Method, device and equipment for determining importance of element characteristics of medium-entropy steel, medium and product

Abstract

The invention discloses a medium-entropy steel element feature importance determination method and device, equipment, a medium and a product. The method comprises the steps that an element feature training data set of medium-entropy steel is acquired; training a limit gradient lifting algorithm based on the element feature training data set to obtain a hardness prediction value set after reaching a set number of iterations; determining element features included in the element feature training data set and influence degree information of the relative hardness prediction value set; and according to the influence degree information, determining a hardness influence analysis result of each element feature on the medium-entropy steel. And determining a hardness influence analysis result by determining the influence degree information of the relative hardness value of each element feature. The influence importance of the element characteristics in the medium-entropy steel on the hardness is efficiently and accurately evaluated, and a scientific basis is provided for optimizing the ratio of the medium-entropy steel and improving the hardness. Compared with a traditional method, the method has the advantages of high efficiency, accuracy and transparency.

Description

Technical Field

[0001] The present invention relates to the technical field of metal material design, and in particular to a method, device, equipment, medium and product for determining the importance of element characteristics of medium-entropy steel. Background Art

[0002] Medium-Entropy Alloys (MEAs) are a new class of multi-principal element alloys, typically composed of three to five primary elemental characteristics. They exhibit high mixing entropy and excellent physical, chemical, and mechanical properties. Compared to traditional alloy design, the concept of using machine learning to assist in the design of medium-entropy steels is more innovative. Its data-driven approach facilitates the search for a wider range of unknown alloy composition spaces, helps predict alloy properties, and designs medium-entropy steels with improved performance. For example, by adjusting the doping ratio of the metallic elements characteristic of medium-entropy steels, the alloy's hardness can be increased.

[0003] Existing technology can use machine learning to assist alloy design technology, which still relies on existing physical experiments to analyze the relationship between the characteristics of each component of medium-entropy steel and its performance.

[0004] Although this method is reliable, it has the following shortcomings: First, the experimental cost is high: the experiment requires a lot of time and resources, including high-purity raw materials, precision instruments and professional technicians; second, it is time-consuming: each experiment requires multiple steps such as batching, smelting, heat treatment and testing, and the cycle is long; finally, the experimental results are limited: due to the limitations of experimental conditions, the experimental results obtained are limited, and it is difficult to fully reflect the importance ranking of a certain component on medium-entropy steel. Summary of the Invention

[0005] The present invention provides a method, device, equipment, medium and product for determining the importance of element characteristics of medium entropy steel, so as to determine the influence of different element characteristics on hardness in medium entropy steel.

[0006] According to a first aspect of the present invention, a method for determining the importance of element characteristics of medium-entropy steel is provided, the method comprising:

[0007] Obtain a training dataset of elemental characteristics of medium-entropy steel;

[0008] Training the extreme gradient boosting algorithm based on the element feature training data set to obtain a hardness prediction value set after reaching a set number of iterations;

[0009] Determining information on the degree of influence of element features included in the element feature training data set on the hardness prediction value set;

[0010] According to the influence degree information, an analysis result of the influence of each element characteristic on the hardness of the medium entropy steel is determined.

[0011] According to a second aspect of the present invention, there is provided a device for determining the importance of elemental characteristics of medium-entropy steel, comprising:

[0012] A data acquisition module is used to obtain a training data set of element characteristics of medium entropy steel;

[0013] A first determination module is used to train the extreme gradient boosting algorithm based on the element feature training data set to obtain a hardness prediction value set after reaching a set number of iterations;

[0014] A second determining module is used to determine the influence degree information of the element features included in the element feature training data set on the hardness prediction value set;

[0015] The third determination module is used to determine the analysis result of the influence of each element characteristic on the hardness of the medium entropy steel according to the influence degree information.

[0016] According to a third aspect of the present invention, there is provided an electronic device, comprising:

[0017] at least one processor; and

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining the importance of element characteristics of medium-entropy steel described in any embodiment of the present invention.

[0020] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining the importance of element characteristics of medium-entropy steel described in any embodiment of the present invention when executed.

[0021] According to the fifth aspect of the present invention, an embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the method for determining the importance of element characteristics of medium-entropy steel of any embodiment of the present invention.

