Heat treatment temperature estimation device and heat treatment temperature estimation method

Abstract

To estimate heat treatment temperature of lastly performed heat treatment of heat treatment performed for a metallic body composed of prescribed alloy components and fabricated through heat treatment.SOLUTION: A heat treatment temperature estimation device comprises: a content data acquisition part acquiring content data about content of alloy components; a first vector generation part generating a first vector representing the distribution state of elements contained in the alloy component on the basis of the content data; a second vector generation part performing thermodynamic analysis calculation for a plurality of temperatures using the content data, and generating for each temperature a second vector representing the distribution state of a physical quantity about the metallic body; and an estimation part calculating similarity between the first vector and the second vector at each temperature, and estimating the last heat treatment temperature on the basis of the obtained similarity.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a heat treatment temperature estimation device and a heat treatment temperature estimation method. [Background technology]

[0002] In recent years, advances in information processing technology have led to the development of techniques for computationally and analytically predicting the structure (metallographic structure) of metallic materials using predetermined physical and mathematical models. By using such techniques, it is possible to predict to some extent what kind of metallographic structure will be obtained without actually manufacturing and verifying the metallic material, and it is therefore expected that the development time and development costs for new metallic materials will be shortened.

[0003] For example, Patent Document 1 below proposes a technology for predicting the structure of steel material through computational analysis using a specific physical model, and Patent Document 2 below proposes a technology for predicting the characteristic values ​​of metal materials using a mathematical model including machine learning. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-7809 [Patent Document 2] Japanese Patent Publication No. 2020-115258 Summary of the Invention [Problem to be solved by the invention]

[0005] The oil well tubular goods (OCTG) used in oil well production and development are often seamless steel pipes, which have no axial seams and are manufactured by drilling round billets. In oil and gas wells, steel pipes are connected with screws and lowered deep underground to form transportation lines for oil and natural gas. These steel pipes are used for different purposes, such as casings that prevent the collapse of the outer layers of the earth, and tubing used in oil and natural gas production. Because the required strength and corrosion resistance vary depending on the depth of laying, various steel pipes manufactured in "lots" with different manufacturing conditions are used.

[0006] When steel pipes are left at installation sites for long periods of time, the markings used to identify the lot may become invisible or disappear, making it impossible to determine which lot the steel pipe came from. In such cases, if the metal structure that makes up the steel pipe could be used to identify the lot, like a human fingerprint, it is expected that confusion at oil and gas well sites would be eliminated.

[0007] In this way, if it were possible to estimate the manufacturing conditions of a metal body using an actually existing metal body as a starting point, rather than using the manufacturing conditions of the metal body as a starting point to predict the structure and physical properties of the resulting metal body as in the techniques of Patent Documents 1 and 2, the manufacturing conditions of the metal body could be used as an index of the metal structure, which is expected to improve convenience for users of metal bodies.

[0008] Here, metal bodies are generally subjected to heat treatment in one or more steps during their manufacture to tailor the structure and various properties of the final metal body. Therefore, the temperature during the heat treatment applied to the metal body (more specifically, the temperature maintained during the heat treatment) is one of the important manufacturing conditions for the metal body and is considered to be useful as an indicator of the metal structure.

[0009] The present invention has been made in consideration of the above-mentioned circumstances, and an object of the present invention is to provide a heat treatment temperature estimation device and a heat treatment temperature estimation method that are capable of estimating the heat treatment temperature of the last heat treatment performed on a metal body that is made of a specified alloy composition and manufactured by undergoing heat treatment. [Means for solving the problem]

[0010] As a result of intensive research into solving the above problems, the inventors came up with the idea of ​​utilizing the spatial non-uniformity of a metal body and performing coupled analysis with a physical model, and have completed the present invention, which will be described in detail below. The gist of the present invention, which was completed based on this idea, is as follows.

[0011] (1) A heat treatment temperature estimation device that estimates a final heat treatment temperature, which is the heat treatment temperature of the last heat treatment performed on a metal body that is made of a predetermined alloy component and manufactured by heat treatment, comprising: a content data acquisition unit that acquires content data related to the content of the alloy component in the metal body; a first vector generation unit that generates a first vector that represents the distribution state of elements contained in the alloy component based on the content data; a second vector generation unit that uses the content data to perform thermodynamic analysis calculations at multiple temperatures and generates, for each of the temperatures, second vectors that represent the distribution state of physical quantities related to the metal body from the results of the thermodynamic analysis calculations; and an estimation unit that calculates the similarity between the first vector and the second vector at each of the temperatures by a similarity analysis process and estimates the final heat treatment temperature based on the obtained similarity. (2) The heat treatment temperature estimation device described in (1), wherein the estimation unit identifies the temperature that gives the maximum or minimum value of the similarity in a plane defined by the temperature and the similarity, and estimates the final heat treatment temperature based on the temperature that gives the maximum or minimum value. (3) The heat treatment temperature estimation device described in (2), wherein the estimation unit identifies the temperature that gives the maximum or minimum value of the similarity in a plane defined by the temperature and the similarity for each of multiple combinations of the first vector and the second vector for the different elements, and estimates the final heat treatment temperature based on the temperatures that give the obtained multiple maximum or minimum values. (4) The heat treatment temperature estimation device according to (3), wherein the estimation unit determines the minimum, maximum, or average value of the obtained plurality of temperatures as the final heat treatment temperature. (5) The heat treatment temperature estimation device according to any one of (1) to (4), wherein the first vector is a vector that represents a distribution state of an element related to the physical quantity in the metal body. (6) The heat treatment temperature estimation device according to any one of (1) to (5), wherein the second vector is a vector that represents a distribution state of the phase fraction of the metal body. (7) The heat treatment temperature estimation device according to any one of (1) to (6), wherein the content data is analytical data obtained by a method of measuring and analyzing the spatial distribution of the alloy components. (8) The heat treatment temperature estimation device according to any one of (1) to (7), wherein the content data is analytical data obtained by analyzing the metal body using an electron probe microanalyzer (EPMA). (9) The heat treatment temperature estimation device according to any one of (1) to (8), wherein the second vector generation unit performs the thermodynamic analysis calculation using a CALPHAD (Computer Coupling of Phase Diagrams and Thermochemistry) method. (10) The heat treatment temperature estimation device according to any one of (1) to (9), wherein the similarity analysis process is an analysis process using any one of EuclideanDistance, SquaredEuclideanDistance, NormalizedSquaredEuclideanDistance, ManhattanDistance, CosineDistance, CorrelationDistance, MeanEuclideanDistance, MeanSquaredEuclideanDistance, RootMeanSquare, MeanReciprocalSquaredEuclideanDistance, MutualInformationVariation, NormalizedMutualInformationVariation, DifferenceNormalizedEntropy, MeanPatternIntensity, GradientCorrelation, MeanReciprocalGradientDistance, or EarthMoverDistance as a function, or a similarity analysis process using deep learning processing. (11) A heat treatment temperature estimation device described in any one of (1) to (10), wherein the first vector generation unit uses the generated first vector to further generate a first mapping image that visualizes the distribution of elements contained in the alloy components, the second vector generation unit uses the generated second vector to further generate a second mapping image that visualizes the distribution of physical quantities related to the metal body, and the estimation unit calculates the similarity using the first mapping image and the second mapping image instead of the first vector and the second vector. (12) The heat treatment temperature estimation device according to (11), wherein the first mapping image and the second mapping image are grayscale images. (13) A heat treatment temperature estimation device according to any one of (1) to (12), further comprising: a heat treatment time change function derivation unit that derives a heat treatment time change function that represents the time change of a heat treatment time dependent quantity, which is an observable quantity or physical quantity that changes depending on the heat treatment time at the estimated final heat treatment temperature; a heat treatment time dependent quantity data acquisition unit that acquires heat treatment time dependent quantity data, which is data regarding the heat treatment time dependent quantity of the metal body; and a final heat treatment time estimation unit that estimates the final heat treatment time, which is the heat treatment time at the final heat treatment temperature of the metal body, based on the acquired heat treatment time dependent quantity of the metal body and the heat treatment time change function. (14) The heat treatment temperature estimation device according to (13), wherein the heat treatment time change function deriving unit derives the time change function using the heat treatment time dependent quantity identified at a plurality of different times. (15) The heat treatment time dependent quantity is a physical quantity representing the mean free energy of the structure of the metal body, and the physical quantity representing the mean free energy is expressed as V with the heat treatment time t as a variable. P (t), and the equilibrium free energy of the structure of the metal body at temperature T calculated from the content data is E Eq T (t), and the free energy of the reference state in the structure of the metal body at temperature T calculated from the content data is E0 T The heat treatment temperature estimation device according to (14), wherein (t) is defined by the following equation (1): (16) The heat treatment temperature estimation device described in (15), wherein the heat treatment time change function derivation unit fits the equation (1) to a function form expressed by a linear combination of one or more terms shown on the right side of the following equation (2), and the obtained fitting result is the heat treatment time change function. In the following formula (2), t is a variable representing the heat treatment time, and V P 0 , A Pi , τ Pi are coefficients, and i is an integer equal to or greater than 1. (17) The heat treatment temperature estimation device according to any one of (13) to (16), further comprising a database creation unit that creates a database of the derived heat treatment time change function, and the final heat treatment time estimation unit estimates the final heat treatment time using the heat treatment time change function that has been created in the database. (18) A heat treatment temperature estimation method for estimating a final heat treatment temperature, which is the heat treatment temperature of the last heat treatment performed on a metal body made of a predetermined alloy component and manufactured by heat treatment, comprising: a content data acquisition step for acquiring content data related to the content of the alloy component in the metal body; a first vector generation step for generating a first vector representing the distribution state of elements contained in the alloy component based on the content data; a second vector generation step for performing a thermodynamic analysis calculation for a plurality of temperatures using the content data, and generating, for each of the temperatures from the results of the thermodynamic analysis calculation, a second vector representing the distribution state of physical quantities related to the metal body; and an estimation step for calculating the similarity between the first vector and the second vector at each of the temperatures by a similarity analysis process, and estimating the final heat treatment temperature based on the obtained similarity. (19) The heat treatment temperature estimation method described in (18) further includes a heat treatment time change function derivation step for deriving a heat treatment time change function that represents the time change of a heat treatment time dependent quantity, which is an observable quantity or physical quantity that changes depending on the heat treatment time at the estimated final heat treatment temperature; a heat treatment time dependent quantity data acquisition step for acquiring heat treatment time dependent quantity data, which is data regarding the heat treatment time dependent quantity of the metal body; and a final heat treatment time estimation step for estimating the final heat treatment time, which is the heat treatment time of the metal body at the final heat treatment temperature, based on the acquired heat treatment time dependent quantity of the metal body and the heat treatment time change function.

