JP7353245B2 - Additive manufacturing equipment and additive manufacturing method - Google Patents

Description

The present disclosure relates to an additive manufacturing apparatus and an additive manufacturing method for manufacturing a three-dimensionally shaped object.

Additive manufacturing (AM) technology is known as one of the technologies for manufacturing three-dimensional objects. According to the Direct Energy Deposition (DED) method, which is one of several methods in additive manufacturing technology, additive manufacturing equipment supplies filler metal to the processing point where the beam is irradiated. A bead is formed by moving the processing point while A bead is a solidified material obtained by solidifying a molten filler material. Additive manufacturing equipment manufactures objects by stacking layers of beads in sequence.

A shaped object may have a portion where an error is likely to occur between the bead shape determined based on design data and the bead shape to be formed. Whether or not an error occurs can be estimated to some extent based on the experience of the user who uses the additive manufacturing device. However, since the occurrence of errors is related to various physical phenomena during the manufacturing of the shaped object, it has been difficult to estimate with high accuracy the location where errors occur or the degree of errors.

Patent Document 1 is equipped with a machining head that sprays metal powder as a filler material at the same time as beam irradiation, and an error that is the difference between the actual distance and the assumed distance between the metal powder spraying surface and the machining head is detected. An additive manufacturing device is disclosed that adjusts the feed rate of metal powder in some cases. When an error is detected in one layer, the additive manufacturing apparatus according to Patent Document 1 calculates processing conditions such as supply amount, laser output, and processing head speed for layers stacked on that layer.

JP 2017-190505 Publication

The conventional additive manufacturing apparatus disclosed in Patent Document 1 performs post-processing to compensate for the lack of height in the next layer to be formed when there is a shape error in which the height of a layer is insufficient. be exposed. For this reason, in conventional additive manufacturing equipment, post-processing is required every time a shape error due to insufficient height occurs, resulting in a problem in that the time required for manufacturing increases. In addition, with conventional additive manufacturing equipment, if a shape error occurs due to a significant deviation of the processing conditions from the proper processing conditions, it is not possible to correct the shape error in the next layer formed, resulting in a decrease in shape accuracy. There was a problem in that it could be reduced.

The present disclosure has been made in view of the above, and an object of the present disclosure is to obtain an additive manufacturing device that can shorten the time required for manufacturing and can manufacture a shaped object with high shape accuracy.

In order to solve the above-mentioned problems and achieve the objectives, the additive manufacturing apparatus according to the present disclosure melts a filler material by beam irradiation and builds by stacking layers that are solidified materials of the melted filler material. Manufacture things. The additive manufacturing apparatus according to the present disclosure updates the objective function value by inputting a measurement value that is a result of measuring a layer of a manufactured object into an objective function that represents a shape error of the object. a search processing section that searches for machining conditions based on evaluation based on the value of the objective function ; and a command generation section that generates a command indicating a group of interpolation points on a beam movement path . The objective function updating unit updates the objective function by inputting into the objective function a measured value at each command position, which is the position of each interpolation point included in the interpolation point group. The search processing section searches for machining conditions for each commanded position.

The additive manufacturing apparatus according to the present disclosure has the advantage that the time required for manufacturing can be shortened and a shaped article with high shape accuracy can be manufactured.

Block diagram showing the configuration of the additive manufacturing apparatus according to the first embodiment A diagram showing an example of a shaped object manufactured by the additive manufacturing apparatus according to Embodiment 1. A diagram for explaining the search for processing conditions by the additive manufacturing apparatus according to the first embodiment Flowchart showing the operation procedure of the additive manufacturing apparatus according to the first embodiment A block diagram showing the configuration of a search processing unit when searching for machining conditions using machine learning in Embodiment 1. A block diagram showing the functional configuration of a learning device included in the additive manufacturing device according to Embodiment 1. Flowchart showing the processing procedure of the learning device in Embodiment 1 A block diagram showing the functional configuration of an inference device included in the additive manufacturing device according to Embodiment 1. Flowchart showing the processing procedure of the inference device in Embodiment 1 A block diagram showing the configuration of a processing condition search device included in the additive manufacturing device according to Embodiment 2. Diagram for explaining the operation of the additive manufacturing apparatus according to Embodiment 3 Diagram for explaining a modification of Embodiment 3 Diagram for explaining the operation of the additive manufacturing apparatus according to Embodiment 4 A diagram showing an example of the hardware configuration of the processing condition search device according to Embodiments 1 to 4.

Below, an additive manufacturing apparatus and an additive manufacturing method according to an embodiment will be described in detail based on the drawings.

Embodiment 1.
FIG. 1 is a block diagram showing the configuration of an additive manufacturing apparatus according to a first embodiment. Additive manufacturing apparatus 100 according to the first embodiment is a machine tool that manufactures a shaped object by adding molten filler material to a workpiece. The additive manufacturing apparatus 100 melts a filler material by beam irradiation and manufactures a shaped object by stacking layers that are solidified materials of the melted filler material. The filler metal is a material such as metal wire or metal powder. The filler metal may be other than metal. In the first embodiment, the beam is a laser beam. The beam may be an electron beam or an arc beam.