[0022] The technical solution of the embodiment of the present invention is to obtain an element characteristic training data set of medium entropy steel; train the extreme gradient boosting algorithm based on the element characteristic training data set to obtain a hardness prediction value set after reaching a set number of iterations; determine the influence degree information of the element characteristics contained in the element characteristic training data set relative to the hardness prediction value set; and determine the analysis results of the influence of each element characteristic on the hardness of medium entropy steel based on the influence degree information. By determining the influence degree information of each element characteristic relative to the hardness value, the hardness influence analysis results are determined. This achieves an efficient and accurate evaluation of the importance of the influence of element characteristics in medium entropy steel on hardness, providing a scientific basis for optimizing the proportion of medium entropy steel and improving hardness. Compared with traditional methods, it has the advantages of high efficiency, precision and transparency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a flow chart of a method for determining the importance of element characteristics of medium-entropy steel provided in accordance with the first embodiment of the present invention;

[0024] Figure 2 1 is an example diagram of the degree of interaction influence in a method for determining the importance of element characteristics of medium-entropy steel according to a first embodiment of the present invention;

[0025] Figure 3 1 is an example diagram of importance in a method for determining the importance of element characteristics of medium-entropy steel according to a first embodiment of the present invention;

[0026] Figure 4 1 is an example diagram of importance in a method for determining the importance of element characteristics of medium-entropy steel according to a first embodiment of the present invention;

[0027] Figure 5 A circular example diagram of a method for determining the importance of element characteristics of medium entropy steel according to embodiment 1 of the present invention is provided.

[0028] Figure 6 2 is a schematic structural diagram of a device for determining the importance of elemental characteristics of medium-entropy steel provided in accordance with a second embodiment of the present invention;

[0029] Figure 7 It is a schematic structural diagram of an electronic device implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The present invention will be further described below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.

[0031] Example 1

[0032] Figure 1A flowchart of a method for determining the importance of elemental characteristics of medium entropy steel is provided for the first embodiment of the present invention. This embodiment is applicable to determining the influence of elemental characteristics contained in medium entropy steel on hardness. The method can be executed by a device for determining the importance of elemental characteristics of medium entropy steel. The device can be implemented in the form of hardware and / or software. The device can be configured in an electronic device. Figure 1 As shown, the method includes:

[0033] S110. Obtain a training dataset of element characteristics of medium-entropy steel.

[0034] In this embodiment, medium-entropy steel can be understood as a multi-principal element alloy typically composed of three to five primary elemental characteristics, where the content of different elemental characteristics affects its hardness. The elemental characteristic training dataset can be understood as a dataset containing the content and hardness of different elemental characteristics of medium-entropy steel.

[0035] The element feature training data in the element feature training dataset includes the hardness of medium-entropy steel and the mole fraction of the element features contained therein. The mole fraction can be understood as the proportion of the element feature in the medium-entropy steel.

[0036] Specifically, the processor may obtain experimental data of medium entropy steel from a relevant database publicly available on the Internet, and form an element feature training data set according to the format.

[0037] For example, in the present invention, the medium entropy steel system is a FeCrNi(AlTi)x two-phase medium entropy alloy system. The data set used is from an online public database. The experimental data of medium entropy steel containing different mole fractions of Fe, Cr, Ni, Al and Ti elemental features are collected from the data. The element feature training data should include input features (i.e., the mole fractions of Fe, Cr, Ni, Al and Ti) and the corresponding output target (Vickers hardness value). The data format is as follows:

[0038] {(x Fe ,x Cr ,x Ni ,x Al ,x Ti ,y n )}

[0039] Among them, x Fe 、x Cr 、x Ni 、x Al 、x Ti Represent the molar fractions of Fe, Cr, Ni, Al, and Ti (i.e., the input eigenvalues), y nIndicates that the nth sample (five eigenvalues and corresponding label values are one sample) corresponds to the Vickers hardness value HV (output label value) of medium entropy steel.

[0040] S120. Training the extreme gradient boosting algorithm based on the element feature training data set to obtain a hardness prediction value set after reaching a set number of iterations.

[0041] In this embodiment, the eXtreme Gradient Boosting (XGBoost) algorithm can be understood as an optimized distributed gradient boosting library that is efficient, flexible, and portable. The set number of iterations can be understood as the number of training runs for the set element feature training data. The hardness prediction value set can be understood as the set of hardness prediction values determined based on different element feature training data.

[0042] Specifically, the processor can preprocess each element feature training data in the element feature training data set, input the processed element feature training data set into the extreme gradient boosting algorithm, predict the hardness of the medium entropy steel, and obtain the hardness prediction value set after reaching the set number of iterations.

[0043] S130: Determine the influence degree information of the element features included in the element feature training data set on the hardness prediction value set.

