Data-driven high-entropy alloy phase composition prediction method and device

Abstract

The present application discloses a data-driven method and device for predicting the phase composition of a high-entropy alloy. The method includes: obtaining at least one phase composition feature of the high-entropy alloy to be measured; screening an optimal feature combination of the high-entropy alloy to be measured from the at least one phase composition feature; and based on the optimal feature combination, using a pre-trained phase composition prediction model to predict the phase composition result of the high-entropy alloy to be measured, where the phase composition result includes one of a solid solution phase, an intermetallic compound, and an amorphous state. Thus, the technical problem in the related art of predicting the phase composition of a high-entropy alloy through high-throughput experimental exploration or by means of semi-empirical phase diagram calculation, which is time-consuming and laborious and difficult to ensure the accuracy of prediction, thereby seriously affecting the development process of high-performance alloys, is solved.

Description

Technical Field

[0001] This application relates to the technical field of metal material phase composition design, and particularly relates to a data-driven high-entropy alloy phase composition prediction method and device. Background Art

[0002] In the past, the development of metal alloys mainly relied on a single main component and the addition of other rare metal elements to improve performance, but this development method has gradually reached a bottleneck.

[0003] In contrast, high-entropy alloys are composed of a mixture of multiple elements, thus having broad development space. At the same time, high-entropy alloys have various advantages such as mechanical properties, wear resistance, and radiation resistance. The types and contents of elements can be adjusted according to usage requirements to meet the service requirements of different scenarios. This unique design concept has led to the development of many high-entropy alloys with application value and prospects.

[0004] Accurately predicting the phase composition of high-entropy alloys is one of the important topics in the development of high-entropy alloys. The phase composition plays an important role in influencing the properties of alloys. The complex thermodynamic evolution and microscopic structure interaction inside pose great challenges to accurately predicting the tissue composition of alloys and then determining high-entropy alloys with development potential.

[0005] In related technologies, predicting the phase composition of high-entropy alloys mainly relies on high-throughput experimental exploration or semi-empirical phase diagram calculation. However, related technologies are time-consuming and laborious, and it is difficult to ensure the accuracy of prediction, seriously affecting the development process of high-performance alloys, and urgent improvement is needed.

[0006] Application Content

[0007] This application provides a data-driven high-entropy alloy phase composition prediction method and device to solve the technical problems in related technologies that predicting the phase composition of high-entropy alloys through high-throughput experimental exploration or semi-empirical phase diagram calculation is time-consuming and laborious, and it is difficult to ensure the accuracy of prediction, thereby seriously affecting the development process of high-performance alloys.

[0008] The first aspect embodiment of this application provides a data-driven high-entropy alloy phase composition prediction method, including the following steps: obtaining at least one phase composition feature of a high-entropy alloy to be measured; screening the optimal feature combination of the high-entropy alloy to be measured from the at least one phase composition feature; and based on the optimal feature combination, using a pre-trained phase composition prediction model to predict the phase composition result of the high-entropy alloy to be measured, where the phase composition result includes one of a solid solution phase, an intermetallic compound, and an amorphous state.

[0009] Optionally, in an embodiment of the present application, before using the pre-trained phase composition prediction model, it further includes: obtaining a phase composition training data set of the high-entropy alloy; calculating at least one feature of the high-entropy alloy from the phase composition training data set; calculating the Pearson correlation coefficient between the features, and generating an optimal feature combination according to the Pearson correlation coefficient between the features; training a machine learning model using the optimal feature combination, and determining the phase composition prediction model based on the target accuracy condition.

[0010] Optionally, in an embodiment of the present application, after determining the phase composition prediction model based on the target accuracy condition, it further includes: obtaining the importance ranking of the features according to the importance analysis attribute in the algorithm, and performing grid division in the feature space to obtain the decision boundary of the phase composition prediction model.

[0011] Optionally, in an embodiment of the present application, the machine learning model includes at least one of a K-nearest neighbor model, a support vector machine model, a decision tree model, a random forest model, and an extreme gradient boosting algorithm model.

