Virtual welding system, method for processing virtual welding system, and program

Abstract

To display a realistic welding state in a virtual welding using a virtual welding system.SOLUTION: The virtual welding system includes: prediction means for predicting a welding state in a virtual welding by using a prediction model for predicting the future welding state; simulation means for estimating the current welding state in the virtual welding by a physics calculation with reference to a predicted welding state predicted by the prediction means; and display means for displaying the welding state in the virtual welding in a screen on the basis of the result of estimation by the simulation means.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a virtual welding system, a processing method for a virtual welding system, and a program. [Background technology]

[0002] In training to learn welding techniques, actual equipment has traditionally been used, and in recent years, training using VR (Virtual Reality) and AR (Augmented Reality) has also been on the rise. Patent Document 1 discloses a technology in which a user undergoing training wears a helmet equipped with a small display device and manipulates a simulated welding object (welding coupon) and mock welding tool, and the display device displays a simulated welding process in a virtual reality space. Using such a mechanism makes it possible to train in a safe environment to improve welding techniques. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-223966 A Summary of the Invention [Problem to be solved by the invention]

[0004] In welding work, it is important to understand the state of the molten pool formed during welding and proceed with the work. In conventional welding technique training using VR, the virtually formed molten pool and weld marks are displayed based on physically calculated information, but there is a problem that they lack reality. The purpose of this invention is to make it possible to display a realistic welding state in virtual welding using a virtual welding system. [Means for solving the problem]

[0005] The virtual welding system according to the present invention is characterized in having a prediction means for predicting a welding state in virtual welding using a prediction model for predicting a future welding state, a simulation means for estimating a current welding state in virtual welding by physical calculation with reference to the welding state predicted by the prediction means, and a display means for displaying an image of the welding state in virtual welding based on a result of estimation by the simulation means. Effect of the Invention

[0006] According to the present invention, it is possible to display a realistic welding state in virtual welding using a virtual welding system. [Brief description of the drawings]

[0007] [Figure 1] FIG. 2 is a diagram illustrating an example of a hardware configuration of the virtual welding system according to the present embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of a functional configuration of the virtual welding system according to the present embodiment. [Diagram 3] FIG. 2 is a diagram illustrating virtual welding in the present embodiment. [Figure 4] 5A to 5C are diagrams illustrating generation of a prediction model of a welding state in the present embodiment. [Diagram 5] 5 is a flowchart illustrating a welding state display process in the present embodiment. [Figure 6] 10 is a flowchart illustrating a flow of a simulation of a welding state. [Figure 7] FIG. 13 is a diagram for explaining division of a target region into meshes. [Figure 8] FIG. 13 is a diagram illustrating stacking of beads. [Figure 9] FIG. 13 is a diagram for explaining heat input calculation. [Figure 10] FIG. 13 is a diagram for explaining prediction of a welding state. [Figure 11] FIG. 13 is a diagram for explaining heat input calculation using a cone heat input model. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the following description, "on-site welding" refers to actual welding performed in an on-site environment, and "virtual welding" refers to virtual welding performed in a simulated virtual space (virtual reality environment).

[0009] 1 is a diagram showing an example of a hardware configuration of a virtual welding system 100 according to one embodiment of the present invention. The virtual welding system 100 according to this embodiment is capable of performing virtual welding in a simulated virtual space (virtual reality environment), and can be used, for example, as a training support system for conducting training in welding techniques based on the movement of a torch and the posture of a worker during welding work.

[0010] The virtual welding system 100 includes a CPU 101, a main memory device 102, an auxiliary memory device 103, an output device 104, an HMD 105, a controller 106, a motion sensor 107, a camera 108, and a torch motion detection device 109. The CPU 101, the main memory device 102, the auxiliary memory device 103, the output device 104, the HMD 105, the controller 106, the motion sensor 107, the camera 108, and the torch motion detection device 109 are connected to each other so as to be able to communicate with each other.

