Forming plan support method and forming plan support device

Abstract

This method uses a mathematical model to correlate input information, including the material of the object, welding conditions for the deposited weld bead, and the welding trajectory, with output information, including the characteristic values ​​of the object when stacked under the conditions of this input information. This mathematical model is then used to create a database. The database is searched to determine the material, welding conditions, and welding trajectory corresponding to the target characteristic values ​​of the object to be manufactured, and the resulting material, welding conditions, and welding trajectory are presented. During the database creation process, each input sub-item of the input information is correlated with the individual characteristic values ​​of the output information using a mathematical model.

Description

Technical Field

[0001] The present invention relates to a molding plan supporting method and a molding plan supporting device when manufacturing a molded object by depositing a weld bead. Background Art

[0002] In recent years, demand for component manufacturing using 3D printers for layered molding has increased significantly, leading to research and development efforts aimed at the practical application of metal molding. 3D printers for metal layered molding use heat sources such as lasers and arcs to melt and solidify metal powder or wire, depositing weld metal (weld beads) to create objects of the desired shape.

[0003] However, in laminated molding using metal materials, material properties such as metal structure and hardness are prone to change depending on manufacturing conditions. Consequently, the properties of the metal material constituting the object can significantly vary from the expected properties. Therefore, conventional welding techniques predict the properties of the object under specified manufacturing conditions based on empirical insights and trial and error, and then adjust the manufacturing conditions to achieve the desired shape and properties.

[0004] Furthermore, in order to materialize the information obtained through the above-mentioned experience and trial and error on a computer, Patent Document 1 discloses the following example: for example, test cross-sectional images of welds and test welds are prepared, and machine learning is utilized in the process of judging the suitability of various specifications of the welds such as strength, ductility, hardness, toughness, and granular structure from these images.

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Publication No. 2019-5809 Summary of the Invention

[0008] Problems to be solved by the invention

[0009] However, because the lamination process is more complex than a simple welding process, predicting the properties of the products produced by lamination based on their materials is considered difficult. Furthermore, the lamination method offers a high degree of freedom in manufacturing conditions, and the combination of product properties involves many factors, requiring extensive computational processing to predict these properties.

[0010] Therefore, an object of the present invention is to provide a molding plan supporting method and a molding plan supporting device that can efficiently predict the characteristics of a molded object with low cost, thereby supporting the creation (creation) of a more appropriate molding plan for the molded object.

[0011] Means for solving problems

[0012] The present invention includes the following structures.

[0013] (1) A shaping plan supporting method for supporting the creation of a shaping plan that represents the material of the shaping object, the welding conditions of the weld beads, and the welding trajectory when a shaping object is manufactured by stacking weld beads formed by melting and solidifying welding filler metal supplied from a welding head into a desired shape, the shaping plan supporting method comprising the following steps: generating a mathematical model that establishes a relationship between input information including the material of the shaping object, the welding conditions, and the welding trajectory, and output information of characteristic values ​​of the shaping object when stacking shaping is performed under the conditions of the items included in the input information; using the mathematical model to create a shaping plan that represents the shaping object; A database of the correspondence between the input information and the output information; searching the database to obtain the conditions of each item of the material of the object, the welding conditions and the welding track corresponding to the target characteristic values ​​of the object to be manufactured; and prompting the material of the object, the welding conditions and the welding track corresponding to the obtained target characteristic values, the items of the input information respectively have a plurality of input sub-items different from each other, and the output information has a plurality of individual characteristic values ​​corresponding to the input sub-items. In the process of generating the mathematical model, the input sub-items of the input information are respectively related to the individual characteristic values ​​using the mathematical model.

[0014] (2) A forming plan supporting method for supporting the creation of a forming plan that respectively characterizes the material of the forming object, the welding conditions of the deposited weld, and the welding track when forming a forming object by stacking deposited welds formed by melting and solidifying welding filler metal supplied from a welding head into a desired shape, the forming plan supporting method comprising the following steps: generating a first mathematical model that establishes a relationship between input information including the material of the forming object, the welding conditions, and the welding track, and intermediate output information including information on the temperature history of the forming object when stacking the forming object under the conditions of the items of the input information, and generating a second mathematical model that establishes a relationship between the intermediate output information and output information including characteristic values ​​of the forming object; and using the first mathematical model and the second mathematical model to create a forming plan that characterizes the input information. A database having a corresponding relationship with the output information; searching the database to obtain the temperature history record, the material of the object, the welding conditions and the welding track corresponding to the target characteristic value of the object to be manufactured; and prompting the material of the object, the welding conditions and the welding track corresponding to the obtained target characteristic value, the items of the input information respectively have a plurality of input sub-items that are different from each other, the intermediate output information has individual intermediate values ​​corresponding to the input sub-items, and the output information has a plurality of individual characteristic values ​​corresponding to the individual intermediate values. In the process of generating the first mathematical model and the second mathematical model, the input sub-items are respectively related to the individual intermediate values ​​using the first mathematical model, and the individual intermediate values ​​are respectively related to the individual characteristic values ​​using the second mathematical model.

[0015] (3) A shaping plan support device that supports the creation of a shaping plan that characterizes the material of the shaping object, the welding conditions of the weld beads, and the welding track when a shaping object is manufactured by stacking weld beads formed by melting and solidifying welding filler metal supplied from a welding head into a desired shape, the shaping plan support device comprising: a mathematical model generating unit that generates a mathematical model that establishes a relationship between input information including the material of the shaping object, the welding conditions, and each item of the welding track, and output information including characteristic values ​​of the shaping object when stacked under the conditions of the input information; and a database generating unit that uses the mathematical model to A database is prepared to characterize the correspondence between the input information and the output information; a retrieval unit is configured to retrieve the material of the object, the welding conditions, and the welding trajectory corresponding to the target characteristic values ​​of the object to be manufactured; and an output unit is configured to prompt the material of the object, the welding conditions, and the welding trajectory corresponding to the obtained target characteristic values, wherein the items of the input information each have a plurality of input sub-items that are different from each other, and the output information has a plurality of individual characteristic values ​​corresponding to the input sub-items, and the mathematical model generation unit uses the mathematical model to establish a relationship between the input sub-items of the input information and the individual characteristic values.

[0016] (4) A shaping plan support device that supports the creation of shaping plans that respectively characterize the material of the shaping object, the welding conditions of the weld beads, and the welding track when a shaping object is manufactured by stacking weld beads formed by melting and solidifying welding filler metal supplied from a welding head into a desired shape, the shaping plan support device comprising: a mathematical model generating unit that respectively generates a first mathematical model that establishes a relationship between input information containing items of the material of the shaping object, the welding conditions, and the welding track and intermediate output information containing information on the temperature history of the shaping object when stacking the shaping object under the conditions of the items of the input information, and a second mathematical model that establishes a relationship between the intermediate output information and output information containing characteristic values ​​of the shaping object; and a database generating unit that uses the first mathematical model and the second mathematical model to generate a mathematical model that establishes a relationship between the input information containing items of the material of the shaping object, the welding conditions, and the welding track and intermediate output information containing information on the temperature history of the shaping object when stacking the shaping object under the conditions of the items of the input information; A database is prepared to characterize the correspondence between the input information and the output information; a retrieval unit, which searches the database to obtain the temperature history record, the material of the object, the welding conditions, and the welding track corresponding to the target characteristic value of the object to be manufactured; and an output unit, which prompts the material of the object, the welding conditions, and the welding track corresponding to the obtained target characteristic value, wherein the items of the input information each have a plurality of input sub-items that are different from each other, the intermediate output information has individual intermediate values ​​corresponding to the input sub-items, and the output information has a plurality of individual characteristic values ​​corresponding to the individual intermediate values, and the mathematical model generation unit establishes a relationship between the input sub-items and the individual intermediate values ​​using the first mathematical model, and establishes a relationship between the individual intermediate values ​​and the individual characteristic values ​​using the second mathematical model.

