Medium-high-entropy aluminum alloy performance prediction system and method and medium

Abstract

The invention relates to the technical field of metal materials, and provides a medium-and-high-entropy aluminum alloy performance prediction method, system and method and a medium, and the method comprises the following steps: obtaining a characteristic data set for medium-and-high-entropy aluminum alloy performance; constructing an MLP prediction model, and obtaining an optimal hyper-parameter combination of the MLP prediction model through ESOA; training the MLP prediction model according to the feature data set and the optimal hyper-parameter combination; and predicting the performance of the medium-high entropy aluminum alloy according to the trained MLP prediction model. According to the method, the advantages of the ESOA and the MLP prediction model are combined, the accuracy of performance prediction of the medium-high entropy aluminum alloy can be effectively improved, the time and experiment cost can be reduced, the prediction cost and the calculation amount can be reduced through the powerful calculation capacity of the MLP prediction model, the prediction precision and the generalization capacity of the model are enhanced, and the method is suitable for popularization and application. The method is suitable for performance prediction of the medium-high-entropy aluminum alloy material.

Description

Technical Field

[0001] The present invention relates to the technical field of metal materials, and in particular to a medium-high entropy aluminum alloy performance prediction system, method and medium. Background Art

[0002] In the field of materials science and engineering, with the development of Industry 4.0, the design and optimization of lightweight materials based on high entropy alloys are gradually becoming the focus of research. High entropy alloys are widely used in ultra-high temperature materials, high yield strength structural materials, and aerospace lightweight materials due to their excellent high strength, high hardness, and corrosion-resistant yield strength. Lightweight high entropy alloys combine lightweight elements (such as Al, Ti, Mg, Li, etc.) with transition elements (such as V, Cr, Zr, Nb, Mn, etc.) to significantly improve the mechanical yield strength of the material while maintaining its low density, becoming a new trend in material design.

[0003] In the prior art, some progress has been made in adjusting the aluminum content in high entropy alloys to optimize the microstructure and mechanical properties. 100-x Taking lightweight high entropy aluminum alloy as an example, research has found that the yield strength of the alloy can be increased by increasing the aluminum content. However, as the precipitate content increases, the plasticity of the alloy decreases significantly, affecting the practical application of the material. This yield strength trade-off makes it a key issue to maintain or even improve plasticity while increasing hardness and strength.

[0004] The traditional method to obtain the correspondence between the alloy composition and mechanical properties of medium- and high-entropy aluminum alloys is through experimental means. However, each experiment of this method requires multiple steps such as batching, smelting, heat treatment and testing, which has the disadvantages of being time-consuming and high in experimental cost. Summary of the invention

[0005] In order to solve the problems of long time consumption and high experimental cost in the prior art of experimentally studying the correspondence between the alloy composition and the mechanical properties of high entropy aluminum alloys, the present application provides a medium-high entropy aluminum alloy performance prediction system that can reduce time cost and experimental cost.

[0006] The medium-high entropy aluminum alloy performance prediction system comprises:

[0007] A data acquisition module, used to acquire characteristic data sets for medium and high entropy aluminum alloy performance;

[0008] A construction module, used to construct an MLP prediction model and obtain an optimal hyperparameter combination of the MLP prediction model through ESOA;

[0009] A training module, used for training the MLP prediction model according to the feature data set and the optimal hyperparameter combination;

[0010] The prediction module is used to predict the properties of medium and high entropy aluminum alloys based on the trained MLP prediction model.

[0011] The medium-high entropy aluminum alloy performance prediction system constructs an MLP prediction model, obtains the optimal hyperparameter combination of the MLP prediction model through ESOA, trains the MLP prediction model according to the feature data set and the optimal hyperparameter combination, and combines the advantages of ESOA and the MLP prediction model. It can not only effectively improve the accuracy of medium-high entropy aluminum alloy performance prediction, but also reduce time and experimental costs through the powerful computing power of the MLP prediction model, thereby solving the problems of long time consumption and high experimental costs in the prior art of studying the correspondence between the alloy composition of high entropy aluminum alloy and its mechanical properties through experimental means.

