JP7550593B2 - Information processing system, method and program - Google Patents

Description

The present invention relates to an information processing system, method, and program.

It is becoming possible to create objects with complex and precise shapes by layering and modeling with a three-dimensional layering device (a so-called 3D printer) that uses metal layering technology. Metal layering technology has attracted a lot of attention, and although most metal materials can basically be melted with a beam, currently only a limited number of metal materials can be handled. To make metal materials usable, it is necessary to develop operating conditions for the beam that melts the metal (hereinafter also referred to as a material recipe). In addition, in actual modeling, it is necessary to simultaneously model support members (sometimes called supporting parts) to prevent deformation, and it is necessary to develop the modeling direction, support conditions, etc. (hereinafter also referred to as a modeling recipe) (see, for example, Patent Document 1).

JP 2017-211977 A

Figure 11 shows a general process from considering a new material to completing a modeled product. First, a prototype of the material is produced (step S910), and then the prototype material is evaluated (step S920). If sufficient material properties are obtained as a result of trial and error, development of a material recipe for additive manufacturing begins (step S930). Here again, trial and error is repeated, and in some cases the material components are readjusted, and a material recipe is developed that can be used as a modeling material.

Next, a modeling recipe for the actual object is developed (step S940), but while the melting conditions are optimized while manufacturing a small test piece during material recipe development, the actual object generally has a complex and larger shape, so there are many cases where the material recipe needs to be improved.
The key to developing a modeling recipe is determining the density of the support members. Deformation can be prevented by attaching strong supports at a high density, but since the support members will need to be removed later, it is necessary to design support members that can suppress deformation and are as easy to remove as possible. After the modeling recipe is completed, the actual modeling begins (step S960). Currently, however, each time deformation occurs, the modeling recipe is frequently improved to prevent deformation, leading to the development of the modeling recipe. In order to reduce trial and error in this modeling, deformation analysis (step S950) may be used.

When using new materials in a series of processes, trial and error is necessary for each process, and a great deal of effort and time is required from material development to the development of the molded product. As a result, although a wide variety of materials should be able to be used in additive manufacturing, the materials that can be used are limited. Also, while the selling point of additive manufacturing is that products can be manufactured in a short time without the need for mold manufacturing, etc., developing a molding recipe can be time-consuming, especially for products with complex shapes, and as a result, it can take a long time to manufacture the product.

The present invention was made in consideration of the above problems, and aims to provide an information processing system, method, and program that can reduce the work related to the molding conditions for each material.

The information processing system according to the first aspect of the present invention includes a modeling difficulty determination unit that determines the modeling difficulty by applying at least one physical property value of a material to a predetermined formula, and a parameter determination unit that determines the density of a support member that can be modeled without deformation in the target overhang by applying the determined modeling difficulty and at least one or more parameters related to the shape of the target overhang portion of the object to be modeled to a first correspondence relationship, or determines at least one or more parameters related to the shape of the overhang portion by applying the determined modeling difficulty and the density of the support member in the target overhang to a second correspondence relationship, the first correspondence relationship being a correspondence relationship between the modeling difficulty, at least one or more parameters related to the shape of the overhang portion, and a limit value of the density of the support member that can be modeled without deformation, and the second correspondence relationship being a correspondence relationship between the modeling difficulty, the density of the support member, and a limit value of at least one or more parameters related to the shape of the overhang portion that can be modeled without deformation.

With this configuration, at least one parameter related to the density of the support member that can be molded without deformation in the target overhang or the shape of the overhang portion is automatically determined, so the density of the support member or the shape of the overhang portion is determined without actually molding, reducing the work related to the molding conditions for each material.

The information processing system according to the second aspect of the present invention is the information processing system according to the first aspect, and includes a modeling recipe determination unit that determines a modeling recipe using at least one parameter related to the determined density of the support member or the determined shape of the overhang portion.

With this configuration, the modeling recipe is automatically determined, reducing the work involved in setting modeling conditions for each material.

The information processing system according to the third aspect of the present invention is the information processing system according to the first or second aspect, and includes a material recipe determination unit that determines a material recipe by performing molten pool analysis using at least one physical property value of the material, and a deformation analysis unit that uses the material recipe and the modeling recipe to analyze whether the object to be modeled will deform.

With this configuration, the object to be molded is automatically analyzed to see if it will deform, reducing the work involved in determining the molding conditions for each material.

The information processing system according to the fourth aspect of the present invention is an information processing system according to any one of the first to third aspects, in which the parameters relating to the shape of the overhang portion are the angle, length and/or thickness of the overhang.

This configuration allows the angle, length and/or thickness of the overhang to be determined automatically.

The information processing system according to the fifth aspect of the present invention includes a receiving unit that receives the limit support conditions under which the support members can be removed, and a material property value determining unit that determines the modeling difficulty by applying the received limit support conditions under which the support members can be removed to the correspondence between the limit support conditions and the modeling difficulty, and determines the value or range of at least one property value for the modeling difficulty by optimization analysis using the determined modeling difficulty and the property values of the target material.

With this configuration, the values or ranges of physical property values can be determined automatically, making it possible to clarify the range of physical property values of a material without actually molding it, and significantly reducing the number of trial-and-error steps required to select new materials and mold them.