[0011] The CPU (Central Processing Unit) 101 is a central processing unit that controls the virtual welding system 100. The main memory 102 is a memory device that functions as a work area for the CPU 101 and a temporary storage location for data. The main memory 102 is implemented using, for example, a random access memory (RAM) or the like. The auxiliary memory 103 is a memory device that stores various data and various programs. The auxiliary memory 103 is implemented using, for example, a read only memory (ROM), a hard disk drive (HDD), a solid state drive (SSD), or the like. The CPU 101 executes processing based on a program stored in, for example, the auxiliary memory 103, thereby realizing various processing operations in the virtual welding system 100. The output device 104 is a device that outputs various information such as the state of virtual welding (virtual welding) performed in a virtual space and evaluation results of welding work, and is used to present various information to a user, etc.

[0012] The HMD (Head Mounted Display) 105 is a head-mounted display device that displays images (VR images) in a virtual space. The HMD 105 may be a non-transparent HMD that prevents the user from checking the outside when worn, or a transparent (see-through) HMD that displays the outside when necessary. The controller 106 is a controller that accompanies the HMD 105 and is used to perform operations in a virtual space displayed on the HMD 105. For example, in virtual welding (virtual welding) performed in a virtual space, a VR image showing the welding state, such as various information and an image during welding, is displayed on the HMD 105, and the virtual welding is performed using the controller 106. At this time, the controller 106 is used to detect, for example, a pseudo torch operation.

[0013] The motion sensor 107 is a sensor for three-dimensionally measuring (motion capture) the user's posture (movement). The motion sensor 107 may be any type capable of measuring the user's posture, etc., and the type of motion capture is not limited. The motion sensor 107 may be, for example, a wearable motion capture suit type also called smart apparel incorporating an IMU sensor (inertial measurement unit), or may be a wearable suit type with a marker to be detected by a tracker, for example. The motion sensor 107 may also be, for example, a photography type (camera type) that captures an image of a subject and detects the distance and position of each part of the subject.

[0014] Camera 108 is a camera that captures images of welding phenomena during actual welding (on-site welding) performed in an on-site environment. Camera 108 captures images of the welding state during actual welding (on-site welding) performed in an on-site environment, such as a molten pool and weld marks formed during welding work. Torch operation detection device 109 detects torch operation during actual welding (on-site welding) performed in an on-site environment. Torch operation detection device 109 detects the movement of the torch based on, for example, a marker attached to the torch. Note that torch operation detection device 109 only needs to detect the movement of the torch during welding, and the detection method is not limited.

[0015] The virtual welding system 100 can acquire various data during actual welding (on-site welding) performed in an on-site environment, for example, using a motion sensor 107, a camera 108, and a torch motion detection device 109. In addition, the virtual welding system 100 can display images relating to virtual welding (virtual welding) performed in a virtual space, for example, using an HMD 105, and can acquire various data during virtual welding (virtual welding) performed in a virtual space, for example, using the motion sensor 107 and a controller 106.

[0016] In addition, when data on the posture of the worker during welding is not acquired, virtual welding system 100 may not include motion sensor 107. In addition, when a prediction model of the welding state described later is performed using photographed images of the welding state in actual welding (on-site welding) performed in an on-site environment and torch operation information, etc., which have been acquired and recorded in advance, virtual welding system 100 may not include camera 108 and torch operation detection device 109.

[0017] 2 is a diagram showing an example of the functional configuration of the virtual welding system 100 in this embodiment. The virtual welding system 100 includes an acquisition unit 201, a storage unit 202, a display unit 203, an operation unit 204, an analysis unit 205, an evaluation unit 206, a learning processing unit 207, a simulation unit 208, an image generation unit 209, a display control unit 210, and a control unit 211.

[0018] The acquisition unit 201 acquires various data used for various processes in the virtual welding system 100. For example, the acquisition unit 201 acquires data related to the operation of a torch and the posture of a user (worker) in on-site welding, captured images of the welding state, etc. Also, for example, the acquisition unit 201 acquires various data related to the operation of a torch and the posture of a user (worker) in virtual welding.