[0012]

number

[0013] As described above, according to the present invention, it is possible to estimate the heat treatment temperature of the last heat treatment performed on a metal body made of a specified alloy composition and manufactured by heat treatment. [Brief explanation of the drawings]

[0014] [Figure 1] 5A to 5C are schematic diagrams for explaining a manufacturing process of the metal body. [Figure 2A] 3 is an explanatory diagram for explaining the heat treatment temperature estimation process performed by the heat treatment temperature estimation device according to each embodiment of the present invention. FIG. [Figure 2B] 3 is an explanatory diagram for explaining the heat treatment temperature estimation process performed by the heat treatment temperature estimation device according to each embodiment of the present invention. FIG. [Figure 3] This is a visualized diagram of the analysis results obtained by analyzing a steel material, which is an example of a metal body, using an electron probe microanalyzer (EPMA). [Figure 4] 1 is a block diagram schematically illustrating the overall configuration of a heat treatment temperature estimation device according to a first embodiment of the present invention. [Figure 5] FIG. 10 is an explanatory diagram for explaining acquisition of content data. [Figure 6] FIG. 10 is a diagram showing an example of a first mapping image. [Figure 7] FIG. 2 is a block diagram showing an example of a configuration of a second image generation unit according to the first embodiment of the present invention. [Figure 8] FIG. 10 is a diagram showing an example of a second mapping image. [Figure 9] FIG. 10 is an explanatory diagram schematically showing the relationship between temperature and similarity corresponding to the second mapping image. [Figure 10] FIG. 10 is a diagram showing an example of a calculation result of similarity. [Figure 11] 3 is a flowchart showing an example of the flow of a heat treatment temperature estimation method according to the first embodiment of the present invention. [Figure 12]FIG. 10 is an explanatory diagram for explaining the heat treatment time estimation process performed by the heat treatment temperature estimation device according to the second embodiment of the present invention. [Figure 13] FIG. 4 is a block diagram schematically illustrating the overall configuration of a heat treatment temperature estimation device according to a second embodiment of the present invention. [Figure 14] FIG. 10 is an explanatory diagram for explaining a heat treatment time estimation process according to a second embodiment of the present invention. [Figure 15] 10A and 10B are explanatory views for explaining an example of heat treatment performed on a metal body sample. [Figure 16] 10A to 10C are diagrams showing a first group of mapping images of the obtained metal body sample. [Figure 17] FIG. 10 is a diagram showing estimated values ​​of final heat treatment temperatures of the obtained metal body samples. [Figure 18] 10A and 10B are diagrams showing a group of mapping images that visualize the distribution of free energy in the metal structure of the obtained metal body sample. [Figure 19] FIG. 10 is a graph showing the change in average free energy over time in the obtained metal body sample. [Figure 20] FIG. 10 is a diagram showing a heat treatment time change function corresponding to the obtained metal body sample. [Figure 21] 6 is a flowchart showing an example of the flow of a heat treatment temperature estimation method according to a second embodiment of the present invention. [Figure 22] FIG. 1 is a block diagram showing an example of a hardware configuration of a heat treatment temperature estimation device according to each embodiment of the present invention. [Figure 23] 10 is a diagram showing estimated values ​​of the final heat treatment time of the metal body sample of Example 2. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0015] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant explanations will be omitted.

[0016] Before describing the heat treatment temperature estimation device and the heat treatment temperature estimation method according to each embodiment of the present invention, a manufacturing process of a metal body will be briefly described with reference to Fig. 1. Fig. 1 is a schematic diagram for explaining the manufacturing process of a metal body.

[0017] Various metal bodies, including steel materials, are manufactured by subjecting the raw material for the metal body (for example, in the case of steel materials, a mass such as a slab, bloom, or billet) to various treatments in one or more steps. For example, as schematically shown in Figure 1, the raw material is subjected to a series of steps, such as step A, step B, step C, etc., to achieve various characteristics, including the shape of the metal body to be manufactured. Specific steps include heating steps, processing steps, and various other steps. Subsequently, heat treatment is generally performed on the metal body to achieve the desired structure (metal structure) and strength (for example, tensile strength).

[0018] For example, when a seamless steel pipe is manufactured using the Mannesmann process, a cylindrical billet is prepared as a raw material, and this billet is subjected to a heating process using a heating furnace and a hot working process using a piercer, a mandrel mill, a sizing mill, etc., to manufacture a tubular structure having a desired shape. Thereafter, this tubular structure is subjected to heat treatment, and a steel pipe having a metal structure capable of achieving a desired strength is manufactured.

[0019] As described above, the microstructure of the final metal body is generally created by heat treatment, and therefore the temperature during the heat treatment applied to the metal body (more specifically, the temperature maintained during the heat treatment) is one of the important manufacturing conditions for the metal body and is considered to be useful as an indicator of the metal microstructure.

[0020] Therefore, in each embodiment of the present invention described below, a technology for estimating the processing temperature in the last heat treatment applied to an actual metal body, starting from information obtained from the actual metal body, is described in detail.

[0021] First Embodiment (About the heat treatment temperature estimation device) A heat treatment temperature estimation device according to an embodiment of the present invention is a device for estimating the heat treatment temperature (holding temperature in heat treatment) of the last heat treatment performed on a metal body made of a predetermined alloy composition and manufactured by heat treatment. In the following description, the "heat treatment temperature of the last heat treatment performed on the metal body" will be referred to as the "final heat treatment temperature."

[0022] <Outline of heat treatment temperature estimation technology> 2A and 2B are explanatory diagrams for explaining the heat treatment temperature estimation process performed by the heat treatment temperature estimation device according to this embodiment. As shown in FIG. 2A, when manufacturing a metal body, the metal body is manufactured by performing heat treatment at a predetermined temperature (the final heat treatment temperature of interest in this embodiment). In the heat treatment temperature estimation device of this embodiment, in contrast to this manufacturing flow, an actually existing metal body is used as a sample, and information obtained from this metal body sample is used to estimate the heat treatment temperature of the last heat treatment performed. This estimated heat treatment temperature becomes the final heat treatment temperature for the metal body sample of interest.

[0023] Furthermore, some metal bodies are manufactured through multiple heat treatment processes. For example, as shown in Figure 2B, if a metal body is manufactured through three heat treatment processes, the final heat treatment temperature obtained by using the final metal body as a sample is the heat treatment temperature of "Heat Treatment 3," which is the last heat treatment performed on the metal body. Furthermore, if we focus on "Intermediate 2" before being subjected to "Heat Treatment 3," the last heat treatment performed on "Intermediate 2" is "Heat Treatment 2." Therefore, by using "Intermediate 2" as a sample, the heat treatment temperature of "Heat Treatment 2," which is the last heat treatment performed on "Intermediate 2," can be obtained. Similarly, by using "Intermediate 1" before being subjected to "Heat Treatment 2," the heat treatment temperature of "Heat Treatment 1," which is the last heat treatment performed on "Intermediate 1," can be obtained.

[0024] In this way, the heat treatment temperature estimation technique described below makes it possible to estimate the heat treatment temperature (final heat treatment temperature) in the last heat treatment applied to a metal body of interest. Furthermore, by tracing the sample used for estimation back in order to a metal body subjected to an even earlier heat treatment step, it becomes possible to estimate the final heat treatment temperature in an even earlier heat treatment step in a stepwise manner.

[0025] <About metal bodies> Next, a metal body that is the focus of the heat treatment temperature estimation device according to this embodiment will be briefly described with reference to Fig. 3. Fig. 3 is a visualized diagram of the analysis results obtained by analyzing a steel material, as an example of a metal body, with an electron probe microanalyzer (EPMA).

[0026] More specifically, Figure 3 shows the results (local analysis data) of an EPMA analysis of the surface of a steel sheet containing 0.3 mass% C, 2.0 mass% Si, 5.0 mass% Mn, and the remainder Fe. The image on the bottom right is a composite image (CP image) obtained when analyzing this steel sheet, and the other three images are area analysis results showing the distribution of C, Si, and Mn, respectively.

[0027] As described above, metal bodies, including steel materials (especially steel materials), exhibit significant spatial nonuniformity in the distribution of contained elements, and the material itself essentially resembles a micro / nanoplate, as used in high-throughput chemistry. This spatial nonuniformity is considered to be a useful feature for estimating the manufacturing conditions of a metal body of interest. Therefore, the heat treatment temperature estimation device described in detail below utilizes information corresponding to the spatial nonuniformity of the contained elements of the metal body to estimate the heat treatment temperature. More specifically, information regarding the content of alloying elements in the metal body of interest is used as the information corresponding to the spatial nonuniformity.

[0028] The heat treatment temperature estimation technology according to the present embodiment is not limited to metal bodies made of steel, but can be applied to any metal body that can be subjected to thermodynamic analysis calculations using the physical model described below. For example, the technology can be suitably applied to metal bodies made of alloys based on magnesium, copper, nickel, and cobalt, as well as non-ferrous metals such as aluminum and titanium. The technology can also be applied to material systems other than metal bodies whose structure (which can also be referred to as composition distribution) is revealed by heat treatment, such as the structure of ceramic materials such as alumina.

[0029] <Configuration of the heat treatment temperature estimation device> Next, the configuration of the heat treatment temperature estimation device according to this embodiment will be described in detail with reference to FIGS.

[0030] FIG. 4 is a block diagram that schematically shows the overall configuration of the heat treatment temperature estimation device 10 according to this embodiment. The heat treatment temperature estimation device 10 of this embodiment includes a content data acquisition unit 101, a first vector generation unit 103, a second vector generation unit 105, an estimation unit 107, an output control unit 109, a display control unit 111, and a memory unit 113.

[0031] The content data acquiring unit 101 is realized by, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), an input device, a communication device, etc. The content data acquiring unit 101 acquires content data relating to the content of alloy components in a metal body of interest.

[0032] The content data acquired by the content data acquisition unit 101 is data on information corresponding to the spatial non-uniformity of the metal body, as shown in FIG. 3, for example. This content data can be acquired by measuring and analyzing the metal body using a method for measuring and analyzing the spatial distribution of alloy components. An example of a method for measuring and analyzing the spatial distribution of alloy components is EPMA. It is also possible to obtain such content data using an analysis method other than EPMA, such as scanning the surface of a metal body using an X-ray measurement probe. However, since content data can be acquired more easily using EPMA, it is preferable to acquire the content data by analyzing the metal body of interest using EPMA.

[0033] FIG. 5 is an explanatory diagram for explaining acquisition of content data. The measurement conditions for EPMA etc. are not particularly limited, and may be set appropriately so that the metal object of interest can be measured appropriately. However, the size of one side of the measurement area of ​​the metal object used for measurement (L in FIG. 5) obs ) is preferably in the range of several μm to several tens of μm. As a result, one measurement area has a size of several μm to several tens of μm square. The size L of one side of the measurement area obs is more preferably in the range of 20 μm to 60 μm. Furthermore, when measuring, the number of measurement data per unit length is preferably 10 / μm or more. By performing the measurement so as to obtain such a number of measurement data, it becomes possible to efficiently obtain content data consisting of a statistically significant number of measurement data when estimating the heat treatment temperature.