[0012] Optionally, in an embodiment of the present application, the at least one includes configurational entropy, mixing enthalpy, average valence electron concentration, valence electron concentration difference, average electronegativity, electronegativity difference, atomic radius difference, melting point difference, Ω parameter, λ parameter, local atomic distortion parameter, γ parameter, δH mix dimensionless mixing enthalpy variance, dimensionless mixing enthalpy variance, dimensionless mixing enthalpy variance and at least one of dimensionless mixing enthalpy variance.

[0013] An embodiment of the second aspect of the present application provides a data-driven high-entropy alloy phase composition prediction device, including: a feature acquisition module for acquiring at least one phase composition feature of a high-entropy alloy to be measured; a screening module for screening an optimal feature combination of the high-entropy alloy to be measured from the at least one phase composition feature; and a prediction module for predicting the phase composition result of the high-entropy alloy to be measured based on the optimal feature combination using a pre-trained phase composition prediction model, where the phase composition result includes one of a solid solution phase, an intermetallic compound, and an amorphous state.

[0014] Optionally, in an embodiment of the present application, the prediction module includes: an acquisition unit configured to acquire a phase composition training data set of a high-entropy alloy; a calculation unit configured to calculate at least one feature of the high-entropy alloy from the phase composition training data set; a generation unit configured to calculate the Pearson correlation coefficient between the features and generate an optimal feature combination according to the Pearson correlation coefficient between the features; and a training unit configured to train a machine learning model by using the optimal feature combination and determine the phase composition prediction model based on the target accuracy condition.

[0015] Optionally, in an embodiment of the present application, the training unit is further configured to: obtain the importance ranking of the features according to the importance analysis attribute in the algorithm, and perform grid division in the feature space to obtain the decision boundary of the phase composition prediction model.

[0016] Optionally, in an embodiment of the present application, the machine learning model includes at least one of a K-nearest neighbor model, a support vector machine model, a decision tree model, a random forest model, and an extreme gradient boosting algorithm model.

[0017] Optionally, in an embodiment of the present application, the at least one includes configurational entropy, mixing enthalpy, average valence electron concentration, valence electron concentration difference, average electronegativity, electronegativity difference, atomic radius difference, melting point difference, Ω parameter, λ parameter, local atomic distortion parameter, γ parameter, δH mix dimensionless mixing enthalpy variance, dimensionless mixing enthalpy variance, dimensionless mixing enthalpy variance and at least one of dimensionless mixing enthalpy variances.

[0018] An embodiment of the third aspect of the present application provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the program to implement the data-driven high-entropy alloy phase composition prediction method as described in the above embodiments.

[0019] An embodiment of the fourth aspect of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause the computer to execute the data-driven high-entropy alloy phase composition prediction method as described in the above embodiments.

[0020] Embodiments of the present application can screen the obtained phase composition features to obtain an optimal feature combination, and use a pre-trained phase composition prediction model to predict the optimal feature combination, thereby accurately predicting the phase composition result of the high-entropy alloy to be measured, improving the prediction accuracy, effectively reducing the workload of experimentally searching for stable high-entropy alloys, facilitating the acceleration of the development process and saving cost resources and time, while effectively meeting the phase composition prediction requirements of high-entropy alloys and enhancing the prediction experience. Thus, the technical problem in the related art of predicting the phase composition of high-entropy alloys through high-throughput experimental exploration or semi-empirical phase diagram calculation, which is time-consuming and laborious and difficult to ensure prediction accuracy, thereby seriously affecting the development process of high-performance alloys, is solved.

[0021] Additional aspects and advantages of the present application will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and / or additional aspects and advantages of the present application will become apparent and be readily understood from the following description of the embodiments in conjunction with the drawings, where:

[0023] Figure 1 is a flowchart of a data-driven high-entropy alloy phase composition prediction method according to an embodiment of the present application;

[0024] Figure 2 is a heat map of the correlation of characteristic parameters of a data-driven high-entropy alloy phase composition prediction method according to an embodiment of the present application;

[0025] Figure 3 is a histogram of the accuracy of each machine learning model of a data-driven high-entropy alloy phase composition prediction method according to an embodiment of the present application on a validation set;

[0026] Figure 4 is a histogram of the importance ranking of each characteristic parameter by the XGBoost model of a data-driven high-entropy alloy phase composition prediction method according to an embodiment of the present application;

