Estimation method, estimation device, estimation system, and computer program

Abstract

To enable estimation of properties of a metallic material containing predetermined ingredients.SOLUTION: An estimation method is provided, comprising an estimation step of using trained model information obtained by performing learning processing in advance using teacher data including condition information showing the conditions regarding metallic materials and information showing mechanical properties of metallic materials meeting the conditions indicated by the condition information to estimate mechanical properties of an estimation target metallic material meeting the condition information.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an estimation method, an estimation device, an estimation system, and a computer program. [Background technology]

[0002] Research has been conducted to improve the material properties of metal materials. For example, the mechanical properties of stainless steel change when alloy elements are contained. However, simply improving the material properties of metal materials can cause other problems. For example, the alloy elements contained in stainless steel are made from metals refined from various minerals. There is a risk that the price of minerals will rise due to speculative or political influences. The burden of rising mineral prices will be borne by material manufacturers, distributors, and consumers. For this reason, it is necessary to minimize raw material costs while ensuring the necessary material properties. Therefore, it is important to clarify the effect of alloy elements contained in metal materials such as stainless steel on material properties. For example, Patent Document 1 shown below is a document related to the effect of alloy elements on the material properties of metal materials. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6297159 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the technique disclosed in Patent Document 1 can only linearly express the effect of alloying elements on the material properties of a metallic material. Therefore, the technique disclosed in Patent Document 1 cannot quantify the synergistic effect when multiple types of alloying elements are contained. In addition, experiments to investigate the effect of alloying elements contained in a metallic material on the material properties of the metallic material must be conducted for each metallic material having a huge variety of chemical compositions. This may impose a heavy burden on the experimenter. For example, if it were possible to estimate the properties of a metallic material having a specified composition, it would be possible to reduce such a burden.

[0005] The present invention has been made in consideration of the above-mentioned circumstances, and provides a technique that makes it possible to estimate the characteristics of a metal material having a specified component. [Means for solving the problem]

[0006] One aspect of the present invention is an estimation method having a control step of estimating mechanical properties of a metallic material corresponding to condition information to be estimated, using learned model information obtained by performing a learning process in advance using training data including condition information that is information indicating conditions related to a metallic material, and information indicating the mechanical properties of the metallic material corresponding to the conditions indicated by the condition information.

[0007] One aspect of the present invention is the above estimation method, wherein the condition information includes information regarding the contents of multiple elements that affect the mechanical properties of the metal material to be estimated.

[0008] One aspect of the present invention is the above-mentioned estimation method, wherein the condition information includes information on the content of each of the plurality of elements, namely, carbon, phosphorus, sulfur, nickel, chromium, molybdenum, copper, cobalt, niobium, aluminum, nitrogen, silicon, manganese, and titanium.

[0009] One aspect of the present invention is the above estimation method, wherein the condition information includes information on the dimensions of the metal material to be estimated.

[0010] One aspect of the present invention is the above estimation method, wherein the condition information includes information regarding the crystal grains of the metal material to be estimated.

[0011] One aspect of the present invention is the above estimation method, wherein the information indicating the mechanical properties includes information regarding the yield strength of the metallic material.

[0012] One aspect of the present invention is the above estimation method, wherein the information indicating the mechanical properties includes information regarding the tensile strength of the metallic material.

[0013] One aspect of the present invention is the above estimation method, wherein the information indicating the mechanical properties includes information regarding the elongation of the metallic material.

[0014] One aspect of the present invention is the above estimation method, in which in the control step, a logarithm is taken of a value included in the condition information, and the mechanical properties of the metal material are estimated using the obtained value.

[0015] One aspect of the present invention is the above-mentioned estimation method, in which, in the control step, a logarithm is taken of the value contained in the information indicating the mechanical properties, and the mechanical properties of the metal material are estimated using the obtained value.

[0016] One aspect of the present invention is an estimation device that includes a control unit that estimates mechanical properties of a metallic material corresponding to condition information to be estimated, using learned model information obtained by performing a learning process in advance using training data including condition information that is information indicating conditions related to a metallic material, and information indicating the mechanical properties of the metallic material corresponding to the conditions indicated by the condition information.

[0017] One aspect of the present invention is an estimation system comprising an estimation device and a terminal device, wherein the estimation device comprises a communication unit that communicates with other devices, and a control unit that acquires condition information of an object to be estimated from the terminal device via the communication unit, estimates mechanical properties of a metallic material corresponding to the condition information of the object to be estimated using learned model information obtained by performing a learning process in advance using teacher data including condition information that is information indicating conditions related to a metallic material and information indicating mechanical properties of the metallic material corresponding to the conditions indicated by the condition information, and transmits the estimation result to the terminal device via the communication unit, and the terminal device comprises a communication unit that communicates with other devices, an input unit that accepts user operations, an output unit that outputs information to the user, and a control unit that acquires condition information of the object to be estimated in response to operations on the input unit, transmits the condition information of the object to be estimated to the estimation device, and outputs the estimation result from the output unit upon receiving it from the estimation device.

[0018] One aspect of the present invention is a computer program for causing a computer to function as an estimation device that includes a control unit that estimates mechanical properties of a metallic material corresponding to condition information to be estimated, using learned model information obtained by performing a learning process in advance using training data including condition information that is information indicating conditions related to a metallic material, and information indicating the mechanical properties of the metallic material corresponding to the conditions indicated by the condition information. Effect of the Invention

[0019] The present invention makes it possible to estimate the properties of a metal material containing a specified component. This makes it possible to reduce the burden involved in determining the chemical composition of a metal material. More specifically, it makes it possible to reduce a part or all of the burden that was previously required for actual experiments. [Brief description of the drawings]