[0017] Effects of the Invention

[0018] According to the present invention, the characteristics of a formed object can be predicted efficiently with less effort, thereby supporting the creation of a more appropriate forming plan for the formed object. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is an overall structural diagram of the shaping system for manufacturing shapes.

[0020] Figure 2 This is a schematic block diagram of a robot control device.

[0021] Figure 3 This is a schematic block diagram of the shaping control device.

[0022] Figure 4 This is an explanatory diagram showing the steps of creating a molding program for performing layered molding.

[0023] Figure 5 This is an explanatory diagram showing the steps of building a database.

[0024] Figure 6 It is a flowchart showing the steps of building a database.

[0025] Figure 7 This is a flowchart showing the procedure for creating the initial database used in the first step.

[0026] Figure 8 (A) is an explanatory diagram showing how the relationship between the input information and the output information is established using a mathematical model. Figure 8 (B) is an explanatory diagram showing a database that links input information and output information.

[0027] Figure 9 This is an explanatory diagram showing the establishment of a relationship between input information and output information using a mathematical model in which the input information includes a plurality of items.

[0028] Figure 10 This is an explanatory diagram showing a process of dividing the shape of a to-be-made object into a plurality of element shapes and obtaining a welding trajectory for each element shape.

[0029] Figure 11 This is a flowchart showing the steps for creating a shaping plan when the shape of a shaped object is decomposed into element shapes.

[0030] Figure 12 This is an explanatory diagram showing the establishment of a relationship using a mathematical model of input information, intermediate output information, and output information.

[0031] Figure 13 This is a graph showing the temperature history at a specific position of the weld bead formed during forming.

[0032] Figure 14 The following are graphs showing the differences in cooling characteristics when weld beads are formed at different heat input levels. (A) shows the temperature change characteristics for a relatively high heat input level, and (B) shows the temperature change characteristics for a relatively low heat input level.

[0033] Figure 15 This is an explanatory diagram showing a state in which a plurality of databases that establish a relationship between input information and output information are selectively used. DETAILED DESCRIPTION

[0034] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0035] Here, the example of a case where a deposited weld bead formed by melting and solidifying a welding filler metal supplied from a welding head is laminated and formed into a desired shape using a forming device is used for explanation. However, the forming method and the structure of the forming device are not limited to this. For example, other forming methods such as powder sintering and lamination can also be used.

[0036] <Structure of the modeling system>

[0037] Figure 1 It is an overall structural diagram of the shaping system for manufacturing shapes.

[0038] The molding system 100 of this configuration includes a molding device 11 and a molding control device 13 that controls the molding device 11. The molding device 11 includes a welding robot 17 having a welding head with a welding torch 15 mounted on its distal shaft; a robot control device 21 that drives the welding robot 17; a welding filler metal supply unit 23 that supplies welding filler metal (welding wire) M to the welding torch 15; and a welding power source 25 that supplies welding current.

[0039] (Shaping device)

[0040] The welding robot 17 is a multi-jointed robot that supports continuously supplied welding filler metal M at the tip of a welding torch 15 mounted on a distal shaft of a robot arm. The position and posture of the welding torch 15 can be arbitrarily set three-dimensionally within the range of the robot arm's freedom according to commands from a robot controller 21.

[0041] A shape sensor 32 and a temperature sensor 30 are provided on the front end shaft of the welding robot 17 so as to move integrally with the welding torch 15 .

[0042] Shape sensor 32 is a non-contact sensor that measures the shape of the formed weld bead 28 and, if necessary, the shape around the weld bead formation location. Measurements by shape sensor 32 can be performed simultaneously with weld bead formation or at different times before or after weld bead formation. Shape sensor 32 can be a laser sensor that detects three-dimensional shape based on the position of reflected light from an irradiated laser beam or the time from irradiation timing to detection of reflected light. Shape sensor 32 is not limited to a laser; other detection methods may also be used.

[0043] The temperature sensor 30 is a contact-type sensor such as a radiation thermometer or a thermal imager, and detects the temperature (temperature distribution) at any position of the object.

[0044] The welding torch 15 is a gas metal arc welding torch having a shielded nozzle (not shown) and supplying shielding gas from the shielded nozzle. The arc welding method may be either a consumable electrode method such as sheathed arc welding or carbon dioxide gas arc welding, or a non-consumable electrode method such as TIG welding or plasma arc welding, and is selected appropriately depending on the laminated object being produced.

[0045] For example, in the case of a consumable electrode type, a contact tip is disposed inside a shield nozzle, and welding filler metal M, to which melting current is supplied, is held at the contact tip. The welding torch 15 holds the welding filler metal M and generates an arc from the tip of the welding filler metal M in a shielding gas atmosphere.

[0046] The welding filler metal supply unit 23 includes a reel 29 for winding the welding filler metal M, and a wire feed sensor 31 that measures the amount of welding filler metal M fed from the reel 29 to the feed mechanism and the welding torch 15. The welding filler metal M is fed from the welding filler metal supply unit 23 to the feed mechanism (not shown) attached to a robot arm or the like, and is fed forward and reversely by the feed mechanism as needed to the welding torch 15.

[0047] Any commercially available welding wire can be used as the welding filler metal M. For example, wires specified in MAG welding and MIG welding solid wires for mild steel, high-tensile steel, and low-temperature steel (JIS Z3312) and arc welding flux wires for mild steel, high-tensile steel, and low-temperature steel (JIS Z3313) can be used. Furthermore, welding filler metals M such as aluminum, aluminum alloys, nickel, and nickel-based alloys can be used depending on the desired properties.

[0048] Then, when the welding filler metal M continuously fed as described above is melted and solidified by the arc, a deposited weld bead 28, which is a molten solidified body of the welding filler metal M, is formed on the base plate 27. The base plate 27 is a metal plate such as a steel plate, but is not limited to a plate shape and may be other shapes such as a block, a rod, or a cylinder.

[0049] (Robot control device)

[0050] The robot control device 21 drives the welding robot 17 to move the welding torch 15 , and melts the continuously supplied welding filler metal M by the welding current and welding voltage from the welding power source 25 .

[0051] Figure 2 It is a schematic block diagram of the robot control device 21.

[0052] The robot control device 21 is a computer device including an input / output interface 33 , a storage unit 35 , and an operation panel 37 .

[0053] The welding robot 17, welding power source 25, and forming control device 13 are connected to the input / output interface 33. The storage unit 35 stores various information, including a driver program described later. The storage unit 35 includes storage media such as ROM, RAM, hard disks, SSDs (Solid State Drives), CDs, DVDs, and various memory cards, and can input and output various information. The operation panel 37 can be an information input unit such as an input operation panel or an input terminal (teaching pendant) for teaching the welding robot 17.

[0054] The molding program corresponding to the object to be produced is sent from the molding control device 13 to the robot control device 21. The molding program contains a large number of command codes and is created based on an appropriate algorithm according to various conditions such as the object's shape data (CAD data, etc.), material, and heat input.

[0055] The robot controller 21 executes the shaping program stored in the storage unit 35, driving the welding robot 17, the welding filler metal supply unit 23, the welding power source 25, and the like to form a deposited weld bead 28 in accordance with the shaping program. Specifically, the robot controller 21 drives the welding robot 19 to move the welding torch 15 along its trajectory (welding trajectory) set in the shaping program. Furthermore, the robot controller 21 drives the welding filler metal supply unit 23 and the welding power source 25 in accordance with the set welding conditions, causing the welding filler metal M at the tip of the welding torch 15 to melt and solidify via the arc. This forms a deposited weld bead 28 on the chassis 27. The deposited weld beads 28 are arranged adjacent to each other to form a layer of deposited weld beads. By repeatedly stacking one layer of deposited weld beads on top of another, the desired three-dimensional object is formed.

[0056] The forming control device 13 is isolated from the forming device 11 and can be remotely connected to the forming device 11 via a network, communication means, storage medium, etc. The forming program can be created in addition to the forming control device 13 or in other devices and sent via communication.

[0057] (Generation of modeling program)

[0058] Next, the configuration of the formation control device 13 and the specific steps until the formation control device 13 generates a formation program will be described.