[0027] Figure 5 is a decision boundary diagram of the XGBoost model of a data-driven high-entropy alloy phase composition prediction method according to an embodiment of the present application;

[0028] Figure 6 is a schematic diagram of the principle of a data-driven high-entropy alloy phase composition prediction method according to an embodiment of the present application;

[0029] Figure 7 is a schematic structural diagram of a data-driven high-entropy alloy phase composition prediction system according to an embodiment of the present application

[0030] Figure 8 It is a schematic structural diagram of an electronic device provided according to an embodiment of the present application. Detailed implementation manners

[0031] The embodiments of the present application will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are intended to explain the present application, and should not be construed as a limitation to the present application.

[0032] The data-driven high-entropy alloy phase composition prediction method and device according to the embodiments of the present application will be described below with reference to the accompanying drawings. Aiming at the technical problem in the related art mentioned in the above background technology that predicting the phase composition of high-entropy alloys through high-throughput experimental exploration or semi-empirical phase diagram calculation is time-consuming and laborious, and it is difficult to ensure the accuracy of prediction, which seriously affects the development process of high-performance alloys, the present application provides a data-driven high-entropy alloy phase composition prediction method. In this method, the obtained phase composition features can be screened to obtain an optimal feature combination, and the pre-trained phase composition prediction model is used to predict the optimal feature combination, so as to accurately predict the phase composition result of the high-entropy alloy to be measured, improve the accuracy of prediction, effectively reduce the workload of experimentally searching for stable high-entropy alloys, facilitate accelerating the development process and saving cost resources and time, while effectively meeting the phase composition prediction requirements of high-entropy alloys and improving the prediction experience. Thus, the technical problem in the related art that predicting the phase composition of high-entropy alloys through high-throughput experimental exploration or semi-empirical phase diagram calculation is time-consuming and laborious, and it is difficult to ensure the accuracy of prediction, which seriously affects the development process of high-performance alloys is solved.

[0033] Specifically, Figure 1 It is a schematic flowchart of a data-driven high-entropy alloy phase composition prediction method provided by an embodiment of the present application.

[0034] As Figure 1 shown, the data-driven high-entropy alloy phase composition prediction method includes the following steps:

[0035] In step S101, at least one phase composition feature of the high-entropy alloy to be measured is obtained.

[0036] In the actual execution process, the embodiment of the present application can obtain at least one phase composition feature of the high-entropy alloy to be measured, providing a basis for subsequent screening of the optimal feature combination of the high-entropy alloy.

[0037] In step S102, the optimal feature combination of the high-entropy alloy to be measured is screened from at least one phase composition feature.

[0038] Specifically, after obtaining at least one phase composition feature of the high-entropy alloy to be measured in the embodiments of the present application, screening can be performed on it, and then an optimal feature combination of the high-entropy alloy to be measured can be obtained. Specifically, in the embodiments of the present application, the obtained phase composition features can be screened to obtain an optimal feature combination, which is convenient for subsequent prediction of the high-entropy alloy phase composition using a pre-trained phase composition prediction model, conducive to improving the accuracy of analysis, reducing the workload of experimentally searching for stable high-entropy alloys, accelerating the development process, and saving cost resources and time.

[0039] In step S103, based on the optimal feature combination, using the pre-trained phase composition prediction model, the phase composition result of the high-entropy alloy to be measured is predicted, where the phase composition result includes one of a solid solution phase, an intermetallic compound, and an amorphous state.

[0040] As a possible implementation manner, in the embodiments of the present application, a pre-trained phase composition prediction model can be used to predict and obtain the phase composition result of the high-entropy alloy to be measured, and the phase composition result includes one of a solid solution phase, an intermetallic compound, and an amorphous state, where the pre-trained phase composition model will be elaborated in detail later. In the embodiments of the present application, the obtained phase composition features can be screened to obtain an optimal feature combination, and the pre-trained phase composition prediction model can be used to predict the optimal feature combination, and then the phase composition result of the high-entropy alloy to be measured can be obtained, which is conducive to improving the accuracy of analysis, reducing the workload of experimentally searching for stable high-entropy alloys, accelerating the development process, and saving cost resources and time.