[0034] Upon acquiring the content data thus acquired, the content data acquiring unit 101 outputs the acquired content data to the first image generating unit 103 and the second image generating unit 105, which will be described later. The content data acquiring unit 101 may also associate the acquired content data with time information regarding the date and time when the data was acquired, and store the data in the storage unit 113 as history information.

[0035] The first vector generation unit 103 is realized by, for example, a CPU, a ROM, a RAM, etc. The first vector generation unit 103 generates a first vector representing the distribution state of elements contained in the alloy components of the metal body based on the content data transmitted from the content data acquisition unit 101. The first vector generation unit 103 may further generate a first mapping image, which is an image that visualizes the distribution state of elements contained in the alloy components of the metal body, using the generated first vector. Generating such a first mapping image makes it possible to grasp the generated first vector as an image, making it easier to understand the generated first vector. The following description will be given taking as an example a case where the first vector generation unit 103 further generates a first mapping image.

[0036] In this embodiment, a vector is information indicating the correspondence between a position in physical space and a quantity, and the quantity is the content of alloy components, a physical quantity related to a metal body described later, a pixel value, etc. In the field of digital information processing, points in physical space can be numbered, so they are numbered 1,...,n, and the distribution of quantity y is expressed as a vector (y1,...,y n ) is commonly expressed. Points on the plane are often numbered using double subscripts ij (i = 1, , m, j = 1, , n), but even in this case, the distribution of a quantity y on the plane can be expressed as a vector (y 11 , , y 1n , , y m1 , , y mn ) and so there is no essential difference.

[0037] The content data is numerical data relating to the content of each alloy component (for example, C, Si, and Mn in the case of the steel plate illustrated in FIG. 3) obtained by measuring the metal body for each measurement unit, as shown schematically in FIG. 5. The first vector generation unit 103 generates a first vector by arranging the content of elements contained in the alloy component for each measurement unit, for example. The first vector generation unit 103 also generates a first mapping image by converting the content of elements contained in the alloy component into pixel values ​​for each measurement unit, for example.

[0038] For example, the length of one side L as shown in Figure 5 obs Assume that in a measurement region having a value of 0.028 to 0.015, the content data representing the Mn content is distributed within a range of 3.028 to 8.315 mass%. In this case, when generating an 8-bit first mapping image, for example, the first vector generation unit 103 divides the numerical range of 3.028 to 8.315 into 256 equal parts and converts the Mn content in each measurement unit into a pixel value. By performing this process for the entire measurement region, a first mapping image can be generated that visualizes the distribution of Mn, among the alloy components of the metal body. Furthermore, since the first mapping image is obtained by converting the content of the element of interest into pixel values, it is preferable that it be a grayscale image.

[0039] An example of the first mapping image generated in this manner is shown in Figure 6. Figure 6 is an image that visualizes the distribution of Mn based on the content data obtained from the EPMA analysis results shown in Figure 3. In Figure 6, the blacker the color of each pixel, the higher the content, and the whiter the color, the lower the content.

[0040] Such first vectors and first mapping images may be generated for some or all of the alloy components contained in the metal body. Also, although the above describes the case where a grayscale image with 8-bit gradation is generated, the number of gradations of the grayscale image is not limited to the above example.

[0041] After generating the first vector and the first mapping image for at least one element of the alloy components as described above, the first vector generation unit 103 outputs the data of the generated first vector and the first mapping image to the estimation unit 107 described later. Furthermore, the first vector generation unit 103 may store the data of the generated first vector and the first mapping image in the storage unit 113 as history information after associating the data with time information regarding the date and time when the data was generated.

[0042] The second vector generation unit 105 is realized by, for example, a CPU, a ROM, a RAM, etc. The second vector generation unit 105 performs thermodynamic analysis calculations for multiple temperatures using the content data transmitted from the content data acquisition unit 101. The second vector generation unit 105 generates second vectors representing the distribution of physical quantities related to the metal body from the results of the thermodynamic analysis calculations. The second vector generation unit 105 may further generate second mapping images for each temperature, using the generated second vectors to visualize the distribution of physical quantities related to the metal body. Generating such second mapping images makes it possible to grasp the generated second vectors as images, making it easier to understand the generated second vectors. The following description will be given taking as an example a case where the second vector generation unit 105 further generates a second mapping image.

[0043] FIG. 7 is a block diagram showing an example of the configuration of the second vector generating unit 105 according to this embodiment. The second vector generation unit 105 having the above-described functions includes a thermodynamic analysis calculation unit 121 and a vector generation unit 123, as shown in FIG.

[0044] The thermodynamic analysis calculation unit 121 is realized by, for example, a CPU, a ROM, a RAM, etc. The thermodynamic analysis calculation unit 121 performs thermodynamic analysis calculations for a metal body of interest at multiple temperatures based on a predetermined physical model, using the content data transmitted from the content data acquisition unit 101. Through such thermodynamic analysis calculations, calculation results for physical quantities related to the metal body of interest can be obtained for each measurement unit of the content data for the metal body of interest.

[0045] Here, any physical quantity related to a metal body whose value changes with temperature can be used as the physical quantity obtained by such thermodynamic analysis calculations. Examples of such physical quantities include the formation energy and Gibbs energy of phases that can be generated in the metal body, and the phase fractions of phases that can be generated in the metal body.

[0046] Of the various calculation results described above, the thermodynamic analysis calculation unit 121 preferably calculates at least the calculation results relating to the phase fractions of the phases constituting the metal body, because such phase fractions are particularly useful as knowledge about the structure constituting the metal body.

[0047] The temperature at which the thermodynamic analysis calculations are performed is not particularly limited, and the calculations should cover at least the temperature range that may be used in the heat treatment performed to manufacture the metal body of interest. When the temperature range for the thermodynamic analysis calculations is set to Ta [°C] to Tb [°C], the interval ΔT at which the calculations are performed is also not particularly limited, and may be set appropriately depending on the accuracy required for the final heat treatment temperature and the computational resources implemented in the information processing device, such as a computer, used. The thermodynamic analysis calculation unit 121 may perform the thermodynamic analysis calculations by setting the temperature at, for example, ΔT = 1°C within the range of Ta [°C] to Tb [°C], or by setting ΔT = approximately 10°C, or by setting ΔT = approximately 0.1°C. However, since it is believed that the smaller the interval ΔT is set, the more accurate the estimation of the final heat treatment temperature will be, it is preferable to set the interval ΔT as small as possible.

[0048] An example of a physical model used in the above-described thermodynamic analysis calculations is a physical model based on the CALPHAD (Computer Coupling of Phase Diagrams and Thermochemistry) method. In addition to the CALPHAD method, other calculation methods can also be used as long as they are capable of analyzing information related to the above-described physical quantities.

[0049] When the thermodynamic analysis calculation unit 121 finishes the thermodynamic analysis calculation based on the acquired content data, it outputs data relating to the obtained calculation results to the downstream vector generation unit 123. Furthermore, the thermodynamic analysis calculation unit 121 may store the obtained data relating to the calculation results as history information in the storage unit 113 after associating it with time information relating to the date and time when the data was generated.

[0050] In the above explanation, the thermodynamic analysis calculation unit 121 is implemented in one heat treatment temperature estimation device 10 as an example, but the thermodynamic analysis calculation unit 121 may also be implemented in various computers, servers, etc. that are interconnected with the heat treatment temperature estimation device 10 via a network, etc.

[0051] The vector generation unit 123 is realized by, for example, a CPU, a ROM, a RAM, etc. The vector generation unit 123 generates, for each temperature at which the calculation results are available, second vectors representing the distribution of physical quantities related to the metal body from the calculation results of the thermodynamic analysis calculation by the thermodynamic analysis calculation unit 121. The vector generation unit 123 also uses the generated second vectors to generate, for each temperature at which the calculation results are available, second mapping images visualizing the distribution of physical quantities related to the metal body. As a result, for the physical quantity of interest, a vector group consisting of a plurality of second vectors representing the distribution at various temperatures and an image group consisting of a plurality of second mapping images visualizing the distribution at various temperatures are generated.

[0052] More specifically, the vector generation unit 123 generates a second vector by arranging the obtained calculation results, for example, for each measurement unit of the content data. The vector generation unit 123 also generates a second mapping image by converting the numerical values ​​of the calculation results to be visualized into pixel values. For example, the vector generation unit 123 equally divides the range in which the numerical values ​​representing a certain calculation result are distributed according to the number of gradations of the second mapping image to be generated (e.g., 256 for 8 bits), and converts the numerical values ​​of the calculation results into pixel values. By performing such processing, the distribution of the numerical values ​​of the calculation results can be visualized as an image. Since the second mapping image is obtained by converting the numerical range indicated by the calculation result of interest into pixel values, it is preferable that the second mapping image be a grayscale image.

[0053] For example, consider visualizing the distribution of phase fractions for phases that may exist in a metal body of interest. In this case, the vector generation unit 123 equally divides the range of possible phase fraction values, from 0 to 100%, according to the number of gradations. Then, for each region of the metal body indicated by the calculation results, the specific phase fraction values ​​are converted into pixel values. By performing this conversion process, the vector generation unit 123 can generate a second mapping image that visualizes the distribution of phase fractions.

[0054] Furthermore, when visualizing the phase fraction, it is preferable to calibrate the division of the phase fraction range. This allows for more accurate visualization of the phase fraction distribution. For example, such calibration can be performed by testing the visualization range in a system for which the correct solution is known and determining the range. A preferable visualization range can be determined in advance by performing a verification to determine in advance what value outside the minimum and maximum values ​​approaches the correct temperature. Furthermore, instead of equally dividing the range from 0 to 100%, it is also possible to perform processing such as equally dividing the range between the minimum and maximum values ​​of the phase fraction in the calculation result, dividing the range from 0 to 100% without a constant division width, or dividing the range between the minimum and maximum values ​​of the phase fraction in the calculation result without a constant division width.

[0055] Figure 8 shows a part of the second mapping image group that visualizes the distribution of the fcc phase fraction, calculated by the CALPHAD method based on the content data obtained from the EPMA analysis results shown in Figure 3. In Figure 8, the blacker the color of each pixel, the higher the phase fraction, and the whiter the color.

[0056] Such second vectors and second mapping images may be generated for some or all of the calculation results calculated by the thermodynamic analysis calculation unit 121. In addition, although the above describes the case where a grayscale image with 8-bit gradation is generated, the number of gradations of the grayscale image is not limited to the above example.

[0057] In addition, since the first vector and first mapping image generated by the first vector generation unit 103 and the second vector and second mapping image generated by the vector generation unit 123 are vectors and images generated using the same content data as a starting point, the number of vector components and image size (the number of pixels that make up the image and the size of the image) are equal to each other.