[0059] Figure 3 It is a schematic block diagram of the shaping control device 13.

[0060] The shaping control device 13 is a computer device similar to the robot control device 21 , and includes a CPU 41 , a storage unit 43 , an input / output interface 45 , an input unit 47 , and an output unit 49 .

[0061] Storage unit 43 includes a ROM as a nonvolatile storage area, a RAM as a volatile storage area, and the like. The aforementioned shape sensor 32, temperature sensor 30, welding filler metal supply unit 23 including wire feed sensor 31, welding power source 25, robot controller 21, input unit 47, and output unit 49 are connected to input / output interface 45.

[0062] The input unit 47 is an input device such as a keyboard and a mouse, and the output unit 49 includes a display device such as a monitor or an output terminal for transmitting an output signal.

[0063] The formation control device 13, which will be described in detail later, further includes a basic information table 51, a mathematical model generation unit 53, a database creation unit 55, a formation planning unit 57, and a search unit 59. Each of the above components is operated by a command from the CPU 41 and performs its respective functions.

[0064] Figure 4 This is an explanatory diagram showing the steps of creating a molding program for performing layered molding.

[0065] First, the operator Figure 3 The input unit 47 of the molding control device 13 shown in the figure inputs data such as the material, shape, and welding conditions of the object to be manufactured. The molding control device 13 creates a molding plan based on the input data so that the object can obtain the desired characteristics. For example, a model is generated from the shape data, and the generated model is divided into layers for each given weld bead height. Various conditions such as the weld bead material, weld bead width, and weld bead formation sequence (weld trajectory) are determined so that each resulting layer is filled with weld bead. There are various methods for determining these weld trajectories, and the method is not limited.

[0066] Next, a database 61 showing the correspondence between various pre-defined manufacturing conditions and the properties of the resulting objects is referenced to predict the property values ​​(metallic structure, average grain size, Vickers hardness, tensile strength, toughness, etc.) of the object produced according to the prepared molding plan. The above-mentioned parameters are preferably used as the property values. This facilitates data collection because the property values ​​can be easily measured using commonly used measuring devices such as metallurgical microscopes, electron microscopes (e.g., SEMs), and Vickers testing machines.

[0067] If the predicted properties of the object do not meet the desired properties, the various manufacturing conditions mentioned above are adjusted and a new forming plan is created. If the properties of the object based on the created forming plan meet the desired properties, a forming program corresponding to the forming plan is created. The thus created forming program is sent to Figure 1The robot control device 21 shown. The robot control device 21 executes the sent forming program to form the object in layers.

[0068] In the molding plan support method and apparatus according to the present invention, a database 61 used for predicting and determining whether a molded object can obtain desired characteristics through a created molding program is efficiently constructed at low cost. This enables accurate and rapid molding plan determination, thereby providing support for smoothly creating a more appropriate molding plan.

[0069] <First Database Structure Example>

[0070] Next, a method of constructing the above-mentioned database 61 will be described.

[0071] Figure 5 This is an explanatory diagram showing the steps of constructing the database 61. Here, input information including the material of the object, welding conditions of the weld bead, and partial weld tracks, and output information including characteristic values ​​of the object formed by stacking under the conditions of the input information are related using a mathematical model. This relationship building process is repeated through machine learning, and a database is created based on the obtained mathematical model. Figure 4 The database 61 is referenced in the prediction and judgment shown.

[0072] Specifically, the forming control device 13 creates a forming plan based on the input data of the material, shape, welding conditions, etc. The characteristic values ​​of the formed object when the formed object is produced according to the forming plan are obtained through the following first and second steps.

[0073] In the first step, the formation control device 13 refers to the initial database 63 in which the relationship between the formation plans and the characteristic values ​​is registered in advance, and predicts the characteristics of the object formed by stacking the prepared formation plans.

[0074] In the second step, the forming control device 13 drives the robot control device 21 according to the prepared forming plan, causing the forming device 11 to stack and form the object. Test samples are cut from the formed objects and their mechanical strength, metal structure, etc. are measured through testing (observation).

[0075] The predicted results of the characteristics of the objects based on the same formation plan are compared with the test results, and a mathematical model 62 is generated so that the difference between the two becomes smaller. The mathematical model 62 is used to create a database 61. The above-mentioned initial database 63 and the database 61 are created by Figure 3 Although the database creation unit 55 shown here performs the operation, it may be performed by a device other than the formation control device 13.

[0076] Here, a series of processing steps will be described: a mathematical model 62 is generated by performing machine learning on the predicted results of the characteristics of the initial database 63 in the first step and the test results in the second step, and a database 61 is created using the mathematical model 62 .

[0077] Figure 6 1 is a flowchart showing the steps of constructing the database 61 .

[0078] First, the object to be manufactured is determined and its shape data (based on 3D-CAD) is created (S11). A forming plan is created from this object's shape data (S12). The forming plan includes multiple slices of the object model obtained by dividing the layers along a given stacking direction axis, the shape of the weld bead in each slice, and the welding conditions for forming the weld bead.

[0079] Next, based on the prepared molding plan, the properties of the molded object are predicted in the first step (S13). The properties of the molded object are predicted using the initial database 63. The initial database 63 is based on the basic information table 51 (which represents the correspondence between various manufacturing conditions and the property values ​​of the molded object produced thereby) based on the experience and knowledge gained from past molding. Figure 3 ) to make.

[0080] Figure 7 This is a flowchart showing the procedure for creating the initial database 63 used in the first step.

[0081] First, information on parameters used in the database (e.g., passes for forming weld beads, the number of passes, the order of forming weld beads (welding tracks), cross-sectional shapes of weld beads, etc.) is extracted from the pre-prepared basic information table 51 and prepared as learning data (S21).

[0082] Next, the prepared learning data and the characteristic values ​​of the object corresponding to the learning data are related through an initial mathematical model (S22). That is, by repeatedly performing machine learning on a plurality of learning data and the characteristic values ​​of the object corresponding thereto, an initial mathematical model that characterizes the relationship between the two parties is generated. The so-called "mathematical model" mentioned here refers to the quantitative behavior of the characteristics of the object being standardized, and its properties can be simulated by calculation. That is, the mathematical model is a computational model made based on a group of experimental data collected in the experiment and associated with a given algorithm. The computational model can assume a given function and be optimized to match the experimental data well, or it can be made by giving input information and output information through machine learning. As examples of specific algorithms, support vector machines, neural networks, or random forests can be cited.

[0083] Then, the generated initial mathematical model is used to predict characteristic values ​​of objects corresponding to a plurality of learning data, and these predicted values ​​are registered in correspondence with the learning data in the table components of the initial database 63 (S23).

[0084] On the other hand, in step 2 ( Figure 6 S14 to S16) of the invention, the object is produced based on the produced shape plan. That is, the shape planning unit 57 ( Figure 3 ) creates a shaping program (S14) corresponding to the shaping plan, and drives the shaping program by executing the shaping program. Figure 1 The forming device 11 shown forms a formed object (S15). Then, a test sample is cut out from the obtained formed object and various properties of the test sample are tested (S16).

[0085] Then, the predicted result of the characteristics of the object obtained in step 1 is compared with the test result obtained in step 2 (S17). If the difference between the predicted result and the test result is large, the correction Figure 5 The mathematical model 62 shown (equivalent to the initial mathematical model used in the creation of the initial database 63) is modified to minimize the difference between the two (S18). That is, the test results for the input information are used as teaching data, and the mathematical model 62 is trained to make the predicted results close to the test results. In addition, if the difference between the predicted results and the test results is small, the mathematical model 62 can be corrected without further machine learning, which helps to improve the accuracy of the mathematical model 62. In this way, the mathematical model 62 becomes a learned model that has learned the relationship between the input information and the output information.

[0086] Then, using mathematical model 62, which is a further machine learning implementation of the initial mathematical model, the characteristic values ​​(output information) of the object corresponding to any number of conditions (input information) are predicted. The set conditions and the predicted characteristic values ​​are linked together and used as table elements in database 61. In this way, initial database 63 is modified using mathematical model 62 to construct database 61 (S19), in which the predicted results for specific conditions and the experimental results are accurately consistent).