[0044] In this embodiment, the element characteristics are the proportions of the characteristics of each element constituting the medium entropy steel. The influence degree information can be understood as reflecting the influence of different element characteristics on the hardness.

[0045] Specifically, the processor analyzes the hardness prediction value set and the corresponding element characteristics to determine the influence of different element characteristics on the hardness prediction value set. For example, the hardness prediction value can be analyzed through game theory or other methods.

[0046] S140. Determine, based on the influence degree information, an analysis result of the influence of each element characteristic on the hardness of the medium entropy steel.

[0047] In this embodiment, the element characteristics can be understood as the element characteristics of the composition of the medium entropy steel, such as Fe, Cr, Ni, Al and Ti. The hardness impact analysis results can be understood as the results used to characterize the degree of impact of different element characteristics on hardness.

[0048] Specifically, the processor can summarize the influence of different element characteristics on hardness through the influence degree information, thereby obtaining the analysis results of the influence of each element characteristic on the hardness of medium entropy steel.

[0049] The technical solution of the embodiment of the present invention is to obtain an element characteristic training data set of medium entropy steel; train the extreme gradient boosting algorithm based on the element characteristic training data set to obtain a hardness prediction value set after reaching a set number of iterations; determine the influence degree information of the element characteristics contained in the element characteristic training data set relative to the hardness prediction value set; and determine the analysis results of the influence of each element characteristic on the hardness of medium entropy steel based on the influence degree information. By determining the influence degree information of each element characteristic relative to the hardness value, the hardness influence analysis results are determined. This achieves an efficient and accurate evaluation of the importance of the influence of element characteristics in medium entropy steel on hardness, providing a scientific basis for optimizing the proportion of medium entropy steel and improving hardness. Compared with traditional methods, it has the advantages of high efficiency, precision and transparency.

[0050] Furthermore, based on the above embodiment, the extreme gradient boosting algorithm can be trained based on the element feature training data set to obtain the hardness prediction value after reaching the set number of iterations, which can be refined as follows:

[0051] The element feature training data in the element feature training data set is standardized to obtain a standardized training data set; the training data contained in the training data set is iteratively trained using an extreme gradient boosting algorithm to obtain a hardness prediction value set after reaching a set number of iterations.

[0052] In this embodiment, the training data set can be understood as a set of training data that can be directly input into the extreme gradient boosting algorithm.

[0053] Specifically, in order to eliminate the influence of data magnitude differences, the processor can first standardize the element feature training data in the element feature training data set to obtain a standardized training data set, and iteratively train the training data contained in the training data set through the extreme gradient boosting algorithm to obtain a hardness prediction value set after reaching the set number of iterations.

[0054] For example, the normalization formula is as follows:

[0055]

[0056] Among them, x i The original feature training data in the element feature training dataset (i.e. x Fe 、x Cr 、x Ni 、x Al 、x Ti ), y n is the original label value (i.e. the above Vickers hardness value HV), μ is the mean, σ is the standard deviation, x i ′ and y n′ is the normalized training data. The mole fractions of Fe, Cr, Ni, Al, and Ti and the Vickers hardness of medium entropy steel are normalized respectively.

[0057]

[0058] Among them, X norm is the original data, μ is the mean, and σ is the standard deviation.

[0059] For example, XGBoost is an additive model composed of k base models. Assuming that the number of iterations is set to t, the tree model to be trained in the tth iteration is f t (x), then:

[0060]

[0061] Among them, x i ′ is the standardized eigenvalue (i.e. x Fe ′、x Cr ′、x Ni ′、x Al ′、x Ti ′), y n ′ is the normalized label value, is the prediction result of sample i after the tth iteration (i.e., the Vickers hardness HV of medium entropy steel), is the prediction result of t-1 tree, f t (x i ′) is the model (function) of the t-th tree.

[0062] Furthermore, based on the above embodiment, the step of determining the influence degree information of the element features contained in the element feature training data set on the hardness prediction value set can be refined as follows:

[0063] Determine the first influence sub-information of each element feature contained in the element feature training data set on the hardness prediction value set; determine the second influence sub-information of each element feature in the element feature training data set on the hardness prediction value set; determine the third influence sub-information of the coordinated effect between the element features on the hardness prediction value set; and use the first influence sub-information, the second influence sub-information and the third influence sub-information as influence information.

[0064] In this embodiment, the first influence sub-information can be understood as representing the impact of different elemental characteristics on hardness, which can include both positive and negative impacts. The second influence sub-information can be understood as representing the importance of different elemental characteristics relative to hardness. The third influence sub-information can be understood as the impact of the interaction between different elemental characteristics on hardness.