[0012] Furthermore, the data acquisition module is also used to clean, screen and standardize the feature data set.

[0013] Furthermore, the medium-high entropy aluminum alloy is Al x (TiVCrNb) 100-x Lightweight high entropy alloy system, the feature data set includes input features and output targets, the input features are the molar fractions of five metals, Al, Ti, V, Cr, and Nb, and the output target is yield strength.

[0014] The present application also provides a medium-high entropy aluminum alloy performance prediction method, which is applied to the medium-high entropy aluminum alloy performance prediction system described in any one of the first aspects, and comprises the following steps:

[0015] S1, obtain the characteristic data set used for medium and high entropy aluminum alloy performance;

[0016] S2, constructing an MLP prediction model, and obtaining an optimal hyperparameter combination of the MLP prediction model through ESOA;

[0017] S3, training the MLP prediction model according to the feature data set and the optimal hyperparameter combination;

[0018] S4, predict the properties of medium and high entropy aluminum alloys based on the trained MLP prediction model.

[0019] 5. A medium-high entropy aluminum alloy performance prediction method as claimed in claim 4, characterized in that the medium-high entropy aluminum alloy performance prediction method further comprises the following steps:

[0020] S5, obtaining a mean square error and a coefficient of determination of the MLP prediction model, and evaluating the accuracy of the MLP prediction model by using the mean square error and the coefficient of determination;

[0021] S6, obtaining predicted values ​​and actual values ​​of the medium-high entropy aluminum alloy performance, and obtaining a scatter plot and a learning rate curve according to the predicted values ​​and the actual values.

[0022] Furthermore, the medium-high entropy aluminum alloy is Al x (TiVCrNb) 100-x Lightweight high entropy alloy system, the feature data set includes input features and output targets, the input features are the molar fractions of five metals, Al, Ti, V, Cr, and Nb, and the output target is yield strength.

[0023] Furthermore, the mean square error The coefficient of determination

[0024]

[0025] Among them, y i is the actual yield strength value, is the predicted yield strength value, N is the sample size, is the mean of the actual yield strength values.

[0026] Furthermore, before training the MLP prediction model according to the feature data set and the optimal hyperparameter combination, the feature data set is first cleaned, screened and standardized.

[0027] The present application also provides a computer-readable storage medium storing a computer program, which implements the medium-high entropy aluminum alloy performance prediction method when the computer program is executed. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The present invention can be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily drawn to scale, but the emphasis is placed on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0029] Figure 1 It is a schematic diagram of the overall process of a method for predicting properties of medium-high entropy aluminum alloys in one embodiment of the present invention;

[0030] Figure 2 It is a schematic diagram of the process of the egret group optimization algorithm in one embodiment of the present invention;

[0031] Figure 3 Schematic diagram of the training process of the MLP prediction model in one embodiment of the present invention;

[0032] Figure 4 is a schematic diagram of a curve showing a change in mean square error in one embodiment of the present invention;

[0033] Figure 5 is a schematic diagram of a variation curve of a determination coefficient in one embodiment of the present invention;

[0034] Figure 6 is a schematic diagram of a learning rate curve in one embodiment of the present invention;

[0035] Figure 7 It is a schematic diagram of a scatter plot in one embodiment of the present invention. DETAILED DESCRIPTION

[0036] 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 its embodiments. It should be understood that the specific implementation methods described herein are only used to explain the present invention and do not limit the scope of protection of the present invention.

[0037] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only and do not represent the only implementation method.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0039] The “first” and “second” mentioned in the present invention do not represent specific quantities and orders, but are merely used to distinguish names.

[0040] The present application provides a medium-high entropy aluminum alloy performance prediction system that can reduce time cost and experimental cost.

[0041] The medium-high entropy aluminum alloy performance prediction system includes a data acquisition module, a construction module, a training module and a prediction module.