[0087] In this way, by predicting output information using the mathematical model 62 based on a plurality of input information, portions for which no experimental results exist can be supplemented, thereby easily increasing the amount of information in the database 61 and improving the accuracy of prediction.

[0088] Next, a specific method of constructing the database 61 using the mathematical model 62 will be described in more detail.

[0089] Figure 8(A) is an explanatory diagram showing how the relationship between input information and output information is established using a mathematical model. Figure 8 (B) is an explanatory diagram showing a database that links input information and output information.

[0090] Here, as one of the input information, the welding filler metal which becomes the material of the object is taken as an example for explanation. Figure 8 As shown in (A), a plurality of welding filler metals A, B, C, ... can be selected as the welding filler metal. When a shape is produced using each of the welding filler metals A, B, C, ..., the properties of the resulting shape are characteristic value A for welding filler metal A, characteristic value B for welding filler metal B, characteristic value C for welding filler metal C, ...

[0091] In this case, a separate mathematical model is used for each type of welding filler metal, so that mathematical model A is used for characteristic value A of the shape of welding filler metal A, mathematical model B is used for characteristic value B of the shape of welding filler metal B, and mathematical model C is used for characteristic value C of the shape of welding filler metal C, to establish a relationship with the characteristic values.

[0092] Therefore, if Figure 8 As shown in (B), the created database 61 establishes relationships so that welding filler metal A is associated with characteristic value A, welding filler metal B is associated with characteristic value B, and welding filler metal C is associated with characteristic value C. Thus, since characteristic values ​​for machine learning are determined for each type of welding filler metal, characteristic values ​​corresponding to the characteristics of the welding filler metal can be accurately and meticulously set. This improves the prediction accuracy of the characteristic values.

[0093] The type of welding filler metal may be identified by a trade name such as MG-51T or MG-S63B (solid welding wire manufactured by Kobe Steel), or may be distinguished by its component composition (eg, carbon content).

[0094] In the above example, the characteristic value relationship is established for each type of welding filler metal, but there are many types of items in actual input information.

[0095] Figure 9 This is an explanatory diagram showing the relationship between input information and output information established using a mathematical model in which input information includes a plurality of items.

[0096] As input information, there are at least the material of the object, welding conditions and a portion of the welding track. As the material of the weld, in addition to the above-mentioned welding filler metal, the chassis 27 ( Figure 1 ), components such as structural parts (not shown) that are joined to the deposited weld to become components of the shaped object.

[0097] Examples of welding conditions include at least one of the following: welding current, welding voltage, welding speed, weld track pitch width, inter-pass time, welding head aiming position, welding head welding posture, and filler metal feed rate, or a combination thereof, during weld bead formation. The welding head aiming position refers to the position of the torch tip for positioning the torch tip at the welding position, and the welding head welding posture refers to the inclination angle between the vertical axis and the torch axis, as well as the circumferential angle of the torch in the inclination direction around the vertical axis. Furthermore, the weld track pitch width refers to the distance between adjacent weld tracks, and the inter-pass time refers to the time between transitioning from one weld pass to the next weld pass in a plurality of weld tracks.

[0098] The above-mentioned inter-pass time affects the metal structure of the formed weld bead.

[0099] When forming a weld bead, the molten mild steel filler metal rapidly cools, resulting in a mixed structure primarily composed of bainite. Furthermore, when the molten mild steel filler metal solidifies naturally, it forms a structure composed of coarse ferrite, pearlite, and bainite. When stacking weld beads, if these structures are heated above the ferrite transformation point by the subsequent weld bead, the pearlite and bainite transform into ferrite, and the coarse ferrite becomes refined.

[0100] By adjusting the interpass time and controlling the interpass time and heat input so that, for example, the interpass temperature remains within the range of 200°C to 550°C, and simultaneously depositing the next weld bead and similarly depositing subsequent weld beads, the weld bead is heated above the transformation point of the ferrite. This results in a homogenized structure containing a fine ferrite phase with an average grain size of 10 μm or less. Such a weld bead achieves a high hardness (e.g., approximately 130 to 180 Hv in Vickers hardness), resulting in a generally uniform hardness with excellent mechanical strength and minimal variation.

[0101] On the other hand, if the interpass temperature is less than 200°C when the next weld bead is stacked, even if the weld bead is heated by stacking subsequent weld beads, it will not exceed the transformation point of the ferrite, and a uniform structure containing fine ferrite phases cannot be achieved. For example, during the initial forming process, the interpass temperature during the stacking of the next weld bead falls below 200°C due to heat removal from the base plate 27. In this case, the weld bead at the beginning of the forming process has a mixed structure primarily composed of bainite. Furthermore, if the interpass temperature exceeds 550°C, the weld bead is heated by stacking the next weld bead, causing flattening and sagging of the weld bead, making it impossible to stack the weld bead into the desired shape. Furthermore, the deposited weld bead in the later stage of forming (the top layer of the formed object) is not heated again because it is not superimposed on the deposited weld bead of the next layer. Therefore, the molten welding filler metal remains in a naturally solidified state, that is, the structure containing coarse ferrite, pearlite and bainite remains unchanged.

[0102] Thus, the metal structure of the weld bead formed during the interpass time changes, and as a result, the properties of the product also change. The above is about the influence of the interpass time on the properties of the product. Similarly, it can be seen that other parameters also have an impact on the properties of the product.

[0103] A partial welding track is a welding track for elemental shapes cut out of a portion of an object. This track is used to create a simple shape (elementary shapes) when breaking down a complex shape. Information on each welding track includes the passes used to form the weld bead, the number of passes, the order in which the weld bead is formed, and the cross-sectional shape of the weld bead.

[0104] Here, the material of the forming material, welding conditions, and partial welding tracks mentioned above are respectively referred to as "items", and the welding filler metals A, B, C,..., as well as the welding current, welding voltage, welding speed,..., element shape, pass, number of passes,... for each item are respectively referred to as "input sub-items".

[0105] By dividing each item of input information into multiple input sub-items, the range of inputs that can be entered can be limited. Specifically, by preventing input data outside of the input sub-items from being set, the recommended ranges for, for example, the welding robot 17 of the forming device 11 and the recommended conditions for welding filler metal can be maintained. This prevents equipment malfunctions and material problems before they occur, and avoids presenting inappropriate conditions.

[0106] like Figure 9 As shown, the input information includes a plurality of items such as the material of the object, welding conditions, and partial welding paths, and each item has a plurality of input sub-items that are different from each other.

[0107] Furthermore, when the content of an item is represented by a numerical value, the input data for each item may be divided into multiple intervals by defining an input sub-item, and a representative value corresponding to each input sub-item may be defined as the input data. The representative value of an input sub-item may be, for example, a median value, an upper limit value, or a lower limit value within the input sub-item.

[0108] The range of input data does not need to be the same as the input information as the performance data. The database creation unit 55 inputs the input data for each input sub-item thus determined into the mathematical model created by the mathematical model generation unit 53 to obtain output data for each input sub-item.

[0109] Furthermore, the output data of the output information is the output value of the mathematical model corresponding to the input data. Here, the input data for each input sub-item can be represented by, for example, the median value of the data interval determined for that input sub-item. The database creation unit 55 then creates a database by storing the output data obtained by inputting the input data of each input sub-item into the mathematical model and the corresponding relationship between the output data and the input data for each input sub-item.

[0110] That is, the database creation unit 55 creates a database in which the output data for each input sub-item is accumulated by partitioning the range of each item of input data into a plurality of sections.

[0111] In this way, the input sub-items of each project are related to the characteristic values ​​of the object through mathematical models. Although it is possible to learn multiple mathematical models in combination as described above, it is preferable to integrate them into a single mathematical model based on specific welding conditions, welding track patterns, etc., and use this mathematical model as the basis for tuning the various parameters. Tuning in this context can include transfer learning, which allows the learning (learned model) in one area to be applied to other areas, thereby promoting efficient learning. This can reduce the amount of learning data and the amount of computation required.

