Information processing device, measurement system, molding system, information processing method, molding method, and information output device

The information processing apparatus and system address defects in additive manufacturing by using measurement and computational models to predict and analyze defects in fabricated objects, enhancing the manufacturing process's accuracy and reliability.

JP7948193B2Active Publication Date: 2026-11-11NIKON CORP
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2023548070
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-09-17
Publication Date
2026-11-11
Estimated Expiration
2041-09-17

AI Technical Summary

Technical Problem

Existing fabrication technologies face challenges in accurately predicting and addressing defects in fabricated objects during the additive manufacturing process, particularly in the formation of molten pools and solidified layers using energy beams.

Method used

An information processing apparatus and system that acquires measurement information on temperature and shape during the fabrication process, generating defect information and characteristic data using computational models and machine learning to predict and analyze defects in the manufacturing of objects through additive processing, including methods like Selective Laser Sintering (SLS) and other fabrication techniques.

Benefits of technology

Enhances the ability to detect and prevent defects in fabricated objects, improving the quality and reliability of the manufacturing process by providing real-time feedback and predictive analytics for defect prevention.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007948193000001
    Figure 0007948193000001
  • Figure 0007948193000002
    Figure 0007948193000002
  • Figure 0007948193000003
    Figure 0007948193000003
Patent Text Reader

Abstract

This information processing device includes: an acquiring unit that acquires measurement information pertaining to molding of a molding object, the measurement information being obtained during a molding period of the molding object molded by solidifying a particulate metallic material melted by being irradiated with an energy beam; and an information processing unit that uses the measurement information and generates defect information pertaining to a defect formed inside the molding object. The measurement information includes information pertaining to the temperature of at least a portion of a molten pool formed by melting the metallic material by irradiating the material with the energy beam.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present invention relates to, for example, the technical field related to fabrication of a fabricated object.

BACKGROUND ART

[0002] An example of a fabrication apparatus capable of fabricating a fabricated object is described in Patent Document 1. One of the technical problems of such a fabrication apparatus is to appropriately fabricate the fabricated object.

PRIOR ART DOCUMENTS

PATENT DOCUMENTS

[0003]

PATENT DOCUMENT 1

SUMMARY OF THE INVENTION

[0004] According to a first aspect, there is provided an information processing apparatus comprising: an acquisition unit that acquires measurement information related to fabrication of a fabricated object, the measurement information being obtained during a fabrication period of the fabricated object that is fabricated by solidification of a powdery metal material melted by irradiation with an energy beam; and an information processing unit that generates defect information related to a defect formed inside the fabricated object, using the measurement information acquired by the acquisition unit, wherein the measurement information includes information related to a temperature of at least a part of a molten pool formed by melting of the metal material due to the irradiation with the energy beam.

[0005] According to a second aspect, there is provided a measurement system comprising: the information processing apparatus provided by the first aspect; and a measurement apparatus, wherein the measurement information is obtained using the measurement apparatus.

[0006] According to a third aspect, there is provided a fabrication system comprising: the information processing apparatus provided by the first aspect; a fabrication apparatus capable of fabricating a fabricated object by solidification of a powdery metal material melted by irradiation with an energy beam; and a measurement apparatus, wherein the measurement information is obtained using the measurement apparatus.

[0007] According to a fourth embodiment, an information processing device is provided, comprising: an acquisition unit that acquires measurement information relating to the molding of a molded object obtained during the molding period of a molded object formed by the solidification of a molding material melted by irradiation with an energy beam; and an information processing unit that generates characteristic information relating to the characteristics of the molded object using the measurement information acquired by the acquisition unit.

[0008] According to a fifth aspect, an information processing device is provided, comprising: an acquisition unit that acquires molding information relating to the molding of a molded object formed by the solidification of a molding material melted by irradiation with an energy beam; and an information processing unit that generates characteristic information relating to the characteristics of the molded object using the molding information acquired by the acquisition unit.

[0009] According to the sixth aspect, an information processing method is provided which includes acquiring measurement information relating to the formation of a molded object obtained during the formation period of a molded object formed by the solidification of a molding material melted by irradiation with an energy beam, and generating characteristic information relating to the characteristics of the molded object using the acquired measurement information.

[0010] According to a seventh aspect, an information processing method is provided which includes acquiring molding information relating to the molding of a molded object formed by the solidification of a molding material melted by irradiation with an energy beam, and generating characteristic information relating to the characteristics of the molded object using the acquired molding information.

[0011] According to the eighth aspect, a molding method is provided which includes molding an object by solidifying a molding material melted by irradiation with an energy beam, performing measurements related to the molding of the object during the molding period of the object, and acquiring characteristic information regarding the properties of the molded object generated using the measurement information obtained from the measurements.

[0012] According to the ninth aspect, a fabrication method is provided which includes fabricating an object by solidifying a fabrication material that has been melted by irradiation with an energy beam, and acquiring characteristic information relating to the properties of the object, which is generated using fabrication information relating to the fabrication of the object.

[0013] According to the tenth embodiment, an information output device is provided which includes an acquisition unit that acquires measurement information relating to the formation of a molded object and characteristic information relating to the characteristics of the molded object, obtained during the formation period of a molded object formed by the solidification of a powdered metal material melted by irradiation with an energy beam, and an output unit that outputs the characteristic information together with the measurement information.

[0014] According to the eleventh embodiment, an information output device is provided which includes an acquisition unit that acquires energy information relating to the amount of energy transmitted from the energy beam to the metal material in the formation of a molded object by solidifying a powdered metal material molten by irradiation with an energy beam, and characteristic information relating to the characteristics of the molded object, and an output unit that outputs the characteristic information together with the energy information.

[0015] The effects and other benefits of the present invention will be revealed from the embodiments described below. [Brief explanation of the drawing]

[0016] [Figure 1] Figure 1 is a block diagram showing the overall configuration of the molding system of this embodiment. [Figure 2] Figure 2 is a block diagram showing the configuration of the molding apparatus in this embodiment. [Figure 3] Figure 3 is a cross-sectional view showing the configuration of the molding apparatus according to this embodiment. [Figure 4] Figure 4 is a block diagram showing the configuration of the measuring device according to this embodiment. [Figure 5] Figure 5 shows an example of a temperature image output by the temperature measuring device of this embodiment. [Figure 6] Figure 6 shows an example of a shape image output by the shape measuring device of this embodiment. [Figure 7] FIG. 7 is a block diagram showing the configuration of the information processing apparatus of the present embodiment. [Figure 8] FIG. 8 is a cross-sectional view showing voids, which are an example of defects generated inside a shaped object. [Figure 9] FIG. 9 is a flowchart showing the flow of defect prediction processing executed by the information processing apparatus. [Figure 10] FIG. 10 shows measurement information, apparatus information, and feature quantity information stored in an information DB in association with position information indicating a shaping position. [Figure 11] FIG. 11 illustrates an output method for outputting defect information associated with position information. [Figure 12] FIG. 12 illustrates an output method for outputting process data together with defect information. [Figure 13] FIG. 13 is a cross-sectional view for explaining the penetration geometry index. [Figure 14] FIG. 14 illustrates an output method for outputting energy information together with defect information. [Figure 15] FIG. 15 shows defect statistical information. [Figure 16] FIG. 16 shows an example of a method for displaying defect information. [Figure 17] FIG. 17 shows an example of a method for displaying defect information. [Figure 18] FIG. 18 is a block diagram showing the configuration of the model generation apparatus of the present embodiment. [Figure 19] FIG. 19 is a flowchart showing the flow of model generation processing executed by the model generation apparatus. [Figure 20] FIG. 20 shows an example of the data structure of a learning dataset. [Figure 21] FIG. 21 shows learning data that satisfies the extraction criteria related to shaping condition information. [Figure 22] FIG. 22 shows learning data that satisfies the extraction criteria related to shaping condition information. [Figure 23] FIG. 23 shows learning data that satisfies the extraction criteria related to design information. [Figure 24]Figure 24 is a perspective view showing multiple divided objects with different shapes, obtained by dividing a fabricated object according to its shape. [Figure 25] Figure 25 shows training data that meets the extraction criteria for material information. [Figure 26] Figure 26 shows another example of characteristic information. [Figure 27] Figure 27 shows another example of characteristic information. [Figure 28] Figures 28(a) through 28(c) are perspective views showing the solidified layer. [Modes for carrying out the invention]

[0017] The following describes embodiments of the information processing device, measurement system, molding system, information processing method, molding method, and information output device with reference to the drawings. The following describes embodiments of the information processing device, measurement system, molding system, information processing method, molding method, and information output device using the SYS molding system, which is capable of molding objects by additive processing. In particular, the following describes embodiments of the information processing device, measurement system, molding system, information processing method, molding method, and information output device using the SYS molding system, which is capable of molding objects by additive processing based on powder bed fusion methods such as Selective Laser Sintering (SLS).

[0018] However, the SYS fabrication system may also fabricate objects by performing additive processing based on a fabrication method other than powder bed fusion. Examples of fabrication methods other than powder bed fusion include at least one of the following: directed energy deposition (DED) methods such as laser metal deposition (LMD), binder jetting, material jetting, stereolithography, and laser metal fusion (LMF). Alternatively, the SYS fabrication system may also fabricate objects by performing any processing other than additive processing (for example, removal processing).

[0019] Furthermore, the following explanation describes the positional relationships of the various components that make up the SYS modeling system using the XYZ Cartesian coordinate system, which is defined by the mutually orthogonal X, Y, and Z axes. For the sake of clarity, the following explanation assumes that the X-axis and Y-axis directions are horizontal (i.e., predetermined directions within the horizontal plane), and the Z-axis direction is vertical (i.e., a direction perpendicular to the horizontal plane, essentially the up and down direction). The rotational directions (in other words, tilt directions) around the X, Y, and Z axes are referred to as the θX direction, θY direction, and θZ direction, respectively. Here, the Z-axis direction may also be defined as the direction of gravity. The XY plane may also be defined as the horizontal direction.

[0020] (1) Configuration of the SYS modeling system First, we will explain the configuration of the SYS modeling system.

[0021] (1-1) Configuration of the SYS modeling system First, we will explain the overall configuration of the SYS 3D printing system, referring to Figure 1. Figure 1 is a block diagram showing the overall configuration of the SYS 3D printing system.

[0022] As shown in Figure 1, the molding system SYS comprises a molding device 1, a measuring device 2, and an information processing device 3. The information processing device 3 may be able to communicate with at least one of the molding device 1 and the measuring device 2 via a communication network or data bus (not shown). In other words, the information processing device 3 may be able to transmit arbitrary information to at least one of the molding device 1 and the measuring device 2 via a communication network or data bus (not shown). The information processing device 3 may also be able to receive arbitrary information from at least one of the molding device 1 and the measuring device 2 via a communication network or data bus (not shown).

[0023] The molding apparatus 1 is a device capable of forming objects by performing additive processing based on powder bed fusion bonding. The configuration of the molding apparatus 1 will be explained in detail later with reference to Figures 2 and 3, but a brief overview will be given. The molding apparatus 1 forms a material layer ML (see Figure 3 below) using unmelted molding material. The molding apparatus 1 irradiates at least a portion of the material layer ML with molding light EL. As a result, the molding apparatus 1 forms a molten pool MP (see Figure 3 below) containing the molding material melted by the irradiation of the molding light EL in at least a portion of the material layer ML. In other words, the molten pool MP is formed by the melting of the molding material due to irradiation of the molding light EL. Subsequently, when the molding light EL is no longer irradiated onto the molten pool MP, the molding material that was melted in the molten pool MP cools and solidifies. As a result, a molded object containing the solidified molding material is formed. The SYS fabrication system repeats a series of fabrication processes, including the formation of a molten pool MP by irradiation with a fabrication light (EL) and the solidification of the molten fabrication material, while moving the irradiation position of the fabrication light (EL) on the material layer (ML). As a result, a layered fabricated object (hereinafter referred to as the "solidified layer STL (see Figure 3 below)") composed of solidified fabrication material is formed. Subsequently, the fabrication apparatus 1 fabricates an object containing multiple solidified layers STL by repeating the formation of the material layer (ML) and the solidified layer STL.

[0024] The measuring device 2 is a device capable of measuring (in other words, measuring) at least a portion of the object being measured. The measuring device 2 is a device capable of measuring at least a portion of the characteristics of the object being measured. For example, the measuring device 2 may be capable of measuring at least a portion of the shape (particularly the three-dimensional shape) of the object being measured as a characteristic of at least a portion of the object being measured. For example, the measuring device 2 may be capable of measuring at least a portion of the temperature of the object being measured as a characteristic of at least a portion of the object being measured.

[0025] The objects to be measured may include any objects related to the creation of the object by the molding device 1.

[0026] For example, the object to be measured may include the object itself that is fabricated by the fabrication device 1. The object fabricated by the fabrication device 1 may include a three-dimensional structure ST having a desired shape that the fabrication device 1 is ultimately supposed to fabricate (i.e., an object composed of multiple solidification layer STLs). The object fabricated by the fabrication device 1 may include intermediate objects fabricated by the fabrication device 1 in the process of fabricating the three-dimensional structure ST. The intermediate objects may include a three-dimensional structure ST in the process of fabrication (for example, an object composed of at least one solidification layer STL). The intermediate objects may include the solidification layer STL itself after fabrication is complete. The intermediate objects may include intermediate objects fabricated by the fabrication device 1 in the process of fabricating the solidification layer STL (i.e., a solidification layer STL in the process of fabrication). In the following explanation, for the sake of clarity, when it is not necessary to distinguish between the types of objects created by the molding device 1, the objects created by the molding device 1 will be collectively referred to as "object BO". In other words, in the following explanation, unless otherwise specified, object BO may mean at least one of the following: a completed three-dimensional structure ST, a three-dimensional structure ST in the process of being created, a completed solidified layer STL, and a solidified layer STL in the process of being created. The object being measured may include at least one of the following: spatter (solidified granular molding material) and fume (vapor of the molding material), which are generated from the molten pool MP during the process in which the molding device 1 melts the unmelted molding material. Object BO may mean at least one of spatter and fume.

[0027] For example, since a material layer ML is formed to create the object BO, the object to be measured may include the material layer ML. For example, since a molten pool MP is formed to create the object BO, the object to be measured may include the molten pool MP.

[0028] The information processing device 3 is a device capable of generating characteristic information regarding the characteristics of the molded object BO manufactured by the molding device 1, using the measurement results of the object to be measured by the measuring device 2. In this embodiment, an example of how the information processing device 3 generates defect information 425 (see Figure 5, etc., described later) regarding defects formed on the molded object BO manufactured by the molding device 1 will be described as an example of characteristic information. In other words, in this embodiment, an example of how the information processing device 3 performs defect prediction processing to generate defect information 425 will be described. Note that the defect prediction processing can also be called a defect prediction operation.

[0029] Furthermore, the information processing device 3 may be considered to generate characteristic information together with the measuring device 2. In this case, the system comprising the information processing device 3 and the measuring device 2 may be referred to as an information processing system or a measurement system.

[0030] (1-2) Configuration of the molding device 1 Next, the configuration of the molding apparatus 1 will be described with reference to Figures 2 and 3. Figure 2 is a block diagram showing the configuration of the molding apparatus 1. Figure 3 is a cross-sectional view showing the configuration of the molding apparatus 1.

[0031] As shown in Figures 2 and 3, the molding apparatus 1 comprises a molding unit 10 and a control unit 16. Furthermore, the molding unit 10 comprises a material supply tank 11, a recoater 12, a recoater drive system 121, a molding tank 13, a lifting stage 14, a stage drive system 141, a molding head 15, and a head drive system 151.

[0032] The material supply tank 11 is a container for holding molding materials. The bottom surface 111 of the material supply tank 11 is movable vertically (i.e., in the Z-axis direction) by a drive system (not shown). The material supply tank 11 can also be called a material supply unit or a material supply chamber.

[0033] The molding material is a powder. In other words, the molding material is a powder. However, the molding material does not have to be a powder. For example, at least one of a wire-type molding material, a gel-type molding material, a liquid-type molding material, a sol-type material, and a gaseous molding material may be used as the molding material. Furthermore, as described above, the molding system SYS forms the molded object BO by melting the molding material using a molding light EL, which is an energy beam. For this reason, the molding material is a material that can be melted by irradiation with a molding light EL of a predetermined intensity or higher. For example, a metal material can be used as such a molding material. However, other materials other than metal materials may be used as the molding material. For example, a resin material may be used as the molding material, or other well-known materials may be used.

[0034] The recoater 12 supplies the molding material contained in the material supply tank 11 to the molding tank 13. Specifically, the recoater 12 supplies the molding material so that a layer of molding material having a predetermined thickness Δd and a nearly flat surface (i.e., material layer ML) is formed on the lifting stage 14 inside the molding tank 13. The molding tank 13 is a working container (molding space) for molding a three-dimensional structure ST in which multiple solidified layers STL are stacked by repeatedly forming material layers ML and solidifying at least a portion of the formed material layers ML to form solidified layers STL. The bottom surface of the molding tank 13 is made up of a lifting stage 14. The lifting stage 14 moves vertically (i.e., in the Z-axis direction) by the stage drive system 141 according to the control of the control unit 16, which will be described later. The molding tank 13 can be called a molding section or a molding chamber.

[0035] The build head 15 irradiates at least a portion of the material layer ML with build light EL generated by a light source (not shown). Furthermore, the build head 15 may move the irradiation position of the build light EL on the surface of the material layer ML using a galvanometer mirror (not shown) as a head drive system 151. Furthermore, the build head 15 may move the irradiation position of the build light EL on the surface of the material layer ML by moving with a motor (not shown) as a head drive system 151.

[0036] The control unit 16 may include, for example, an arithmetic unit and a memory device. The arithmetic unit may include, for example, at least one of a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The memory device may include, for example, memory. The control unit 16 functions as a device that controls the operation of the molding unit 10 by having the arithmetic unit execute a computer program. This computer program is a computer program that causes the arithmetic unit to perform (i.e., execute) the processing that the control unit 16 should perform, as described later. In other words, this computer program is a computer program that causes the control unit 16 to function in such a way that the molding unit 10 performs the operations described later. The computer program executed by the arithmetic unit may be recorded in a memory device (i.e., recording medium) provided by the control unit 16, or it may be recorded in any storage medium (e.g., a hard disk or semiconductor memory) that is built into or can be attached to the control unit 16. Alternatively, the arithmetic unit may download the computer program to be executed from an external device of the control unit 16 via a network interface.

[0037] The control unit 16 does not have to be located inside the molding apparatus 1. For example, the control unit 16 may be located outside the molding apparatus 1 as a server or the like. In this case, the control unit 16 and the molding apparatus 1 may be connected by a wired and / or wireless network (or a data bus and / or communication line). As a wired network, a network using a serial bus interface, such as at least one of IEEE1394, RS-232x, RS-422, RS-423, RS-485, and USB, may be used. As a wired network, a network using a parallel bus interface may be used. As a wired network, a network using an Ethernet® compliant interface, such as at least one of 10BASE-T, 100BASE-TX, and 1000BASE-T, may be used. As a wireless network, a network using radio waves may be used. An example of a network using radio waves is a network compliant with IEEE802.1x (for example, at least one of wireless LAN and Bluetooth®). A wireless network using infrared light may be used. A wireless network using optical communication may be used. In this case, the control unit 16 and the molding device 1 may be configured to enable the transmission and reception of various types of information via the network. The control unit 16 may also be able to transmit information such as commands and control parameters to the molding device 1 via the network. The molding device 1 may be equipped with a receiving device that receives information such as commands and control parameters from the control unit 16 via the network. The molding device 1 may be equipped with a transmitting device (i.e., an output device that outputs information to the control unit 16) that transmits information such as commands and control parameters to the control unit 16 via the network. Alternatively, a first control device that performs some of the processing performed by the control unit 16 may be provided inside the molding device 1, while a second control device that performs other parts of the processing performed by the control unit 16 may be provided outside the molding device 1.

[0038] The control unit 16 may implement a computational model that can be constructed by machine learning by executing a computer program on the computing device. An example of a computational model that can be constructed by machine learning is a computational model that includes a neural network (so-called artificial intelligence (AI)). In this case, the learning of the computational model may include learning the parameters of the neural network (for example, at least one of the weights and biases). The control unit 16 may use the computational model to control the operation of the molding apparatus 1. That is, the operation of controlling the operation of the molding apparatus 1 may include the operation of controlling the operation of the molding apparatus 1 using the computational model. The control unit 16 may implement a computational model that has been constructed by offline machine learning using training data. In addition, the computational model implemented in the control unit 16 may be updated by online machine learning on the control unit 16. Alternatively, the control unit 16 may use a computational model implemented in an external device (i.e., a device provided outside the molding apparatus 1) in addition to or instead of the computational model implemented in the control unit 16 to control the operation of the molding apparatus 1.

[0039] Furthermore, the recording medium for recording the computer program executed by the control unit 16 may be at least one of the following: optical discs such as CD-ROM, CD-R, CD-RW, flexible disk, MO, DVD-ROM, DVD-RAM, DVD-R, DVD+R, DVD-RW, DVD+RW, and Blu-ray®; magnetic media such as magnetic tape; magneto-optical disks; semiconductor memory such as USB memory; and any other medium capable of storing a program. The recording medium may also include equipment capable of recording the computer program (for example, a general-purpose or dedicated device on which the computer program is implemented in a state in which it can be executed in at least one form such as software and firmware). Moreover, each process and function included in the computer program may be realized by logical processing blocks realized within the control unit 16 when the control unit 16 (i.e., the computer) executes the computer program, or by hardware such as a predetermined gate array (FPGA, ASIC) provided in the control unit 16, or in a form in which logical processing blocks and partial hardware modules that realize some elements of the hardware are mixed.

[0040] The control unit 16 may control the operation of the molding apparatus 1 (particularly the molding unit 10) so that the molding apparatus 1 (particularly the molding unit 10) performs the operations described below to create the molded object BO.

[0041] First, the recoater 12 supplies the molding material from the material supply tank 11 to the molding tank 13, thereby forming a material layer ML inside the molding tank 13 (for example, on the lifting stage 14). Specifically, the stage drive system 141 moves the lifting stage 14 of the molding tank 13 downward by the thickness Δd of the material layer ML (towards -Z in the example shown in Figure 3). At the same time, the bottom surface 111 of the material supply tank 11 moves upward by a predetermined amount (towards +Z in the example shown in Figure 3). Subsequently, the recoater 12 supplies the molding material from the material supply tank 11 to the molding tank 13, and forms the material layer ML by leveling the supplied layer of molding material to a predetermined thickness Δd.

[0042] Subsequently, the build head 15 irradiates at least a portion of the formed material layer ML with build light EL to form a molten pool MP in at least a portion of the material layer ML. Here, as described above, the build head 15 moves the irradiation position of the build light EL on the surface of the material layer ML. In other words, the build head 15 scans at least a portion of the surface of the material layer ML with the build light EL. Specifically, the path that the build head 15 scans with the build light EL is set from tomographic model information generated based on the design information of the three-dimensional structure ST to be built by the build apparatus 1 (e.g., three-dimensional shape data such as CAD data and STL (Stereolithography) data). Accordingly, the control unit 16 sets a target build region in the material layer ML where the build material should be melted and then solidified, based on the shape information of the tomographic model corresponding to the solidification layer STL that the build apparatus 1 is trying to build. The build head 15 moves the irradiation position of the build light EL so that the build light EL is irradiated onto the target build region. As a result, a solidified layer STL is fabricated in the material layer ML at the location where the fusion light EL is irradiated. In other words, the solidified layer STL is fabricated from the material layer ML. To put it another way, the solidified layer STL is formed by the melting and solidification of at least a portion of the material layer ML. The thickness of the solidified layer STL is typically the same as the thickness Δd of the material layer ML, but it may be different from the thickness Δd of the material layer ML.

[0043] When a solidified layer STL is formed from the fabrication material (i.e., material layer ML) supplied onto the lifting stage 14, the lifting stage 14 moves downward (towards -Z in the example shown in Figure 3). Subsequently, the recoater 12 forms a new material layer ML on the upper surface of the solidified layer STL (on the +Z side in the example shown in Figure 3). A new solidified layer STL is formed when at least a portion of this new material layer ML solidifies. This process of forming material layers ML and solidified layers STL is repeated, resulting in the stacking of multiple solidified layers STL. As a result, a three-dimensional structure ST containing multiple stacked solidified layers STL is fabricated.