[0020] [Figure 1] 1 is a schematic block diagram showing a system configuration of an estimation system 100 according to the present invention. [Diagram 2]2 is a schematic block diagram showing a specific example of the functional configuration of a terminal device 10. FIG. [Diagram 3] 2 is a schematic block diagram showing a specific example of the functional configuration of a learning device 20. FIG. [Figure 4] 2 is a diagram showing information relating to specific examples of teacher data used in the learning device 20. FIG. [Diagram 5] FIG. 2 is a diagram showing a more specific example of teacher data used in the learning device 20. [Figure 6] 4 is a flowchart showing a specific example of processing by the learning device 20. [Figure 7] 2 is a schematic block diagram showing a specific example of a functional configuration of an estimation device 30. FIG. [Figure 8] 4 is a flowchart showing a specific example of the process of the estimation device 30. [Figure 9] 1 is a diagram showing an example of a relationship between an experimentally determined 0.2% proof stress and a 0.2% proof stress estimated using learned model information obtained by the learning device 20. FIG. [Figure 10] 1 is a diagram showing an example of the relationship between an experimentally determined tensile strength and a tensile strength estimated using learned model information obtained by the learning device 20. [Figure 11] 13 is a diagram showing an example of the relationship between an experimentally determined elongation and an elongation estimated using learned model information obtained by the learning device 20. FIG. [Figure 12] A figure showing an example of the relationship between the carbon content contained in the metal material to be estimated and the estimated yield strength estimated by the estimation device 30 using the learned model information obtained by the learning device 20. [Figure 13] A figure showing an example of the relationship between the molybdenum content contained in the metal material to be estimated and the estimated yield strength estimated by the estimation device 30 using the learned model information obtained by the learning device 20. [Figure 14] A figure showing an example of the relationship between the nitrogen content contained in the metal material to be estimated and the estimated yield strength estimated by the estimation device 30 using the learned model information obtained by the learning device 20. [Figure 15]A figure showing an example of the relationship between the carbon content contained in the metal material to be estimated and the estimated tensile strength estimated by the estimation device 30 using the learned model information obtained by the learning device 20. [Figure 16] A figure showing an example of the relationship between the molybdenum content contained in the metal material to be estimated and the estimated tensile strength estimated by the estimation device 30 using the learned model information obtained by the learning device 20. [Figure 17] A figure showing an example of the relationship between the nitrogen content contained in the metal material to be estimated and the estimated tensile strength estimated by the estimation device 30 using the learned model information obtained by the learning device 20. [Figure 18] A figure showing an example of the relationship between the carbon content contained in the metal material to be estimated and the estimated elongation estimated by the estimation device 30 using the learned model information obtained by the learning device 20. [Figure 19] A figure showing an example of the relationship between the molybdenum content contained in the metal material to be estimated and the estimated elongation estimated by the estimation device 30 using the learned model information obtained by the learning device 20. [Figure 20] A figure showing an example of the relationship between the nitrogen content contained in the metal material to be estimated and the estimated elongation estimated by the estimation device 30 using the learned model information obtained by the learning device 20. [Figure 21] 1 is a diagram showing an example of a relationship between an experimentally determined 0.2% proof stress and a 0.2% proof stress estimated using learned model information obtained by the learning device 20. FIG. [Figure 22] 1 is a diagram showing an example of the relationship between an experimentally determined tensile strength and a tensile strength estimated using learned model information obtained by the learning device 20. [Diagram 23] 13 is a diagram showing an example of the relationship between an experimentally determined elongation and an elongation estimated using learned model information obtained by the learning device 20. FIG. [Figure 24] FIG. 2 is a diagram illustrating an outline of a hardware configuration example of an information processing device 90 applied to the present embodiment. [Diagram 25] FIG. 13 is a diagram illustrating a modified example of the estimation device 30. [Figure 26] FIG. 13 is a diagram illustrating a modified example of the estimation device 30. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] [Summary] Hereinafter, specific configuration examples of the present invention will be described with reference to the drawings. FIG. 1 is a schematic block diagram showing a system configuration of an estimation system 100 of the present invention. First, an outline of the estimation system 100 will be described. The estimation system 100 is used when estimating the characteristics of a metal material having a predetermined component. Hereinafter, a person who performs an operation for performing such estimation is called a user. The user inputs information indicating conditions related to the metal material to be estimated (hereinafter referred to as "condition information") to the terminal device 10. Specific examples of such condition information include information on the chemical composition of the metal material to be estimated and information on factors that affect the characteristics of the metal material to be estimated. Information on the dimensions of the metal material to be estimated may also be used as the condition information. A specific example of the information on the dimensions may be information indicating the thickness (hereinafter referred to as "plate thickness") of the metal material when it is a plate-shaped material. A specific example of the information on the dimensions may be information indicating the thickness (e.g., diameter) of the metal material when it is a rod-shaped material. Information on the crystal grains of the metal material to be estimated (e.g., crystal grain size) may also be used as the condition information.

[0022] The terminal device 10 transmits the input condition information to the estimation device 30. The estimation device 30 estimates the properties of the metal material according to the conditions indicated by the condition information. The type of the property to be estimated may be one type or multiple types. In this embodiment, a specific example of the property to be estimated is a mechanical property. Specific examples of information indicating the mechanical property are proof stress, tensile strength, and elongation. A more specific example of proof stress is 0.2% proof stress. The units of the values ​​of proof stress and tensile strength may be, for example, MPa (megapascals), and the unit of the value of elongation may be, for example, % (percent). The estimation device 30 transmits information indicating the estimated property (property information) to the terminal device 10. When the terminal device 10 receives the property information, it outputs the received property information. In this way, the user can easily obtain an estimation result for the property of the metal material having the composition to be estimated, without performing an experiment using metal materials having various actual chemical compositions.

[0023] [System Details] Next, the details of the estimation system 100 will be described. The estimation system 100 includes a terminal device 10, a learning device 20, and an estimation device 30. The terminal device 10 and the estimation device 30 are communicatively connected via a network 70. The learning device 20 and the estimation device 30 may be communicatively connected via the network 70. The network 70 may be a network using wireless communication, or may be a network using wired communication. The network 70 may be configured using, for example, the Internet, or may be configured using a local area network (LAN). The network 70 may be configured by combining a plurality of networks.

[0024] 2 is a schematic block diagram showing a specific example of the functional configuration of the terminal device 10. The terminal device 10 is configured using information devices such as a smartphone, a tablet, a personal computer, a dedicated device, etc. The terminal device 10 includes a communication unit 11, an operation unit 12, an output unit 13, a storage unit 14, and a control unit 15.

[0025] The communication unit 11 is a communication device. The communication unit 11 may be configured as, for example, a network interface. The communication unit 11 performs data communication with other devices via the network 70 in response to the control of the control unit 15. The communication unit 11 may be a device that performs wireless communication or a device that performs wired communication.

[0026] The operation unit 12 is configured using existing input devices such as a keyboard, a pointing device (mouse, tablet, etc.), buttons, and a touch panel. The operation unit 12 is operated by a user when inputting a user's instruction to the terminal device 10. The operation unit 12 may be an interface for connecting the input device to the terminal device 10. In this case, the operation unit 12 inputs an input signal generated in the input device in response to a user's input to the terminal device 10. The operation unit 12 may be configured using a microphone and a voice recognition device. In this case, the operation unit 12 recognizes a word spoken by the user and inputs character string information of the recognition result to the terminal device 10. In this case, the operation unit 12 may only input voice, and the voice recognition may be performed by the control unit 15. The operation unit 12 may be configured in any way as long as it is configured to input a user's instruction to the terminal device 10.

[0027] The output unit 13 outputs information in a form that can be recognized by the user. The output unit 13 may be, for example, an image display device such as a liquid crystal display or an organic EL (Electro Luminescence) display. The output unit 13 may be an interface for connecting the image display device to the terminal device 10. In this case, the output unit 13 generates a video signal for displaying image data and outputs the video signal to the image display device connected to the output unit 13. The output unit 13 may be a device for outputting sound such as a speaker. The output unit 13 may be an interface for connecting an audio output device such as a speaker or a headphone to the terminal device 10. In this case, the output unit 13 generates an audio signal for reproducing audio data and outputs the audio signal to the audio output device connected to the output unit 13.

[0028] The storage unit 14 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. The storage unit 14 stores data used by the control unit 15. The storage unit 14 stores data required when the control unit 15 performs processing.