[0112] Next, the element shapes when obtaining a partial welding trajectory and the welding trajectory for each element shape will be described together with a specific example of a shaped object.

[0113] Figure 10 This is an explanatory diagram showing a process of dividing the shape of a to-be-made object into a plurality of element shapes and obtaining a welding trajectory for each element shape.

[0114] Here, an example of a structure 65 is shown, comprising a main body 65A, a first protrusion 65B connected to one surface of the main body 65A, and a second protrusion 65C connected to the other surface of the main body 65A. If this structure 65 is divided into simple elemental shapes, the structure comprises a cylindrical first protrusion 65B, a cubic main body 65A, and a U-shaped second protrusion 65C. This division into elemental shapes can be performed manually or by pattern matching with pre-registered simple shapes.

[0115] A welding trajectory representing the order in which the deposited weld bead is formed is determined for each of the divided element shapes. In other words, a welding trajectory is determined for each divided element shape. The welding trajectory for each element shape is designed and determined sequentially each time the element shape is divided into element shapes. However, since element shapes are simple shapes, it is also possible to pre-register multiple simple welding trajectories (reference welding trajectories) in an element database and determine a welding trajectory corresponding to the element shape by referring to this element database.

[0116] For example, in the case of a cylindrical element, the cylindrical body is divided into multiple layers, and a reference welding trajectory is used to determine the formation pass (torch trajectory) of the deposited weld bead for each of the divided layers. By using this reference welding trajectory for the first protrusion 65B, the formation step, i.e., the welding trajectory B, when forming the first protrusion 65B with the deposited weld bead can be easily determined.

[0117] The same is true for the main body 65A and the second protrusion 65C. By searching and finding a reference welding trajectory of a similar shape from the element database, the welding trajectory A of the main body 65A and the welding trajectory C of the second protrusion 65C can be easily determined based on the found reference welding trajectory. In this way, even a complex shaped object can be divided into element shapes and viewed as an aggregate of simple shapes, thereby simplifying the shape planning.

[0118] Figure 11 This is a flowchart showing the steps for creating a shaping plan when the shape of a shaped object is decomposed into element shapes.

[0119] If the shape data of the object to be made is input into Figure 3 The shaping planning unit 57 of the shaping control device 13 shown in FIG31 decomposes the model created from the shape data into a plurality of element shapes (S32). Then, the shaping planning unit 57 searches a pre-prepared element database (not shown) to extract information such as the reference welding trajectory and welding conditions corresponding to each decomposed element shape (S33). The element database used here contains information on the reference welding trajectory and welding conditions set corresponding to the element shapes, and this information is pre-registered in the element database.

[0120] Welding trajectories are obtained by applying the extracted reference welding trajectories to the corresponding element shapes (S34), and a formation plan for the entire object is prepared in accordance with the welding conditions (S35).

[0121] The resulting shape plan becomes Figure 7 Therefore, the shaping plan of S12 is decomposed into element shapes to determine the welding trajectory and welding conditions for each element shape, and the mathematical model is generated and the database 61 is constructed in the same way as above to support Figure 4 The shape plan shown.

[0122] Furthermore, information on welding conditions can be easily collected from the forming device 11, the wire feed sensor 31, the shape sensor 32, and the drive signal of the welding robot 17. These values ​​can also be used in the shape of the feedback control model as needed.

[0123] <Second Database Structure Example>

[0124] Next, a description will be given of a case where intermediate output information is provided in addition to the input information and output information of the aforementioned database 61 .

[0125] Figure 12 This is an explanatory diagram showing the establishment of a relationship using a mathematical model of input information, intermediate output information, and output information.

[0126] Each of the input information items, including the object's material, welding conditions, and partial weld path, has multiple input sub-items. Intermediate output information is associated with each combination of input sub-items via a separate first mathematical model. Furthermore, each input sub-item in the intermediate output information is associated with each input sub-item in the output information via a second mathematical model. Here, information on the object's temperature history is used as an example of intermediate output information.

[0127] When the material of a structure, such as a filler metal, is heated according to welding conditions and melts and solidifies along a predetermined weld path, the temperature history of the resulting structure (weld bead) varies depending on the conditions described above. Consequently, the mechanical strength, metallurgical structure, and other properties of the resulting structure also vary depending on the conditions.

[0128] When estimating the characteristic values ​​of an object, even when it's difficult to directly infer the object's characteristics from each item (condition) of the input information, it's sometimes easier to infer the object's characteristics from the temperature history if the temperature history corresponding to each item is known. To this end, when establishing a relationship between the input information and the object's characteristics as output information, a two-stage relationship building process is performed: first, each item of the input information is associated with the temperature history as intermediate output information, and then, the temperature history is associated with the object's characteristic values ​​as output information. This allows for more accurate relationship building and estimation than when directly relating the input and output information.

[0129] Figure 13 This is a graph showing the temperature history at a specific location of the weld bead formed during forming. Figure 13 As shown, after the repeatedly stacked weld beads melt and solidify to form a weld bead, heat is again input through the weld bead stacked above them, repeating heating (and sometimes melting in adjacent layers) and cooling. Each peak in the temperature history decreases in temperature as the weld bead moves further away from the specific location as it moves above the layer above it.

[0130] If the melting point Tw of the weld bead is set to 1534°C, the melting point of iron (carbon steel), and the transformation point t (A1 transformation point of carbon steel) is set to 723°C, the material quality of the weld bead after solidification is roughly determined by the temperature history within the range above the transformation point Tt and below the melting point Tw. In other words, during laminate molding, repeated heating and cooling occur, but the temperature history within the aforementioned range Aw is the factor that affects the structure of the molded object. Therefore, by extracting the characteristic values ​​of the temperature history within the range Aw above the transformation point Tt and below the melting point Tw (the inspection temperature range), the properties of the molded object can be predicted.

[0131] For example, ignoring Figure 13 The peaks above the melting point Tw and below the transformation point Tt are selected from the multiple peaks shown. Next, the temperature of the lower-side maximum point Pk2 closest to the transformation point Tt and the temperature of the second higher-side maximum point Pk1 closest to the transformation point Tt are extracted from the peaks in the inspection temperature range Aw above the transformation point Tt and below the melting point Tw. The temperatures of these higher-side maximum points Pk1 and lower-side maximum points Pk2 are set as feature quantities of the temperature history record, i.e., intermediate output information.

[0132] Figure 14The following are graphs showing the differences in cooling characteristics when forming weld beads at different heat input levels. (A) shows the temperature change characteristics for a relatively high heat input level, and (B) shows the temperature change characteristics for a relatively low heat input level.

[0133] like Figure 14 As shown in (A), even if the heat input is increased from Qa to Qb, Qc, and Qd in the order of up to 350°C, the time until cooling is almost unchanged, and in this case is about 15 seconds (see Pend). Figure 14 As shown in (B), when the heat input is relatively low, cooling to approximately 300°C is achieved with a cooling time of approximately 15 seconds (see Pend). In other words, the higher the heat input, the slower the cooling rate, while the lower the heat input, the faster the cooling rate. Therefore, the cooling rate depends on the heat input, and knowing the temperature of the low-temperature maximum point Pk2 allows prediction of the weld bead microstructure. Furthermore, by combining the temperatures of the low-temperature maximum point Pk2 and the high-temperature maximum point Pk1 to predict the microstructure, prediction accuracy is improved compared to using only either temperature alone.

[0134] In this way, if the temperature history that determines the material of the weld bead can be determined from the aforementioned characteristic quantities, the material of the weld bead formed using this temperature history can be predicted with relatively high accuracy. To this end, the items in the intermediate processing information are set as factors that determine the material of the forming material, the input information and the intermediate output information are linked using a first mathematical model, and the intermediate output information and the output information are linked using a second mathematical model. This approach can be expected to more accurately link the input and output information, compared to directly linking them.

