Method for designing alloy by utilizing artificial intelligence optimized phase diagram

Abstract

The invention relates to the technical field of alloy design, in particular to a method for designing alloy by utilizing an artificial intelligence optimized phase diagram, which comprises the following steps: S1, obtaining all phase structures of a Pt-Ir-Al-Cr system; s2, optimizing a phase structure and calculating phase energy data, namely entropy, enthalpy, heat capacity and Gibbs free energy data; s3, determining a sublattice dot matrix model based on the phase structure, and establishing a thermodynamic model file based on calculated energy data; s4, collecting existing experimental data, and optimizing the thermodynamic model file based on an MCMC artificial intelligence algorithm; s5, drawing a phase diagram according to the optimized thermodynamic model, calculating phase composition and phase fraction according to the phase diagram, and determining an alloy design range; according to the method, phase diagram optimization is rapidly completed by using an MCMC artificial intelligence optimization algorithm, a thermodynamic model calculation file is obtained, an alloy design range is obtained based on the model, and target alloy components are obtained through grouping experiments.

Description

Technical Field

[0001] The present invention relates to the technical field of alloy design, and in particular to a method for designing alloys by optimizing phase diagrams using artificial intelligence. Background Art

[0002] Alloy design refers to the process of rationally selecting and adjusting the alloy's composition, organizational structure and production process to obtain performance that meets specific application requirements. This process aims to optimize the alloy's mechanical properties, physical properties and chemical properties to meet the requirements of various industrial applications. Alloy design is a complex and sophisticated process that involves multiple factors, including the selection of alloy composition, regulation of organizational structure and optimization of production process.

[0003] The design of traditional alloys relies on the "trial and error method", which is a method of approaching or achieving the desired effect through continuous testing and elimination of errors. It has problems such as long R&D cycle and high cost, which is not enough to meet the needs of the current rapid development of the electronics industry. Therefore, new methods should be found to improve work efficiency. Summary of the invention

[0004] In order to solve the current problems of high experimental trial and error cost and long cycle, the purpose of the present invention is to provide a method for designing alloys using artificial intelligence to optimize phase diagrams, which can optimize phase diagrams based on artificial intelligence, obtain accurate and reliable computational thermodynamic model files, and obtain the alloy design range based on the thermodynamic model.

[0005] To achieve the above-mentioned purpose, the present invention provides a method for designing alloys by optimizing phase diagrams using artificial intelligence, comprising the following steps: S1: obtaining all phase structures of the Pt-Ir-Al-Cr system; S2: optimizing the phase structure and calculating the energy data of the phase, namely, entropy, enthalpy, heat capacity and Gibbs free energy data; S3: determining the sublattice lattice model based on the phase structure, and establishing a thermodynamic model file based on the calculated energy data; S4: collecting existing experimental data, and optimizing the thermodynamic model file based on the MCMC artificial intelligence algorithm; S5: drawing a phase diagram according to the optimized thermodynamic model, calculating the phase composition and phase fraction according to the phase diagram, and determining the alloy design range.

[0006] As a preferred embodiment of the present invention, S1 includes the following steps: S101: Based on an open material database website, obtaining the physical phase structure of the Pt-Ir-Al-Cr system on the website; S102: Consulting relevant literature, for physical phase structures not provided in the literature, using structure generation software to establish the physical phase structure.

[0007] As a preferred embodiment of the present invention, S2 includes the following steps: S201: Based on the S1 phase structure, perform a full relaxation calculation of the structure based on the first principles; use the fully relaxed structure as the input structure to calculate the static cold energy, Fermi energy and energy state density; S202: Based on the static cold energy, Fermi energy and energy state density files, calculate the entropy, enthalpy, heat capacity and Gibbs free energy data of the phase structure.

[0008] As a preferred embodiment of the present invention, S3 includes the following steps: S301: Determine the sublattice lattice model of the phase based on the structure, create a phase structure file, and create a phase energy file based on the energy data; S302: Based on the phase structure file and the phase energy file, generate a thermodynamic model file based on the first principles calculation data.

[0009] As a preferred embodiment of the present invention, S4 includes the following steps: S401: Collect experimental data by consulting literature and store the experimental data as a standard experimental data file; S402: Optimize the thermodynamic model file using the MCMC artificial intelligence algorithm based on the standard experimental data file.

[0010] As a preferred embodiment of the present invention, S5 includes the following steps: S501: drawing a phase diagram according to an optimized thermodynamic model file; S502: according to the phase diagram of the thermodynamic model, selecting a suitable temperature and element content range to calculate and determine the phase composition and phase fraction of the phase structure, and analyzing to obtain the alloy design range.

[0011] As a preferred embodiment of the present invention, after step S5, the method further includes: S6: preparing alloy samples by taking multiple groups of alloy components according to the alloy design range; S7: cutting the alloy samples and testing them.