[0029] The control unit 15 is configured using a processor such as a CPU (Central Processing Unit) and a memory (main storage device). The control unit 15 functions by the processor executing a program. All or part of the functions of the control unit 15 may be realized using hardware such as an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array). The above program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a semiconductor storage device (e.g., an SSD: Solid State Drive), or a storage device such as a hard disk or a semiconductor storage device built into a computer system. The above program may be transmitted via an electric communication line.

[0030] The control unit 15 may execute, for example, an application that is preinstalled on the device (terminal device 10). A specific example of such an application is an application provided to the terminal device 10 as a dedicated application for the estimation system 100. Another specific example of such an application is a web browser application. The control unit 15 operates according to the program of the application being executed.

[0031] The control unit 15 controls the terminal device 10 in response to a user's operation or information received from the estimation device 30. For example, the control unit 15 transmits information input by the user operating the operation unit 12 to the estimation device 30 by using the communication unit 11. For example, when the information transmitted from the estimation device 30 is received by the communication unit 11 via the network 70, the control unit 15 generates screen data based on the received information and causes the output unit 13 to display the screen data. Such screen data includes images and characters transmitted from the estimation device 30. For example, when the information transmitted from the estimation device 30 is received by the communication unit 11 via the network 70, the control unit 15 generates voice data based on the received information and causes the output unit 13 to output the voice data.

[0032] Below, a specific example of the operation of the control unit 15 will be described. In the following example, an image display device is used as a specific example of the output unit 13. However, as described above, the output unit 13 does not need to be configured using an image display device, and may be configured using an audio output device, or may be configured using both an image display device and an audio output device.

[0033] The control unit 15 generates screen data having characters and images instructing the user to input condition information, for example. The control unit 15 causes the output unit 13 to display the generated screen data. The control unit 15 may instruct the input of information on, for example, carbon, phosphorus, sulfur, nickel, chromium, molybdenum, copper, cobalt, niobium, aluminum, nitrogen, silicon, manganese, and titanium as the condition information. The control unit 15 may instruct the input of information on the dimensions of the metallic material and information on the crystal grains of the metallic material to be estimated, in addition to information on, for example, carbon, phosphorus, sulfur, nickel, chromium, molybdenum, copper, cobalt, niobium, aluminum, nitrogen, silicon, manganese, and titanium as the condition information. Specific examples of the information on the dimensions of the metallic material include, for example, plate thickness and diameter. Specific examples of the information on the crystal grains of the metallic material include crystal grain size and crystal grain size.

[0034] A user inputs condition information into the terminal device 10 by operating the operation unit 12. The control unit 15 transmits the input condition information to the estimation device 30 using the communication unit 11. The control unit 15 receives an estimation result corresponding to the condition information from the estimation device 30. The control unit 15 generates screen data showing information indicating the estimation result. The control unit 15 causes the output unit 13 to display the generated screen data.

[0035] 3 is a schematic block diagram showing a specific example of the functional configuration of learning device 20. Learning device 20 is configured using an information processing device such as a personal computer or a server device. Learning device 20 includes a communication unit 21, a storage unit 22, and a control unit 23.

[0036] The communication unit 21 is a communication device. The communication unit 21 may be configured as, for example, a network interface. The communication unit 21 performs data communication with other devices via the network 70 in accordance with the control of the control unit 23. The communication unit 21 may be a device that performs wireless communication or a device that performs wired communication.

[0037] The storage unit 22 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. The storage unit 22 stores data used by the control unit 23. The storage unit 22 may function as, for example, a teacher data storage unit 221, a preprocessed teacher data storage unit 222, and a trained model information storage unit 223.

[0038] The teacher data storage unit 221 stores teacher data used in the supervised learning process executed in the learning device 20. The teacher data stored in the teacher data storage unit 221 is data including condition information (explanatory variables) and characteristic information (objective variables) of the metal material according to the condition information. The condition information and characteristic information will be described below.

[0039] The condition information may be, for example, information on the chemical composition of the metal material whose mechanical properties are to be estimated. A specific example of the information on the chemical composition of the metal material is information on elements (e.g., carbon (C), phosphorus (P), sulfur (S), nickel (Ni), chromium (Cr), molybdenum (Mo), copper (Cu), cobalt (Co), niobium (Nb), aluminum (Al), nitrogen (N), silicon (Si), manganese (Mn), titanium (Ti), and other elements) contained in the metal material to be estimated. Among these, information on only some of the elements may be used. For example, information on the contents of a plurality of elements that affect the mechanical properties contained in the metal material to be estimated may be used. For example, information on the respective contents of carbon, nickel, chromium, molybdenum, and nitrogen may be used. Furthermore, for example, information on the respective contents of phosphorus, sulfur, copper, cobalt, niobium, and aluminum may be used in addition to carbon, nickel, chromium, molybdenum, and nitrogen. Further, for example, in addition to carbon, phosphorus, sulfur, nickel, chromium, molybdenum, copper, cobalt, niobium, aluminum, and nitrogen, information on the respective contents of silicon, manganese, and titanium may be used. The unit of the contents referred to here is mass percent. Furthermore, the condition information may include, in addition to information on the chemical composition of the metallic material (content of alloy components), information on factors that affect other mechanical properties (for example, information on precipitates and inclusions contained in the metallic material (for example, type, size (grain size, etc.), amount (number density, etc.) of precipitates and inclusions)). The metallic material referred to here is, for example, steel such as stainless steel, but the metallic material is not limited to steel and may be any metal. For example, it may be a nickel alloy, etc.

[0040] The condition information may further include information on the dimensions of the metallic material to be estimated. A specific example of the information on dimensions may be information indicating the thickness (plate thickness) of the metallic material when it is plate-shaped. A specific example of the information on dimensions may be information indicating the thickness (diameter) of the metallic material when it is rod-shaped. Information on the crystal grains of the metallic material to be estimated (for example, crystal grain size (diameter) or crystal grain size (grain size number)) may also be used as the condition information.

[0041] The property information is information that indicates mechanical properties. Examples of such property information include 0.2% yield strength (YS), tensile strength (TS), and elongation (EL).

[0042] The teacher data having such condition information and characteristic information may be obtained, for example, by conducting an experiment to measure the characteristic information using a metal material according to the condition information. Specifically, the teacher data may be obtained by manufacturing a metal material according to the condition information as a test material and measuring the characteristic information of the manufactured test material, or by cutting a sample from an actual product material instead of or in addition to the test material and measuring the characteristic information. For example, when 0.2% proof stress, tensile strength, and elongation, which are information indicating mechanical properties, are used as teacher data as characteristic information, the 0.2% proof stress, tensile strength, and elongation may be obtained by a tensile test using such a test material or sample. More specifically, the tensile test may be a tensile test method conforming to the provisions of JIS Z 2241:2011, for example. In addition, the teacher data is preferably data created using experimental results on a metal material having a chemical composition relatively close to the chemical composition of the metal material to be estimated for the characteristic information.

[0043] The preprocessed teacher data storage unit 222 stores preprocessed teacher data obtained by executing preprocessing on the teacher data by the preprocessing control unit 232. The trained model information storage unit 223 stores trained model information obtained by executing a learning process by the learning control unit 233.