The information processing system according to the sixth aspect of the present invention is the information processing system according to the fifth aspect, in which the reception unit receives an upper limit and/or a lower limit of a first physical property value determined from a design requirement, and the material physical property value determination unit determines a value or range of a second physical property value such that the first physical property value falls within the range of the upper limit and/or the lower limit of the first physical property value received by the reception unit, using the relationship between the first physical property value and the second physical property value under conditions where the modeling difficulty level is the determined modeling difficulty level.

With this configuration, even if there are physical property values required by design requirements, such as material strength, it is possible to take these conditions into account and determine the values or ranges of physical property values that satisfy the design requirements and can be molded.

The information processing system according to the seventh aspect of the present invention is the information processing system according to the fifth or sixth aspect, further comprising a material composition determination unit that determines the material composition using the values or ranges of each physical property value determined by the material property value determination unit.

With this configuration, by combining the determined material property values with materials informatics techniques, it is possible to automatically propose material compositions for additive manufacturing.

The information processing system according to the eighth aspect of the present invention is the information processing system according to the seventh aspect, in which the material composition determination unit inputs the physical property values determined by the material property value determination unit into a trained machine learning model that inputs the material property values and outputs the material composition, thereby outputting the material composition.

With this configuration, the accuracy of predicting optimal material composition can be improved by using a trained machine learning model.

The method according to the ninth aspect of the present invention includes a modeling difficulty determination step of determining the modeling difficulty by applying at least one physical property value of the material to a predetermined formula, and a parameter determination step of determining the density of a support member that can be modeled without deformation in the target overhang by applying the determined modeling difficulty and at least one or more parameters related to the shape of the target overhang portion of the object to be modeled to a first correspondence relationship, or determining at least one or more parameters related to the shape of the overhang portion by applying the determined modeling difficulty and the density of the support member in the target overhang to a second correspondence relationship, the first correspondence relationship being a correspondence relationship between the modeling difficulty, at least one or more parameters related to the shape of the overhang portion, and a limit value of the density of the support member that can be modeled without deformation, and the second correspondence relationship being a correspondence relationship between the modeling difficulty, the density of the support member, and a limit value of at least one or more parameters related to the shape of the overhang portion that can be modeled without deformation.

With this configuration, at least one parameter related to the density of the support member that can be molded without deformation in the target overhang or the shape of the overhang portion is automatically determined, so the density of the support member or the shape of the overhang portion is determined without actually molding, reducing the work related to the molding conditions for each material.

The method according to the tenth aspect of the present invention includes a receiving step for receiving the limit support conditions under which the support members can be removed, and a material property value determination step for determining the modeling difficulty by applying the received limit support conditions under which the support members can be removed to the correspondence between the limit support conditions and the modeling difficulty, and determining the value or range of at least one property value for the modeling difficulty through optimization analysis using the determined modeling difficulty and the physical property values of the target material.

With this configuration, the values or ranges of physical property values can be determined automatically, making it possible to clarify the range of physical property values of a material without actually molding it, and significantly reducing the number of trial-and-error steps required to select new materials and mold them.

The program according to the eleventh aspect of the present invention is a program for causing a computer to execute a modeling difficulty determination procedure for determining the modeling difficulty by applying at least one physical property value of a material to a predetermined formula, and a parameter determination procedure for determining the density of a support member that can be modeled without deformation in a target overhang by applying the determined modeling difficulty and at least one or more parameters related to the shape of a target overhang portion of an object to be modeled to a first correspondence relationship, or determining at least one or more parameters related to the shape of the overhang portion by applying the determined modeling difficulty and the density of the support member in the target overhang to a second correspondence relationship, the first correspondence relationship being a correspondence relationship between the modeling difficulty, at least one or more parameters related to the shape of the overhang portion, and a limit value of the density of the support member that can be modeled without deformation, and the second correspondence relationship being a correspondence relationship between the modeling difficulty, the density of the support member, and a limit value of at least one or more parameters related to the shape of the overhang portion that can be modeled without deformation.

With this configuration, at least one parameter related to the density of the support member that can be molded without deformation in the target overhang or the shape of the overhang portion is automatically determined, so the density of the support member or the shape of the overhang portion is determined without actually molding, reducing the work related to the molding conditions for each material.

The program according to the twelfth aspect of the present invention is a program for causing a computer to execute a receiving procedure for receiving limit support conditions under which a support member can be removed, and a material property value determination procedure for determining the modeling difficulty by applying the received limit support conditions under which the support member can be removed to the correspondence between the limit support conditions and the modeling difficulty, and determining the value or range of at least one property value for the modeling difficulty through optimization analysis using the determined modeling difficulty and the physical property values of the target material.

With this configuration, the values or ranges of physical property values can be determined automatically, making it possible to clarify the range of physical property values of a material without actually molding it, and significantly reducing the number of trial-and-error steps required to select new materials and mold them.

According to one aspect of the present invention, at least one parameter related to the density of the support member that can be molded without deformation in the target overhang or the shape of the overhang portion is automatically determined. Since the density of the support member or the shape of the overhang portion is determined without actually molding, the work related to the molding conditions for each material can be reduced.
According to another aspect of the present invention, the value or range of physical property values can be automatically determined, making it possible to clarify the range of physical property values of a material without actually molding it, and significantly reducing the number of times that new materials need to be selected and the trial and error required to mold them.