[0042] The data acquisition module is used to acquire a feature data set for the performance of a medium-high entropy aluminum alloy. The feature data set includes an input feature and an output target, wherein the input feature is the molar fraction of different metals in the medium-high entropy aluminum alloy, and the output target is the performance of the medium-high entropy aluminum alloy corresponding to the input feature, such as hardness, tensile strength, elongation, impact toughness, and electrochemical corrosion.

[0043] The construction module is used to construct an MLP (Multi-Layer Perceptron) prediction model, and obtain the optimal hyperparameter combination of the MLP prediction model through ESOA (Egret Swarm Optimization Algorithm).

[0044] The MLP prediction model includes an input layer, a hidden layer and an output layer, and the optimal hyperparameter combination includes a learning rate, a batch size, a number of training rounds and the number of neurons in each hidden layer.

[0045] The training module is used to train the MLP prediction model according to the feature data set and the optimal hyperparameter combination.

[0046] Preferably, the MLP prediction model is trained according to the feature data set and the optimal hyperparameter combination, and the feature data set is cleaned, screened and standardized by the data acquisition module.

[0047] The purpose of cleaning and screening the feature data set is to remove noise data and outliers to ensure the quality of the data; the purpose of standardizing the feature data set is to avoid deviations caused by differences in data dimensions and to enhance the adaptability and stability of the MLP prediction model for different data ranges.

[0048] The prediction module is used to predict the performance of medium- and high-entropy aluminum alloys based on the trained MLP prediction model.

[0049] The egret group optimization algorithm simulates the foraging behavior of egrets and has the advantages of simple algorithm implementation, fast convergence speed, and low possibility of local optimality. The MLP neural network has powerful data processing capabilities and wide applicability. Therefore, by combining the advantages of ESOA and MLP neural networks, optimizing the MLP prediction model through ESOA, and obtaining the optimal hyperparameter combination of the MLP prediction model, the accuracy and reliability of the MLP prediction model can be significantly improved, thereby improving the accuracy and reliability of the prediction of medium and high entropy aluminum alloy performance. At the same time, using the powerful data processing capabilities of the MLP prediction module to complete the prediction of medium and high entropy aluminum alloy performance can also save multiple repeated experimental steps such as batching, smelting, heat treatment and testing, saving time and experimental costs.

[0050] That is to say, the medium-high entropy aluminum alloy performance prediction system constructs an MLP prediction model, obtains the optimal hyperparameter combination of the MLP prediction model through ESOA, and trains the MLP prediction model according to the feature data set and the optimal hyperparameter combination. By combining the advantages of ESOA and the MLP prediction model, it can not only effectively improve the accuracy of medium-high entropy aluminum alloy performance prediction, but also reduce time and experimental costs through the powerful computing power of the MLP prediction model, thereby solving the problems of long time consumption and high experimental costs in the prior art of studying the correspondence between the alloy composition of high entropy aluminum alloy and its mechanical properties through experimental means.

[0051] In one embodiment, the medium-high entropy aluminum alloy is Al x (TiVCrNb) 100-x Lightweight high entropy alloy system, the feature data set includes input features and output targets, the input features are the molar fractions of five metals, Al, Ti, V, Cr, and Nb, and the output target is yield strength.

[0052] In the prediction of material yield strength, the Multi-Layer Perceptron (MLP) model, as a commonly used neural network method, has been widely used in the prediction of alloy yield strength. However, the yield strength of MLP is highly dependent on its hyperparameter settings and network structure. It is difficult to optimize and is prone to fall into local optimality, which limits the accuracy and generalization ability of the prediction. At present, many researchers have tried to use different optimization algorithms to improve the performance of MLP, but these methods are usually cumbersome steps, complex parameters, and low convergence efficiency, which are difficult to meet the needs of efficient and accurate prediction.

