Design method, device and equipment of multi-main-element high-temperature alloy and storage medium

Abstract

The invention is applicable to the technical field of material science and metallurgical engineering, and provides a design method of a multi-principal-element high-temperature alloy, which comprises the following steps: randomly generating candidate alloys according to a preset alloy generation rule based on a candidate element set and the content range of each element in the candidate element set to obtain a candidate alloy set, and predicting the high-temperature compression yield strength of each candidate alloy in the candidate alloy set through the target prediction model to obtain a prediction result, screening out candidate alloys for alloy synthesis from the candidate alloy set according to the prediction result, and preparing the multi-principal-element high-temperature alloy based on the chemical components of the screened-out candidate alloys. Therefore, the development speed of the novel high-strength multi-main-element high-temperature alloy is increased, the material development period is shortened, and the development cost is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of material science and metallurgical engineering, and in particular relates to a design method, device, equipment and storage medium for a multi-main-element high-temperature alloy. Background Art

[0002] Over the past few decades, the development of nickel-based high-temperature alloys has greatly benefited the power generation industry and aerospace engineering. However, the development of nickel-based high-temperature alloys has approached their operating temperature limit, and they themselves are expensive and dense. In order to further improve the efficiency of aircraft engines and gas turbine engines and protect the environment, it is necessary to develop new high-temperature structural materials that are low in cost, light in weight, and resistant to high temperatures.

[0003] The discovery of multi-principal element alloys (MPEAs) provides a new paradigm for designing new high-performance structural materials. Unlike most traditional alloys developed based on one principal element, MPEAs are usually composed of three or more different elements, with no one element dominating. This multi-component element alloying approach provides a wide range of opportunities for composition design to obtain the desired combination of properties. Among them, MPEAs strengthened by L12 precipitation have an excellent combination of strength and ductility. In addition, due to the high entropy and slow diffusion effects, L12 precipitation-strengthened MPEAs exhibit remarkable thermal stability under long-term aging treatment at high temperatures. In addition to high stability, high concentrations of aluminum, iron, chromium, and titanium can also reduce the density and cost of MPEAs. Most reported MPEAs have a density below 8.0 g / cm 3 The microstructure of L12 precipitation-strengthened MPEAs is similar to that of nickel-based superalloys, and an abnormal increase in yield stress with increasing temperature is also observed. Therefore, such MPEAs are also called high-entropy superalloys or multi-principal-element superalloys (MPESAs). Based on these advantages, MPESAs show great potential in high-temperature engineering applications.

[0004] In the process of developing such materials, target-oriented key design strategies are prerequisites, including low-cost raw materials, low density, high L12 phase dissolution temperature, high melting temperature, good microstructural stability, strong oxidation resistance, excellent creep resistance, and high strength and acceptable ductility. Among them, the yield strength of the alloy at high temperature is the most critical evaluation index of high-temperature alloys. Therefore, exploring suitable chemical compositions with high yield strength and low density at high temperatures is crucial for developing the required materials. However, the chemical composition space of MPESAs is almost infinite, and it is challenging to find suitable compositions with excellent properties. Therefore, it is necessary to find methods that can quickly and accurately predict the high-temperature yield strength of MPESAs. Existing theoretical calculation methods such as first-principles calculations and molecular dynamics simulations cannot directly predict the high-temperature yield strength of alloys. Machine Learning (ML) is a simple and effective technology that has been applied in the field of materials science. It can establish potential mapping relationships between material composition, process, empirical parameters and material properties, and then help to quickly search the vast composition space through effective adjustment and optimization, thereby accelerating the development of materials with excellent properties. ML has been applied to predict the phase formation of MPEAs and superalloys, the L12 phase dissolution temperature and L12 phase volume fraction of superalloys, the hardness and room temperature yield strength of MPEAs, etc. Previous research has laid an important foundation and proved that ML can promote the accelerated design of alloys, thus saving a lot of manpower and time. However, there are no reports on the prediction of high temperature yield strength of MPESAs. Therefore, using ML technology to build a model that can quickly and accurately predict the high temperature yield strength of MPESAs is of great significance for accelerating the development of new high-strength MPESAs. Summary of the invention

[0005] The purpose of the present invention is to provide a design method, device, equipment and storage medium for a multi-main element high-temperature alloy, aiming to solve the problems of slow development speed and high development cost of new high-strength multi-main element high-temperature alloys caused by existing technologies.