[0044] The control unit 16 may control the operation of the molding apparatus 1 (in particular, the molding unit 10) to fabricate the object BO using the fabrication conditions that the molding apparatus 1 uses to fabricate the object BO. In other words, the control unit 16 may control the operation of the molding apparatus 1 (in particular, the molding unit 10) to fabricate the object BO using the fabrication conditions that are set to define the operation of the molding apparatus 1 for fabricating the object BO. The fabrication conditions may include irradiation conditions relating to the irradiation manner of the EL (electroluminescent light) by the molding head 15. The irradiation conditions may include at least one of the following: the output of the light source that generates the EL (electroluminescent beam), the intensity of the EL emitted from the light source, the intensity of the EL on the material layer ML, the intensity distribution of the EL on the material layer ML, the irradiation timing of the EL, the irradiation position of the EL, the movement rule of the EL irradiation position (a so-called scanning strategy, which is essentially information indicating the scanning path), the movement trajectory of the EL irradiation position, the hatching distance (hatch pitch) which is the distance between two adjacent movement trajectories, the movement speed of the EL irradiation position (i.e., scanning speed), the spot diameter of the EL emitted from the light source, and the diameter of the spot formed by the EL on the material layer ML. The molding conditions may also include recoater movement conditions relating to the movement mode of the recoater 12 by the recoater drive system 121. The molding conditions may also include stage movement conditions that define the movement mode of the lifting stage 14 by the stage drive system 141. The molding conditions may include head movement conditions that define the movement mode of the molding head 15 by the head drive system 151. The movement mode may include, for example, at least one of the amount of movement, the speed of movement, the direction of movement, and the timing of movement. The molding conditions may also include material conditions that define the properties of the molding material and / or the supply mode of the molding material. The conditions defining the properties of the molding material may include at least one of the type of molding material supplied, the particle size of the molding material supplied, the particle size distribution of the molding material supplied, the fluidity of the molding material supplied, and the angle of repose of the molding material supplied. The supply mode of the molding material may include at least one of the amount of molding material supplied, the timing of the supply of the molding material supplied, the supply speed of the molding material supplied, and the thickness Δd of the material layer ML formed by the supplied molding material.In other words, the material conditions may include at least one of the following: the amount of material supplied for the molding process, the timing of the material supply, the rate of the material supply, the thickness Δd of the material layer ML formed by the supplied material, the type of material supplied, the particle size of the material supplied, the particle size distribution of the material supplied, the fluidity of the material supplied, and the angle of repose of the material supplied. The molding conditions may also include environmental conditions relating to the molding environment in which the molding apparatus 1 molds the object BO. The molding environment may include at least one of the following: the oxygen concentration of the molding space, which is the space inside the molding unit 10 (in particular the molding tank 13) where the material layer ML is melted and the solidified layer STL is formed, the inert gas concentration of the molding space, the temperature of the molding space, the type of inert gas supplied to the molding space, and the temperature of the lifting stage 14 (in particular the temperature of the +Z side of the lifting stage 14). The molding conditions may also include design conditions for the object BO (in particular the three-dimensional structure ST) that the molding apparatus 1 is to mold. The design conditions for the printed object BO may include design data that shows the three-dimensional shape of the printed object BO. The design data may include information about the design size of a part of the printed object BO (e.g., at least one of length, width, depth, height, area, and volume). The design data may include information about the design orientation of a part of the printed object BO (e.g., inclination angle). The design data may include three-dimensional model data that shows the three-dimensional model of the printed object BO. The three-dimensional model data may include CAD (Computer Aided Design) data. The CAD data may include STL (Standard Triangulated Language) data. The design data may include slice model data that shows a sliced ​​model obtained by slicing the three-dimensional model of the printed object BO.

[0045] (1-3) Configuration of measuring device 2 Next, the configuration of the measuring device 2 will be explained with reference to Figure 4. Figure 4 is a block diagram showing the configuration of the measuring device 2.

[0046] As shown in Figure 4, the measuring device 2 comprises a temperature measuring device 21 and a shape measuring device 22. The temperature measuring device 21 is a device capable of measuring the temperature of at least a portion of the object to be measured. The shape measuring device 22 is a device capable of measuring the shape of at least a portion of the object to be measured (for example, the shape of the surface, and in particular, the three-dimensional shape). As described above, the measuring device 2 is a device capable of measuring any characteristic of at least a portion of the object to be measured. Figure 4 shows an example in which the measuring device 2 can measure the temperature of at least a portion of the object to be measured and the shape of at least a portion of the object to be measured as an example of the characteristics of at least a portion of the object to be measured. However, the measuring device 2 may be capable of measuring any characteristic of the object to be measured other than temperature and shape, in addition to or instead of temperature and shape.

[0047] The temperature measuring device 21 can be any measuring device as long as it is capable of measuring the temperature of the object to be measured. For example, the temperature measuring device 21 may measure the temperature of the object by detecting infrared radiation (i.e., light in the infrared band) emitted by the object. For example, the temperature measuring device 21 may measure the temperature of the object by detecting two types of light with different wavelengths emitted by the object. In other words, the temperature measuring device 21 may measure the temperature of the object using a two-color method (e.g., near-infrared two-color method). The temperature measuring device 21 may include an optical detection device (not shown) that includes a plurality of optical detection elements for detecting light from the object to be measured. The temperature measuring device 21 may measure the temperature of the object based on the detection results of the optical detection device. The temperature measuring device 21 may measure the temperature of the object based on the detection results of a plurality of optical detection elements of the optical detection device. For example, the temperature measuring device 21 may measure the temperature for each detection result of a plurality of optical detection elements. In other words, the temperature measuring device 21 may measure the temperature at a plurality of different locations on the object to be measured. The temperature measuring device 21 may include a CCD (Charge Coupled Device) or CMOS (Complementary Metal-Oxide-Semiconductor) as a photodetector (not shown). The temperature measuring device 21 may also include a single photodetector (not shown) for detecting light from the object being measured. The temperature measuring device 21 may measure the temperature of the object based on the detection result of the single photodetector.

[0048] The temperature measuring device 21 outputs measurement information 411 containing information about the measurement results obtained by the temperature measuring device 21. For example, the temperature measuring device 21 may output measurement information 411 containing information about the temperature of at least a portion of the object being measured. The temperature of at least a portion of the object being measured may include the temperature of one part of the object being measured. The temperature of at least a portion of the object being measured may include the temperatures of multiple parts of the object being measured. In this embodiment, an example is described in which the temperature measuring device 21 outputs temperature image data D_T showing a temperature image IMG_T as measurement information 411, as shown in Figure 5. The temperature image IMG_T is an image showing the temperature distribution of at least a portion of the object being measured. The temperature image IMG_T may be an image showing the temperature of at least a portion of the object being measured by showing the temperature differences of the object being measured by differences in color or gradation. In this case, the temperature image data D_T may be data showing the color or gradation of each of the multiple pixels constituting the temperature image IMG_T. However, the output of the temperature measuring device 21 is not limited to temperature image data showing a temperature image IMG_T. For example, the temperature measuring device 21 may output temperature data as measurement information 411, which is numerical data indicating the temperature of at least a part of the object being measured.

[0049] In this embodiment, an example is described in which the temperature measuring device 21 measures the temperature of at least a portion of a molten pool MP, which is an example of an object to be measured. In this case, the temperature measuring device 21 may output a temperature image IMG_T (specifically, temperature image data D_T) showing the temperature distribution of at least a portion of the molten pool MP as measurement information 411, as shown in Figure 5.

[0050] However, the temperature measuring device 21 may measure the temperature of at least a portion of an object to be measured that is different from the molten pool MP, in addition to or instead of the temperature of at least a portion of the molten pool MP. In other words, the temperature image IMG_T may include an object to be measured that is different from the molten pool MP. For example, when a molten pool MP is formed, sputtered particles (SPT) may be generated above (+Z direction) or around the molten pool MP due to irradiation with the fusion light (EL). In this case, as shown in Figure 5, the temperature image IMG_T may include sputtered particles (SPT). For example, when a molten pool MP is formed, fumes, which are evaporated materials of the molten material, may be generated above (+Z direction) or around the molten pool MP due to irradiation with the fusion light (EL). In this case, the temperature image IMG_T may include fumes. For example, the object to be measured may include at least one of the molten pool MP, sputtered particles (SPT), and fumes. The temperature measuring device 21 may measure the temperature of at least one of the melt pool MP, sputter SPT, and fume, or at least a portion thereof.

[0051] Furthermore, as mentioned above, the object to be measured, including the molten pool MP, is any object related to the formation of the fabricated object BO. For this reason, the measurement information 411 may be considered as measurement information related to the formation of the fabricated object BO.

[0052] The temperature measuring device 21 measures the temperature of at least a portion of the object to be measured during the molding period when the molding device 1 is molding the object BO. The molding period may mean a series of periods during which the molding device 1 repeatedly forms the material layer ML and the solidification layer STL in order to mold the three-dimensional structure ST, which is the final object BO. The molding period may mean a series of periods during which the formation of the material layer ML and the solidification layer STL are repeatedly performed in the molding tank 13 of the molding device 1 in order to mold the three-dimensional structure ST. The molding period may mean a period during which the control unit 16 controls the molding unit 10 so that the molding device 1 repeatedly forms the material layer ML and the solidification layer STL. The molding period may mean a period during which the control unit 16 controls the molding unit 10 so that the molding device 1 repeatedly forms the material layer ML and the solidification layer STL in order to mold the three-dimensional structure ST, which is the final object BO. The molding period may refer to the period from when the molding device 1 supplies material to the molding tank 13 to form the first solidified layer STL, through irradiation of each material layer ML with molding light EL, until the formation of all solidified layer STLs to be molded is completed. In the case of only one solidified layer STL to be molded, the molding period may refer to the period from when the molding device 1 supplies material to the molding tank 13 to form the first solidified layer STL, through irradiation of the first material layer ML with molding light EL, until the formation of the first solidified layer STL is completed.

[0053] The molding period may mean the period during which the molding device 1 repeatedly supplies the molding material, melts the molding material, and molds the molded object BO (i.e., the solidified object formed by the solidification of the melted molding material) resulting from the solidification of the melted molding material. The molding period may also mean the period during which the molding device 1 repeatedly supplies the molding material, melts the molding material, and molds the molded object BO resulting from the solidification of the melted molding material within the molding space where the molded object BO is molded. The molding period may also mean the period during which the control unit 16 controls the molding unit 10 so that the molding device 1 repeatedly supplies the molding material, melts the molding material, and molds the molded object BO resulting from the solidification of the melted molding material. The molding period may refer to the period during which the control unit 16 controls the molding unit 10 so that the molding apparatus 1 repeatedly supplies molding material, melts the molding material, and molds the molding material BO due to the solidification of the molten molding material in order to mold the three-dimensional structure ST, which is the final molded object BO.

[0054] Here, if the temperature measuring device 21 measures the temperature of at least a portion of the object to be measured during the molding process, the object to be measured may move relative to the measurement field of view (measurement range) of the temperature measuring device 21 as the molding of the object BO progresses. For example, the molten pool MP, which is the object to be measured by the temperature measuring device 21, moves on the material layer ML as the irradiation position of the molding light EL moves. As a result, the molten pool MP may move relative to the measurement field of view of the temperature measuring device 21. For this reason, the temperature measuring device 21 may be aligned with respect to the molding apparatus 1 so that the entire area in which the molten pool MP may be formed is included in the measurement field of view of the temperature measuring device 21. For example, because the molten pool MP is formed in the material layer ML, the temperature measuring device 21 may be aligned with respect to the molding apparatus 1 so that the entire material layer ML is included in the measurement field of view of the temperature measuring device 21. Alternatively, the temperature measuring device 21 may move its measurement field of view so that the molten pool MP, whose position is changing, is included in the measurement field of view of the temperature measuring device 21. For example, if the build head 15 moves the irradiation position of the build light EL on the surface of the material layer ML using a galvanometer mirror (not shown) as a head drive system 151, the temperature measuring device 21 may be positioned in the build apparatus 1 so as to detect light from at least a portion of the molten pool MP via the same galvanometer mirror, thereby moving the measurement field of view of the temperature measuring device 21 along with the movement of the irradiation position of the build light EL. For example, the temperature measuring device 21 may move its measurement field of view by moving itself. The temperature measuring device 21 may be positioned relative to the build apparatus 1 so as to include at least one of the molten pool MP, sputter SPT, and fume in its measurement field of view. The temperature measuring device 21 may also move its measurement field of view so as to include at least one of the molten pool MP, sputter SPT, and fume whose positions are changing.

[0055] Next, the shape measuring device 22 may be any measuring device as long as it is capable of measuring the shape of the object to be measured. For example, the shape measuring device 22 may optically measure the three-dimensional shape of the object to be measured. An example of a measuring device that optically measures the three-dimensional shape of an object to be measured is a 3D scanner. A 3D scanner may measure the shape of an object by projecting patterned light onto at least a part of the object to be measured and imaging at least a part of the object onto which the patterned light is projected. Other examples of measuring devices that optically measure the three-dimensional shape of an object to be measured include at least one of an interferometric measuring device, an optical coherence tomography (OCT) measuring device, a measuring device equipped with a white confocal displacement meter, a phase modulation measuring device, and an intensity modulation measuring device. An example of an OCT measuring device is described in Japanese Patent Application Publication No. 2020-101499. An example of a white confocal displacement meter is described in Japanese Patent Application Publication No. 2020-085633. An example of a phase-modulation type measuring device is described in Japanese Patent Publication No. 2010-025922. An example of an intensity-modulation type measuring device is described in Japanese Patent Publication No. 2016-510415 and U.S. Patent Application Publication No. 2014 / 226145. The shape measuring device 22 is not limited to the above measuring device that optically measures the three-dimensional shape of the object to be measured, but may be other existing measuring devices that can measure the three-dimensional shape of the object to be measured non-contact. The shape measuring device 22 may be a measuring device that measures the three-dimensional shape of the object to be measured using ultrasound. For example, the shape measuring device 22 may optically measure the two-dimensional shape of the object to be measured. An example of a measuring device that optically measures the two-dimensional shape of the object to be measured is an image measuring machine. The image measuring machine may measure the two-dimensional shape (dimensions) of the object to be measured from the acquired image by imaging at least a part of the object to be measured. Furthermore, the shape measuring device 22 is not limited to an image measuring machine, but may be any other existing measuring device capable of measuring the two-dimensional shape of an object without contact. Furthermore, the shape measuring device 22 is not limited to a measurement method that optically measures the shape of an object, but may be any other measurement method. For example, the shape measuring device 22 may be a contact-type measuring device that measures the shape of an object by bringing a touch probe into contact with at least a part of the object.

[0056] The shape measuring device 22 outputs measurement information 412 which includes information about the measurement results obtained by the shape measuring device 22. For example, the shape measuring device 22 may output measurement information 412 which includes information about the shape of at least a part of the object being measured. The shape of at least a part of the object being measured may include the shape of a part of the surface of the object being measured. The shape of at least a part of the object being measured may include the shapes of multiple parts of the surface of the object being measured. In this embodiment, an example is described in which the shape measuring device 22 outputs shape image data D_3D, which shows a shape image IMG_3D, as the measurement information 412, as shown in Figure 6. The shape image IMG_3D is an image that shows the shape of at least a part of the object being measured. The shape image IMG_3D may be an image that shows the shape of at least a part of the object being measured by indicating, for example, differences in the position of the surface of the object being measured (i.e., differences in distance from the shape measuring device 22) with differences in color or gradation. Alternatively, the shape image IMG_3D may be an image that includes so-called depth information, for example. In this case, the shape image data D_3D may be data indicating the color or gradation of each of the multiple pixels that make up the shape image IMG_3D. However, the output of the shape measuring device 22 is not limited to the shape image data D_3D. For example, the shape measuring device 22 may output point cloud data indicating the shape of at least a part of the object being measured as measurement information 412. For example, the shape measuring device 22 may output distance data, which is numerical data indicating the distance from the shape measuring device 22 to at least a part of the object being measured, as measurement information 412.

[0057] In this embodiment, an example is described in which the shape measuring device 22 measures the shape of at least one of the molded object BO and the material layer ML, each of which is an example of an object to be measured. In this case, the shape measuring device 22 may output a shape image IMG_3D (specifically, shape image data D_3D) showing the shape of at least one of the molded object BO and the material layer ML as measurement information 412, as shown in Figure 6.

[0058] However, the shape measuring device 22 may measure the shape of at least a part of an object to be measured that is different from the molded object BO and the material layer ML, in addition to or instead of the shape of at least one of the molded object BO and the material layer ML. In other words, the shape image IMG_3D may include an object to be measured that is different from the molded object BO and the material layer ML. For example, the object to be measured may include at least one of the molded object BO, the material layer ML, and the sputtered SPT. The shape measuring device 22 may measure the shape of at least a part of at least one of the molded object BO, the material layer ML, and the sputtered SPT.

[0059] As mentioned above, the object to be measured, which includes at least one of the molded object BO and the material layer ML, is any object related to the molding of the molded object BO. For this reason, the measurement information 412 may be considered, like the measurement information 411, to be measurement information related to the molding of the molded object BO.

[0060] The shape measuring device 22 measures the shape of at least a portion of the object to be measured during the printing period while the printing device 1 is printing the object BO. Here, if the shape measuring device 22 measures the shape of at least a portion of the object to be measured during the printing period, the object to be measured may move within the measurement field of the shape measuring device 22 as the printing of the object BO progresses. For this reason, the shape measuring device 22 may be aligned with the printing device 1 so that the entire area in which the object to be measured, the object BO and the material layer ML, may exist is included in the measurement field of the shape measuring device 22. For example, since the object BO is printed from the material layer ML, the shape measuring device 22 may be aligned with the printing device 1 so that the entire material layer ML is included in the measurement field of the shape measuring device 22. Alternatively, the shape measuring device 22 may move its measurement field so that the object to be measured, the object BO and the material layer ML, are included in the measurement field of the shape measuring device 22. For example, the shape measuring device 22 may move its measurement field of view by moving itself.

[0061] (1-4) Configuration of the information processing device 3 Next, the configuration of the information processing device 3 will be explained with reference to Figure 7. Figure 7 is a block diagram showing the configuration of the information processing device 3.

[0062] As shown in Figure 7, the information processing device 3 comprises an arithmetic unit 31, a storage device 32, and a communication device 33. Furthermore, the information processing device 3 may also include an input device 34 and an output device 35. However, the information processing device 3 does not need to include at least one of the input device 34 and the output device 35. The arithmetic unit 31, the storage device 32, the communication device 33, the input device 34, and the output device 35 may be connected via a data bus 36.

[0063] The arithmetic unit 31 includes, for example, at least one of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and an FPGA (Field Programmable Gate Array). The arithmetic unit 31 reads a computer program. For example, the arithmetic unit 31 may read a computer program stored in the storage device 32. For example, the arithmetic unit 31 may read a computer program stored in a computer-readable and non-temporary recording medium using a recording medium reading device (not shown) provided by the information processing device 3. The arithmetic unit 31 may obtain a computer program from a device (not shown) located outside the information processing device 3 via a communication device 33 (or other communication device) (i.e., it may download or read the program). The arithmetic unit 31 executes the read computer program. As a result, a logical functional block for executing the processing that the information processing device 3 should perform (for example, the defect prediction processing described above) is realized within the arithmetic unit 31. In other words, the arithmetic unit 31 can function as a controller for realizing a logical functional block for executing the processing that the information processing device 3 should perform.

[0064] Figure 7 shows an example of a logical functional block implemented within the arithmetic unit 31 to perform defect prediction processing. As shown in Figure 7, the arithmetic unit 31 implements an information acquisition unit 311, a defect prediction unit 312, and a defect output unit 313. The processing performed by the information acquisition unit 311, the defect prediction unit 312, and the defect output unit 313 will be described in detail later with reference to Figure 9, etc., but an overview is given below.

[0065] The information acquisition unit 311 acquires information necessary to generate defect information 425 regarding defects formed in the molded object BO manufactured by the molding apparatus 1. For example, the information acquisition unit 311 may acquire at least one of the measurement information 411 and 412 related to the molding of the molded object BO. In other words, the information acquisition unit 311 may acquire at least one of the measurement information 411 and 412 obtained by actually measuring the object to be measured during the molding of the molded object BO. The information acquired by the information acquisition unit 311 may be stored, for example, in the information DB (Database) 321 in the storage device 32.

[0066] The defect prediction unit 312 generates defect information 425 using at least a portion of the information acquired by the information acquisition unit 311. In this embodiment, an example is described in which the defect information 425 includes information about defects formed inside the molded object BO. Defects formed inside the molded object BO may include defects that are not exposed to the outside of the molded object BO. Defects formed inside the molded object BO may include defects that are exposed to the outside of the molded object BO. The defect information 425 may also include information about defects formed on the surface (i.e., outer surface) of the molded object BO.

[0067] The information acquired by the information acquisition unit 311 is not information obtained by directly measuring defects formed on the molded object BO. Therefore, the process by which the defect prediction unit 312 generates defect information 425 can be considered to be substantially equivalent to the process of predicting defects formed on the molded object BO.

[0068] In this embodiment, the defect prediction unit 312 may generate defect information 425 using at least a portion of the information acquired by the information acquisition unit 311 and a prediction model PM. The prediction model PM is, for example, a learning model generated by machine learning so that it can generate defect information 425 based on the information acquired by the information acquisition unit 311. Information regarding such a prediction model PM may be stored in the storage device 32 as model information 322.

[0069] An example of a defect predicted by the defect prediction unit 312 is a void SP formed inside the fabricated object BO. The void SP may also be referred to as a hole, pore, void, borosity, or crack. In this case, the defect information 425 may include void information relating to the void SP. For example, Figure 8 shows an example of a void SP formed inside a three-dimensional structure ST (i.e., the final fabricated object BO), which is an example of a fabricated object BO. The void SP may include a void SP1 that forms an opening on the surface of the fabricated object BO (i.e., is exposed on the surface of the fabricated object BO). The void SP1 may also be referred to as an open void (in other words, an open pore). The void SP may also include a void SP2 that does not form an opening on the surface of the fabricated object BO (i.e., is not exposed on the surface of the fabricated object BO). The void SP2 may also be referred to as a closed void (in other words, a closed pore).

[0070] Such voids SP may be generated, for example, when the amount of energy transferred from the printing light EL to the material layer ML (i.e., the amount of energy transferred to the printing material) during the fabrication of a printed object BO is excessive or insufficient compared to the target amount. For example, if the amount of energy transferred from the printing light EL to the material layer ML is insufficient compared to the target amount, the printing material that should be melted may not melt properly. Typically, the amount of melted printing material may be insufficient. As a result, voids SP may be formed inside the printed object BO. For example, if the amount of energy transferred from the printing light EL to the material layer ML is excessive compared to the target amount, voids SP may be formed from keyholes generated during the process of melting the printing material (material layer ML) due to irradiation with the printing light EL. In this embodiment, the defect prediction unit 312 can generate defect information 425 that includes void information regarding such voids SP.

[0071] In Figure 7 again, the defect output unit 313 outputs the defect information 425 generated by the defect prediction unit 312. For example, the defect output unit 313 may output the defect information 425 using the output device 35. For example, the defect output unit 313 may output the defect information 425 using the communication device 33 which can function as an output device. Since the defect output unit 313 outputs the defect information 425, the information processing device 3 equipped with the defect output unit 313 may be referred to as an information output device.

[0072] The storage device 32 is capable of storing desired data. For example, the storage device 32 may temporarily store a computer program executed by the arithmetic unit 31. The storage device 32 may temporarily store data that the arithmetic unit 31 uses temporarily when it is executing a computer program. The storage device 32 may store data that the information processing device 3 stores long-term. The storage device 32 may include at least one of the following: RAM (Random Access Memory), ROM (Read Only Memory), hard disk drive, magneto-optical disk drive, SSD (Solid State Drive), and disk array device. In other words, the storage device 32 may include a non-temporary recording medium.

[0073] The communication device 33 can communicate with the molding device 1 and the measuring device 2, respectively, via a communication network (not shown). Alternatively, the communication device 33 may communicate with other devices different from the molding device 1 and the measuring device 2, in addition to or instead of at least one of the molding device 1 and the measuring device 2, via a communication network (not shown).

[0074] The input device 34 is a device that receives information input to the information processing device 3 from outside the information processing device 3. For example, the input device 34 may include an operating device that can be operated by the user of the information processing device 3 (for example, at least one of a keyboard, mouse, and touch panel). For example, the input device 34 may include a recording medium reader that can read information recorded as data on a recording medium that can be attached externally to the information processing device 3.

[0075] Furthermore, information can be input as data from an external device to the information processing device 3 via the communication device 33. In this case, the communication device 33 may function as an input device that receives information input to the information processing device 3 from an external source.