[0012] As a preferred embodiment of the present invention, S6 includes the following steps: S601: According to the alloy design range calculated and optimized by the calculation phase diagram method, Pt-Ir-Al-Cr alloys with different proportions are selected as research objects; S602: Preparing metal powders with a purity of ≥99.99% of Pt, Ir, Al, and Cr, the powder particle size of the metal element powders Ir and Al is <100 mesh, and the powder particle size of the metal element powders Pt and Cr is <200 mesh, and the ingredients are prepared according to the proportion of the metal element powders, and the raw material powders are preliminarily evenly mixed and pressed into tablets; S603: Smelting in a vacuum arc melting furnace, requiring the vacuum degree in the furnace to be >2.0X10 -3Pa, fill with argon as protective gas, the smelting temperature is 1500-1700℃, the smelting times are ≥12 times, and the molten state is maintained for >5min each time; S604: take out the ingot after cooling, seal the ingot in a vacuum tube and put it into a tubular furnace, set the heating rate to 1min / ℃, the temperature to 1300℃, keep warm for 60h, and take out the sample after the tubular furnace cools to room temperature.

[0013] As a preferred embodiment of the present invention, S7 includes the following steps: S701: slicing the prepared Pt-Ir-Al-Cr alloy sample, characterizing the sample by X-ray diffraction, and observing and analyzing the phase composition of the sample; S702: performing morphology analysis using a scanning electron microscope; S703: testing the hardness of the sample using a digital microhardness tester; S704: testing the tensile strength, compressive strength and other properties of the sample using an electronic universal mechanical properties testing machine.

[0014] The beneficial effects of the present invention are as follows: the present invention can quickly and automatically complete the phase diagram optimization based on Bayesian statistics by using the MCMC (Markov Chain Monte Carlo) artificial intelligence optimization algorithm, obtain an accurate and reliable computational thermodynamic model file, obtain the alloy design range based on the thermodynamic model, and obtain the target alloy composition through group experiments, which greatly reduces the cost and design cycle compared to the trial and error method. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0016] Figure 1 It is a design flow chart of the method for designing alloys by optimizing phase diagrams using artificial intelligence according to the present invention.

[0017] Figure 2 It is the phase diagram of the thermodynamic model Pt-Al.

[0018] Figure 3 The thermodynamic model Pt 0.88 Al 0.12 The phase fraction.

[0019] Figure 4 The thermodynamic model Pt 0.88 Al 0.12 phase composition.

[0020] Figure 5 The thermodynamic model Pt 0.82 Al 0.12 Cr 0.06 The phase fraction.

[0021] Figure 6 The thermodynamic model Pt 0.82 Al 0.12 Cr 0.06 phase composition.

[0022] Figure 7 The thermodynamic model Pt 0.82 Ir 0.03 Al 0.12 Cr 0.03 The phase fraction.

[0023] Figure 8 The thermodynamic model Pt 0.82 Ir 0.03 Al 0.12 Cr 0.03 phase composition.

[0024] Fig. 9 The thermodynamic model Pt 0.79 Ir 0.03 Al 0.12 Cr 0.06 phase composition.

[0025] Fig.10 The thermodynamic model Pt 0.79 Ir 0.03 Al 0.12 Cr 0.06 The phase fraction.

[0026] Fig.11 The thermodynamic model Pt 0.76 Ir 0.06 Al 0.12 Cr 0.06 The phase fraction.

[0027] Fig.12 The thermodynamic model Pt 0.76 Ir 0.06 Al 0.12 Cr 0.06 phase composition.

[0028] Fig.13 Example 1 (Pt 0.82 Ir 0.03 Al 0.12 Cr 0.03 )’s SEM-BSE test results.

[0029] Fig.14 Example 2 (Pt 0.79 Ir 0.03 Al 0.12 Cr 0.06 )’s SEM-BSE test results.

[0030] Fig.15 Example 3 (Pt 0.76 Ir 0.06 Al 0.12 Cr 0.06 )’s SEM-BSE test results. DETAILED DESCRIPTION

[0031] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the embodiments of the present invention, and should not be construed as limiting the present invention.

[0032] In the description of the embodiments of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0033] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0034] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0035] Principle explanation: Calphad (CALculation of PHAse Diagrams) method is a method for calculating material phase diagrams, which is a product of the combination of thermodynamics and computer science; MCMC (Markov Chain Monte Carlo) is a powerful tool for resampling from Bayesian probability distribution; the core idea of ​​MCMC algorithm is: 1. Design a Markov chain to make its stable distribution the target distribution; 2. Starting from a certain initial state, use Monte Carlo random sampling to approximate the required statistics to achieve the transition rule of the Markov chain for iteration; 3. After enough iterations, the state distribution of the Markov chain converges to the standard distribution, and enough samples are drawn from the Markov chain to approximate the target distribution to complete the optimization.

