Lamination planning apparatus, lamination molding system, and lamination planning method

Abstract

To stabilize the shape of a weld bead when shaping an object by laminating the weld bead.SOLUTION: The control device (lamination planning device) drafts a lamination plan for building an object by laminating weld beads. The control device includes a shape data acquisition unit that acquires three dimensional shape data of the object, and a path planning unit that plans a path that is a movement path of the melting unit for shaping the object. The path planning unit sets paths so as to minimize the number of paths based on the shape data.SELECTED DRAWING: Figure 9

Description

[Technical Field]

[0001] The present disclosure relates to a stacking planning device, an additive manufacturing system, and a stacking planning method for an object to be manufactured by stacking weld beads. [Background technology]

[0002] For example, Japanese Patent No. 6997044 (Patent Document 1) discloses a lamination planning method for an object to be manufactured by laminating weld beads formed by melting and solidifying a filler metal. This lamination planning method sets the lamination direction of the weld beads using three-dimensional shape data of the object to be manufactured, places the base material at a neutral position in the lamination direction that minimizes thermal shrinkage of the entire object, and creates a lamination plan to build the object by laminating weld beads on both the front and back sides of the base material. As a result, in the object manufactured according to the lamination plan, thermal shrinkage is distributed in a balanced manner to both the front and back regions of the base material, thereby suppressing uneven thermal shrinkage of the entire object. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6997044 Summary of the Invention [Problem to be solved by the invention]

[0004] When forming an object by stacking weld beads, it is important to stabilize the shape of the weld bead in the stacking direction, but the shape of the start (arc start) and end (arc end) of the weld bead in the stacking direction tends to be unstable. Specifically, the size of the start of the weld bead tends to increase in the stacking direction, while the size of the start of the weld bead tends to decrease in the stacking direction. However, the stacking planning method disclosed in Japanese Patent No. 6997044 (Patent Document 1) does not take into consideration the instability of the shape of the start and end of the weld bead.

[0005] The present disclosure has been made to solve the above-mentioned problems, and its purpose is to stabilize the shape of a weld bead when forming an object by stacking weld beads. [Means for solving the problem]

[0006] A stacking planning device according to the present disclosure creates a stacking plan for building an object by stacking weld beads, and includes an acquisition unit that acquires three-dimensional shape data of the object, and a path planning unit that plans paths that are movement routes of molten parts to build the object. The path planning unit sets paths based on the shape data so as to minimize the number of paths.

[0007] An additive manufacturing system according to the present disclosure includes the above-described stacking planning device and a manufacturing device that manufactures an object by stacking weld beads in accordance with the stacking plan formulated by the stacking planning device.

[0008] A stacking planning method according to the present disclosure is a stacking planning method for building an object by stacking weld beads, and includes a step of acquiring three-dimensional shape data of the object and a step of planning paths that are movement paths of molten parts to build the object. The step of planning paths includes a step of setting paths based on the shape data so as to minimize the number of paths. [Effects of the Invention]

[0009] According to the present disclosure, when forming an object by stacking weld beads, the shape of the weld beads can be stabilized. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating an example of a configuration of an additive manufacturing system including a stacking planning device. [Figure 2] 1A and 1B are diagrams illustrating an example of a top view and a side view of an object formed in a rectangular shape on a base material. [Figure 3]FIG. 2 is a block diagram schematically showing functions related to stacking planning of the control device. [Figure 4] FIG. 10 is a diagram schematically illustrating an example of a method for creating a graph showing a layered region of an object. [Figure 5] 10A and 10B are diagrams illustrating an example of path setting when a graph can be drawn in one stroke. [Figure 6] 10A and 10B are diagrams illustrating an example of setting a path when a graph cannot be drawn in one stroke. [Figure 7] FIG. 10 is a diagram illustrating a process of generating path candidates through a search process. [Figure 8] FIG. 8 is a diagram showing an example of score setting for path candidate A shown in FIG. 7. [Figure 9] 10 is a flowchart illustrating an example of a processing procedure executed by a path planning unit when setting a path. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are designated by the same reference numerals, and description thereof will not be repeated.