[0076] The output device 35 is a device that outputs information to the outside of the information processing device 3. For example, the output device 35 may output information as an image. That is, the output device 35 may include a display device (so-called display) capable of displaying images. For example, the output device 35 may output information as sound. That is, the output device 35 may include an audio device (so-called speaker) capable of outputting sound. For example, the output device 35 may output information onto paper. That is, the output device 35 may include a printing device (so-called printer) capable of printing desired information onto paper. For example, the output device 35 may output information as data to a recording medium that can be attached externally to the information processing device 3.

[0077] Furthermore, the information processing device 3 can output information as data to external devices via the communication device 33. In this case, the communication device 33 may function as an output device that outputs information to an external device of the information processing device 3.

[0078] (2) Defect prediction processing performed by the information processing device 3 Next, we will explain the defect prediction process performed by the information processing device 3, referring to Figure 9. Figure 9 is a flowchart showing the flow of the defect prediction process performed by the information processing device 3.

[0079] As shown in Figure 9, the information acquisition unit 311 acquires at least one of the measurement information 411 and 412 (step S11). However, the information acquisition unit 311 does not have to acquire at least one of the measurement information 411 and 412.

[0080] For example, the information acquisition unit 311 may use the communication device 33 to acquire measurement information 411 from the measurement device 2 that generates measurement information 411. For example, the information acquisition unit 311 may use the communication device 33 to acquire measurement information 412 from the measurement device 2 that generates measurement information 412. For example, if at least one of the measurement information 411 and 412 is stored in an external device different from the measurement device 2 (e.g., a cloud server), the information acquisition unit 311 may use the communication device 33 to acquire at least one of the measurement information 411 and 412 from the external device. For example, if at least one of the measurement information 411 and 412 is recorded on a recording medium, the information acquisition unit 311 may use an input device 34 that can function as a recording medium reader to acquire at least one of the measurement information 411 and 412 from the recording medium.

[0081] The information acquisition unit 311 may acquire at least one of the measurement information 411 and 412 during the molding period when the molding apparatus 1 is molding the molded object BO (especially the three-dimensional structure ST). An example of the definition of the molding period has already been described above. In this case, the information acquisition unit 311 may acquire the measurement information 411 obtained during the molding period by the temperature measuring device 21 measuring the temperature of at least a part of the molten pool MP during the molding period. Similarly, the information acquisition unit 311 may acquire the measurement information 412 obtained during the molding period by the shape measuring device 22 measuring the shape of at least one of the molded object BO and the material layer ML during the molding period.

[0082] The information acquisition unit 311 may acquire at least one of the measurement information 411 and 412 after the molding apparatus 1 has finished molding the molded object BO (especially the three-dimensional structure ST). In other words, the information acquisition unit 311 may acquire at least one of the measurement information 411 and 412 after the molding period has ended. In this case, the information acquisition unit 311 may acquire the measurement information 411 obtained during the molding period by the temperature measuring device 21 measuring the temperature of at least a part of the molten pool MP during the molding period, after the molding period has ended. Similarly, the information acquisition unit 311 may acquire the measurement information 412 obtained by the shape measuring device 22 measuring the shape of at least one of the molded object BO and the material layer ML during the molding period, after the molding period has ended.

[0083] The information acquisition unit 311 may associate the acquired measurement information 411 with the molding position where the molding device 1 was performing the molding process at the time the measurement information 411 was generated. Alternatively, the temperature measuring device 21 may associate the measurement information 411 with the molding position where the molding device 1 was performing the molding process at the time the temperature measuring device 21 measured the temperature, and output the measurement information 411 associated with the molding position. The information acquisition unit 311 may store the measurement information 411 associated with the molding position in the information DB 321 of the storage device 32. For this reason, the information acquisition unit 311 may acquire information regarding the molding position at the time the measurement information 411 was generated, along with the measurement information 411. Typically, the molding position may mean the position where the molding device 1 was irradiating the molding light EL. In other words, the molding position may mean the position where the molding device 1 was forming the molten pool MP. For example, if temperature image data D_T is used as measurement information 411, the information acquisition unit 311 may associate the acquired temperature image data D_T with the fabrication position at the time the temperature image data D_T was generated (i.e., the position where the molten pool MP, which is captured in the temperature image IMG_T shown by the temperature image data D_T, was formed). For example, the information acquisition unit 311 may associate the acquired measurement information 411 with the fabrication position at the time when light (light for measuring temperature) from the object to be measured (e.g., molten pool MP) was detected by the temperature measuring device 21. For example, the information acquisition unit 311 may associate the measurement information 411 calculated based on the light from the object to be measured detected by the temperature measuring device 21 at the first time point with the first molding position where the molding device 1 was performing the molding process at the first time point, and may associate the measurement information 411 calculated based on the light from the object to be measured detected by the temperature measuring device 21 at a second time point different from the first time point with the second molding position where the molding device 1 was performing the molding process at the second time point.

[0084] Similarly, the information acquisition unit 311 may associate the acquired measurement information 412 with the molding position where the molding device 1 was performing the molding process. Specifically, the information acquisition unit 311 may associate the measurement information 412 relating to the shape of one part of the object to be measured with the molding position where the molding device 1 was performing the molding process on that part of the object to be measured (i.e., the position of that part of the object to be measured). The information acquisition unit 311 may also associate the measurement information 412 relating to the shape of other parts of the object to be measured with the molding position where the molding device 1 was performing the molding process on other parts of the object to be measured (i.e., the position of other parts of the object to be measured). Alternatively, the shape measuring device 22 may associate the measurement information 412 with the molding position where the molding device 1 was performing the molding process and output the measurement information 412 associated with the molding position. The information acquisition unit 311 may store the measurement information 412 associated with the molding position in the information DB 321 of the storage device 32. Therefore, the information acquisition unit 311 may acquire information regarding the molding position at the time the measurement information 412 was generated, along with the measurement information 412. For example, if shape image data D_3D is used as the measurement information 412, the information acquisition unit 311 may associate the acquired shape image data D_3D with the molding position at the time the shape image IMG_3D was generated (that is, the molding position at the time when at least one of the molded object BO and the material layer ML reflected in the shape image IMG_3D shown by the shape image data D_3D was measured by the shape measuring device 22).

[0085] The fabrication position may be specified by position coordinates including the position in the X-axis direction, the Y-axis direction, and the Z-axis direction. Here, the fabrication position in the direction along the surface of the material layer ML (i.e., at least one of the X-axis and Y-axis directions) corresponds to the irradiation position of the fabrication light EL in the direction along the surface of the material layer ML. On the other hand, the fabrication position in the direction intersecting the surface of the material layer ML (i.e., the Z-axis direction) corresponds to the position of the surface of the material layer ML because the fabrication light EL is irradiated onto the surface of the material layer ML. However, since the lifting stage 14 moves downward by the thickness Δd of the material layer ML each time a solidification layer STL is formed, it can be said that, apparently, the position of the surface of the material layer ML in the direction intersecting the surface of the material layer ML (i.e., the Z-axis direction) does not change. For this reason, when specifying the fabrication position using a coordinate system that does not consider the movement of the lifting stage 14, there is a possibility that the fabrication position for forming one solidification layer STL and the fabrication position for forming another solidification layer STL different from one solidification layer STL may coincide in the Z-axis direction (i.e., become indistinguishable). In this case, the printing position in the direction intersecting at least one surface of the material layer ML and the solidification layer STL (i.e., the Z-axis direction) may be determined by the position in a coordinate system based on the moving lifting stage 14. For example, the printing position in the direction intersecting at least one surface of the material layer ML and the solidification layer STL (i.e., the Z-axis direction) may be determined by the distance from the surface of the lifting stage 14 to the surface of the material layer ML in the direction intersecting the surface of the material layer ML (i.e., the Z-axis direction). Alternatively, the printing position may be determined within a coordinate system used to determine the position of each part of the three-dimensional structure ST that the printing device 1 is to ultimately print. The printing position may also be expressed in coordinates in a three-dimensional coordinate system based on the moving lifting stage 14. The printing position may also be expressed in coordinates on the XY plane for each layer (a layer containing at least one of the material layer ML and the solidification layer STL).In this case, for example, measurement information 411 and 412 indicating the temperature and shape of each part of the first solidified layer STL, which is an example of the object to be measured, may be associated with a molding position expressed in coordinates relative to the first solidified layer STL, and measurement information 411 and 412 indicating the temperature and shape of each part of the second solidified layer STL, which is different from the first solidified layer STL, which is an example of the object to be measured, may be associated with a molding position expressed in coordinates relative to the second solidified layer STL. Furthermore, measurement information 411 or 412 of each part of the first solidified layer STL, which is an example of the object to be measured, may be associated with a molding position expressed in coordinates relative to the first solidified layer STL, and measurement information 411 or 412 of each part of the second solidified layer STL, which is different from the first solidified layer STL, which is an example of the object to be measured, may be associated with a molding position expressed in coordinates relative to the second solidified layer STL. Furthermore, the molding position may mean the position where the molding material was supplied in the molding tank 13. Furthermore, the fabrication location may refer to the location where the material layer ML is formed.

[0086] Around the same time as or in parallel with the processing in step S11, the information acquisition unit 311 acquires molding information 413 related to the molding of the molded object BO (step S12). The molding information 413 is information related to the molding of the molded object BO and includes information different from the information generated by the actual measurement of the object by the measuring device 2. However, the information acquisition unit 311 does not have to acquire the molding information 413.

[0087] The molding information 413 may include device information relating to the molding apparatus 1 that molds the object BO. The device information may include molding condition information (so-called recipe information) relating to the molding conditions used by the molding apparatus 1 to mold the object BO. Specific examples of molding conditions have already been explained. The device information may also include operation status information relating to the operating status of the molding apparatus 1. In other words, the device information may include operation status information relating to the actual operating status of the molding apparatus 1 operating according to preset molding conditions. The operation status information may include information relating to the actual irradiation pattern of the molding light EL by the molding head 15 operating according to the molding conditions. The operation status information may include information relating to the actual movement pattern of the recoater 12 by the recoater drive system 121 operating according to the molding conditions. The operation status information may include information relating to the actual movement pattern of the lifting stage 14 by the stage drive system 141 operating according to the molding conditions. The operational status information may include information about the actual movement of the build head 15 by the head drive system 151 operating according to the build conditions. The operational status information may include information about the actual supply of build material by the recoater 12 operating according to the build conditions. The operational status information may include information about the actual conditions of the build environment in which the build apparatus 1 builds the build object BO.

[0088] The information acquisition unit 311 may acquire the molding information 413 from the molding device 1 using the communication device 33. If the molding information 413 is stored in an external device different from the molding device 1 (for example, a cloud server), the information acquisition unit 311 may acquire the molding information 413 from the external device using the communication device 33. If the molding information 413 is recorded on a recording medium, the information acquisition unit 311 may acquire the molding information 413 from the recording medium using an input device 34 that can function as a recording medium reader.

[0089] The information acquisition unit 311 may acquire the molding information 413 during the molding period when the molding device 1 is molding the molded object BO (especially the three-dimensional structure ST). The information acquisition unit 311 may acquire the molding information 413 after the molding period has ended. The information acquisition unit 311 may acquire the molding information 413 before the molding period (before the molding of the molded object BO begins).

[0090] The information acquisition unit 311 may associate the acquired molding information 413 with the molding position where the molding device 1 was performing the molding process, similar to the measurement information 411 and 412. Specifically, the information acquisition unit 311 may associate the molding information 413 relating to the molding of one part of the molded object BO with the molding position where the molding device 1 was performing the molding process for that part of the molded object BO (i.e., the position of that part of the molded object BO). The information acquisition unit 311 may also associate the molding information 413 relating to the molding of other parts of the molded object BO with the molding position where the molding device 1 was performing the molding process for other parts of the molded object BO (i.e., the position of other parts of the molded object BO). Alternatively, a device other than the information processing device 3 may associate the molding information 413 with the molding position where the molding device 1 performs the molding process, and the information acquisition unit 311 may acquire the molding information 412 associated with the molding position. The information acquisition unit 311 may store the molding information 413 associated with the molding position in the information DB 321 of the storage device 32.

[0091] For example, the information acquisition unit 311 may associate the molding condition information included in the molding information 413 with the molding position where the molding device 1 was performing the molding process using the molding conditions indicated by the molding condition information. For example, if the molding information 413 includes irradiation conditions, which are an example of molding conditions, the information acquisition unit 311 may associate the information regarding the irradiation conditions included in the molding information 413 with the molding position where the molding device 1 was performing the molding process using molding light EL corresponding to the irradiation conditions. For example, if the molding information 413 includes recoater movement conditions, which are an example of molding conditions, the information acquisition unit 311 may associate the information regarding the recoater movement conditions included in the molding information 413 with the molding position where the molding device 1 was performing the molding process while moving the recoater 12 according to the recoater movement conditions. For example, if the molding information 413 includes stage movement conditions, which are an example of molding conditions, the information acquisition unit 311 may associate the information regarding the stage movement conditions included in the molding information 413 with the molding position where the molding device 1 was performing the molding process while moving the lifting stage 14 according to the stage movement conditions. For example, if the molding information 413 includes head movement conditions, which are an example of molding conditions, the information acquisition unit 311 may associate the information regarding the head movement conditions included in the molding information 413 with the molding position where the molding device 1 was performing the molding process while moving the molding head 15 according to the head movement conditions. For example, if the molding information 413 includes material conditions, which are an example of molding conditions, the information acquisition unit 311 may associate the information regarding the material conditions included in the molding information 413 with the molding position where the molding device 1 was performing the molding process while supplying molding material according to the material conditions. For example, if the molding information 413 includes environmental conditions, which are an example of molding conditions, the information acquisition unit 311 may associate the information regarding the environmental conditions included in the molding information 413 with the molding location where the molding device 1 was performing the molding process under the environmental conditions indicated by the environmental conditions. For example, if the molding information 413 includes design conditions, which are an example of molding conditions, the information acquisition unit 311 may associate the information regarding the design conditions included in the molding information 413 with the molding location where the molding device 1 was performing the molding process to mold the object BO indicated by the design conditions.

[0092] For example, the information acquisition unit 311 may associate the operation status information included in the molding information 413 with the molding position where the molding device 1 was performing the molding process in the operation status indicated by the operation status information. For example, if the operation status information includes information about the actual irradiation pattern of the molding light EL, the information acquisition unit 311 may associate the operation status information included in the molding information 413 with the molding position where the molding device 1 was performing the molding process by actually irradiating the molding light EL in the irradiation pattern indicated by the operation status information. For example, if the operation status information includes information about the actual movement pattern of the recoater 12, the information acquisition unit 311 may associate the operation status information included in the molding information 413 with the molding position where the molding device 1 was performing the molding process by actually moving the recoater 12 in the movement pattern indicated by the operation status information. For example, if the operation status information includes information about the actual movement pattern of the lifting stage 14, the information acquisition unit 311 may associate the operation status information included in the molding information 413 with the molding position where the molding device 1 was performing the molding process by actually moving the lifting stage 14 in the movement pattern indicated by the operation status information. For example, if the operation status information includes information about the actual movement pattern of the molding head 15, the information acquisition unit 311 may associate the operation status information included in the molding information 413 with the molding position where the molding device 1 was performing the molding process by actually moving the molding head 15 in the movement pattern indicated by the operation status information. For example, if the operation status information includes information about the actual supply pattern of the molding material, the information acquisition unit 311 may associate the operation status information included in the molding information 413 with the molding position where the molding device 1 was performing the molding process by actually supplying the molding material in the supply pattern indicated by the operation status information.

[0093] Subsequently, the defect prediction unit 312 calculates (in other words, extracts) the feature quantities of the object to be measured based on at least one of the measurement information 411 and 412 acquired in step S11 (step S13). However, the defect prediction unit 312 does not have to calculate the feature quantities of the object to be measured.

[0094] The defect prediction unit 312 may calculate feature quantities during the molding period when the molding apparatus 1 is molding the object BO (especially the three-dimensional structure ST). In this case, the defect prediction unit 312 may calculate feature quantities during the molding period based on measurement information 411 and 412 obtained during the molding period by the measurement device 2 measuring the object to be measured during the molding period.

[0095] The defect prediction unit 312 may calculate feature quantities after the molding period has ended. In this case, the defect prediction unit 312 may calculate feature quantities after the molding period has ended based on the measurement information 411 and 412 obtained during the molding period by the measurement device 2 measuring the object to be measured during the molding period.

[0096] Furthermore, since the characteristic quantities of the object being measured can be calculated based on at least one of the measurement information 411 and 412, at least one of the measurement information 411 and 412 may be considered to contain information regarding the characteristic quantities of the object being measured. In other words, at least one of the measurement information 411 and 412 may be considered to contain information regarding the characteristic quantities of the object being measured obtained during the molding period in which the molding apparatus 1 molds the molded object BO.

[0097] For example, the temperature image IMG_T, which is measurement information 411, contains information about the temperature of at least a portion of the molten pool MP. The temperature of the molten pool MP is much higher than the temperature of other objects (e.g., material layer ML or molded object BO). Therefore, within the temperature image IMG_T, the molten pool MP and other objects can be distinguished by their temperature distribution. For this reason, the defect prediction unit 312 may calculate feature quantities of the molten pool MP based on the measurement information 411. For example, the defect prediction unit 312 may calculate an index value indicating the shape of the molten pool MP as a feature quantity of the molten pool MP. For example, the defect prediction unit 312 may calculate an index value indicating the size of the molten pool MP as a feature quantity of the molten pool MP. Examples of size include length, which is the size along a first direction; width, which is the size along a second direction intersecting the first direction; depth (in other words, height or thickness), which is the size along a third direction intersecting the first and second directions; surface area; and at least one of volume. For example, the defect prediction unit 312 may calculate an index value indicating the temperature of the molten pool MP (for example, at least one of the average temperature, maximum temperature, minimum temperature, temperature variation, and temperature distribution pattern) as a feature quantity of the molten pool MP.

[0098] For example, if the measurement information 411 includes information regarding the temperature of at least a part of the object being measured that is different from the molten pool MP, the defect prediction unit 312 may calculate feature quantities of the object being measured that is different from the molten pool MP based on the measurement information 411. For example, as described above, the temperature image IMG_T may include sputter SPT generated from the molten pool MP due to irradiation by the fusion light EL. In this case, the defect prediction unit 312 may calculate feature quantities of the sputter SPT based on the measurement information 411. For example, the defect prediction unit 312 may calculate the number of sputters captured in the temperature image IMG_T as a feature quantity of the sputter SPT. For example, the defect prediction unit 312 may calculate an index value indicating the size of the sputters captured in the temperature image IMG_T (for example, at least one of the average size, maximum size, and minimum size) as a feature quantity of the sputter SPT. For example, the defect prediction unit 312 may calculate an index value indicating the temperature of the sputter captured in the temperature image IMG_T (for example, at least one of the average temperature, maximum temperature, minimum temperature, temperature variation, and temperature distribution pattern) as a feature quantity of the sputter SPT. Also, as described above, the temperature image IMG_T may capture fumes generated due to irradiation with the EL (electroluminescent beam). In this case, the defect prediction unit 312 may calculate a feature quantity of the fume based on the measurement information 411. For example, the defect prediction unit 312 may calculate an index value indicating the size of the fume captured in the temperature image IMG_T (for example, at least one of the area and volume) as a feature quantity of the fume. The defect prediction unit 312 may calculate an index value indicating the temperature of the fume captured in the temperature image IMG_T (for example, at least one of the average temperature, maximum temperature, minimum temperature, temperature variation, and temperature distribution pattern) as a feature quantity of the fume.

[0099] Furthermore, the feature quantities calculated based on the temperature image data D_T (i.e., feature quantities related to the temperature of the object being measured) may be calculated by the temperature measuring device 21 in addition to, or instead of, the defect prediction unit 312. In this case, the temperature measuring device 21 may output measurement information 411 including the calculated feature quantities.

[0100] For example, the shape image data D_3D, which is measurement information 412, contains information about the shape of the molded object BO. Therefore, the defect prediction unit 312 may calculate the feature quantities of the molded object BO based on the measurement information 412.

[0101] As an example, the fabricated object BO may include a three-dimensional structure ST that has been fabricated or is in the process of being fabricated. In this case, the defect prediction unit 312 may calculate feature quantities of the three-dimensional structure ST that has been fabricated or is in the process of being fabricated based on the measurement information 412. For example, the defect prediction unit 312 may calculate an index value indicating the surface condition of the three-dimensional structure ST as a feature quantity of the three-dimensional structure ST. An example of an index value indicating the surface condition is surface roughness (in other words, surface texture) which indicates the degree of surface irregularities. An example of surface roughness is at least one of the arithmetic mean height Sa, the root mean square height Sq, and the maximum height Sz. For example, the defect prediction unit 312 may calculate an index value indicating the size of the three-dimensional structure ST (for example, at least one of length, width, height (depth or thickness), surface area, and volume, as described above) as a feature quantity of the three-dimensional structure ST.

[0102] As another example, as mentioned above, the molded object BO may include a solidified layer STL that has been molded or is in the process of being molded. In this case, the defect prediction unit 312 may calculate feature quantities of the solidified layer STL that has been molded or is in the process of being molded based on the measurement information 412. For example, the defect prediction unit 312 may calculate an index value indicating the surface condition of the solidified layer STL (e.g., the surface roughness mentioned above) as a feature quantity of the solidified layer STL. For example, the defect prediction unit 312 may calculate an index value indicating the size of the solidified layer STL (e.g., at least one of length, width, height (depth), thickness, surface area, and volume, as mentioned above) as a feature quantity of the solidified layer STL. For example, the defect prediction unit 312 may calculate the thickness of the nth solidified layer STL molded on the n-1th solidified layer STL by subtracting the surface shape (height) of the nth solidified layer STL from the surface shape (height) of the nth solidified layer STL. Note that n represents an integer greater than or equal to 2.

[0103] For example, the shape data D_3D, which is measurement information 412, contains information about the shape of the material layer ML. Therefore, the defect prediction unit 312 may calculate feature quantities of the material layer ML based on the measurement information 412. For example, the defect prediction unit 312 may calculate an index value indicating the surface condition of the material layer ML (e.g., the surface roughness described above) as a feature quantity of the material layer ML. For example, the defect prediction unit 312 may calculate an index value indicating the size of the material layer ML (e.g., at least one of length, width, height (depth), thickness, surface area, and volume, as described above) as a feature quantity of the material layer ML. For example, the defect prediction unit 312 may calculate the thickness of the material layer ML formed on the (n-1)th solidified layer STL in order to fabricate the nth solidified layer STL by subtracting the surface shape (height) of the (n-1)th solidified layer STL from the surface shape (height) of the material layer ML.

[0104] Furthermore, the feature quantities calculated based on the shape image data D_3D (i.e., feature quantities related to the shape of the object being measured) may be calculated by the shape measuring device 22 in addition to, or instead of, the defect prediction unit 312. In this case, the shape measuring device 22 may output measurement information 412 including the calculated feature quantities.

[0105] As described above, if at least one of the measurement information 411 and 412 is associated with the molding position, the defect prediction unit 312 may associate feature information 414, which indicates a feature calculated from at least one of the measurement information 411 and 412, with the molding position. The defect prediction unit 312 may store the feature information 414 associated with the molding position in the information DB 321 of the storage device 32.

[0106] Here, Figure 10 shows an example of measurement information 411, measurement information 412, molding information 413, and feature information 414 stored in the information DB 321 in association with position information 416 indicating the molding position. As shown in Figure 10, the information DB 321 may store multiple unit data 410, each containing position information 416 and process data 415 that includes measurement information 411, measurement information 412, molding information 413, and feature information 414 associated with the position information 416. In the example shown in Figure 10, the process data 415 includes all of the measurement information 411, measurement information 412, molding information 413, and feature information 414. However, the process data 415 does not have to include at least one of the measurement information 411, measurement information 412, molding information 413, and feature information 414.

[0107] Here, an example of process data 415 associated with position information 416 will be described. For example, temperature image data D_T as measurement information 411 may be associated with position information 416 indicating the center of the temperature image IMG_T shown by the temperature image data D_T (for example, the position where the aforementioned galvanometer mirror, which irradiates the material layer ML with building light EL and guides the light from the molten pool MP to the temperature measuring device 21, irradiates the building light EL). For example, shape image data D_3D as measurement information 412 may be associated with position information 416 indicating the center of the shape image IMG_3D shown by the shape image data D_3D. For example, feature information 414 may include information about the size of the molten pool MP calculated from the temperature image IMG_T whose center is the position indicated by the position information 416 associated with feature information 414. For example, feature information 414 may include information about the number of sputter SPTs calculated from the temperature image IMG_T whose center is the position indicated by the position information 416 associated with feature information 414. For example, feature information 414 may include information regarding the thickness of the solidified layer STL formed at the position indicated by the position information 416 associated with feature information 414. Because the positions in the Z-axis direction of multiple solidified layer STLs are different, the Z-axis position indicated by the position information 416 associated with feature information 414 may be used as information to distinguish multiple solidified layer STLs from each other. For example, feature information 414 may include information regarding the surface roughness of the solidified layer STL formed at the position indicated by the position information 416 associated with feature information 414 (i.e., the solidified layer STL at a position that can be uniquely identified by the Z-axis position indicated by the position information 416).