[0065] Specifically, the processor can determine the first level of influence of each element feature in the element feature training dataset on the hardness prediction set using SHAP values. The SHAP value uses game theory to measure the marginal contribution of a feature to model predictions in different combinations. The SHAP value provides an explanation of the model prediction results for a single feature and can be used to interpret the prediction results of the extreme gradient boosting algorithm. The processor can determine the second level of influence of each element feature in the element feature training dataset on the hardness prediction set using the average absolute SHAP value. The average absolute SHAP value represents the average contribution of a feature to the model prediction across all samples. By sorting the average absolute SHAP values, the importance of each feature can be determined. The average absolute SHAP value helps identify which features are most important to the overall prediction results. The processor can determine the third level of influence of each element feature on the hardness prediction set using the feature interaction SHAP value. The feature interaction SHAP value reveals the synergy between features, helping to understand the complex interactions between features and their joint contribution to the prediction results. The feature interaction SHAP value describes the synergy and interaction between features, thereby improving the interpretability of the model, especially when features are highly correlated. The processor may analyze the first influence degree sub-information, the second influence degree sub-information, and the third influence degree sub-information respectively, determine the influence of different element characteristics on hardness under different sub-information, and use the analysis results as influence degree information.

[0066] For example, the SHAP value can be calculated by the following formula:

[0067]

[0068] φ ij : SHAP value of element feature i, indicating the contribution of element feature j to the hardness prediction value; S: element feature subset, excluding element feature i; M: total number of element features; f(S∪{i}): hardness prediction value calculated on element feature subset S containing element feature i; f(S): hardness prediction value calculated on element feature subset S excluding element feature i; Characterize the weights to ensure fairness of the SHAP value for the feature subset (i = {Fe, Cr, Ni, Al, Ti}). The processor can use the different SHAP values of each element feature as the first influence sub-information.

[0069] For example, the mean absolute SHAP value (Mean(|SHAPValue|)) can be calculated by the following formula:

[0070]

[0071] |φij |: the absolute value of the SHAP value of the i-th feature in the j-th sample; n: the number of samples.

[0072] For example, the SHAP Interaction Value can be calculated using the following formula:

[0073] SHAP Interaction ij =φ ij +φ ji

[0074] φ ij : The interaction effect between features i and j; the SHAP interaction value calculates the joint impact of two features on model prediction.

[0075] The step of determining the analysis result of the influence of each element characteristic on the hardness of the medium entropy steel according to the influence degree information may include:

[0076] The first influence degree sub-information in the influence degree information is analyzed, and a diagram showing the influence degree of each element characteristic on the hardness of medium entropy steel is drawn; the second influence degree sub-information in the influence degree information is analyzed, and a diagram showing the importance of the influence of each element characteristic on the hardness of medium entropy steel is drawn; the third influence degree sub-information in the influence degree information is analyzed, and a diagram showing the interactive influence degree of different element characteristics on hardness is drawn; based on the influence degree diagram, the importance diagram, and the interactive influence degree diagram, the analysis results of the influence of each element characteristic on the hardness of medium entropy steel are determined.

[0077] In this embodiment, the influence map can be understood as a visualization of the impact of different elemental features on the predicted hardness value. The importance map can be understood as a visualization of the importance of different elemental features relative to hardness. The interaction influence map can be understood as a visualization of the impact of interactions between different elemental features on hardness.

[0078] Specifically, the processor can analyze the first influence degree sub-information in the influence degree information according to the different SHAP values corresponding to different element characteristics, and draw an influence degree diagram of each element characteristic on the hardness of medium entropy steel; the processor can analyze the second influence degree sub-information in the influence degree information according to the different average absolute SHAP values corresponding to different element characteristics, and draw an importance diagram of the influence of each element characteristic on the hardness of medium entropy steel; the processor can analyze the feature interaction SHAP values corresponding to the interaction of two different element characteristics in the third influence degree sub-information in the influence degree information, and draw an interactive influence degree diagram of the hardness under the interaction of different element characteristics; based on the influence degree diagram, the importance diagram and the interactive influence degree diagram, determine the analysis results of the influence of each element characteristic on the hardness of medium entropy steel.