1 is a schematic configuration diagram of an information processing system according to a first embodiment. FIG. 4 is a schematic vertical cross-sectional view illustrating an overhang portion of the object. 13 is a graph showing an example of a relationship between parameters related to the shape of an overhang portion and modeling difficulty; 1 is an example of a process according to the present embodiment. FIG. 13 is a schematic diagram of a process using deformation analysis according to a comparative example. FIG. 2 is a schematic diagram of the process of the present embodiment. FIG. 11 is a schematic diagram for explaining a process for determining a range of a second physical property value of a material. 1 is a flowchart showing an example of a process flow for determining a material composition by using a data scientific method. FIG. 2 is a schematic diagram for explaining a method for determining a material composition. FIG. 11 is a schematic configuration diagram of an information processing system according to a second embodiment. FIG. 11 is a schematic configuration diagram of a computer system according to a second embodiment. FIG. 1 is a diagram showing a general process from considering a new material to completing a molded product. FIG. 1 illustrates an example of an optimization analysis. FIG. 13 is a diagram illustrating an example of optimization analysis. FIG. 1 is a diagram showing the results of the optimization analysis expressed as a response surface.

Each embodiment will be described below with reference to the drawings. However, more detailed explanation than necessary may be omitted. For example, detailed explanation of already well-known matters or duplicate explanation of substantially identical configurations may be omitted. This is to avoid the following explanation becoming unnecessarily redundant and to make it easier for those skilled in the art to understand.

As mentioned above, although a wide range of materials can in principle be used in metal additive manufacturing, only a limited number of materials have been put to practical use. One reason for this is that the degree of difficulty of modeling varies for each material, and modeling conditions must be established for each material. However, in the consideration required for practical use, trial and error must be repeated as described above at each stage, making the work complicated. In response to this, this embodiment provides a method for reducing the number of trial and error steps required by indexing the modeling difficulty of the material, and for easily establishing a modeling process for a new material.

<Definition of terms>
A material recipe is a beam condition during printing that allows the material to be melted and molded properly, and there is an appropriate range for each material. Beam output, scan speed, beam spot diameter, etc. It is like welding conditions. Material recipes differ depending on the type of equipment. In general, test pieces are printed under various beam conditions, and the printed objects are evaluated to determine the appropriate beam condition range.
A modeling recipe is a modeling method that is determined for each object. It determines the placement of the object and the support conditions, and is a know-how unique to additive manufacturing. Support is required in areas with a high risk of deformation, but if the support is made too strong or the blade cannot reach the area, it cannot be removed. It is necessary to design a support that does not deform and can be removed.

Each embodiment also solves the following specific problems.
<Challenge 1: Reducing trial and error in developing modeling recipes>
Since the degree of difficulty of modeling differs depending on the material, for example, even for products with the same shape, the modeling recipe must be remade when changing the material. This requires trial and error in developing the modeling recipe, but the goal is to reduce this trial and error.

<Challenge 2: Reducing the repeated trial and error of changing materials>
In the additive manufacturing process, materials such as Ti alloys, which are difficult to process using conventional manufacturing methods such as rolling and machining, are frequently used. One of the reasons for their use is that Ti alloys are easy to mold using additive manufacturing. Thus, materials suitable for molding differ between conventional manufacturing methods and additive manufacturing. Currently, it is necessary to create raw materials and actually mold them to see whether they are suitable for molding. Generally, when using different raw materials in metal additive manufacturing machines, contamination occurs, so the equipment needs to be cleaned every time the raw materials are replaced. Therefore, it is extremely difficult to repeat trial and error in molding with different materials. Therefore, a method for determining whether molding is possible without actually performing a molding trial is desired.

<Challenge 3: Development of materials suitable for modeling>
Conventional plate materials, forged materials, cast materials, etc., are required to have various characteristics such as rolling, forging, and casting properties, as well as machinability, bending, and welding properties, and there are limitations on the material composition that can be used. On the other hand, a major feature of additive manufacturing is that it is possible to create objects that are close to the final shape of the product from powder raw materials. In other words, the material to be used can be selected without being restricted by processability. In addition, the composition of the raw metal powder is easy to adjust, and the alloy composition can be freely adjusted. However, as mentioned above, it takes a lot of effort to try new materials, which is a major hindrance to the development of new materials for additive manufacturing and is a major factor limiting the potential of additive manufacturing.

<Special terms related to this embodiment>
Modeling difficulty is an index that expresses the difficulty of modeling a material (ease of deformation during modeling). The difficulty of modeling differs even with the same material when the modeling process (laser or electron beam) is different. The main cause of deformation during modeling is shrinkage during the melting, solidification, and cooling processes. To prevent deformation at this time, support members are installed in areas with a high risk of deformation. If the deformation is large, the support members will tear off, leading to major deformation of the entire model, and ultimately resulting in modeling failure. For example, materials with a high linear expansion coefficient are prone to deformation and are difficult to model.

