prediction model for the weld shape. The measured value of the arc pressure distribution may be calculated based on the brightness of an image captured of the arc light reflected on the surface of the work-piece 4 during arc welding, and may be stored in association with the arc welding conditions.

[0026] The arc pressure distribution set as the boundary condition is calculated from the surface brightness, so that the arc pressure distribution can be easily calculated. The easy calculation of the arc pressure distribution reduces the time required to obtain a predicted result of the weld shape using a prediction model of the weld shape. The reduction in the time required to obtain a predicted result of the weld shape reduces the time required to select arc welding conditions. In other words, the method for predicting a weld shape according to the present disclosure allows for efficient selection of arc welding conditions.

[0027] (Configuration example of weld shape prediction system 1) As shown in FIG. 2, a prediction system 1 according to an embodiment of the present disclosure includes a prediction device 10, a welding device 20, and a measurement device 30.

[0028] As shown in FIG. 1 , welding device 20 includes welding torch 3, and performs arc welding by generating an arc between welding torch 3 and workpiece 4. Welding device 20 is configured to be able to control at least one of the arc current and arc voltage during welding. Welding device 20 may be configured to be able to move at least one of workpiece 4 or welding torch 3. Welding device 20 may be configured so that a stage on which workpiece 4 is mounted can move relative to welding torch 3, or so that welding torch 3 can move relative to the stage on which workpiece 4 is mounted.

[0029] The measuring device 30 includes a camera that captures the arc light reflected on the surface of the work-piece 4 during arc welding by the welding device 20.

[0030] Prediction device 10 inputs welding conditions and an arc pressure distribution calculated from the brightness of an image of the surface of workpiece 4 into prediction model 12. Prediction device 10 acquires the predicted results of the weld shape output from prediction model 12 and outputs them to the outside.

[0031] The prediction device 10 and the welding device 20 or the measuring device 30 are connected to each other so that they can communicate with each other. The prediction device 10, the welding device 20, or the measuring device 30 may have a communication interface based on a wired or wireless communication standard. For example, the wireless communication standard may include a cellular phone communication standard such as 3G, 4G, or 5G. Furthermore, for example, the wireless communication standard may include IEEE 802.11 or Bluetooth (registered trademark). The communication interface may support one or more of these communication standards. The communication interface is not limited to these examples and may communicate with other devices or input and output data based on various standards. The prediction device 10, the welding device 20, or the measuring device 30 may be connected via a network or may be directly connected, for example, peer-to-peer (P2P).

[0032] The prediction device 10 may be communicatively connected to an external device so as to transmit the predicted weld shape result to the external device. The prediction device 10 may include an output device that outputs the predicted weld shape result. The prediction device 10 may include a display device for displaying the predicted weld shape result as the output device. The display device may include, for example, an LCD (Liquid Crystal Display), an organic EL (Electro-Luminescence) display or an inorganic EL display, or a PDP (Plasma Display Panel). The display device is not limited to these displays and may include various other types of displays.

[0033] The prediction device 10 may be configured to include at least one processor, such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). The prediction device 10 may be configured with one processor or multiple processors. The processor constituting the prediction device 10 may realize the functions of the prediction device 10 by reading and executing a program stored in a storage unit described later.

[0034] The prediction device 10 may include a storage unit. The storage unit stores various types of information or data. The storage unit may store, for example, a program executed in the prediction device 10, or data or processing results used in processing executed in the prediction device 10. The storage unit may also function as a work memory for the prediction device 10. The storage unit may be configured to include, for example, a semiconductor memory, but is not limited to this. For example, the storage unit may be configured as an internal memory of a processor used as the prediction device 10, or as a hard disk drive (HDD) accessible from the prediction device 10. The storage unit may be configured as a non-transitory readable medium. The storage unit may be configured integrally with the prediction device 10, or may be configured separately from the prediction device 10.

[0035] The prediction device 10 may generate the prediction model 12, or may acquire the prediction model 12 from an external device. The prediction system 1 may not include the welding device 20. If the prediction system 1 does not include the welding device 20, the prediction device 10 may acquire the welding conditions from an external device that performs the welding. The prediction system 1 may not include the measuring device 30. If the prediction system 1 does not include the measuring device 30, the prediction device 10 may acquire an image of the surface of the workpiece 4 from an external imaging device.

[0036] (About the weld shape prediction model 12) Hereinafter, the prediction model 12 used in the method for predicting the shape of a weld according to the present disclosure calculates the following equation (1) using the arc pressure distribution as a boundary condition: ρ represents density, H represents enthalpy, t represents time, u represents velocity, λ represents thermal conductivity, T represents temperature, and q Arc represents the arc heat input flux, and n→ represents the normal vector. Note that symbols with → above them to represent vectors in mathematical formulas are substituted with, for example, u→ in the description of the specification.

[0037]

number

[0038] The velocity u→ used in equation (1) is calculated by solving the following equations for conservation of mass (2) and momentum (3). Among the symbols used in equations (2) and (3), P represents pressure, μ represents the viscosity coefficient, and F→ EX represents external forces including gravity, surface tension, or arc pressure.