The additive manufacturing device 100 includes a processing condition search device 10 that searches for processing conditions, a numerical control device 11 that controls the additive manufacturing device 100, a processing section 12 that performs processing for manufacturing a shaped object, and a processing section 12 that performs processing for manufacturing a shaped object. It has a measuring device 13 for measuring the shape of. In the first embodiment, the processing condition search device 10 is built into the additive manufacturing device 100. The processing condition search device 10 may be a device external to the additive manufacturing device 100. The processing condition search device 10 may be a device that can be connected to the additive manufacturing device 100 via a network. The processing condition search device 10 may be a device existing on a cloud server. The processing conditions searched by the processing condition search device 10 may be applied to a plurality of additive manufacturing devices 100 connected to the processing condition search device 10 via a network.

The processing section 12 includes a laser oscillator 20 that oscillates a laser beam, a material supply section 21 that supplies filler material to the irradiation position of the laser beam, and a shaft drive section 22 that drives the processing head. Illustration of the processing head is omitted. The processing head emits a laser beam toward the workpiece. The shaft drive unit 22 moves the processing head in each of the X-axis direction, Y-axis direction, and Z-axis direction. The X-axis, Y-axis, and Z-axis are three axes that are perpendicular to each other. The X axis and the Y axis are horizontal axes. The Z axis is a vertical axis. In the additive manufacturing apparatus 100, the shaft drive unit 22 moves the processing head, thereby moving the irradiation position of the laser beam on the workpiece.

The measuring device 13 measures the three-dimensional shape of the layer by measuring the height of the layer in the Z-axis direction and the shape of the layer in the X-axis direction and the Y-axis direction. For example, a camera is used as the measuring device 13. The measuring device measures the three-dimensional shape of the layer by analyzing the captured image. The measuring device 13 may be a laser displacement meter, an optical coherence tomography (OCT), or the like that performs optical coherence tomography.

Numerical control device 11 controls additive manufacturing device 100 according to a processing program. The numerical control device 11 includes a machining program analysis section 16 that analyzes a machining program, a command generation section 17 that generates commands to be sent to the machining section 12, and a machining condition output section 18 that outputs machining conditions to the machining section 12. have

The machining program analysis unit 16 determines the movement path of the irradiation position based on the contents of the processing described in the machining program. The machining program analysis section 16 outputs data on the movement route to the command generation section 17. The machining program analysis unit 16 sets machining conditions by acquiring information for setting machining conditions from the machining program. The machining program analysis unit 16 may acquire the machining program by reading machining conditions described in the machining program from the machining program.

The command generation unit 17 generates a position command based on the movement route data. The position command is a command indicating a group of interpolation points on the movement path of the laser beam. The command generating section 17 controls the shaft driving section 22 by sending the generated position command to the shaft driving section 22 .

The command generation unit 17 generates a beam output command according to the machining program and machining conditions. The beam output command is a command indicating the output of the laser beam. The command generation unit 17 controls the laser oscillator 20 by sending the generated laser output command to the laser oscillator 20.

The command generation unit 17 generates a supply command according to a machining program and machining conditions. The supply command is a command indicating the supply speed of the filler metal. The command generation unit 17 controls the material supply unit 21 by sending the generated supply command to the material supply unit 21 .

The machining condition output unit 18 receives machining condition information that is a search result by a machining condition search device 10, which will be described later. The initial machining conditions are machining conditions that are initially set according to the machining program, and are machining conditions that are set in advance. The switching unit 19 switches between the search results and the initial machining conditions. Through this switching, the machining condition output unit 18 outputs one of the machining condition information that is the search result and the initial machining condition information to the machining unit 12. Thereby, the machining conditions input to the machining section 12 are switched between the search results and the initial machining conditions.

The processing condition search device 10 includes an objective function update section 14 and a search processing section 15. The objective function updating unit 14 maintains an objective function that uses the shape error of the layer as a variable, and updates the value of the objective function by inputting the measured value of the shape error in the manufactured object into the objective function. The search processing unit 15 searches for machining conditions based on evaluation based on the calculation results.

The command generation unit 17 outputs command position information 24, which is information indicating the command position, to the objective function update unit 14 and the search processing unit 15. The commanded position is the position of each interpolation point that constitutes the interpolation point group. The measuring device 13 outputs a measured value 25 as a measurement result to the objective function updating section 14 and the search processing section 15. The machining condition output unit 18 outputs machining condition information 26, which is one of the machining condition information that is the search result and the initial machining condition information, to the machining unit 12 and the search processing unit 15.

In the first embodiment, the objective function is expressed as “objective function=Σ|(target shape of layer)−(measured shape of layer)|”. “|(Target shape of layer)−(Measured shape of layer)|” represents the shape error of the layer. The objective function updating unit 14 updates the objective function by inputting the measured value 25 for each commanded position into the objective function. Further, the objective function updating unit 14 updates the objective function by inputting the measured values 25 for each of the plurality of layers constituting the modeled object into the objective function.