[0044] The control unit 23 is configured using a processor such as a CPU and a memory. The control unit 23 functions as an information control unit 231, a pre-processing control unit 232, and a learning control unit 233 by the processor executing a program. All or part of the functions of the control unit 23 may be realized using hardware such as an ASIC, a PLD, or an FPGA. The above program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a semiconductor storage device (e.g., SSD), or a storage device such as a hard disk or a semiconductor storage device built into a computer system. The above program may be transmitted via an electric communication line.

[0045] The information control unit 231 controls the input and output of information. For example, the information control unit 231 acquires teacher data from another device (an information processing device or a storage medium) and records the teacher data in the teacher data storage unit 221. For example, the information control unit 231 transmits the trained model information stored in the trained model information storage unit 223 to another device (for example, the estimation device 30).

[0046] The preprocessing control unit 232 generates preprocessed teacher data by performing a predetermined preprocessing on the teacher data. The preprocessing control unit 232 may perform, for example, a process of taking a logarithm of a numerical value included in the teacher data (logarithmic conversion process). More specifically, a process of taking a logarithm of a value included in the condition information (for example, information indicating the chemical composition of a metal material) may be performed. More specifically, a process of taking a logarithm of the contents (mass percent) of a plurality of elements (for example, all elements) included in the condition information may be performed. Also, for example, when the condition information further includes information regarding the dimensions of a metal material, a process of taking a logarithm of a value included in the information regarding the dimensions (for example, specific values ​​of plate thickness and diameter) may be performed. Also, for example, when the condition information further includes information regarding crystal grains, a process of taking a logarithm of a value included in the information regarding crystal grains (for example, crystal grain size) may be performed. Also, for example, a process of taking a logarithm of a value included in the information indicating a mechanical property, which is the characteristic information, (for example, 0.2% proof stress, tensile strength, elongation) may be performed. By carrying out such preprocessing, even when the content of each element differs greatly (for example, by one digit, several digits, or around 10 digits), it is possible to perform more accurate estimation processing by bringing the numerical values ​​representing the content of each element closer together.

[0047] The pre-processing control unit 232 may perform scaling on explanatory variables included in the teacher data. As a specific example of such scaling, standardization or normalization may be performed. As a specific example of such standardization, a process of dividing the deviation from the average value of the data by the standard deviation may be performed.

[0048] When the preprocessing is completed, the preprocessing control unit 232 records the teacher data converted by the preprocessing in the preprocessed teacher data storage unit 222 as preprocessed teacher data.

[0049] The learning control unit 233 executes a supervised learning process using the preprocessed teacher data stored in the preprocessed teacher data storage unit 222. As a specific example of such a learning process, for example, support vector regression (SVR) may be used, or other learning techniques such as neural networks and deep learning may be used. Although support vector regression is a complex model and has low model interpretability, it has the advantage that both linear regression and nonlinear regression are possible and high estimation accuracy can be achieved with less teacher data than deep learning. The learning control unit 233 records the learned model information obtained by executing the learning process in the learned model information storage unit 223. Such learned model information obtained by the learning control unit 233 may be transmitted to the estimation device 30 and recorded in the learned model information storage unit 321 of the estimation device 30. Such learned model information can obtain an estimated value of characteristic information as an output by providing condition information as an input.

[0050] FIG. 4 is a diagram showing information on a specific example of teacher data used in the learning device 20. In FIG. 4, the condition information is information on the content of multiple elements contained in the metal material to be estimated, information on the dimensions of the metal material, and information on crystal grains. In FIG. 4, the characteristic information is information on the mechanical properties of the metal material to be estimated. The information on the content of multiple elements is the content (mass percent) of each of carbon (C), phosphorus (P), sulfur (S), nickel (Ni), chromium (Cr), molybdenum (Mo), copper (Cu), cobalt (Co), niobium (Nb), aluminum (Al), nitrogen (N), silicon (Si), manganese (Mn), and titanium (Ti). As a specific example of the information on dimensions, plate thickness (millimeters) is used, and as a specific example of the information on crystal grains, crystal grain size (grain size number specified in JIS G 0551 (2020)) is used. The information on the mechanical properties is 0.2% yield strength, tensile strength and elongation, the units of the values ​​of 0.2% yield strength and tensile strength are both MPa, and the unit of the value of elongation is %.

[0051] The teacher data was created by manufacturing a metal material according to the condition information shown in FIG. 4, specifically, a metal material containing the alloy elements shown in FIG. 4, and measuring the mechanical properties of the manufactured metal material. The remainder of the metal material manufactured to obtain the teacher data shown in FIG. 4 is iron (Fe) and impurities. Impurities are components that are mixed in due to various factors in raw materials such as ores and scraps and manufacturing processes when manufacturing metal materials industrially. The metal material used to obtain the teacher data may contain trace amounts of elements other than the alloy elements (carbon, phosphorus, sulfur, nickel, chromium, molybdenum, copper, cobalt, niobium, aluminum, nitrogen, silicon, manganese, and titanium) contained in the teacher data, as well as iron (Fe) and impurities. The 0.2% proof stress, tensile strength, and elongation were measured in accordance with the tensile test method specified in JIS Z 2241:2011, for example.

[0052] 4 shows the minimum, average and maximum values ​​of each value. Each value of the training data used in the learning device 20 may be distributed so as to satisfy the statistical values ​​shown in FIG.

[0053] FIG. 5 is a diagram showing a more specific example of the teacher data used in the learning device 20 shown in FIG. 4. In FIG. 5, information about stainless steel is mainly shown as the teacher data. In FIG. 5, the information about the mechanical properties is 0.2% proof stress, tensile strength, and elongation, which are represented as "YS", "TS", and "EL", respectively. The measurement conditions for 0.2% proof stress, tensile strength, and elongation are as described above. In addition, the information about the dimensions of the metal material is the plate thickness, which is represented as "Thickness". In addition, the information about the crystal grains is the crystal grain size (grain size number specified in JIS G 0551 (2020)), which is represented as "GS".

[0054] FIG. 5(A) is a specific example of training data, showing the content (mass percent) of each of carbon (C), silicon (Si), manganese (Mn), nickel (Ni), chromium (Cr), phosphorus (P), sulfur (S), copper (Cu), molybdenum (Mo), cobalt (Co), titanium (Ti), aluminum (Al), niobium (Nb), and nitrogen (N). In addition, “YS” in FIG. 5(A) represents the 0.2% proof stress value (MPa), “TS” represents the tensile strength value (MPa), and “EL” represents the elongation value (%). In addition, “GS” in FIG. 5(A) represents the grain size value, and “Thickness” represents the plate thickness value (mm). In this way, by using information about grains and information about dimensions as condition information (explanatory variables), it is possible to improve the estimation accuracy. However, these values ​​do not necessarily have to be used. FIG. 5(B) shows the values ​​of training data after preprocessing. In the pre-processing, a logarithmic conversion process (a process of calculating common logarithms) is performed on each numerical value included in the teacher data. In the learning device 20, a learning process may be performed using the teacher data thus obtained as shown in Fig. 5(B). In this case, it is desirable that the estimation device 30 also performs a similar pre-processing.