[0036] Embodiment 1

[0037] like Figure 1 As shown, a method for designing alloys by optimizing phase diagrams using artificial intelligence includes the following steps: S1: Obtain all phase structures of the Pt-Ir-Al-Cr system, retrieve all phase structures of the Pt-Ir-Al-Cr system based on open material database websites (such as: Material Project, OQMD, ICSD, Calphad, etc.), consult relevant literature, ATAT alloy calculation toolkit SQS and SAE software package; take the Pt-Al system as an example, retrieve Al21Pt5, Al21Pt8, Al2Pt, Al3Pt2, AlPt, Al3Pt5, AlPt2, disordered solid solution FCC_A1, BCC_A2 and ordered L12 and B2 phases of the Pt-Al system; the specific steps include: S101: Based on open material database websites (such as: Material Project, OQMD, ICSD, Calphad, etc.), consult relevant literature, ATAT alloy calculation toolkit SQS and SAE software package Project, OQMD, ICSD, Calphad, etc.) to obtain the phase structure of the Pt-Ir-Al-Cr system on the website; S102: Consult relevant literature, and use structure generation software (ATAT alloy calculation toolkit SQS and SAE software package) to establish the phase structure if the phase structure is not provided in the literature.

[0038] S2: Optimize the phase structure and calculate the energy data of the phase, namely entropy, enthalpy, heat capacity and Gibbs free energy data; the specific steps include: S201: Based on the phase structure of S1, perform a full relaxation calculation of the structure based on the first principles, and use the full relaxation structure as the input structure to calculate the static cold energy, Fermi energy and energy state density; the first principle calculation software VASP can be used, and the specific settings are: the cutoff energy is 400eV, the electronic self-consistent convergence energy is 10 -6 eV / atom, the convergence accuracy of the force is less than A 6×6×6 Monkhorst-Pack K-point grid was used, ISIF=3; S202: Based on the static cold energy, Fermi energy and energy state density files, the entropy, enthalpy, heat capacity and Gibbs free energy data of the phase structure were calculated. The calculation formula is as follows:

[0039] G(T,P)=F(T,V)+PV

[0040]

[0041] H=G+TS

[0042]

[0043] Where F(T,V) is the Helmholtz energy, G is the Gibbs free energy, S is the entropy, H is the enthalpy, and C is the Pis the isobaric heat capacity, P is the pressure, T is the temperature, and V is the volume.

[0044] The Gibbs free energy of a single phase is generally described as follows:

[0045]

[0046] in, is the Gibbs free energy of the α phase, represents the Gibbs free energy of the unreacted mixture in the composition; represents the Gibbs free energy contribution due to the physical model; represents the configurational entropy of the phase; Excess Gibbs free energy; or calculated using MFP calculation software.

[0047] S3: Determine the sublattice lattice model based on the phase structure, and establish a thermodynamic model file based on the calculated energy data; the specific steps include: S301: Determine the sublattice lattice model of the phase based on the structure, and create a phase structure file, that is, create a phase structure file based on the phase description information of the sublattice lattice model. The phase structure file includes element types, sublattice lattices and corresponding proportions. The compound phase generally uses 2SL (2sublattice, such as [A][B]), and the FCC_A1, BCC_A2, and HCP_A3 phases use 1SL (1sublattice, such as [A X ,B 1-X ]), BCC_B2 phase uses 2SL (2sublattice, such as [A X ,B 1-X ]A X ,B 1-X ]) and FCC_L12 phase uses 4SL (4sublattice, such as [A X ,B 1-X ][A X ,B 1-X ][A X ,B 1-X ][A X ,B 1-X ]); the physical phase energy file refers to an energy data file calculated based on the physical phase structure, that is, for each determined phase, the energy value of the corresponding fixed structure is calculated and converted into a standard output format file. The energy values ​​include entropy, enthalpy and mixing enthalpy; S302: according to the physical phase structure file and the physical phase energy file, a thermodynamic model file based on the first-principles calculation data is generated, that is, the physical phase structure file reads the corresponding physical phase energy file to generate a thermodynamic model file.

[0048] S4: Collect existing experimental data and optimize the thermodynamic model file based on the MCMC artificial intelligence algorithm; the specific steps include: S401: Collect experimental data by consulting literature, such as experimental phase boundary data, single-phase region and multi-phase region experimental data, and store the experimental data as a standard experimental data file; S402: According to the standard experimental data file, use the MCMC artificial intelligence algorithm to optimize the thermodynamic model file, that is, the Markov chain Monte Carlo algorithm based on Bayesian statistics, and use the data optimized by the Monte Carlo algorithm each time as the input data of the next Markov chain, and optimize the thermodynamic model file in a chain iterative manner; the specific operation can use AutoCalphad software to read the experimental data file in a standard format, and use the MCMC (MarkovChain Monte Carlo) artificial intelligence algorithm to optimize the thermodynamic model file.

[0049] S5: Draw a phase diagram based on the optimized thermodynamic model, calculate the phase composition and phase fraction based on the phase diagram, and determine the alloy design range. The phase diagram is as follows: Figure 2 As shown; the specific steps include: S501: According to the phase diagram of the optimized thermodynamic model file, the AutoCalphad software can be used to draw it; S502: According to the phase diagram of the thermodynamic model, the appropriate temperature and element content range are selected to calculate the phase composition and phase fraction, and the alloy design range is obtained by analysis.