[0006] In one aspect, the present invention provides a design method for a multi-main-element high-temperature alloy, the method comprising the following steps:

[0007] Based on the candidate element set and the content range of each element in the candidate element set, randomly generating candidate alloys according to a preset alloy generation rule to obtain a candidate alloy set;

[0008] Predicting the high temperature compressive yield strength of each candidate alloy in the candidate alloy set by using a target prediction model to obtain a prediction result;

[0009] According to the prediction result, a candidate alloy for alloy synthesis is screened out from the candidate alloy set;

[0010] A multi-main-element high-temperature alloy is prepared based on the chemical composition of the screened candidate alloy.

[0011] Preferably, before the step of predicting the high temperature compressive yield strength of each candidate alloy in the candidate alloy set by using the target prediction model, the method further comprises:

[0012] Constructing a training data set related to predicting high temperature compressive yield strength;

[0013] According to the training data set, a prediction model constructed by each regression algorithm in the regression algorithm set is trained and performance evaluated by 10-fold cross validation to obtain evaluation results, wherein each prediction model is used to predict high temperature compressive yield strength;

[0014] According to the evaluation result, a regression algorithm for constructing the target prediction model is determined from the regression algorithm set.

[0015] Preferably, the step of constructing a training data set related to predicting high temperature compressive yield strength comprises:

[0016] According to the preset phase type, a number of training samples are collected;

[0017] Performing a high temperature test on each of the training samples within a preset high temperature threshold range to obtain the high temperature compressive yield strength of each of the training samples at different test temperatures;

[0018] Collecting and constructing a series of characteristic variables for mapping the high temperature compressive yield strength for each training sample;

[0019] The training data set is constructed using the training samples and the chemical formula, test temperature, high temperature compressive yield strength and characteristic variables of the training samples.

[0020] Preferably, the step of randomly generating candidate alloys according to a preset alloy generation rule based on a candidate element set and a content range of each element in the candidate element set comprises:

[0021] According to a preset range of element types, randomly selecting an element for generating the candidate alloy from the candidate element set;

[0022] Determining the element content of each of the elements from the content ranges corresponding to each of the selected elements according to a preset first content threshold and a preset element content value step;

[0023] The candidate alloys are generated based on the elements and the content of the elements.

[0024] Preferably, the step of determining the element content of each of the elements from the content ranges corresponding to each of the selected elements according to the preset first content threshold and the preset element content value step size comprises:

[0025] When all the selected elements contain Ni and Co at the same time, the element content of each element is determined from the content range corresponding to each element according to the preset second content threshold, the first content threshold and the element content value step.

[0026] In another aspect, the present invention provides a design device for a multi-main-element high-temperature alloy, the device comprising:

[0027] A candidate alloy generation unit, configured to randomly generate candidate alloys according to a preset alloy generation rule based on a candidate element set and a content range of each element in the candidate element set, to obtain a candidate alloy set;

[0028] A yield strength prediction unit, used to predict the high temperature compressive yield strength of each candidate alloy in the candidate alloy set by using a target prediction model to obtain a prediction result;

[0029] A candidate alloy screening unit, used for screening candidate alloys for alloy synthesis from the candidate alloy set according to the prediction result;

[0030] The high-temperature alloy preparation unit is used to prepare a multi-main-element high-temperature alloy based on the chemical composition of the screened candidate alloy.

[0031] Preferably, the device further comprises:

[0032] A data set construction unit, used for constructing a training data set related to the prediction of high temperature compressive yield strength;

[0033] A prediction model training unit, used to train and evaluate the performance of the prediction model constructed by each regression algorithm in the regression algorithm set according to the training data set by using 10-fold cross validation to obtain an evaluation result, wherein each prediction model is used to predict the high temperature compressive yield strength;

[0034] A regression algorithm determination unit is used to determine a regression algorithm for constructing the target prediction model from the regression algorithm set according to the evaluation result.

[0035] Preferably, the data set construction unit comprises:

[0036] The alloy sample collection unit is used to collect a number of training samples according to a preset phase type;

[0037] A sample high temperature testing unit, used to perform a high temperature test on each of the training samples within a preset high temperature threshold range to obtain the high temperature compressive yield strength of each of the training samples at different test temperatures;

[0038] A characteristic variable collection unit, used for collecting and constructing a series of characteristic variables for mapping the high temperature compressive yield strength for each training sample;

[0039] The data set construction subunit is used to construct the training data set using the training sample and the chemical formula, test temperature, high temperature compressive yield strength and characteristic variables of the training sample.

[0040] On the other hand, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps described in the above-mentioned method for designing a multi-main-element high-temperature alloy are implemented.

[0041] On the other hand, the present invention further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps described in the above-mentioned method for designing a multi-main-element high-temperature alloy.