[0135] The temperature history can also be recorded by the temperature sensor 30 ( Figure 1 ) monitors the temperature of the object and obtains temperature data at a specific location. The temperature sensor 30 can work in conjunction with the shape sensor 32 to detect temperature. Specifically, the shape sensor 32 detects the shape of the object, while the temperature sensor 30 detects the temperature at a specific location on the object.

[0136] In addition, temperature simulation calculations can also be performed based on the type of welding filler metal or welding conditions.

[0137] An example of a basic formula used in temperature simulation is shown below.

[0138] [Mathematical formula 1]

[0139] t+Δt {H}= t {H}-Δt〔C〕〔K〕t {T}-Δt〔C〕 t {F}+Δt t {Q}…(1)

[0140] The basic formula (1) is a formula for heat transfer analysis based on the so-called explicit solution method FEM (Finite Element Method). The parameters of the basic formula (1) are as follows.

[0141] H: enthalpy

[0142] C: the inverse of the node volume

[0143] K: Thermal Conductivity Matrix

[0144] F: heat flux

[0145] Q: Volumetric heat

[0146] By using enthalpy as an unknown, nonlinear phenomena such as latent heat release can be accurately calculated. Furthermore, the heat input during welding is input into the volumetric heat generation or heat flux parameters.

[0147] In the three-dimensional heat conduction equation, basic equation (1), the heat input during forming (welding) can be applied to the weld area along with the welding speed. In addition, when the weld bead is short, the heat input can be applied to the entire weld bead.

[0148] <Other database structure examples>

[0149] Figure 15 This is an explanatory diagram showing how a plurality of databases are selectively used to establish a relationship between input information and output information.

[0150] In the first database structure example described above, input information and output information are related using a mathematical model I, and the mathematical model I constitutes the database DB1 (the aforementioned database 61).

[0151] In addition, the second database structure example uses mathematical model IIa to establish a relationship between input information and intermediate output information, and uses mathematical model IIb to establish a relationship between intermediate output information and output information. These mathematical models IIa and IIb constitute the database DB2 (the aforementioned database 61).

[0152] Then, the constructed database DB1 and database DB2 are compared, and the database whose output information is more correct relative to the input information is selected as Figure 4 The database 61 shown uses a comparison between the databases DB1 and DB2, for example, by using a pair of input information and output information (teaching data) with a known correspondence relationship to determine the correctness of the output with respect to the input.

[0153] According to this, by constructing a plurality of databases and selectively using a database with higher accuracy, the accuracy of prediction of characteristic values ​​of a formed object is improved, and creation of a more appropriate forming plan can be supported.

[0154] Thus, the present invention is not limited to the above-mentioned embodiments. Combining the various structures of the embodiments with each other and making changes and applications by those skilled in the art based on the description in the specification and known technologies are also designed aspects of the present invention and are included in the scope of protection.

[0155] As described above, the following matters are disclosed in this specification.

[0156] (1) A forming plan supporting method for supporting the creation of a forming plan that represents the material of the forming object, the welding conditions of the weld beads, and the welding trajectory when a forming object is manufactured by stacking weld beads formed by melting and solidifying welding filler metal supplied from a welding head into a desired shape, comprising the following steps: generating a mathematical model that establishes a relationship between input information including the material of the forming object, the welding conditions, and the welding trajectory, and output information of characteristic values ​​of the forming object when stacking forming is performed under the conditions of the items in the input information; using the mathematical model to create a forming plan that represents the input information A database of corresponding relationships with the output information; searching the database to obtain the conditions of each item of the material of the object, the welding conditions, and the welding track corresponding to the target characteristic values ​​of the object to be manufactured; and prompting the material of the object, the welding conditions, and the welding track corresponding to the obtained target characteristic values, wherein the items of the input information each have a plurality of input sub-items that are different from each other, and the output information has a plurality of individual characteristic values ​​corresponding to the input sub-items. In the process of generating the mathematical model, the input sub-items of the input information are respectively related to the individual characteristic values ​​using the mathematical model.

[0157] According to this molding plan support method, even in stacked molding where welding is complicated and the number of passes often becomes large, it is possible to support the creation of an appropriate molding plan by preparing a database in advance.

[0158] (2) A forming plan supporting method for supporting the preparation of a forming plan that respectively characterizes the material of the forming object, the welding conditions of the deposited weld and the welding track when forming a forming object by stacking the deposited weld formed by melting and solidifying the welding filler metal supplied from the welding head into a desired shape, comprising the following steps: respectively generating a first mathematical model that establishes a relationship between input information including the material of the forming object, the welding conditions and the welding track and intermediate output information including information on the temperature history of the forming object when stacking the forming under the conditions of the items of the input information, and a second mathematical model that establishes a relationship between the intermediate output information and output information including characteristic values ​​of the forming object; using the first mathematical model and the second mathematical model to prepare a forming plan that characterizes the relationship between the input information and the input information. a database of corresponding information; searching the database to obtain the temperature history records, the material of the object, the welding conditions and the welding track corresponding to the target characteristic values ​​of the object to be manufactured; and prompting the material of the object, the welding conditions and the welding track corresponding to the obtained target characteristic values, wherein the items of the input information respectively have a plurality of input sub-items that are different from each other, the intermediate output information has individual intermediate values ​​corresponding to the input sub-items, and the output information has a plurality of individual characteristic values ​​corresponding to the individual intermediate values. In the process of generating the first mathematical model and the second mathematical model, the input sub-items are respectively related to the individual intermediate values ​​using the first mathematical model, and the individual intermediate values ​​are respectively related to the individual characteristic values ​​using the second mathematical model.

[0159] This modeling planning support method uses temperature history records, a representative process characteristic of the modeled object, as intermediate output information, making it easier to correlate input information with properties such as the hardness of the modeled object. Furthermore, temperature history records can be used not only for actual modeling measurements but also for calculations using temperature simulations. This makes data addition simple and facilitates database construction.

[0160] (3) In the molding planning support method described in (1) or (2), the material information in the input information includes information on the type of welding filler metal.

[0161] According to this forming plan support method, since the viscosity of the deposited weld bead during melting varies depending on the type of welding filler metal, and the cross-sectional shape of the deposited weld bead tends to vary accordingly, by creating a mathematical model for each type of welding filler metal, it is possible to set welding conditions and trajectory plans suitable for each welding filler metal.

[0162] (4) Based on the forming plan support method described in any one of (1) to (3), the information on the welding conditions in the input information includes information on at least any one of the welding current, welding voltage, welding speed, the spacing width between adjacent welding tracks, the time between passes from a specific welding track among the plurality of welding tracks to other welding tracks, the aiming position of the welding head, the welding posture of the welding head, and the supply speed of the welding filler metal, or a combination thereof.

[0163] According to this modeling plan supporting method, since all the information can be monitored during modeling, data can be easily collected.

[0164] (5) Based on the forming planning support method described in any one of (1) to (4), the information of the welding track in the input information includes information on at least any one of the passes for forming the deposited weld bead, the number of passes, the order in which the deposited weld bead is formed, and the cross-sectional shape of the deposited weld bead.

[0165] According to this modeling plan supporting method, since all the information can be monitored during modeling, data can be easily collected.

[0166] (6) In the molding planning support method according to any one of (1) to (5), the welding track is a welding track corresponding to a portion of an element shape cut out of a portion of the overall shape of the molding.

[0167] This modeling planning support method allows for simple modeling by cutting the object's shape into several elemental shapes and planning welding conditions and weld paths for each elemental shape. By creating weld paths corresponding to each elemental shape in various variations, modeling can be performed without requiring complex processing, even for complex shapes.

[0168] (7) Based on the molding planning support method described in any one of (1) to (6), the output information includes information on at least one of an index representing the state of the metal structure of the molded object, hardness, and mechanical strength.

[0169] According to this molding plan supporting method, the database can be easily constructed by using the state of metal structure, hardness (Vickers hardness, etc.), and mechanical strength, which can be tested relatively simply and in a short time.