[0041] Optionally, in an embodiment of the present application, before using the pre-trained phase composition prediction model, it further includes: obtaining a phase composition training data set of the high-entropy alloy; calculating at least one feature of the high-entropy alloy from the phase composition training data set; calculating the Pearson correlation coefficient between the features, and generating an optimal feature combination according to the Pearson correlation coefficient between the features; training a machine learning model using the most characteristic combination, and determining a phase composition prediction model based on the target accuracy condition.

[0042] Here, the establishment of the pre-trained phase composition prediction model will be elaborated in detail.

[0043] The establishment of the pre-trained phase composition prediction model mainly includes the following steps:

[0044] 1. Obtain a high-entropy alloy phase composition data set. Among them, the high-entropy alloy phase composition is one of a solid solution phase, an intermetallic compound, and an amorphous state.

[0045] 2. Design a feature parameter calculation program according to the high-entropy alloy chemical formula. In the embodiments of the present application, a feature parameter calculation program can be designed according to the high-entropy alloy chemical formula, which is conducive to improving the applicability and practicability of the phase composition prediction model, and is conducive to improving the accuracy of the prediction analysis result of the phase composition prediction model.

[0046] 3. Screen the optimal feature combination according to feature engineering. During the actual execution process, the embodiments of the present application can screen at least one phase composition feature obtained and obtain the optimal feature combination. The embodiments of the present application can calculate the Pearson correlation coefficient between all features. As Figure 2 shown, generally speaking, for features with a relatively large correlation, the content they describe is closer. Therefore, the embodiments of the present application can choose to eliminate these three features δ r , δ χ , to ensure that the Pearson coefficient between the remaining 13 features is less than 0.9, which helps to reduce the training difficulty and prevent dimensional explosion.

[0047] 4. Train a machine learning model and determine the best model according to the accuracy of each model on the validation set. It can be understood that the embodiments of the present application can randomly select 80% of the data from the data set as the training set, and the remaining 20% as the validation set, use 5-fold cross-validation to improve the data set utilization efficiency, and determine the best model according to the accuracy of each model on the validation set.

[0048] 5. Evaluate the applicability of the model and the importance of each feature parameter. The embodiments of the present application can obtain the importance ranking of features according to the importance analysis attribute in the algorithm, and then realize the evaluation of the applicability of the model and the importance of each feature parameter, so as to ensure the accuracy of the model output result.

[0049] The embodiments of the present application can use the above-mentioned pre-established and trained phase composition prediction model to predict the phase composition of unknown high-entropy alloys, which is beneficial to improving the accuracy of analysis, reducing the workload of experimental search for stable high-entropy alloys, accelerating the development process, and saving cost resources and time.

[0050] Optionally, in an embodiment of the present application, at least one includes configurational entropy, mixing enthalpy, average valence electron concentration, valence electron concentration difference, average electronegativity, electronegativity difference, atomic radius difference, melting point difference, Ω parameter, λ parameter, local atomic distortion parameter, γ parameter, δH mix dimensionless mixing enthalpy variance, dimensionless mixing enthalpy variance, dimensionless mixing enthalpy variance and at least one of the dimensionless mixing enthalpy variances.

[0051] Specifically, the embodiments of the present application can design a feature parameter calculation program according to the high-entropy alloy chemical formula, where the feature parameters are shown in Table 1.

[0052] Table 1

[0053]

[0054] In summary, the characteristic parameters include: configurational entropy, mixing enthalpy, average valence electron concentration, valence electron concentration difference, average electronegativity, electronegativity difference, atomic radius difference, melting point difference, Ω parameter, λ parameter, local atomic distortion parameter, γ parameter, δH mix Dimensionless mixing enthalpy variance, Dimensionless mixing enthalpy variance, Dimensionless mixing enthalpy variance sum, and Dimensionless mixing enthalpy variance.

[0055] It should be noted that an alloy composition identification program can be introduced before calculating the characteristic parameters in the embodiments of the present application to prevent data duplication.

[0056] Optionally, in an embodiment of the present application, the machine learning model includes at least one of a K-nearest neighbor model, a support vector machine model, a decision tree model, a random forest model, and an extreme gradient boosting algorithm model.