[0079] For example, Figure 2 An example diagram of the degree of interaction in a method for determining the importance of element characteristics of medium entropy steel is provided for the first embodiment of the present invention, as shown in FIG. Figure 2 As shown, a SHAP honeycomb plot can be used as an example. In this plot, color represents the magnitude of eigenvalues. Pink dots indicate that the element's eigenvalue is high in this prediction model, meaning it has a high eigenvalue; blue dots indicate that the element's eigenvalue is low in this prediction model, meaning it has a low eigenvalue. The horizontal axis shows the SHAP value, which displays the impact of each feature on the prediction result. The farther the dot is from the center line (zero), the greater the impact of the feature on the model output. SHAP values to the right of zero indicate a positive impact, while SHAP values to the left of zero indicate a negative impact. The vertical axis shows the arrangement of the element features: The vertically arranged features in the plot are sorted from top to bottom in terms of influence. Features at the top have the greatest overall impact on the model output, while features at the bottom have a smaller impact. The top feature displays a large number of positive and negative effects, indicating that its impact on the model prediction varies significantly across different observations. The middle feature also displays dots of both colors, but the distribution is more concentrated, indicating a relatively smaller impact. The bottom feature has the smallest impact on the model, with most of its impact close to zero, indicating that these features contribute little to the model's prediction. It can be seen from the figure that Al has the most significant positive effect on hardness. A high value of Al corresponds to a positive hardness increase. The effects of other characteristics (such as Fe and Cr) are weaker and have a positive and negative alternating trend, indicating that these characteristics have a more complex effect on hardness.

[0080] For example, Figure 3 An example diagram of importance in a method for determining the importance of element characteristics of medium entropy steel is provided for the first embodiment of the present invention, such as Figure 3 As shown in the figure, the horizontal axis represents the average absolute SHAP value, and the vertical axis represents the different element characteristics. It can be seen intuitively that Al is the element characteristic that most affects hardness, and its importance far exceeds that of other element characteristics. Fe and Ni follow closely, and also have a certain impact on hardness.

[0081] For example, Figure 4 An example diagram of importance in a method for determining the importance of element characteristics of medium entropy steel is provided for the first embodiment of the present invention, such as Figure 4As shown in the figure, feature ranking: Features are sorted by importance, with the most important features at the top of the graph. The total importance of each feature is the sum of the importance values of its interactions with all other features. SHAP interaction value distribution: Each point represents the interaction value of a sample. The larger the interaction value, the greater the impact of the interaction between the feature and the other feature on the model prediction. The color simply represents the size of the feature value, with blue representing a smaller feature value and pink representing a larger feature value. The position on the horizontal axis shows the strength and direction of the impact of the interaction between two features on the model output. To the right of the horizontal axis (positive direction): The interaction between the two features jointly drives the predicted value to increase. To the left of the horizontal axis (negative direction): The interaction between the two features jointly drives the predicted value to decrease. Near zero (center of the horizontal axis): The interaction is very small and has almost no impact on the prediction result. In this SHAP interaction plot, the position on the horizontal axis and the color distribution can be used to analyze the interaction between features. For example, the interaction between "Al" and "Fe" shows that the points are primarily distributed to the right of the horizontal axis (positive values), indicating that their interaction has a positive impact on the prediction results. Pink points (high "Al" values) are concentrated on the right, indicating that the interaction between high "Al" and "Fe" values significantly increases the predicted value, while blue points (low "Al" values) are distributed on the left, indicating that the interaction between low "Al" and "Fe" values may decrease the predicted value. Similarly, the interaction between "Cr" and "Ti" shows that most points are close to zero, indicating a weak interaction, but a small number of points are far from zero, indicating significant positive or negative effects in some samples. Pink points are concentrated in the positive region, and blue points are concentrated in the negative region, corresponding to the different interaction directions for high and low values, respectively. In summary, the distribution range and color changes of the points can be used to interpret the strength and direction of the impact of feature interactions on model prediction results, as well as the impact of feature value size.

[0082] Furthermore, the average absolute SHAP value in the second influence sub-information can be normalized to calculate the proportion of the importance of the element features, and the normalized proportion value of each element feature can be used to generate a ring chart to represent the relative contribution ratio of each element feature to hardness.

[0083] For element feature f, the calculation formula for its proportion is:

[0084]

[0085] Among them, j represents all element features in the training dataset.