[0058] After generating the second vector and the second mapping image as described above, the second vector generation unit 105 outputs the data of the generated second vector and the second mapping image to the estimation unit 107, which will be described later. Furthermore, the second vector generation unit 105 may store the data of the second vector and the second mapping image generated in this manner in the storage unit 113 as history information after associating the data with time information regarding the date and time when the data was generated.

[0059] The estimation unit 107 is realized by, for example, a CPU, a ROM, a RAM, etc. The estimation unit 107 calculates the similarity between the first vector and the first mapping image generated by the first vector generation unit 103 and the second vector and the second mapping image at each temperature generated by the second vector generation unit 105 by a similarity analysis process, and estimates the final heat treatment temperature based on the obtained similarity.

[0060] Conventionally, similarity used to determine the degree of similarity between two vectors or two images has often been used to quantify the degree of similarity between vectors or images of the same type (vectors or images known to be somewhat similar). However, the estimation unit 107 according to this embodiment deliberately calculates similarity between completely different types of vectors or images, namely, vectors or images (first vectors, first mapping images) relating to the distribution state of "element content" and vectors or images (second vectors, second mapping images) relating to the distribution state of "physical quantities of a metal body at a certain temperature." In this case, a specific numerical value representing the similarity is calculated for a combination of a first vector or first mapping image and one second vector or second mapping image.

[0061] Here, when calculating the similarity between a first vector or a first mapping image for a certain element of the alloy components and a second vector or a second mapping image for a certain physical quantity at each temperature, the estimation unit 107 preferably uses, as the first vector or the first mapping image, a vector or a mapping image for an element related to the physical quantity of interest in the second vector or the second mapping image. This makes it possible to associate different types of vectors or images, such as the first vector and the second vector or the first mapping image and the second mapping image, in the physical sense of "elements related to the physical quantity of interest," thereby further enhancing the significance of calculating the similarity.

[0062] For example, when the second vector or the second mapping image focuses on the phase fraction of a certain phase at each temperature, it is preferable that the element whose distribution state is expressed in the first vector or the first mapping image is a “former of the phase of interest.” For example, when the phase fraction of the fcc phase illustrated in Fig. 8 is focused on, it is preferable that the first mapping image of Mn, which is a former of the fcc phase, as illustrated in Fig. 6, be used as the target for calculating the similarity.

[0063] In addition to the above examples, examples of combinations with a high correlation between elements that can be contained in steel materials and physical quantities of interest include the following. Regarding ferrite-stabilizing elements and austenite-stabilizing elements, reference was made to Ishida Kiyohito and Nishizawa Taiji, Journal of the Japan Institute of Metals, 36, 270 (1972). Regarding elements that dissolve in cementite and elements that easily form unique carbides, reference was made to Table 5 in Kojima Akihiko, Journal of the Japan Welding Society, 77, 33 (2008). Regarding nitrides and borides, reference was made to Ishiguro Yasuhide, "Study on Precipitation Behavior in Steel Materials," Doctoral Dissertation, Nagoya University (No. 8507) (2009).

[0064] [Table 1]

[0065] In addition to steel materials, the following are examples of combinations with a high correlation between the elements that make up non-ferrous metals and physical quantities of interest. For Ti alloys, we referred to Kimura Keizo, Iron and Steel, 72, 113 (1986). For Al alloys, we referred to Masahashi Naoya, "Aluminum Fundamentals," Manufacturing Basics Lecture Series (30th Technical Seminar) (2012). For Mg alloys, we referred to Kamado Shigeharu, "Development of Heat-Treated Wrought Magnesium Alloys with Excellent Workability," ALCA New Technology Briefing (2018).

[0066] [Table 2]

[0067] The estimation unit 107 can use various types of similarity analysis processing when calculating the similarity, as long as it is an analysis processing that can calculate the similarity between vectors or images. Examples of such similarity analysis processing include various types of processing that use functions corresponding to the similarity, such as those shown in Table 3 below. Such functions can be used by using various numerical calculation applications, such as Mathematica.

[0068] [Table 3]

[0069] In selecting the above-mentioned function, it is sufficient to carry out a verification in advance according to the metal body to be applied, and select an appropriate one from the obtained results. In some cases, the method for calculating the similarity may be determined by a calibration operation such as averaging the final heat treatment temperatures estimated from multiple functions.

[0070] Furthermore, the estimation unit 107 may use a similarity analysis process that utilizes deep learning when calculating the similarity. Deep learning is a type of machine learning method in which a neural network is constructed in multiple layers. In this embodiment, too, the use of deep learning is effective in combination with metric learning.

[0071] Metric learning is a method for learning a feature space that takes into account the relationships between data, controlling which relationships to emphasize for information retrieval, clustering, individual recognition, etc. Deep metric learning, which combines deep learning and lightweight learning, is a method for training a deep neural network (DNN) model so that the "distance" between two feature vectors reflects the "similarity" of the data.

[0072] For example, a DNN model can be trained so that the distance between feature vectors obtained from samples belonging to the same class (i.e., similar) is small, while the distance between feature vectors obtained from samples belonging to different classes (i.e., dissimilar) is large. Here, the "distance" used refers to a metric defined between two vectors, such as Euclidean distance or cosine similarity shown in Table 3. Because a DNN model is a model that learns the "relationship between two feature vectors" rather than the "feature vectors" themselves, it has the advantage of being able to handle feature vectors that do not exist in the training data. Therefore, applying this method makes it possible to obtain a more reliable similarity function.

[0073] By using the above-described method, the estimation unit 107 calculates the similarity between the first vector and each of the second vectors included in the second vector group, or between the first mapping image and each of the second mapping images included in the second mapping image group, and thereby it is possible to obtain a curve showing the relationship between the temperature (range: Ta to Tb) corresponding to the second vector or the second mapping image and the similarity, as schematically shown in FIG. 9, for example.

[0074] The estimation unit 107 identifies the temperature that gives the maximum or minimum value of the similarity in a plane defined by the temperature and the similarity as shown in Fig. 9, and estimates the final heat treatment temperature based on the temperature that gives this maximum or minimum value. For example, in the example shown in Fig. 9, the higher the similarity value, the more similar the first vector or the first mapping image is to the first vector or the second mapping image at a certain temperature. Therefore, the estimation unit 107 estimates the temperature Tf that gives the maximum value as the final heat treatment temperature.

[0075] Fig. 10 shows the results of calculating each function shown in Table 3 above for the first mapping image shown in Fig. 6 and the second mapping image group shown in Fig. 8. All of the functions shown in Table 3 are functions whose calculated values ​​decrease as the similarity increases, so the temperature Tf that gives the minimum value is estimated to be the final heat treatment temperature.

[0076] As shown in Figure 10, most of the functions shown in Table 3 show minimum values ​​for similarity at roughly the same temperature. Furthermore, some functions have a minimum value rather than a minimum value, and some functions do not allow for a specified minimum or minimum value. For this reason, it is advisable to conduct verification in advance and use a function appropriate for the metal object of interest.

[0077] Furthermore, instead of estimating the final heat treatment temperature from a combination of a first vector or a first mapping image related to a certain element and a second vector or a second mapping image related to a certain physical quantity, as described above, the estimation unit 107 may estimate the final heat treatment temperature using multiple combinations of first vectors or first mapping images related to different elements and second vectors or second mapping images related to the physical quantities corresponding to those elements.

[0078] For example, information about three elements, C, Si, and Mn, can be obtained from the content data based on the EPMA shown in Fig. 3. Therefore, the estimation unit 107 may estimate the final heat treatment temperature using two or more of the following three combinations: "a combination of a first vector or a first mapping image for C and a second vector or a second mapping image for the cementite phase fraction," "a combination of a first vector or a first mapping image for Si and a second vector or a second mapping image for the bcc phase fraction," and "a combination of a first vector or a first mapping image for Mn and a second vector or a second mapping image for the fcc phase fraction."

[0079] In this case, the estimation unit 107 identifies the temperature that gives the maximum or minimum value of similarity for each of a plurality of combinations of first vectors or first mapping images for different elements and corresponding second vectors or second mapping images. Then, the final heat treatment temperature can be estimated based on the temperatures that give the obtained maximum or minimum values. Specifically, the estimation unit 107 can determine the minimum, maximum, or average value of the obtained temperatures as the final heat treatment temperature.

[0080] After estimating the final heat treatment temperature for the metal body of interest as described above, the estimation unit 107 outputs the obtained data of the estimation result to the output control unit 109, which will be described later. Furthermore, the estimation unit 107 may associate the obtained data of the estimation result with time information regarding the date and time when the data was generated, and store the data as history information in the storage unit 113.

[0081] Returning to FIG. 4, the output control unit 109 will be described. The output control unit 109 is realized by, for example, a CPU, a ROM, a RAM, an output device, a communication device, etc. The output control unit 109 outputs information about the final heat treatment temperature of the metal body of interest output from the estimation unit 107 to a user of the heat treatment temperature estimation device 10. Specifically, the output control unit 109 associates data about the estimation result output from the estimation unit 107 with time data about the date and time when the data was generated, and outputs the data to various servers or control devices, or outputs the data on paper media using an output device such as a printer. The output control unit 109 may also output the data about the estimation result to various information processing devices such as external computers or various recording media.

[0082] Furthermore, the output control unit 109 can output data relating to the estimation result by the estimation unit 107 to the display control unit 111, which will be described later.

[0083] The display control unit 111 is realized by, for example, a CPU, a ROM, a RAM, an output device, a communication device, etc. The display control unit 111 controls the display of the estimation result output from the output control unit 109 on an output device such as a display provided in the heat treatment temperature estimation device 10 or on an output device provided external to the heat treatment temperature estimation device 10. This allows the user of the heat treatment temperature estimation device 10 to immediately grasp the estimation result of the final heat treatment temperature of the metal body of interest.

[0084] The memory unit 113 is an example of a memory device included in the heat treatment temperature estimation device 10, and is realized by, for example, a ROM, a RAM, a storage device, etc. This memory unit 113 appropriately records various parameters and intermediate processing progress (e.g., various pre-stored data, databases, programs, etc.) that need to be saved when the heat treatment temperature estimation device 10 according to this embodiment performs some processing. This memory unit 113 allows the content data acquisition unit 101, the first vector generation unit 103, the second vector generation unit 105, the estimation unit 107, the output control unit 109, the display control unit 111, etc. to freely read / write data.

[0085] The heat treatment temperature estimation device 10 according to this embodiment has been described in detail above. In the above description, an example has been given in which each processing unit having the above-described functions is implemented within a single device. However, one or more of the processing units described above may be distributed and implemented in multiple devices, such as various computers, connected via a network. In this case, the system consisting of the multiple devices operates in cooperation with each other, thereby realizing the functions of the heat treatment temperature estimation device 10 as a whole. In other words, the functions of the heat treatment temperature estimation device 10 according to this embodiment may be distributed and implemented in multiple computers, etc., and exist in the form of a heat treatment temperature estimation system.