[0057] For example, the machine learning model participating in the training can be at least one of K-nearest neighbor (KNN), support vector machine (SVM), decision tree (DT), random forest (RF), and extreme gradient boosting algorithm (XGBoost). In the embodiments of the present application, the best model can be determined according to the accuracy of each model on the validation set. As Figure 3 shown, among them, the best performing is the XGBoost algorithm model with an accuracy of 0.85. Hyperparameter tuning is performed by the grid search method. The XGBoost model parameters are: the number of weak learners n_estimators: 90, the depth of the tree max_depth: 12, and the learning rate controlling the weight reduction system of each weak learner: 0.2.

[0058] Optionally, in an embodiment of the present application, after determining the phase composition prediction model based on the target accuracy condition, it further includes: obtaining the importance ranking of features according to the importance analysis attribute in the algorithm, and performing grid division in the feature space to obtain the decision boundary of the phase composition prediction model.

[0059] Furthermore, as Figure 4 shown, it can be seen that different classification models have different sensitivities to features. Geometric features represented by atomic radius, such as λ, and thermodynamic features represented by entropy and enthalpy, such as ΔH mix , these features have a high influence factor when training the model. However, chemical features represented by electronegativity, such as δ χ , these features have a relatively small influence on training the model. In the embodiments of the present application, two parameters, ΔH, representing the geometric features and thermodynamic features of high-entropy alloys, can be selected according to the above feature importance analysis. mix, γ, using them as reference coordinates for the distribution in the feature space, and according to the trained XGBoost model, performing grid division calculations in the space to obtain the decision boundaries for each category in the model, and then plotting the distribution map of the validation set data in the decision space. As Figure 5 shown, the XGBoost model has a large number of decision tree calculations. By adding a regularization term to prevent overfitting, its decision boundaries are usually regular and can still be distinguished even when the sample space distributions are similar. The decision boundaries are fine and specific, indicating that the XGBoost model demonstrates good applicability and effectiveness in the prediction of high-entropy alloy phase compositions.

[0060] Next, in combination with Figures 2 to 6 shown, a specific example is used to elaborate in detail the working principle of the data-driven high-entropy alloy phase composition prediction method according to the embodiments of the present application. As Figure 6 shown, the embodiments of the present application include the following steps:

[0061] Step S601: Obtain a high-entropy alloy phase composition data set. The embodiments of the present application can obtain at least one phase composition feature of the high-entropy alloy to be measured, providing a basis for subsequent screening of the optimal feature combination of the high-entropy alloy. Among them, the high-entropy alloy phase composition is one of a solid solution phase, an intermetallic compound, and an amorphous state.

[0062] Step S602: Design a feature parameter calculation program according to the high-entropy alloy chemical formula. Specifically, the embodiments of the present application can design a feature parameter calculation program according to the high-entropy alloy chemical formula, where the feature parameters include: configurational entropy, mixing enthalpy, average valence electron concentration, valence electron concentration difference, average electronegativity, electronegativity difference, atomic radius difference, melting point difference, Ω parameter, λ parameter, local atomic distortion parameter, γ parameter, δH mix dimensionless mixing enthalpy variance, dimensionless mixing enthalpy variance, dimensionless mixing enthalpy variance, and dimensionless mixing enthalpy variance.

[0063] It should be noted that the embodiments of the present application can introduce an alloy composition identification program before calculating the feature parameters to prevent data duplication.

[0064] Step S603: Screen the optimal feature combination according to feature engineering. The embodiments of the present application can calculate the Pearson correlation coefficients between all features. As Figure 2 shown, generally speaking, for features with a relatively large correlation, the content they describe is closer. Therefore, the embodiments of the present application can choose to eliminate these three features δ r , δ χ , ensuring that the Pearson coefficients between the remaining 13 features are less than 0.9, which helps to reduce the training difficulty and prevent dimensional explosion.

[0065] Step S604: Train and select a machine learning model. In an embodiment of the present application, 80% of the data can be randomly selected from the data set as the training set, and the remaining 20% as the validation set. 5-fold cross-validation is adopted to improve the data set utilization efficiency, and the best model is determined according to the accuracy of each model on the validation set.