The degree of deformation during modeling is determined by the physical properties of the material (hereafter referred to as material physical properties), so theoretically, if the stress generated by deformation is less than the tensile strength of the support, the support will not tear and a sound modeling is possible. The degree of deformation can be predicted from the physical properties and turned into an index, which is the difficulty of modeling.

Support conditions are conditions related to the support members, such as the necessity, position, width, thickness and/or density of the support members.

The limit support conditions are the conditions for supports that can be printed without deformation, and are the limit value (also called limit density) or range of the density of the support member that can be printed without deformation for the shape of the object, such as the overhang angle, length, and thickness, or the object conditions, such as the overhang angle that can be printed without deformation when the density of the support member is constant. The limit support conditions differ depending on the material and the model of the printing device. Specifically, the object conditions are, for example, the limit value or range of at least one or more parameters related to the shape of the overhang part that can be printed without deformation.

<Basic Concept>
Using multiple types of materials, actually perform modeling using commonly used support members, and obtain in advance the limit density of the support members that can be modeled for the modeling shape. Alternatively, obtain in advance the modeling conditions, such as the overhang angle that can be modeled without deformation while keeping the density of the support members constant. Then, obtain in advance the relationship between the limit density of the support members and the modeling difficulty, or the relationship between the modeling conditions and the modeling difficulty. From this, if the modeling difficulty is known from the obtained relationship, it becomes possible to obtain the limit density of the support members or the modeling conditions.

1 is a schematic diagram of an information processing system according to a first embodiment. The information processing system S1 according to the first embodiment includes, as an example, an information processing device 1 and a display 17 connected to the information processing device 1. The information processing device 1 includes, for example, an input interface 11, a communication module 12, a storage 13, a memory 14, an output interface 15, and a processor 16.
The input interface 11 receives a user's operation and outputs an input signal corresponding to the received operation to the processor 16. In this embodiment, the input interface 11 is, for example, a keyboard and a mouse.
The communications module 12 is connected to a communications network for communicating with other computers connected to the communications network, the communication being either wired or wireless.

The storage 13 stores programs and various data to be read and executed by the processor 16 .
The memory 14 temporarily stores data and programs. The memory 14 is a volatile memory, such as a random access memory (RAM).
The output interface 15 is connected to an external display 17 , and outputs a video signal to the display 17 in accordance with a command from the processor 16 .

The processor 16 loads the program according to the first embodiment from the storage 13 into the memory 14 and executes a series of commands contained in the program, thereby functioning as a modeling difficulty determination unit 160, a material recipe determination unit 161, a parameter determination unit 162, a modeling recipe determination unit 163, a deformation analysis unit 164, a reception unit 165, a material property value determination unit 166, and a material composition determination unit 167. The material composition determination unit 167 has an AI unit 168. The processing of each unit will be described in detail later.

Example 1
In the first embodiment, a means for solving the above problems 1 and 2 is provided.

<Calculation of modeling difficulty>
Material factors that affect molding characteristics include various physical properties such as thermal expansion coefficient, thermal conductivity, and yield strength. In other words, molding difficulty is the amount of deformation caused by molding. Therefore, by evaluating the molding of a simple model and the amount of deformation, and by evaluating and indexing the molding difficulty for each material, it becomes possible to estimate the molding difficulty of a new material.

Therefore, a numerical analysis is performed by varying the material factors in various ways to determine the amount of deformation under each condition, and the correlation between the physical properties and the amount of deformation is determined. Then, by performing a multiple regression analysis of the amount of deformation and the material physical properties, it is possible to index the amount of deformation as the difficulty of modeling, for example in the form of the following formula 1. The modeling characteristics of this material differ depending on whether a laser or an electron beam is used as the heat source, so they must be determined for each device.

Modeling difficulty = F (material properties, linear expansion coefficient, Young's modulus, yield strength, specific heat, density, thermal conductivity)... (Formula 1)

Here, F is a function. Figure 2 is a schematic vertical cross-sectional view explaining an overhang portion of a molded object. Figure 2 shows the overhang angle θ, which is the angle between the overhang portion and another molded object, the overhang length L, which is the length of the overhang portion, and the thickness D of the overhang portion P for the overhang portion P of the molded object Z.

Figure 3 shows three graphs showing an example of the relationship between parameters related to the shape of the overhang portion and the difficulty of printing. Line W1 shows the relationship between the overhang angle θ and the difficulty of printing, with support members not required when the line is above W1 and support members required when the line is below W1. Similarly, line W2 shows the relationship between the overhang length L and the difficulty of printing, with support members not required when the line is above W2 and support members required when the line is below W2. Similarly, line W3 shows the relationship between the thickness D of the overhang portion P and the difficulty of printing, with support members required when the line is above W3 and support members not required when the line is below W3.

For example, using a modelable material, the limit values of at least one or more parameters related to the shape of the overhang portion that can be modeled without deformation with a predetermined standard material are found. Specifically, for example, the overhang portion is where support members are required, and the relationship is found such that for material A, which has low modeling difficulty, the overhang angle θ is up to 25°, and for material B, which has high modeling difficulty, the overhang angle θ is up to 40°. The parts where support members are required differ depending on the modeling device, so they need to be found for each device. Using this relationship between modeling difficulty and overhang angle, it is possible to immediately calculate the overhang angle when new material C is used. This makes it possible to significantly reduce the trial and error required.