[0012] 1 is a diagram schematically illustrating an example of the configuration of an additive manufacturing system 1 equipped with a layering planning device according to this embodiment. The additive manufacturing system 1 is configured to manufacture an object 100 by layering weld beads 70. The additive manufacturing system 1 includes a welding power supply 10, a welding torch 20, a robot arm 40, a wire feeder 50, and a control device 30.

[0013] Robot arm 40 is a multi-joint arm, for example a six-axis multi-joint arm. Welding torch 20 is fixed to the tip of robot arm 40. Robot arm 40 functions as a moving device that moves welding torch 20 at a set welding speed. As robot arm 40 moves welding torch 20, weld beads 70 are formed by melting and fixing consumable electrode wire 51 (hereinafter also simply referred to as "wire 51"), and are laid up. Robot arm 40 is controlled by control device 30 so that object 100 is formed by laying up weld beads 70.

[0014] A welding current is supplied to wire 51 from a power feed tip (not shown) located inside nozzle 21 of welding torch 20. Wire 51 is melted by arc 22 generated between wire 51 and base metal 80 or already-laid weld bead 70. Arc 22 melts base metal 80 or already-laid weld bead 70 simultaneously with wire 51, forming molten zone (molten pool) 23, which cools and solidifies to form weld bead 70. Hereinafter, the lamination direction of weld bead 70 (the normal direction of base metal 80) will be referred to as the Z-axis direction, a direction perpendicular to the Z-axis direction (the longitudinal direction of base metal 80 in the example shown in FIG. 1) will be referred to as the X-axis direction, and a direction perpendicular to the Z-axis direction and the X-axis direction will be referred to as the Y-axis direction.

[0015] Welding power supply 10 supplies a welding current to wire 51. The welding current may be either direct current or alternating current.

[0016] Wire feeder 50 includes a roller and a motor (not shown). Wire feeder 50 feeds wire 51 to welding torch 20 by driving the motor to rotate the roller.

[0017] The control device 30 includes a CPU (Central Processing Unit), a memory, and an input / output port for inputting and outputting various signals (none of which are shown). These elements are connected via a bus.

[0018] Control device 30 controls robot arm 40, wire feeder 50, and welding power supply 10 in accordance with a stacking plan for building object 100. As a result, weld beads 70 are stacked in accordance with the stacking plan to build object 100. Control for realizing the functions of control device 30 may be processed by software or dedicated hardware (electronic circuitry). Note that while FIG. 1 illustrates an example of a configuration in which a command to wire feeder 50 is output from control device 30, a configuration in which a command to wire feeder 50 is output from welding power supply 10 rather than from control device 30 may also be used.

[0019] [Lamination planning] Control device 30 according to the present embodiment has a function of planning a lamination plan in addition to a function of laminating weld beads 70 according to the lamination plan. Specifically, control device 30 acquires three-dimensional shape data of target object 100 from the outside, and sets a movement path (hereinafter also referred to as a "path") of molten portion 23 for forming target object 100 based on the acquired shape data.

[0020] In order to form object 100 according to the lamination plan, it is desirable to stabilize the shape of each layer of weld bead 70. However, the shape of the start portion (arc start portion) and end portion (arc end portion) of weld bead 70 in the lamination direction is prone to become unstable.

[0021] Fig. 2 shows an example of a bird's-eye view and a side view of an object formed in a rectangular shape on a base material. The object shown in Fig. 2 was formed by performing the operation of moving the molten zone in the X-axis direction to form linear weld beads extending in the X-axis direction multiple times while moving the start position of each weld bead by the width of the weld bead from the negative direction to the positive direction in the Y-axis direction.

[0022] 2, the start portion (arc start portion) of the weld bead tends to increase in size in the stacking direction (Z-axis direction), while the end portion (arc end portion) of the weld bead tends to decrease in size in the stacking direction. Therefore, the greater the number of weld beads 70, the greater the number of start portions and end portions, which makes the shape of each layer of weld bead 70 less stable.

[0023] In view of this, when creating a stacking plan for the object 100, the control device 30 according to this embodiment uses graph theory to set paths so as to minimize the number of paths.

[0024] 3 is a block diagram showing functions related to stacking planning of the control device 30. The control device 30 has a shape data acquisition unit 31, a path planning unit 32, and a stacking command unit 33.