[0019] The storage unit 202 stores various data acquired by the acquisition unit 201 and various models generated by the learning processing unit 207 including a prediction model of the welding state. The storage unit 202 may also store data such as processing results acquired by the analysis unit 205 and the evaluation unit 206. The display unit 203 displays various information to be presented to a user or the like under the control of the display control unit 209. For example, the display unit 203 displays various information when performing virtual welding, VR images showing the state of welding, and a result screen showing the evaluation result of the welding work by the evaluation unit 206. The operation unit 204 is used to perform operations in a virtual space. The operation unit 204 is used, for example, when performing various operations in virtual welding.

[0020] The analysis unit 205 analyzes the torch movement during welding, the user (worker) posture, welding phenomena, and the like, based on various data acquired by the acquisition unit 201. For example, the analysis unit 205 analyzes the torch movement (speed, position, and angle) during welding, based on data acquired from the torch movement detection device 109 during on-site welding and data acquired from the controller 106 during virtual welding. The analysis unit 205 also analyzes the posture (movement) of the user (worker) during welding, based on data acquired from the motion sensor 107 during on-site welding and virtual welding.

[0021] The evaluation unit 206 evaluates the welding work based on the movement of the torch during the virtual welding work, the posture of the user (worker), and the welding state. For example, the evaluation unit 206 compares the movement of the torch during the virtual welding work with a torch model generated in advance regarding the movement of the torch of an expert, and evaluates the torch movement of the user (worker) during the virtual welding work. The evaluation unit 206 also compares the posture of the user (worker) during the virtual welding work with a welding posture model generated in advance regarding the posture of an expert during the work, and evaluates the posture of the user (worker) during the virtual welding work. The evaluation unit 206 also compares the welding state simulated based on the movement of the torch during the virtual welding work, etc., with a welding state model, described later, generated in advance regarding the welding state in welding by an expert, and evaluates the welding state during the virtual welding work.

[0022] The learning processing unit 207 performs machine learning (e.g., deep learning) using as input images of the welding state captured during on-site welding and feature amount data related to the welding work, and generates a prediction model of the welding state for predicting the future welding state. In addition, for example, the learning processing unit 207 performs machine learning using as input various data obtained during on-site welding by an expert or virtual welding work, and generates models for evaluating the welding work, such as a torch model related to the movement of the torch, a welding posture model related to the posture during the welding work, and a welding state model related to the welding state (e.g., a molten pool model related to the molten pool).

[0023] The simulation unit 208 simulates the welding state in virtual welding. The simulation unit 208 simulates the welding state during welding work in virtual welding based on various data acquired in virtual welding. The simulation unit 208 combines a simulation based on a physical calculation model and a prediction model of the welding state generated by machine learning by the learning processing unit 207 to simulate the welding state in virtual welding. The simulation unit 208 is an example of a prediction means and a simulation means.

[0024] The image generation unit 209 generates images to be displayed in the virtual welding system 100. The image generation unit 209 generates, for example, a VR image to be displayed when performing virtual welding, and an image showing an evaluation result of welding work, etc. The display control unit 210 controls the display on the display unit 203. The display control unit 210 controls, for example, the display position, size, etc. of the image generated by the image generation unit 209 on the display unit 203. The control unit 211 controls each functional unit of the virtual welding system 100.

[0025] FIG. 3 is a diagram for explaining virtual welding (virtual welding) performed in a virtual space in this embodiment. FIG. 3(a) shows an example of a VR image 301 displayed on the HMD 105 when virtual welding is performed. In the VR image 301, a torch (welding rod) 302 simulating actual welding (on-site welding) performed in a field environment, a welding mark (bead) 303, and a molten pool 304 formed by a generated arc are displayed as elements related to the welding work itself. In addition, the VR image 301 can also display an evaluation 306 of the welding work (for example, a four-level display, etc.), operation information 307 of the torch (welding rod) 302, and posture information 308 during the welding work. In addition, it is also possible to output sounds during welding in virtual welding, sounds that beat rhythms, etc. together with the VR image 301. In addition, precautions (reminders) regarding the operation and posture of the torch during virtual welding may be displayed by image or output by sound.