4 shows an example of the process of this embodiment. By evaluating the modeling difficulty after prototyping the material (step S10) (step S30), it becomes possible to determine whether modeling is possible before starting to develop a difficult material recipe, and it becomes possible to significantly reduce trial and error involved in the material recipe development (step S40).
As a specific example of the determination of whether or not a material can be molded, the degree of difficulty of molding a prototype alloy is calculated, and if the degree of difficulty is lower than that of a material with a proven track record, it can be determined that the material can be molded. In the case of a material with no proven track record, the amount of deformation required, for example, by deformation analysis can be determined to determine whether or not the material can be molded.
However, currently, research is being conducted on numerical analysis technology for the molten state, and it is possible that the difficulty of developing material recipes will be significantly reduced in the future. In that case, it is not necessary to follow this procedure, and the process may be changed as appropriate.

Next, the modeling difficulty of the new material is applied to the relationship between modeling difficulty and support conditions determined for a predetermined standard material, and the support conditions required for the new material are determined. Because the necessary support materials can be estimated before modeling, the need for repeated trial and error in modeling is significantly reduced.

By reproducing each modeling process on an information processing device, it becomes possible to consider whether modeling is possible at a desk, making it possible to significantly reduce modeling trials. Commercially available software has been developed for deformation analysis, and software such as Magics is widely used for recipe creation. Molten pool analysis (step S401), also known as material recipe development (step S40), is also currently being widely developed, and its practical application is being considered.

The key points of this embodiment are the modeling difficulty calculation and support analysis. The material property values are input to find the modeling difficulty (step S30), and the modeling difficulty is used to determine the above-mentioned limit support conditions (step S50). At that time, a modeling recipe is developed based on this (step S60). Finally, a deformation analysis is performed (step S701) to confirm that there are no problems with the modeling recipe. After that, modeling is performed (step S70), and a modeled product is obtained.

In the information processing device 1, the following process is executed.
(Step S301) The modeling difficulty determination unit 160 determines the modeling difficulty by applying at least one physical property value of a material to a predetermined formula.

(Step S401) The material recipe determination unit 161 determines the material recipe by performing molten pool analysis using at least one physical property value of the material.

(Step S501) The parameter determination unit 162 determines the density of the support member that can be formed without deformation in the target overhang by applying the modeling difficulty determined in step S301 and at least one parameter related to the shape of the target overhang portion of the object to be modeled to a first correspondence relationship. Alternatively, the parameter determination unit 162 determines at least one parameter related to the shape of the overhang portion by applying the modeling difficulty determined in step S301 and the density of the support member in the target overhang to a second correspondence relationship.

Here, the first correspondence relationship is a correspondence relationship between the modeling difficulty, at least one or more parameters related to the shape of the overhanging portion, and a limit value of the density of the supporting member at which the shape can be modeled without deformation, and the second correspondence relationship is a correspondence relationship between the modeling difficulty, the density of the supporting member, and a limit value of at least one or more parameters related to the shape of the overhanging portion at which the shape can be modeled without deformation.
Moreover, the parameters relating to the shape of the overhang portion are the angle, length and/or thickness of the overhang.

(Step S601) The modeling recipe determination unit 163 determines a modeling recipe using at least one parameter related to the density of the support member determined in step S501 or the shape of the overhanging portion determined in step S501. This allows the modeling recipe to be determined automatically, reducing the work related to the modeling conditions for each material.

(Step S701) The deformation analysis unit 164 uses the material recipe and the modeling recipe to analyze whether the object to be modeled will deform. If the analysis shows that the object will not deform, the modeling shape is determined. This automatically analyzes whether the object to be modeled will deform, reducing the work involved in determining the modeling conditions for each material.

5A and 5B, the difference between the process using deformation analysis according to the comparative example and the process of this embodiment will be described. Fig. 5A is a schematic diagram of the process using deformation analysis according to the comparative example. Fig. 5B is a schematic diagram of the process of this embodiment.
5A, in the method of using the deformation analysis in the comparative example, the support shape (molding recipe) and the material property values are input data, and the shape after molding is calculated. If deformation occurs, trial and error is repeated to correct the support shape (molding recipe).

In contrast, in this embodiment, as shown in FIG. 5B, the necessary support conditions can be determined based on the modeling difficulty, so deformation analysis can be used to check the modeling recipe, significantly reducing the trial and error involved in deformation analysis.

The key point of this embodiment is that it is possible to propose the support conditions necessary to prevent deformation from the material property values. Since it is difficult to directly grasp the relationship with support members from the multiple material property values (multiple factors) of a single material, the inventor came up with the idea of first converting the multiple material property values into a single factor, the modeling difficulty, and then grasping the relationship with support members.

Example 2: Proposal of optimal material
Example 2 solves the above problem 3. In Example 2, material property values are proposed using the concept of this embodiment. Since a tool is generally used to remove a support member, if a support is attached to an area that the tool cannot reach, such as inside a flow path, it cannot be removed, and as a result, modeling cannot be completed. From such a structural viewpoint, a removable support range is required.