[0025] The shape data acquisition unit 31 communicates with the outside via wire or wirelessly to acquire from the outside three-dimensional shape data of the target object 100 to be manufactured.

[0026] The path planning unit 32 plans a path for forming the object 100 based on shape data of the object 100. The path planning unit 32 includes a path candidate generating unit 32a, a score setting unit 32b, and a path setting unit 32c.

[0027] The path candidate generating unit 32a generates path candidates capable of forming the object 100, based on the shape data of the object 100. For example, the path candidate generating unit 32a generates path candidates using the following method.

[0028] First, path candidate generating unit 32a identifies the layered region of object 100 based on the shape data of object 100. Then, path candidate generating unit 32a defines the layered region of object 100 as a collection of multiple cells separated into a mesh shape by the width of weld bead 70, and creates a graph connecting the centers of the multiple cells.

[0029] Fig. 4 is a diagram schematically showing an example of a method for creating a graph showing the layered region of the object 100. Note that, although Fig. 4 simply shows the layered region of the object 100 in two dimensions, the actual layered region of the object 100 is three-dimensional.

[0030] Path candidate generation unit 32a grasps the layered region of object 100 as shown in Fig. 4(A) based on the shape data of object 100. Next, as shown in Fig. 4(B), path candidate generation unit 32a defines the layered region of object 100 as a collection of multiple cells separated into a mesh by the width of weld bead 70, and creates a graph in which the center of each of the multiple cells is a node and adjacent nodes are connected by edges.

[0031] Once a graph is created, the path candidate generation unit 32a performs a division determination to determine whether or not to divide the created graph. Specifically, the path candidate generation unit 32a checks whether or not the created graph can be drawn in one stroke. Whether or not the graph can be drawn in one stroke can be determined by known graph theory. Specifically, the number of edges connected to each node in the graph is calculated as the degree, and if the number of nodes with an odd degree is 0 (Eulerian graph) or 2 (quasi-Eulerian graph), it can be determined that the graph can be drawn in one stroke.

[0032] If the created graph can be drawn in one stroke, the path candidate generation unit 32a leaves the created graph as is without dividing it. On the other hand, if the created graph cannot be drawn in one stroke, the path candidate generation unit 32a divides the created graph into the minimum number of subgraphs, each of which can be drawn in one stroke. By setting paths to routes that can be drawn in one stroke in the graph or each subgraph thus set, it is possible to minimize the number of paths.

[0033] Furthermore, when there are multiple division patterns each having the same number of subgraphs, the path candidate generation unit 32a calculates the number of nodes in each of the multiple subgraphs for each division pattern and selects the division pattern that maximizes the minimum value of the calculated number of nodes. This makes it possible to prevent the length of the path after division from becoming extremely short.

[0034] Once the graph or subgraph is determined, the path candidate generation unit 32a performs a search process to generate path candidates. In the search process, a path that can pass through all the nodes in the graph or subgraph by moving the molten part at a constant movement speed (welding speed) is generated as a path candidate. Note that, if there are multiple routes that can be drawn in one stroke through the graph or each subgraph, the path candidate generation unit 32a generates multiple path candidates in the graph or each subgraph by repeating the search process multiple times while changing the movement start position and movement route of the molten part.

[0035] Next, the score setting unit 32b will be described. When multiple path candidates are generated in a graph or each subgraph, the score setting unit 32b sets a score for each of the multiple generated path candidates. The score is an index representing the amount of heat of cells in the base part (hereinafter also referred to as "contact cells") that the molten part comes into contact with in the process of moving the molten part along each path candidate.

[0036] The score setting unit 32b sets a cost for the cells of the base portion that are not in contact with the molten part (hereinafter also referred to as "non-contact cells") each time the molten part is moved to an adjacent cell along each path candidate. The cost is an index representing the temperature of the non-contact cells.

[0037] Specifically, the score setting unit 32b reduces the cost of the non-contact cell by 1 each time the molten part is moved to the next adjacent cell along each path candidate (i.e., each time the time for moving the molten part to the next cell has elapsed). This reduces the cost of the non-contact cell in response to the temperature decrease due to heat dissipation over time.