[0108] In the example shown in Figure 10, all of the measurement information 411, measurement information 412, molding information 413, and feature information 414 are associated with the position information 416 indicating the molding position. However, at least one of the measurement information 411, measurement information 412, molding information 413, and feature information 414 does not need to be associated with the position information 416.

[0109] In Figure 9 again, the defect prediction unit 312 then generates defect information 425 using process data 415 (i.e., measurement information 411, measurement information 412, molding information 413, and feature information 414) (step S14). In this embodiment, an example will be described in which the defect prediction unit 312 generates defect information 425 using process data 415 (i.e., measurement information 411, measurement information 412, and molding information 413 acquired by the information acquisition unit 311, and feature information generated from this information acquired by the information acquisition unit 311) and a prediction model PM.

[0110] Furthermore, the information acquisition unit 311 does not need to acquire at least one of the measurement information 411, measurement information 412, and molding information 413, and the defect prediction unit 312 does not need to calculate the feature quantities of the object to be measured (i.e., it does not need to generate feature quantity information 414, as described above). In this case, the defect prediction unit 312 may generate defect information 425 without using at least one of the measurement information 411, measurement information 412, molding information 413, and feature quantity information 414. In other words, the defect prediction unit 312 does not need to acquire at least one of the measurement information 411, measurement information 412, molding information The defect information 425 may be generated using at least one of the report 413 and feature information 414, while not using at least one of the measurement information 411, measurement information 412, molding information 413, and feature information 414. The defect prediction unit 312 may generate the defect information 425 using a portion of the process data 415, while not using other portions of the process data 415. The process data 415 does not necessarily have to include at least one of the measurement information 411, measurement information 412, molding information 413, and feature information 414.

[0111] The predictive model PM is a machine learning model generated to enable the generation of defect information 425 based on process data 415. Typically, the predictive model PM is a machine learning model that uses process data 415 as explanatory variables and defect information 425 as the target variable. Examples of such machine learning models include at least one of the following: regression models (e.g., at least one of linear regression models and nonlinear regression models), decision tree models, clustering models (in other words, classification models), and neural network models. The method for generating the predictive model PM using machine learning will be described in detail later with reference to Figure 18, etc.

[0112] As described above, process data 415 is associated with location information 416 indicating a molding location. In this case, the defect prediction unit 312 may use the process data 415 associated with location information 416 indicating a molding location to generate defect information 425 that includes information about a defect formed at that molding location. Conversely, the prediction model PM may be generated by machine learning so that it can generate defect information 425 that includes information about a defect formed at that molding location using the process data 415 associated with location information 416 indicating a molding location. The defect prediction unit 312 may use the process data 415 associated with location information 416 indicating a molding location to generate defect information 425 that includes information about a defect formed at at least one of the molding location and a location near that molding location. Conversely, the prediction model PM may be generated by machine learning so that it can generate defect information 425, which includes information about defects formed at at least one of the molding location and a neighboring location of the molding location, using process data 415 associated with location information 416 indicating a molding location. Note that a neighboring location of a molding location may mean a location whose distance from the molding location is less than or equal to a predetermined distance. The defect prediction unit 312 may generate defect information 425, which includes information about defects formed at the molding location, using process data 415 associated with location information 416 indicating at least one of the molding location and a neighboring location of the molding location. Conversely, the prediction model PM may be generated by machine learning so that it can generate defect information 425, which includes information about defects formed at the molding location, using process data 415 associated with location information 416 indicating at least one of the molding location and a neighboring location of the molding location. The defect prediction unit 312 may generate defect information 425, which includes information about a defect formed at at least one of the molding location and the vicinity of the molding location, using process data 415 associated with position information 416 indicating at least one of the molding location and the vicinity of the molding location.Conversely, the predictive model PM may be generated by machine learning so as to be able to generate defect information 425, which includes information about defects formed at at least one of the build location and the neighboring location, using process data 415 associated with location information 416 indicating at least one of the build location and the neighboring location of the build location.

[0113] When defect information 425 is generated that includes information about a defect formed at a certain location, the defect information 425 may be associated with information about the location where the defect is formed (for example, information about the fabrication location where the defect is formed). In other words, the defect prediction unit 312 may generate defect information 425 associated with information about the location where the defect is formed. Conversely, the prediction model PM may be generated by machine learning so that it can generate defect information 425 associated with information about the location where the defect is formed (for example, information about the fabrication location where the defect is formed).

[0114] The defect prediction unit 312 may generate defect information 425 that includes information regarding the presence or absence of defects. Specifically, the defect prediction unit 312 may generate defect information 425 that includes information indicating whether or not a defect is formed at a certain location on the molded object BO. Conversely, the prediction model PM may be generated by machine learning so that it can generate defect information 425 that includes information regarding the presence or absence of defects.

[0115] If the defect is the aforementioned void SP, the density of the fabricated object BO when the defect is present (i.e., when the void SP is formed) is likely to be different from the density of the fabricated object BO when the defect is not present (i.e., when the void SP is not formed). Typically, the density of the fabricated object BO when the defect is present is likely to be lower than the density of the fabricated object BO when the defect is not present. Therefore, the density of the fabricated object BO can be used as information that indirectly indicates the presence or absence of a defect. Accordingly, the defect prediction unit 312 may generate defect information 425 that includes information about the density of the fabricated object BO (density information). Conversely, the prediction model PM may be generated by machine learning so that it can generate defect information 425 that includes information about the density of the fabricated object BO. Furthermore, if defects occur in two printed objects BO, the density of one object BO with a larger volume of internal defects (i.e., the volume of voids SP) is likely to be lower than the density of the other object BO with a smaller volume of internal defects (i.e., the volume of voids SP). Furthermore, if defects occur in two printed objects BO, the density of one object BO with a larger total volume of multiple internal defects (i.e., the total volume of multiple voids SP) is likely to be lower than the density of the other object BO with a smaller total volume of multiple internal defects (i.e., the total volume of multiple voids SP).

[0116] The density of the printed object BO may include the density of each part of the printed object BO (for example, each part contained in a unit space centered on the location indicated by location information 426). In this case, the predictive model PM may be generated by machine learning so as to be able to generate information about the density of a part of the printed object BO contained in a unit space centered on one printing location, using process data 415 associated with location information 416 indicating one printing location. The predictive model PM may be generated by machine learning so as to be able to generate information about the density of a part of the printed object BO contained in a unit space centered on another printing location, using process data 415 associated with location information 416 indicating another printing location. Alternatively, the predictive model PM may be generated by machine learning so as to be able to generate information about the density of the printed object BO using process data 415 related to the printed object BO. Note that the density of the printed object BO may include the average density of the printed object BO. As described above, the fabricated object BO means at least one of the following: a completed three-dimensional structure ST, a three-dimensional structure ST in the process of being fabricated, a completed solidified layer STL, and a solidified layer STL in the process of being fabricated. Therefore, the density of the fabricated object BO may mean the density of at least one of the following: a completed three-dimensional structure ST, a three-dimensional structure ST in the process of being fabricated, a completed solidified layer STL, and a solidified layer STL in the process of being fabricated.

[0117] The defect prediction unit 312 may generate defect information 425, which includes information regarding the presence or absence of defects, based on information regarding the density of the molded object BO. For example, the defect prediction unit 312 may determine that there are defects in the molded object BO if its density falls below a predetermined density threshold. For example, the defect prediction unit 312 may determine that there are no defects in the molded object BO if its density does not fall below a predetermined density threshold.

[0118] However, even if the defect is not the aforementioned void SP, if the density of the fabricated object BO differs from the design density, there is a possibility that some kind of defect has occurred in the fabricated object BO. Therefore, even if the defect is not the aforementioned void SP, the defect prediction unit 312 may generate defect information 425 that includes information regarding the density of the fabricated object BO.

[0119] Furthermore, the defect prediction unit 312 may generate information regarding the density of the molded object BO separately from the defect information 425. This is because the density of the molded object BO itself may not be considered a defect of the molded object BO. In this case, the information regarding the density of the molded object BO (density information) may be included in the characteristic information as an example of characteristic information regarding the characteristics of the molded object BO, separately from the defect information 425.

[0120] The defect prediction unit 312 may generate defect information 425 that includes information about the characteristics of the defect (e.g., a void SP). Conversely, the prediction model PM may be generated by machine learning so that it can generate defect information 425 that includes information about the characteristics of the defect. The characteristics of the defect may include at least one of the following: the size of the defect, the location of the defect, and the shape of the defect. When the defect is a void SP, the shape of the defect is often spherical. In this case, at least one of sphericity and ellipticity may be used as an index value to represent the shape of the defect.

[0121] If the defect is the aforementioned void SP, the defect prediction unit 312 may consider void SP having a size exceeding the allowable size as a defect, while not considering void SP having a size not exceeding the allowable size as a defect. In this case, the defect prediction unit 312 may generate information regarding the presence or absence of defects based on information regarding the characteristics of the defects predicted using the prediction model PM. Regardless of the size of the void SP, the defect prediction unit 312 may consider void SP having characteristics that do not satisfy the allowable conditions as a defect, while not considering void SP having characteristics that satisfy the allowable conditions as a defect.

[0122] The defect prediction unit 312 may generate defect information 425 during the molding period when the molding apparatus 1 is molding an object BO (especially a three-dimensional structure ST). In this case, the defect prediction unit 312 may generate defect information 425 during the molding period using process data 415 that includes at least one of the measurement information 411 and 412 acquired during the molding period. The defect prediction unit 312 may generate defect information 425 during the molding period using process data 415 that includes molding information 413 acquired during the molding period. The defect prediction unit 312 may generate defect information 425 during the molding period using process data 415 that includes feature information 414 generated during the molding period.

[0123] The defect prediction unit 312 may generate defect information 425 after the molding period has ended. In this case, the defect prediction unit 312 may generate defect information 425 after the molding period has ended using process data 415 which includes at least one of the measurement information 411 and 412 acquired during or after the molding period. The defect prediction unit 312 may generate defect information 425 after the molding period has ended using process data 415 which includes molding information 413 acquired during or after the molding period. The defect prediction unit 312 may generate defect information 425 after the molding period has ended using process data 415 which includes feature information 414 generated during or after the molding period.

[0124] Subsequently, the defect output unit 313 outputs the defect information 425 generated by the defect prediction unit 312 in step S14 (step S15). However, the defect output unit 313 does not have to output the defect information 425.

[0125] The defect output unit 313 may output defect information 425 during the molding period when the molding apparatus 1 is molding an object BO (especially a three-dimensional structure ST). In this case, the defect output unit 313 may output defect information 425 generated during the molding period. Alternatively, the defect output unit 313 may output defect information 425 after the molding period has ended. In this case, the defect output unit 313 may output defect information 425 generated during or after the molding period has ended after the molding period has ended.

[0126] The defect output unit 313 may output defect information 425 to the user of the molding apparatus 1. In this case, the defect output unit 313 may output the defect information 425 using an output device 35. For example, the defect output unit 313 may output (i.e., display) the defect information 425 as an image using an output device 35 that can function as a display device. For example, the defect output unit 313 may output the defect information 425 as sound using an output device 35 that can function as an audio device (speaker). For example, the defect output unit 313 may output a paper page on which the defect information 425 is printed using an output device 35 that can function as a printing device. Alternatively, in addition to or instead of using the output device 35, the defect output unit 313 may output (i.e., transmit) the defect information 425 as data using a communication device 33. For example, the defect output unit 313 may use the communication device 33 to output (transmit) the defect information 425 as data to an information terminal owned by the user (for example, a personal computer, smartphone, or tablet device).

[0127] In either case, the user can obtain information about defects formed on the printed object BO. That is, the user can recognize whether or not defects are formed on the printed object BO. Furthermore, the user can recognize what kind of defects are formed on the printed object BO. In this case, the user may take action to reduce the impact caused by the defects. For example, the user may control a processing device (e.g., printing device 1 or another processing device different from printing device 1) to perform additional processing to reduce the impact caused by the defects (e.g., reduce or remove the defects). For example, the user may manually perform additional processing to remove the defects.

[0128] As described above, if defect information 425 is output during the molding period, if the molding device 1 continues to mold the object BO, there is a possibility that an object BO (especially a three-dimensional structure ST) with defects will be molded. Therefore, the user may control the molding device 1 to stop molding the object BO when defect information 425 is output. For example, the user may control the molding device 1 to stop molding the object BO uniformly when defect information 425 is output. Alternatively, the user may control the molding device 1 to stop molding the object BO if the defect information 425 satisfies predetermined stop conditions. For example, the user may control the molding device 1 to stop molding the object BO if the defect indicated by the defect information 425 adversely affects the proper molding of the object BO. On the other hand, the user may control the molding device 1 to continue molding the object BO if the defect information 425 does not satisfy predetermined stop conditions. For example, the user may control the molding apparatus 1 to continue molding the object BO if the defect indicated by the defect information 425 does not adversely affect the proper molding of the object BO.

[0129] If defects are formed in the printed object BO, the printing conditions set to define the operation of the printing device 1 that prints the object BO may not be appropriate. Therefore, the user may modify the printing conditions. Specifically, the user may modify the printing conditions based on the defect information 425 so that no defects are formed in the printed object BO. In this case, the user may modify the entire printing conditions. Alternatively, the user may modify the printing conditions partially. Note that modifying the printing conditions may be considered equivalent to changing the printing conditions.

[0130] When modifying the build conditions partially, for example, the user may modify some of the build conditions while leaving other parts of the build conditions unchanged. As an example, the user may modify at least one of the following: irradiation conditions, recoater movement conditions, stage movement conditions, head movement conditions, material conditions, environmental conditions, and design conditions, while leaving at least one of the following unmodified conditions. For example, a user may modify at least one of the multiple conditions included in the irradiation conditions (e.g., the output of the light source that generates the EL, the intensity of the EL emitted from the light source, the intensity of the EL on the material layer ML, the intensity distribution of the EL on the material layer ML, the irradiation timing of the EL, the irradiation position of the EL, the movement rules for the irradiation position of the EL, the movement trajectory of the irradiation position of the EL, the hatching distance which is the distance between two adjacent movement trajectories, the movement speed of the irradiation position of the EL (i.e., scanning speed), the spot diameter of the EL emitted from the light source, and the diameter of the spot formed by the EL on the material layer ML) while not modifying at least one of the multiple conditions included in the irradiation conditions. A user may modify at least one of the multiple conditions included in the recoater movement conditions while not modifying at least one of the multiple conditions included in the recoater movement conditions. A user may modify at least one of the multiple conditions included in the head movement conditions while not modifying at least one of the multiple conditions included in the head movement conditions. The user may modify at least one of the multiple conditions included in the stage transition conditions, while not modifying at least one of the multiple conditions included in the stage transition conditions. The user may modify at least one of the multiple conditions included in the material conditions, while not modifying at least one of the multiple conditions included in the material conditions. The user may modify at least one of the multiple conditions included in the environment conditions, while not modifying at least one of the multiple conditions included in the environment conditions.A user may modify at least one of the multiple conditions included in the design conditions, while not having to modify at least one of the other multiple conditions included in the design conditions.

[0131] When modifying the molding conditions partially, for example, the user may modify the molding conditions used to print one part of the printed object BO, while not modifying the molding conditions used to print other parts of the printed object BO. For example, if defect information 425 indicates that a defect has formed in one part of the printed object BO, the user may modify the molding conditions used to print one part of the printed object BO, while not modifying the molding conditions used to print other parts of the printed object BO that are different from that part. Alternatively, the user may modify the molding conditions used to print one part of the printed object BO and its vicinity, while not modifying the molding conditions used to print other parts of the printed object BO that are different from one part and its vicinity. Note that the vicinity of one part may mean a part whose distance from one part is less than or equal to a predetermined distance. As a result, the time required to modify the molding conditions can be reduced.

[0132] If the molding conditions are modified, the molding device 1 may use the modified molding conditions to fabricate the object BO. For example, if defect information 425 is output after the molding period has ended (i.e., after the fabrication of a certain three-dimensional structure ST is completed), the user may modify the molding conditions used when fabricating the next three-dimensional structure ST having the same shape, etc. In this case, the molding device 1 may use the modified molding conditions when fabricating the next three-dimensional structure ST having the same shape, etc. As another example, if defect information 425 is output during the molding period (i.e., before the fabrication of a certain three-dimensional structure ST is completed), the user may modify the molding conditions used to complete the three-dimensional structure ST that is in the process of being fabricated. In this case, the molding device 1 may use the modified molding conditions to resume fabricating the three-dimensional structure ST that is in the process of being fabricated. Furthermore, if defect information 425 indicating that a defect has formed in a part of the printed object BO is output during the printing period, the user may repair the defect in that part of the printed object BO. For example, the user may control the printing apparatus 1 to re-irradiate a part of the printed object BO with the printing light EL to remelt it. At this time, the user may control the printing apparatus 1 to re-irradiate a part of the printed object BO with the printing light EL using the modified printing conditions.

[0133] The defect output unit 313 may output defect information 425 to an external device of the information processing device 3. In this case, the defect output unit 313 may use the communication device 33 to output (i.e., transmit) the defect information 425 to the external device. Specifically, the defect output unit 313 may transmit the defect information 425 to the molding device 1. As a result, the molding device 1 may receive (acquire) the defect information 425.

[0134] In this case, the molding device 1 may output the defect information 425 using an output device (not shown) provided by the molding device 1. For example, the molding device 1 may output the defect information 425 as an image using a display device provided by the molding device 1. For example, the molding device 1 may output the defect information 425 as audio using an audio device provided by the molding device 1. For example, the molding device 1 may output a printed page with the defect information 425 using a printing device provided by the molding device 1. The molding device 1 may output (i.e., transmit) the defect information 425 as data using a communication device (not shown) provided by the molding device 1. For example, the molding device 1 may output (transmit) the defect information 425 as data to an information terminal (e.g., a personal computer, smartphone, or tablet) owned by the user.

[0135] The control unit 16 of the molding apparatus 1 may control the molding unit 10 based on the received defect information 425. For example, the control unit 16 may control the molding unit 10 (or other processing apparatus) based on the defect information 425 to perform additional processing to reduce the impact caused by the defect (for example, to reduce or remove the defect). In this case, it becomes possible to reduce the impact caused by the defect without requiring any action from the user.

[0136] Furthermore, the control unit 16 may control the molding unit 10 to stop the molding of the molded object BO if defect information 425 is output. For example, the control unit 16 may control the molding unit 10 to uniformly stop the molding of the molded object BO if defect information 425 is output. For example, the control unit 16 may control the molding unit 10 to stop the molding of the molded object BO if the defect information 425 satisfies predetermined stop conditions. For example, the control unit 16 may control the molding unit 10 to stop the molding of the molded object BO if the defect indicated by the defect information 425 adversely affects the proper molding of the molded object BO. On the other hand, the control unit 16 may control the molding unit 10 to continue molding the molded object BO if the defect information 425 does not satisfy predetermined stop conditions. For example, the control unit 16 may control the printing unit 10 to continue printing the object BO if the defect indicated by the defect information 425 does not adversely affect the proper printing of the object BO. In this case, it is possible to minimize the impact caused by the defect without requiring any intervention from the user.

[0137] Furthermore, the control unit 16 may modify the molding conditions based on the defect information 425. Specifically, the control unit 16 may modify the molding conditions based on the defect information 425 so that no defects are formed in the molded object BO. In this case, the control unit 16 may modify the entire molding conditions. Alternatively, the control unit 16 may partially modify the molding conditions, similar to how a user might partially modify the molding conditions.

[0138] When modifying the molding conditions in part, for example, the control unit 16 may modify some of the molding conditions while leaving other parts of the molding conditions untouched. As an example, the control unit 16 may modify at least one of the irradiation conditions, recoater movement conditions, stage movement conditions, head movement conditions, material conditions, environmental conditions, and design conditions while leaving at least one of the other of the irradiation conditions, recoater movement conditions, stage movement conditions, head movement conditions, material conditions, environmental conditions, and design conditions untouched. As an example, the control unit 16 may modify at least one of the multiple conditions included in the irradiation conditions (for example, the output of the light source that generates the EL, the intensity of the EL emitted from the light source, the intensity of the EL on the material layer ML, the intensity distribution of the EL on the material layer ML, the irradiation timing of the EL, the irradiation position of the EL, the movement rule of the irradiation position of the EL, the movement trajectory of the irradiation position of the EL, the hatching distance which is the interval between two adjacent movement trajectories, the movement speed of the irradiation position of the EL (i.e., scanning speed), the spot diameter of the EL emitted from the light source, and the diameter of the spot formed by the EL on the material layer ML), while not modifying at least one of the multiple conditions included in the irradiation conditions. The control unit 16 may modify at least one of the multiple conditions included in the recoater movement conditions, while not modifying at least one of the multiple conditions included in the recoater movement conditions. The control unit 16 may modify at least one of the multiple conditions included in the head movement conditions, while not modifying at least one of the multiple conditions included in the head movement conditions. The control unit 16 may modify at least one of the multiple conditions included in the stage movement conditions, while not modifying at least one of the multiple conditions included in the stage movement conditions. The control unit 16 may modify at least one of the multiple conditions included in the material conditions, while not modifying at least one of the multiple conditions included in the material conditions. The control unit 16 may modify at least one of the multiple conditions included in the environmental conditions, while not modifying at least one of the multiple conditions included in the environmental conditions.The control unit 16 may modify at least one of the multiple conditions included in the design conditions, while not modifying at least one of the multiple conditions included in the design conditions.

[0139] When modifying the molding conditions partially, for example, the control unit 16 may modify the molding conditions used to fabricate one part of the fabricated object BO, while not modifying the molding conditions used to fabricate other parts of the fabricated object BO. For example, if defect information 425 indicates that a defect has formed in one part of the fabricated object BO, the control unit 16 may modify the molding conditions used to fabricate one part of the fabricated object BO, while not modifying the molding conditions used to fabricate other parts of the fabricated object BO that are different from that part. Alternatively, the control unit 16 may modify the molding conditions used to fabricate one part of the fabricated object BO and its vicinity, while not modifying the molding conditions used to fabricate other parts of the fabricated object BO that are different from that part and its vicinity. As a result, the time required to modify the molding conditions can be reduced.

[0140] If the molding conditions are modified, the control unit 16 may control the molding unit 10 to mold the molded object BO using the modified molding conditions. For example, if defect information 425 is input from the information processing device 3 to the molding device 1 (particularly the control unit 16) after the molding period has ended (i.e., after the molding of a certain three-dimensional structure ST is completed), the control unit 16 may modify the molding conditions used when molding the next three-dimensional structure ST having the same shape, etc. In this case, the control unit 16 may control the molding unit 10 using the modified molding conditions when molding the next three-dimensional structure ST having the same shape, etc. As another example, if defect information 425 is input from the information processing device 3 to the molding device 1 (particularly the control unit 16) during the molding period (i.e., before the molding of a certain three-dimensional structure ST is completed), the control unit 16 may modify the molding conditions used to complete the three-dimensional structure ST that is in the process of being molded. In this case, the control unit 16 may control the printing unit 10 to resume printing the three-dimensional structure ST that is in the process of being printed using the modified printing conditions. Furthermore, if defect information 425 indicating that a defect has formed in a part of the printed object BO is input during the printing period, the printing unit 10 may repair the defect in the part of the printed object BO. For example, the control unit 16 may control the printing unit 10 to re-irradiate a part of the printed object BO with the printing light EL to re-melt it. At this time, the control unit 16 may control the printing unit 10 to re-irradiate a part of the printed object BO with the printing light EL using the modified printing conditions.

[0141] When outputting defect information 425, the defect output unit 313 may output (for example, display or transmit) defect information 425 associated with location information 426 indicating the location where the defect is formed, as shown in Figure 11. When defect information 425 is displayed, Figure 11 shows an example of a display format in which defect information 425 and location information 426 are associated. On the other hand, when defect information 425 is transmitted as data, Figure 11 shows an example of a data structure in which defect information 425 and location information 426 are associated. In this case, the user or control unit 16 can easily identify where the defect is formed on the molded object BO.