Figure 6 is a schematic diagram for explaining the process of determining the range of the second physical property value of a material. In Figure 6, a curve W11 is shown in a graph in which the vertical axis represents the limit support condition (e.g., the limit value of the overhang angle) and the horizontal axis represents the difficulty of modeling, indicating the correspondence between the limit support condition and the difficulty of modeling. The limit support condition here is, for example, the limit value of at least one or more parameters (e.g., the limit value of the overhang angle) related to the shape of the overhang portion that can be modeled without deformation when the density of the support member is kept constant. The limit support condition under which the support member can be removed is determined manually depending on the shape of the target object.

Here, the receiving unit 165 receives the limit support condition under which the support member can be removed. Then, the material property value determining unit 166 determines the modeling difficulty level at which modeling is possible by applying the limit support condition under which the received support member can be removed to the correspondence relationship between the limit support condition (e.g., overhang angle) and the modeling difficulty level.
The material physical property value determination unit 166 determines the value or range of each physical property value for the modeling difficulty level by optimization analysis using the determined modeling difficulty level and the physical property values of the target material. Specifically, for example, the material physical property value determination unit 166 determines a Pareto solution (non-inferior solution).
FIG. 12 is a diagram showing an example of optimization analysis. For convenience of explanation, two physical property values are used in the explanation, but in reality, analysis may be performed using multidimensional physical property values. The physical property values A and B are arbitrarily set, and the modeling difficulty is calculated. Then, in the set physical property value range, the relationship between the physical property value and the modeling difficulty (the relationship between each curve in FIG. 12) is obtained, and the range of each physical property value, for example, X1, X2, etc., is calculated. A modeling difficulty value, for example, X3, obtained by applying the limit support condition to the relationship between the limit support condition and the modeling difficulty (see FIG. 6) is selected by the material physical property value determination unit 166, and this modeling difficulty value (for example, X3) is applied to the relationship between the obtained physical property value and the modeling difficulty to obtain the possible range of each physical property value by the material physical property value determination unit 166.
The range of physical property values set in the optimization analysis may be set arbitrarily, or the values may be determined using experimental design, etc. Furthermore, a method such as finding a response surface using commercially available optimization analysis software such as modeFRONTIER may also be used.
This makes it possible to automatically determine the value or range of each physical property, making it possible to clarify the range of physical property values of a material without actually molding it, significantly reducing the number of times it is necessary to select new materials and perform trial and error in molding them.

6 shows a curve W12 showing the relationship between the modeling difficulty and the first physical property value, a curve W13 showing the relationship between the modeling difficulty and the second physical property value, and a curve W14 showing the relationship between the modeling difficulty and the third physical property value. The material physical property value determination unit 166 may determine the first physical property value, the second physical property value, and the third physical property value by applying the determined modeling difficulty to the relationship between the modeling difficulty and the first physical property value, the relationship between the modeling difficulty and the second physical property value, and the relationship between the modeling difficulty and the third physical property value.

For example, the relationship between the first physical property value and the second physical property value under the condition of the determined modeling difficulty level is represented by the curve W15 in FIG. 6. There are physical property values required from the design requirements, such as material strength. For example, if an upper limit of the first physical property value is given from the design requirements, the range of the second physical property value will be B1 or more. In this case, the reception unit 165 receives the upper limit and/or lower limit of the first physical property value determined from the design requirements. The material physical property value determination unit 166 may then use the relationship between the first physical property value and the second physical property value under the condition of the determined modeling difficulty level to determine the value or range of the second physical property value in which the first physical property value falls within the range of the upper limit and/or lower limit of the first physical property value received by the reception unit 165. As a result, even if there are physical property values required from the design requirements, such as material strength, it is possible to consider these conditions and determine the value or range of the physical property value that satisfies the design requirements and is modelable.

As a result, it is possible to clarify the range of material property values without actually creating a model, significantly reducing the number of trial-and-error steps required to select new materials and create them.

Example 3: Combination with data science
Next, in Example 3, a means for solving the above-mentioned problem 3 is provided. Currently, research and development on materials informatics is a major topic in the field of materials. In Example 3, efficient material development is made possible by incorporating data science into material development. A common calculation method is to input the material composition and calculate the material properties, but in this example, the material composition is calculated from the material properties by using a data scientific method.

Figure 7 is a flowchart showing an example of the process flow for determining material composition using data scientific methods. Figure 7 shows a process that incorporates this method. By combining the process of calculating material property values from material composition and combining the whole with data scientific methods, it becomes possible to propose the optimal material composition for molding, significantly shortening the time required to consider the process from material development to molding, and as a result, facilitating the development of new materials for additive manufacturing.

(Step S310) As described above, the material physical property value determination unit 166 determines the modeling difficulty level at which the accepted support member can be removed by applying the limit support conditions to the correspondence relationship between the limit support conditions and the modeling difficulty level, and determines the value or range of at least one physical property value for the modeling difficulty level through optimization analysis using both the determined modeling difficulty level and the physical property values of the target material.
13 is a diagram showing an example of optimization analysis. This is an example in which a data set consisting of six material property values, namely, linear expansion coefficient, Young's modulus, yield strength, density, specific heat, and thermal conductivity, is created, and the material difficulty level is calculated for each data set. The material property value determination unit 166 can determine an adoptable data set from the data sets showing a material difficulty level lower than the upper limit of the required material difficulty level.
Fig. 14 is a diagram showing the above results expressed by a response surface. For convenience of explanation, two physical property values are used for the explanation. It represents the degree of influence of physical property value 1 and physical property value 2 on the modeling difficulty, and when the required material difficulty is at a specific level and is in the region shown in black in Fig. 14, the material physical property value determination unit 166 can determine the range of physical property values 1 and 2 from these regions.