[0042] The present invention randomly generates candidate alloys based on a candidate element set and the content range of each element in the candidate element set according to a preset alloy generation rule to obtain a candidate alloy set, predicts the high-temperature compressive yield strength of each candidate alloy in the candidate alloy set through a target prediction model to obtain a prediction result, and screens out candidate alloys for alloy synthesis from the candidate alloy set according to the prediction result, and prepares a multi-main-element high-temperature alloy based on the chemical composition of the screened candidate alloy, thereby improving the development speed of new high-strength multi-main-element high-temperature alloys, shortening the material development cycle, and reducing development costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a flow chart for implementing the design method of a multi-main-element high-temperature alloy provided in the first embodiment of the present invention;

[0044] Figure 2 is a flow chart for implementing a design method for a multi-main-element high-temperature alloy provided in Embodiment 2 of the present invention;

[0045] Figure 3 It is a schematic diagram of the experimental true value and predicted value fitting results of the 10-fold cross validation of the ERTR algorithm in the design method of the multi-main element high-temperature alloy provided in the second embodiment of the present invention;

[0046] Figure 4: is a schematic diagram of the structure of a design device for a multi-main-element high-temperature alloy provided in Embodiment 3 of the present invention;

[0047] Figure 5 2 is a schematic diagram of the structure of a design device for a multi-main-element high-temperature alloy provided in a fourth embodiment of the present invention;

[0048] Figure 6 It is a schematic diagram of the structure of an electronic device provided in Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0050] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments:

[0051] Embodiment 1:

[0052] Figure 1 The implementation process of the design method of the multi-main element high-temperature alloy provided by the first embodiment of the present invention is shown. For the convenience of description, only the part related to the embodiment of the present invention is shown, which is described in detail as follows:

[0053] In step S101, based on the candidate element set and the content range of each element in the candidate element set, candidate alloys are randomly generated according to a preset alloy generation rule to obtain a candidate alloy set.

[0054] In an embodiment of the present invention, each candidate alloy in the candidate alloy set is a multi-principal element high-temperature alloy (i.e., MPESAs). Here, based on practical experience, the elements in Table 1 are selected as candidate elements of new MPESAs, and a reasonable value range is set for each element, as shown in Table 1. Based on the candidate element set composed of the elements in Table 1 and the content range corresponding to each element (lower limit (i.e., minimum content) and upper limit (i.e., maximum content)), candidate alloys are randomly generated according to preset alloy generation rules to obtain a candidate alloy set.

[0055] Table 1

[0056]

[0057] In one feasible embodiment, the generation of candidate alloys is achieved by the following steps:

[0058] (1) randomly selecting an element for generating a candidate alloy from a set of candidate elements according to a preset range of element types;

[0059] In an embodiment of the present invention, the alloy generation rule sets the number range of element types for each alloy chemical formula. According to the number range of element types, elements for generating candidate alloys are randomly selected from the candidate element set. Specifically, the alloy generation rule sets 4 to 7 elements for each alloy chemical formula (i.e., at least 4 and at most 7 elements constituting the alloy chemical formula), and 4 to 7 elements are randomly selected from the candidate element set {Ni, Co, Cr, Fe, Al, Ti, Ta, Zr, Nb, Hf, Mo, V} for generating candidate alloys. As an example, six elements, Al, Co, Fe, Nb, Ni, and Ti, are randomly selected from the candidate element set for generating candidate alloys.

[0060] (2) determining the element content of each element from the content range corresponding to each selected element according to a preset first content threshold and a preset element content value step;

[0061] In the embodiment of the present invention, the alloy generation rule further sets a first content threshold and an element content value step, wherein the first content threshold is the total content of all elements in each alloy chemical formula, and the element content value step refers to the minimum change amount of each adjustment when the element content is continuously adjusted. Here, the content range corresponding to each selected element is determined from the content range of each element in the candidate element set, and within the content range corresponding to each selected element, the content of each selected element is adjusted with the element content value step, so that the total content of all selected elements reaches the first content threshold.

[0062] In a feasible embodiment, the total content of all elements in each alloy chemical formula is set to be equal to 100 at %, and the step length of the element content is 0.1 at %.

[0063] In another feasible embodiment, when all the selected elements include Ni and Co at the same time, the element content of each element is determined from the content range corresponding to each element according to the preset second content threshold, the first content threshold and the element content value step.

[0064] In the embodiment of the present invention, a second content threshold is also set in the alloy generation rule, wherein the second content threshold is the maximum content of Ni+Co when Ni and Co are both included in the alloy chemical formula. Here, when both Ni and Co exist in all the selected elements, the element content of each element is determined from the content range corresponding to each element according to the second content threshold, the first content threshold and the element content value step. Specifically, the second content threshold is 80at%, that is, when Ni and Co are both included, the sum of the two does not exceed 80at%.