[0038] Furthermore, the score setting unit 32b increases the cost of the non-contact cells that are adjacent to the contact cell by 5. This is to increase the cost of the cells that are adjacent to the contact cell in response to the temperature rise of the cells that are adjacent to the contact cell due to the heat that is transmitted from the contact cell that is in contact with the molten part to the surroundings.

[0039] Then, the score setting unit 32b stores the cost of the abutting cell each time the molten part is moved to the next cell along each path candidate, and sets the total value of the multiple costs stored until the molten part has passed through all the cells of each path candidate as the score of each path candidate.

[0040] Next, the path setting unit 32c will be described. The path setting unit 32c sets the path candidate having the smallest score set by the score setting unit 32b as the official path from among the plurality of path candidates generated by the path candidate generating unit 32a.

[0041] Next, we will explain stacking command unit 33. Stacking command unit 33 controls robot arm 40, wire feeder 50, and welding power supply 10 so as to move fusion zone 23 along the path set by path setting unit 32c and stack weld beads 70. In this way, object 100 planned in the stacking plan is formed by stacking weld beads 70.

[0042] Figure 5 shows an example of setting a path when the graph can be drawn in one stroke. Note that for the sake of simplicity, Figure 5 shows a two-dimensional graph with three nodes vertically and four nodes horizontally, but the actual graph is a more complex three-dimensional graph.

[0043] As shown in Figure 5, if the graph can be drawn in one stroke, the graph is not divided but remains as is, and the path that allows the graph to be drawn in one stroke is set as the path. This sets the number of paths to the minimum number of "1." Therefore, when building an object by layering weld beads, the number of start and end points of the path (weld bead) can be kept to the minimum number of "1." As a result, the shape of the weld bead can be stabilized.

[0044] Since the graph shown in FIG. 5 has multiple paths that can be drawn in one stroke, the above-mentioned search process generates multiple path candidates, each of which can be drawn in one stroke, and a score is set for each path candidate. FIG. 5 illustrates an example in which three path candidates A to C are generated and the scores for the three path candidates A to C are set to "65," "63," and "66," respectively. In this case, path candidate B, which has the smallest score, is set as the official path. As a result, of the multiple path candidates that can be drawn in one stroke, the path candidate that has the smallest heat quantity in the base portion that contacts the molten part is set as the official path. As a result, it is possible to easily suppress burn-through due to the heat of the molten part when building an object by layering weld beads.

[0045] Figure 6 shows an example of setting a path when the graph cannot be drawn in one stroke. Note that, for the sake of simplicity, a two-dimensional graph is shown in Figure 6, but the actual graph is a more complex three-dimensional graph.

[0046] As shown in Figure 6, if a graph cannot be drawn in one stroke, the graph is divided into the minimum number of subgraphs, each of which can be drawn in one stroke. The graph shown in Figure 6 can be divided into two subgraphs, each of which can be drawn in one stroke, so the graph is divided into two subgraphs. As a result, the number of passes is set to "2," which is the division number. Therefore, when building an object by layering weld beads, the number of start and end parts of the weld beads can be kept to the minimum necessary. As a result, the shape of the weld bead can be stabilized.

[0047] Since there are multiple division patterns for the graph shown in FIG. 6 that can be divided into two, the number of nodes in multiple subgraphs is calculated for each division pattern, and the minimum of the calculated numbers of nodes is calculated as the minimum number of nodes for each division pattern. The division pattern with the largest minimum number of nodes is then selected as the official division pattern. FIG. 6 illustrates an example in which the minimum numbers of nodes for four division patterns A to D are "2," "4," "7," and "8," respectively. In this case, division pattern D with the largest minimum number of nodes is selected as the official division pattern. This makes it possible to prevent the length of the shortest path after division from becoming extremely short.

[0048] Figure 7 is a diagram that shows a schematic diagram of the process of generating path candidates through search processing. Note that for the sake of simplicity, Figure 7 shows a two-dimensional graph with three nodes vertically and four nodes horizontally, but the actual graph is a more complex three-dimensional graph.

[0049] First, an initial graph is created at time t0, just before welding begins. Next, one of the multiple nodes contained in the graph is selected as the starting point. The starting point is the position of the molten part at time t1, when lamination begins. Below, the cell containing the starting point is also referred to as the "starting cell." If the initial graph is an Eulerian graph, it can be drawn in one stroke starting from any node, so any node is selected as the starting point. If the initial graph is a quasi-Eulerian graph, it can be drawn in one stroke starting from either node with an odd degree, so either node with an odd degree is selected as the starting point. The starting point is saved in a list of path candidates.