The search processing unit 15 searches for machining conditions for each commanded position, which are machining conditions that minimize the value of the objective function. The search processing unit 15 searches processing conditions such as material supply speed, laser beam output, processing head feed speed, and pitch. The pitch is the distance between the laser beam emission position and the workpiece.

The search processing unit 15 obtains machining conditions as a search result by adding the correction amount to the machining conditions used in the current machining. The search processing unit 15 searches for machining conditions based on the initial machining conditions in the first search after starting manufacturing of the shaped object. After performing the first search, the search processing unit 15 searches for machining conditions based on the machining conditions that are the search results.

A randomly determined correction amount can be used as the correction amount. The amount of correction may be determined based on the correlation between the shape of the object and the processing conditions. For the correlation, a relationship that has been empirically obtained in additive manufacturing can be used. The amount of correction may be determined based on a model of physical phenomena associated with modeling. A model of the physical phenomenon refers to a model in which the supply of filler metal is increased to increase the volume of the layer.

When one layer is formed when manufacturing one modeled object, the measuring device 13 measures the formed layer and sends the measured value 25 to the processing condition search device 10. The processing condition search device 10 acquires command position information 24 corresponding to the measured value 25 from the command generation unit 17 . The processing condition search device 10 also acquires target shape data indicating the target shape of the layer. The processing condition search device 10 acquires target shape data from, for example, design data of a shaped object. The objective function updating unit 14 updates the value of the objective function by inputting the target shape data and the measured value 25 into the objective function.

The objective function updating unit 14 acquires the measured value 25 linked to the command position. The objective function updating unit 14 acquires target shape data and measured values 25 for each layer of the object, and updates the value of the objective function. Furthermore, the objective function updating unit 14 updates the value of the objective function in manufacturing a plurality of objects having the same shape. The same shape means that the three-dimensional shapes are the same. The plurality of objects having the same shape are, for example, the plurality of objects manufactured based on the same design data.

The objective function updating unit 14 determines whether to finish updating the value of the objective function by evaluating whether the value of the objective function has been minimized. When the objective function updating unit 14 determines that the value of the objective function has been minimized, it determines to end the update. The objective function updating unit 14 sends a signal 23 indicating whether or not there is an update to the search processing unit 15. Note that in the first embodiment, the value of the objective function is minimized refers to a state in which the difference between the target shape and the formed shape is no longer detected by the measuring instrument 13.

The search processing unit 15 searches for machining conditions every time the value of the objective function is updated. The search processing unit 15 ends the search for processing conditions when the objective function updating unit 14 finishes updating the value of the objective function. In this way, the search processing unit 15 searches for machining conditions based on evaluation based on the value of the objective function.

Next, with reference to FIGS. 2 and 3, searching for machining conditions by the machining condition search device 10 will be described. FIG. 2 is a diagram showing an example of a shaped object manufactured by the additive manufacturing apparatus according to the first embodiment. FIG. 3 is a diagram for explaining the search for processing conditions by the additive manufacturing apparatus according to the first embodiment. The shaped object 30 shown in FIG. 2 has a three-layer wall shape manufactured by sequentially stacking a first layer 31, a second layer 32, and a third layer 33. Here, a search for processing conditions in the case of manufacturing a plurality of shaped objects 30 shown in FIG. 2 will be described. The height of the first layer 31 is z1, the height of the second layer 32 is z2, and the height of the third layer 33 is z3. This height is defined as the height in the Z-axis direction.

The object 30-1 is the object 30 manufactured by the first manufacturing process among the plural objects 30. The object 30-1 is a object 30 manufactured using the initial processing conditions. FIG. 3 shows the shape of the object 30-1 in the XZ plane and the amount of filler material supplied when forming each layer. The supply amount is one example of processing conditions that are searched. The vertical axis of the graph representing the supply amount is the supply amount W, and the horizontal axis represents the command position in the X-axis direction.

When the first layer 31 of the object 30-1 is formed, the measuring device 13 measures the shape of the first layer 31 at each of the command positions x1, x2, . . . , xn. When the second layer 32 of the object 30-1 is formed, the measuring device 13 measures the shape of the second layer 32 at each of the command positions x1, x2, . . . , xn. When the third layer 33 of the object 30-1 is formed, the measuring device 13 measures the shape of the third layer 33 at each of the command positions x1, x2, . . . , xn. n is an integer.

The objective function updating unit 14 updates the objective function using the measured value 25 for the object 30-1. The objective function updating unit 14 inputs the absolute value of Δz1 at each command position x1, x2, . . . , xn to the objective function. Δz1 is a shape error of the first layer 31, and is the difference between the height of the formed first layer 31 and z1. The objective function updating unit 14 inputs the absolute value of Δz2 at each command position x1, x2, . . . , xn to the objective function. Δz2 is a shape error of the second layer 32, and is the difference between the height of the formed second layer 32 and z2. The objective function updating unit 14 inputs the absolute value of Δz3 at each command position x1, x2, . . . , xn to the objective function. Δz3 is a shape error of the third layer 33, and is the difference between the height of the formed third layer 33 and z3. The search processing unit 15 searches for machining conditions for each command position x1, x2, . . . , xn as the objective function is updated.