[0086] The above describes an example of the functions of the heat treatment temperature estimation device 10 according to this embodiment. Each of the above components may be configured using general-purpose components and circuits, or may be configured using hardware specialized for the function of each component. Furthermore, the functions of each component may all be performed by a CPU or the like. Therefore, the configuration used can be changed as appropriate depending on the technical level at the time of implementing this embodiment.

[0087] It is possible to create a computer program for implementing each function of the heat treatment temperature estimation device according to the present embodiment as described above and install it on a personal computer or the like. A computer-readable recording medium storing such a computer program can also be provided. Examples of the recording medium include a magnetic disk, an optical disk, a magneto-optical disk, and a flash memory. The computer program may also be distributed, for example, via a network without using a recording medium.

[0088] (Heat treatment temperature estimation method) Next, an example of the flow of a heat treatment temperature estimation method using the heat treatment temperature estimation device 10 according to this embodiment will be briefly described with reference to Fig. 11. Fig. 11 is a flow chart showing an example of the flow of a heat treatment temperature estimation method according to this embodiment.

[0089] In the heat treatment temperature estimation method according to this embodiment, first, the content data acquisition unit 101 acquires data (content data) on the content of alloy components of a metal body of interest (step S101). The content data acquisition unit 101 outputs the acquired content data to the first vector generation unit 103 and the second vector generation unit 105, respectively.

[0090] The first vector generation unit 103 generates a first vector and a first mapping image for at least one element of the alloy components based on the transmitted content data (step S103). Thereafter, the first vector generation unit 103 outputs the generated first vector and first mapping image to the estimation unit 107.

[0091] Furthermore, the second vector generation unit 105 generates a second vector and a second mapping image for at least one of the physical quantities of the metal body based on the transmitted content data (step S105). More specifically, the thermodynamic analysis calculation unit 121 of the second vector generation unit 105 performs thermodynamic analysis calculations at multiple temperatures for the metal body of interest based on the transmitted content data. Thereafter, the vector generation unit 123 uses the obtained thermodynamic calculation results to generate multiple second vectors and second mapping images for at least one of the physical quantities of the metal body. Thereafter, the vector generation unit 123 outputs the generated second vectors and second mapping images to the estimation unit 107.

[0092] The estimation unit 107 estimates the final heat treatment temperature based on the first vector and first mapping image generated by the first vector generation unit 103 and the second vector and second mapping image at each temperature generated by the second vector generation unit 105 (step S107). More specifically, the estimation unit 107 calculates the similarity between the first vector or first mapping image and each second vector or each second mapping image, and estimates the final heat treatment temperature for the metal body of interest based on the temperature that gives the maximum or minimum value of the similarity. After estimating the final heat treatment temperature for the metal body of interest, the estimation unit 107 outputs the obtained estimation result to the output control unit 109.

[0093] The output control unit 109 outputs the final heat treatment temperature for the metal body estimated by the estimation unit 107 (step S109). This allows a user of the heat treatment temperature estimation method to understand the estimated result of the final heat treatment temperature for the metal body of interest.

[0094] An example of the flow of the heat treatment temperature estimation method according to this embodiment has been briefly described above.

[0095] In this embodiment, the distribution of elements contained in the alloy components and the physical quantities related to the metal body are mainly represented (i.e., visualized) as images to be used as the first mapping image and the second mapping image, respectively. However, the distribution of elements contained in the alloy components and the physical quantities related to the metal body do not necessarily need to be represented as images, and may be represented as general vectors as described above to be used as the first vector and the second vector, respectively.

[0096] The first vector may be, for example, a vector consisting of the content of elements contained in the alloy components as described above, or a vector consisting of quantities other than pixel values ​​converted from the content of elements contained in the alloy components. The second vector may be, for example, a vector consisting of physical quantities related to the metal body as described above, or a vector consisting of quantities other than pixel values ​​converted from physical quantities related to the metal body. Without the requirement of pixel values, for example, there is no constraint such as having a finite gradation. Even if a general vector is used, the similarity can be calculated using the similarity analysis process described in this embodiment, and the final heat treatment temperature can be estimated.

[0097] <Summary> As described above, the heat treatment temperature estimation method and heat treatment temperature estimation method according to this embodiment make it possible to estimate the heat treatment temperature of a metal body (more specifically, the final heat treatment temperature that characterizes the metal body) from only the content data of the alloy components of the metal body of interest. This makes it possible to provide a technology that can estimate the final heat treatment temperature of a metal body of interest as a contraction index that characterizes its structure. This makes it possible to identify the manufacturing lot of a metal body product with an estimation accuracy of about 50°C, as shown in the following examples, thereby improving user convenience.

[0098] Second Embodiment Next, a heat treatment temperature estimating device according to a second embodiment of the present invention will be described in detail. As explained above, the heat treatment temperature estimation device according to the first embodiment is a device that estimates the heat treatment temperature of the last heat treatment performed on a metal body made of a predetermined alloy composition and manufactured by heat treatment. The heat treatment temperature estimation device according to the second embodiment, which will be explained below, is a device that uses the estimated heat treatment temperature (final heat treatment temperature) to further estimate the heat treatment time of the last heat treatment performed on the metal body.

[0099] FIG. 12 is an explanatory diagram for explaining the heat treatment time estimation process performed by the heat treatment temperature estimation device according to this embodiment. In the heat treatment performed on a metal body, the metal body is heated to a desired final heat treatment temperature, held at the final heat treatment temperature for a desired time, and then cooled to the desired temperature, as shown schematically in Figure 12. This time during which the final heat treatment temperature is held will be referred to as the "final heat treatment time" below.

[0100] (About the heat treatment temperature estimation device) <Configuration of the heat treatment temperature estimation device> Next, the configuration of the heat treatment temperature estimation device according to this embodiment will be described in detail with reference to Fig. 13 and Fig. 14. Fig. 13 is a block diagram that schematically shows the overall configuration of the heat treatment temperature estimation device according to this embodiment, and Fig. 14 is an explanatory diagram for explaining the heat treatment time estimation process according to this embodiment.

[0101] As shown in FIG. 13, the heat treatment temperature estimation device 10A of this embodiment includes a content data acquisition unit 101, a first vector generation unit 103, a second vector generation unit 105, an estimation unit 107, an output control unit 109, a display control unit 111, a memory unit 113, a heat treatment time change function derivation unit 151, a heat treatment time dependent quantity data acquisition unit 153, a final heat treatment time estimation unit 155, and a database creation unit 157.

[0102] Here, the content data acquisition unit 101, the first vector generation unit 103, the second vector generation unit 105, the estimation unit 107, the output control unit 109, the display control unit 111, and the memory unit 113 have the same configuration as each processing unit in the heat treatment temperature estimation device 10 according to the first embodiment and have the same effects, so detailed explanations thereof will be omitted below.

[0103] The heat treatment time change function derivation unit 151 is realized by, for example, a CPU, a ROM, a RAM, an input device, a communication device, etc. Hereinafter, an observable quantity or physical quantity related to the metal body that changes depending on the length of heat treatment time in heat treatment performed at the final heat treatment temperature estimated by the estimation unit 107 will be referred to as a "heat treatment time dependent quantity." The heat treatment time change function derivation unit 151 derives a heat treatment time change function that represents the time change of such heat treatment time dependent quantity.

[0104] During heat treatment of metals, including steel, various substances escape from the metal and, in some cases, enter the metal. During this heat treatment, various substances (e.g., atoms, molecules, etc.) and various forms of energy, such as thermal energy, are exchanged between the metal being heat treated and the outside world. Therefore, the system of consideration when considering heat treatment of metals can be said to be what is known as an "open system" in statistical thermodynamics.

[0105] During heat treatment, various substances may escape from or penetrate into the metal structure that constitutes the metal body, depending on the reaction that occurs in the metal body. Therefore, when considering changes in the metal structure over time in an open system, for example, when considering observable quantities such as the content of alloying elements, the objects to be considered are too case-by-case, making it inconvenient to treat them in a unified manner.

[0106] Therefore, the heat treatment temperature estimation device 10A according to this embodiment introduces a scientific index that can be handled uniformly regardless of the reaction occurring in the metal body. The scientific index is then treated as a quantity dependent on the heat treatment time. With such a scientific index that can be handled uniformly, it becomes possible to estimate the heat treatment time by tracking the change in the index over time.

[0107] As an index that can be handled in a unified manner regardless of the reaction occurring in the metal body, it is conceivable to focus on various physical quantities related to energy. As a result of extensive research by the present inventors into such an index, they came up with the idea of ​​focusing on the mean free energy of the metal structure.

[0108] Now, let us consider the heat treatment time t as a variable and define the physical quantity that expresses the mean free energy of the metal structure (i.e., the heat treatment time dependent quantity of interest) as V P (t). At this time, the physical quantity V that expresses the average free energy is P (t) is defined as the following equation (101).

[0109]

number

[0110] Here, the first term on the right side of the above formula (101) is the equilibrium free energy of the metal body's structure at temperature T, calculated by thermodynamic analysis calculations such as the CALPHAD method based on the content data of the metal body, with the heat treatment time t as a variable. Also, the second term on the right side of the above formula (101) is the free energy of the reference state of the metal body's structure at temperature T, calculated by thermodynamic analysis calculations such as the CALPHAD method based on the content data of the metal body, with the heat treatment time t as a variable. In other words, the right side of the above formula (101) is the relative difference in free energy (ΔE Eq T (t)).

[0111] Here, the reference state in the structure of the metal body at temperature T may be set appropriately taking into consideration the characteristics of the metal body in question, but it is preferable to set it to the physical quantity focused on when generating the second vector in the second vector generation unit 105, for example, as shown in Tables 1 and 2 above.

[0112] Also, indicator V P The free energy values ​​used in calculating (t) may be calculated using content data of the metal body for which the heat treatment time is to be estimated, or content data of a metal body similar to the metal body for which the heat treatment time is to be estimated, and the final heat treatment temperature estimated by the estimation unit 107, using analytical calculation results obtained from thermodynamic analytical calculations such as the CALPHAD method or analytical calculations based on physical models such as the Phase-Field method.

[0113] Alternatively, a plurality of metal body samples may be actually produced by changing the heat treatment temperature using an alloy material having alloy components corresponding to the metal body for which the heat treatment time is to be estimated or a metal body similar to the metal body for which the heat treatment time is to be estimated. Then, the content of each metal body sample is measured by EPMA, and the obtained EPMA measurement results (i.e., content data, which are data showing the distribution of the content) are used to calculate the index V by thermodynamic analysis calculations such as the CALPHAD method. P The value of each free energy used in calculating (t) may be calculated.

[0114] The heat treatment time dependent quantities obtained as described above at different times are plotted on a coordinate plane with the value of the heat treatment time dependent quantity on the vertical axis and the heat treatment time on the horizontal axis, thereby obtaining a graph as shown schematically in FIG. 14.

[0115] Here, the above average free energy V P The calculation process of (t) may be performed by the thermodynamic analysis calculation unit 121 included in the second vector generation unit 105, or may be performed separately by an information processing device such as a computer provided outside the heat treatment temperature estimation device 10A.