[0050] Once the starting point is selected, the node at the next time t2 is selected from among the nodes adjacent to the starting point, and the position of the molten part is moved to the next node. At this time, the node after the movement is registered in a list of path candidates, and the previous node is deleted from the graph. This series of processes of moving the molten part, registering it in the list, and deleting the node from the graph is repeated. A path that allows the molten part to pass through all nodes is a path that can be drawn in one stroke, and becomes a path candidate. Note that if the molten part cannot move to the next node, a path that can be drawn in one stroke is searched for by retracing the path and proceeding to another node until another node can be selected (so-called backtracking). Note that a path that can be drawn in one stroke may also be searched for by solving an optimization problem (such as the traveling salesman problem).

[0051] FIG. 7 shows an example in which a path candidate A is generated that allows the molten part to pass through all nodes by moving the molten part at a constant moving speed (welding speed) from time t1 to time t12.

[0052] The above search process is repeated multiple times while changing the starting cell and route, thereby generating at least one path candidate.

[0053] Fig. 8 is a diagram showing an example of score setting for path candidate A shown in Fig. 7. Fig. 8 shows an example in which, among the cells in the base part, a cell located vertically or horizontally of the adjacent cell is determined to be a cell adjacent to the adjacent cell.

[0054] As shown in FIG. 8, first, the initial cost ("0" in the example shown in FIG. 8) of all cells in the base portion at time t0 immediately before the start of welding is set. After that, each time the molten part is moved to the next cell along path candidate A, the cost of non-contact cells (cells other than the abutting cells) is reduced by "1". Note that FIG. 8 shows an example in which the lower limit of the cost is set to "0". Furthermore, the cost of non-contact cells that are adjacent to the abutting cells (cells located vertically or horizontally of the abutting cells) is increased by "5".

[0055] Then, each time the fusion zone is moved to the next cell along path candidate A, the cost of the adjacent cell at that time is stored. The total value of the multiple costs stored until the fusion zone has passed through all the cells of path candidate A is set as the score of path candidate A.

[0056] 8 shows an example in which a cost is assigned only to the base portion that exists at time t0 immediately before the start of welding, for the sake of simplicity, and as a result, the final score of path candidate A is "65." However, in reality, the cells that the molten part has passed through after time t1 (the cells that make up path candidate A) also become base portions and are assigned costs, so the actual score of path candidate A is expected to be higher than "65."

[0057] FIG. 9 is a flowchart showing an example of a processing procedure executed by the path planning unit 32 when setting up a path.

[0058] First, as shown in FIG. 4 above, the path planning unit 32 grasps the layered area of ​​the object based on the shape data of the object, and divides the grasped layered area into a mesh by the weld bead width to create a graph (step S10).

[0059] Next, the path planning unit 32 determines whether the graph can be drawn in one stroke by using graph theory or by solving an optimization problem, as shown in FIG. 5 above (step S20).

[0060] If the graph can be drawn in one stroke (YES in step S20), the path planning unit 32 sets the number of paths to the minimum number "1" (step S21). On the other hand, if the graph cannot be drawn in one stroke (NO in step S20), the path planning unit 32 divides the graph into the minimum number of subgraphs, each of which can be drawn in one stroke (step S22), and sets the number of paths to the division number (step S23).

[0061] Next, the path planning unit 32 performs the above-mentioned search process to generate path candidates (step S30).

[0062] Next, the path planning unit 32 determines whether multiple path candidates exist in the graph or each subgraph (step S31). If only one path candidate exists in the graph or each subgraph (NO in step S31), the path planning unit 32 sets the generated path candidate as a formal path (step S32). If multiple path candidates exist in the graph or each subgraph (YES in step S31), the path planning unit 32 sets scores for the multiple path candidates and sets the path candidate with the smallest set score as the formal path (step S33).