[0026] The display (output) of each element in virtual welding can be set to be displayed (output) or not via a setting screen such as that shown in Fig. 3(b). Fig. 3(b) shows an example in which it is possible to set whether to display the arc light, spatter, fume, and molten pool as elements to be displayed in virtual welding, and whether to display (output) the ruler, metronome (rhythmic sound), molten pool frame, and torch movement angle / operation angle as guides to support the welding work, and whether to display real-time evaluation. Note that the settings shown in Fig. 3(b) are merely examples and are not limited to these.

[0027] The virtual welding system 100 in this embodiment executes a simulation based on a physical calculation model using a prediction model of a welding state for predicting a future welding state in virtual welding, and displays a VR image in virtual welding based on the simulation result. Here, the welding state includes a molten pool formed during welding, or both a molten pool and a welding mark. The virtual welding system 100 predicts a welding state in virtual welding using a prediction model of a welding state, and executes a simulation based on a physical calculation model with reference to the prediction result, thereby estimating a welding state related to a molten pool, a welding mark, etc. virtually formed in virtual welding. The virtual welding system 100 simulates, for example, a phenomenon occurring in a welding base material after an arc is generated in welding, such as a melting phenomenon of the welding base material, a molten pool, a fluid, rust, etc., and displays an image according to the simulation result.

[0028] The operation of the virtual welding system 100 related to image display in virtual welding will be described below. Note that, among the image displays in virtual welding, the following mainly describes the display of the welding state related to the molten pool, welding marks, etc. As for the image display of the pseudo torch etc. operated by the user (worker) in virtual welding, it is sufficient to display the image by a general method, and the description thereof will be omitted.

[0029] First, a prediction model of a welding state for predicting a welding state in virtual welding will be described. Fig. 4 is a diagram for explaining generation of a prediction model of a welding state. Learning processing unit 207 receives as input images 401 of a time series of welding states captured by camera 108 during on-site welding and feature amount data 402 indicating feature amounts in welding work, and performs machine learning to generate a prediction model 403 of a welding state.

[0030] The machine learning is performed, for example, by using a plurality of images (images taken at times t, (t-1), . . . , (tn-1)) 401 of the welding state up to time t (for example, n images, where n is an integer of 2 or more) and a set of feature amount data 402 at time t and the welding state image 401 as one teacher data. The learning processing unit 207 performs machine learning (for example, deep learning using a DNN: Deep Neural Network) using a plurality of teacher data with different times t, for example, to generate a prediction model 403 of the welding state. In this way, a prediction model 403 of the welding state that estimates the welding state at time (t+1) is generated based on the n images of the welding state taken up to time t and the feature amount data at time t. The generated prediction model 403 of the welding state is stored in, for example, the storage unit 202. Note that in the early stage of the welding work when there are not a predetermined number of images of the welding state, the machine learning may use images of the welding state taken from the start of the welding work.

[0031] Next, a description will be given of the display operation by the virtual welding system 100 when virtual welding is performed. Fig. 5 is a flowchart illustrating an example of a display process of the welding state in virtual welding by the virtual welding system 100. Note that data acquired in each step of the process shown in Fig. 5 and processing results such as simulation results are appropriately stored in the storage unit 202.

[0032] When an instruction is given to start virtual welding with a VR image displayed on the HMD 105, the virtual welding system 100 estimates and determines the output in virtual welding based on the setting conditions of the virtual welding, etc. The setting conditions of the virtual welding include the diameter of the wire, which is the welding material, the wire feed speed, the maximum wire length, the minimum, standard and maximum arc length, the standard voltage, the minimum, standard and maximum current, etc. The output in virtual welding includes the estimated values ​​of the wire length, the arc length, the current, the voltage, and the penetration amount.

[0033] After estimating and determining the output in virtual welding, virtual welding is started, and in step S501, virtual welding system 100 acquires various data related to the virtual welding being performed. In this step S501, acquisition unit 201 of virtual welding system 100 acquires data for acquiring information related to the pseudo torch operation (speed, position, and angle) in virtual welding, which is detected by, for example, controller 106 or the like. For example, analysis unit 205 analyzes the data acquired by acquisition unit 201 to obtain torch operation data indicating the torch operation (speed, position, and angle) in virtual welding.