[0116] As described above, the heat treatment time change function derivation unit 151 calculates V P After obtaining the value of (t), the above formula (101) is fitted with a function form expressed by a linear combination of one or more terms shown on the right side of the following formula (102), and the fitting result is used as the heat treatment time change function. In the following formula (102), t is a variable representing the heat treatment time, and V P 0 , A Pi , τ Pi are coefficients, and i is an integer equal to or greater than 1.

[0117]

number

[0118] In statistical thermodynamics, the process of starting from a state where an external force is applied, dispersing and randomizing the broad-sense strain in the system under consideration throughout the entire population, and then being discharged outside the system, until a certain equilibrium is reached, is called a "relaxation phenomenon." The time transition of free energy, which is the focus of this embodiment, is also a typical relaxation phenomenon. The reason why the second term on the right side of the above equation (102) uses a function form expressed as a linear combination of multiple terms is that, since the formation process of a metal structure is generally a process in which multiple reaction processes can be involved, multiple relaxation times τ corresponding to each reaction process are used. Pi In equation (102), the number of terms i in the linear combination may be set appropriately depending on the number of heat treatment time-dependent quantities used in the fitting, the fitting results, etc. Such fitting processing can be performed, for example, by implementing a nonlinear least-squares method using a commercially available numerical calculation application.

[0119] After deriving the heat treatment time change function as described above, the heat treatment time change function derivation unit 151 outputs the derived heat treatment time change function to the final heat treatment time estimation unit 155 and the database creation unit 157. Furthermore, the heat treatment time change function derivation unit 151 may associate data on the derived heat treatment time change function with time information on the date and time when the data was generated, and store the data as history information in the storage unit 113. When the heat treatment time change function derivation unit 151 outputs the derived heat treatment time change function, it is preferable to associate attribute information that characterizes the heat treatment time change function, such as content data used in calculating the heat treatment time dependent amount, with the function.

[0120] The heat treatment time dependent quantity data acquisition unit 153 is realized by, for example, a CPU, a ROM, a RAM, an input device, a communication device, etc. The heat treatment time dependent quantity data acquisition unit 153 acquires heat treatment time dependent quantity data, which is data related to the heat treatment time dependent quantity of the metal body of interest, from inside or outside the heat treatment temperature estimation device 10A.

[0121] For example, as a heat treatment time dependent quantity, the mean free energy V of the metal structure as explained earlier is P When focusing on (t), the heat treatment time dependent quantity data acquisition unit 153 calculates the mean free energy V P Data on the specific value of (t) is obtained from the thermodynamic analysis calculation unit 121 included in the second vector generation unit 105, or from various information processing devices such as computers provided outside the heat treatment temperature estimation device 10A.

[0122] When the heat treatment time dependent quantity data acquisition unit 153 acquires the heat treatment time dependent quantity data of the metal body of interest in this manner, it outputs the acquired heat treatment time dependent quantity data to the final heat treatment time estimation unit 155, which will be described later. In addition, the heat treatment time dependent quantity data acquisition unit 153 may associate the acquired heat treatment time dependent quantity data with time information regarding the date and time when the data was generated, and store the data as history information in the storage unit 113.

[0123] The final heat treatment time estimation unit 155 is realized by, for example, a CPU, a ROM, a RAM, etc. The final heat treatment time estimation unit 155 estimates a final heat treatment time, which is the heat treatment time at the final heat treatment temperature of the metal body of interest, based on the heat treatment time dependent quantities of the metal body acquired by the heat treatment time dependent quantity data acquisition unit 153 and the heat treatment time change function derived by the heat treatment time change function derivation unit 151.

[0124] The heat treatment time change function derived by the heat treatment time change function derivation unit 151 can be used as a so-called calibration curve when estimating the final heat treatment time, as schematically shown in Fig. 14. Therefore, the final heat treatment time estimation unit 155 uses the heat treatment time dependence of the metal body and the heat treatment time change function to estimate the heat treatment time t that gives a value corresponding to the heat treatment time dependence of the metal body. Then, the final heat treatment time estimation unit 155 treats the heat treatment time t estimated in this way as the final heat treatment time.

[0125] After estimating the final heat treatment time as described above, the final heat treatment time estimation unit 155 outputs the obtained data on the final heat treatment time to the output control unit 109. Furthermore, the final heat treatment time estimation unit 155 may store the obtained data on the final heat treatment time as history information in the storage unit 113 after associating it with time information on the date and time when the data was generated.

[0126] Furthermore, if the heat treatment time change function derived by the heat treatment time change function derivation unit 151 is compiled into a database by the database creation unit 157 (described later), the final heat treatment time estimation unit 155 may refer to the created database to identify a heat treatment time change function suitable for the metal body of interest. Then, the final heat treatment time estimation unit 155 estimates the final heat treatment time using the identified heat treatment time change function.

[0127] The database creation unit 157 is realized by, for example, a CPU, a ROM, a RAM, etc. The database creation unit 157 creates a database of the heat treatment time change functions derived by the heat treatment time change function derivation unit 151 and stores the created database in, for example, the storage unit 113. In this case, it is preferable that the database creation unit 157 creates the database after associating attribute information that characterizes the heat treatment time change functions, such as the types of alloy components contained in the metal body and detailed content data. This enables the final heat treatment time estimation unit 155 to search for a desired heat treatment time change function based on such attribute information.

[0128] The heat treatment temperature estimation device 10A according to this embodiment has been described in detail above. In the above description, the processing units having the above-described functions are implemented in a single device. However, one or more of the processing units may be distributed and implemented in multiple devices, such as various computers, connected via a network. In this case, the system consisting of the multiple devices operates in cooperation with each other, thereby realizing the functions of the heat treatment temperature estimation device 10A as a whole. In other words, the functions of the heat treatment temperature estimation device 10A according to this embodiment may be distributed and implemented in multiple computers, etc., to exist in the form of a heat treatment temperature estimation system.

[0129] The above describes an example of the functions of the heat treatment temperature estimation device 10A according to this embodiment. Each of the above components may be configured using general-purpose components and circuits, or may be configured using hardware specialized for the function of each component. Furthermore, the functions of each component may all be performed by a CPU or the like. Therefore, the configuration used can be changed as appropriate depending on the technical level at the time of implementing this embodiment.

[0130] It is possible to create a computer program for implementing each function of the heat treatment temperature estimation device according to the present embodiment as described above and install it on a personal computer or the like. A computer-readable recording medium storing such a computer program can also be provided. Examples of the recording medium include a magnetic disk, an optical disk, a magneto-optical disk, and a flash memory. The computer program may also be distributed, for example, via a network without using a recording medium.

[0131] <Example of final heat treatment time estimation process> An example of the above-described final heat treatment time estimation process will be specifically described below with reference to FIGS.

[0132] Fig. 15 is an explanatory diagram for explaining an example of heat treatment performed on a metal body sample. In the following specific example, seamless steel pipes were produced by the Mannesmann process using a billet whose average alloy composition was 0.3 mass% C-5.0 mass% Mn-Fe and a billet whose average alloy composition was 0.2 mass% C-1.5 mass% Si-1.5 mass% Mn-Fe. In producing such seamless steel pipes, the resulting steel pipes were heated from room temperature to a holding temperature at an average heating rate of 10°C / s, and then held at the holding temperature for 100 seconds, 1000 seconds, 20 hours, or 336 hours, followed by water cooling. All of these heat treatments were performed in an Ar atmosphere to suppress decarburization reactions during short-term heat treatments. The holding temperature was 680°C for the 0.3 mass% C-5.0 mass% Mn-Fe billet, and 780°C for the 0.2 mass% C-1.5 mass% Si-1.5 mass% Mn-Fe billet.

[0133] Steel materials containing these alloying elements have a metallurgical structure in which the fcc phase is the main phase. In this example, the alloying elements (C, Si, Mn) of the obtained seamless steel pipes were analyzed using an EPMA (JXA-8530F manufactured by JEOL Corporation). Among the obtained results, the distribution range of the Mn content data was 0 to 7.5 mass% for 0.3 mass% C-5.0 mass% Mn-Fe, and 0 to 2.5 mass% for 0.2 mass% C-1.5 mass% Si-1.5 mass% Mn-Fe. Figure 16 shows a first group of mapping images showing the distribution of the Mn content in each of the obtained metal body samples.

[0134] Using the content data as described above, the heat treatment temperature estimation device 10A according to this embodiment was used to estimate the heat treatment temperature in the heat treatment that was finally performed. In this example, a mapping image showing the distribution of Mn content was generated as the first mapping image, and a mapping image relating to the fcc phase fraction in the temperature range of 600 to 800°C was generated as the second mapping image (ΔT = 1°C). Using these mapping images, the final heat treatment temperature was estimated using the method for estimating the final heat treatment temperature according to this embodiment.

[0135] The obtained results are shown in Figure 17. Figure 17 is a diagram showing the estimated values ​​of the final heat treatment temperatures of the obtained metal body samples. As is clear from Figure 17, the final heat treatment temperatures estimated by the method for estimating the final heat treatment temperature according to this embodiment are very close to the actual holding temperature for each heat treatment time.

[0136] Using the data on each content obtained by EPMA and the final heat treatment temperature shown in Figure 17, the free energy of the metallographic structure at each measurement position of the EPMA was calculated for each heat treatment time using the CALPHAD method in accordance with the above formula (101), and the distribution of the energy values ​​was visualized. In this case, since the metal body sample having the alloy components of interest is a dual-phase steel in which the fcc phase and the bcc phase are expected, the fcc phase was selected as the reference state of the metallographic structure. A group of mapping images showing the distribution of the obtained free energy is shown in Figure 18. Figure 18 is a diagram showing a group of mapping images visualizing the distribution of the free energy of the metallographic structure in the obtained metal body sample. By calculating the average value of the free energy for each diagram in Figure 18, the average free energy of the metallographic structure, V P (t).

[0137] For each alloy component, the resulting average free energy V P The transition of the value of (t) (t=100 seconds, 1000 seconds, 20 hours, 336 hours) is shown in FIG. 19. FIG. 19 is a graph showing the time change of the average free energy in the obtained metal body sample. The vertical axis of FIG. 19 is the average free energy V P The horizontal axis in FIG. 19 represents the heat treatment time expressed in common logarithm.

[0138] A nonlinear least squares method was performed using a commercially available numerical calculation application on the transition of the mean free energy shown in Figure 19 to derive a function of change in heat treatment time. In this fitting, the number of terms on the right side of equation (102) was set to i = 1. In equation (102) where i = 1, there are three unknowns, and the calculated mean free energy V P Since the number of (t) is four, each coefficient in equation (102) can be determined. The results are shown in Figure 20, which shows the heat treatment time change function corresponding to the obtained metal body sample. Using the heat treatment time change function shown in Figure 20 and the value of the average free energy at each heat treatment time t, it is possible to estimate the heat treatment time t.