The end of the bead at the machining start position and the end of the bead at the machining end position tend to sag in shape. In the modeled object 30-1, since the machining conditions have not been changed from the initial machining conditions, a shape error occurs between the bead machining start position and the bead machining end position in each layer. Each of the first layer 31, second layer 32, and third layer 33 of the shaped object 30-1 has a shape error that reduces the height in the Z-axis direction. When the additive manufacturing apparatus 100 determines that the height of the modeled object 30-1 is insufficient at the time when the production of the modeled object 30-1 is completed, the additive manufacturing apparatus 100 supplies materials 34-2, 34 to compensate for the lack of height. Perform post-processing to add -3.

In this way, the additive manufacturing apparatus 100 supplements the material in a post-process when there is a shortage of material added in manufacturing the modeled object 30. The additive manufacturing apparatus 100 may remove a part of the added material by post-processing when the material added in manufacturing the modeled object 30 is excessive.

The additive manufacturing apparatus 100 performs the second manufacturing process using the processing conditions that are the search results when the manufacturing of the shaped object 30-1 is completed, and manufactures the shaped object 30-2. The measuring device 13 performs measurements in the same manner as in the case of the shaped object 30-1. When the manufacturing of the object 30-2 is completed, the objective function updating unit 14 updates the objective function by inputting the shape error into the objective function as in the case of the object 30-1. The search processing unit 15 searches for machining conditions for each command position x1, x2, . . . , xn as the objective function is updated.

Since the processing conditions used for the modeled object 30-2 are more suitable than the initial processing conditions, the shape errors in each layer of the modeled object 30-2 are smaller than those for the modeled object 30-1. . Specifically, by increasing the supply amount W at the bead processing start position and bead processing end position of each layer, the insufficient height of each layer is compensated for. The amount of material 34-3 added to the object 30-2 through post-processing is smaller than that for the object 30-1.

The shaped article 30-m is the shaped article 30 manufactured by the m-th manufacturing process by the additive manufacturing apparatus 100. m is an integer. The processing condition search device 10 searches for processing conditions every time a molded article 30 of the same shape is manufactured repeatedly. The additive manufacturing apparatus 100 can reduce shape errors by searching for appropriate processing conditions each time the manufacturing of the shaped object 30 is repeated. The additive manufacturing apparatus 100 can efficiently reduce shape errors at locations where shape errors are likely to occur by optimizing the processing conditions for each commanded position. By repeating the manufacturing process a plurality of times, it is possible to obtain a shaped object 30-m in which shape errors are eliminated, as shown in FIG.

The additive manufacturing apparatus 100 searches for machining conditions each time a molded object 30 having the same shape is repeatedly manufactured, thereby eliminating the need for shape correction by post-processing. The additive manufacturing apparatus 100 can shorten the time required for manufacturing by eliminating the need for shape correction by post-processing and eliminating the time required for manufacturing.

FIG. 4 is a flowchart showing the operation procedure of the additive manufacturing apparatus according to the first embodiment. In step S1, the additive manufacturing apparatus 100 forms layers of the shaped object 30. In step S2, the additive manufacturing apparatus 100 measures the shape of the layer using the measuring device 13. In step S3, the additive manufacturing apparatus 100 determines whether or not the modeling of the object 30 has been completed.

The additive manufacturing apparatus 100 determines that the modeling of the modeled object 30 has been completed when all the layers of the modeled object 30 have been formed. If there is a layer of the object 30 that has not been formed yet, the additive manufacturing apparatus 100 determines that the modeling of the object 30 has not been completed. When the modeling of the object 30 is completed (step S3, Yes), the additive manufacturing apparatus 100 advances the procedure to step S4, which will be described below. If the modeling of the object 30 has not been completed (Step S3, No), the additive manufacturing apparatus 100 returns the procedure to Step S3 and forms a layer that has not been formed yet.

In step S4, the additive manufacturing apparatus 100 updates the objective function using the measured value 25 obtained by the measurement in step S2. In step S5, the additive manufacturing apparatus 100 searches for processing conditions in the search processing section 15. Thereby, the additive manufacturing apparatus 100 ends the operation according to the procedure shown in FIG. When manufacturing objects 30 of the same shape, the additive manufacturing apparatus 100 updates the objective function and searches for processing conditions by operating according to the procedure shown in FIG.

The search processing unit 15 may search for machining conditions using machine learning. FIG. 5 is a block diagram showing the configuration of a search processing section when searching for machining conditions using machine learning in the first embodiment. The search processing unit 15 includes a learning device 41 that performs machine learning, an inference device 42 that infers processing conditions that reduce the shape error of the object 30, and a model storage unit 43 that stores a learned model. The learning device 41 learns processing conditions that minimize the objective function and generates a learned model. The inference device 42 uses the learned model to infer machining conditions for each commanded position that minimize the objective function.