[0086] For example, Figure 5 A circular example diagram of a method for determining the importance of element characteristics of medium entropy steel is provided for the first embodiment of the present invention, such as Figure 5As shown, the circular chart illustrates the relative contribution of each characteristic to hardness. Al has the highest contribution, followed by Fe, Ni, Cr, and Ti. Al's contribution reaches 56%, significantly higher than the other characteristics, indicating that Al is the core driver of hardness. Fe (19.8%) and Ni (10.1%) also contribute significantly to hardness; Cr and Ti contribute relatively less. Analysis of the impact of five elemental characteristics (Al, Fe, Ni, Cr, and Ti) on hardness reveals the following conclusions: Al is the most important factor influencing hardness, significantly contributing 56% to the hardness, significantly higher than the other characteristics. Furthermore, Fe and Ni also have a certain influence on hardness, accounting for 19.8% and 10.1%, respectively, possibly further influencing hardness through inter-characteristic interactions. Cr and Ti contribute less significantly to hardness, though their impact on hardness is less pronounced than that of the other characteristics, although they may play a role in specific circumstances. Overall, Al is the core driver of hardness, with the other elemental characteristics playing synergistic or secondary roles.

[0087] Furthermore, based on the above embodiment, the steps of determining the analysis results of the influence of each element characteristic on the hardness of medium entropy steel according to the influence degree diagram, importance diagram and interactive influence degree diagram can be refined as follows:

[0088] From the influence degree diagram, determine the first target element feature with the greatest positive impact on hardness; from the importance diagram, determine the second target element feature with the greatest importance; from the interactive influence degree diagram, determine the target element feature group that increases hardness; and based on the first target element feature, the second target element feature, and the target element feature group, determine the hardness impact analysis results.

[0089] In this embodiment, the first target element feature can be understood as the element feature with the greatest impact determined from the influence map. The second target element feature can be understood as the element feature with the greatest impact determined from the importance map. The target element feature group can be understood as a combination of multiple element features that have the greatest impact on hardness through interaction. The first target element feature and the second target element feature can be the same or different.

[0090] Specifically, the processor may determine, from the influence map, the element feature with the greatest positive impact on hardness as the first target element feature. The processor may determine, from the importance map, the element feature with the greatest importance as the second target element feature. The processor may determine, from the interaction influence map, a target element feature group that increases hardness. The processor may use the first target element feature, the second target element feature, the target element feature group, and the aforementioned maps as a hardness impact analysis result.

[0091] For example, as shown in the drawings in the above example, it can be determined that the first target element feature and the second target element feature are both Al, and the target element feature group is Al and Al.

[0092] Through the above four visualization diagrams, the processor can identify the key role of Al as the core driver of hardness, and also reveal the synergistic effect of other elements (such as Fe and Ni), providing a scientific basis for optimizing the proportion of medium-entropy steel and improving hardness.

[0093] The technical solution of the embodiment of the present invention is to standardize the element feature training data set, and then predict the hardness of the element features under different mole fractions through the extreme gradient boosting algorithm to obtain a hardness prediction value set. With the help of the powerful feature learning ability of the extreme gradient boosting algorithm, combined with the feature standardization process, the accuracy and stability of the hardness prediction model are significantly improved, thereby ensuring the accuracy of the feature importance analysis. By calculating different SHAP values, different influence degree sub-information is determined, which can effectively quantify the influence of the mole fractions of the five elements Fe, Cr, Ni, Al, and Ti on the hardness of medium entropy steel, and clearly identify the influence of each element on the hardness. By drawing a visualization diagram of the different influence degree sub-information and determining the target element feature with the greatest influence, the hardness influence analysis result is obtained, and the importance and interaction are intuitively displayed, helping researchers to intuitively understand the contribution of each element to the hardness and its interactive effect, thereby enhancing the interpretability and decision support capabilities of the method, being able to deeply explore the synergistic effect between elements, and having a more comprehensive understanding of the influence law of the hardness of medium entropy steel, providing data support for the multivariate optimization of material composition, and improving the interpretability of the model prediction results.

[0094] Example 2

[0095] Figure 6 This is a schematic diagram of a device for determining the importance of elemental characteristics of medium entropy steel provided in Example 2 of the present invention. Figure 6 As shown, the device includes: a data acquisition module 61 , a first determination module 62 , a second determination module 63 and a third determination module 64 .

[0096] A data acquisition module 61 is used to obtain a training data set of element characteristics of medium entropy steel;

[0097] A first determination module 62 is configured to train an extreme gradient boosting algorithm based on the element feature training data set to obtain a hardness prediction value set after a set number of iterations.

[0098] A second determining module 63 is configured to determine information on the degree of influence of the element features included in the element feature training data set on the hardness prediction value set;

[0099] The third determining module 64 is configured to determine, based on the influence degree information, an analysis result of the influence of each element characteristic on the hardness of the medium entropy steel.