[0066] For example, the machine learning models participating in the training can be at least one of K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). In an embodiment of the present application, the best model can be determined according to the accuracy of each model on the validation set. As Figure 3 shown, among them, the best-performing is the XGBoost algorithm model, with an accuracy of 0.85. Hyperparameter tuning is performed by the grid search method. The XGBoost model parameters are: the number of weak learners n_estimators: 90, the depth of the tree max_depth: 12, and the weight reduction coefficient learning_rate: 0.2 that controls the weight of each weak learner.

[0067] Step S605: Conduct feature importance and model applicability analysis. In an embodiment of the present application, the importance ranking of features can be obtained according to the importance analysis attributes in the algorithm, and then the applicability of the XGBoost model and the importance of each feature parameter are evaluated.

[0068] As Figure 4 shown, it can be seen from this that different classification models have different sensitivities to features. Geometric features represented by atomic radius, such as λ, and thermodynamic features represented by entropy and enthalpy, such as ΔH mix , these features have a high influence factor when training the model. However, chemical features represented by electronegativity, such as δ χ , these features have a relatively small influence on training the model. In an embodiment of the present application, according to the above feature importance analysis, two parameters ΔH mix and γ that respectively represent the geometric features and thermodynamic features of high-entropy alloys can be selected as the reference coordinates for the feature space distribution. And according to the trained XGBoost model, grid division calculation is performed in the space to obtain the decision boundary of each category in the model, and then the distribution map of the validation set data in the decision space is drawn. As Figure 5 shown, the XGBoost model has a large number of decision tree calculations. By adding a regularization term to prevent overfitting, its decision boundary is usually regular and can still be distinguished even when the sample space distributions are similar. The decision boundary is fine and specific. The XGBoost model shows good applicability and effectiveness in the prediction of high-entropy alloy phase composition.

[0069] Step S606: Predict the unknown high-entropy alloy according to the trained model.

[0070] It should be noted that in the embodiment of the present application, a prediction model is established and trained through steps S601 - S605. In the actual implementation process, in the embodiment of the present application, only the chemical formula of the high-entropy alloy to be predicted needs to be input into the prediction model, and the prediction analysis of the phase composition of the high-entropy alloy can be obtained.

[0071] Specifically, in the embodiment of the present application, the chemical formula of the high-entropy alloy to be predicted can be input into the program, and the trained XGBoost model is used for prediction and output. Here, taking three different types of high-entropy alloys as examples, as shown in Table 2, the comparison with the actual results shows that the model can correctly predict the phase compositions of these three high-entropy alloys, achieving the prediction effect. Among them, Table 2 is the phase composition prediction table.

[0072] Table 2

[0073] alloy prediction result actual result FeMnCoNi solid solution phase solid solution phase AlCoCrCuMnFe intermetallic compound intermetallic compound AlCrTaTiZr amorphous amorphous

[0074] According to the data-driven high-entropy alloy phase composition prediction method proposed by the embodiment of the present application, the obtained phase composition features can be screened to obtain the optimal feature combination, and the pre-trained phase composition prediction model is used to predict the optimal feature combination, thereby accurately predicting the phase composition result of the high-entropy alloy to be tested, improving the prediction accuracy, effectively reducing the workload of experimentally searching for stable high-entropy alloys, facilitating the acceleration of the development process and saving cost resources and time, while effectively meeting the phase composition prediction requirements of high-entropy alloys and enhancing the prediction experience. Thus, the technical problem in the related art of predicting the phase composition of high-entropy alloys through high-throughput experimental exploration or semi-empirical phase diagram calculation, which is time-consuming and laborious and difficult to ensure the prediction accuracy, and thus seriously affects the development process of high-performance alloys, is solved.

[0075] Next, a data-driven high-entropy alloy phase composition prediction device according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0076] Figure 7 It is a block diagram of the data-driven high-entropy alloy phase composition prediction device according to the embodiment of the present application.

[0077] As Figure 7 shown, the data-driven high-entropy alloy phase composition prediction device 10 includes: a feature acquisition module 100, a screening module 200, and a prediction module 300.

[0078] Specifically, the feature acquisition module 100 is used to acquire at least one phase composition feature of the high-entropy alloy to be tested.

[0079] The screening module 200 is used to screen the optimal feature combination of the high-entropy alloy to be measured from at least one phase composition feature.