FIG. 6 is a block diagram showing the functional configuration of a learning device included in the additive manufacturing apparatus according to the first embodiment. The learning device 41 includes a data acquisition section 44 and a model generation section 45. The data acquisition unit 44 acquires command position information 24, measured values 25, and processing condition information 26. Further, the data acquisition unit 44 acquires target shape data indicating the target shape. The data acquisition unit 44 acquires command position information 24, measured values 25, processing condition information 26, and target shape data as learning data.

The data acquisition unit 44 creates a data set in which the measured value 25 for each command position, the machining condition information 26 for each command position, and the target shape data for each command position are combined. That is, the data acquisition unit 44 creates a data set including the measured value 25 linked to the command position, the machining condition information 26, and the target shape data. The data acquisition unit 44 outputs the created data set to the model generation unit 45.

The model generation unit 45 generates a trained model using the data set. Any learning algorithm may be used by the model generation unit 45. As an example, a case where reinforcement learning is applied will be described. Reinforcement learning is a method in which an agent in an environment observes the current state and decides what action to take. Agents obtain rewards from the environment by selecting actions, and through a series of actions, they learn strategies that will yield the most rewards. Q-Learning and TD-Learning are known as typical methods of reinforcement learning. For example, in the case of Q-learning, the action value table, which is a general update formula for the action value function Q(s, a), is expressed by the following equation (1). The action value function Q(s, a) represents the action value Q, which is the value of the action of selecting action "a" under environment "s".

The update formula expressed by the above equation (1) is that if the action value of the best action "a" at time "t+1" is greater than the action value Q of the action "a" executed at time "t", then , the action value Q is increased, and in the opposite case, the action value Q is decreased. In other words, the action value function Q(s, a) is updated so that the action value Q of action "a" at time "t" approaches the best action value at time "t+1". As a result, the best action value in a certain environment is sequentially propagated to the action value in the previous environment.

The model generation section 45 includes a reward calculation section 46 and a function update section 47. The reward calculation unit 46 calculates the reward based on the value of the objective function. The function updating unit 47 updates the function for determining processing conditions according to the remuneration calculated by the remuneration calculation unit 46. The function update unit 47 outputs the learned model created by updating the function to the model storage unit 43.

The reward calculation unit 46 calculates the value of the target function based on the measured value 25 linked to the command position and the target shape data. The reward calculation unit 46 calculates the reward "r" based on the value of the objective function. The reward calculation unit 46 increases the reward "r" when the value of the objective function becomes small. The reward calculation unit 46 increases the reward "r" by giving a reward value of "1". Note that the reward value is not limited to "1". Moreover, the reward calculation unit 46 reduces the reward "r" when the difference becomes large. The reward calculation unit 46 reduces the reward "r" by giving a reward value of "-1". Note that the reward value is not limited to "-1".

FIG. 7 is a flowchart showing the processing procedure of the learning device in the first embodiment. A reinforcement learning method for updating the action value function Q(s,a) will be described with reference to the flowchart in FIG.

In step S11, the learning device 41 acquires learning data. The data acquisition unit 44 creates a dataset using the learning data. In step S12, the learning device 41 calculates a reward. In step S13, the learning device 41 updates the action value function Q(s, a) based on the reward. In step S14, the learning device 41 determines whether the action value function Q(s, a) has converged. The learning device 41 determines that the action value function Q(s, a) has converged because the action value function Q(s, a) is no longer updated in step S14.

If it is determined that the action value function Q(s, a) has not converged (step S14, No), the learning device 41 returns the operation procedure to step S11. If it is determined that the action value function Q(s, a) has converged (step S14, Yes), the learning device 41 ends the learning according to the procedure shown in FIG. 7. Note that the learning device 41 may continue learning by returning the operating procedure from step S13 to step S11 without making the determination in step S14. The model storage unit 43 stores the learned model that is the generated action value function Q(s, a).

FIG. 8 is a block diagram showing the functional configuration of an inference device included in the additive manufacturing apparatus according to the first embodiment. The inference device 42 infers machining conditions for each commanded position, which are machining conditions that minimize the value of the objective function, based on the learned model. The inference device 42 includes a data acquisition section 48 and an inference section 49.

The data acquisition unit 48 acquires the command position information 24 as inference data. The data acquisition section 48 outputs the command position information 24 to the inference section 49 . The inference unit 49 inputs the command position information 24 to the trained model read out from the model storage unit 43 to infer machining conditions for each command position that minimize the value of the objective function.

FIG. 9 is a flowchart showing the processing procedure of the inference device in the first embodiment. In step S15, the inference device 42 uses the data acquisition unit 48 to acquire command position information 24, which is inference data. In step S16, the inference unit 49 of the inference device 42 inputs the command position information 24, which is the acquired data, to the learned model, and infers processing conditions for each command position. In step S17, the inference device 42 outputs machining condition information indicating the machining conditions, which is the inference result, to the numerical control device 11. Thereby, the inference device 42 ends the processing according to the procedure shown in FIG.

The additive manufacturing apparatus 100 can efficiently search for processing conditions that minimize the value of the objective function using machine learning techniques.