[0065] As an example, the elements selected from Table 1 are Al, Co, Fe, Nb, Ni, and Ti, and the content ranges corresponding to the selected elements are 0-15, 20-50, 0-25, 0-5, 20-50, and 0-15, respectively. Within the content range of these selected elements, the content of each selected element is adjusted by 0.1 at% so that the total content of all selected elements reaches 100 at%, and the content of Ni+Co does not exceed 80 at%. Here, a content scheme is obtained as 11.5, 26.6, 4.9, 3.4, 48.6, and 5, that is, the content of Al is 11.5 at%, the content of Co is 26.6 at%, the content of Fe is 4.9 at%, the content of Nb is 3.4 at%, the content of Ni is 48.6 at%, and the content of Ti is 5 at%.

[0066] (3) Generate candidate alloys based on the elements and their contents.

[0067] In the embodiment of the present invention, the corresponding candidate alloys are generated according to the selected elements and the determined element contents corresponding to the elements. For example, according to the elements Al, Co, Fe, Nb, Ni, Ti and the corresponding element contents of 11.5, 26.6, 4.9, 3.4, 48.6, 5, the candidate alloy Al11.5Co26.6Fe4.9Nb3.4Ni48.6Ti5 is generated.

[0068] The generation of candidate alloys is achieved through the above steps (1)-(3), thereby increasing the diversity of the candidate alloy set while ensuring the performance of the candidate alloys.

[0069] In step S102, the high temperature compressive yield strength of each candidate alloy in the candidate alloy set is predicted by the target prediction model to obtain a prediction result.

[0070] In an embodiment of the present invention, the target prediction model is a regression model pre-constructed based on a machine learning method for predicting the high-temperature compressive yield strength of multi-main-element high-temperature alloys. The high-temperature compressive yield strength of each candidate alloy in the candidate alloy set is predicted by the target prediction model to obtain a prediction result, which includes the predicted high-temperature compressive yield strength of each candidate alloy.

[0071] In step S103, candidate alloys for alloy synthesis are screened out from the candidate alloy set according to the prediction result.

[0072] In an embodiment of the present invention, according to the magnitude of each high temperature compressive yield strength in the prediction result, a candidate alloy corresponding to the maximum high temperature compressive yield strength is screened out from the candidate alloy set for use in subsequent alloy preparation.

[0073] In step S104, a multi-main-element high-temperature alloy is prepared based on the chemical composition of the screened candidate alloys.

[0074] In an embodiment of the present invention, the chemical composition of the candidate alloy includes various elements constituting the candidate alloy and the corresponding content of each element. When preparing a multi-main element high-temperature alloy based on the chemical composition of the screened candidate alloy, specifically, first, according to the chemical composition of the screened candidate alloy, a vacuum arc melting method is used to melt the elements to obtain a uniform alloy ingot. Thereafter, the ingot is placed in a high-temperature furnace and subjected to a solution treatment for a preset solution time at a preset high temperature threshold to eliminate or reduce the internal stress and structural defects generated by the alloy during the casting process. After the solution treatment is completed, the ingot is quickly put into water for quenching to quickly cool and retain the structural state at high temperature. Subsequently, according to a preset compression ratio, the quenched ingot is cold rolled to improve the microstructure of the alloy and improve its strength and hardness. Finally, the cold-rolled ingot is placed in a constant temperature environment of a preset constant temperature threshold for aging treatment for a preset aging time to improve the mechanical properties and stability of the alloy, thereby finally obtaining a novel high-strength multi-main element high-temperature alloy.

[0075] As an example, the candidate alloy screened out is Al11.5Co26.6Fe4.9Nb3.4Ni48.6Ti5. According to the chemical composition of the candidate alloy, the vacuum arc melting method is used to melt the elements to obtain a uniform alloy ingot with a total weight of 60 g. The ingot is solution treated at 1250°C for 12 hours, then water quenched, and then cold rolled with a reduction (i.e., compression ratio) of 30%. The cold rolled sample is aged at 800°C for 48 hours to obtain the final multi-main element high-temperature alloy.