[0055] 6 is a flowchart showing a specific example of processing of the learning device 20. First, the information control unit 231 acquires teacher data (step S101). The teacher data may be input by a user, acquired by communication from another information device, or acquired from a recording medium connected to the learning device 20. The preprocessing control unit 232 executes a predetermined preprocessing on the teacher data (step S102). The learning control unit 233 executes a learning process using the preprocessed teacher data, and records the learned model information in the learned model information storage unit 223 (step S103).

[0056] 7 is a schematic block diagram showing a specific example of a functional configuration of the estimation device 30. The estimation device 30 is configured using an information processing device such as a personal computer or a server device. The estimation device 30 includes a communication unit 31, a storage unit 32, and a control unit 33.

[0057] The communication unit 31 is a communication device. The communication unit 31 may be configured as, for example, a network interface. The communication unit 31 performs data communication with other devices via the network 70 according to the control of the control unit 33. The communication unit 31 may be a device that performs wireless communication or a device that performs wired communication.

[0058] The storage unit 32 is configured using a storage device such as a magnetic hard disk device or a semiconductor storage device. The storage unit 32 stores data used by the control unit 33. The storage unit 32 may function as a trained model information storage unit 321, for example.

[0059] The trained model information storage unit 321 stores information on trained models that have been generated in advance by a learning process. Such a learning process may be executed by, for example, another device (for example, the learning device 20) or may be executed by the device itself (the estimation device 30).

[0060] The control unit 33 is configured using a processor such as a CPU and a memory. The control unit 33 functions as an information control unit 331 and an estimation unit 332 by the processor executing a program. All or part of the functions of the control unit 33 may be realized using hardware such as an ASIC, a PLD, or an FPGA. The above program may be recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a semiconductor storage device (e.g., SSD), or a storage device such as a hard disk or a semiconductor storage device built into a computer system. The above program may be transmitted via an electric communication line.

[0061] The information control unit 331 acquires information from other devices such as the terminal device 10. A specific example of the acquired information is, for example, condition information input to the terminal device 10. The information control unit 331 records the acquired information in a storage device such as a memory. The information control unit 331 transmits information to other devices such as the terminal device 10. A specific example of the transmitted information is, for example, characteristic information estimated by the estimation unit 332. Such exchange of information between the information control unit 331 and other devices may be performed by, for example, communication by the communication unit 31.

[0062] The estimation unit 332 estimates characteristic information according to the condition information obtained from the user. This estimation process is performed using the trained model information stored in the trained model information storage unit 321. The estimation unit 332 executes an inference process using the trained model information and the condition information obtained from the user, thereby estimating characteristic information of the metal material according to the conditions indicated by the condition information.

[0063] FIG. 8 is a flowchart showing a specific example of the process of the estimation device 30. First, the information control unit 331 acquires condition information (step S201). The condition information may be, for example, information input by a user to the terminal device 10 and acquired via the network 70. The estimation unit 332 executes a predetermined preprocessing on the input condition information (step S202). The preprocessing executed by the estimation unit 332 is the same as the preprocessing executed on the teacher data when the trained model information is acquired. For example, the preprocessing executed by the estimation unit 332 is the same as the preprocessing executed on the condition information included in the teacher data by the preprocessing control unit 232 of the learning device 20. The estimation unit 332 executes the estimation process using the preprocessed condition information and the trained model information (step S203). The estimation unit 332 executes the estimation process to acquire characteristic information as an estimation result by executing the estimation process. The information control unit 331 outputs the acquired characteristic information. For example, the information control unit 331 transmits the acquired characteristic information to the terminal device 10 (step S204).

[0064] The results of the experiments are shown below in Figures 9 to 23. Unless otherwise specified, the metal materials used in the experiments were as follows: carbon: 0.05 mass%, silicon: 0.5 mass%, manganese: 1.2 mass%, nickel: 8 mass%, chromium: 18 mass%, phosphorus: 0.03 mass%, sulfur: 0.004 mass%, copper: 0.01 mass%, molybdenum: 0.01 mass%, cobalt: 0.01 mass%, titanium: 0.001 mass%, aluminum: 0.001 mass%, niobium: 0.001 mass%, nitrogen: 0.03, grain size: 8, plate thickness: 1 mm.

[0065] FIG. 9 is a diagram showing an example of the relationship between the experimentally determined 0.2% proof strength (YS) and the 0.2% proof strength (YS) estimated using the trained model information obtained by the learning device 20. The horizontal axis of FIG. 9 indicates the experimentally determined 0.2% proof strength (MPa). The vertical axis of FIG. 9 indicates the 0.2% proof strength (MPa) obtained by the estimation device 30. In FIG. 9, the teacher data (1667 data) shown in FIG. 4 and FIG. 5 are shown as white circles. The hatched circles shown in FIG. 9 are test data different from the teacher data.

[0066] The white and black circles shown in FIG. 9 are both distributed in an approximately linear fashion. Therefore, FIG. 9 shows that the trained model information obtained by the learning device 20 can be obtained with high accuracy, and mechanical property data showing a highly reliable 0.2% proof stress (YS) can be output. In other words, it can be seen that by using the trained model information obtained by the learning device 20, it is possible to obtain property information from condition information with high accuracy. Thus, the experimental results shown in FIG. 9 show that there is a correlation between condition information and property information.

[0067] FIG. 10 is a diagram showing an example of the relationship between the experimentally determined tensile strength (TS) and the tensile strength (TS) estimated using the trained model information obtained by the learning device 20. The horizontal axis of FIG. 10 indicates the experimentally determined tensile strength (MPa). The vertical axis of FIG. 10 indicates the tensile strength (MPa) obtained by the estimation device 30. In FIG. 10, the teacher data shown in FIG. 4 and FIG. 5 are shown as white circles. The hatched circles shown in FIG. 10 are test data different from the teacher data.

[0068] The white and black circles shown in FIG. 10 are both distributed in an approximately linear fashion. Therefore, FIG. 10 shows that the trained model information obtained by the learning device 20 can be obtained with high accuracy, and mechanical property data showing highly reliable tensile strength (TS) can be output. In other words, it can be seen that by using the trained model information obtained by the learning device 20, it is possible to obtain property information from condition information with high accuracy. Thus, the experimental results shown in FIG. 10 show that there is a correlation between condition information and property information.

[0069] FIG. 11 is a diagram showing an example of the relationship between the elongation (EL) obtained experimentally and the elongation (EL) estimated using the trained model information obtained by the learning device 20. The horizontal axis of FIG. 11 indicates the elongation (%) obtained experimentally. The vertical axis of FIG. 11 indicates the elongation (%) obtained by the estimation device 30. In FIG. 11, the teacher data shown in FIG. 4 and FIG. 5 are shown as white circles. The hatched circles shown in FIG. 11 are test data separate from the teacher data.

[0070] The white and black circles shown in FIG. 11 are both distributed in an approximately linear fashion. Therefore, FIG. 11 shows that the trained model information obtained by the learning device 20 can be obtained with high accuracy, and mechanical property data showing highly reliable elongation (EL) can be output. In other words, it can be seen that by using the trained model information obtained by the learning device 20, it is possible to obtain accurate property information from condition information. Thus, the experimental results shown in FIG. 11 show that there is a correlation between condition information and property information.