In the first embodiment, a case has been described in which reinforcement learning is applied to the learning algorithm used by the learning device 41, but learning other than reinforcement learning may be applied to the learning algorithm. The learning device 41 may perform machine learning using a known learning algorithm other than reinforcement learning, for example, a learning algorithm such as deep learning, neural network, genetic programming, functional logic programming, or support vector machine. good.

In the first embodiment, the learning device 41 is built into the additive manufacturing device 100. The learning device 41 is not limited to a device included in the additive manufacturing device 100, but may be a device external to the additive manufacturing device 100. The learning device 41 may be a device connectable to the additive manufacturing device 100 via a network. The learning device 41 may be a device existing on a cloud server.

The learning device 41 may learn parameters according to data sets created for the plurality of additive manufacturing devices 100. The learning device 41 may acquire data sets from multiple additive manufacturing devices 100 used at the same site, or may acquire datasets from multiple additive manufacturing devices 100 used at different sites. Also good. The data set may be collected from multiple additive manufacturing devices 100 operating independently from each other at multiple sites. After starting the collection of data sets from a plurality of additive manufacturing apparatuses 100, a new additive manufacturing apparatus 100 may be added to the targets for which data sets are collected. Furthermore, after starting the collection of data sets from the plurality of additive manufacturing apparatuses 100, some of the plurality of additive manufacturing apparatuses 100 may be excluded from the targets for which data sets are collected.

The learning device 41 that has learned about one additive manufacturing device 100 may also learn about other additive manufacturing devices 100 other than the additive manufacturing device 100. The learning device 41 that performs learning on the other additive manufacturing device 100 can update the output prediction model by relearning in the other additive manufacturing device 100.

According to the first embodiment, the additive manufacturing apparatus 100 updates the value of the objective function by inputting the measured value 25 into the objective function whose variable is the shape error of the layer, and performs evaluation based on the value of the objective function. Search for machining conditions based on In the additive manufacturing apparatus 100, processing conditions are searched based on evaluation based on the value of the target function, so that post-processing is not necessary, so that the time required for manufacturing can be shortened. Further, the additive manufacturing apparatus 100 is able to manufacture a shaped object with high shape accuracy by searching for processing conditions based on evaluation based on the value of the objective function. As described above, the additive manufacturing apparatus 100 has the advantage of being able to shorten the time required for manufacturing and to manufacture a shaped object with high shape accuracy.

Embodiment 2.
In the second embodiment, an example will be described in which the value of the objective function is updated and the processing conditions are searched based on the results of simulating the manufacture of the shaped object 30. FIG. 10 is a block diagram showing the configuration of a processing condition search device included in the additive manufacturing apparatus according to the second embodiment. In Embodiment 2, the same components as in Embodiment 1 described above are given the same reference numerals, and configurations that are different from Embodiment 1 will be mainly explained.

The processing condition search device 50 includes an objective function update section 14, a search processing section 15, and a simulation section 51. The simulation unit 51 simulates the manufacturing of the shaped object 30.
In the second embodiment, the simulation section 51 is built into the processing condition search device 50. The simulation unit 51 may be a device external to the processing condition search device 50. The simulation unit 51 may be a device external to the additive manufacturing device 100.

The simulation unit 51 outputs shape data representing the shape of each layer constituting the modeled object 30 to the objective function updating unit 14 each time the simulation unit 51 simulates the manufacturing of the modeled object 30. The search processing unit 15 searches for machining conditions used in the simulation, and outputs machining condition information representing the machining conditions as a search result to the simulation unit 51. The simulation unit 51 simulates the manufacturing of the shaped object 30 according to inputted processing condition information. When the simulation by the simulation unit 51 is completed, the search processing unit 15 outputs machining condition information representing the machining conditions as a search result to the numerical control device 11.

According to the second embodiment, the additive manufacturing apparatus 100 can search for processing conditions by performing simulation without consuming filler metal, electric power, etc. required for actual modeling. can.

Embodiment 3.
In Embodiment 3, an example will be described in which processing conditions for a target shape are searched for using processing condition search results for a shaped object having a shape similar to the target shape. FIG. 11 is a diagram for explaining the operation of the additive manufacturing apparatus according to the third embodiment. In Embodiment 3, the same components as in Embodiment 1 or 2 described above are given the same reference numerals, and configurations that are different from Embodiment 1 or 2 will be mainly explained.

The shaped object 30b shown in FIG. 11 is the first shaped object that is subject to processing conditions. As for the shaped object 30a, it is assumed that the search for processing conditions has already been performed as in the case shown in the first or second embodiment. The object 30a is a second object whose processing conditions have already been searched. It is assumed that the target shape SA of the object 30b and the shape of the object 30a are similar to each other, and the target shape SA is a scaled-up version of the shape of the object 30a. The similarity ratio R1 is the similarity ratio between the shape of the object 30a and the target shape SA.