[0142] Here, an example of defect information 425 associated with location information 426 will be described. For example, defect information 425 may include information about the density of defects in an arbitrary unit space centered at the location indicated by the location information 426 associated with defect information 425, as information about the density of defects. In other words, defect information 425 that includes information about the density of defects may be associated with location information 426 indicating the position of the center of this unit space. Note that the density of defects in an arbitrary unit space may include the average density of defects in the arbitrary unit space. For example, defect information 425 may include information about the characteristics of defects formed in an arbitrary unit space centered at the location indicated by the location information 426 associated with defect information 425, as information about the characteristics of defects. In other words, defect information 425 that includes information about the characteristics of defects may be associated with location information 426 indicating the position of the center of this unit space. Alternatively, for example, defect information 425 may include information about the characteristics of a defect (e.g., a void SP) whose centroid is at the location indicated by the location information 426 associated with defect information 425, as information about the characteristics of defects. In other words, defect information 425, which includes information about the characteristics of the defect, may be associated with position information 426 that indicates the position of the centroid of the defect.

[0143] When outputting defect information 425, as shown in Figure 12, the defect output unit 313 may output (for example, display or transmit) the process data 415 described above along with the defect information 425. The defect output unit 313 may output (for example, display or transmit) the defect information 425 and the process data 415 in an output format in which the defect information 425 and the process data 415 are associated. In other words, the defect output unit 313 may output (for example, display or transmit) at least one of the measurement information 411, measurement information 412, molding information 413, and feature information 414 that constitute the process data 415 along with the defect information 425. The defect output unit 313 may output (for example, display or transmit) the defect information 425 and at least one of the measurement information 411, measurement information 412, molding information 413, and feature information 414 that constitute the process data 415 in an output format in which the defect information 425 is associated with at least one of the measurement information 411, measurement information 412, molding information 413, and feature information 414 that constitute the process data 415. In the example shown in Figure 12, the defect information 425 and the process data 415 are associated via position information 416(426). However, the defect information 425 and the process data 415 may be associated in a way other than the method using position information 416(426). When the defect information 425 is displayed, Figure 12 shows an example of a display format in which the process data 415 is displayed together with the defect information 425. On the other hand, when the defect information 425 is transmitted as data, Figure 12 shows an example of a data structure in which the process data 415 is transmitted together with the defect information 425.

[0144] In this case, the defect output unit 313 may output defect information 425 and process data 415 in an output format in which defect information 425 (referred to as "defect information 425a" for convenience of explanation) relating to a defect formed on a part of the molded object BO and process data 415 (referred to as "process data 415a" for convenience of explanation) relating to the molding of a part of the molded object BO are associated. In the example shown in Figure 12, the defect information 425a and process data 415a are associated via position information 416 and 426 (referred to as "position information 416a and 426a" for convenience of explanation) indicating the location of a part of the molded object BO. In this case, the user or control unit 16 can easily understand the cause of the defect formed on a part of the molded object BO by referring to the process data 415 relating to the molding of a part of the molded object BO.

[0145] As mentioned above, if the defect is the aforementioned void SP, one of the causes of the defect is that the amount of energy transferred from the build light EL to the material layer ML is insufficient or excessive compared to the target amount. Therefore, if the amount of energy transferred from the build light EL to the material layer ML is output along with the defect information 425, the user or control unit 16 can more appropriately understand the cause of the defect indicated by the defect information 425.

[0146] Therefore, the defect prediction unit 312 may calculate an estimated value regarding the amount of energy transferred from the printing light EL to the material layer ML based on the process data 415. Specifically, the defect prediction unit 312 may calculate an estimated value regarding the amount of energy transferred from the printing light EL to the material layer ML during the printing of a part of the printed object BO based on the process data 415 regarding the printing of a part of the printed object BO. For example, the defect prediction unit 312 may calculate an estimated value regarding the amount of energy based on at least one of the measurement information 411 and 412 from the process data 415.

[0147] As an example, the defect prediction unit 312 may use the width of the molten pool MP, which is an example of a feature quantity calculated from the measurement information 411, and the thickness of the solidified layer STL, which is an example of a feature quantity calculated from the measurement information 412, to calculate an estimated value of energy density (i.e., energy per unit volume), which is an example of an estimated value related to energy quantity. Specifically, if the width of the molten pool MP is Δy, the thickness of the solidified layer STL is Δz, the output of the light source that generates the build light EL is P, the movement speed of the irradiation position of the build light EL on the material layer ML (i.e., scanning speed) is v, and the estimated value of energy density is ED, the defect prediction unit 312 may use the formula ED = P / (v × Δy × Δz) to calculate the estimated value of energy density. Note that the output of the light source (P) and the scanning speed of the build light EL (v) may be the values ​​from the build information 413.

[0148] As another example, the defect prediction unit 312 may use the width of the molten pool MP, which is an example of a feature quantity calculated from the measurement information 411, to calculate an estimated value of power density (i.e., energy per unit area), which is an example of an estimated value related to energy quantity. Specifically, the defect prediction unit 312 calculates the diameter of the spot formed by the fusion light EL on the material layer ML based on the width of the molten pool MP. Subsequently, the defect prediction unit 312 may calculate the power density based on the calculated spot diameter. Specifically, if the spot diameter is d, the output of the light source that generates the fusion light EL is P, the movement speed of the irradiation position of the fusion light EL on the material layer ML (i.e., scanning speed) is v, and the estimated power density is PD, the estimated power density may be calculated using the formula PD = P / (d × v). Note that the output of the light source (P) and the scanning speed of the fusion light EL (v) may be the values ​​from the fusion information 413. Furthermore, the spot diameter (d) of the EL light used in the build process may be the value from build information 413.

[0149] As another example, the defect prediction unit 312 may use the size of the molten pool MP, which is an example of a feature quantity calculated from the measurement information 411, to calculate an estimated value of an index value called the penetration geometry, which is an example of an estimated value related to the amount of energy. An example of the penetration geometry is shown in Figure 13. As shown in Figure 13, the penetration geometry is defined by the depth dp and do of the molten material in the molten pool MP and the thickness tL of the material layer ML. The depth dp corresponds to the distance from the surface of the material layer ML to the bottom of the molten material in the molten pool MP (i.e., the depth of the molten pool MP). The depth do corresponds to the distance from the surface of the material layer ML to the position where two adjacent molten pool MPs (specifically, two molten pool MPs formed on two adjacent scanning lines) intersect. Both the depth dp and do can be calculated (estimated) from the size of the molten pool MP. In this case, a first penetration geometry calculated from the formula dp / tL and a second penetration geometry calculated from the formula do / tL may be used as the penetration geometry. In this case, if at least one of the first and second penetration geometry is less than 1, it is estimated that the melting of the molding material is insufficient (i.e., the amount of energy is insufficient). If at least one of the first and second penetration geometry is 1 or greater and less than the allowable upper limit, it is estimated that the melting of the molding material is appropriate (i.e., the amount of energy is appropriate). If at least one of the first and second penetration geometry is greater than or equal to the allowable upper limit, it is estimated that the melting of the molding material is excessive (i.e., the amount of energy is excessive).

[0150] As described above, if at least a portion of the process data 415 is associated with the fabrication location (i.e., location information 416), the defect prediction unit 312 may associate the energy information 435 relating to the estimated energy amount calculated by the defect prediction unit 312 with the fabrication location. The defect prediction unit 312 may generate energy information 435 associated with the fabrication location. For example, the defect prediction unit 312 may associate energy information 435 containing information regarding an estimated energy amount transferred from the printing light EL to the material layer ML for one part of the fabricated object BO with the fabrication location where the fabrication device 1 was performing the fabrication process (i.e., one part of the fabricated object BO). The defect prediction unit 312 may also associate energy information 435 containing information regarding an estimated energy amount transferred from the printing light EL to the material layer ML for another part of the fabricated object BO with the fabrication location where the fabrication device 1 was performing the fabrication process (i.e., another part of the fabricated object BO).

[0151] Subsequently, the defect output unit 313 may output (e.g., display or transmit) energy information 435 related to an estimated value of the energy amount calculated by the defect prediction unit 312, along with the defect information 425, as shown in Figure 14. In other words, the defect output unit 313 may output (e.g., display or transmit) the defect information 425 and the energy information 435 in an output format in which the defect information 425 and the energy information 435 are associated. In the example shown in Figure 14, the defect information 425 and the energy information 435 are associated via position information 416 (426). In this case, energy information 435 containing information about an estimated value of the energy amount transmitted from the printing light EL to the material layer ML may be associated with one part of the fabricated object BO, and defect information 425 containing information about a defect that occurred in one part of the fabricated object BO. Energy information 435 containing information about an estimated value of the energy amount transmitted from the printing light EL to the material layer ML may be associated with another part of the fabricated object BO, and defect information 425 containing information about a defect that occurred in another part of the fabricated object BO. However, the defect information 425 and the energy information 435 may be associated in a way other than the method using the position information 416 (426). When the defect information 425 is displayed, Figure 14 shows an example of a display format in which the energy information 435 is displayed together with the defect information 425. On the other hand, when the defect information 425 is transmitted as data, Figure 14 shows an example of a data structure in which the energy information 435 is output together with the defect information 425. In addition, although process data 415 is output along with the defect information 425 and the energy information 435 in Figure 14, the process data 415 is not necessarily output.

[0152] As a result, the user or control unit 16 can more appropriately determine the cause of the defect indicated by the defect information 425. Specifically, the user or control unit 16 can more appropriately determine whether the cause of the defect indicated by the defect information 425 is that the amount of energy transferred from the EL printer to the material layer ML is insufficient compared to the target amount, or that the amount of energy transferred from the EL printer to the material layer ML is excessive compared to the target amount. As a result, the user or control unit 16 can more appropriately modify the above-mentioned printing conditions (in particular, the irradiation conditions related to the EL printer). For example, if the amount of energy transferred from the EL printer to the material layer ML is insufficient compared to the target amount, the user or control unit 16 may modify the printing conditions so that the amount of energy increases. As an example, the user or control unit 16 may modify the printing conditions to increase the amount of energy transferred from the printing light EL to the material layer ML by increasing the output of the light source that generates the printing light EL, reducing the path through which the energy transferred from the printing light EL to the material layer ML diffuses by reducing the thickness of the support members located below the material layer ML, and / or slowing down the movement speed (i.e., scanning speed) of the irradiation position of the printing light EL on the material layer ML. On the other hand, if, for example, the amount of energy transferred from the printing light EL to the material layer ML is excessive compared to the target amount, the user or control unit 16 may modify the printing conditions to decrease the amount of energy. As an example, the user or control unit 16 may modify the printing conditions to reduce the amount of energy transferred from the printing light EL to the material layer ML by reducing the output of the light source that generates the printing light EL, increasing the thickness of the support members located below the material layer ML to increase the paths through which the energy transferred from the printing light EL to the material layer ML diffuses, and / or increasing the speed at which the irradiation position of the printing light L moves (i.e., the scanning speed) on the material layer ML.

[0153] The defect output unit 313 may output defect information 425 and energy information 435 in an output format in which defect information 425 (referred to as "defect information 425a" for convenience of explanation) relating to a defect formed on a part of the fabricated object BO is associated with energy information 435 (referred to as "energy information 435a" for convenience of explanation) relating to the amount of energy transmitted from the fabrication light EL to the material layer ML during the fabrication of a part of the fabricated object BO. In the example shown in Figure 14, the defect information 425a and energy information 435a are associated via position information 416 and 426 (referred to as "position information 416a and 426a" for convenience of explanation) indicating the location of a part of the fabricated object BO. In this case, the user or control unit 16 can more easily grasp the cause of a defect formed at a certain location.

[0154] In addition to outputting the defect information 425 as is, the defect output unit 313 may also output (for example, display or transmit) defect statistics information 445 that shows the statistical values ​​of the defects indicated by the defect information 425. An example of defect statistics information 445 is shown in Figure 15. As shown in Figure 15, the defect statistics information 445 may include at least one of the following: the total number of defects, the maximum volume of defects (i.e., the volume of the largest defect), the maximum length of defects (i.e., the length of the longest defect), the average volume of defects, and the average length of defects.

[0155] In the examples shown in Figures 11 to 13, the defect output unit 313 outputs (specifically displays) the defect information 425 in table format. However, if the defect information 425 is to be displayed to the user, the defect output unit 313 may display the defect information 425 in a display format that makes it easier for the user to intuitively understand the information about the defect indicated in the defect information 425.

[0156] For example, as shown in Figure 16, the defect output unit 313 may display defect information 425 along with the three-dimensional model of the fabricated object BO (in the example shown in Figure 16, the three-dimensional structure ST). In this case, the defect output unit 313 may superimpose a mark MI indicating the formation of a defect on the part of the three-dimensional model of the fabricated object BO where the defect is formed. In this case, if the user selects a mark MI using the input device 34, detailed information about the defect indicated by the mark MI may be displayed (for example, displayed as a pop-up). The detailed information about the defect may include information on at least one of the defect's characteristics and density. For example, the detailed information about the defect may include a numerical value for at least one of the defect's characteristics and density. Furthermore, the defect output unit 313 may change the display format of the mark MI indicating the formation of a defect depending on the cause of the defect. For example, the defect output unit 313 may change the display format of the mark MI indicating a defect formed due to an excess of energy and the display format of the mark MI indicating a defect formed due to an insufficient amount of energy. For example, the display format may be at least one of the following: the color of Mark MI, the brightness of Mark MI, the size of Mark MI, and the shape of Mark MI.

[0157] The defect output unit 313 may change the display format of the Mark MI based on at least a portion of the defect information 425 and process data 415. For example, the defect output unit 313 may change the display format of the Mark MI indicating a defect predicted to occur at a certain location based on the temperature of the molten pool MP formed at a certain location calculated from the measurement information 411. For example, the defect output unit 313 may change the display format of the Mark MI indicating a defect predicted to occur at a certain location based on the difference between the temperature of the molten pool MP formed at a certain location calculated from the measurement information 411 and the reference temperature of the molten pool MP. For example, the defect output unit 313 may change the display format of the Mark MI indicating a defect predicted to occur at a certain location based on the difference between the thickness of the solidified layer STL formed at a certain location calculated from the measurement information 412 and the design thickness of the solidified layer STL. If the user selects a Mark MI using the input device 34, detailed information of the process data 415 indicated by the Mark MI may be displayed (for example, in a pop-up). Furthermore, the detailed information of the process data 415 may include at least one numerical value from the measurement information 411, measurement information 412, molding information 413, and feature information 414 within the process data 415.

[0158] The defect output unit 313 may superimpose a mark indicating that no defects have been formed on the portion of the three-dimensional model of the molded object BO where no defects have been formed. In this case, the defect output unit 313 may use different display formats for the mark MI indicating that a defect has been formed and the mark indicating that no defects have been formed. The defect output unit 313 may change the display format of the mark indicating that no defects have been formed based on at least a portion of the defect information 425 and process data 415.

[0159] The defect output unit 313 may display energy information 435 in addition to or instead of defect information 425. For example, the defect output unit 313 may display an energy information 435 as a display object that shows an estimated value regarding the amount of energy transmitted to a part of the molded object BO. In this case as well, the defect output unit 313 may change the display format of the display object corresponding to the energy information 435 based on at least a part of the defect information 425 and process data 415. For example, the defect output unit 313 may change the display format of the display object corresponding to the energy information 435 based on an estimated value regarding the amount of energy. Furthermore, the defect output unit 313 may change the display format of a mark MI indicating a defect predicted to occur in a part of the molded object BO based on the energy information 435 of that part. Furthermore, if the user selects a mark MI using the input device 34, detailed information of the energy information 435 indicated by the mark MI may be displayed (for example, displayed in a pop-up). Furthermore, the detailed information in energy information 435 may include at least one numerical value for energy density, power density, and fusion geometry. In this case, the user can more easily intuitively understand the amount of energy transferred to each part of the printed object BO.

[0160] The defect output unit 313 may display process data 415 in addition to or instead of defect information 425. For example, the defect output unit 313 may display a display object showing process data 415 related to the molding of a part of the molded object BO. In this case as well, the defect output unit 313 may change the display format of the display object showing process data 415 based on at least a portion of the defect information 425 and process data 415. In this case, the user will be able to intuitively understand the process data 415 related to the molding of each part of the molded object BO.

[0161] The defect output unit 313 may display density information relating to the density of the molded object BO in addition to or instead of the defect information 425. For example, the defect output unit 313 may display a display object indicating the density of a part of the molded object BO as density information. In this case as well, the defect output unit 313 may change the display format of the display object indicating density based on at least a part of the defect information 425 and process data 415. For example, the defect output unit 313 may change the display format of the display object corresponding to the energy information 435 based on the density. Furthermore, the defect output unit 313 may change the display format of the mark MI indicating a defect predicted to occur in a part of the molded object BO based on the density information of that part. Furthermore, the part of the molded object BO in which a defect is predicted to occur may be a part where the density (i.e., the density estimated as described above) is lower than a threshold. In other words, the mark MI may be a display object indicating a part where the density is lower than a threshold as a defect. Furthermore, if the user selects a mark MI using the input device 34, detailed information about the density indicated by that mark MI (e.g., a numerical value of the density) may be displayed (e.g., displayed as a pop-up). In this case, the user will be able to intuitively understand the density of each part of the molded object BO.

[0162] The defect output unit 313 may simultaneously display at least two of the defect information 425, process data 415, energy information 435, and density information. For example, the defect output unit 313 may simultaneously display energy information 435 and defect information 415. In this case, the defect output unit 313 may change the display format of energy information 435 based on an estimated value of energy quantity using a first modification method, and change the display format of defect information 415 based on defect characteristics using a second modification method. The first modification method may be, for example, a method of changing a first characteristic (e.g., size) of a display object such as a mark MI, and the second modification method may be, for example, a method of changing a second characteristic (e.g., color) of a display object such as a mark MI. Alternatively, for example, the defect output unit 313 may simultaneously display density information and process data 415. In this case, the defect output unit 313 may change the display format of density information based on density using a first modification method, and change the display format of process data 415 based on process data 415 using a second modification method. Alternatively, for example, the defect output unit 313 may display density information and defect information 425 simultaneously. In this case, the defect output unit 313 may change the display format of the density information based on the density using the first modification method, and change the display format of the defect information 425 based on the defect characteristics using the second modification method.

[0163] For example, as shown in Figure 17, the defect output unit 313 may display defect information 425 in association with the molding time. Specifically, as described above, the molding apparatus 1 sequentially molds each part of the molded object BO (typically a three-dimensional structure ST). Therefore, the position of each part of the molded object BO (i.e., the molding position) can be converted into molding time. Thus, the defect output unit 313 may display defect information 425 in association with the molding time to indicate whether or not a defect was formed at any of the molding times in the series of molding periods for molding the three-dimensional structure ST. In other words, the defect output unit 313 may display defect information 425 in association with the molding time to indicate at which of the molding times in the series of molding periods for molding the three-dimensional structure ST the molding process that caused the formation of the defect occurred. Figure 17 shows an example of displaying defect information 425 indicating that a defect was formed at molding times t1, t2, and t3 in the series of molding periods. In addition, the defect output unit 313 may display energy information 435 in association with the molding time. Figure 17 shows an example in which energy information 435 is displayed using a graph showing the change in energy over time. In this case, the user or control unit 16 can more appropriately determine whether the cause of defect formation is insufficient energy compared to the target amount, or in excess energy compared to the target amount. The molding apparatus 1 fabricates the object BO by sequentially stacking solidification layer STLs. Therefore, the position of each solidification layer STL in the object BO in the stacking direction (for example, the Z-axis direction) can be converted into molding time. For example, the defect output unit 313 may display defect information 425 in association with each solidification layer STL (for example, each solidification layer STL from the 1st layer to the Nth layer).

[0164] Figure 17 shows an example in which the defect output unit 313 displays energy information 435 using the vertical axis in a two-dimensional plane defined by two axes. However, the defect output unit 313 may also display at least one of the defect information 425, process data 415, and density information using the vertical axis. Alternatively, the defect output unit 313 may display at least two of the energy information 435, defect information 425, process data 415, and density information in a three-dimensional space defined by three axes.

[0165] (3) Technical effects As described above, in this embodiment, the information processing device 3 generates defect information 425 regarding defects formed in the molded object BO using at least one of the measurement information 411 and 412 obtained by the measurement device 2 actually measuring the object to be measured during the molding of the molded object BO. In other words, the information processing device 3 predicts defects formed in the molded object BO using at least one of the measurement information 411 and 412. Therefore, compared to the information processing device of the first comparative example, which predicts defects formed in the molded object BO without using the measurement information 411 and 412 obtained by actually measuring the object to be measured, the information processing device 3 can predict defects formed in the molded object BO with higher accuracy.

[0166] Furthermore, the information processing device 3 predicts defects to be formed on the fabricated object BO using a predictive model PM that has been generated in advance by machine learning. Therefore, compared to the information processing device of the second comparative example, which predicts defects to be formed on the fabricated object BO without using the predictive model PM, the information processing device 3 can predict defects to be formed on the fabricated object BO with higher accuracy.

[0167] (4) Generation of the predictive model PM Next, we will explain how to generate the predictive model PM.

[0168] (4-1) Configuration of Model Generation Device 5 First, the configuration of the model generation device 5 that generates the predictive model PM will be explained with reference to Figure 18. Figure 18 is a block diagram showing the configuration of the model generation device 5.

[0169] As shown in Figure 18, the model generation device 5 comprises an arithmetic unit 51, a storage device 52, and a communication device 53. Furthermore, the model generation device 5 may also include an input device 54 and an output device 55. However, the model generation device 5 does not need to include at least one of the input device 54 and the output device 55. The arithmetic unit 51, the storage device 52, the communication device 53, the input device 54, and the output device 55 may be connected via a data bus 56.

[0170] The arithmetic unit 51 includes, for example, at least one of a CPU, GPU, and FPGA. The arithmetic unit 51 reads a computer program. For example, the arithmetic unit 51 may read a computer program stored in the storage device 52. For example, the arithmetic unit 51 may read a computer program stored in a computer-readable and non-temporary recording medium using a recording medium reader (not shown) provided by the model generation device 5. The arithmetic unit 51 may obtain (i.e., download or read) a computer program from a device (not shown) located outside the model generation device 5 via a communication device 53 (or other communication device). The arithmetic unit 51 executes the read computer program. As a result, logical functional blocks for executing the processing that the model generation device 5 should perform (e.g., model generation processing for generating a predictive model PM) are realized within the arithmetic unit 51. In other words, the arithmetic unit 51 can function as a controller for realizing logical functional blocks for executing the processing that the model generation device 5 should perform.

[0171] Figure 18 shows an example of a logical functional block implemented within the arithmetic unit 51 to execute the model generation process. As shown in Figure 18, the arithmetic unit 51 implements an information acquisition unit 511 and a model generation unit 512. The processes performed by the information acquisition unit 511 and the model generation unit 512 will be described in detail later with reference to Figure 19, etc., but an overview is given below. The information acquisition unit 511 acquires multiple (typically large) training data 610 necessary for machine learning to generate a predictive model PM. The multiple training data 610 acquired by the information acquisition unit 511 are stored in the storage device 52 as a training dataset 600. The model generation unit 512 uses at least a portion of the training dataset 600 to perform machine learning to generate the predictive model PM. The model generation process can also be called the model generation operation.

[0172] The storage device 52 is capable of storing desired data. For example, the storage device 52 may temporarily store a computer program executed by the arithmetic unit 51. The storage device 52 may temporarily store data that the arithmetic unit 51 uses temporarily when it is executing a computer program. The storage device 52 may store data that the model generation device 5 stores long-term. The storage device 52 may include at least one of the following: RAM (Random Access Memory), ROM (Read Only Memory), hard disk drive, magneto-optical disk drive, SSD (Solid State Drive), and disk array device. In other words, the storage device 52 may include a non-temporary recording medium.

[0173] The communication device 53 can communicate with devices outside the model generation device 5 via a communication network (not shown).

[0174] The input device 54 is a device that receives information input to the model generation device 5 from outside the model generation device 5. For example, the input device 54 may include an operating device that can be operated by the user of the model generation device 5 (for example, at least one of a keyboard, mouse, and touch panel). For example, the input device 54 may include a recording medium reader that can read information recorded as data on a recording medium that can be attached externally to the model generation device 5.

[0175] Furthermore, the model generation device 5 can receive data from external devices via the communication device 53. In this case, the communication device 53 may function as an input device that receives information from external sources to the model generation device 5.

[0176] The output device 55 is a device that outputs information to the outside of the model generation device 5. For example, the output device 55 may output information as an image. That is, the output device 55 may include a display device (so-called display) capable of displaying images. For example, the output device 55 may output information as sound. That is, the output device 55 may include an audio device (so-called speaker) capable of outputting sound. For example, the output device 55 may output information onto paper. That is, the output device 55 may include a printing device (so-called printer) capable of printing desired information onto paper. For example, the output device 55 may output information as data to a recording medium that can be attached externally to the model generation device 5.