[0076] Based on the multi-main element high-temperature alloy prepared in the above example, the high-temperature compression performance of the multi-main element high-temperature alloy was tested. Specifically, the sample after aging for 48 hours was cut into a preset size (for example, 4*4*6mm) of rectangular parallelepiped for testing the compression performance at different high temperatures. The compression test was carried out using a thermal simulation tester G3180, and the compression rate was set to 0.36mm / min. The high-temperature compression yield strength of the alloy at 600 / 700 / 800 / 900°C was tested respectively. The test results are shown in Table 2, wherein the predicted value in Table 2 is the high-temperature compression yield strength of the alloy predicted by the target prediction model, and the experimental test value is the experimental result obtained by the compression test of the alloy. In addition, the density of the alloy was tested by the Archimedes drainage method, and the density of the alloy was 7.83g / cm 3From the experimental results in Table 2, it can be seen that a new multi-principal element superalloy with a high yield strength prediction value at high temperature was screened based on the prediction results of the target prediction model, and a high-strength and lightweight multi-principal element superalloy with a high yield strength of 929 MPa at 700°C and a density of 7.83 g / cm3 was synthesized, which proves that the target prediction model of the present application can quickly and accurately predict the high-temperature compressive yield strength of MPESAs, which is of great significance for accelerating the development of new high-strength MPESAs.

[0077] Table 2

[0078] Test temperature (℃) Experimental test value (MPa) Predicted value (MPa) 600 825 774.25 700 929 807.97 800 713 728.17 900 550 550.66

[0079] In an embodiment of the present invention, based on a candidate element set and the content range of each element in the candidate element set, candidate alloys are randomly generated according to preset alloy generation rules to obtain a candidate alloy set, and the high-temperature compressive yield strength of each candidate alloy in the candidate alloy set is predicted by a target prediction model to obtain a prediction result. Based on the prediction result, candidate alloys for alloy synthesis are screened out from the candidate alloy set, and a multi-main-element high-temperature alloy is prepared based on the chemical composition of the screened candidate alloy, thereby improving the development speed of new high-strength multi-main-element high-temperature alloys, shortening the material development cycle, and reducing development costs.

[0080] Embodiment 2:

[0081] Figure 2 The implementation process of the design method of the multi-main element high-temperature alloy provided by the second embodiment of the present invention is shown. For the convenience of explanation, only the part related to the embodiment of the present invention is shown, which is described in detail as follows:

[0082] In step S201, a training data set related to predicting high temperature compressive yield strength is constructed.

[0083] In an embodiment of the present invention, a training data set is constructed for training a model for predicting high temperature compressive yield strength.

[0084] In a feasible embodiment, the construction of the training data set is achieved through the following steps:

[0085] (1) Collect several training samples according to the preset phase type;

[0086] In the embodiment of the present invention, the phase type is set to FCC type, and here, a number (for example, 67) of FCC type multi-main element alloys are collected as training samples.

[0087] (2) performing a high temperature test on each training sample within a preset high temperature threshold range to obtain the high temperature compressive yield strength of each training sample at different test temperatures;

[0088] In the embodiment of the present invention, the high temperature threshold range is set to 500-1100° C. Here, a high temperature test is performed on each multi-main element alloy within 500-1100° C. to obtain the high temperature compressive yield strength of each multi-main element alloy at different test temperatures.

[0089] (3) Collect and construct a series of characteristic variables for mapping high temperature compressive yield strength for each training sample;

[0090] In the embodiment of the present invention, the characteristic variables are key parameters describing the properties of the alloy sample, which include the heat treatment process parameters of the alloy, the content of various elements in the alloy (i.e., chemical composition), and commonly used empirical parameters, which should be able to fully reflect the physical, chemical and mechanical properties of the alloy. Specifically, Table 3 shows the representation and meaning of each characteristic variable used to map the high temperature compressive yield strength. Here, according to the influencing factors of the high temperature compressive yield strength of each training sample, the characteristic variables that are most likely to affect the high temperature compressive yield strength of the training sample are selected from all the features shown in Table 3, and the data of these characteristic variables are collected.

[0091] Table 3

[0092]

[0093] (4) Construct a training data set using the training samples and their chemical formulas, test temperatures, high temperature compressive yield strengths, and characteristic variables.

[0094] In the embodiment of the present invention, the training data set is constructed using all the collected training samples, and each training sample includes its alloy chemical formula, test temperature, high temperature compressive yield strength corresponding to the test temperature, and characteristic variables mapping the high temperature compressive yield strength.

[0095] In another feasible embodiment, after the training data set is constructed, the data in the training data set is preprocessed. Specifically, the text data is digitized. For example, in the Cooling Method category, "water quenching" is numbered 1, "air cooling" is numbered 2, "furnace cooling" is numbered 3, "oil quenching" is numbered 4, and other cooling methods or unmarked cooling methods are recorded as 0.