[0050] like Figure 2 to Figure 6 As shown, for example, the appearance range of FCC_L12 (Pt3Al) phase is Pt (0.76-0.90), temperature 300-2000℃, based on the Pt-Al phase diagram information, calculate Pt 0.88 Al 0.12 Phase fraction and image composition; add Cr element and calculate Pt 0.82 Al 0.12 Cr 0.06 phase fraction and phase composition.

[0051] like Figure 7 and Figure 8 As shown, S6: according to the alloy design range, multiple sets of alloy components are selected to prepare alloy samples; the specific steps include: S601: according to the alloy design range calculated and optimized by the calculation phase diagram method, Pt-Ir-Al-Cr alloys with different proportions are selected as research objects. In this embodiment, Pt 0.82 Ir 0.03 Al 0.12 Cr 0.03; S602: prepare metal powders of Pt, Ir, Al, and Cr with a purity of ≥99.99%, the powder particle size of metal element powders Ir and Al is <100 mesh, and the powder particle size of metal element powders Pt and Cr is <200 mesh, and the ingredients are prepared according to the proportion of metal element powders: 82at% Pt, 3at% Ir, 12at% Al, 3at% Cr, and the raw material powders are initially evenly mixed and pressed into tablets; S603: smelting in a vacuum arc melting furnace, requiring the vacuum degree in the furnace to be >2.0X10 -3 Pa, fill with argon as protective gas, the smelting temperature is 1500-1700℃, the smelting times are ≥12 times, and the molten state is maintained for >5min each time; S604: take out the ingot after cooling, seal the ingot in a vacuum tube and put it into a tubular furnace, set the heating rate to 1min / ℃, the temperature to 1300℃, keep warm for 60h, and take out the sample after the tubular furnace cools to room temperature.

[0052] like Fig.13 As shown, S7: cutting the alloy sample and testing it; the specific steps include: S701: slicing the prepared Pt-Ir-Al-Cr alloy sample, characterizing the sample with X-ray diffraction (XRD), and observing and analyzing the phase composition of the sample. Grind the sample into a cross section with sandpaper, and use 3μm diamond polishing agent to polish and clean it to make the cross section smooth. The XRD test conditions are: scanning diffraction angle range of 10-70°, scanning speed of 10min, step length of 0.02deg, and use MDIjade data analysis software to calibrate the XRD phase analysis results; S702: Use scanning electron microscope (SEM) for morphology analysis. The sample for point scanning analysis is polished with 3μm diamond polishing agent, and an acceleration voltage of 30kV is used during analysis. EDS point scanning is performed on a single particle, such as Fig. 9 As shown; S703: Use digital microhardness tester to test sample hardness. Analyze sample results, 3 samples are made for each alloy, Pt 0.82 Ir 0.03 Al 0.12 Cr 0.03 The average hardness is 320.022HV; S704: The tensile strength of the samples was tested using an electronic universal mechanical properties testing machine. 0.82 Ir 0.03 Al 0.12 Cr 0.03 The maximum tensile strength is 743MPa.

[0053] Embodiment 2

[0054] like Figure 1As shown, a method for designing alloys by optimizing phase diagrams using artificial intelligence includes the following steps: S1: Obtain all phase structures of the Pt-Ir-Al-Cr system, retrieve all phase structures of the Pt-Ir-Al-Cr system based on open material database websites (such as: Material Project, OQMD, ICSD, Calphad, etc.), consult relevant literature, ATAT alloy calculation toolkit SQS and SAE software package; take the Pt-Al system as an example, retrieve Al21Pt5, Al21Pt8, Al2Pt, Al3Pt2, AlPt, Al3Pt5, AlPt2, disordered solid solution FCC_A1, BCC_A2 and ordered L12 and B2 phases of the Pt-Al system; the specific steps include: S101: Based on open material database websites (such as: Material Project, OQMD, ICSD, Calphad, etc.), consult relevant literature, ATAT alloy calculation toolkit SQS and SAE software package Project, OQMD, ICSD, Calphad, etc.) to obtain the phase structure of the Pt-Ir-Al-Cr system on the website; S102: Consult relevant literature, and use structure generation software (ATAT alloy calculation toolkit SQS and SAE software package) to establish the phase structure if the phase structure is not provided in the literature.

[0055] S2: Optimize the phase structure and calculate the energy data of the phase, namely entropy, enthalpy, heat capacity and Gibbs free energy data; the specific steps include: S201: Based on the phase structure of S1, perform a full relaxation calculation of the structure based on the first principles, and use the full relaxation structure as the input structure to calculate the static cold energy, Fermi energy and energy state density; the first principle calculation software VASP can be used, and the specific settings are: the cutoff energy is 400eV, the electronic self-consistent convergence energy is 10 -6 eV / atom, the convergence accuracy of the force is less than A 6×6×6 Monkhorst-Pack K-point grid was used, ISIF=3; S202: Based on the static cold energy, Fermi energy and energy state density files, the entropy, enthalpy, heat capacity and Gibbs free energy data of the phase structure were calculated. The calculation formula is as follows:

[0056] G(T,P)=F(T,V)+PV

[0057]

[0058] H=G+TS

[0059]

[0060] Where F(T,V) is the Helmholtz energy, G is the Gibbs free energy, S is the entropy, H is the enthalpy, and C is the P is the isobaric heat capacity, P is the pressure, T is the temperature, and V is the volume.