[0063] As described above, control device 30 according to this embodiment includes shape data acquisition unit 31 that acquires three-dimensional shape data of an object, and path planning unit 32 that plans paths, which are movement paths of the molten parts for forming the object. Path planning unit 32 sets paths based on the shape data of the object so as to minimize the number of paths. This minimizes the number of paths when forming the object by stacking weld beads, and can keep the number of start and end points of paths (weld beads) to a necessary minimum. As a result, the shape of the weld bead can be stabilized.

[0064] Furthermore, the path planning unit 32 defines the layered area of ​​the object as a collection of multiple cells divided into a mesh by the width of the weld bead, creates a graph connecting the centers of the multiple cells, determines whether the graph can be drawn in one stroke using graph theory or by solving an optimization problem, and if the graph can be drawn in one stroke, sets the route that allows the graph to be drawn in one stroke as a path. This makes it possible to minimize the number of paths to "1".

[0065] Furthermore, if the graph cannot be drawn in one stroke, the path planning unit 32 divides the graph into the minimum number of subgraphs that can each be drawn in one stroke, and sets a path that can draw each subgraph in one stroke, thereby minimizing the number of paths.

[0066] Furthermore, when there are multiple path candidates that can draw the graph or subgraph in one stroke, the path planning unit 32 assigns a score to each path candidate, which is an index representing the heat amount of the base portion that the molten part comes into contact with as the molten part moves along each path candidate, and sets the path candidate with the smallest score as the path. When there are multiple path candidates, the path planning unit 32 stores the cost of the abutting cells that the molten part comes into contact with among the cells in the base portion, and sets the cost of the non-abutting cells that the molten part does not come into contact with among the cells in the base portion, and sets the score of each path candidate as the total or maximum value of the multiple costs stored until the molten part moves through all the cells in each path candidate. This makes it possible to set a path that is less likely to cause burn-through due to the heat of the molten part, taking into account the temperature of the base portion that changes from moment to moment as the molten part moves.

[0067] Furthermore, each time the molten part is moved to the next cell along each path candidate, the path planning unit 32 decreases the cost of the non-contact cells and increases the cost of the non-contact cells that are close to the abutting cell. This makes it possible to appropriately set the cost of the non-contact cells, taking into account the temperature decrease due to heat radiation over time and the temperature increase due to heat propagated from the molten part.

[0068] Furthermore, control device 30 according to the present embodiment includes stacking command unit 33 that moves fusion zone 23 along the path set by path setting unit 32c to stack weld bead 70. This makes it possible to form an object with a stable weld bead shape.

[0069] [Variations] (A) In the above-described embodiment, the cost of a cell is set as a linear function of time t (the cost of a cell is decreased by "1" as time passes), but the cost of a cell may also be set as an nth-order function of time t (n is an integer greater than or equal to 2) or a function of a multidimensional variable that includes variables other than time t.

[0070] (B) When setting the cost of a cell, taking into account defects caused by excessive temperature drops, the cost of cells that have been stacked for a certain amount of time can be set to zero or a negative value. This allows for a path that passes through the base part that has been stacked for a certain amount of time to be set more quickly.

[0071] (C) The cost increase / decrease amount may be a variable value instead of a fixed value, which makes it easier to optimize the cost.

[0072] (D) The cost of the cell may be increased as the number of times the molten parts approach each other increases.

[0073] (E) In view of the fact that heat tends to accumulate more easily when layers are stacked in the weld bead, the cost of the cell may be increased as the weld bead layer becomes higher.

[0074] (F) When the welding current is AC, the higher the EN ratio, the lower the net heat input to the molten part. Therefore, the higher the EN ratio, the lower the cost. The EN ratio is the ratio of the time during which the wire is negative polarity to the total time in one AC cycle. Alternatively, the EN ratio may be the ratio of the time-integrated value of the negative polarity current in one AC cycle to the time-integrated value of the current in one AC cycle (the sum of the time-integrated value of the positive polarity current and the time-integrated value of the negative polarity current).

[0075] (G) The temperature may be actually monitored during lamination, and the cost may be changed based on the results.

[0076] (H) The cost may be changed during one layering. For example, the cost may be reduced near the starting point.

[0077] (I) The range of cells whose costs are to be increased may be changed. For example, the costs of cells located diagonally from the abutting cell, or cells located within a predetermined distance from the abutting cell, may be increased in addition to the cells located vertically or horizontally from the abutting cell.