FIG. 11 shows a graph of the supply amount W, which is the searched processing condition for the object 30a, and a graph of the supply amount W, which is the processing condition for the object 30b. The search processing unit 15 determines the association between the supply amount W and the command position under the initial processing conditions by changing the association between the supply amount W and the command position under the searched processing conditions using the similarity ratio R1. do. In this way, the search processing unit 15 uses the processing conditions obtained by adjusting the searched processing conditions for the object 30a based on the similarity ratio R1, as the initial processing conditions when searching for the processing conditions for the object 30b. The search processing unit 15 searches for processing conditions for the object 30b based on the searched processing conditions for the object 30a.

According to the third embodiment, additive manufacturing apparatus 100 can shorten the time required for searching by searching for processing conditions based on searched processing conditions. Further, the additive manufacturing apparatus 100 can search for processing conditions for the modeled object 30a in advance after the modeled object 30a is scaled down from the actual size. It is possible to reduce the number of times the full-size modeled object 30a is manufactured. Thereby, the additive manufacturing apparatus 100 can reduce consumption of filler material, electric power, and the like.

In the third embodiment, even if the shape of the first object and the shape of the second object are not similar to each other, the shape of the first object is determined based on the searched processed shape of the second object. It is also possible to search for a processed shape for a modeled object. Even if the shape of the first object is different from the shape of the second object in height in the Z-axis direction or length in the X-axis direction, the search for the second object can be performed. A processed shape for the first shaped object may be searched based on the completed processed shape. The shape of the first object is a shape obtained by changing the shape of the second object at a constant size ratio. In such a case, the shape of the first shaped object is assumed to be similar to the shape of the second shaped object.

FIG. 12 is a diagram for explaining a modification of the third embodiment. A shaped object 30d shown in FIG. 12 is a first shaped object that is subject to processing conditions. As for the shaped object 30c, it is assumed that the search for processing conditions has already been performed, similar to the case shown in Embodiment 1 or 2. The object 30c is a second object whose processing conditions have already been searched. The shape of the modeled object 30d and the shape of the modeled object 30c are similar to each other. The shape of the modeled object 30d is obtained by reducing the shape SB of the modeled object 30c by a size ratio R2 in the X-axis direction and the Z-axis direction.

The search processing unit 15 determines the link between the supply amount W and the command position in the initial processing conditions by changing the link between the supply amount W and the command position in the searched processing conditions using the size ratio R2. do. In this way, the search processing unit 15 uses the processing conditions obtained by adjusting the searched processing conditions for the object 30c based on the size ratio R2, as the initial processing conditions when searching for the processing conditions for the object 30d. The search processing unit 15 searches for processing conditions for the object 30d based on the searched processing conditions for the object 30c. In the case of the modified example as well, the additive manufacturing apparatus 100 can shorten the time required for the search by searching for processing conditions based on the searched processing conditions.

Embodiment 4.
In Embodiment 4, an example will be described in which the search for machining conditions is terminated when the shape error falls within an allowable range. FIG. 13 is a diagram for explaining the operation of the additive manufacturing apparatus according to the fourth embodiment. In Embodiment 4, the same components as in Embodiments 1 to 3 described above are given the same reference numerals, and configurations that are different from Embodiments 1 to 3 will be mainly explained.

In the fourth embodiment, the objective function updating unit 14 finishes updating the objective function value when the objective function value is converged within a preset range. The search processing unit 15 starts searching for machining conditions when each of the shape errors Δz1, Δz2, and Δz3 becomes less than a preset threshold in response to the objective function updating unit 14 completing the update. finish. If it is assumed that the value of the objective function is minimized by repeating m-times of manufacturing and searching for processing conditions, then the object 30-(m-k) which is the object 30 manufactured the m-kth time When it is determined that updating of the value of the objective function is to be completed for the object, the additive manufacturing apparatus 100 ends the search for processing conditions for the object 30. k is an integer less than m.

According to the fourth embodiment, the additive manufacturing apparatus 100 can shorten the time required for the search by ending the search for processing conditions when the shape error falls within the allowable range.

Next, the hardware configuration of the machining condition search devices 10 and 50 according to the first to fourth embodiments will be described. FIG. 14 is a diagram showing an example of the hardware configuration of the processing condition search device according to the first to fourth embodiments. FIG. 14 shows a hardware configuration in which the functions of the machining condition search devices 10 and 50 are realized by using hardware that executes a program.

The machining condition search devices 10 and 50 include a processor 61 that executes various processes, a memory 62 that is a built-in memory, a storage device 63 that stores information, and input of information to the machining condition search devices 10 and 50 and processing conditions. It has an interface circuit 64 for outputting information from the search devices 10 and 50.

The processor 61 is a CPU (Central Processing Unit). The processor 61 may be a processing device, a microprocessor, a microcomputer, or a DSP (Digital Signal Processor). The memory 62 is a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), or an EEPROM (registered trademark) (Electrically Erasable Programmable Read Only Memory).

The storage device 63 is an HDD (Hard Disk Drive) or an SSD (Solid State Drive). A program that causes the computer to function as the processing condition search device 10 is stored in the storage device 63. The processor 61 reads the program stored in the storage device 63 into the memory 62 and executes it.