[0039]

number

[0040]

number

[0041] The formulas used in the above explanation are merely examples and may be modified as appropriate.

[0042] On the other hand, in the prediction model 12 used in the method for predicting the shape of a weld according to the present disclosure, the arc pressure distribution is calculated by F → EX The arc pressure distribution reflected in the equation of the prediction model 12 can be easily calculated from the surface brightness, thereby reducing the time required to predict the weld shape.

[0043] (Example of procedure for predicting weld shape) An example of the procedure of the weld shape prediction method according to the present disclosure will be described below with reference to the flowchart shown in Fig. 3. The weld shape prediction method may be executed by prediction device 10. The weld shape prediction method may be realized as a weld shape prediction program executed by a processor included in prediction device 10. The weld shape prediction program may be stored in a non-transitory computer-readable medium.

[0044] The prediction device 10 acquires welding conditions from the welding device 20 when the welding device 20 performs arc welding (step S1). The welding conditions may include an arc current or an arc voltage. The welding conditions may include a speed at which the workpiece 4 moves relative to the welding torch 3 of the welding device 20 during arc welding.

[0045] Prediction device 10 acquires the arc pressure distribution of welded portion 5 of workpiece 4 during arc welding from measurement device 30 (step S2). The arc pressure distribution is calculated from the brightness of an image obtained by measuring device 30 by capturing the reflected light of the arc light on the surface of welded portion 5, as shown in FIG. 4. The image in FIG. 4 includes welding torch 3, welded portion 5, and a deformed portion 7 within welded portion 5. Welded portion 5 is a portion where arc welding has been performed or a portion where arc welding is currently being performed. Welding torch 3 of welding device 20 is located near deformed portion 7. Deformed portion 7 is a portion that has been deformed due to the physical action of the arc pressure from welding torch 3 on the surface of welded portion 5. Within welded portion 5, the brightness of deformed portion 7 is higher than the brightness of portions not deformed by the arc pressure, i.e., portions other than deformed portion 7.

[0046] Specifically, as illustrated in FIG. 5, the image of FIG. 4 may be binarized based on brightness. The brightness threshold used when binarizing the image of FIG. 4 may be set appropriately. For example, if the brightness of the image is 256 grayscale levels, the brightness threshold may be set to 128. Pixels in the image of FIG. 4 that have a brightness equal to or greater than the brightness threshold are replaced with white in the image of FIG. 5. Pixels in the image of FIG. 4 that have a brightness less than the brightness threshold are replaced with black in the image of FIG. 5. The range represented by white pixels in the image of FIG. 5 corresponds to the range affected by the arc pressure. The arc pressure distribution is identified as the range in which the surface of the workpiece 4 is affected by the arc pressure.

[0047] The relationship between the brightness of the deformed portion 7 and the brightness of the non-deformed portion can change depending on the position of the measurement device 30. If the measurement device 30 is positioned so that the brightness of the deformed portion 7 is lower than the brightness of the non-deformed portion, the area affected by the arc pressure corresponds to the area represented by black pixels in the binarized image.

[0048] The radius of the arc pressure distribution may be calculated as half the width (W) of the range affected by the arc pressure. Therefore, the radius of the arc pressure distribution is calculated as the width (W), which is the length of deformed portion 7 in the X-axis direction in FIG. 5. When measuring device 30 is located in the X-axis direction relative to welding torch 3, the radius of the arc pressure distribution is calculated as the length of deformed portion 7 in the Y-axis direction. Note that the image in FIG. 4 was taken when measuring device 30 was located in the negative Y-axis direction relative to welding torch 3.

[0049] As described above, the measuring device 30 calculates the arc pressure distribution based on an image of the surface of the work-piece 4. The measuring device 30 may calculate the arc pressure distribution by binarizing the image of the surface of the work-piece 4. The prediction device 10 may acquire the arc pressure distribution calculated by the measuring device 30. The prediction device 10 may acquire the image from the measuring device 30 and calculate the arc pressure distribution by analyzing the image.

[0050] Returning to Fig. 3, prediction device 10 inputs the arc pressure distribution of welded portion 5 and the welding conditions for arc welding into prediction model 12 (step S3). Prediction device 10 outputs the shape of welded portion 5, i.e., the predicted result of the welded portion shape, from prediction model 12 (step S4). After executing the procedure of step S4, prediction device 10 ends execution of the flowchart in Fig. 3.

[0051] The predicted weld shape is compared with the weld shape expected when arc welding is performed on the workpiece 4. If the predicted weld shape is sufficiently close to the expected weld shape, the welding conditions used when the predicted weld shape was obtained may be used. If there is a large difference between the expected weld shape and the predicted weld shape, the welding conditions are changed and the weld shape is predicted again. The change in welding conditions and the prediction of the weld shape are repeated until the predicted weld shape is sufficiently close to the expected weld shape.