[0071] FIG. 12 is a diagram showing an example of the relationship between the carbon content contained in the metal material to be estimated and the 0.2% proof stress (YS) (hereinafter referred to as "estimated proof stress") estimated by the estimation device 30 using the learned model information obtained by the learning device 20. The horizontal axis of FIG. 12 indicates the carbon content (mass percent) contained in the metal material to be estimated. The contents of each element other than carbon contained in the metal material to be estimated are as described above. The vertical axis of FIG. 12 indicates the estimated proof stress (MPa). FIG. 12 shows the relationship between the carbon content in steel material and the estimated proof stress.

[0072] 12 shows the relationship between the carbon content and the estimated yield strength for a given composition. As shown in the plot of FIG. 12, the carbon content was changed in the range of 0.003 to 0.1 mass%.

[0073] The graph of the estimation results shown in Figure 12 indicates that as the carbon content increases, the 0.2% proof stress gradually increases, and that the higher the carbon content, the higher the 0.2% proof stress. In other words, it shows how the 0.2% proof stress changes depending on the carbon content. This type of information on the trends in characteristics can be obtained at less cost and in less time than if the metal material were actually produced and experimental results were obtained.

[0074] FIG. 13 is a diagram showing an example of the relationship between the molybdenum content contained in the metal material to be estimated and the 0.2% proof stress (YS) (estimated proof stress) estimated by the estimation device 30 using the learned model information obtained by the learning device 20. The horizontal axis of FIG. 13 indicates the molybdenum content (mass percent) contained in the metal material to be estimated. The contents of each element other than molybdenum contained in the metal material to be estimated are as described above. The vertical axis of FIG. 13 indicates the estimated proof stress (MPa). FIG. 13 shows the relationship between the molybdenum content in steel material and the estimated proof stress.

[0075] 13 shows the relationship between the molybdenum content and the estimated yield strength for a given composition. As shown in the plot of FIG. 13, the molybdenum content was changed in the range of 0.01 to 3.5 mass%.

[0076] The graph of the estimation results shown in Figure 13 indicates that when a trace amount of molybdenum is added, the 0.2% proof stress decreases, but when the molybdenum content exceeds approximately 0.7%, the 0.2% proof stress gradually increases, and the higher the molybdenum content, the higher the 0.2% proof stress. In other words, it shows how the 0.2% proof stress changes depending on the molybdenum content. This information on the trends in characteristics can be obtained at less cost and in less time than if an actual metal material were produced and experimental results were obtained.

[0077] FIG. 14 is a diagram showing an example of the relationship between the nitrogen content contained in the metal material to be estimated and the 0.2% proof stress (YS) (estimated proof stress) estimated by the estimation device 30 using the learned model information obtained by the learning device 20. The horizontal axis of FIG. 14 indicates the nitrogen content (mass percent) contained in the metal material to be estimated. The contents of each element other than nitrogen contained in the metal material to be estimated are as described above. The vertical axis of FIG. 14 indicates the estimated proof stress (MPa). FIG. 14 shows the relationship between the nitrogen content in steel material and the estimated proof stress.

[0078] 14 shows the relationship between the nitrogen content and the estimated yield strength for a given composition. As shown in the plot of FIG. 14, the nitrogen content is changed in the range of 0.01 to 0.055 mass%.

[0079] The graph of the estimation results shown in Figure 14 indicates that as the nitrogen content increases, the 0.2% proof stress gradually increases, and that the higher the nitrogen content, the higher the 0.2% proof stress. In other words, it shows how the 0.2% proof stress changes depending on the nitrogen content. This type of information on the trends in characteristics can be obtained at less cost and in less time than if the metal material were actually produced and experimental results were obtained.

[0080] FIG. 15 is a diagram showing an example of the relationship between the carbon content contained in the metal material to be estimated and the tensile strength (TS) (hereinafter referred to as "estimated tensile strength") estimated by the estimation device 30 using the learned model information obtained by the learning device 20. The horizontal axis of FIG. 15 indicates the carbon content (mass percent) contained in the metal material to be estimated. The contents of each element other than carbon contained in the metal material to be estimated are as described above. The vertical axis of FIG. 15 indicates the estimated tensile strength (MPa). FIG. 15 shows the relationship between the carbon content and the estimated tensile strength in a steel material.

[0081] 15 shows the relationship between the carbon content and the estimated tensile strength for a given composition. As shown in the plot of FIG. 15, the carbon content was varied in the range of 0.003 to 0.1 mass%.

[0082] The graph of the estimation results shown in Figure 15 indicates that as the carbon content increases, the tensile strength gradually increases, and that the higher the carbon content, the higher the tensile strength obtained. In other words, it shows how tensile strength changes depending on the carbon content. This type of information on the trends in characteristics can be obtained at less cost and in less time than if the metal material were actually produced and experimental results were obtained.

[0083] FIG. 16 is a diagram showing an example of the relationship between the molybdenum content contained in the metal material to be estimated and the tensile strength (TS) (estimated tensile strength) estimated by the estimation device 30 using the learned model information obtained by the learning device 20. The horizontal axis of FIG. 16 indicates the molybdenum content (mass percent) contained in the metal material to be estimated. The contents of each element other than molybdenum contained in the metal material to be estimated are as described above. The vertical axis of FIG. 16 indicates the estimated tensile strength (MPa). FIG. 16 shows the relationship between the molybdenum content in a steel material and the estimated tensile strength.

[0084] 16 shows the relationship between the molybdenum content and the estimated tensile strength for a given composition. The molybdenum content was varied in the range of 0.01 to 3.5 mass %, as shown in the plot of FIG.

[0085] The graph of the estimation results shown in Figure 16 indicates that as the molybdenum content increases from a value near 0 (for example, 0.01), the tensile strength gradually increases up to a certain value (a value near 0.5 in the case of Figure 16), after which the tensile strength gradually decreases as the molybdenum content increases, and the higher the molybdenum content, the lower the tensile strength. In other words, it shows how tensile strength changes according to the molybdenum content. This information on the trends in characteristics can be obtained at less cost and in less time than if the metal material were actually produced and experimental results were obtained.

[0086] FIG. 17 is a diagram showing an example of the relationship between the nitrogen content contained in the metal material to be estimated and the tensile strength (TS) (estimated tensile strength) estimated by the estimation device 30 using the learned model information obtained by the learning device 20. The horizontal axis of FIG. 17 indicates the nitrogen content (mass percent) contained in the metal material to be estimated. The contents of each element other than nitrogen contained in the metal material to be estimated are as described above. The vertical axis of FIG. 17 indicates the estimated tensile strength (MPa). FIG. 17 shows the relationship between the nitrogen content in steel material and the estimated tensile strength.

[0087] 17 shows the relationship between the nitrogen content and the estimated tensile strength for a given composition. As shown in the plot of FIG. 17, the nitrogen content was varied in the range of 0.01 to 0.055 mass%.

[0088] The graph of the estimation results shown in Figure 17 indicates that as the nitrogen content increases, the tensile strength gradually decreases, and that the lower the nitrogen content, the higher the tensile strength. In other words, it shows how tensile strength changes depending on the nitrogen content. This type of information on the trends in characteristics can be obtained at less cost and in less time than if the metal material were actually produced and experimental results were obtained.