[0080] The prediction module 300 is used to predict the phase composition result of the high-entropy alloy to be measured based on the optimal feature combination by using a pre-trained phase composition prediction model, where the phase composition result includes one of a solid solution phase, an intermetallic compound, and an amorphous state.

[0081] Optionally, in an embodiment of the present application, the prediction module 300 includes: an acquisition unit, a calculation unit, a generation unit, and a training unit.

[0082] Among them, the acquisition unit is used to acquire the phase composition training data set of the high-entropy alloy.

[0083] The calculation unit is used to calculate at least one feature of the high-entropy alloy from the phase composition training data set.

[0084] The generation unit is used to calculate the Pearson correlation coefficient between features and generate an optimal feature combination according to the Pearson correlation coefficient between features.

[0085] The training unit is used to train a machine learning model by using the most characteristic combination and determine a phase composition prediction model based on the target accuracy condition.

[0086] Optionally, in an embodiment of the present application, the training unit is further used to obtain the importance ranking of features according to the importance analysis attribute in the algorithm and perform grid division in the feature space to obtain the decision boundary of the phase composition prediction model.

[0087] Optionally, in an embodiment of the present application, the machine learning model includes at least one of a K-nearest neighbor model, a support vector machine model, a decision tree model, a random forest model, and an extreme gradient boosting algorithm model.

[0088] Optionally, in an embodiment of the present application, at least one includes configurational entropy, mixing enthalpy, average valence electron concentration, valence electron concentration difference, average electronegativity, electronegativity difference, atomic radius difference, melting point difference, Ω parameter, λ parameter, local atomic distortion parameter, γ parameter, δH mix Dimensionless mixing enthalpy variance, Dimensionless mixing enthalpy variance, Dimensionless mixing enthalpy variance and At least one of the dimensionless mixing enthalpy variances.

[0089] It should be noted that the foregoing explanation of the embodiments of the data-driven high-entropy alloy phase composition prediction method also applies to the data-driven high-entropy alloy phase composition prediction device of this embodiment, and will not be repeated here.

[0090] The data-driven high-entropy alloy phase composition prediction device proposed according to the embodiments of the present application can screen the obtained phase composition features, obtain the optimal feature combination, and use the pre-trained phase composition prediction model to predict the optimal feature combination, thereby accurately predicting the phase composition result of the high-entropy alloy to be tested, improving the prediction accuracy, effectively reducing the workload of experimentally searching for stable high-entropy alloys, facilitating the acceleration of the development process and saving cost resources and time, while effectively meeting the phase composition prediction requirements of high-entropy alloys and enhancing the prediction experience. Thus, it solves the technical problem in the related art that predicting the phase composition of high-entropy alloys through high-throughput experimental exploration or semi-empirical phase diagram calculation is time-consuming and laborious, and it is difficult to ensure the prediction accuracy, which seriously affects the development process of high-performance alloys.

[0091] Figure 8 It is a schematic structural diagram of the electronic device provided by the embodiments of the present application. The electronic device may include:

[0092] A memory 801, a processor 802, and a computer program stored on the memory 801 and executable on the processor 802.

[0093] When the processor 802 executes the program, it implements the data-driven high-entropy alloy phase composition prediction method provided in the above embodiments.

[0094] Furthermore, the electronic device further includes:

[0095] A communication interface 803 for communication between the memory 801 and the processor 802.

[0096] The memory 801 is used to store the computer program executable on the processor 802.

[0097] The memory 801 may include a high-speed RAM memory, and may also include non-volatile memory, such as at least one disk memory.

[0098] If the memory 801, the processor 802, and the communication interface 803 are implemented independently, the communication interface 803, the memory 801, and the processor 802 may be interconnected through a bus and complete communication with each other. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8It is represented only by a thick line, but it does not mean that there is only one bus or one type of bus.

[0099] Optionally, in a specific implementation, if the memory 801, the processor 802, and the communication interface 803 are integrated on a single chip, the memory 801, the processor 802, and the communication interface 803 can communicate with each other through an internal interface.

[0100] The processor 802 may be a central processing unit (CPU for short), or an application specific integrated circuit (ASIC for short), or one or more integrated circuits configured to implement the embodiments of the present application.

[0101] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the above data-driven high-entropy alloy phase composition prediction method is implemented.