[0096] In step S202, based on the training data set, 10-fold cross validation is used to train and evaluate the performance of the prediction model constructed by each regression algorithm in the regression algorithm set to obtain an evaluation result, wherein each prediction model is used to predict the high temperature compressive yield strength.

[0097] In an embodiment of the present invention, each regression algorithm in a pre-provided regression algorithm set is used to construct a corresponding prediction model for predicting high-temperature compressive yield strength. When all regression algorithms select default parameters, 10-fold cross validation is used to train and evaluate the performance of each prediction model on a training data set to obtain an evaluation result.

[0098] Specifically, Table 4 shows the fourteen regression algorithms provided and the average results of 10-fold cross validation of each regression algorithm, where R is the average Pearson product-moment correlation coefficient (Pearson correlation coefficient) for predicting high temperature compressive yield strength, and R 2 is the average determination coefficient for predicting high temperature compressive yield strength, and RMSE is the root mean square error (RMSE) for predicting high temperature compressive yield strength.

[0099] Table 4

[0100]

[0101]

[0102] In step S203, based on the evaluation result, a regression algorithm for constructing a target prediction model is determined from the regression algorithm set.

[0103] In the embodiment of the present invention, according to the evaluation results shown in Table 4, a regression algorithm for constructing a target prediction model is determined from the 14 regression algorithms shown in Table 4. Specifically, it can be seen from Table 4 that the ERTR algorithm predicts R and R of high temperature compressive yield strength. 2 are the highest, reaching 0.9076 and 0.8236 respectively, and RMSE is the lowest among all algorithms. In addition, Figure 3 The experimental true value and predicted value fitting results of the 10-fold cross validation of the ERTR algorithm are shown. Figure 3 It can be seen that the true value and the predicted value have a good fit, indicating that the prediction model constructed by the ERTR algorithm has high accuracy. Therefore, the prediction model constructed by the ERTR algorithm and trained with the training data set is used as the target prediction model for subsequent alloy design and screening.

[0104] In an embodiment of the present invention, a training data set related to the prediction of high-temperature compressive yield strength is constructed. Based on the training data set, a prediction model constructed by each regression algorithm in a regression algorithm set is trained and performance evaluated using 10-fold cross validation to obtain an evaluation result, wherein each prediction model is used to predict the high-temperature compressive yield strength. Based on the evaluation result, a regression algorithm for constructing a target prediction model is determined from the regression algorithm set, so that the target prediction model can quickly and accurately predict the high-temperature compressive yield strength of the new multi-main element high-temperature alloy.

[0105] Embodiment three:

[0106] Figure 4 The structure of the design device for multi-main-element high-temperature alloy provided by the third embodiment of the present invention is shown. For the convenience of description, only the part related to the embodiment of the present invention is shown, including:

[0107] A candidate alloy generation unit 41 is used to randomly generate candidate alloys according to a preset alloy generation rule based on the candidate element set and the content range of each element in the candidate element set to obtain a candidate alloy set;

[0108] The yield strength prediction unit 42 is used to predict the high temperature compressive yield strength of each candidate alloy in the candidate alloy set by using a target prediction model to obtain a prediction result;

[0109] A candidate alloy screening unit 43 is used to screen candidate alloys for alloy synthesis from the candidate alloy set according to the prediction result;

[0110] The high temperature alloy preparation unit 44 is used to prepare a multi-main element high temperature alloy based on the chemical composition of the screened candidate alloys.

[0111] Wherein, preferably, the candidate alloy generation unit 41 includes:

[0112] An element selection unit, used for randomly selecting elements for generating candidate alloys from a set of candidate elements according to a preset range of element types;

[0113] A content determination unit, used to determine the element content of each element from the content range corresponding to each selected element according to a preset first content threshold and a preset element content value step;

[0114] The alloy generation unit is used to generate candidate alloys according to elements and element contents.

[0115] Still preferably, the content determination unit comprises:

[0116] The content determination subunit is used to determine the element content of each element from the content range corresponding to each element according to the preset second content threshold, the first content threshold and the element content value step when all the selected elements contain Ni and Co at the same time.

[0117] In the embodiment of the present invention, each unit of the multi-primary element high temperature alloy design device can be implemented by a corresponding hardware or software unit, and each unit can be an independent software or hardware unit, or can be integrated into a software or hardware unit, which is not intended to limit the present invention. Specifically, the implementation of each unit can refer to the description of the aforementioned embodiment 1, which will not be repeated here.