[0139] (Heat treatment temperature estimation method) Next, the flow of the heat treatment temperature estimation method performed by the heat treatment temperature estimation device 10A according to this embodiment will be briefly described with reference to Fig. 21. Fig. 21 is a flow chart showing an example of the flow of the heat treatment temperature estimation method according to this embodiment.

[0140] In the heat treatment temperature estimation method according to this embodiment, first, the first vector generation unit 103, the second vector generation unit 105, and the estimation unit 107 cooperate with each other to estimate the final heat treatment temperature of the metal body of interest in accordance with the heat treatment temperature estimation method described in the first embodiment (step S201). The obtained estimation result of the final heat treatment temperature is output to the heat treatment time change function derivation unit 151.

[0141] The heat treatment time change function deriving unit 151 derives a heat treatment time change function using the acquired final heat treatment temperature and the content data used to estimate the final heat treatment temperature (step S203). Thereafter, the heat treatment time change function deriving unit 151 outputs the obtained heat treatment time change function to the final heat treatment time estimation unit 155.

[0142] In addition, the heat treatment time dependent quantity data acquisition unit 153 acquires heat treatment time dependent quantity data for the metal body of interest at multiple different heat treatment times, such as the mean free energy of the metal structure (step S205), and outputs the acquired data to the final heat treatment time estimation unit 155.

[0143] The final heat treatment time estimation unit 155 estimates the final heat treatment time of the metal body of interest using the acquired heat treatment time dependent quantity data of the metal body of interest and the heat treatment time change function (step S207). Thereafter, the final heat treatment time estimation unit 155 outputs the estimated final heat treatment time to the output control unit 109.

[0144] The output control unit 109 outputs the final heat treatment time for the metal body estimated by the final heat treatment time estimation unit 155 (step S209). This allows the user of the heat treatment temperature estimation method to understand the estimated result of the final heat treatment time for the metal body of interest.

[0145] An example of the flow of the heat treatment temperature estimation method according to this embodiment has been briefly described above.

[0146] (summary) According to the heat treatment temperature estimation device and the heat treatment temperature estimation method of this embodiment, which also include the heat treatment time estimation method described above, the heat treatment temperature and the heat treatment time of a metal body can be estimated from only the content data of the alloy components of the metal body of interest. This makes it possible to provide a technology that can estimate the final heat treatment temperature and the final heat treatment time of the metal body of interest as contraction indices that characterize its structure.

[0147] (Hardware configuration of the heat treatment temperature estimation device) Next, the hardware configuration of the heat treatment temperature estimation device 10, 10A according to each embodiment of the present invention will be described in detail with reference to Fig. 22. Fig. 22 is a block diagram for explaining the hardware configuration of the heat treatment temperature estimation device 10, 10A according to the embodiment of the present invention.

[0148] The heat treatment temperature estimation device 10, 10A mainly includes a CPU 901, a ROM 903, and a RAM 905. The heat treatment temperature estimation device 10 further includes a bus 907, an input device 909, an output device 911, a storage device 913, a drive 915, a connection port 917, and a communication device 919.

[0149] The CPU 901 functions as a central processing unit and control unit, and controls all or part of the operation of the heat treatment temperature estimation device 10, 10A in accordance with various programs recorded in the ROM 903, the RAM 905, the storage device 913, or the removable recording medium 921. The ROM 903 stores programs and calculation parameters used by the CPU 901. The RAM 905 temporarily stores programs used by the CPU 901 and parameters that change as appropriate during program execution. These are interconnected by a bus 907, which is an internal bus such as a CPU bus.

[0150] The bus 907 is connected to an external bus such as a PCI (Peripheral Component Interconnect / Interface) bus via a bridge.

[0151] The input device 909 is an operating means operated by a user, such as a mouse, keyboard, touch panel, button, switch, or lever. The input device 909 may be, for example, a remote control means (so-called remote control) using infrared or other radio waves, or an externally connected device 923 such as a PDA that supports operation of the heat treatment temperature estimation device 10, 10A. The input device 909 may also include, for example, an input control circuit that generates an input signal based on information input by the user using the operating means and outputs the signal to the CPU 901. By operating the input device 909, the user can input various data and instruct processing operations to the heat treatment temperature estimation device 10.

[0152] The output device 911 is a device capable of visually or audibly notifying the user of acquired information. Examples of such devices include display devices such as CRT displays, liquid crystal displays, plasma displays, EL displays, and lamps; audio output devices such as speakers and headphones; printers, mobile phones, and facsimiles. The output device 911 outputs, for example, results obtained from various processes performed by the heat treatment temperature estimation device 10, 10A. Specifically, the display device displays the results obtained from various processes performed by the heat treatment temperature estimation device 10, 10A as text or images. On the other hand, the audio output device converts audio signals, such as reproduced voice data or acoustic data, into analog signals and outputs them.

[0153] The storage device 913 is a data storage device configured as an example of a storage unit of the heat treatment temperature estimation device 10, 10A. The storage device 913 is configured, for example, by a magnetic storage device such as a hard disk drive (HDD), a semiconductor storage device, an optical storage device, or a magneto-optical storage device. The storage device 913 stores programs and various data executed by the CPU 901, as well as various data acquired from the outside.

[0154] The drive 915 is a reader / writer for a recording medium, and is built into or externally attached to the heat treatment temperature estimation device 10, 10A. The drive 915 reads information recorded on a removable recording medium 921, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, and outputs the information to the RAM 905. The drive 915 can also write information to the removable recording medium 921, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory. The removable recording medium 921 may be, for example, a CD medium, a DVD medium, or a Blu-ray (registered trademark) medium. The removable recording medium 921 may also be, for example, a CompactFlash (registered trademark) card, a flash memory, or an SD (Secure Digital memory card) memory card. The removable recording medium 921 may also be, for example, an IC (Integrated Circuit Card) or an electronic device equipped with a contactless IC chip.

[0155] The connection port 917 is a port for directly connecting a device to the heat treatment temperature estimation device 10, 10A. Examples of the connection port 917 include a USB (Universal Serial Bus) port, an IEEE 1394 port, a SCSI (Small Computer System Interface) port, an RS-232C port, and an HDMI (High-Definition Multimedia Interface) port. By connecting an external device 923 to the connection port 917, the heat treatment temperature estimation device 10, 10A can directly obtain various data from the external device 923 or provide various data to the external device 923.

[0156] The communication device 919 is, for example, a communication interface configured with a communication device or the like for connecting to a communication network 925. The communication device 919 is, for example, a communication card for a wired or wireless LAN (Local Area Network), Bluetooth (registered trademark), or WUSB (Wireless USB). The communication device 919 may also be a router for optical communication, a router for ADSL (Asymmetric Digital Subscriber Line), or a modem for various types of communication. This communication device 919 can transmit and receive signals, for example, between the Internet and other communication devices in accordance with a predetermined protocol such as TCP / IP. The communication network 925 connected to the communication device 919 is configured with a network connected by wire or wirelessly, and may be, for example, the Internet, a home LAN, an in-house LAN, infrared communication, radio wave communication, satellite communication, or the like.

[0157] The above describes an example of a hardware configuration capable of realizing the functions of the heat treatment temperature estimation device 10, 10A according to the embodiment of the present invention. Each of the above components may be configured using general-purpose components, or may be configured using hardware specialized for the function of each component. Therefore, the hardware configuration used can be changed as appropriate depending on the technical level at the time of implementing this embodiment. [Example]

[0158] The heat treatment temperature estimation device and the heat treatment temperature estimation method according to the present invention will be specifically described below with reference to examples. Note that the examples shown below are merely examples of the heat treatment temperature estimation device and the heat treatment temperature estimation method according to the present invention, and the heat treatment temperature estimation device and the heat treatment temperature estimation method according to the present invention are not limited to the examples below.

[0159] Example 1 Hereinafter, examples corresponding to the heat treatment temperature estimation device and the heat treatment temperature estimation method according to the first embodiment will be specifically described. In this example, a seamless steel pipe was produced by the Mannesmann process using a billet with an average alloy composition of 0.2% by mass C-5.0% by mass Mn-2.0% by mass Si-Fe. During the production of this seamless steel pipe, the resulting steel pipe was subjected to a heat treatment at 800°C for 60 seconds, followed by a subsequent heat treatment at 700°C for 60 seconds.

[0160] The alloy composition of the obtained seamless steel pipe was analyzed using an EPMA (JXA-8530F manufactured by JEOL Corp.) As a result, the distribution ranges of the data for the C content, Mn content, and Si content were C: -0.211 to 2.299 mass%, Mn: 3.028 to 8.315 mass%, and Si: 1.291 to 2.663 mass%.

[0161] Using the content data as described above, a heat treatment temperature estimation device was used to estimate the heat treatment temperature (700°C in this example) in the heat treatment that was finally performed. In this example, a mapping image showing the distribution state of the Mn content was generated as the first mapping image, and a mapping image relating to the phase fraction of the fcc phase in the temperature range of 600 to 800°C was generated as the second mapping image (ΔT=1°C).

[0162] The Euclidean Distance and Squared Euclidean Distance, distance functions implemented in the numerical calculation application Mathematica, were used to calculate the similarity, and the temperature that gave the minimum distance value for each function was identified. The temperatures obtained from each function were averaged to determine the final heat treatment temperature.

[0163] As a result, the final heat treatment temperature obtained was 658° C. Since the actual heat treatment temperature was 700° C., it was found that the estimation error was approximately 50° C. (specifically, 42° C.).

[0164] Just to be sure, estimations were also made separately for two combinations of the C content and cementite phase fraction, and the Si content and bcc phase fraction, and the final heat treatment temperatures were estimated to be 668°C and 666°C, respectively.

[0165] From the above, it has been found that by using the heat treatment temperature estimation device and heat treatment temperature estimation method according to the present invention, it is possible to accurately estimate the final heat treatment temperature of a metal body of interest.

[0166] Example 2 Hereinafter, examples corresponding to the heat treatment temperature estimation device and the heat treatment temperature estimation method according to the second embodiment will be specifically described. In this example, first, the estimation process for the final heat treatment time, which was described with reference to Figures 16 to 19, was carried out for a seamless steel pipe having 0.3 mass% C-5.0 mass% Mn-Fe as alloy components, as described with reference to Figure 15. As a result, a heat treatment time change function for 0.3 mass% C-5.0 mass% Mn-Fe was derived, as shown in Figure 20.

[0167] Next, seamless steel pipes were produced by heat treatment as described with reference to Fig. 15 using a billet of 0.2 mass% C-1.5 mass% Si-1.5 mass% Mn-Fe. Thereafter, in accordance with the process of estimating the final heat treatment time described with reference to Figs. 16 to 19, data on the heat treatment time dependence (more specifically, the mean free energy V of the metallographic structure) for 0.2 mass% C-1.5 mass% Si-1.5 mass% Mn-Fe was calculated. P The final heat treatment time was estimated by estimating the heat treatment time using the heat treatment time-dependent quantity data, and the actual heat treatment time and the degree of dissociation were verified.