[0102] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, without conflict, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0103] In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "N" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0104] Any process or method description represented in a flowchart or otherwise described herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logical function or process. The scope of the preferred embodiments of the present application includes additional implementations where functions may be executed in a substantially simultaneous manner or in a reverse order according to the functions involved, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present application pertain.

[0105] The logic and / or steps represented in a flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing a logical function, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in conjunction with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, a computer-readable medium can even be paper or other suitable media on which the program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0106] It should be understood that the various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one or a combination of the following techniques known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0107] Those of ordinary skill in the art can understand that all or part of the steps carried out in implementing the above-described embodiment methods can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0108] In addition, each functional unit in various embodiments of the present application can be integrated in a processing module, can exist separately physically for each unit, or two or more units can be integrated in a module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. When the above-mentioned integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0109] The above-mentioned storage medium can be a read-only memory, a magnetic disk, an optical disc, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A data-driven method for predicting the phase composition of high-entropy alloys, characterized in that, It includes the following steps: Obtain at least one phase composition feature of the high-entropy alloy to be measured; Screen the optimal feature combination of the high-entropy alloy to be measured from the at least one phase composition feature, including: Obtain the phase composition training data set of the high-entropy alloy; Calculate at least one feature of the high-entropy alloy from the phase composition training data set; Calculate the Pearson correlation coefficient between features, and generate an optimal feature combination according to the Pearson correlation coefficient between the features; Train a machine learning model using the optimal feature combination, and determine the phase composition prediction model based on the target accuracy condition; Use the parameters ΔHmix and γ representing the geometric feature and thermodynamic feature of the high-entropy alloy respectively as the reference coordinates of the feature space distribution, perform grid division in the feature space to obtain the decision boundary of the phase composition prediction model, and then draw the distribution map of the validation set data in the decision space; and Based on the optimal feature combination, use the pre-trained phase composition prediction model to predict the phase composition result of the high-entropy alloy to be measured, where the phase composition result includes one of solid solution phase, intermetallic compound and amorphous state.

2. The method according to claim 1, characterized in that, The machine learning model includes at least one of K-nearest neighbor model, support vector machine model, decision tree model, random forest model and extreme gradient boosting algorithm model.

3. The method according to any one of claims 1-2, characterized in that, Said at least one includes configurational entropy, enthalpy of mixing, average valence electron concentration, valence electron concentration difference, average electronegativity, electronegativity difference, atomic radius difference, melting point difference, Ω parameter, λ parameter, local atomic distortion parameter, γ parameter, δH mix dimensionless enthalpy of mixing variance, dimensionless enthalpy of mixing variance, dimensionless enthalpy of mixing variance, and at least one of dimensionless enthalpy of mixing variances.

4. A data-driven high-entropy alloy phase composition prediction device, characterized in that, It includes: A feature acquisition module for obtaining at least one phase composition feature of the high-entropy alloy to be measured; A screening module for screening the optimal feature combination of the high-entropy alloy to be measured from the at least one phase composition feature; And A prediction module for predicting the phase composition result of the high-entropy alloy to be measured based on the optimal feature combination using the pre-trained phase composition prediction model, where the phase composition result includes one of solid solution phase, intermetallic compound and amorphous state; The prediction module includes: An acquisition unit for obtaining the phase composition training data set of the high-entropy alloy; A calculation unit for calculating at least one feature of the high-entropy alloy from the phase composition training data set; A generation unit for calculating the Pearson correlation coefficient between features and generating an optimal feature combination according to the Pearson correlation coefficient between the features; A training unit for training a machine learning model using the optimal feature combination and determining the phase composition prediction model based on the target accuracy condition; The training unit is further configured to: use the parameters ΔHmix and γ representing the geometric feature and thermodynamic feature of the high-entropy alloy respectively as the reference coordinates of the feature space distribution, perform grid division in the feature space to obtain the decision boundary of the phase composition prediction model, and then draw the distribution map of the validation set data in the decision space.

5. An electronic device, characterized in that, It includes: A memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the program to implement the data-driven high-entropy alloy phase composition prediction method according to any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to be used to implement the data-driven high-entropy alloy phase composition prediction method according to any one of claims 1-3.

Images