[0061] The Gibbs free energy of a single phase is generally described as follows:

[0062]

[0063] in, is the Gibbs free energy of the α phase, represents the Gibbs free energy of the unreacted mixture in the composition; represents the Gibbs free energy contribution due to the physical model; represents the configurational entropy of the phase; Excess Gibbs free energy; or calculated using MFP calculation software.

[0064] S3: Determine the sublattice lattice model based on the phase structure, and establish a thermodynamic model file based on the calculated energy data; the specific steps include: S301: Determine the sublattice lattice model of the phase based on the structure, and create a phase structure file, that is, create a phase structure file based on the phase description information of the sublattice lattice model. The phase structure file includes element types, sublattice lattices and corresponding proportions. The compound phase generally uses 2SL (2sublattice, such as [A][B]), and the FCC_A1, BCC_A2, and HCP_A3 phases use 1SL (1sublattice, such as [A X ,B 1-X ]), BCC_B2 phase uses 2SL (2sublattice, such as [A X ,B 1-X ]A X ,B 1-X ]) and FCC_L12 phase uses 4SL (4sublattice, such as [A X ,B 1-X ][A X ,B 1-X ][A X ,B 1-X ][A X ,B 1-X ]); the physical phase energy file refers to an energy data file calculated based on the physical phase structure, that is, for each determined phase, the energy value of the corresponding fixed structure is calculated and converted into a standard output format file. The energy values ​​include entropy, enthalpy and mixing enthalpy; S302: according to the physical phase structure file and the physical phase energy file, a thermodynamic model file based on the first-principles calculation data is generated, that is, the physical phase structure file reads the corresponding physical phase energy file to generate a thermodynamic model file.

[0065] S4: Collect existing experimental data and optimize the thermodynamic model file based on the MCMC artificial intelligence algorithm; the specific steps include: S401: Collect experimental data by consulting literature, such as experimental phase boundary data, single-phase region and multi-phase region experimental data, and store the experimental data as a standard experimental data file; S402: According to the standard experimental data file, use the MCMC artificial intelligence algorithm to optimize the thermodynamic model file, that is, the Markov chain Monte Carlo algorithm based on Bayesian statistics, and use the data optimized by the Monte Carlo algorithm each time as the input data of the next Markov chain, and optimize the thermodynamic model file in a chain iterative manner; the specific operation can use AutoCalphad software to read the experimental data file in a standard format, and use the MCMC (MarkovChain Monte Carlo) artificial intelligence algorithm to optimize the thermodynamic model file.

[0066] S5: Draw a phase diagram based on the optimized thermodynamic model, calculate the phase composition and phase fraction based on the phase diagram, and determine the alloy design range. The phase diagram is as follows: Figure 2 As shown; the specific steps include: S501: According to the phase diagram of the optimized thermodynamic model file, the AutoCalphad software can be used to draw it; S502: According to the phase diagram of the thermodynamic model, the appropriate temperature and element content range are selected to calculate the phase composition and phase fraction, and the alloy design range is obtained by analysis.

[0067] like Figure 2 to Figure 6 As shown, for example, the appearance range of FCC_L12 (Pt3Al) phase is Pt (0.76-0.90), temperature 300-2000℃, based on the Pt-Al phase diagram information, calculate Pt 0.88 Al 0.12 Phase fraction and image composition; add Cr element and calculate Pt 0.82 Al 0.12 Cr 0.06 phase fraction and phase composition.

[0068] like Fig. 9 and Fig.10 As shown, S6: according to the alloy design range, multiple sets of alloy components are selected to prepare alloy samples; the specific steps include: S601: according to the alloy design range calculated and optimized by the calculation phase diagram method, Pt-Ir-Al-Cr alloys with different proportions are selected as research objects. In this embodiment, Pt 0.79 Ir 0.03 Al 0.12 Cr 0.06; S602: prepare metal powders of Pt, Ir, Al, and Cr with a purity of ≥99.99%. The powder particle size of metal element powders Ir and Al is <100 mesh, and the powder particle size of metal element powders Pt and Cr is <200 mesh. The ingredients are prepared according to the proportion of metal element powders: 79at% Pt, 3at% Ir, 12at% Al, and 6at% Cr. The raw material powders are initially evenly mixed and pressed into tablets; S603: smelting is carried out in a vacuum arc melting furnace, and the vacuum degree in the furnace is required to be >2.0X10 -3 Pa, fill with argon as protective gas, the smelting temperature is 1500-1700℃, the smelting times are ≥12 times, and the molten state is maintained for >5min each time; S604: take out the ingot after cooling, seal the ingot in a vacuum tube and put it into a tubular furnace, set the heating rate to 1min / ℃, the temperature to 1300℃, keep warm for 60h, and take out the sample after the tubular furnace cools to room temperature.