[0034] Next, in step S502, the virtual welding system 100 simulates the welding state in virtual welding based on the output in virtual welding estimated and determined at the start of virtual welding and the various data acquired in step S501. In the simulation of the welding state in step S502, the simulation unit 208 of the virtual welding system 100 estimates the size, shape, etc. of the molten pool and the weld mark by a prediction model of the welding state based on the feature amount of the welding work and the past welding state, and performs a simulation of the welding state based on a physical calculation model using the estimated size, shape, etc. of the molten pool and the weld mark.

[0035] In the next step S503, the virtual welding system 100 displays the welding state in the virtual welding on the HMD 105 based on the simulation result in step S502, etc. At this time, the virtual welding system 100 displays the welding state such as the molten pool and the welding marks, as well as the arc light, spatter (sparks), fumes, etc. in the virtual welding, and outputs the welding sound, etc. In this step S503, for example, the image generating unit 209 of the virtual welding system 100 generates a VR image showing the welding state in the virtual welding including the welding state, in which the display color according to the temperature is set, based on the simulation result in step S502, etc., and the display control unit 210 causes the display unit 203 to display the VR image generated by the image generating unit 209.

[0036] Next, in step S504, the virtual welding system 100 determines whether or not to end the virtual welding. The virtual welding system 100 determines whether or not to end the virtual welding based on, for example, a user (operator) operation on the controller 106. If it is determined that the virtual welding is to be ended, the process shown in Fig. 5 is ended, and if it is determined that the virtual welding is not to be ended, i.e., that the virtual welding is to be continued, the process returns to step S501.

[0037] Fig. 6 is a flow chart for explaining the flow of the simulation of the welding state in step S502 in Fig. 5. In the simulation of the welding state shown in Fig. 6, the simulation based on the physical calculation model is performed by dividing the target area of ​​the simulation into a plurality of meshes as shown in Fig. 7, and using known physical calculations (such as the finite volume method, the finite element method, and the boundary tracking method). Fig. 7 shows an example in which the target area shown in Fig. 7(a) is divided into 10 as shown in Fig. 7(b), but the size of the mesh and the number of divisions are arbitrary and may be appropriately determined depending on the accuracy of the simulation result, the calculation load (calculation cost), etc.

[0038] When the simulation of the welding state is started, in step S601, the simulation unit 208 determines the melting position due to the generation of an arc in virtual welding. The simulation unit 208 determines the melting position in the current simulation by calculating the start position and end position (position where the arc falls) of the generated arc based on the estimated arc length in virtual welding and information on the pseudo torch operation (speed, position, and angle) in virtual welding acquired in step S501 of Fig. 5.

[0039] In step S602, simulation unit 208 acquires a prediction result of the welding state from storage unit 202. The prediction result of the welding state acquired in step S602 is predicted in the previous simulation using a prediction model of the welding state in step S608 described later and stored in storage unit 202. Note that in the first simulation, for example, a welding state obtained by physical calculation may be used as the prediction result, or a welding state based on an image captured during on-site welding may be used as the prediction result.

[0040] In step S603, the simulation unit 208 piles up beads in the mesh-divided target area. In the mesh-divided target area, the simulation unit 208 piles up beads according to the penetration amount and the like in a predetermined range based on the end position of the arc (position where the arc falls) obtained in step S601. Here, for example, the bead pile-up is calculated using a cone model in which the distribution of the piled-up beads has a cone shape. FIG. 8 shows an example of the processing result in step S603. FIG. 8 shows an example in which beads are piled up as shown in FIG. 8(b) using a cone model in the mesh-divided target area shown in FIG. 8(a). The predetermined range in which the beads are piled up may be determined based on the end position of the arc (position where the arc falls) obtained in step S601 and the predicted result of the welding state obtained in step S602.