[0089] FIG. 18 is a diagram showing an example of the relationship between the carbon content contained in the metal material to be estimated and the elongation (EL) (hereinafter referred to as "estimated elongation") estimated by the estimation device 30 using the learned model information obtained by the learning device 20. The horizontal axis of FIG. 18 indicates the carbon content (mass percent) contained in the metal material to be estimated. The contents of each element other than carbon contained in the metal material to be estimated are as described above. The vertical axis of FIG. 18 indicates the estimated elongation (%). Note that FIG. 18 shows the relationship between the carbon content and estimated elongation in steel material.

[0090] 18 shows the relationship between the carbon content and the estimated elongation for a given composition. As shown in the plot of FIG. 18, the carbon content was changed in the range of 0.003 to 0.1 mass %.

[0091] The graph of the estimation results shown in Figure 18 indicates that as the carbon content increases, the elongation gradually decreases, and the lower the carbon content, the greater the elongation. In other words, it shows how elongation changes depending on the carbon content. This type of information on the tendency of characteristics can be obtained at less cost and in less time than if the metal material were actually produced and experimental results were obtained.

[0092] FIG. 19 is a diagram showing an example of the relationship between the molybdenum content contained in the metal material to be estimated and the elongation (EL) (estimated elongation) estimated by the estimation device 30 using the learned model information obtained by the learning device 20. The horizontal axis of FIG. 19 indicates the molybdenum content (mass percent) contained in the metal material to be estimated. The contents of each element other than molybdenum contained in the metal material to be estimated are as described above. The vertical axis of FIG. 19 indicates the estimated elongation (%). FIG. 19 shows the relationship between the molybdenum content in a steel material and the estimated elongation.

[0093] 19 shows the relationship between the molybdenum content and the estimated elongation for a given composition. As shown in the plot of FIG. 19, the molybdenum content was varied in the range of 0.01 to 3.5 mass%.

[0094] The graph of the estimation results shown in Figure 19 indicates that as the molybdenum content increases, the elongation gradually decreases, and the lower the molybdenum content, the greater the elongation. In other words, it shows how elongation changes depending on the molybdenum content. This information on the tendency of characteristics can be obtained at a lower cost and in less time than if the metal material were actually produced and experimental results were obtained.

[0095] FIG. 20 is a diagram showing an example of the relationship between the nitrogen content contained in the metal material to be estimated and the elongation (EL) (estimated elongation) estimated by the estimation device 30 using the learned model information obtained by the learning device 20. The horizontal axis of FIG. 20 indicates the nitrogen content (mass percent) contained in the metal material to be estimated. The contents of each element other than nitrogen contained in the metal material to be estimated are as described above. The vertical axis of FIG. 20 indicates the estimated elongation (%). Note that FIG. 20 shows the relationship between the nitrogen content and the estimated elongation in a steel material.

[0096] 20 shows the relationship between the nitrogen content and the estimated elongation for a given component. As shown in the plot of FIG. 20, the nitrogen content was changed in the range of 0.01 to 0.055 mass%.

[0097] The graph of the estimation results shown in Figure 20 indicates that as the nitrogen content increases, the elongation gradually decreases, and that the lower the nitrogen content, the greater the elongation. In other words, it shows how elongation changes depending on the nitrogen content. This type of information on the tendency of characteristics can be obtained at less cost and in less time than if the metal material were actually produced and experimental results were obtained.

[0098] FIG. 21 is a diagram showing an example of the relationship between the 0.2% proof stress (YS) obtained experimentally and the 0.2% proof stress (YS) estimated using the trained model information obtained by the learning device 20. The experimental results shown in FIG. 21 are obtained by performing the learning process and the estimation process without using the value of the grain size as the condition information (explanatory variable), unlike FIG. 9. The horizontal axis of FIG. 21 shows the 0.2% proof stress (MPa) obtained experimentally. The vertical axis of FIG. 21 shows the 0.2% proof stress (MPa) obtained by the estimation device 30. In FIG. 21, the teacher data (1667 data: not including the grain size (GS)) shown in FIG. 4 and FIG. 5 are shown as white circles. The hatched circles shown in FIG. 21 are test data other than the teacher data.

[0099] The white and black circles shown in FIG. 21 are both distributed in an approximately linear fashion. For this reason, FIG. 21 shows that even without using grain size as an explanatory variable, the learned model information obtained by the learning device 20 can be obtained with high accuracy, and mechanical property data showing a highly reliable 0.2% proof strength (YS) can be output. The squared value (R 2 The value was 0.9019 in FIG. 9 and 0.8577 in FIG.

[0100] FIG. 22 is a diagram showing an example of the relationship between the experimentally obtained tensile strength (TS) and the tensile strength (TS) estimated using the trained model information obtained by the learning device 20. The experimental results shown in FIG. 22 are obtained by performing the learning process and the estimation process without using the value of the grain size as the condition information (explanatory variable), unlike FIG. 10. The horizontal axis of FIG. 22 indicates the experimentally obtained tensile strength (MPa). The vertical axis of FIG. 22 indicates the tensile strength (MPa) obtained by the estimation device 30. In FIG. 22, the teacher data (not including the grain size (GS)) shown in FIG. 4 and FIG. 5 are shown as white circles. The hatched circles shown in FIG. 22 are test data other than the teacher data.

[0101] The white and black circles shown in FIG. 22 are both distributed in an approximately linear fashion. For this reason, FIG. 22 shows that the trained model information obtained by the learning device 20 can be obtained with high accuracy even without using the grain size as an explanatory variable, and mechanical property data showing a highly reliable tensile strength (TS) can be output. The squared value (R 2 The value was 0.9878 in FIG. 10 and 0.9846 in FIG.

[0102] FIG. 23 is a diagram showing an example of the relationship between the elongation (EL) experimentally obtained and the elongation (EL) estimated using the trained model information obtained by the learning device 20. The experimental results shown in FIG. 23 are obtained by performing the learning process and the estimation process without using the value of the grain size as the condition information (explanatory variable), unlike FIG. 11. The horizontal axis of FIG. 23 shows the elongation (%) experimentally obtained. The vertical axis of FIG. 23 shows the elongation (%) obtained by the estimation device 30. In FIG. 23, the teacher data shown in FIG. 4 and FIG. 5 (but not including the grain size (GS)) is shown as white circles. The hatched circles shown in FIG. 23 are test data other than the teacher data.

[0103] The white and black circles shown in FIG. 23 are both distributed in an approximately linear fashion. For this reason, FIG. 23 shows that even without using grain size as an explanatory variable, the learned model information obtained by the learning device 20 can be obtained with high accuracy, and mechanical property data showing highly reliable elongation (EL) can be output. The squared value (R 2 The value was 0.9702 in FIG. 11 and 0.9598 in FIG.

[0104] 21 to 23, the grain size does not necessarily have to be used as condition information (explanatory variable). However, by further using the value of the information on the grain size as condition information (explanatory variable), it is possible to improve the estimation accuracy.