[0177] Furthermore, the model generation device 5 can output information as data to external devices via the communication device 53. In this case, the communication device 53 may function as an output device that outputs information to an external device of the model generation device 5.

[0178] (4-2) Flow of the model generation process performed by the model generation device 5 Next, we will explain the model generation process performed by the model generation device 5, referring to Figure 19. Figure 19 is a flowchart showing the flow of the model generation process performed by the model generation device 5.

[0179] Typically, the model generation device 5 performs the model generation process before the information processing device 3 described above starts the defect prediction process. This is because the information processing device 3 uses the prediction model PM generated by the model generation device 5 to perform the defect prediction process. However, the model generation device 5 may also perform the model generation process at a desired timing after the information processing device 3 has started the defect prediction process using the prediction model PM. In this case, the model generation process can be considered equivalent to updating (in other words, improving or relearning) the prediction model PM already implemented in the information processing device 3.

[0180] As shown in Figure 19, the information acquisition unit 511 acquires multiple training data 610 necessary for machine learning to generate the predictive model PM (steps S21 to S24).

[0181] In this embodiment, the learning data 610 used is data that associates the above-mentioned process data 415, which is acquired and / or calculated when the molded object BO is actually fabricated, with the actual measurement results of defects (i.e., characteristics of the molded object BO) formed on the fabricated object BO. For this reason, the molded object BO is actually fabricated in order to allow the information acquisition unit 511 to acquire the learning data 610 (i.e., to perform the model generation process). Specifically, the above-mentioned molding device 1 may fabricate the molded object BO. Alternatively, in addition to or instead of the molding device 1, another molding device different from the molding device 1 may fabricate the molded object BO. In the following description, the molded object BO fabricated for the purpose of performing the model generation process will be referred to as "molded object BO*" to distinguish it from the molded object BO that is the target of the defect prediction process. However, the molded object BO that is the target of the defect prediction process may also be used as the molded object BO fabricated for the purpose of performing the model generation process.

[0182] As mentioned above, process data 415 does not necessarily have to include at least one of measurement information 411, measurement information 412, molding information 413, and feature information 414. Therefore, process data 415 included in training data 610 also does not necessarily have to include at least one of measurement information 411, measurement information 412, molding information 413, and feature information 414. In other words, training data does not necessarily have to include at least one of measurement information 411, measurement information 412, molding information 413, and feature information 414.

[0183] During the molding period in which the object BO* is fabricated for the purpose of model generation processing, the measurement device 2 measures the object to be measured, similar to the case in which the object BO subject to defect prediction processing is fabricated. As a result, at least one of the measurement information 411 and 412 is generated. The information acquisition unit 511 acquires at least one of the measurement information 411 and 412 generated during the molding period in which the object BO* is fabricated as part of the training data 610 (step S21). In other words, the information acquisition unit 511 acquires at least one of the measurement information 411 and 412, similar to the information acquisition unit 311 described above. For this reason, the processing in step S21 in Figure 19 may be the same as the processing in step S11 in Figure 9 described above. In the following description, the measurement information 411 and 412 acquired for the purpose of model generation processing will be referred to as "measurement information 611 and 612" respectively to distinguish them from the measurement information 411 and 412 acquired for the purpose of defect prediction processing. However, the information acquisition unit 511 does not need to acquire at least one of the measurement information 611 and 612 as part of the learning data 610.

[0184] Furthermore, in addition to or instead of measuring device 2, another measuring device different from measuring device 2 may measure the object to be measured. In other words, another measuring device different from measuring device 2 may generate at least one of the measurement information 611 and 612.

[0185] Furthermore, the information acquisition unit 511 acquires molding information 413 related to the molding of the molded object BO* as part of the training data 610 (step S22). In other words, the information acquisition unit 511 acquires molding information 413 in the same way as the information acquisition unit 311 described above. For this reason, the process in step S22 of Figure 19 may be the same as the process in step S12 of Figure 9 described above. In the following explanation, the molding information 413 acquired for the purpose of model generation processing will be referred to as "molding information 613" to distinguish it from the molding information 413 acquired for the purpose of defect prediction processing. However, the information acquisition unit 511 does not have to acquire the molding information 613 as part of the training data 610.

[0186] Furthermore, the information acquisition unit 511 calculates the feature quantities of the object to be measured based on at least one of the measurement information 611 and 612 acquired in step S21 (step S23), similar to the defect prediction unit 312 described above, which calculates the feature quantities of the object to be measured based on at least one of the measurement information 411 and 412. In other words, the information acquisition unit 511 generates feature quantity information 414, similar to the defect prediction unit 312 described above. For this reason, the processing in step S23 of Figure 19 may be the same as the processing in step S13 of Figure 9 described above. The generated feature quantity information 414 is used as part of the training data 610. In the following description, the feature quantity information 414 generated for model generation processing will be referred to as "feature quantity information 614" to distinguish it from the feature quantity information 414 generated for defect prediction processing. However, the information acquisition unit 511 does not have to generate the feature quantity information 614 as part of the training data 610.

[0187] Furthermore, as explained in the description of the defect prediction process, the feature quantities calculated based on the temperature image data D_T may be calculated by the temperature measuring device 21, and the feature quantities calculated based on the shape image data D_3D may be calculated by the shape measuring device 22. In this case, at least one of the measurement information 611 and 612 may include feature quantity information 614.

[0188] As mentioned above, the measurement information 411, measurement information 412, molding information 413, and feature information 414 obtained in the defect prediction process may be associated with location information 416 indicating the molding location. The measurement information 611, measurement information 612, molding information 613, and feature information 614 obtained in the model generation process may also be associated with location information 416 indicating the molding location where the molding process was performed to create the molded object BO*. In the following explanation, the location information 416 used for the model generation process will be referred to as "location information 616" to distinguish it from the location information 416 used for the defect prediction process.

[0189] Furthermore, the information acquisition unit 511 acquires the actual measurement results of defects formed in the molded object BO* (i.e., the characteristics of the molded object BO*) (step S24). For example, after the molding apparatus 1 has finished molding the molded object BO*, the molded object BO* may be transported from the molding apparatus 1 to a defect measurement device capable of measuring defects (e.g., an X-ray CT scanner). The defect measurement device may then actually measure the defects formed in the molded object BO*. The information acquisition unit 511 may then acquire the measurement results from the defect measurement device as the actual measurement results of the defects. The acquired defect measurement information 617, which shows the actual measurement results of the defects, is used as part of the training data 610. The defect measurement information 617 is used as a label for machine learning (i.e., a correct label or training data), as will be described later.

[0190] The defect measurement information 617 may include information similar to that of the defect information 425 described above. For example, the defect measurement information 617 may include information regarding the presence or absence of defects in each part of the molded object BO*. The defect measurement information 617 may include information regarding the density of the molded object BO*. The defect measurement information 617 may include information regarding the characteristics of defects formed on the molded object BO* (e.g., at least one of size, location, and shape).

[0191] Furthermore, as mentioned above, the density of the molded object BO itself may not be considered a defect of the molded object BO. Therefore, the information acquisition unit 511 may acquire, separately from the actual defect measurement information 617, the measurement results from a density measuring device that actually measured the density of the molded object BO* as actual density measurement information.

[0192] An example of the data structure of a training dataset 600 that stores the training data 610 acquired in this way is shown in Figure 20. As shown in Figure 20, the training dataset 600 includes multiple training data 610. Each training data 610 includes location information 616 indicating a certain molding location, process data 615 including measurement information 611, measurement information 612, molding information 613, and feature information 614 associated with the location information 616, and defect measurement information 617 indicating the actual measurement result of a defect at the molding location indicated by the location information 616. For this reason, typically, when one molded object BO* is fabricated, the information acquisition unit 511 may acquire a dataset containing multiple training data 610 corresponding to multiple different molding locations. Furthermore, when multiple molded objects BO* are fabricated, the information acquisition unit 511 may acquire multiple datasets corresponding to multiple different molded objects BO*.

[0193] In Figure 19 again, the information acquisition unit 511 continues to acquire training data 610 until the required number of training data 610 has been acquired (step S25). If the required number of training data 610 has been acquired (step S25: Yes), the model generation unit 512 performs machine learning to generate a predictive model PM using at least a portion of the training dataset 600 (step S26). In other words, the model generation unit 512 performs machine learning to generate a predictive model PM using at least a portion of the multiple training data 610 acquired in steps S21 to S24 (step S26).

[0194] Specifically, the model generation unit 512 extracts N (where N is an integer greater than or equal to 1) training data 610 from the training dataset 600 to be used for machine learning. Then, the model generation unit 512 uses the prediction model PM (for example, the prediction model PM before training or the prediction model PM that has already been trained but is subject to retraining) and the process data 615 contained in each training data 610 to generate defect information 425 regarding defects formed in the molded object BO*. In other words, the model generation unit 512 uses the prediction model PM and the process data 615 to predict defects formed in the molded object BO*. Note that the process of generating defect information 425 using the prediction model PM and the process data 615 may be the same as the process of generating defect information 425 using the prediction model PM and the process data 415 (i.e., the process in step S14 of Figure 9 above). Furthermore, in the following explanation, the defect information 425 generated in the model generation process will be referred to as "defect information 625" to distinguish it from the defect information 425 generated in the defect prediction process. The model generation unit 512 repeats the process of generating defect information 625 N times, using the N process data 615 contained in each of the N training data 610. As a result, N pieces of defect information 625 are generated.

[0195] Subsequently, the model generation unit 512 updates the prediction model PM using a loss function related to the error (i.e., difference) between the N defect information 625 and the N measured defect information 617 contained in each of the N training data 610. An example of a loss function is one that uses at least one of the following: Mean Absolute Error (MAE), Mean Square Error (MSE), Root Mean Square Error (RMSE), and coefficient of determination. In this case, the model generation unit 512 may update the prediction model PM so that the loss calculated using the loss function becomes small (preferably minimized). As a result, a prediction model PM is generated with improved defect prediction accuracy compared to the prediction model PM before training.

[0196] Furthermore, if the prediction model PM generates defect information 425 that includes information about the presence or absence of defects, it can be said that the prediction model PM is performing binary classification. In this case, the model generation unit 512 may update the prediction model PM using precision indicators in the confusion matrix, in addition to or instead of the loss function described above. Examples of precision indicators in the confusion matrix include at least one of recall, precision, and F-measure.

[0197] The model generation unit 512 may generate defect information 625 using all of the process data 615 included in the training data 610. Alternatively, the model generation unit 512 may generate defect information 625 using only a portion of the process data 615 included in the training data 610. In other words, the model generation unit 512 does not need to use other portions of the process data 615 included in the training data 610 to generate defect information 625. For example, the model generation unit 512 may generate defect information 625 using one portion of the model data 615 that is assumed to be necessary for efficiently generating the predictive model PM, while not using other portions of the process data 615 that are not assumed to be necessary for efficiently generating the predictive model PM. As an example, if the training data 610 includes process data 615 which contains all of the measurement information 611, measurement information 612, molding information 613, and feature information 614, the model generation unit 512 may generate defect information 625 using at least one of the measurement information 611, measurement information 612, molding information 613, and feature information 614, while not using at least one of the other of the measurement information 611, measurement information 612, molding information 613, and feature information 614. As another example, the model generation unit 512 may generate defect information 625 using part of the measurement information 611, part of the measurement information 612, part of the molding information 613, and / or part of the feature information 614, while not using other parts of the measurement information 611, other parts of the measurement information 612, other parts of the molding information 613, and / or other parts of the feature information 614.

[0198] The model generation unit 512 may generate multiple predictive model PMs corresponding to different combinations of information elements by changing the combination of information elements used to generate the predictive model PM. For example, the model generation unit 512 may generate a first predictive model PM using a first combination of information elements included in the process data 615, and generate a second predictive model PM different from the first predictive model PM using a second combination of information elements included in the process data 615 but different from the first combination. As an example, the model generation unit 512 may generate a first predictive model PM using all of the measurement information 611, measurement information 612, molding information 613, and feature information 614, and generate a second predictive model PM using the measurement information 611, measurement information 612, and feature information 614, but without using the molding information 613. As another example, the model generation unit 512 may generate a first prediction model PM using all of the measurement information 611, all of the measurement information 612, all of the molding information 613 and / or all of the feature information 614, and generate a second prediction model PM using some of the measurement information 611, some of the measurement information 612, some of the molding information 613 and / or some of the feature information 614, while not using the other parts of the measurement information 611, some of the measurement information 612, some of the molding information 613 and / or the other parts of the feature information 614.

[0199] The model generation unit 512 may perform validation of the generated predictive model PM after generating the predictive model PM. For example, the model generation unit 512 may extract M (where M is an integer greater than or equal to 1) training data 610 from the training dataset 600 to be used for validation. In this case, the model generation unit 512 may use a cross-validation method to extract the M training data 610. Subsequently, the model generation unit 512 may generate M defect information 625 using the generated predictive model PM and the M process data 615 contained in each of the M training data 610. Subsequently, the model generation unit 512 may evaluate the accuracy of the generated predictive model PM using a loss function (or the accuracy index value in the confusion matrix described above) relating to the error between the M defect information 625 and the M measured defect information 617 contained in each of the M training data 610. If the accuracy of the predictive model PM falls below the acceptable accuracy, the model generation unit 512 may repeat the machine learning process to generate the predictive model PM as described above.

[0200] (5) Variant (5-1) Modified model generation process When the model generation unit 512 extracts at least one training data 610 from the training dataset 600 to be used for machine learning, it may extract at least one training data 610 that satisfies predetermined extraction criteria.

[0201] The extraction criteria may include criteria related to measurement information 611. In this case, the model generation unit 512 may extract training data 610 containing measurement information 611 that satisfies the extraction criteria. The extraction criteria may include criteria related to measurement information 612. In this case, the model generation unit 512 may extract training data 610 containing measurement information 612 that satisfies the extraction criteria. The extraction criteria may include criteria related to molding information 613. In this case, the model generation unit 512 may extract training data 610 containing molding information 613 that satisfies the extraction criteria. The extraction criteria may include criteria related to feature information 614. In this case, the model generation unit 512 may extract training data 610 containing feature information 614 that satisfies the extraction criteria. The extraction criteria may include criteria related to location information 616. In this case, the model generation unit 512 may extract training data 610 containing location information 616 that satisfies the extraction criteria. The extraction criteria may include criteria related to actual defect measurement information 617. In this case, the model generation unit 512 may extract training data 610 that includes actual defect measurement information 617 that satisfies the extraction criteria.

[0202] As an example, the model generation unit 512 may extract at least one training data 610 that satisfies an extraction criterion regarding the molding condition information contained in the molding information 613. As an extraction criterion regarding the molding condition information, for example, the criterion "the molding conditions are set to predetermined conditions" may be used. In this case, training data 610 relating to a molded object BO* molded using specific molding conditions is extracted. As a result, the model generation unit 512 can generate a predictive model PM that is tailored to a molded object BO molded using certain specific molding conditions. In other words, the model generation unit 512 can generate a predictive model PM that is suitable for predicting defects formed in a molded object BO molded using certain specific molding conditions. Therefore, the defect prediction unit 312 can use such a predictive model PM to predict with greater accuracy the defects formed in a molded object BO when the molded object BO is molded using specific molding conditions.

[0203] For example, Figure 21 shows an example where the criterion "the molding condition is set to molding condition A" is used as an extraction criterion for molding condition information. In this case, the model generation unit 512 can generate a predictive model PM that matches the molded object BO molded using molding condition A. In other words, the model generation unit 512 can generate a predictive model PM that is suitable for predicting defects formed in the molded object BO molded using molding condition A. Therefore, the defect prediction unit 312 can use such a predictive model PM to predict with greater accuracy the defects formed in the molded object BO when the molded object BO is molded using molding condition A.

[0204] For example, Figure 22 shows an example where the criterion used for extracting molding condition information is "the molding condition is set to a throughput-priority condition, which prioritizes the throughput for molding the molded object BO* over the quality of the molded object BO*." In this case, the model generation unit 512 can generate a predictive model PM that matches the molded object BO molded using the throughput-priority condition. In other words, the model generation unit 512 can generate a predictive model PM that is suitable for predicting defects formed in the molded object BO molded using the throughput-priority condition. Therefore, the defect prediction unit 312 can use such a predictive model PM to predict defects formed in the molded object BO with greater accuracy when the molded object BO is molded using the throughput-priority condition. Note that the throughput for molding the molded object BO* can be said to be the time required to mold the molded object BO*, or the molding time of the molded object BO*.

[0205] Furthermore, in the example shown in Figure 22, the criterion for extracting molding condition information may be that "the molding conditions are set to quality-priority conditions, where the quality of the molded object BO* takes precedence over the throughput required to mold the molded object BO*." In this case, the model generation unit 512 can generate a predictive model PM that matches the molded object BO molded using quality-priority conditions. In other words, the model generation unit 512 can generate a predictive model PM that is suitable for predicting defects formed in the molded object BO molded using quality-priority conditions. Therefore, the defect prediction unit 312 can use such a predictive model PM to predict with even greater accuracy the defects formed in the molded object BO when the molded object BO is molded using quality-priority conditions.

[0206] Similarly, in the example shown in Figure 22, the criterion for extracting molding condition information may be that "the molding conditions are set to balance conditions that achieve both the quality of the molded object BO* and the throughput for molding the molded object BO*." In this case, the model generation unit 512 can generate a predictive model PM that matches the molded object BO molded using the balance conditions. In other words, the model generation unit 512 can generate a predictive model PM that is suitable for predicting defects formed in the molded object BO molded using the balance conditions. Therefore, the defect prediction unit 312 can use such a predictive model PM to predict with greater accuracy the defects formed in the molded object BO when the molded object BO is molded using the molding conditions of balance conditions.

[0207] As another example, the model generation unit 512 may extract at least one training data 610 that satisfies an extraction criterion for design information contained in the molding information 613. As an extraction criterion for design information, for example, a criterion such as "the three-dimensional shape of the molded object BO* is a predetermined shape" may be used. In this case, training data 610 relating to a molded object BO* having a specific three-dimensional shape is extracted. As a result, the model generation unit 512 can generate a prediction model PM that is tailored to a molded object BO having a specific three-dimensional shape. In other words, the model generation unit 512 can generate a prediction model PM that is suitable for predicting defects formed on a molded object BO having a specific three-dimensional shape. Therefore, the defect prediction unit 312 can use such a prediction model PM to predict with greater accuracy the defects formed on a molded object BO when a molded object BO having a specific three-dimensional shape is manufactured.

[0208] For example, Figure 23 shows an example where the criterion "the shape of the molded object BO* is cylindrical" is used as an extraction criterion for design information. In this case, the model generation unit 512 can generate a predictive model PM that matches the cylindrical molded object BO. In other words, the model generation unit 512 can generate a predictive model PM that is suitable for predicting defects formed in the cylindrical molded object BO. Therefore, the defect prediction unit 312 can use such a predictive model PM to predict defects formed in the molded object BO with greater accuracy when a cylindrical molded object BO is manufactured.

[0209] However, it is not always the case that a molded object BO* having the same shape as the molded object BO that is the target of the defect prediction process is molded in order to perform the model generation process. This is because the molded object BO that is the target of the defect prediction process does not necessarily have a simple shape. In this case, in the defect prediction process, the defect prediction unit 312 may divide the molded object BO into at least two divided objects BOd with different shapes, and use at least two prediction models PM corresponding to at least two different shapes to predict the defects formed in at least two divided objects BOd. For example, Figure 24 shows an example in which the molded object BO can be divided into a divided object BOd1 having a cylindrical shape and a divided object BOd2 having a rectangular parallelepiped shape. In this case, the defect prediction unit 312 may use a prediction model PM corresponding to the cylindrical molded object BO (i.e., a prediction model PM generated using training data 610 that satisfies the extraction criterion that "the shape of the molded object BO* is cylindrical") to predict the defects formed in the divided object BOd1 having a cylindrical shape. On the other hand, the defect prediction unit 312 may use a prediction model PM corresponding to a rectangular parallelepiped-shaped object BO (i.e., a prediction model PM generated using training data 610 that satisfies the extraction criterion that "the shape of the object BO* is a rectangular parallelepiped") to predict defects formed in the rectangular parallelepiped-shaped divided object BOd2. As a result, even if an object BO* having the same shape as the object BO is not fabricated in order to perform the model generation process, the defect prediction unit 312 can predict defects formed in the object BO with reasonable accuracy.

[0210] Considering that multiple prediction models PM corresponding to multiple different shapes may be used, the molding apparatus 1 may fabricate a small number of simple, distinct shapes as the fabricated objects BO* to be fabricated for the model generation process. For example, the molding apparatus 1 may fabricate a spherical object BO*, a rectangular object BO*, a cylindrical object BO*, and a conical object BO*. In other words, the molding apparatus 1 does not need to fabricate a large number of complex, distinct shapes BO*. Even in this case, the defect prediction unit 312 can predict defects formed on complex objects BO with reasonable accuracy by using multiple prediction models PM corresponding to multiple different shapes.

[0211] However, if a molded object BO* having the same shape as the molded object BO subject to defect prediction processing is molded, a prediction model PM that can predict defects in the molded object BO with greater accuracy can be generated compared to the case where a molded object BO* having a different shape from the molded object BO subject to defect prediction processing is molded. For this reason, the molding device 1 may mold a molded object BO* having the same shape as the molded object BO. In this case, the molding device 1 may mold a molded object BO* having the same size as the molded object BO (i.e., a molded object BO having a shape congruent to the molded object BO). Alternatively, the molding device 1 may mold a molded object BO* having a different size from the molded object BO (i.e., a molded object BO having a shape similar to the molded object BO).

[0212] Furthermore, the use of multiple prediction models PM to predict defects formed in a given molded object BO is not limited to cases where multiple prediction models PM corresponding to different shapes are used. The use of multiple prediction models PM to predict defects formed in a given molded object BO is not limited to cases where prediction models PM are generated using extraction criteria related to design information. In other words, the defect prediction unit 312 may, in any given scenario, predict defects formed in a given molded object BO using a prediction model PM generated using training data 610 that satisfies a first extraction criterion, and a prediction model PM generated using training data 610 that satisfies a second criterion different from the first extraction criterion.

[0213] As another example, the model generation unit 512 may extract at least one training data 610 that satisfies the extraction criteria for material information contained in the molding information 613. For example, the extraction criteria for material information may be "the molded object BO* is manufactured using a specific molding material." In this case, training data 610 relating to molded objects BO* manufactured using a specific molding material is extracted. As a result, the model generation unit 512 can generate a predictive model PM that is tailored to molded objects BO manufactured using a specific molding material. In other words, the model generation unit 512 can generate a predictive model PM that is suitable for predicting defects formed in molded objects BO manufactured using a specific molding material. Therefore, the defect prediction unit 312 can use such a predictive model PM to predict with greater accuracy the defects formed in molded objects BO when molded using a specific molding material.

[0214] For example, Figure 25 shows an example where the criterion "The object BO* is fabricated using fabrication material A" is used as an extraction criterion for material information. In this case, the model generation unit 512 can generate a predictive model PM that matches the object BO fabricated using fabrication material A. In other words, the model generation unit 512 can generate a predictive model PM that is suitable for predicting defects formed in the object BO fabricated using fabrication material A. Therefore, the defect prediction unit 312 can use such a predictive model PM to predict with greater accuracy the defects formed in the object BO when the object BO is fabricated using fabrication material A.

[0215] However, it is not always the case that the object BO* used for model generation processing is fabricated using the same fabrication material as the object BO that is subject to defect prediction processing. On the other hand, the fabrication material used to fabricate object BO may be approximated by a combination of two or more fabrication materials used to fabricate object BO*. For example, in a situation where a fabrication material containing metal a and a fabrication material containing metal b are used to fabricate object BO*, if the fabrication material used to fabricate object BO is an alloy c of metal a and metal b, then the fabrication material used to fabricate object BO (alloy c) can be approximated by a combination of the two fabrication materials (metal a and metal b) used to fabricate object BO*. In this case, the model generation unit 512 may generate a prediction model PM using both training data 610 that satisfy extraction criterion a, which states that "the object BO* is fabricated using metal a as the fabrication material," and training data 610 that satisfy extraction criterion b, which states that "the object BO* is fabricated using metal b as the fabrication material." In this case, the ratio of the number of training data 610 that satisfy extraction criterion a to the number of training data 610 that satisfy extraction criterion b may be set according to the ratio of metal a to metal b in alloy c. Using such a prediction model PM, the defect prediction unit 312 can predict with reasonable accuracy the defects that will be formed on the object BO when the object BO is fabricated using alloy c as the fabrication material.