[0069] like Fig.14 As shown, S7: cutting the alloy sample and testing it; the specific steps include: S701: slicing the prepared Pt-Ir-Al-Cr alloy sample, characterizing the sample with X-ray diffraction (XRD), and observing and analyzing the phase composition of the sample. Grind the sample into a cross section with sandpaper, and use 3μm diamond polishing agent to polish and clean it to make the cross section smooth. The XRD test conditions are: scanning diffraction angle range of 10-70°, scanning speed of 10min, step length of 0.02deg, and use MDIjade data analysis software to calibrate the XRD phase analysis results; S702: Use scanning electron microscope (SEM) for morphology analysis. The sample for point scanning analysis is polished with 3μm diamond polishing agent, and an acceleration voltage of 30kV is used during analysis. EDS point scanning is performed on a single particle, such as Fig.10 As shown; S703: Use a digital microhardness tester to test the hardness of the sample. ; Analyze the sample results, 3 samples were prepared for each alloy, Pt 0.79 Ir 0.03 Al 0.12 Cr 0.06 The average hardness is 356.902HV; S704: The tensile strength performance of the sample is tested using an electronic universal mechanical performance testing machine. Analysis of the sample results, Pt 0.79 Ir 0.03 Al 0.12 Cr 0.06 The maximum tensile strength is 836MPa.

[0070] Embodiment 3

[0071] like Figure 1As shown, a method for designing alloys by optimizing phase diagrams using artificial intelligence includes the following steps: S1: Obtain all phase structures of the Pt-Ir-Al-Cr system, retrieve all phase structures of the Pt-Ir-Al-Cr system based on open material database websites (such as: Material Project, OQMD, ICSD, Calphad, etc.), consult relevant literature, ATAT alloy calculation toolkit SQS and SAE software package; take the Pt-Al system as an example, retrieve Al21Pt5, Al21Pt8, Al2Pt, Al3Pt2, AlPt, Al3Pt5, AlPt2, disordered solid solution FCC_A1, BCC_A2 and ordered L12 and B2 phases of the Pt-Al system; the specific steps include: S101: Based on open material database websites (such as: Material Project, OQMD, ICSD, Calphad, etc.), consult relevant literature, ATAT alloy calculation toolkit SQS and SAE software package Project, OQMD, ICSD, Calphad, etc.) to obtain the phase structure of the Pt-Ir-Al-Cr system on the website; S102: Consult relevant literature, and use structure generation software (ATAT alloy calculation toolkit SQS and SAE software package) to establish the phase structure if the phase structure is not provided in the literature.

[0072] S2: Optimize the phase structure and calculate the energy data of the phase, namely entropy, enthalpy, heat capacity and Gibbs free energy data; the specific steps include: S201: Based on the phase structure of S1, perform a full relaxation calculation of the structure based on the first principles, and use the full relaxation structure as the input structure to calculate the static cold energy, Fermi energy and energy state density; the first principle calculation software VASP can be used, and the specific settings are: the cutoff energy is 400eV, the electronic self-consistent convergence energy is 10 -6 eV / atom, the convergence accuracy of the force is less than A 6×6×6 Monkhorst-Pack K-point grid was used, ISIF=3; S202: Based on the static cold energy, Fermi energy and energy state density files, the entropy, enthalpy, heat capacity and Gibbs free energy data of the phase structure were calculated. The calculation formula is as follows:

[0073] G(T,P)=F(T,V)+PV

[0074]

[0075] H=G+TS

[0076]

[0077] Where F(T,V) is the Helmholtz energy, G is the Gibbs free energy, S is the entropy, H is the enthalpy, and C is the P is the isobaric heat capacity, P is the pressure, T is the temperature, and V is the volume.

[0078] The Gibbs free energy of a single phase is generally described as follows:

[0079]

[0080] in, is the Gibbs free energy of the α phase, represents the Gibbs free energy of the unreacted mixture in the composition; represents the Gibbs free energy contribution due to the physical model; represents the configurational entropy of the phase; Excess Gibbs free energy; or calculated using MFP calculation software.

[0081] S3: Determine the sublattice lattice model based on the phase structure, and establish a thermodynamic model file based on the calculated energy data; the specific steps include: S301: Determine the sublattice lattice model of the phase based on the structure, and create a phase structure file, that is, create a phase structure file based on the phase description information of the sublattice lattice model. The phase structure file includes element types, sublattice lattices and corresponding proportions. The compound phase generally uses 2SL (2sublattice, such as [A][B]), and the FCC_A1, BCC_A2, and HCP_A3 phases use 1SL (1sublattice, such as [A X ,B 1-X ]), BCC_B2 phase uses 2SL (2sublattice, such as [A X ,B 1-X ]A X ,B 1-X ]) and FCC_L12 phase uses 4SL (4sublattice, such as [A X ,B 1-X ][A X ,B 1-X ][A X ,B 1-X ][A X ,B 1-X ]); the physical phase energy file refers to an energy data file calculated based on the physical phase structure, that is, for each determined phase, the energy value of the corresponding fixed structure is calculated and converted into a standard output format file. The energy values ​​include entropy, enthalpy and mixing enthalpy; S302: according to the physical phase structure file and the physical phase energy file, a thermodynamic model file based on the first-principles calculation data is generated, that is, the physical phase structure file reads the corresponding physical phase energy file to generate a thermodynamic model file.