[0053] The egret flock optimization algorithm simulates the foraging behavior of egrets and has the advantages of simple algorithm implementation, fast convergence speed, and low possibility of local optimality. By optimizing the MLP prediction model through ESOA and obtaining the optimal hyperparameter combination of the MLP prediction model, the accuracy and reliability of the MLP prediction model can be significantly improved.

[0054] Adopting the ESOA-optimized MLP prediction model and using the trained MLP prediction model to predict the yield strength of medium- and high-entropy aluminum alloys can more accurately capture the complex relationship between alloy composition and yield strength, reduce time and labor costs, and improve the reliability and accuracy of the prediction.

[0055] An embodiment of the present application also provides a medium-high entropy aluminum alloy performance prediction method, which is applied to the medium-high entropy aluminum alloy performance prediction system described above, such as Figure 1 As shown, it includes the following steps:

[0056] S1, obtain the characteristic data set used to measure the properties of medium and high entropy aluminum alloys.

[0057] The feature data set includes input features and output targets, wherein the input features are the molar fractions of different metals in the medium-high entropy aluminum alloy, and the output targets are the properties of the medium-high entropy aluminum alloy corresponding to the input features, such as hardness, tensile strength, elongation, impact toughness, and electrochemical corrosion.

[0058] S2, constructing an MLP prediction model, and obtaining the optimal hyperparameter combination of the MLP prediction model through ESOA.

[0059] The MLP prediction model includes an input layer, a hidden layer and an output layer, and the optimal hyperparameter combination includes a learning rate, a batch size, a number of training rounds and the number of neurons in each hidden layer.

[0060] S3, training the MLP prediction model according to the feature data set and the optimal hyperparameter combination.

[0061] Preferably, before training the MLP prediction model according to the feature data set and the optimal hyperparameter combination, the feature data set is first cleaned, screened and standardized.

[0062] Cleaning and screening the feature data set can remove noise data and outliers to ensure data quality. Standardizing the feature data set can prevent the MLP prediction model from causing deviations due to dimensional differences when processing different features, thereby enhancing the adaptability and stability of the MLP prediction model for different ranges of data.

[0063] S4, predict the properties of medium and high entropy aluminum alloys based on the trained MLP prediction model.

[0064] The medium-high entropy aluminum alloy performance prediction method constructs an MLP prediction model, obtains the optimal hyperparameter combination of the MLP prediction model through ESOA, trains the MLP prediction model according to the feature data set and the optimal hyperparameter combination, and combines the advantages of ESOA and the MLP prediction model. It can not only effectively improve the accuracy of medium-high entropy aluminum alloy performance prediction, but also reduce time and experimental costs through the powerful computing power of the MLP prediction model, reduce the prediction cost and calculation amount, enhance the prediction accuracy and generalization ability of the model, and is suitable for the performance prediction of medium-high entropy aluminum alloy materials. It solves the problems of long time consumption and high experimental cost in the prior art of studying the correspondence between the alloy composition of high entropy aluminum alloy and its mechanical properties through experimental means.

[0065] In addition, by optimizing the MLP prediction model through ESOA, the complex relationship between alloy composition and properties such as hardness and yield strength can be captured more accurately. By reducing time and labor costs and improving the reliability and accuracy of prediction, the medium-high entropy aluminum alloy performance prediction method provides strong support for the design and application of medium-high entropy aluminum alloys, promotes the digitalization and intelligentization process in the field of materials science, and provides a new technical approach for performance prediction and composition optimization of medium-high entropy aluminum alloys.

[0066] In one embodiment, the medium-high entropy aluminum alloy performance prediction method further includes the following steps:

[0067] S5, obtaining the mean square error and the determination coefficient of the MLP prediction model, and evaluating the accuracy of the MLP prediction model through the mean square error and the determination coefficient.

[0068] S6, obtaining predicted values ​​and actual values ​​of the medium-high entropy aluminum alloy performance, and obtaining a scatter plot and a learning rate curve according to the predicted values ​​and the actual values.

[0069] The mean square error The coefficient of determination Among them, y i is the actual performance value, is the prediction performance value, N is the number of samples, is the mean of the actual performance values.