[0216] In other words, the model generation unit 512 may generate a predictive model PM using both a dataset containing at least one training data 610 that satisfies the first extraction criterion and a dataset containing at least one training data 610 that satisfies the second extraction criterion. Alternatively, the model generation unit 512 may generate a predictive model PM using three or more datasets corresponding to three or more extraction criteria. As a result, even if the object BO* is not fabricated using the same fabrication material as the fabrication material used to fabricate the object BO, the defect prediction unit 312 can accurately predict defects that will be formed on the object BO.

[0217] Furthermore, the use of multiple datasets corresponding to multiple extraction criteria to generate a certain predictive model PM is not limited to cases where the material used to fabricate the object BO can be approximated by a combination of two or more materials used to fabricate the object BO*. The use of multiple datasets corresponding to multiple extraction criteria to generate a certain predictive model PM is not limited to cases where the predictive model PM is generated using extraction criteria related to material information. In other words, the model generation unit 512 may generate a single predictive model PM using multiple datasets corresponding to multiple extraction criteria in any given scenario.

[0218] (5-2) Modification of the prediction model In the above description, the information processing device 3 generates defect information 425 regarding defects formed on the object BO manufactured by the molding device 1, as an example of characteristic information regarding the characteristics of the object BO manufactured by the molding device 1. However, the information processing device 3 may generate characteristic information different from the defect information 425.

[0219] For example, the information processing device 3 may generate microstructure state information relating to the state of the metal structure constituting at least a part of the fabricated object BO, as an example of characteristic information. In other words, the information processing device 3 may predict the state of the metal structure constituting at least a part of the fabricated object BO. An example of microstructure state information is shown in Figure 26. As shown in Figure 26, the microstructure state information may include information about the orientation of the metal structure (so-called crystal orientation). The microstructure state information may include information about the Miller indices of the metal structure. The microstructure state information may include information about the size of the metal structure. The information about the size of the metal structure may include information about at least one of the length, width, and grain size of the metal structure. The microstructure state information may include information about the shape of the metal structure. The information about the shape of the metal structure may include information about the aspect ratio of the metal structure. The microstructure state information may include information about the type of crystal lattice of the metal structure.

[0220] For example, the information processing device 3 may generate mechanical property information relating to at least some of the mechanical properties of the molded object BO as an example of characteristic information. In other words, the information processing device 3 may predict at least some of the mechanical properties of the molded object BO. An example of mechanical property information is shown in Figure 27. As shown in Figure 27, the mechanical property information may include information regarding the tensile strength of the molded object BO. The mechanical property information may include information regarding the tensile strength of the molded object BO. The mechanical property information may include information regarding the 0.2% proof stress of the molded object BO. The mechanical property information may include information regarding the yield point of the molded object BO. The mechanical property information may include information regarding the elongation of the molded object BO. The mechanical property information may include information regarding the reduction of area of ​​the molded object BO. The mechanical property information may include information regarding the Young's modulus of the molded object BO. The mechanical property information may include information regarding the hardness (i.e., rigidity) of the molded object BO. The mechanical property information may include information regarding the fracture toughness value of the molded object BO. The mechanical properties information may include information regarding the high-temperature strength of the fabricated object BO. The mechanical properties information may also include information regarding the coefficient of thermal expansion of the fabricated object BO.

[0221] Even when the information processing device 3 generates characteristic information different from the defect information 425, the information processing device 3 may generate the characteristic information using process data 415 and a predictive model PM, similar to the case where the defect information 425 is generated. However, in this case, the information processing device 3 uses a learning model generated by machine learning to enable the generation of characteristic information based on the process data 415 as the predictive model PM. Similarly, the model generation device 5 performs machine learning to generate a predictive model PM capable of generating characteristic information based on the process data 415.

[0222] (5-3) Modified Defect Prediction Process In the above description, the information acquisition unit 311 associates the molding position with the measurement information 411. In other words, the information acquisition unit 311 associates the molding position (i.e., the position of the target part) where the molding device 1 was performing the molding process on the target part with information regarding the temperature of a part of the object to be measured (hereinafter referred to as the "target part"). In a modified example, the information acquisition unit 311 may use information regarding the temperature of the target part, calculated from the temperature of the target part and the temperature of a neighboring part located temporally and / or spatially near the target part, as the information regarding the temperature of the target part associated with the molding position. The neighboring part located temporally near the target part may be a part that satisfies the condition that the time difference between the time when the molding light EL was irradiated on the target part and the time when the molding light EL was irradiated on the neighboring part is within a predetermined time. The neighboring part located spatially near the target part may be a part that satisfies the condition that the distance between the target part and the neighboring part is within a predetermined distance.

[0223] For example, as shown in Figure 28(a), the information acquisition unit 311 may use a temperature calculated from the temperature of the target portion P11 and the temperatures of at least one neighboring portion (in the example shown in Figure 28(a), four neighbors P12 to P15) located temporally near the target portion P11 as the temperature of the target portion P11 of the solidified layer STL, which is an example of an object to be measured (in particular, a temperature associated with the position of the target portion P11, which is the fabrication position). Neighboring portions P12 and P13 are the portions that were irradiated by the fabrication light EL immediately before the fabrication light EL was irradiated onto the target portion. Neighboring portions P14 and P15 are the portions that were irradiated by the fabrication light EL immediately after the fabrication light EL was irradiated onto the target portion. In this case, neighboring portions P12 to P15 may be considered to be located near the target portion P11 in the scanning order (scanning direction) of the fabrication light EL. As an example, the information acquisition unit 311 may associate the temperature of the target portion P11 with the temperature obtained by applying a predetermined filtering process to the neighboring portions P12 to P15 as the temperature associated with the location of the target portion P11. The predetermined filtering process may be a filtering process for performing an averaging process. In this case, the predetermined filtering process may be a filtering process that uses at least one of an averaging filter and a median filter.

[0224] For example, as shown in Figure 28(b), the information acquisition unit 311 may use the temperature of the target portion P21 and the temperatures of at least one neighboring portion (eight neighbors P22 to P29 in the example shown in Figure 28(b)) that are spatially close to the target portion P21 as the temperature of the target portion P21 of the solidified layer STL (in particular, the temperature associated with the position of the target portion 21, which is the fabrication position). Neighboring portions P12 and P13 are the portions that were irradiated by the fabrication light EL immediately before the fabrication light EL was irradiated onto the target portion. Neighboring portions P14 and P15 are the portions that were irradiated by the fabrication light EL immediately after the fabrication light EL was irradiated onto the target portion. As an example, the information acquisition unit 311 may associate the temperature of the target portion P21 with the temperature obtained by applying the predetermined filtering process described in the previous paragraph to neighboring portions P22 to P29 as the temperature associated with the position of the target portion P21.

[0225] For example, as shown in Figure 28(c), the information acquisition unit 311 may use a temperature calculated from the temperature of the target portion P31, the temperature of the neighboring portion P32 of the (n-1)th solidified layer STL that at least partially overlaps with the target portion P31 in the stacking direction of the solidified layer STL, and the temperature of the neighboring portion P33 of the (n+1)th solidified layer STL that at least partially overlaps with the target portion P31 in the stacking direction of the solidified layer STL. In this case, the neighboring portions P32 to P33 may be considered to be located near the target portion P31 in the stacking direction of the solidified layer STL. As an example, the information acquisition unit 311 may associate the temperature obtained by applying the predetermined filtering process described in the previous paragraph to the temperature of the target portion P31 and the neighboring portions P32 to P33 with the temperature associated with the position of the target portion P31.

[0226] Although detailed explanations are omitted to avoid repetition, the information acquisition unit 311 may use information about the shape of the target part and the shape of a neighboring part located temporally and / or spatially near the target part as information about the shape of the target part associated with the build position (i.e., measurement information 412). The information acquisition unit 311 may use information calculated from information about the shape of the target part and the shape of a neighboring part located temporally and / or spatially near the target part as information about the build of the target part associated with the build position (i.e., build information 413). The information acquisition unit 311 may use information about the feature quantities of the target part and the feature quantities calculated from the feature quantities of a neighboring part located temporally and / or spatially near the target part as information about the feature quantities of the target part associated with the build position (i.e., feature quantity information 414).

[0227] In the above description, the temperature measuring device 21 outputs temperature image data D_T, which represents the temperature image IMG_T, to the information processing device 3 as measurement information 411. However, the temperature measuring device 21 may also output temperature video data, which represents a video corresponding to a plurality of temporally or spatially continuous temperature images IMG_T, to the information processing device 3 as measurement information 411. Similarly, the shape measuring device 22 may output shape video data, which represents a video corresponding to a plurality of temporally continuous shape images IMG_3D, to the information processing device 3 as measurement information 412. In this case, the defect prediction unit 312 may calculate the time change of the feature quantities of the object being measured based on at least one of the measurement information 411 and 412. That is, the defect prediction unit 312 may generate feature quantity information 414, which represents the time change of the feature quantities of the object being measured, based on at least one of the measurement information 411 and 412. For example, the defect prediction unit 312 may generate feature quantity information 414, which represents the time change of temperature of each part of the object being measured, based on the measurement information 411. For example, the defect prediction unit 312 may generate feature information 414 showing the time change in the number of sputter SPTs occurring in each part of the object being measured, based on the measurement information 411. For example, the defect prediction unit 312 may generate feature information 414 showing the time change in the thickness of the solidified layer STL, based on the measurement information 412. For example, the defect prediction unit 312 may generate feature information 414 showing the time change in the surface roughness (or arbitrary surface properties) of the solidified layer STL, based on the measurement information 412.

[0228] (5-3) Other variations In the above description, the molding system SYS includes an information processing device 3 in addition to the molding device 1. However, the molding device 1 (for example, the control unit 16) may include the information processing device 3. In other words, the control unit 16 may also function as the information processing device 3. Alternatively, the molding device 1 may include the information processing device 3 in addition to the control unit 16. Furthermore, the control unit 1 may perform some of the functions of the information processing device 3.

[0229] In the above description, the molding system SYS is equipped with an information processing device 3 in addition to the measuring device 2. However, the measuring device 2 may include the information processing device 3. For example, the information processing device 3 may be included in the measuring device 2. In other words, the measuring device 2 may also function as the information processing device 3. Furthermore, the measuring device 2 may perform some of the functions of the information processing device 3.

[0230] The molding system SYS may include a model generation device 5. For example, the information processing device 3 may include a model generation device 5. That is, the information processing device 3 may also function as a model generation device 5. For example, the molding device 1 (e.g., control unit 16) may include a model generation device 5. The molding device 1 (e.g., control unit 16) may include an information processing device 3 that includes a model generation device 5. That is, the control unit 16 may also function as a model generation device 5. Alternatively, the molding device 1 may include a model generation device 5 separately from the control unit 16. Furthermore, the control unit 1 may perform some of the functions of the model generation device 5. The molding device 1 may include an information processing device 3 that includes a model generation device 5 separately from the control unit 16. For example, the measuring device 2 may include a model generation device 5. The measuring device 2 may include an information processing device 3 that includes a model generation device 5. That is, the measuring device 2 may also function as a model generation device 5. Furthermore, the measuring device 2 may perform some of the functions of the model generation device 5.

[0231] In the above description, the SYS fabrication system melts the fabrication material using EL (electroluminescent beam). However, the SYS fabrication system may also melt the fabrication material using any energy beam. Examples of any energy beam include at least one of charged particle beams and electromagnetic waves. Examples of charged particle beams include at least one of electron beams and ion beams.