[0082] S4: Collect existing experimental data and optimize the thermodynamic model file based on the MCMC artificial intelligence algorithm; the specific steps include: S401: Collect experimental data by consulting literature, such as experimental phase boundary data, single-phase region and multi-phase region experimental data, and store the experimental data as a standard experimental data file; S402: According to the standard experimental data file, use the MCMC artificial intelligence algorithm to optimize the thermodynamic model file, that is, the Markov chain Monte Carlo algorithm based on Bayesian statistics, and use the data optimized by the Monte Carlo algorithm each time as the input data of the next Markov chain, and optimize the thermodynamic model file in a chain iterative manner; the specific operation can use AutoCalphad software to read the experimental data file in a standard format, and use the MCMC (MarkovChain Monte Carlo) artificial intelligence algorithm to optimize the thermodynamic model file.

[0083] S5: Draw a phase diagram based on the optimized thermodynamic model, calculate the phase composition and phase fraction based on the phase diagram, and determine the alloy design range. The phase diagram is as follows: Figure 2 As shown; the specific steps include: S501: According to the phase diagram of the optimized thermodynamic model file, the AutoCalphad software can be used to draw it; S502: According to the phase diagram of the thermodynamic model, the appropriate temperature and element content range are selected to calculate the phase composition and phase fraction, and the alloy design range is obtained by analysis.

[0084] like Figure 2 to Figure 6 As shown, for example, the FCC_L12 (Pt3Al) phase appears in the range of Pt (0.76-0.90) and the temperature is 300-2000 ° C. Based on the Pt-Al phase diagram information, the Pt 0.88 Al 0.12 Phase fraction and image composition; add Cr element and calculate Pt 0.82 Al 0.12 Cr 0.06 phase fraction and phase composition.

[0085] like Fig.11 and Fig.12 As shown, S6: according to the alloy design range, multiple sets of alloy components are selected to prepare alloy samples; the specific steps include: S601: according to the alloy design range calculated and optimized by the calculation phase diagram method, Pt-Ir-Al-Cr alloys with different proportions are selected as research objects. In this embodiment, Pt 0.76 Ir 0.06 Al0.15Cr 0.06; S602: prepare metal powders of Pt, Ir, Al, and Cr with a purity of ≥99.99%. The powder particle size of metal element powders Ir and Al is <100 mesh, and the powder particle size of metal element powders Pt and Cr is <200 mesh. The ingredients are prepared according to the proportion of metal element powders: 76at% Pt, 6at% Ir, 12at% Al, and 6at% Cr. The raw material powders are initially evenly mixed and pressed into tablets; S603: smelting is carried out in a vacuum arc melting furnace, and the vacuum degree in the furnace is required to be >2.0X10 -3 Pa, fill with argon as protective gas, the smelting temperature is 1500-1700℃, the smelting times are ≥12 times, and the molten state is maintained for >5min each time; S604: take out the ingot after cooling, seal the ingot in a vacuum tube and put it into a tubular furnace, set the heating rate to 1min / ℃, the temperature to 1300℃, keep warm for 60h, and take out the sample after the tubular furnace cools to room temperature.

[0086] like Fig.15 As shown, S7: cutting the alloy sample and testing it; the specific steps include: S701: slicing the prepared Pt-Ir-Al-Cr alloy sample, characterizing the sample with X-ray diffraction (XRD), and observing and analyzing the phase composition of the sample. Grind the sample into a cross section with sandpaper, and use 3μm diamond polishing agent to polish and clean it to make the cross section smooth. The XRD test conditions are: scanning diffraction angle range of 10-70°, scanning speed of 10min, step length of 0.02deg, and use MDIjade data analysis software to calibrate the XRD phase analysis results; S702: Use scanning electron microscope (SEM) for morphology analysis. The sample for point scanning analysis is polished with 3μm diamond polishing agent, and an acceleration voltage of 30kV is used during analysis. EDS point scanning is performed on a single particle, such as Fig.11 As shown; S703: Use digital microhardness tester to test sample hardness. Analyze sample results, 3 samples are made for each alloy, Pt 0.76 Ir 0.06 Al0.15Cr 0.06 The average hardness is 302.743HV; S704: The tensile strength of the samples was tested using an electronic universal mechanical performance testing machine. 0.76 Ir 0.06 Al0.15Cr 0.06 The maximum tensile strength is 689MPa.

[0087] Table 1. Types and amounts of elements in alloys of samples 1-3

[0088]

[0089] Table 2. Hardness test results of samples 1-3

[0090]

[0091] Table 3. Compressive strength test results of samples 1-3

[0092]

[0093] The preferred implementation modes of the present application are described in detail above in conjunction with the accompanying drawings. Typical known structures and common knowledge technologies in the preferred implementation modes are not described in detail here. Ordinary technicians in the relevant field can improve and implement the technical solutions of the present invention based on their own abilities under the inspiration given by the present implementation modes. Some typical known structures, known methods or common knowledge technologies should not become obstacles for ordinary technicians in the relevant field to implement the present application.