[0070] The performance values ​​here include but are not limited to hardness, tensile strength, elongation, impact toughness and electrochemical corrosion of medium-high entropy aluminum alloys. For different medium-high entropy aluminum alloy properties, the corresponding mean square error and determination coefficient are obtained through the MLP prediction model.

[0071] The accuracy of the MLP prediction model is evaluated by the mean square error and the coefficient of determination. Generally speaking, a lower MSE value indicates that the error between the predicted value and the actual value is smaller, and the coefficient of determination R 2 The closer the value is to 1, the better the prediction performance of the MLP prediction model.

[0072] The scatter plot and learning rate curve can help users observe the prediction accuracy and model fit more intuitively.

[0073] In one embodiment, the medium-high entropy aluminum alloy performance prediction method comprises the following steps:

[0074] S1, obtain the characteristic data set used to measure the properties of medium and high entropy aluminum alloys.

[0075] The medium-high entropy aluminum alloy Al x (TiVCrNb)100-x Lightweight high entropy alloy system, the feature data set comes from an online public database, including input features and output targets, the input features are the component proportions (i.e., molar fractions) of five metals, Al, Ti, V, Cr, and Nb, and the output target (i.e., the label corresponding to the input target) is the yield strength.

[0076] After obtaining the feature data set, the data set is cleaned and screened to remove noise data and outliers to ensure the quality of the data.

[0077] In addition, the feature data set is also standardized. Specifically, the data of each feature is standardized by subtracting its mean (average) and dividing by the standard deviation. This method is usually called Z-score standardization and can be expressed as the following formula:

[0078]

[0079] Among them, x represents the original data point, μ and σ are the mean and standard deviation of all data of this feature, respectively. This process ensures that each feature in the data set has zero mean (μ=0) and unit variance (σ=1), so that the MLP prediction model will not cause deviations due to dimensional differences when processing different features, and enhances the adaptability and stability of the model for different ranges of data. In this way, the contribution of each feature is treated equally, which helps the optimization algorithm converge faster and reduces the numerical calculation problems that may occur in the training process of the MLP prediction model.

[0080] S2, constructing an MLP prediction model, and obtaining the optimal hyperparameter combination of the MLP prediction model through ESOA.

[0081] The MLP prediction model includes an input layer, a hidden layer and an output layer, and the optimal hyperparameter combination includes a learning rate, a batch size, a number of training rounds and the number of neurons in each hidden layer.

[0082] like Figure 2 As shown, in step S2, the specific method of obtaining the optimal hyperparameter combination of the MLP prediction model through ESOA includes the following steps:

[0083] S21, initialize the egret population, including the position, speed and fitness value of each egret.

[0084] The fitness value of each egret is determined by the objective function.

[0085] S22, calculate the fitness of each individual.

[0086] Evaluate the fitness of each egret and determine the optimal egret for the current population based on the fitness value.

[0087] S23, updating the position and speed of the egret population.

[0088] The update formula is as follows:

[0089]

[0090] Among them, v i (t) and x i (t) represents the speed and position of the egret at the time; w is the inertia weight; c1 and c2 are acceleration coefficients; r1 and r2 are random numbers between [0,1]; p i is the optimal position of the individual egret; g i is the global optimal position.

[0091] S24, determining whether the upper limit of the number of iterations is reached or the convergence condition is met, if so, outputting the optimal hyperparameter combination, otherwise continuing to update the individual position and speed until the upper limit of the number of iterations is reached or the convergence condition is met.

[0092] After each iteration, the fitness of the new position is evaluated and the optimal position is updated. If the predetermined number of iterations is reached or the convergence condition is met, the iteration is terminated and the optimal egret position is output as the final result.

[0093] In this way, the ESOA algorithm can effectively obtain the optimal hyperparameter combination of the MLP prediction model.

[0094] S3, training the MLP prediction model according to the feature data set and the optimal hyperparameter combination.