[0078] (J) The cost may be derived using actually measured temperatures or temperature simulation values.

[0079] (K) In the above embodiment, the total value of the multiple costs stored until the fusion section passes through all the cells of each path candidate is set as the score for each path candidate. However, it is also possible to set the maximum value of the multiple costs stored until the fusion section passes through all the cells of each path candidate as the score for each path candidate.

[0080] (L) In the above embodiment, the cost is used only to calculate the final score, but when selecting a node, it is also possible to select a node that has the smallest cost.

[0081] (M) The width of the mesh (cell width) that separates the layered area of ​​the object is set to the width of the weld bead, but since the width of the weld bead can vary depending on the layering conditions, the mesh width may be varied depending on the layering conditions.

[0082] (N) In the layer immediately below the fusion zone where the cost fluctuations are large, the mesh size may be made finer.

[0083] (O) In areas where there is a large temperature gradient, such as where heat has accumulated, the mesh size may be made finer.

[0084] (P) In areas where heat conduction is likely to occur, the mesh size may be made finer.

[0085] (Q) If multiple bead widths can be set in welding power supply 10, the maximum bead width that can be drawn in one stroke may be selected from the settable bead widths.

[0086] (R) For example, if the number of paths varies depending on the bead width, you can select a bead width that results in fewer paths. For example, if the graph of the buildup area is not drawn in one stroke when the bead width is 6 mm (there will be two or more paths), but the graph of the buildup area can be drawn in one stroke when the bead width is 5 mm (there will be one path), you can set the bead width to 5 mm, which results in fewer paths.

[0087] (S) The shape of the cells separating the stacked regions does not have to be rectangular, and may be a shape other than rectangular, such as a triangle.

[0088] [Aspect] It will be understood by those skilled in the art that the above-described embodiments and their modifications are specific examples of the following aspects.

[0089] (Item 1) A stacking planning device according to the present disclosure creates a stacking plan for building an object by stacking weld beads, and includes an acquisition unit that acquires three-dimensional shape data of the object, and a path planning unit that plans paths that are movement routes of molten parts to build the object. The path planning unit sets paths based on the shape data so as to minimize the number of paths.

[0090] According to the lamination planning device of the first paragraph, the number of passes is minimized when laminating weld beads to form an object, so the number of start and end portions of the weld beads can be minimized. As a result, the shape of the weld bead can be stabilized.

[0091] (2) In the stacking planning device described in paragraph 1, the path planning unit defines the stacking area of ​​the object grasped based on the shape data as a collection of multiple cells divided into a mesh by the width of the weld bead, creates a graph connecting the centers of the multiple cells, determines whether the graph can be drawn in one stroke using graph theory, and if the graph can be drawn in one stroke, sets a route that allows the graph to be drawn in one stroke as the path.

[0092] According to the stacking planning device of the second paragraph, the number of passes can be reduced to the minimum number, "1."

[0093] (Item 3) In the stacking planning device described in item 2, if the graph cannot be drawn in one stroke, the path planning unit divides the graph into the minimum number of subgraphs, each of which can be drawn in one stroke, and sets a path that can draw each subgraph in one stroke.

[0094] According to the stacking planning device of the third item, the number of passes can be reduced to the minimum necessary.

[0095] (4) In the stacking planning device described in paragraph 2, when there are multiple path candidates that can be drawn in one stroke on the graph, the path planning unit sets a score for each path candidate, which is an index representing the heat amount of the base part that the molten part comes into contact with in the process of moving the molten part along each path candidate, and sets the path candidate with the smallest score as the path.

[0096] (Item 5) In the stacking planning device described in item 4, when there are multiple path candidates, the path planning unit stores the cost of the abutting cells, which are cells in the base part that the molten part abuts, each time it moves the molten part to the next cell along each path candidate, and sets the cost of the non-abutting cells, which are cells in the base part that the molten part does not abut, and sets the score of each path candidate to the total or maximum value of the multiple costs stored until the molten part has moved through all the cells of each path candidate.

[0097] According to the lamination planning device of items 4 and 5, it is possible to set a path that is less likely to cause melt-through due to the heat of the molten part, taking into consideration the temperature of the base part that changes from moment to moment as the molten part moves.