[0105] The metal material whose mechanical properties have been estimated by the estimation device 30 is provided to the market as an industrial product. Furthermore, a structure at least partially formed of the metal material whose mechanical properties have been estimated by the estimation device 30 is provided to the market as an industrial product.

[0106] FIG. 24 is a diagram showing an outline of a hardware configuration example of an information processing device 90 applied to the present embodiment. The information processing device 90 includes a processor 91, a main storage device 92, a communication interface 93, an auxiliary storage device 94, an input / output interface 95, and an internal bus 96. The processor 91, the main storage device 92, the communication interface 93, the auxiliary storage device 94, and the input / output interface 95 are communicably connected to each other via the internal bus 96. The information processing device 90 may be applied to, for example, the learning device 20 and the estimation device 30. In this case, for example, the communication unit 21 and the communication unit 31 may be configured using the communication interface 93. For example, the memory unit 22 and the memory unit 32 may be configured using the auxiliary storage device 94. In addition, the control unit 23 and the control unit 33 may be configured using the processor 91 and the main storage device 92.

[0107] The estimation system 100 of the present embodiment makes it possible to estimate the properties of a metallic material having a predetermined component. This makes it possible to reduce the burden incurred in determining the chemical composition of a metallic material. More specifically, it becomes possible to reduce a part or all of the burden that has been required for actual experiments up to now.

[0108] (Modification) In this embodiment, the terminal device 10 and the estimation device 30 are configured as different devices, but may be configured as an integrated device. FIG. 25 is a diagram showing a modified example of the estimation device 30 configured in this way. The estimation device 30 shown in FIG. 25 includes an operation unit 34 and an output unit 35. The operation unit 34 and the output unit 35 of the estimation device 30 shown in FIG. 25 function in the same manner as the operation unit 12 and the output unit 13 of the terminal device 10, respectively. The control unit 33 operates in response to an operation on the operation unit 34, and outputs information using the output unit 35.

[0109] In this embodiment, the learning device 20 and the estimation device 30 are configured as different devices, but may be configured as an integrated device. FIG. 26 is a diagram showing a modified example of the estimation device 30 configured in this manner. The storage unit 32 of the estimation device 30 shown in FIG. 26 also functions as a teacher data storage unit 322 and a preprocessed teacher data storage unit 323. The control unit 33 of the estimation device 30 shown in FIG. 26 also functions as a preprocessing control unit 333 and a learning control unit 334. The teacher data storage unit 322 and the preprocessed teacher data storage unit 323 function in the same manner as the teacher data storage unit 221 and the preprocessed teacher data storage unit 222 of the learning device 20, respectively. The preprocessing control unit 333 and the learning control unit 334 function in the same manner as the preprocessing control unit 232 and the learning control unit 233 of the learning device 20, respectively.

[0110] The learning device 20 may be implemented using a plurality of information processing devices. For example, the learning device 20 may be implemented using a device such as a cloud. For example, in the learning device 20, the storage unit 22 and the control unit 23 may be implemented in different information processing devices. For example, the storage unit 22 of the learning device 20 may be distributed and implemented in a plurality of information processing devices. The estimation device 30 may be implemented using a plurality of information processing devices. For example, the estimation device 30 may be implemented using a device such as a cloud. For example, in the estimation device 30, the storage unit 32 and the control unit 33 may be implemented in different information processing devices. For example, the storage unit 32 of the estimation device 30 may be distributed and implemented in a plurality of information processing devices.

[0111] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of the present invention are also included. [Explanation of symbols]

[0112] 100... estimation system, 10... terminal device, 11... communication unit, 12... operation unit, 13... output unit, 14... memory unit, 15... control unit, 20... learning device, 21... communication unit, 22... memory unit, 221... teacher data memory unit, 222... preprocessed teacher data memory unit, 223... trained model information memory unit, 23... control unit, 231... information control unit, 232... preprocessing control unit, 233... learning control unit, 30... estimation device, 31... communication unit, 32... memory unit, 321... trained model information memory unit, 33... control unit, 331... information control unit, 332... estimation unit

Claims

1. An estimation method having a control step of estimating mechanical properties of a metallic material corresponding to the condition information of the estimation target, using learned model information obtained by performing a learning process in advance using training data including condition information that is information indicating conditions related to a metallic material, and information indicating the mechanical properties of the metallic material corresponding to the conditions indicated by the condition information.

2. The estimation method according to claim 1 , wherein the condition information includes information regarding contents of a plurality of elements that are contained in the metal material to be estimated and that affect mechanical properties.

3. 3. The estimation method according to claim 2, wherein the condition information includes information regarding the respective contents of carbon, phosphorus, sulfur, nickel, chromium, molybdenum, copper, cobalt, niobium, aluminum, nitrogen, silicon, manganese, and titanium as the contents of the plurality of elements.

4. The estimation method according to claim 2 , wherein the condition information includes information regarding a dimension of the metal material to be estimated.

5. The estimation method according to claim 2 , wherein the condition information includes information regarding crystal grains of the metal material to be estimated.

6. The method according to claim 1 , wherein the information indicative of the mechanical properties includes information regarding a yield strength of the metallic material.

7. The method of claim 1 , wherein the information indicative of the mechanical properties includes information regarding a tensile strength of the metallic material.

8. The method according to claim 1 , wherein the information indicative of the mechanical properties includes information regarding the elongation of the metallic material.

9. The estimation method according to claim 1 , wherein in said control step, a logarithm is taken for a value included in said condition information, and the mechanical property of said metallic material is estimated using the obtained value.

10. The estimation method according to claim 1 , wherein in the control step, a logarithm is taken of a value included in the information indicating the mechanical property, and the mechanical property of the metallic material is estimated using the obtained value.

11. a control unit that estimates mechanical properties of a metallic material corresponding to the condition information of an estimation target, using learned model information obtained by performing a learning process in advance using training data including condition information that is information indicating conditions related to a metallic material, and information indicating mechanical properties of the metallic material corresponding to the conditions indicated by the condition information; An estimation device comprising:

12. An estimation system including an estimation device and a terminal device, The estimation device includes: A communication unit for communicating with other devices; a control unit that acquires condition information of an object to be estimated from the terminal device via the communication unit, estimates mechanical properties of the metallic material corresponding to the condition information of the object to be estimated using trained model information obtained by performing a learning process in advance using teacher data including condition information indicating conditions related to the metallic material and information indicating mechanical properties of the metallic material corresponding to the conditions indicated by the condition information, and transmits an estimation result to the terminal device via the communication unit; Equipped with The terminal device A communication unit for communicating with other devices; an input unit for accepting user operations; an output unit for outputting information to a user; a control unit that acquires condition information of the estimation object in response to an operation on the input unit, transmits the condition information of the estimation object to the estimation device, and outputs the estimation result from the output unit upon receiving the estimation result from the estimation device.

13. A computer program for causing a computer to function as an estimation device having a control unit that estimates the mechanical properties of a metallic material corresponding to the condition information of the object to be estimated, using trained model information obtained by performing a learning process in advance using training data including condition information that is information indicating conditions related to a metallic material, and information indicating the mechanical properties of the metallic material corresponding to the conditions indicated by the condition information.

Images