[0100] The technical solution of the embodiment of the present invention is to obtain an element characteristic training data set of medium entropy steel; train the extreme gradient boosting algorithm based on the element characteristic training data set to obtain a hardness prediction value set after reaching a set number of iterations; determine the influence degree information of the element characteristics contained in the element characteristic training data set relative to the hardness prediction value set; and determine the analysis results of the influence of each element characteristic on the hardness of medium entropy steel based on the influence degree information. By determining the influence degree information of each element characteristic relative to the hardness value, the hardness influence analysis results are determined. This achieves an efficient and accurate evaluation of the importance of the influence of element characteristics in medium entropy steel on hardness, providing a scientific basis for optimizing the proportion of medium entropy steel and improving hardness. Compared with traditional methods, it has the advantages of high efficiency, precision and transparency.

[0101] The element characteristic training data in the element characteristic training data set includes: the hardness of the medium entropy steel and the molar fraction of the element characteristics contained therein.

[0102] Furthermore, the first determining module 62 is specifically configured to:

[0103] Standardizing the element feature training data in the element feature training data set to obtain a standardized training data set;

[0104] The training data contained in the training data set is iteratively trained using an extreme gradient boosting algorithm to obtain a hardness prediction value set after reaching a set number of iterations.

[0105] Furthermore, the second determining module 63 is specifically configured to:

[0106] Determining first influence sub-information of each element feature included in the element feature training data set on the hardness prediction value set;

[0107] Determining second influence sub-information of each element feature in the element feature training data set on the hardness prediction value set;

[0108] Determining third influence sub-information on the hardness prediction value set under the coordinated effect between the element characteristics;

[0109] The first influence level sub-information, the second influence level sub-information, and the third influence level sub-information are used as influence level information.

[0110] Furthermore, the third determining module 64 includes:

[0111] a first drawing unit, configured to analyze the first influence degree sub-information in the influence degree information, and draw a diagram showing the influence degree of each element characteristic on the hardness of the medium entropy steel;

[0112] a second drawing unit, configured to analyze the second influence degree sub-information in the influence degree information, and draw an importance diagram of the influence of each element characteristic on the hardness of the medium entropy steel;

[0113] a third drawing unit, configured to analyze the third influence degree sub-information in the influence degree information, and draw a diagram of the interactive influence degree of the hardness under the interaction of different element characteristics;

[0114] A result determination unit is used to determine an analysis result of the influence of each element characteristic on the hardness of the medium entropy steel according to the influence degree diagram, the importance diagram and the interactive influence degree diagram.

[0115] The result determination unit is specifically configured to:

[0116] determining, from the influence degree map, a first target element characteristic having the greatest positive influence on the hardness;

[0117] Determining a second target element feature with the greatest importance from the importance map;

[0118] determining a target element feature group for increasing the hardness from the interactive influence degree map;

[0119] A hardness impact analysis result is determined according to the first target element feature, the second target element feature, and the target element feature group.

[0120] The video tracking device provided in the embodiment of the present invention can execute the method for determining the importance of element characteristics of medium-entropy steel provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0121] Example 3

[0122] Figure 7 A schematic diagram of the structure of an electronic device 70 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0123] like Figure 7 As shown, the electronic device 70 includes at least one processor 71 and a memory connected to the at least one processor 71, such as a read-only memory (ROM) 72, a random access memory (RAM) 73, etc., wherein the memory stores a computer program that can be executed by the at least one processor, and the processor 71 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 72 or the computer program loaded from the storage unit 78 to the random access memory (RAM) 73. Various programs and data required for the operation of the electronic device 70 can also be stored in the RAM 73. The processor 71, ROM 72 and RAM 73 are connected to each other via a bus 74. An input / output (I / O) interface 75 is also connected to the bus 74.

[0124] Multiple components in the electronic device 70 are connected to the I / O interface 75, including an input unit 76, such as a keyboard, a mouse, etc.; an output unit 77, such as various types of displays, speakers, etc.; a storage unit 78, such as a magnetic disk, an optical disk, etc.; and a communication unit 79, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 79 allows the electronic device 70 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0125] The processor 71 may be any general-purpose and / or specialized processing component with processing and computing capabilities. Examples of the processor 71 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors for running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 71 executes the various methods and processes described above, such as the method for determining the importance of elemental characteristics of medium-entropy steel.