[0098] (Item 6) In the stacking planning device described in Item 5, the path planning unit reduces the cost of non-contact cells and increases the cost of cells that are close to the contact cells among the non-contact cells each time the molten part is moved to the next cell along each path candidate.

[0099] According to the lamination planning device of item 6, the cost of the non-contacting cells can be appropriately set by taking into consideration the temperature decrease due to heat radiation over time and the temperature increase due to heat propagated from the molten part.

[0100] (Item 7) An additive manufacturing system according to the present disclosure includes the stacking planning device according to item 1, and a manufacturing device that manufactures an object by stacking weld beads in accordance with the stacking plan formulated by the stacking planning device.

[0101] According to the additive manufacturing system of item 7, it is possible to manufacture an object with a stable weld bead shape.

[0102] (Item 8) A stacking planning method according to the present disclosure is a stacking planning method for building an object by stacking weld beads, and includes the steps of acquiring three-dimensional shape data of the object and planning paths, which are movement paths of molten parts to build the object. The path planning step includes the step of setting paths based on the shape data so as to minimize the number of paths.

[0103] According to the lamination planning method of Section 8, the number of passes is minimized when laminating weld beads to form an object, so the number of start and end portions of the weld bead can be kept to a necessary minimum. As a result, the shape of the weld bead can be stabilized.

[0104] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The technical scope of the present disclosure is defined by the claims, not by the description of the above embodiments, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0105] 1 Additive manufacturing system, 10 welding power supply, 20 welding torch, 21 nozzle, 22 arc, 23 molten part, 30 control device, 31 shape data acquisition unit, 32 path planning unit, 32a path candidate generation unit, 32b score setting unit, 32c path setting unit, 33 stacking command unit, 40 robot arm, 50 wire supply device, 51 consumable electrode wire, 70 weld bead, 80 base material, 100 object.

Claims

1. A stacking planning device that creates a stacking plan for building an object by stacking weld beads, an acquisition unit that acquires three-dimensional shape data of the object; a path planning unit that plans a path that is a movement path of a molten part for forming the object, The path planning unit sets the paths based on the shape data so as to minimize the number of paths.

2. The path planning unit defining the layered region of the object grasped based on the shape data as an aggregate of a plurality of cells divided into a mesh shape by the width of the weld bead, and creating a graph connecting the centers of the plurality of cells; determining whether the graph can be drawn in one stroke using graph theory; The stacking planning device according to claim 1 , wherein, when the graph can be drawn in one stroke, a route that allows the graph to be drawn in one stroke is set as the path.

3. When the graph cannot be drawn in one stroke, the path planning unit Dividing the graph into a minimum number of subgraphs, each of which is unicursible; The stacking planning device according to claim 2 , wherein a path that allows each of the subgraphs to be traced in one stroke is set as the path.

4. When there are a plurality of path candidates that can be drawn in one stroke on the graph, the path planning unit a score is set for each of the path candidates, the score being an index representing the amount of heat of a base portion that comes into contact with the molten portion during the process of moving the molten portion along each of the path candidates; The stacking planning device according to claim 2 , wherein the path candidate with the smallest score is set as the path.

5. When there are a plurality of path candidates, the path planning unit Each time the molten part is moved to the next cell along each of the path candidates, a cost of an abutting cell, which is a cell of the base part with which the molten part is in contact, is stored, and a cost of a non-abutting cell, which is a cell of the base part with which the molten part is not in contact, is set; 5. The stacking planning device according to claim 4, wherein a score for each of the path candidates is set to a total or maximum value of the plurality of costs stored until the fusion portion has moved through all the cells of each of the path candidates.

6. 6. The stacking planning device according to claim 5, wherein the path planning unit decreases a cost of the non-contact cells and increases a cost of a cell of the non-contact cells that is adjacent to the contact cell, each time the molten portion is moved to a next cell along each of the path candidates.

7. The stacking planning device according to claim 1 ; and a modeling device that models the object by stacking weld beads in accordance with the stacking plan created by the stacking planning device.

8. A stacking planning method for creating a stacking plan for building an object by stacking weld beads, comprising: acquiring three-dimensional shape data of the object; a step of planning a path that is a movement path of a molten portion for forming the object; The step of planning the paths includes a step of setting the paths based on the shape data so that the number of paths is minimized.

Images