The program may be stored in a storage medium that can be read by a computer system. The machining condition search devices 10 and 50 may store a program recorded on a storage medium in the memory 62. The storage medium may be a portable storage medium such as a flexible disk, or a flash memory such as semiconductor memory. The program may be installed onto the computer system from another computer or server device via a communications network.

The functions of the objective function update section 14, search processing section 15, and simulation section 51 are realized by a combination of the processor 61 and software. Each of the functions may be realized by a combination of the processor 61 and firmware, or may be realized by a combination of the processor 61, software, and firmware. Software or firmware is written as a program and stored in the storage device 63. Each function of the model storage section 43 is realized by using the storage device 63. Interface circuit 64 receives signals from external equipment connected to the hardware. The external devices are a numerical control device 11, a processing section 12, and a measuring device 13.

The configurations shown in each of the embodiments above are examples of the contents of the present disclosure. The configuration of each embodiment can be combined with other known techniques. The configurations of each embodiment may be combined as appropriate. It is possible to omit or change a part of the configuration of each embodiment without departing from the gist of the present disclosure.

10, 50 machining condition search device, 11 numerical control device, 12 machining unit, 13 measuring device, 14 objective function update unit, 15 search processing unit, 16 machining program analysis unit, 17 command generation unit, 18 machining condition output unit, 19 Switching unit, 20 Laser oscillator, 21 Material supply unit, 22 Axis drive unit, 23 Signal, 24 Command position information, 25 Measured value, 26 Processing condition information, 30, 30-1, 30-2, 30-(m-k ), 30-m, 30a, 30b, 30c, 30d Modeled object, 31 1st layer, 32 2nd layer, 33 3rd layer, 34-2, 34-3 Material, 41 Learning device, 42 Inference device, 43 Model Storage unit, 44, 48 Data acquisition unit, 45 Model generation unit, 46 Reward calculation unit, 47 Function update unit, 49 Inference unit, 51 Simulation unit, 61 Processor, 62 Memory, 63 Storage device, 64 Interface circuit, 100 Additive manufacturing Device.

Claims (9)

An additive manufacturing device that manufactures a shaped object by melting a filler material by beam irradiation and stacking layers that are solidified products of the melted filler material,
an objective function updating unit that updates a value of the objective function by inputting a measurement value that is a result of measuring a layer of the manufactured object into an objective function representing a shape error of the object;
a search processing unit that searches for machining conditions based on evaluation based on the value of the objective function ;
a command generation unit that generates a command indicating a group of interpolation points on the movement path of the beam ,
The objective function updating unit updates the objective function by inputting the measured value at each command position, which is the position of each interpolation point included in the interpolation point group, to the objective function,
The additive manufacturing apparatus is characterized in that the search processing section searches for the processing conditions for each of the commanded positions . 2. The objective function updating unit updates the value of the objective function by inputting the measured values for each of the plurality of layers constituting the object into the objective function. Additive manufacturing equipment as described. If the shape of the first object for which the processing conditions are being searched and the shape of the second object for which the processing conditions have already been searched are similar to each other, the search processing section 3. The additive manufacturing apparatus according to claim 1, wherein the processing conditions for the first object are searched based on the search results for the processing conditions for the second object. The additive manufacturing apparatus according to any one of claims 1 to 3 , wherein the search processing unit searches for the processing conditions that minimize the value of the objective function. Any one of claims 1 to 3 , wherein the objective function updating unit finishes updating the value of the objective function when the value of the objective function is converged within a preset range. Additive manufacturing equipment described in. The search processing unit is characterized by having an inference device that infers the machining conditions for each commanded position, which are the machining conditions that minimize the objective function, based on commanded position information indicating the commanded position. The additive manufacturing apparatus according to any one of claims 1 to 5 . The search processing unit includes a learning device that generates a trained model for inferring the machining conditions for each commanded position, which are the machining conditions that minimize the objective function,
Additive manufacturing according to claim 6 , wherein the inference device uses the learned model to infer the machining conditions for each of the commanded positions, which are the machining conditions that minimize the objective function. Device. The learning device includes a data acquisition unit that acquires the command position information, the measured value, processing condition information indicating the processing conditions, and target shape data indicating the target shape of the object;
a model generation unit that generates the learned model using a data set that includes the measured value for each commanded position, the machining condition information for each commanded position, and the target shape data for each commanded position; The additive manufacturing apparatus according to claim 7 , characterized in that it has the following. An additive manufacturing method for manufacturing a shaped object using an additive manufacturing device that melts a filler material by beam irradiation and stacks layers that are solidified products of the melted filler material, the method comprising:
updating the value of the objective function by inputting a measurement value that is a result of measuring a layer of the manufactured object into an objective function representing a shape error of the object ;
searching for processing conditions based on evaluation based on the value of the objective function ;
generating a command indicating a group of interpolation points on the movement path of the beam ,
In the step of updating the value of the objective function, the objective function is updated by inputting the measurement value at each command position, which is the position of each interpolation point included in the interpolation point group, into the objective function;
An additive manufacturing method characterized in that, in the step of searching for the processing conditions, the processing conditions are searched for each of the commanded positions .

Images