[0095] like Figure 3 As shown, first, the feature data set is input into the MLP prediction model. The input layer maps the feature data to the hidden layer, and through linear transformation and activation function processing, the output of each neuron is calculated by the following formula: Among them, x i is the input feature, w ij is the weight, b j is the bias and σ is the activation function (Leaky ReLU).

[0096] The learnable Leaky ReLU activation function is adopted to enhance the robustness and generalization ability of the MLP prediction model.

[0097] Secondly, in the hidden layer, repeat the above calculations and pass them layer by layer until the output layer. The calculation formula of the output layer is:

[0098]

[0099] Among them, y k is the output predicted yield strength value.

[0100] Finally, during the training process, the training set data and labels are used to calculate the loss function and determine whether the loss meets the requirements. If so, the loss MLP prediction model is saved. If not, the weights and biases are adjusted through the back-propagation algorithm to minimize the loss function.

[0101] The loss function is calculated as:

[0102]

[0103] Among them, y i is the actual yield strength value, is the predicted yield strength value and N is the sample size.

[0104] Preferably, before training the MLP prediction model according to the feature data set and the optimal hyperparameter combination, the feature data set is first cleaned, screened and standardized.

[0105] Cleaning and screening the feature data set can remove noise data and outliers to ensure data quality. Standardizing the feature data set can prevent the MLP prediction model from causing deviations due to dimensional differences when processing different features, thereby enhancing the adaptability and stability of the MLP prediction model for different ranges of data.

[0106] S4, predict the properties of medium and high entropy aluminum alloys based on the trained MLP prediction model.

[0107] In order to quantify the prediction accuracy of the MLP prediction model, the mean squared error (MSE) and the coefficient of determination (R 2 ) is used as the evaluation index, and the accuracy of the MLP prediction model is evaluated by the mean square error and the determination coefficient.

[0108] Preferably, by obtaining the actual yield strength value and the predicted yield strength value of the medium-high entropy aluminum alloy, a scatter plot and a learning rate curve are obtained according to the actual yield strength value and the predicted yield strength value, so as to help users observe the prediction accuracy and model fit more intuitively.

[0109] The mean square error The coefficient of determination

[0110] Among them, y i is the actual yield strength value, is the predicted yield strength value, N is the sample size, is the mean of the actual yield strength values.

[0111] MSE change curve is as follows Figure 4 As shown. Figure 4 It can be seen that, with the increase in the number of training rounds, the mean square error (MSE) decreases rapidly and stabilizes after about 20 rounds, indicating that the MLP prediction model quickly learns and effectively fits the training data in the initial stage.

[0112] R 2 The change curve of Figure 5 As shown. It can be seen that R 2 The score rises rapidly in the early stage and then stabilizes at a high value close to 1, indicating that the MLP prediction model has strong explanatory power and fitting effect.

[0113] The learning rate curve is as follows Figure 6 As shown in the figure. From the learning curve, it can be seen that with the increase of the number of training iterations (Epoch), the training error and the test error both decrease significantly and tend to stabilize after about 20 Epochs. Finally, the training error and the test error converge to a lower level, indicating that the MLP prediction model performs well on both the training set and the test set, without overfitting or underfitting.

[0114] The scatter plot of actual yield strength value and predicted yield strength value is shown in Figure 7 As shown. Figure 7 In the figure, the red points represent the predicted yield strength values, and the blue points represent the actual yield strength values. As can be seen from the figure, the predicted yield strength values ​​and the actual yield strength values ​​are roughly distributed on a diagonal line, indicating that the predicted values ​​of the model are highly consistent with the actual values.

[0115] Combining the MSE curve, R 2 Curve, learning rate curve, scatter plot of actual yield strength value and predicted yield strength value, the performance of the MLP prediction model in the prediction of yield strength of medium and high entropy aluminum alloys is satisfactory, and it has high prediction accuracy and stability.