[0118] Embodiment 4:

[0119] Figure 5 The structure of the design device for multi-main-element high-temperature alloy provided by the fourth embodiment of the present invention is shown. For the convenience of description, only the part related to the embodiment of the present invention is shown, including:

[0120] A data set construction unit 51 is used to construct a training data set related to predicting high temperature compressive yield strength;

[0121] A prediction model training unit 52 is used to train and evaluate the performance of the prediction model constructed by each regression algorithm in the regression algorithm set according to the training data set by using 10-fold cross validation to obtain an evaluation result, wherein each prediction model is used to predict the high temperature compressive yield strength;

[0122] A regression algorithm determination unit 53, used to determine a regression algorithm for constructing a target prediction model from a set of regression algorithms according to the evaluation result;

[0123] A candidate alloy generation unit 54 is used to randomly generate candidate alloys according to a preset alloy generation rule based on the candidate element set and the content range of each element in the candidate element set to obtain a candidate alloy set;

[0124] The yield strength prediction unit 55 is used to predict the high temperature compressive yield strength of each candidate alloy in the candidate alloy set by using the target prediction model to obtain a prediction result;

[0125] A candidate alloy screening unit 56 is used to screen candidate alloys for alloy synthesis from the candidate alloy set according to the prediction result;

[0126] The high temperature alloy preparation unit 57 is used to prepare a multi-main element high temperature alloy based on the chemical composition of the screened candidate alloys.

[0127] Wherein, preferably, the data set construction unit 51 includes:

[0128] The alloy sample collection unit is used to collect a number of training samples according to a preset phase type;

[0129] A sample high temperature testing unit is used to perform a high temperature test on each training sample within a preset high temperature threshold range to obtain the high temperature compressive yield strength of each training sample at different test temperatures;

[0130] A characteristic variable collection unit, used for collecting and constructing a series of characteristic variables for mapping high temperature compressive yield strength for each training sample;

[0131] The data set construction subunit is used to construct a training data set using training samples and chemical formulas, test temperatures, high temperature compressive yield strengths, and characteristic variables of the training samples.

[0132] In the embodiment of the present invention, each unit of the multi-primary element high temperature alloy design device can be implemented by a corresponding hardware or software unit, and each unit can be an independent software or hardware unit, or can be integrated into a software or hardware unit, which is not intended to limit the present invention. Specifically, the implementation of each unit can refer to the description of the aforementioned embodiments 1 and 2, and will not be repeated here.

[0133] Embodiment five:

[0134] Figure 6 The structure of an electronic device provided by Embodiment 5 of the present invention is shown. For the convenience of description, only the part related to the embodiment of the present invention is shown.

[0135] The electronic device 6 of the embodiment of the present invention includes a processor 60, a memory 61, and a computer program 62 stored in the memory 61 and executable on the processor 60. When the processor 60 executes the computer program 62, the steps in the embodiment of the design method of the multi-main element high-temperature alloy are implemented, such as Figure 1 Alternatively, when the processor 60 executes the computer program 62, the functions of each unit in the above-mentioned device embodiments are realized, for example Figure 4 The functions of the units 41 to 44 are shown.

[0136] In an embodiment of the present invention, based on a candidate element set and the content range of each element in the candidate element set, candidate alloys are randomly generated according to preset alloy generation rules to obtain a candidate alloy set, and the high-temperature compressive yield strength of each candidate alloy in the candidate alloy set is predicted by a target prediction model to obtain a prediction result. Based on the prediction result, candidate alloys for alloy synthesis are screened out from the candidate alloy set, and a multi-main-element high-temperature alloy is prepared based on the chemical composition of the screened candidate alloy, thereby improving the development speed of new high-strength multi-main-element high-temperature alloys, shortening the material development cycle, and reducing development costs.

[0137] The steps implemented when the processor 60 in the electronic device 6 of the embodiment of the present invention executes the computer program 62 to implement the design method of the multi-main-element high-temperature alloy can refer to the description of the aforementioned method embodiment, which will not be repeated here.

[0138] Embodiment six:

[0139] In an embodiment of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned design method of multi-main-element high-temperature alloy are implemented, for example, Figure 1 Alternatively, when the computer program is executed by a processor, the functions of each unit in the above-mentioned device embodiments are realized, for example Figure 4 The functions of the units 41 to 44 are shown.

[0140] In an embodiment of the present invention, based on a candidate element set and the content range of each element in the candidate element set, candidate alloys are randomly generated according to preset alloy generation rules to obtain a candidate alloy set, and the high-temperature compressive yield strength of each candidate alloy in the candidate alloy set is predicted by a target prediction model to obtain a prediction result. Based on the prediction result, candidate alloys for alloy synthesis are screened out from the candidate alloy set, and a multi-main-element high-temperature alloy is prepared based on the chemical composition of the screened candidate alloy, thereby improving the development speed of new high-strength multi-main-element high-temperature alloys, shortening the material development cycle, and reducing development costs.