(Step S320) Next, the material composition determination unit 167 determines the material composition using the values or ranges of the physical property values determined by the material physical property value determination unit 166.

Figure 8 is a schematic diagram for explaining a method for determining a material composition. The AI unit 168 of the material composition determination unit 167 executes machine learning using a learning data set in which material property values are input in advance and material composition is output, and stores information about the learned machine learning model obtained as a result of the execution in the storage 13. This allows the AI unit 168 to read and execute the learned machine learning model. Here, the material property values may be experimental data and/or calculated physical properties. In this case, the AI unit 168 of the material composition determination unit 167 inputs the property values determined by the material property value determination unit 166 into a learned machine learning model that inputs material property values and outputs material composition, and outputs the material composition. This allows the use of a learned machine learning model to improve the prediction accuracy of the optimal material composition.

By using the method of the second embodiment, the required material property values are determined by the material property value determination unit 166. By combining the determined material property values with a materials informatics method, it is possible to automatically propose a material composition for additive manufacturing.
In addition, the input of the learning data set may further include data on the usage environment and/or corrosion resistance. By incorporating the usage environment and corrosion resistance data, it becomes possible to propose a material composition that also satisfies the corrosion resistance characteristics.

As described above, the information processing system S1 according to this embodiment includes a modeling difficulty determination unit 160 that determines the modeling difficulty by applying at least one physical property value of the material to a predetermined formula, and a parameter determination unit 162 that determines the density of a support member that can be modeled without deformation in the target overhang by applying the determined modeling difficulty and at least one or more parameters related to the shape of the target overhang portion of the object to be modeled to a first correspondence relationship, or determines at least one or more parameters related to the shape of the overhang portion by applying the determined modeling difficulty and the density of the support member in the target overhang to a second correspondence relationship. The first correspondence relationship is a correspondence relationship between the modeling difficulty, at least one or more parameters related to the shape of the overhang portion, and the limit value of the density of the support member that can be modeled without deformation, and the second correspondence relationship is a correspondence relationship between the modeling difficulty, the density of the support member, and the limit value of at least one or more parameters related to the shape of the overhang portion that can be modeled without deformation.

With this configuration, the density of the support member that can be molded without deforming the target overhang, or at least one parameter related to the shape of the overhang portion, is automatically determined, reducing the work related to the molding conditions for each material.

Second Embodiment
Next, a second embodiment will be described. In the first embodiment, the processing is executed by an information processing device 1 used by a user, but in the second embodiment, the processing is executed by a computer system to which a terminal device used by a user is connected via a communication network.
9 is a schematic diagram of an information processing system according to a second embodiment. As shown in FIG. 9, the information processing system S2 includes, as an example, terminal devices 3-1, ..., 3-N (N is a natural number) and a computer system 2 connected to each of the terminal devices 3-1 to 3-N via a communication network NW. The computer system 2 executes processing in response to requests from the terminal devices 3-1, ..., 3-N. Here, the computer system 2 is described as being one server as an example, but is not limited to this and may be composed of multiple computers like a cloud service.

The terminal devices 3-1 to 3-N are terminal devices used by different users, and may be, for example, mobile phones such as multi-function mobile phones (so-called smartphones), mobile devices such as tablets, e-book readers, and notebook computers, or desktop computers. The terminal devices 3-1 to 3-N may display information sent from the computer system 2 using, for example, a web browser, or may display information sent from the computer system 2 in an application installed on the terminal devices 3-1 to 3-N.

Fig. 10 is a schematic configuration diagram of a computer system according to the second embodiment. As shown in Fig. 10, the computer system 2 includes, for example, an input interface 21, a communication module 22, a storage 23, a memory 24, and a processor 25.
The input interface 11 accepts operations by an administrator of the computer system 2 , and outputs an input signal corresponding to the accepted operation to the processor 25 .
The communication module 22 is connected to the communication network NW and communicates with the terminal devices 3-1 to 3-N connected to the communication network NW. This communication may be wired or wireless.

The storage 23 stores programs and various data to be read and executed by the processor 16 .
The memory 24 temporarily stores data and programs. The memory 24 is a volatile memory, such as a RAM (Random Access Memory).

The processor 25 loads the program according to the first embodiment from the storage 23 into the memory 24 and executes a series of instructions contained in the program, thereby functioning as a modeling difficulty determination unit 160, a material recipe determination unit 161, a parameter determination unit 162, a modeling recipe determination unit 163, a deformation analysis unit 164, a reception unit 165, a material property value determination unit 166, and a material composition determination unit 167. The material composition determination unit 167 has an AI unit 168. These functions are the same as those in the first embodiment, so their description will be omitted.

In addition, some of the functions of the computer system 2 may be realized by the terminal devices 3-1 to 3-N.