[0168] Here, as explained previously, both the 0.3 mass% C-5.0 mass% Mn-Fe alloy and the 0.2 mass% C-1.5 mass% Si-1.5 mass% Mn-Fe alloy are dual-phase steels that are expected to have fcc and bcc phases. Therefore, the relaxation time τ PIt was determined that the relaxation time τ for 0.3 mass% C-5.0 mass% Mn-Fe was not significantly different between the two alloys. P (1 / τ P =2.36383×10 -5 ) and the relaxation time τ of the heat treatment time change function for 0.2 mass%C-1.5 mass%Si-1.5 mass%Mn-Fe P It was decided to use it as such.

[0169] In deriving the heat treatment time change function for 0.2 mass%C-1.5 mass%Si-1.5 mass%Mn-Fe, the relaxation time τP was used from other alloy systems, so the unknown quantity to be found in equation (102) is V P 0 and A P The four calculated average free energies V P The heat treatment time change function of 0.2 mass% C-1.5 mass% Si-1.5 mass% Mn-Fe was derived using the values ​​of t=100 seconds and t=336 hours (specific values ​​are shown in FIG. 23) of (t). V P 0 =-1.67460, A P That is, the heat treatment time change function obtained in this example is different from that shown in FIG. P (t)=-1.67460+0.95913×exp(2.36383×10 -5 ×t).

[0170] The derived heat treatment time change function and the four calculated mean free energies V P The final heat treatment time was estimated using the values ​​of t = 1000 seconds and 20 hours in (t). The results are summarized in Figure 23.

[0171] 23, the estimated final heat treatment time for the heat treatment at t = 1000 seconds was 4027 seconds, which was 4.0 times the actual heat treatment time, and the estimated final heat treatment time for the heat treatment at t = 20 hours was 14 hours, which was 0.7 times the actual heat treatment time.

[0172] The estimated result for t = 1000 seconds showed a value 4.0 times the actual heat treatment time, but since the change in free energy of the metal structure is small when the heat treatment time is short, the error tends to be large, and it was presumed that the result was in line with this tendency. Also, in this example, the heat treatment time t = 4 types (i.e., the EPMA measurement data was also only t = 4 types) and the number of measurements N was small to begin with, which is thought to have resulted in a large error.

[0173] On the other hand, the estimation results for t = 20 hours show that the estimation was performed with an error of less than a factor of 2. This suggests that by appropriately adjusting the number of heat treatment times t used in the estimation, it is possible to estimate the heat treatment time with higher accuracy.

[0174] Although the preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings, the present invention is not limited to these examples. It is clear that a person skilled in the art to which the present invention pertains can conceive of various modifications and alterations within the scope of the technical ideas set forth in the claims, and it is understood that these also naturally fall within the technical scope of the present invention. [Explanation of symbols]

[0175] 10, 10A Heat treatment temperature estimation device 101 Content data acquisition section 103 First Vector Generator 105 Second Vector Generator 107 Estimation part 109 Output control section 111 Display control unit 113 Storage section 121 Thermodynamic analysis calculation section 123 Vector Generation Unit 151 Heat treatment time change function derivation part 153 Heat treatment time dependent quantity data acquisition section 155 Final heat treatment time estimation section

Claims

1. A heat treatment temperature estimation device that estimates a final heat treatment temperature, which is a heat treatment temperature in the last heat treatment performed on a metal body that is made of a predetermined alloy component and manufactured by heat treatment, a content data acquisition unit that acquires content data regarding the content of the alloy component in the metal body; a first vector generating unit that generates a first vector representing a distribution state of elements contained in the alloy components based on the content data; a second vector generation unit that performs a thermodynamic analysis calculation for a plurality of temperatures using the content data, and generates, for each of the temperatures, a second vector that represents a distribution state of a physical quantity related to the metal body from a calculation result of the thermodynamic analysis calculation; an estimation unit that calculates a similarity between the first vector and the second vector at each of the temperatures by a similarity analysis process and estimates the final heat treatment temperature based on the obtained similarity; A heat treatment temperature estimation device comprising:

2. 2. The heat treatment temperature estimation device according to claim 1, wherein the estimation unit identifies the temperature that gives a maximum or minimum value of the similarity in a plane defined by the temperature and the similarity, and estimates the final heat treatment temperature based on the temperature that gives the maximum or minimum value.

3. 3. The heat treatment temperature estimation device of claim 2, wherein the estimation unit identifies the temperature that gives the maximum or minimum value of the similarity in a plane defined by the temperature and the similarity for each of a plurality of combinations of the first vector and the second vector for the different elements, and estimates the final heat treatment temperature based on the temperatures that give the obtained maximum or minimum values.

4. The heat treatment temperature estimation device according to claim 3 , wherein the estimation unit determines the minimum, maximum, or average value of the obtained plurality of temperatures as the final heat treatment temperature.

5. 5. The heat treatment temperature estimation device according to claim 1, wherein the first vector is a vector that represents a distribution state of an element related to the physical quantity in the metal body.

6. 6. The heat treatment temperature estimation device according to claim 1, wherein the second vector is a vector that represents a distribution state of a phase fraction of the metal body.

7. 7. The heat treatment temperature estimation device according to claim 1, wherein the content data is analytical data obtained by a method of measuring and analyzing the spatial distribution of the alloy components.

8. The heat treatment temperature estimation device according to any one of claims 1 to 7, wherein the content data is analysis data obtained by analyzing the metal body with an electron probe microanalyzer (EPMA).

9. 9. The heat treatment temperature estimation device according to claim 1, wherein the second vector generation unit performs the thermodynamic analysis calculation using a CALPHAD (Computer Coupling of Phase Diagrams and Thermochemistry) method.

10. The similarity analysis process uses functions such as EuclideanDistance, SquaredEuclideanDistance, NormalizedSquaredEuclideanDistance, ManhattanDistance, CosineDistance, CorrelationDistance, MeanEuclideanDistance, MeanSquaredEuclideanDistance, RootMeanSquare, and MeanReciprocalSquaredEuclideanDistance. , MutualInformationVariation, NormalizedMutualInformationVariation, DifferenceNormalizedEntropy, MeanPatternIntensity, GradientCorrelation, MeanReciprocalGradientDistance, or EarthMoverDistance. The heat treatment temperature estimation device according to any one of claims 1 to 9, which is an analysis process using any one of these, or a similarity analysis process using deep learning processing.

11. the first vector generation unit further generates a first mapping image that visualizes a distribution state of elements included in the alloy components, using the generated first vector; the second vector generation unit further generates a second mapping image that visualizes a distribution state of a physical quantity related to the metal body using the generated second vector; The heat treatment temperature estimation device according to any one of claims 1 to 10, wherein the estimation unit calculates the similarity using the first mapping image and the second mapping image instead of the first vector and the second vector.

12. The heat treatment temperature estimation device of claim 11 , wherein the first mapping image and the second mapping image are grayscale images.

13. a heat treatment time change function deriving unit that derives a heat treatment time change function representing a time change of a heat treatment time dependent quantity, which is an observable quantity or a physical quantity that changes depending on the heat treatment time at the estimated final heat treatment temperature; a heat treatment time dependent quantity data acquisition unit that acquires heat treatment time dependent quantity data, which is data regarding the heat treatment time dependent quantity of the metal body; a final heat treatment time estimation unit that estimates a final heat treatment time, which is a heat treatment time of the metal body at the final heat treatment temperature, based on the acquired heat treatment time dependency of the metal body and the heat treatment time change function; Further provided with the heat treatment time dependent quantity is a physical quantity that represents the average free energy of the structure of the metal body, The heat treatment temperature estimation device according to any one of claims 1 to 12, wherein the heat treatment time t is a variable, the physical quantity expressing the average free energy is V P (t), the equilibrium free energy of the structure of the metal body at temperature T calculated from the content data is E Eq T (t), and the free energy of a reference state in the structure of the metal body at temperature T calculated from the content data is E 0 T (t), is defined by the following formula (1): [Equation 1]

14. The heat treatment temperature estimation device according to claim 13 , wherein the heat treatment time change function deriving unit derives the time change function using the heat treatment time dependent quantity identified at a plurality of different times.

15. 14. The heat treatment temperature estimation device according to claim 13, wherein the heat treatment time change function derivation unit fits the equation (1) to a function form expressed by a linear combination of one or more terms shown on the right side of the following equation (2), and uses the obtained fitting result as the heat treatment time change function. [Equation 2] In the above formula (2), t is a variable representing the heat treatment time, and V Ｐ ０ , A Ｐｉ , τ Ｐｉ are coefficients, and i is an integer equal to or greater than 1.

16. a database creation unit that creates a database of the derived heat treatment time change function; 16. The heat treatment temperature estimation device according to claim 13, wherein the final heat treatment time estimation unit estimates the final heat treatment time using the heat treatment time change function stored in a database.

17. A heat treatment temperature estimation method for estimating a final heat treatment temperature, which is a heat treatment temperature in the last heat treatment performed on a metal body made of a predetermined alloy composition and manufactured by heat treatment, comprising: a content data acquisition step of acquiring content data regarding the content of the alloy component in the metal body; a first vector generating step of generating a first vector representing a distribution state of elements contained in the alloy components based on the content data; a second vector generation step of performing a thermodynamic analysis calculation for a plurality of temperatures using the content data, and generating, for each of the temperatures, a second vector representing a distribution state of a physical quantity related to the metal body from a calculation result of the thermodynamic analysis calculation; an estimation step of calculating a similarity between the first vector and the second vector at each of the temperatures by a similarity analysis process, and estimating the final heat treatment temperature based on the obtained similarity; A method for estimating a heat treatment temperature, comprising:

18. a heat treatment time change function deriving step of deriving a heat treatment time change function that represents a time change of a heat treatment time dependent quantity, which is an observable quantity or a physical quantity that changes depending on the heat treatment time at the estimated final heat treatment temperature; a heat treatment time dependent quantity data acquisition step of acquiring heat treatment time dependent quantity data which is data relating to the heat treatment time dependent quantity of the metal body; a final heat treatment time estimation step of estimating a final heat treatment time, which is the heat treatment time of the metal body at the final heat treatment temperature, based on the acquired heat treatment time dependence of the metal body and the heat treatment time change function; Further comprising: the heat treatment time dependent quantity is a physical quantity that represents the average free energy of the structure of the metal body, 18. The heat treatment temperature estimation method according to claim 17, wherein the heat treatment time t is a variable, the physical quantity expressing the average free energy is V P (t), the equilibrium free energy of the structure of the metal body at temperature T calculated from the content data is E Eq T (t), and the free energy of a reference state in the structure of the metal body at temperature T calculated from the content data is E 0 T (t), is defined by the following equation (1): [Equation 3]

Images