[0170] (8) Based on the shaping planning support method described in any one of (1) to (7), the mathematical model is a learned model that has machine-learned the relationship between the input information and the output information.

[0171] This modeling planning support method uses machine learning to construct a mathematical model, making it possible to supplement areas where experimental data is lacking, and as this data is added, prediction accuracy improves. Furthermore, because data corresponding to inputs and outputs can be collected using basic structures such as wall structures and block structures, machine learning data can be easily prepared.

[0172] (9) In the modeling planning support method according to any one of (1) to (8), the input range of the input information is limited to a range defined based on predetermined conditions.

[0173] According to this molding plan support method, by setting limits on the input range so as not to deviate from the recommended range for driving the molding device and the recommended conditions for using the welding filler metal, input of conditions that are likely to cause problems due to device failure or material problems can be avoided.

[0174] (10) A shaping plan supporting device for supporting the creation of shaping plans that respectively characterize the material of the shaping object, the welding conditions of the fused weld bead, and the welding track when a shaping object is manufactured by stacking the fused weld bead formed by melting and solidifying the welding filler metal supplied from the welding head into a desired shape, comprising: a mathematical model generating unit that generates a mathematical model that establishes a relationship between input information including the material of the shaping object, the welding conditions, and the welding track, and output information including characteristic values ​​of the shaping object when the shaping object is stacked under the conditions of the input information; and a database generating unit that uses the mathematical model to generate a shaping plan that characterizes the shaping object. A database of the correspondence between the input information and the output information; a retrieval unit, which searches the database to obtain the material of the object, the welding conditions and the welding track corresponding to the target characteristic values ​​of the object to be manufactured; and an output unit, which prompts the material of the object, the welding conditions and the welding track corresponding to the obtained target characteristic values, the items of the input information respectively have a plurality of input sub-items different from each other, the output information has a plurality of individual characteristic values ​​corresponding to the input sub-items, and the mathematical model generation unit establishes a relationship between the input sub-items of the input information and the individual characteristic values ​​using the mathematical model.

[0175] According to this molding plan support device, even in stacked molding where welding is complicated and the number of passes often becomes large, it is possible to support the creation of an appropriate molding plan by preparing a database in advance.

[0176] (11) A forming plan supporting device for supporting the preparation of a forming plan that respectively characterizes the material of the forming object, the welding conditions of the deposited weld and the welding track when forming the forming object by stacking the deposited weld formed by melting and solidifying the welding filler metal supplied from the welding head into a desired shape, comprising: a mathematical model generating unit that respectively generates a first mathematical model that establishes a relationship between input information containing the material of the forming object, the welding conditions and the welding track and intermediate output information containing information on the temperature history of the forming object when the forming is stacked under the conditions of the items of the input information, and a second mathematical model that establishes a relationship between the intermediate output information and output information containing characteristic values ​​of the forming object; and a database creating unit that uses the first mathematical model and the second mathematical model to create a forming plan that characterizes the material of the forming object, the welding conditions and the welding track. A database of the correspondence between the input information and the output information; a retrieval unit, which searches the database to obtain the temperature history record, the material of the shape, the welding conditions and the welding track corresponding to the target characteristic value of the shape to be manufactured; and an output unit, which prompts the material of the shape, the welding conditions and the welding track corresponding to the obtained target characteristic value, the items of the input information respectively have multiple input sub-items different from each other, the intermediate output information has individual intermediate values ​​corresponding to the input sub-items, and the output information has multiple individual characteristic values ​​corresponding to the individual intermediate values, the mathematical model generation unit establishes a relationship between the input sub-items and the individual intermediate values ​​using the first mathematical model, and establishes a relationship between the individual intermediate values ​​and the individual characteristic values ​​using the second mathematical model.

[0177] This molding planning support device processes temperature history records, a representative process characteristic of the molded object, as intermediate output information, making it easy to correlate input information with properties such as the object's hardness. Furthermore, temperature history records can be used not only for actual mold measurements but also for calculations using temperature simulations. This makes data addition simple and facilitates database construction.

[0178] In addition, this application is based on the Japanese patent application (Japanese Patent Application No. 2020-123860) filed on July 20, 2020, the contents of which are incorporated herein by reference.

[0179] Description of Reference Signs

[0180] 11 Shaping Device

[0181] 13Shaping control device

[0182] 15 welding torch

[0183] 17 Welding Robot

[0184] 21Robot control device

[0185] 23 Welding filler metal supply unit

[0186] 25 welding power supply

[0187] 27 chassis

[0188] 29 reels

[0189] 30 temperature sensors

[0190] 31 Wire feed sensor

[0191] 32 shape sensors

[0192] 33 input and output interfaces

[0193] 35 Storage Department

[0194] 37 Operation Panel

[0195] 41CPU

[0196] 43 Storage Department

[0197] 45 input and output interfaces

[0198] 47 Input unit

[0199] 49 output unit

[0200] 51 Basic Information Form

[0201] 53 Mathematical Model Generation Department

[0202] 55 Database Creation Department

[0203] 57 Modeling Planning Department

[0204] 59 Search Department

[0205] 61 Database

[0206] 63 Initial Database

[0207] 65, 64A shaped objects

[0208] 65A main body (element shape)

[0209] 65B first protrusion (element shape)

[0210] 65C second protrusion (elementary shape)

Claims

1. A method for supporting a formation plan, wherein, when manufacturing a formed object by laminating a deposited weld bead formed by melting and solidifying a welding filler metal supplied from a welding head into a desired shape, the method supports the creation of a formation plan that specifies the material of the formed object, welding conditions of the deposited weld bead, and a welding trajectory. The shaping plan support method is characterized by comprising the following steps: generating a mathematical model that relates input information including items such as the material of the object, the welding conditions, and the welding trajectory, and output information including characteristic values ​​of the object when lamination is performed under the conditions of the items in the input information; Creating a shaping plan based on the input information of the shape; The characteristic values ​​of the object are obtained by the following method: The forming control device predicts the characteristics of the object to be formed in layers according to the prepared forming plan with reference to an initial database in which the relationship between the forming plan and the characteristic values ​​is pre-registered, and The shaping control device drives the robot control device according to the prepared shaping plan, and the shaping control device causes the shaping device to stack and shape the object. Using the mathematical model to create a database representing the corresponding relationship between the input information and the output information; Searching the database to obtain the material of the object, the welding conditions, and the conditions of each item of the welding track corresponding to the target characteristic value of the object; and The output unit presents the material of the object, the welding conditions, and the welding trajectory corresponding to the obtained target characteristic value. Each of the items of input information has a plurality of different input sub-items, and the input sub-items are welding filler metal, welding current, welding voltage, welding speed, element shape, pass, and / or number of passes for each item. The output information has a plurality of individual characteristic values ​​corresponding to the input sub-items, In the step of generating the mathematical model, the input sub-items of the input information are associated with the individual characteristic values ​​using the mathematical model.

2. A shaping plan support method for supporting the creation of a shaping plan that includes the material of the object, welding conditions for the weld bead, and a welding trajectory when manufacturing an object by stacking weld beads formed by melting and solidifying a welding filler metal supplied from a welding head into a desired shape. The shaping plan support method is characterized by comprising the following steps: generating a first mathematical model for establishing a relationship between input information including items such as the material of the object, the welding conditions, and the welding trajectory and intermediate output information including information on the temperature history of the object when laminating the object under the conditions of the items in the input information, and a second mathematical model for establishing a relationship between the intermediate output information and output information including characteristic values ​​of the object; Creating a shaping plan based on the input information of the shape; The characteristic values ​​of the object are obtained by the following method: The forming control device predicts the characteristics of the object to be formed in layers according to the prepared forming plan with reference to an initial database in which the relationship between the forming plan and the characteristic values ​​is pre-registered, and The shaping control device drives the robot control device according to the prepared shaping plan, and the shaping control device causes the shaping device to stack and shape the object. Creating a database representing the correspondence between the input information and the output information using the first mathematical model and the second mathematical model; Searching the database to obtain the temperature history corresponding to the target characteristic value of the object, the material of the object, the welding conditions, and the welding trajectory; and The output unit presents the material of the object, the welding conditions, and the welding trajectory corresponding to the obtained target characteristic value. Each of the items of input information has a plurality of different input sub-items, and the input sub-items are welding filler metal, welding current, welding voltage, welding speed, element shape, pass, and / or number of passes for each item. The intermediate output information has individual intermediate values ​​corresponding to the input sub-items, The output information has a plurality of individual characteristic values ​​corresponding to the individual intermediate values, In the steps of generating the first mathematical model and the second mathematical model, the first mathematical model is used to establish a relationship between the input sub-items and the individual intermediate values, and the second mathematical model is used to establish a relationship between the individual intermediate values ​​and the individual characteristic values.