[0126] In some embodiments, the method for determining the importance of elemental characteristics of medium entropy steel may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 78. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 70 via the ROM 72 and / or the communication unit 79. When the computer program is loaded into the RAM 73 and executed by the processor 71, one or more steps of the method for determining the importance of elemental characteristics of medium entropy steel described above may be performed. Alternatively, in other embodiments, the processor 71 may be configured to execute the method for determining the importance of elemental characteristics of medium entropy steel in any other appropriate manner (e.g., by means of firmware).

[0127] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system comprising at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0128] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0129] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0130] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0131] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0132] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0133] In one embodiment, the present invention further includes a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the method for determining the importance of element characteristics of medium-entropy steel of any embodiment of the present invention.

[0134] The computer program product may be implemented by writing computer program code for performing the operations of the present invention in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0135] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0136] The above embodiments are only preferred embodiments of the present invention and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and replacements made by technicians in this field on the basis of the present invention fall within the scope of protection required by the present invention.

Claims

1. A method for determining the importance of element characteristics of medium entropy steel, characterized in that: include: Obtain a training dataset of elemental characteristics of medium-entropy steel; Training the extreme gradient boosting algorithm based on the element feature training data set to obtain a hardness prediction value set after reaching a set number of iterations; Determining information on the degree of influence of element features included in the element feature training data set on the hardness prediction value set; According to the influence degree information, an analysis result of the influence of each element characteristic on the hardness of the medium entropy steel is determined.

2. The method according to claim 1, characterized in that The element characteristic training data in the element characteristic training data set includes: the hardness of the medium entropy steel and the molar fraction of the element characteristics contained therein.

3. The method according to claim 1, characterized in that The extreme gradient boosting algorithm is trained based on the element feature training data set to obtain a hardness prediction value after reaching a set number of iterations, including: Standardizing the element feature training data in the element feature training data set to obtain a standardized training data set; The training data contained in the training data set is iteratively trained using an extreme gradient boosting algorithm to obtain a hardness prediction value set after reaching a set number of iterations.

4. The method according to claim 1, wherein The determining of the influence degree information of the element features included in the element feature training data set on the hardness prediction value set includes: Determining first influence sub-information of each element feature included in the element feature training data set on the hardness prediction value set; Determining second influence sub-information of each element feature in the element feature training data set on the hardness prediction value set; Determining third influence sub-information on the hardness prediction value set under the coordinated effect between the element characteristics; The first influence level sub-information, the second influence level sub-information, and the third influence level sub-information are used as influence level information.

5. The method according to claim 1, characterized in that Determining, based on the influence degree information, an analysis result of the influence of each element characteristic on the hardness of the medium entropy steel includes: Analyzing the first influence degree sub-information in the influence degree information, and drawing a graph showing the influence degree of each element characteristic on the hardness of the medium entropy steel; Analyzing the second influence degree sub-information in the influence degree information, and drawing an importance diagram of the influence of each element characteristic on the hardness of the medium entropy steel; Analyzing the third influence degree sub-information in the influence degree information, and drawing a diagram of the interactive influence degree of different element characteristics on the hardness; According to the influence degree diagram, the importance diagram and the interactive influence degree diagram, an analysis result of the influence of each element characteristic on the hardness of the medium entropy steel is determined.

6. The method according to claim 5, characterized in that The determining of the analysis results of the influence of each element characteristic on the hardness of the medium entropy steel according to the influence degree diagram, the importance diagram, and the interactive influence degree diagram includes: determining, from the influence degree map, a first target element characteristic having the greatest positive influence on the hardness; Determining a second target element feature with the greatest importance from the importance map; determining a target element feature group for increasing the hardness from the interactive influence degree map; A hardness impact analysis result is determined according to the first target element feature, the second target element feature, and the target element feature group.

7. A device for determining the importance of element characteristics of medium entropy steel, characterized in that: include: A data acquisition module is used to obtain a training data set of element characteristics of medium entropy steel; A first determination module is used to train the extreme gradient boosting algorithm based on the element feature training data set to obtain a hardness prediction value set after reaching a set number of iterations; A second determining module is used to determine the influence degree information of the element features included in the element feature training data set on the hardness prediction value set; The third determination module is used to determine the analysis result of the influence of each element characteristic on the hardness of the medium entropy steel according to the influence degree information.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the method for determining the importance of element characteristics of medium-entropy steel according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining the importance of element characteristics of medium-entropy steel according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The computer program product comprises a computer program, which, when executed by a processor, implements the method for determining the importance of element characteristics of medium-entropy steel according to any one of claims 1 to 6.

Images