Note that at least a part of the information processing device 1 described in the above embodiment may be configured with hardware or software. If configured with software, a program that realizes at least a part of the functions of the information processing device 1 may be stored on a recording medium such as a flexible disk or CD-ROM, and may be read and executed by a computer. The recording medium is not limited to removable recording media such as magnetic disks and optical disks, but may also be fixed recording media such as a hard disk device or memory.

The program that realizes at least a part of the functions of the information processing device 1 may be distributed via a communication line (including wireless communication) such as the Internet. Furthermore, the program may be encrypted, modulated, or compressed and distributed via a wired or wireless line such as the Internet, or stored on a recording medium.

Furthermore, the information processing device 1 may be operated by one or more information processing devices. When multiple information processing devices are used, one of the information processing devices may be a computer, and the computer may execute a predetermined program to realize the functions as at least one of the means of the information processing device 1.

In addition, in the method invention, all processes (steps) may be realized by automatic control using a computer. Furthermore, each process may be performed by a computer, while progress control between processes may be performed manually. Furthermore, at least some of the processes may be performed manually.

As mentioned above, the present invention is not limited to the above-mentioned embodiment as it is, and in the implementation stage, the components can be modified and embodied without departing from the gist of the invention. Furthermore, various inventions can be formed by appropriately combining the multiple components disclosed in the above-mentioned embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, components from different embodiments may be appropriately combined.

REFERENCE SIGNS LIST 1 Information processing device 11 Input interface 12 Communication module 13 Storage 14 Memory 15 Output interface 16 Processor 160 Modeling difficulty determination unit 161 Material recipe determination unit 162 Parameter determination unit 163 Modeling recipe determination unit 164 Deformation analysis unit 165 Reception unit 166 Material property value determination unit 167 Material composition determination unit 168 AI unit 17 Display 2 Computer system 21 Input interface 22 Communication module 23 Storage 24 Memory 25 Processor 3 Terminal device S1, S2 Information processing system

Claims (6)

a modeling difficulty determination unit that determines modeling difficulty by applying at least one physical property value of the material to a predetermined formula;
a parameter determination unit that determines a density of a supporting member that can be formed in the target overhang without deformation by applying the determined modeling difficulty level and at least one or more parameters related to a shape of the target overhang portion of the object to be modeled to a first correspondence relationship, or determines at least one or more parameters related to the shape of the overhang portion by applying the determined modeling difficulty level and a density of a supporting member in the target overhang to a second correspondence relationship;
Equipped with
the first correspondence relationship is a correspondence relationship between a modeling difficulty level, at least one parameter related to a shape of the overhang portion, and a limit value of a density of the supporting member at which the modeling can be performed without deformation,
The second correspondence relationship is a correspondence relationship between the modeling difficulty, the density of the supporting members, and a limit value of at least one parameter related to a shape of the overhang portion that can be modeled without deformation. The information processing system according to claim 1 , further comprising a modeling recipe determination unit that determines a modeling recipe using at least one parameter related to the determined density of the support member or the determined shape of the overhang portion. a material recipe determination unit that determines a material recipe by analyzing a molten pool using at least one physical property value of a material;
a deformation analysis unit that analyzes whether or not the object to be molded is deformed by using the material recipe and the molding recipe;
The information processing system according to claim 2 . The information processing system according to claim 1 , wherein the parameter relating to the shape of the overhang portion is an angle, a length and/or a thickness of the overhang. A modeling difficulty determination step for determining modeling difficulty by applying at least one physical property value of the material to a predetermined formula;
a parameter determination step of determining a density of a supporting member that can be formed in the target overhang without deformation by applying the determined modeling difficulty level and at least one or more parameters related to a shape of the target overhang portion of the object to be modeled to a first correspondence relationship, or determining at least one or more parameters related to the shape of the overhang portion by applying the determined modeling difficulty level and a density of a supporting member in the target overhang to a second correspondence relationship;
A method comprising:
the first correspondence relationship is a correspondence relationship between a modeling difficulty level, at least one parameter related to a shape of the overhang portion, and a limit value of a density of the supporting member at which the modeling can be performed without deformation,
The second correspondence relationship is a correspondence relationship between the difficulty of modeling, the density of the support member, and the limit value of at least one parameter related to the shape of the overhang portion that can be modeled without deformation. On the computer,
A modeling difficulty determination step for determining modeling difficulty by applying at least one physical property value of the material to a predetermined formula;
a parameter determination step of determining a density of a supporting member that can be formed in the target overhang without deformation by applying the determined modeling difficulty level and at least one or more parameters related to a shape of the target overhang portion of the object to be modeled to a first correspondence relationship, or determining at least one or more parameters related to the shape of the overhang portion by applying the determined modeling difficulty level and a density of a supporting member in the target overhang to a second correspondence relationship;
A program for executing
the first correspondence relationship is a correspondence relationship between a modeling difficulty level, at least one parameter related to a shape of the overhang portion, and a limit value of a density of the supporting member at which the modeling can be performed without deformation,
The second correspondence relationship is a correspondence relationship between the modeling difficulty, the density of the supporting member, and the limit value of at least one parameter related to the shape of the overhang portion that can be modeled without deformation.