[0141] The computer-readable storage medium of the embodiment of the present invention may include any entity or device or recording medium capable of carrying computer program code, for example, ROM / RAM, magnetic disk, optical disk, flash memory and other memories.

[0142] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A design method for a multi-main-element high-temperature alloy, characterized in that: The method comprises the following steps: Based on the candidate element set and the content range of each element in the candidate element set, randomly generating candidate alloys according to a preset alloy generation rule to obtain a candidate alloy set; Predicting the high temperature compressive yield strength of each candidate alloy in the candidate alloy set by using a target prediction model to obtain a prediction result; According to the prediction result, a candidate alloy for alloy synthesis is screened out from the candidate alloy set; A multi-main-element high-temperature alloy is prepared based on the chemical composition of the screened candidate alloy.

2. The method according to claim 1, characterized in that Before the step of predicting the high temperature compressive yield strength of each candidate alloy in the candidate alloy set by using the target prediction model, the method further includes: Constructing a training data set related to predicting high temperature compressive yield strength; According to the training data set, a prediction model constructed by each regression algorithm in the regression algorithm set is trained and performance evaluated by 10-fold cross validation to obtain evaluation results, wherein each prediction model is used to predict high temperature compressive yield strength; According to the evaluation result, a regression algorithm for constructing the target prediction model is determined from the regression algorithm set.

3. The method according to claim 2, characterized in that The steps to construct a training dataset related to predicting high temperature compressive yield strength include: According to the preset phase type, a number of training samples are collected; Performing a high temperature test on each of the training samples within a preset high temperature threshold range to obtain the high temperature compressive yield strength of each of the training samples at different test temperatures; Collecting and constructing a series of characteristic variables for mapping the high temperature compressive yield strength for each training sample; The training data set is constructed using the training samples and the chemical formula, test temperature, high temperature compressive yield strength and characteristic variables of the training samples.

4. The method according to claim 1, characterized in that Based on the candidate element set and the content range of each element in the candidate element set, the step of randomly generating candidate alloys according to a preset alloy generation rule comprises: According to a preset range of element types, randomly selecting an element for generating the candidate alloy from the candidate element set; Determining the element content of each of the elements from the content ranges corresponding to each of the selected elements according to a preset first content threshold and a preset element content value step; The candidate alloys are generated based on the elements and the content of the elements.

5. The method according to claim 4, characterized in that The step of determining the element content of each of the elements from the content ranges corresponding to each of the selected elements according to a preset first content threshold and a preset element content value step size comprises: When all the selected elements contain Ni and Co at the same time, the element content of each element is determined from the content range corresponding to each element according to the preset second content threshold, the first content threshold and the element content value step.

6. A design device for multi-main-element high-temperature alloys, characterized in that: The device comprises: A candidate alloy generation unit, configured to randomly generate candidate alloys according to a preset alloy generation rule based on a candidate element set and a content range of each element in the candidate element set, to obtain a candidate alloy set; A yield strength prediction unit, used to predict the high temperature compressive yield strength of each candidate alloy in the candidate alloy set by using a target prediction model to obtain a prediction result; A candidate alloy screening unit, used for screening candidate alloys for alloy synthesis from the candidate alloy set according to the prediction result; The high-temperature alloy preparation unit is used to prepare a multi-main-element high-temperature alloy based on the chemical composition of the screened candidate alloy.

7. The device according to claim 6, characterized in that The device also includes: A data set construction unit, used for constructing a training data set related to the prediction of high temperature compressive yield strength; A prediction model training unit, used to train and evaluate the performance of the prediction model constructed by each regression algorithm in the regression algorithm set according to the training data set by using 10-fold cross validation to obtain an evaluation result, wherein each prediction model is used to predict the high temperature compressive yield strength; A regression algorithm determination unit is used to determine a regression algorithm for constructing the target prediction model from the regression algorithm set according to the evaluation result.

8. The device according to claim 7, characterized in that The data set construction unit comprises: The alloy sample collection unit is used to collect a number of training samples according to a preset phase type; A sample high temperature testing unit, used to perform a high temperature test on each of the training samples within a preset high temperature threshold range to obtain the high temperature compressive yield strength of each of the training samples at different test temperatures; A characteristic variable collection unit, used for collecting and constructing a series of characteristic variables for mapping the high temperature compressive yield strength for each training sample; The data set construction subunit is used to construct the training data set using the training sample and the chemical formula, test temperature, high temperature compressive yield strength and characteristic variables of the training sample.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images