[0094] The scope of protection required by this application shall be based on the contents of its claims, and the contents recorded in the invention content, specific implementation methods and drawings of the specification shall be used to interpret the claims.

[0095] Within the technical concept of the present application, several modifications may be made to the specific implementation methods of the present application, and these modified specific implementation methods should also be regarded as within the protection scope of the present application.

Claims

1. A method for designing alloys by optimizing phase diagrams using artificial intelligence, characterized in that: The following steps are involved: S1: Obtain all phase structures of the Pt-Ir-Al-Cr system; S2: Optimize the phase structure and calculate the energy data of the phase, namely entropy, enthalpy, heat capacity and Gibbs free energy data; S3: Determine the sub-lattice lattice model based on the physical phase structure and establish a thermodynamic model file based on the calculated energy data; S4: Collect existing experimental data and optimize the thermodynamic model file based on MCMC artificial intelligence algorithm; S5: Draw a phase diagram based on the optimized thermodynamic model, calculate the phase composition and phase fraction based on the phase diagram, and determine the alloy design range.

2. The method for designing alloys by optimizing phase diagrams using artificial intelligence according to claim 1, characterized in that: S1 includes the following steps: S101: Based on the open material database website, obtain the physical structure of the Pt-Ir-Al-Cr system on the website; S102: Consult relevant literature, and use structure generation software to establish the physical structure if it is not provided in the literature.

3. The method for designing alloys by optimizing phase diagrams using artificial intelligence according to claim 1, characterized in that: S2 includes the following steps: S201: Based on the S1 phase structure, perform a fully relaxed structural calculation based on first principles; use the fully relaxed structure as the input structure to calculate the static cold energy, Fermi energy and energy state density; S202: Calculate the entropy, enthalpy, heat capacity and Gibbs free energy data of the phase structure based on the static cold energy, Fermi energy and energy state density files.

4. The method for designing alloys by optimizing phase diagrams using artificial intelligence according to claim 1, characterized in that: S3 includes the following steps: S301: Determine a sub-lattice lattice model of a phase based on the structure, create a phase structure file, and create a phase energy file based on energy data; S302: Generate a thermodynamic model file based on first-principles calculation data according to the physical phase structure file and the physical phase energy file.

5. The method for designing alloys by optimizing phase diagrams using artificial intelligence according to claim 1, characterized in that: S4 includes the following steps: S401: Collect experimental data by consulting literature and save the experimental data as a standard experimental data file; S402: Optimize the thermodynamic model file using the MCMC artificial intelligence algorithm based on the standard experimental data file.

6. The method for designing alloys by optimizing phase diagrams using artificial intelligence according to claim 1, characterized in that: S5 includes the following steps: S501: Draw a phase diagram according to the optimized thermodynamic model file; S502: According to the phase diagram of the thermodynamic model, select the appropriate temperature and element content range to calculate the phase composition and phase fraction, and analyze to obtain the alloy design range.

7. The method for designing alloys by optimizing phase diagrams using artificial intelligence according to claim 1, characterized in that: After step S5, the method further includes: S6: According to the alloy design range, multiple sets of alloy components are taken to prepare alloy samples; S7: Cut the alloy sample and perform the test.

8. The method for designing alloys by optimizing phase diagrams using artificial intelligence according to claim 7, characterized in that: S6 includes the following steps: S601: According to the alloy design range calculated and optimized by the calculation phase diagram method, Pt-Ir-Al-Cr alloys with different composition ratios are selected as research objects; S602: prepare metal powders of Pt, Ir, Al, and Cr with a purity of ≥99.99%, the powder particle size of metal element powders Ir and Al is less than 100 mesh, and the powder particle size of metal element powders Pt and Cr is less than 200 mesh, and the ingredients are prepared according to the proportion of the metal element powders, and the raw material powders are preliminarily evenly mixed and pressed into tablets; S603: Vacuum arc melting furnace is used for melting, and the vacuum degree in the furnace is required to be >2.0X10 -3 Pa, argon is filled as a protective gas, the melting temperature is 1500-1700℃, the number of melting times is ≥12 times, and the time of each melting to maintain the molten state is >5min; S604: After cooling, take out the ingot, seal the ingot in a vacuum tube and put it into a tube furnace, set the heating rate to 1 min / °C, set the temperature to 1300°C, keep it warm for 60 hours, and take out the sample after the tube furnace cools to room temperature.

9. The method for designing alloys by optimizing phase diagrams using artificial intelligence according to claim 7, characterized in that: S7 includes the following steps: S701: Slice the prepared Pt-Ir-Al-Cr alloy sample, characterize the sample by X-ray diffraction, and observe and analyze the phase composition of the sample; S702: morphology analysis using scanning electron microscopy; S703: Use a digital microhardness tester to test the hardness of the sample; S704: Use an electronic universal mechanical properties testing machine to test the tensile strength, compressive strength and other properties of the sample.

Images