[0041] In step S604, the simulation unit 208 performs calculation of heat input (input of temperature parameters) for the target region divided into meshes. The simulation unit 208 performs calculation of heat input by applying a predetermined heat input model to the range of the molten pool having the shape and size indicated by the predicted result of the welding state obtained in step S602, which is formed at the end position of the arc obtained in step S601 (position where the arc falls). At this time, if the range of the molten pool indicated by the predicted result of the welding state is different from the heat input range in the heat input model, the heat input model is applied only to the range of the molten pool indicated by the predicted result of the welding state. For example, if the range of the molten pool indicated by the predicted result of the welding state is the region 901 shown in FIG. 9(a) and the heat input range in the heat input model is the region 902 shown in FIG. 9(b), the heat input is calculated by applying the heat input model only to the region 901 shown in FIG. 9(c), excluding the range other than the molten pool.

[0042] In step S605, the simulation unit 208 performs a fluid calculation for the target region divided into meshes. The simulation unit 208 performs deformation due to a flow phenomenon for each mesh by moving meshes that show a predetermined temperature or higher as a fluid based on physical calculations. The simulation unit 208 performs a fluid calculation, for example, by approximating a diffusion equation on a two-dimensional mesh using a central difference method and solving it using an explicit method. This method basically results in changing the value so as to approach the average value of the values ​​of the meshes on the top, bottom, left, and right, so that the components in adjacent directions increase or decrease by the same amount with opposite signs, and a result equivalent to a movement of volume is obtained.

[0043] In step S606, the simulation unit 208 performs a heat dissipation calculation (heat dissipation of temperature parameters) for the target region divided into meshes. The simulation unit 208 performs a heat dissipation calculation by applying a predetermined heat dissipation model to each mesh.

[0044] In step S607, the simulation unit 208 performs calculations for the chorus for the target region divided into meshes. The simulation unit 208 defines a chorus generation range in which the outer edge of the range in which the beads were piled up in step S603 is set as the start position of the chorus generation, and a predetermined edge away from the range in which the beads were piled up is set as the end position of the chorus generation, and performs calculations for the chorus to be generated.

[0045] In this manner, a simulation using physical calculations is performed by the simulation unit 208, and a welding state in virtual welding at a certain time is estimated. The simulation result (estimated result) of the welding state obtained by the above processing is stored in the storage unit 202, for example.

[0046] In the next step S608, the simulation unit 208 predicts the welding state to be used in the next simulation using the prediction model of the welding state. The simulation unit 208 predicts the next welding state using the prediction model of the welding state based on the time-series images of the welding state in the virtual welding up to this time, which are based on the simulation results obtained as described above, and on the feature amount data indicating the feature amount in the virtual welding work. When the processing of step S608 ends, the simulation of the welding state shown in FIG. 6 ends, and the processing returns to the processing of the flowchart shown in FIG. 5.

[0047] Fig. 10 is a diagram for explaining prediction of the welding state executed in step S608. Fig. 10 shows an example of predicting the welding state at time (t+1) based on data up to time t and the like. As shown in Fig. 10, an image 1001 of the welding state based on a plurality of (e.g., n) simulation results of the welding state up to time t (simulation results of the welding state at times t, (t-1), . . . , (tn-1)) and feature amount data 1002 at time t are input to simulation unit 208. Simulation unit 208 acquires a prediction result of the welding state at time (t+1) using prediction model 403 of the welding state generated as shown in Fig. 4 described above based on image 1001 of the welding state up to time t and feature amount data 1002 at time t.

[0048] The virtual welding system 100 in this embodiment predicts the welding state in virtual welding using a prediction model of the welding state, executes a simulation based on a physical calculation model by referring to the predicted result of the welding state, and displays the welding state in virtual welding using a VR image based on the simulation result. This makes it possible to display a more realistic welding state in virtual welding compared to a display of the welding state based on physically calculated information. In addition, by predicting and processing the welding state in virtual welding using the prediction model of the welding state, it is possible to display and reproduce the welding state in real time according to the operation of the user (operator) of the virtual welding.