[0116] In this embodiment, the medium-high entropy aluminum alloy prediction method combines the MLP neural network and ESOA, fully utilizes the advantages of each algorithm, effectively improves the accuracy of the yield strength prediction of the medium-high entropy alloy, reduces time and experimental costs, and can efficiently and accurately obtain data reflecting the regular changes between the composition proportions of the five metal elements (Al, Ti, V, Cr, Nb) of the medium-high entropy aluminum alloy and their corresponding yield strengths. It has the characteristics of high computational efficiency, high accuracy and strong applicability; by obtaining the scatter plot and the learning rate curve, the prediction method has a significant visualization effect, which is convenient for users to observe the prediction accuracy and model fit more intuitively.

[0117] An embodiment of the present application also provides a computer-readable storage medium storing a computer program, which implements the above-mentioned medium-high entropy aluminum alloy performance prediction method when the computer program is executed.

[0118] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0119] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A medium-high entropy aluminum alloy performance prediction system, characterized in that: The medium-high entropy aluminum alloy performance prediction system comprises: A data acquisition module, used to acquire characteristic data sets for medium and high entropy aluminum alloy performance; A construction module, used to construct an MLP prediction model and obtain an optimal hyperparameter combination of the MLP prediction model through ESOA; A training module, used for training the MLP prediction model according to the feature data set and the optimal hyperparameter combination; The prediction module is used to predict the properties of medium and high entropy aluminum alloys based on the trained MLP prediction model.

2. A medium-high entropy aluminum alloy performance prediction system as claimed in claim 1, characterized in that: The data acquisition module is also used to clean, screen and standardize the feature data set.

3. A medium-high entropy aluminum alloy performance prediction system as claimed in claim 2, characterized in that: The medium-high entropy aluminum alloy is Al x (TiVCrNb) 100-x Lightweight high entropy alloy system, the feature data set includes input features and output targets, the input features are the molar fractions of five metals, Al, Ti, V, Cr, and Nb, and the output target is yield strength.

4. A medium-high entropy aluminum alloy performance prediction method, applied to the medium-high entropy aluminum alloy performance prediction system as described in any one of claims 1-3, characterized in that: The medium-high entropy aluminum alloy performance prediction method comprises the following steps: S1, obtain the characteristic data set used for medium and high entropy aluminum alloy performance; S2, constructing an MLP prediction model, and obtaining an optimal hyperparameter combination of the MLP prediction model through ESOA; S3, training the MLP prediction model according to the feature data set and the optimal hyperparameter combination; S4, predict the properties of medium and high entropy aluminum alloys based on the trained MLP prediction model.

5. A medium-high entropy aluminum alloy performance prediction method as claimed in claim 4, characterized in that: The medium-high entropy aluminum alloy performance prediction method further comprises the following steps: S5, obtaining a mean square error and a coefficient of determination of the MLP prediction model, and evaluating the accuracy of the MLP prediction model by using the mean square error and the coefficient of determination; S6, obtaining predicted values ​​and actual values ​​of the medium-high entropy aluminum alloy performance, and obtaining a scatter plot and a learning rate curve according to the predicted values ​​and the actual values.

6. A method for predicting properties of medium-high entropy aluminum alloys as claimed in claim 5, characterized in that: The medium-high entropy aluminum alloy is Al x (TiVCrNb) 100-x Lightweight high entropy alloy system, the feature data set includes input features and output targets, the input features are the molar fractions of five metals, Al, Ti, V, Cr, and Nb, and the output target is yield strength.

7. A method for predicting properties of medium-high entropy aluminum alloys as claimed in claim 6, characterized in that: The mean square error The coefficient of determination Among them, y i is the actual yield strength value, is the predicted yield strength value, N is the sample size, is the mean of the actual yield strength values.

8. A method for predicting properties of medium-high entropy aluminum alloys as claimed in claim 5, characterized in that: Before training the MLP prediction model according to the feature data set and the optimal hyperparameter combination, the feature data set is first cleaned, screened and standardized.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed, implements the medium-high entropy aluminum alloy performance prediction method as described in any one of claims 4-8.

Images