[0232] (6) Addendum [Note 1] An acquisition unit that acquires measurement information related to the formation of a molded object obtained during the formation period of a molded object, which is formed by the solidification of a molding material melted by irradiation with an energy beam, An information processing unit generates characteristic information relating to the characteristics of the molded object using the measurement information acquired by the acquisition unit. An information processing device equipped with the following features. [Note 2] The characteristic information includes at least one of the following: defect information relating to defects formed inside the molded object; density information relating to the density of at least a part of the molded object; microstructure information relating to the state of the metal structure constituting at least a part of the molded object; and mechanical characteristic information relating to the mechanical properties of at least a part of the molded object. The information processing device described in Appendix 1. [Note 3] The aforementioned defect includes a void formed inside the molded object. The defect information includes void information relating to the void. The information processing device described in Appendix 2. [Note 4] The defect information includes at least one of the following: information regarding the presence or absence of the defect and information regarding the characteristics of the defect. The information processing device described in Appendix 2 or 3. [Note 5] The characteristics of the defect include at least one of the size of the defect, the location of the defect, and the shape of the defect. The information processing device described in Appendix 4. [Note 6] The microstructure information includes information regarding at least one of the following: the orientation of the metal structure, the Miller index of the metal structure, the size of the metal structure, the shape of the metal structure, and the type of crystal lattice of the metal structure. An information processing device as described in any one of the items 2 to 5 of the appendix. [Note 7] The mechanical property information includes information regarding at least one of the following: the tensile strength of the molded object, the 0.2% proof stress of the molded object, the yield point of the molded object, the elongation of the molded object, the reduction of area of ​​the molded object, the Young's modulus of the molded object, the hardness of the molded object, the fracture toughness of the molded object, the fatigue limit of the molded object, the high-temperature strength of the molded object, and the coefficient of linear expansion of the molded object. An information processing device as described in any one of the items 2 to 6 of the appendix. [Note 8] The system further comprises an output unit that outputs the aforementioned characteristic information. An information processing device as described in any one of the items 1 to 7 of the appendix. [Note 9] The output unit includes a display unit that displays the characteristic information. The information processing device described in Appendix 8. [Note 10] The output unit includes a transmission unit that transmits the characteristic information to a molding apparatus that fabricates the molded object. The information processing device described in Appendix 8 or 9. [Note 11] The output unit outputs the measurement information along with the characteristic information. An information processing device as described in any one of the appendices 8 to 10. [Note 12] The characteristic information includes first information relating to the characteristics of a part of the molded object, The measurement information includes second information obtained from measurements relating to the molding of the one part of the molded object, The output unit outputs the characteristic information and the measurement information in an output format in which the first information and the second information are associated. The information processing device described in Appendix 11. [Note 13] The output unit includes a display unit that displays the characteristic information and the measurement information in an output format in which the first information and the second information are associated. The information processing device described in Appendix 12. [Note 14] The output unit includes a transmission unit that transmits the characteristic information and the measurement information to a molding device that fabricates the molded object. An information processing device as described in any one of the appendices 11 to 13. [Note 15] The information processing unit further generates energy information relating to the amount of energy transmitted from the energy beam to the molding material based on the measurement information. The output unit outputs the energy information along with the characteristic information. An information processing device as described in any one of the appendices 8 to 14. [Note 16] The characteristic information includes first information relating to the characteristics of a part of the molded object, The energy information includes second information relating to the amount of energy transferred from the energy beam to the molding material in the molding of the one part of the molded object, The output unit outputs the characteristic information and the energy information in an output format in which the first information and the second information are associated. The information processing device described in Appendix 15. [Note 19] The output unit includes a display unit that displays the characteristic information and the energy information in an output format in which the first information and the second information are associated. The information processing device described in Appendix 16. [Note 18] The output unit includes a transmission unit that transmits the characteristic information and the energy information to a molding apparatus that fabricates the object. An information processing device as described in any one of the appendices 15 to 17. [Note 19] The output unit outputs the characteristic information during the molding period in which the molded object is formed and / or after the molding of the molded object is completed. An information processing device as described in any one of the appendices 8 to 18. [Note 20] The information processing unit generates the characteristic information using the measurement information and a learning model generated by machine learning to enable the generation of the characteristic information based on the measurement information. An information processing device as described in any one of the appendices 1 to 19. [Note 21] The measurement information includes information about the shape of at least a part of the molded object obtained during the molding period. An information processing device as described in any one of the items 1 to 20 of the appendix. [Note 22] The aforementioned molded object is formed by irradiating at least a portion of the layered material layer formed by the supplied molding material with the energy beam, causing at least a portion of the melted material layer to solidify. The measurement information includes information regarding the shape of at least a portion of the material layer obtained during the molding period. An information processing device as described in any one of the items from Appendix 1 to 21. [Note 23] The molded object includes a solidified material formed by the solidification of at least a portion of the molded material that has been melted by irradiation of at least a portion of the supplied molded material with the energy beam, The measurement information includes information regarding the shape of at least a portion of the solidified material obtained during the molding period. An information processing device as described in any one of the items 1 to 22 of the appendix. [Note 24] The measurement information includes information about at least a portion of the molten pool formed by the melting of the metal material due to irradiation with the energy beam. An information processing device as described in any one of the items 1 to 23 of the appendix. [Note 25] The acquisition unit further acquires molding information relating to the molding of the molded object, The information processing unit generates the characteristic information using the measurement information and the molding information. An information processing device as described in any one of the items 1 to 24 of the appendix. [Note 26] The information processing unit generates the characteristic information using the measurement information, the molding information, and a learning model generated by machine learning to enable the generation of the characteristic information based on the measurement information and the molding information. Information processing device described in Appendix 25 [Note 27] The modeling information includes at least one of apparatus information related to a modeling apparatus that models the modeled object, material information related to a modeling material used for modeling the modeled object, and design information of the modeled object The information processing apparatus according to Supplementary Note 25 or 26. [Supplementary Note 28] The apparatus information includes at least one of information related to modeling conditions used by the modeling apparatus to model the modeled object, and information related to the operation status of the modeling apparatus The information processing apparatus according to Supplementary Note 27. [Supplementary Note 29] An acquisition unit that acquires modeling information related to modeling of a modeled object formed by solidification of a modeling material melted by irradiation of an energy beam, An information processing unit that generates characteristic information related to characteristics of the modeled object using the modeling information acquired by the acquisition unit An information processing apparatus comprising: [Supplementary Note 30] The characteristic information includes at least one of defect information related to defects formed inside the modeled object, density information related to the density of at least a part of the modeled object, microstructure state information related to the state of a metal microstructure constituting at least a part of the modeled object, and mechanical characteristic information related to mechanical characteristics of at least a part of the modeled object The information processing apparatus according to Supplementary Note 29. [Supplementary Note 31] The defects include voids formed inside the modeled object The information processing apparatus according to Supplementary Note 30. [Supplementary Note 32] The defect information includes void information related to the voids The information processing apparatus according to Supplementary Note 31. [Supplementary Note 33] The defect information includes at least one of information related to the presence or absence of the defects and information related to characteristics of the defects The information processing apparatus according to any one of Supplementary Notes 30 to 32. [Supplementary Note 34] The characteristics of the defects include at least one of the size of the defects, the position of the defects, and the shape of the defects The information processing device described in Appendix 33. [Note 35] The microstructure information includes information regarding at least one of the orientation of the metal structure, the size of the metal structure, the shape of the metal structure, and the type of crystal lattice of the metal structure. An information processing device as described in any one of the appendices 30 to 34. [Note 36] The mechanical property information includes information regarding at least one of the following: the tensile strength of the molded object, the 0.2% proof stress of the molded object, the yield point of the molded object, the elongation of the molded object, the reduction of area of ​​the molded object, the Young's modulus of the molded object, the hardness of the molded object, the fracture toughness of the molded object, the fatigue limit of the molded object, the high-temperature strength of the molded object, and the coefficient of linear expansion of the molded object. An information processing device as described in any one of the appendices 30 to 35. [Note 37] The system further comprises an output unit that outputs the aforementioned characteristic information. An information processing device as described in any one of the items 29 to 36 of the appendix. [Note 38] The output unit includes a display unit that displays the characteristic information. The information processing device described in Appendix 37. [Note 39] The output unit includes a transmission unit that transmits the characteristic information to a molding apparatus that fabricates the molded object. The information processing device described in Appendix 37 or 38. [Note 40] The output unit outputs the molding information along with the characteristic information. An information processing device as described in any one of the appendices 37 to 39. [Note 41] The characteristic information includes first information relating to the characteristics of a part of the molded object, The molding information includes second information relating to the one part of the molded object, The output unit outputs the characteristic information and the molding information in an output format in which the first information and the second information are associated. The information processing device described in Appendix 40. [Note 42] The output unit includes a display unit that displays the characteristic information and the molding information in an output format in which the first information and the second information are associated. The information processing device described in Appendix 41. [Note 43] The output unit includes a transmission unit that transmits the characteristic information and the molding information to a molding apparatus that fabricates the molded object. An information processing device as described in any one of the appendices 40 to 42. [Note 44] The information processing unit further generates energy information relating to the amount of energy transferred from the energy beam to the molding material based on the measurement information relating to the molding of the molded object obtained during the molding period of the molded object. The output unit outputs the energy information along with the characteristic information. An information processing device as described in any one of the items 37 to 43 of the appendix. [Note 45] The characteristic information includes first information relating to the characteristics of a part of the molded object, The energy information includes second information relating to the amount of energy transferred from the energy beam to the molding material in the molding of the one part of the molded object, The output unit outputs the characteristic information and the energy information in an output format in which the first information and the second information are associated. The information processing device described in Appendix 44. [Note 46] The output unit includes a display unit that displays the characteristic information and the energy information in an output format in which the first information and the second information are associated. The information processing device described in Appendix 45. [Note 47] The output unit includes a transmission unit that transmits the characteristic information and the energy information to a molding apparatus that fabricates the object. An information processing device as described in any one of the appendices 44 to 46. [Note 48] The output unit outputs the characteristic information during the molding period in which the molded object is formed and / or after the molding of the molded object is completed. The information processing apparatus according to any one of appendices 37 to 47. [Appendix 49] The information processing unit generates the characteristic information using the modeling information and a learned model generated by machine learning so as to be capable of generating the characteristic information based on the modeling information The information processing apparatus according to any one of appendices 29 to 48 [Appendix 50] The modeling information includes at least one of apparatus information related to a modeling apparatus that models the modeled object, material information related to a modeling material used for modeling the modeled object, and design information of the modeled object The information processing apparatus according to any one of appendices 29 to 49. [Appendix 51] The apparatus information includes at least one of information related to modeling conditions used by the modeling apparatus to model the modeled object, and information related to an operation status of the modeling apparatus The information processing apparatus according to appendix 50. [Appendix 52] The acquisition unit acquires measurement information related to modeling of the modeled object obtained during a modeling period of the modeled object, The information processing unit generates the characteristic information using the measurement information and the modeling information acquired by the acquisition unit The information processing apparatus according to any one of appendices 29 to 51. [Appendix 53] The information processing unit generates the characteristic information using the modeling information, the measurement information, and a learned model generated by machine learning so as to be capable of generating the characteristic information based on the modeling information and the measurement information The information processing apparatus according to appendix 52. [Appendix 54] The measurement information includes information related to a shape of at least a part of the modeled object, obtained during the modeling period The information processing apparatus according to appendix 52 or 53. [Appendix 55] The aforementioned molded object is formed by irradiating at least a portion of the layered material layer formed by the supplied molding material with the energy beam, causing at least a portion of the melted material layer to solidify. The measurement information includes information regarding the shape of at least a portion of the material layer obtained during the molding period. An information processing device as described in any one of the appendices 52 to 54. [Note 56] The molded object includes a solidified material formed by the solidification of at least a portion of the molded material that has been melted by irradiation of at least a portion of the supplied molded material with the energy beam, The measurement information includes information regarding the shape of at least a portion of the solidified material obtained during the molding period. An information processing device as described in any one of the appendices 52 to 55. [Note 57] The measurement information includes information about at least a portion of the molten pool formed by the melting of the metal material due to irradiation with the energy beam. An information processing device as described in any one of the appendices 52 to 56. [Note 58] An information processing device described in any one of the items from Appendix 1 to 57, A molding apparatus capable of creating objects by solidifying molten powdered metal material upon irradiation with an energy beam, Measuring device and Equipped with, The measurement information is a molding system obtained using the measurement device. [Note 59] To acquire measurement information related to the formation of a molded object obtained during the formation period of a molded object, which is formed by the solidification of a molding material melted by irradiation with an energy beam, Using the acquired measurement information, characteristic information regarding the properties of the molded object is generated. Information processing methods including [Note 60] To obtain information about the shape of an object formed by the solidification of a molten material irradiated with an energy beam, Using the acquired molding information, characteristic information relating to the characteristics of the molded object is generated. Information processing methods including [Note 61] A computer program that causes a computer to execute the information processing methods described in Appendix 59 or 60. [Note 62] A recording medium on which the computer program described in Appendix 61 is recorded. [Note 63] The process involves creating an object by solidifying a molten material that has been irradiated with an energy beam, During the molding period of the aforementioned molded object, measurements related to the molding of the aforementioned molded object shall be taken. To acquire characteristic information regarding the properties of the molded object generated using the measurement information obtained by the above measurement, A molding method that includes this. [Note 64] The characteristic information includes at least one of the following: defect information relating to defects formed inside the molded object; density information relating to the density of at least a part of the molded object; microstructure information relating to the state of the metal structure constituting at least a part of the molded object; and mechanical characteristic information relating to the mechanical properties of at least a part of the molded object. The molding method described in Appendix 63. [Note 65] The aforementioned defect includes a void formed inside the molded object. The molding method described in Appendix 64. [Note 66] The defect information includes void information relating to the void. The molding method described in Appendix 65. [Note 67] The defect information includes at least one of the following: information regarding the presence or absence of the defect and information regarding the characteristics of the defect. The molding method described in any one of the appendices 64 to 66. [Note 68] The characteristics of the defect include at least one of the size of the defect, the location of the defect, and the shape of the defect. The molding method described in Appendix 67. [Note 69] The microstructure information includes information regarding at least one of the orientation of the metal structure, the size of the metal structure, the shape of the metal structure, and the type of crystal lattice of the metal structure. The molding method described in any one of the appendices 64 to 68. [Note 70] The mechanical property information includes information regarding at least one of the following: the tensile strength of the molded object, the 0.2% proof stress of the molded object, the yield point of the molded object, the elongation of the molded object, the reduction of area of ​​the molded object, the Young's modulus of the molded object, the hardness of the molded object, the fracture toughness of the molded object, the fatigue limit of the molded object, the high-temperature strength of the molded object, and the coefficient of linear expansion of the molded object. The molding method described in any one of the appendices 64 to 69. [Note 71] Further includes outputting the aforementioned characteristic information. The molding method described in any one of the appendices 63 to 70. [Note 72] Outputting the aforementioned information includes displaying the characteristic information. The molding method described in Appendix 71. [Note 73] Outputting the aforementioned information includes transmitting the characteristic information to a molding apparatus that fabricates the aforementioned object. The molding method described in Appendix 71 or 72. [Note 74] The output described above includes outputting the measurement information along with the characteristic information. The molding method described in any one of the appendices 71 to 73. [Note 75] The characteristic information includes first information relating to the characteristics of a part of the molded object, The measurement information includes second information obtained from measurements relating to the molding of the one part of the molded object, The output described above includes outputting the characteristic information and the measurement information in an output format in which the first information and the second information are associated. The molding method described in Appendix 74. [Note 76] The output described above includes displaying the characteristic information and the measurement information in an output format in which the first information and the second information are associated. The molding method described in Appendix 75. [Note 77] The output includes transmitting the characteristic information and the measurement information to a molding device that fabricates the molded object. The molding method described in any one of the appendices 74 to 76. [Note 78] The method further includes generating energy information relating to the amount of energy transferred from the energy beam to the molding material based on the measurement information, The output includes outputting the energy information along with the characteristic information. The molding method described in any one of the appendices 71 to 77. [Note 79] The characteristic information includes first information relating to the characteristics of a part of the molded object, The energy information includes second information relating to the amount of energy transferred from the energy beam to the molding material in the molding of the one part of the molded object, The output described above includes outputting the characteristic information and the energy information in an output format in which the first information and the second information are associated. The molding method described in Appendix 78. [Note 80] The output described above includes displaying the characteristic information and the energy information in an output format in which the first information and the second information are associated. The molding method described in Appendix 79. [Note 81] The output includes transmitting the characteristic information and the energy information to a molding apparatus that fabricates the object. The molding method described in any one of the appendices 78 to 80. [Note 82] The output includes outputting the characteristic information during the printing period in which the object is being printed and / or after the printing of the object is completed. The molding method described in any one of the appendices 71 to 81. [Note 83] The acquisition described above includes acquiring the characteristic information generated using the measurement information and a learning model generated by machine learning to enable the generation of the characteristic information based on the measurement information. The molding method described in any one of the appendices 63 to 82. [Note 84] The measurement information includes information about the shape of at least a part of the molded object obtained during the molding period. The molding method described in any one of the appendices 63 to 83. [Note 85] The aforementioned molded object is formed by irradiating at least a portion of the layered material layer formed by the supplied molding material with the energy beam, causing at least a portion of the melted material layer to solidify. The measurement information includes information regarding the shape of at least a portion of the material layer obtained during the molding period. The molding method described in any one of the appendices 63 to 84. [Note 86] The molded object includes a solidified material formed by the solidification of at least a portion of the molded material that has been melted by irradiation of at least a portion of the supplied molded material with the energy beam, The measurement information includes information regarding the shape of at least a portion of the solidified material obtained during the molding period. The molding method described in any one of the appendices 63 to 85. [Note 87] The measurement information includes information about at least a portion of the molten pool formed by the melting of the metal material due to irradiation with the energy beam. The molding method described in any one of the appendices 63 to 86. [Note 88] The acquisition described above includes acquiring the characteristic information generated using the measurement information and the molding information relating to the molding of the molded object. The molding method described in any one of the appendices 63 to 87. [Note 89] The acquisition described above includes acquiring the characteristic information generated using the measurement information, the molding information, and a learning model generated by machine learning to enable the generation of the characteristic information based on the measurement information and the molding information. The molding method described in Appendix 88 [Note 90] The molding information includes at least one of the following: device information relating to a molding apparatus for molding the molded object; material information relating to the molding material used for molding the molded object; and design information of the molded object. The molding method described in Appendix 88 or 89. [Note 91] The device information includes at least one of the following: information regarding the molding conditions used by the molding apparatus to mold the object, and information regarding the operating status of the molding apparatus. The molding method described in Appendix 90. [Note 92] The process involves creating an object by solidifying a molten material that has been irradiated with an energy beam, To obtain characteristic information relating to the characteristics of the aforementioned molded object, which is generated using the molding information relating to the molding of the aforementioned molded object. A molding method that includes this. [Note 93] The characteristic information includes at least one of the following: defect information relating to defects formed inside the molded object; density information relating to the density of at least a part of the molded object; microstructure information relating to the state of the metal structure constituting at least a part of the molded object; and mechanical characteristic information relating to the mechanical properties of at least a part of the molded object. The molding method described in Appendix 92. [Note 94] The aforementioned defect includes a void formed inside the molded object. The molding method described in Appendix 93. [Note 95] The defect information includes void information relating to the void. The molding method described in Appendix 94. [Note 96] The defect information includes at least one of the following: information regarding the presence or absence of the defect and information regarding the characteristics of the defect. The molding method described in any one of the appendices 93 to 95. [Note 97] The characteristics of the defect include at least one of the size of the defect, the location of the defect, and the shape of the defect. The molding method described in Appendix 96. [Note 98] The microstructure information includes information regarding at least one of the orientation of the metal structure, the size of the metal structure, the shape of the metal structure, and the type of crystal lattice of the metal structure. The molding method described in any one of the appendices 93 to 97. [Note 99] The mechanical property information includes information regarding at least one of the following: the tensile strength of the molded object, the 0.2% proof stress of the molded object, the yield point of the molded object, the elongation of the molded object, the reduction of area of ​​the molded object, the Young's modulus of the molded object, the hardness of the molded object, the fracture toughness of the molded object, the fatigue limit of the molded object, the high-temperature strength of the molded object, and the coefficient of linear expansion of the molded object. The molding method described in any one of the appendices 93 to 98. [Note 100] Further includes outputting the aforementioned characteristic information. The molding method described in any one of the appendices 92 to 99. [Note 101] Outputting the aforementioned information includes displaying the characteristic information. The molding method described in Appendix 100. [Note 102] Outputting the aforementioned information includes transmitting the characteristic information to a molding apparatus that fabricates the aforementioned object. The molding method described in Appendix 100 or 101. [Note 103] The output described above includes outputting the molding information along with the characteristic information. The molding method described in any one of the appendices 100 to 102. [Note 104] The characteristic information includes first information relating to the characteristics of a part of the molded object, The molding information includes second information relating to the one part of the molded object, The output described above includes outputting the characteristic information and the molding information in an output format in which the first information and the second information are associated. The molding method described in Appendix 103. [Note 105] The output described above includes displaying the characteristic information and the molding information in an output format in which the first information and the second information are associated. The molding method described in Appendix 104. [Note 106] The output includes transmitting the characteristic information and the molding information to a molding apparatus that fabricates the molded object. The molding method described in any one of the appendices 103 to 105. [Note 107] The method further includes generating energy information relating to the amount of energy transferred from the energy beam to the molding material based on measurement information relating to the molding of the molded object obtained during the molding period of the molded object, The output includes outputting the energy information along with the characteristic information. The molding method described in any one of the appendices 100 to 106. [Note 108] The characteristic information includes first information relating to the characteristics of a part of the molded object, The energy information includes second information relating to the amount of energy transferred from the energy beam to the molding material in the molding of the one part of the molded object, The output described above includes outputting the characteristic information and the energy information in an output format in which the first information and the second information are associated. The molding method described in Appendix 107. [Note 109] The output described above includes displaying the characteristic information and the energy information in an output format in which the first information and the second information are associated. The molding method described in Appendix 108. [Note 110] The output includes transmitting the characteristic information and the energy information to a molding apparatus that fabricates the object. The molding method described in any one of the appendices 107 to 109. [Note 111] The output includes outputting the characteristic information during the printing period in which the object is being printed and / or after the printing of the object is completed. The molding method described in any one of the appendices 100 to 110. [Note 112] The acquisition described above includes acquiring the characteristic information generated using the molding information and a learning model generated by machine learning to enable the generation of the characteristic information based on the molding information. The molding method described in any one of the appendices 92 to 111 [Note 113] The molding information includes at least one of the following: device information relating to a molding apparatus for molding the molded object; material information relating to the molding material used for molding the molded object; and design information of the molded object. The molding method described in any one of the appendices 92 to 112. [Note 114] The device information includes at least one of the following: information regarding the molding conditions used by the molding apparatus to mold the object, and information regarding the operating status of the molding apparatus. The molding method described in Appendix 113. [Note 115] The acquisition unit acquires measurement information related to the molding of the molded object obtained during the molding period of the molded object, The acquisition described above includes acquiring the characteristic information generated using the measurement information and the measurement information related to the molding of the molded object obtained during the molding period of the molded object. The molding method described in any one of the appendices 92 to 114. [Note 116] The acquisition described above includes acquiring the characteristic information generated using the molding information, the measurement information, and a learning model generated by machine learning to enable the generation of the characteristic information based on the molding information and the measurement information. The molding method described in Appendix 115. [Note 117] The measurement information includes information about the shape of at least a part of the molded object obtained during the molding period. The molding method described in Appendix 115 or 116. [Note 118] The aforementioned molded object is formed by irradiating at least a portion of the layered material layer formed by the supplied molding material with the energy beam, causing at least a portion of the melted material layer to solidify. The measurement information includes information regarding the shape of at least a portion of the material layer obtained during the molding period. The molding method described in any one of the appendices 115 to 117. [Note 119] The molded object includes a solidified material formed by the solidification of at least a portion of the molded material that has been melted by irradiation of at least a portion of the supplied molded material with the energy beam, The measurement information includes information regarding the shape of at least a portion of the solidified material obtained during the molding period. The molding method described in any one of the appendices 115 to 118. [Note 120] The measurement information includes information about at least a portion of the molten pool formed by the melting of the metal material due to irradiation with the energy beam. The molding method described in any one of the appendices 115 to 119. [Note 121] An acquisition unit that acquires measurement information related to the formation of a molded object obtained during the formation period of a molded object formed by the solidification of a powdered metal material melted by irradiation with an energy beam, and characteristic information related to the characteristics of the molded object. An output unit that outputs the characteristic information along with the measurement information. An information output device equipped with the following features. [Note 122] The characteristic information includes first information relating to the characteristics of a part of the molded object, The measurement information includes second information obtained from measurements relating to the molding of the one part of the molded object, The output unit outputs the characteristic information and the measurement information in an output format in which the first information and the second information are associated. The information output device described in Appendix 121. [Note 123] The output unit includes a display unit that displays the characteristic information and the measurement information in an output format in which the first information and the second information are associated. The information output device described in Appendix 122. [Note 124] The output unit includes a transmission unit that transmits the characteristic information and the measurement information to a molding device that fabricates the molded object. An information output device as described in any one of the appendices 121 to 123. [Note 125] The output unit outputs the characteristic information along with the measurement information during the molding period in which the molded object is formed and / or after the molding of the molded object is completed. An information output device as described in any one of the appendices 121 to 124. [Note 126] The characteristic information includes at least one of the following: defect information relating to defects formed inside the molded object; density information relating to the density of at least a part of the molded object; microstructure information relating to the state of the metal structure constituting at least a part of the molded object; and mechanical characteristic information relating to the mechanical properties of at least a part of the molded object. An information output device as described in any one of the appendices 121 to 125. [Note 127] The acquisition unit acquires energy information relating to the amount of energy transferred from the energy beam to the metal material during the formation of the molded object by the solidification of the powdered metal material melted by irradiation with the energy beam. The output unit outputs the energy information along with the measurement information. An information output device as described in any one of the appendices 121 to 126. [Note 128] An acquisition unit that acquires energy information relating to the amount of energy transferred from the energy beam to the metal material in the process of creating an object by solidifying a powdered metal material that has been molten by irradiation with an energy beam, and characteristic information relating to the characteristics of the object. An output unit that outputs the characteristic information along with the energy information. An information output device equipped with the following features. [Note 129] The characteristic information includes first information relating to the characteristics of a part of the molded object, The energy information includes second information relating to the amount of energy transferred from the energy beam to the metal material in the formation of the one part of the molded object, The output unit outputs the characteristic information and the energy information in an output format in which the first information and the second information are associated. The information output device according to claim 128. [Note 130] The output unit includes a display unit that displays the characteristic information and the energy information in an output format in which the first information and the second information are associated. The information output device described in Appendix 129. [Note 131] The output unit includes a transmission unit that transmits the characteristic information and the energy information to a molding apparatus that fabricates the object. An information output device as described in any one of the appendices 128 to 130. [Note 133] The output unit outputs the characteristic information along with the energy information during the molding period in which the molded object is formed and / or after the molding of the molded object is completed. An information output device as described in any one of the appendices 128 to 132. [Note 134] The characteristic information includes at least one of the following: defect information relating to defects formed inside the molded object; density information relating to the density of at least a part of the molded object; microstructure information relating to the state of the metal structure constituting at least a part of the molded object; and mechanical characteristic information relating to the mechanical properties of at least a part of the molded object. An information output device as described in any one of the items 128 to 133 of the appendix. [Note 135] The acquisition unit acquires measurement information related to the formation of the formed object obtained during the formation period of the formed object, which is formed by the solidification of the powdered metal material melted by irradiation with the energy beam. The output unit outputs the measurement information along with the energy information. An information output device as described in any one of the items 128 to 134 of the appendix.

[0233] At least some of the constituent elements of each embodiment described above can be appropriately combined with at least some other constituent elements of each embodiment described above. Some of the constituent elements of each embodiment described above may not be used. Furthermore, to the extent permitted by law, all of the published patents and U.S. patent disclosures cited in each embodiment described above shall be incorporated into the text.

[0234] The present invention is not limited to the embodiments described above, and can be modified as appropriate without contradicting the gist or idea of ​​the invention as can be read from the claims and specification as a whole. Information processing devices, measurement systems, molding systems, information processing methods, molding methods, and information output devices that involve such modifications are also included in the technical scope of the present invention. [Explanation of Symbols]

[0235] SYS Modeling System 1 Modeling equipment 2. Measuring device 21 Temperature measuring device 22 Shape measuring device 3. Information Processing Device 31 Arithmetic unit 311 Information Acquisition Department 312 Defect Prediction Unit 313 Defect Output Section 35 Output device 411, 412 Measurement Information 413 Modeling Information 414 Feature Information 415 Process Data 416 Location information 425 Defect Information 435 Energy Information 5 Model Generator PM Prediction Model

Claims

1. An acquisition unit that acquires measurement information related to the formation of a molded object obtained during the formation period of a molded object, which is formed by the solidification of a powdered metal material melted by irradiation with an energy beam, An information processing unit generates defect information relating to defects formed inside the molded object and energy information relating to the amount of energy transmitted from the energy beam to the metal material, using the measurement information acquired by the acquisition unit. Includes, The measurement information includes information regarding the temperature of at least a portion of the molten pool formed by the melting of the metal material due to irradiation with the energy beam. The defect information includes information regarding the density of at least a portion of the molded object. Information processing device.

2. The aforementioned defect includes a void formed inside the molded object. The defect information includes void information relating to the void. The information processing apparatus according to claim 1.

3. The void information includes information regarding the density. The information processing apparatus according to claim 2.

4. The defect information includes at least one of the following: information regarding the presence or absence of the defect and information regarding the characteristics of the defect. The information processing apparatus according to any one of claims 1 to 3.

5. The characteristics of the defect include at least one of the size of the defect, the location of the defect, and the shape of the defect. The information processing apparatus according to claim 4.

6. The system further comprises an output unit that outputs the aforementioned defect information. The information processing apparatus according to any one of claims 1 to 5.

7. The output unit includes a display unit that displays the defect information. The information processing apparatus according to claim 6.

8. The output unit includes a transmission unit that transmits the defect information to the molding apparatus that fabricates the molded object. The information processing apparatus according to claim 6 or 7.

9. The output unit outputs the measurement information along with the defect information. The information processing apparatus according to any one of claims 6 to 8.

10. The defect information includes first information relating to a defect formed on a part of the molded object, The measurement information includes second information obtained from measurements relating to the shape of the one part of the molded object, The output unit outputs the defect information and the measurement information in an output format in which the first information and the second information are associated. The information processing apparatus according to claim 9.

11. The output unit includes a display unit that displays the defect information and the measurement information in an output format in which the first information and the second information are associated. The information processing apparatus according to claim 10.

12. The output unit includes a transmission unit that transmits the defect information and the measurement information to the molding apparatus that fabricates the molded object. The information processing apparatus according to any one of claims 9 to 11.

13. The information processing unit further generates the energy information based on the temperature information, The output unit outputs the energy information along with the defect information. The information processing apparatus according to any one of claims 6 to 12.

14. The defect information includes first information relating to a defect formed on a part of the molded object, The energy information includes second information relating to the amount of energy transferred from the energy beam to the metal material in the formation of the one part of the molded object, The output unit outputs the defect information and the energy information in an output format in which the first information and the second information are associated. The information processing apparatus according to claim 13.

15. The output unit includes a display unit that displays the defect information and the energy information in an output format in which the first information and the second information are associated. The information processing apparatus according to claim 14.

16. The output unit includes a transmission unit that transmits the defect information and the energy information to the molding apparatus that fabricates the object. The information processing apparatus according to any one of claims 13 to 15.

17. The output unit outputs the defect information during the molding period in which the molded object is formed and / or after the molding of the molded object is completed. The information processing apparatus according to any one of claims 6 to 16.

18. The information processing unit generates the defect information using the temperature information and a learning model generated by machine learning to enable the generation of the defect information based on the temperature information. The information processing apparatus according to any one of claims 1 to 17.

19. The measurement information includes information about the shape of at least a part of the molded object obtained during the molding period. The information processing apparatus according to any one of claims 1 to 18.

20. The aforementioned molded object is formed by irradiating at least a portion of the layered material layer formed by the supplied metal material with the energy beam, causing at least a portion of the molten material layer to solidify. The measurement information includes information regarding the shape of at least a portion of the material layer obtained during the molding period. The information processing apparatus according to any one of claims 1 to 19.

21. The acquisition unit further acquires molding information relating to the molding of the molded object, The information processing unit generates the defect information using the measurement information and the molding information. The information processing apparatus according to any one of claims 1 to 20.

22. The molding information includes at least one of the following: device information relating to a molding apparatus for molding the molded object; material information relating to the metal material used for molding the molded object; and design information of the molded object. The information processing apparatus according to claim 21.

23. The device information includes at least one of the following: information regarding the molding conditions used by the molding apparatus to mold the object, and information regarding the operating status of the molding apparatus. The information processing apparatus according to claim 22.

24. The information processing unit uses the temperature information to generate at least one of the following: information regarding the state of the metallic structure constituting at least a part of the molded object, and information regarding the mechanical properties of at least a part of the molded object. The information processing apparatus according to any one of claims 1 to 23.

25. An information processing device according to any one of claims 1 to 24, Measuring device and Equipped with, The measurement information is obtained using the measurement device.

26. An information processing device according to any one of claims 1 to 24, A molding apparatus capable of creating objects by solidifying molten powdered metal material upon irradiation with an energy beam, Measuring device and Equipped with, The measurement information is a molding system obtained using the measurement device.

27. Based on the defect information, control the molding apparatus. The molding system according to claim 26.

28. If the defect information satisfies predetermined stop conditions, the molding apparatus stops the operation of molding the object. The molding system according to claim 26 or 27.

29. The molding apparatus molds the object based on the defect information. The molding system according to any one of claims 26 to 28.

30. The molding apparatus molds the object based on the molding conditions for which the molding apparatus molds the object. The molding apparatus molds the molded object or a new molded object based on the molding conditions that have been modified based on the defect information generated by the information processing device. The molding system according to any one of claims 26 to 29.

31. The molding apparatus modifies the molding conditions for molding the object using the defect information. A molding system according to any one of claims 26 to 30.

32. The molding apparatus uses the void information to fabricate the object based on the fabrication conditions modified by the user of the molding system. The molding system according to claim 30 or 31.

33. An acquisition unit that acquires measurement information related to the formation of a molded object obtained during the formation period of a molded object, which is formed by the solidification of a molding material melted by irradiation with an energy beam, An information processing unit that uses the measurement information acquired by the acquisition unit to generate characteristic information relating to the characteristics of the molded object and energy information relating to the amount of energy transmitted from the energy beam to the molding material. Equipped with, The characteristic information includes information relating to the density of at least a portion of the molded object.

34. An acquisition unit that acquires molding information related to the molding of an object formed by the solidification of a molding material melted by irradiation with an energy beam, An information processing unit generates characteristic information relating to the properties of the fabricated object and energy information relating to the amount of energy transmitted from the energy beam to the fabrication material, using the fabrication information acquired by the acquisition unit. Equipped with, The characteristic information includes information relating to the density of at least a portion of the molded object.

35. To acquire measurement information related to the formation of a molded object obtained during the formation period of a molded object, which is formed by the solidification of a molding material melted by irradiation with an energy beam, Using the acquired measurement information, characteristic information relating to the properties of the fabricated object and energy information relating to the amount of energy transmitted from the energy beam to the fabrication material are generated. Includes, The characteristic information includes an information processing method that includes information regarding the density of at least a portion of the molded object.

36. To obtain information about the shape of an object formed by the solidification of a molten material irradiated with an energy beam, Using the acquired molding information, characteristic information relating to the properties of the molded object and energy information relating to the amount of energy transmitted from the energy beam to the molding material are generated. Includes, The characteristic information includes an information processing method that includes information regarding the density of at least a portion of the molded object.

37. The process involves creating an object by solidifying a molten material that has been irradiated with an energy beam, During the molding period of the aforementioned molded object, measurements related to the molding of the aforementioned molded object shall be taken. To acquire characteristic information relating to the properties of the fabricated object generated using the measurement information obtained by the aforementioned measurement, and energy information relating to the amount of energy transmitted from the energy beam to the fabrication material. Includes, The characteristic information includes information regarding the density of at least a portion of the molded object.

38. The process involves creating an object by solidifying a molten material that has been irradiated with an energy beam, To obtain characteristic information relating to the properties of the fabricated object, generated using the fabrication information relating to the fabrication of the fabricated object, and energy information relating to the amount of energy transmitted from the energy beam to the fabrication material. Includes, The characteristic information includes information regarding the density of at least a portion of the molded object.

39. An acquisition unit that acquires measurement information related to the formation of a molded object obtained during the formation period of a molded object formed by the solidification of a powdered metal material melted by irradiation with an energy beam, characteristic information related to the properties of the molded object, and energy information related to the amount of energy transferred from the energy beam to the metal material. An output unit that outputs the characteristic information along with the measurement information. Equipped with, The characteristic information includes an information output device that includes information regarding the density of at least a portion of the molded object.

40. An acquisition unit that acquires energy information relating to the amount of energy transferred from the energy beam to the metal material in the process of creating an object by solidifying a powdered metal material that has been molten by irradiation with an energy beam, and characteristic information relating to the characteristics of the object. An output unit that outputs the characteristic information along with the energy information. Equipped with, The characteristic information includes an information output device that includes information regarding the density of at least a portion of the molded object.

Citation Information

Patent Citations

  • Machine Learning to Accelerate Alloy Design

    US20200257933A1

  • Apparatus for manufacturing three-dimensional shaped object, and method for manufacturing structure

    WO2016143137A1

  • Arithmetic device, detection system, modeling apparatus, arithmetic method, detection method, modeling method, arithmetic program, detection program, and modeling program

    WO2019239530A1