[0049] Here, the feature quantities of the welding operation used in generating the prediction model of the welding state and predicting the welding state using the prediction model include, for example, the welding speed (torch speed), the distance between the welding base material and the torch (tip), and the motion angle and movement angle of the torch during the welding operation. These feature quantities can be acquired based on data indicating the motion of the torch during the welding operation detected by, for example, the torch motion detection device 109 or the controller 106. Note that the feature quantities of the welding operation are not limited to these, and other feature quantities may be used.

[0050] In the above embodiment, the calculation of heat input in the simulation of the welding state is performed by applying a predetermined heat input model, and an example of the heat input model is a cone heat input model that adds heat based on the distribution of the shape of a cone 1101 as shown in Fig. 11(a). When the applied heat input model is a cone heat input model and the range of the molten pool shown in the prediction result of the welding state is different from the heat input range in the heat input model, as shown in Fig. 11(b), for example, in the same manner as described above, it is sufficient to apply the heat input model 1103 only to the range 1102 of the molten pool shown in the prediction result of the welding state. Another example of the heat input model is a cylinder heat input model that adds heat based on the distribution of the shape of a cylinder having a bottom surface similar to that of the cone 1101 shown in Fig. 11(a).

[0051] In addition, in the simulation of the welding state, instead of simulating each phenomenon by physical calculation, a surrogate model that predicts the phenomenon by a machine learning model may be applied. By performing prediction using the surrogate model, it is possible to reduce the calculation load and the calculation cost.

[0052] It should be noted that the above-mentioned embodiments are merely examples of the implementation of the present invention, and the technical scope of the present invention should not be interpreted as being limited by these embodiments. In other words, the present invention can be implemented in various forms without departing from its technical concept or main features. [Explanation of symbols]

[0053] 100 Virtual Welding System 201 Acquisition Department 202 Storage section 203 Display section 204 Operation section 205 Analysis Department 206 Evaluation Department 207 Learning Processing Unit 208 Simulation Department 209 Image Generation Unit 210 Display control unit 211 Control Unit

Claims

1. A prediction means for predicting a welding state in a virtual welding process using a prediction model for predicting a future welding state; a simulation means for estimating a current welding state in virtual welding by physical calculation with reference to the welding state predicted by the prediction means; and display means for displaying an image of the welding state in virtual welding based on a result of estimation by the simulation means.

2. The virtual welding system according to claim 1 , wherein the welding state includes a state of a molten pool formed during welding.

3. 3. The virtual welding system according to claim 1, further comprising an image generating means for generating an image showing the welding state in virtual welding, the image having a display color set according to temperature based on the estimation result by the simulation means.

4. 3. The virtual welding system according to claim 1, wherein the prediction means predicts at least one of a shape and a size of the welding state.

5. 3. The virtual welding system according to claim 1, wherein the prediction means predicts a future welding state based on a feature amount of a welding operation and a welding state in a virtual welding operation that has already been predicted.

6. The virtual welding system according to claim 5 , wherein the feature amount includes at least one of a welding speed in the virtual welding, a distance between the welding base material and the torch, and an operating angle and a moving angle of the torch.

7. The virtual welding system according to claim 1 or 2, further comprising a learning means for performing machine learning using, as input, an image of a welding state captured during actual welding performed in an on-site environment and feature quantities of the actual welding work, to generate the predictive model.

8. a learning means for performing machine learning using, as input, an image of a welding state captured during actual welding performed in an on-site environment and feature quantities of the actual welding work, to generate a prediction model for predicting a future welding state; and a storage means for storing the prediction model generated by the learning means.

9. a prediction step of predicting a welding state in a virtual welding process using a prediction model for predicting a future welding state; a simulation step of estimating a current welding state in virtual welding by physical calculation with reference to the welding state predicted in the prediction step; and a display step of displaying an image of the welding state in virtual welding on a display means based on a result of estimation in the simulation step.

10. The virtual welding system computer a prediction step of predicting a weld state in a virtual weld using a prediction model for predicting a future weld state; a simulation step of estimating a current welding state in virtual welding by physical calculation with reference to the welding state predicted in the prediction step; and a display step of displaying an image of the welding state in virtual welding on a display means based on a result of estimation in the simulation step.

Images