3. The modeling planning support method according to claim 1, wherein: The material information in the input information includes information on the type of welding filler metal.

4. The modeling planning support method according to claim 2, wherein: The material information in the input information includes information on the type of welding filler metal.

5. The modeling planning support method according to claim 1, wherein: The information on the welding conditions in the input information includes information on at least any one of the welding current, welding voltage, welding speed, a pitch width between adjacent welding tracks, an inter-pass time from a specific welding track among the plurality of welding tracks to another welding track, an aiming position of the welding head, a welding posture of the welding head, and a feed rate of the welding filler metal, or a combination thereof, when forming the deposited weld bead.

6. The modeling planning support method according to claim 2, wherein: The information on the welding conditions in the input information includes information on at least any one of the welding current, welding voltage, welding speed, a pitch width between adjacent welding tracks, an inter-pass time from a specific welding track among the plurality of welding tracks to another welding track, an aiming position of the welding head, a welding posture of the welding head, and a feed rate of the welding filler metal, or a combination thereof, when forming the deposited weld bead.

7. The modeling planning support method according to claim 3, wherein: The information on the welding conditions in the input information includes information on at least any one of the welding current, welding voltage, welding speed, a pitch width between adjacent welding tracks, an inter-pass time from a specific welding track among the plurality of welding tracks to another welding track, an aiming position of the welding head, a welding posture of the welding head, and a feed rate of the welding filler metal, or a combination thereof, when forming the deposited weld bead.

8. The modeling planning support method according to claim 4, wherein: The information on the welding conditions in the input information includes information on at least any one of the welding current, welding voltage, welding speed, a pitch width between adjacent welding tracks, an inter-pass time from a specific welding track among the plurality of welding tracks to another welding track, an aiming position of the welding head, a welding posture of the welding head, and a feed rate of the welding filler metal, or a combination thereof, when forming the deposited weld bead.

9. The shaping plan supporting method according to any one of claims 1 to 8, wherein: The information on the welding track in the input information includes at least any one of the number of passes for forming the weld bead, the number of passes, the order in which the weld bead is formed, and the cross-sectional shape of the weld bead.

10. The shaping plan supporting method according to any one of claims 1 to 8, wherein: The welding track is a welding track of a portion corresponding to an element shape cut out of a portion of the overall shape of the object.

11. The shaping plan supporting method according to claim 9, wherein: The welding track is a welding track of a portion corresponding to an element shape cut out of a portion of the overall shape of the object.

12. The shaping plan supporting method according to any one of claims 1 to 8, wherein: The output information includes information on at least one of an index representing a state of a metal structure of the object, hardness of the object, and mechanical strength of the object.

13. The shaping plan supporting method according to claim 9, wherein: The output information includes information on at least one of an index representing a state of a metal structure of the object, hardness of the object, and mechanical strength of the object.

14. The shaping plan supporting method according to claim 10, wherein: The output information includes information on at least one of an index representing a state of a metal structure of the object, hardness of the object, and mechanical strength of the object.

15. The shaping plan supporting method according to claim 11, wherein: The output information includes information on at least one of an index representing a state of a metal structure of the object, hardness of the object, and mechanical strength of the object.

16. The shaping plan supporting method according to any one of claims 1, 3, 5 and 7, The mathematical model is a learned model that has been machine-learned to understand the relationship between the input information and the output information.

17. The shaping plan supporting method according to any one of claims 2, 4, 6 and 8, The first mathematical model is a learned model that has learned the relationship between the input information and the intermediate output information. The second mathematical model is a learned model that has been machine-learned to learn the relationship between the intermediate output information and the output information.

18. The modeling planning support method according to any one of claims 1 to 8, An input range of the input information is limited to a range defined based on a predetermined condition.

19. A shaping plan support device for supporting the creation of a shaping plan that represents the material of the object, welding conditions of the weld bead, and welding trajectory when manufacturing an object by stacking weld beads formed by melting and solidifying welding filler metal supplied from a welding head into a desired shape. The aforementioned shaping plan support device is characterized by comprising: a mathematical model generating unit for generating a mathematical model that relates input information including items such as the material of the object, the welding conditions, and the welding trajectory, and output information including characteristic values ​​of the object when laminating the object under the conditions of the input information; A shaping control device for generating a shaping plan based on the input information of the shaping object; The characteristic values ​​of the object are obtained by the following method: The forming control device predicts the characteristics of the object to be formed in layers according to the prepared forming plan with reference to an initial database in which the relationship between the forming plan and the characteristic values ​​is pre-registered, and The shaping control device drives the robot control device according to the prepared shaping plan, and the shaping control device causes the shaping device to stack and shape the object. a database creation unit that uses the mathematical model to create a database representing the correspondence between the input information and the output information; a search unit that searches the database to obtain the material of the object, the welding conditions, and the welding trajectory corresponding to the target characteristic value of the object; and an output unit for presenting the material of the object, the welding conditions, and the welding trajectory corresponding to the obtained target characteristic value; Each of the items of input information has a plurality of different input sub-items, and the input sub-items are welding filler metal, welding current, welding voltage, welding speed, element shape, pass, and / or number of passes for each item. The output information has a plurality of individual characteristic values ​​corresponding to the input sub-items, The mathematical model generation unit establishes a relationship between the input sub-items of the input information and the individual characteristic values ​​using the mathematical model.

20. A shaping plan support device for supporting the creation of a shaping plan that represents the material of the object, welding conditions of the weld bead, and welding trajectory when manufacturing an object by stacking weld beads formed by melting and solidifying welding filler metal supplied from a welding head into a desired shape. The aforementioned shaping plan support device is characterized by comprising: a mathematical model generating unit for generating a first mathematical model for establishing a relationship between input information including items such as the material of the object, the welding conditions, and the welding trajectory and intermediate output information including information on the temperature history of the object when laminating and forming the object under the conditions of the items in the input information, and a second mathematical model for establishing a relationship between the intermediate output information and output information including characteristic values ​​of the object; A shaping control device for generating a shaping plan based on the input information of the shaping object; The characteristic values ​​of the object are obtained by the following method: The forming control device predicts the characteristics of the object to be formed in layers according to the prepared forming plan with reference to an initial database in which the relationship between the forming plan and the characteristic values ​​is pre-registered, and The shaping control device drives the robot control device according to the prepared shaping plan, and the shaping control device causes the shaping device to stack and shape the object. a database creation unit that uses the first mathematical model and the second mathematical model to create a database representing the correspondence between the input information and the output information; a search unit that searches the database to obtain the temperature history corresponding to the target characteristic value of the object, the material of the object, the welding conditions, and the welding trajectory; and an output unit for presenting the material of the object, the welding conditions, and the welding trajectory corresponding to the obtained target characteristic value; Each of the items of input information has a plurality of different input sub-items, and the input sub-items are welding filler metal, welding current, welding voltage, welding speed, element shape, pass, and / or number of passes for each item. The intermediate output information has individual intermediate values ​​corresponding to the input sub-items, The output information has a plurality of individual characteristic values ​​corresponding to the individual intermediate values, The mathematical model generation unit establishes a relationship between each of the input sub-items and the individual intermediate values ​​using the first mathematical model, and establishes a relationship between each of the individual intermediate values ​​and the individual characteristic values ​​using the second mathematical model.

Images