[0052] (Small summary) As described above, the prediction device 10 according to the present disclosure predicts the shape of a weld using the arc pressure distribution that physically acts on the surface of the work-piece 4. The arc pressure distribution can be easily calculated using the brightness of an image of the surface of the work-piece 4. As a comparative example, when the arc pressure distribution is expressed as a Gaussian distribution and set as the boundary condition of the prediction model 12 to predict the shape of a weld, it takes a lot of time to calculate an arc pressure distribution that matches the actual situation, and it also takes a lot of time to predict the shape of the weld. On the other hand, the prediction device 10 according to the present disclosure can easily calculate the arc pressure distribution. As a result, the shape of the weld is predicted efficiently.

[0053] (Example) Specific examples are described below. The weld shape prediction model 12 was configured so that the size of the workpiece 4 to be welded was 25 mm thick, 150 mm wide, and 200 mm long as geometry. The prediction model 12 was also configured to be able to calculate the arc welding phenomenon in which a wire with a diameter of 1.2 mm is dropped. The temperature-dependent physical properties of the workpiece 4 to be welded, equivalent to those of SM490 steel, were used. The welding method was bead-on-plate.

[0054] In this embodiment, the arc pressure distribution to be set as a boundary condition in the prediction model 12 was calculated by binarizing an image of the weld 5 taken by the camera of the measuring device 30 during arc welding of the workpiece 4.

[0055] In this example, the shape of the weld was predicted by setting an assumed value for the arc heat input distribution, which is one of the boundary conditions of the prediction model 12. The arc heat input distribution was then adjusted by repeatedly assuming the arc heat input distribution and predicting the shape of the weld based on the assumed boundary conditions. On the other hand, as a comparative example, the shape of the weld was predicted by assuming not only the arc heat input distribution but also the arc pressure distribution as boundary conditions of the prediction model 12. The shape of the weld was then adjusted by repeatedly assuming both the arc heat input distribution and the arc pressure distribution and predicting the shape of the weld based on the assumed boundary conditions.

[0056] In both the present example and the comparative example, the convergence condition for the fitting was set to be that the difference in width and depth between the predicted weld shape and the actual weld shape be within 10%. The number of times the fitting process converged was compared between the present example and the comparative example. To reduce the influence of personal factors when fitting the arc heat input distribution, the peak value was appropriately set based on a Gaussian distribution approximating the arc heat input distribution or a distribution similar to literature values. Then, if the calculated value was smaller than the experimental value, the peak value of the arc heat input set as the boundary condition was increased by 10%. Conversely, if the calculated value was larger than the experimental value, the peak value of the arc heat input set as the boundary condition was decreased by 10%. After reaching an intermediate state, the peak value of the arc heat input was adjusted in 1% increments.

[0057] Table 1 shows the number of attempts required until convergence when fitting was performed in each of the present embodiment and the comparative example using the method described above. In Table 1, Nos. 1 to 5 show the results of the present embodiment when measured values ​​calculated from brightness were applied to the arc pressure distribution. Nos. 6 to 10 show the results of the comparative example when a hypothetical Gaussian distribution was applied to the arc pressure distribution. The current value, voltage value, and welding speed were set as welding conditions. The welding conditions were set to the same values ​​for Nos. 1 to 5 of the present embodiment and Nos. 6 to 10 of the comparative example. As described above, the measured values ​​were applied to the arc pressure distribution in the present embodiment, and the Gaussian distribution was applied in the comparative example. In addition, if the number of attempts required until convergence was 9 or less, it was evaluated as good, and if it was 10 or more, it was evaluated as bad.

[0058] [Table 1]

[0059] According to Table 1, under all of the conditions Nos. 1 to 5 of this example, the number of matching trials was 9 or less, and was judged as good. On the other hand, under all of the conditions Nos. 6 to 10 of the comparative examples, the number of matching trials was 10 or more, and was judged as bad. According to this result, by using the measured value calculated from the brightness as the arc pressure distribution in this example, the number of matching trials can be reduced. As a result, the time required to select welding conditions is shortened. In other words, the welding conditions can be selected efficiently.

[0060] Although the embodiments of the present disclosure have been described based on the drawings and examples, it should be noted that those skilled in the art could make various modifications or alterations based on the present disclosure. Therefore, it should be noted that these modifications and alterations are included within the scope of the present disclosure. For example, the functions included in each component or step can be rearranged so as not to cause logical inconsistencies, and multiple components or steps can be combined or divided into one. The embodiments of the present disclosure can also be realized as a program executed by a processor included in an apparatus or a storage medium on which a program is recorded. It should be understood that these are also included within the scope of the present disclosure. [Explanation of symbols]

[0061] 1. Prediction System 3 welding torches 4. Welding target 5 Welded parts 7 Deformation section 10 Prediction Device (12: Prediction Model) 20 Welding equipment 30 Measuring Equipment

Claims

1. A method for predicting a shape of a weld in a workpiece for arc welding, comprising: A method for predicting a shape of a weld, comprising: applying an arc pressure distribution calculated from the brightness of the arc light reflected on the surface of the workpiece to be welded.

2. The method for predicting a shape of a welded portion according to claim 1 , further comprising the step of binarizing an image of a surface of the workpiece to be welded and calculating the arc pressure distribution.

3. A model for predicting a shape of a weld in a workpiece for arc welding, A prediction model for a weld shape configured to input an arc pressure distribution calculated from the brightness of the arc light reflected on the surface of the workpiece.

Images