CN-121983187-A - Alloy material component screening method, system, equipment and medium

CN121983187ACN 121983187 ACN121983187 ACN 121983187ACN-121983187-A

Abstract

The invention relates to the technical field of alloy materials, and discloses a method, a system, equipment and a medium for screening alloy material components. The method comprises the steps of constructing an alloy component performance data set, constructing a multi-model performance prediction model, training the multi-model performance prediction model according to the alloy component performance data set to screen a performance prediction reference model, performing global search according to a genetic algorithm and the performance prediction reference model to obtain an alloy component scheme and a corresponding performance prediction result, performing weighted sorting according to the alloy component scheme and the performance prediction result to screen a target alloy combination, and performing simulation verification and process analysis on the target alloy combination to optimize the performance prediction reference model. According to the invention, deviation of a single model is avoided through a multi-model performance prediction model, a genetic algorithm is introduced to ensure that global search avoids local optimization, a closed-loop process design is formed by combining simulation verification and process analysis, a target alloy combination meeting requirements is rapidly and accurately screened, and development efficiency is improved.

Inventors

CHEN HAOYUAN
YANG ZHAN
ZHOU CHUNXIAO
LI XIAOYANG
YI HAITAO
ZHOU WEI
GAO SHENG
LI MING
HU HONGWEI
LI LONGYUN
YIN HONG
LIU CHUNTANG
XIA LIWEI
WANG XINGCHAO
TU SIYU

Assignees

国网湖北省电力有限公司超高压公司

Dates

Publication Date: 20260505
Application Date: 20251208

Claims (10)

1. A method for screening alloy material components, comprising the steps of: Constructing an alloy component performance data set; constructing a multi-model performance prediction model, and training the multi-model performance prediction model according to the alloy component performance data set so as to screen a performance prediction reference model; Performing global search according to a genetic algorithm and the performance prediction reference model to obtain an alloy composition scheme and a corresponding performance prediction result; Weighting, scoring and sorting are carried out according to the alloy component scheme and the performance prediction result so as to screen out target alloy combinations; And performing simulation verification and process analysis on the target alloy combination to optimize the performance prediction reference model.
2. The method of claim 1, wherein the constructing an alloy composition property dataset comprises: collecting alloying element composition, mechanical property and heat treatment process data of the target alloy; Analyzing the alloying element composition, the mechanical property and the heat treatment process data to obtain element quality and an optimization target; carrying out standardization treatment on the element quality and the optimization target; And constructing an alloy component performance data set according to the standardized element quality and the optimization target.
3. The method of claim 1, wherein said constructing a multi-model performance prediction model, training said multi-model performance prediction model based on said alloy composition performance dataset to screen out a performance prediction basis model, comprises: constructing a multi-model performance prediction model based on a plurality of machine learning models; Training the multi-model performance prediction model according to the alloy component performance data set to obtain a trained model; Constructing an evaluation index according to the average absolute value error, the mean square error and the decision coefficient; And screening the trained model based on the evaluation index to obtain a performance prediction reference model.
4. The method of claim 1, wherein the multi-model performance prediction model comprises a back propagation neural network, a support vector machine model, a random forest model, and the performance prediction reference model comprises a back propagation neural network.
5. The method of claim 1, wherein performing a global search based on a genetic algorithm and the performance prediction benchmark model to obtain an alloy composition scheme and corresponding performance prediction results comprises: Setting a chromosome coding mode by adopting a genetic algorithm, determining a component search boundary and configuring genetic algorithm control parameters; calculating the predicted performance of each alloy individual component by taking the performance prediction reference model as an evaluation tool, and constructing a multi-objective comprehensive score function as an adaptability function of the genetic algorithm; performing selection, crossover and mutation genetic operations; and performing iterative search based on the genetic algorithm to obtain an alloy composition scheme and a corresponding performance prediction result of each individual.
6. The method of claim 1, wherein said weighting and scoring the alloy composition scheme and the performance prediction result to screen out a target alloy combination comprises: Normalizing the performance prediction result to obtain a normalized index score; Acquiring the weight of each performance index; Calculating a weighted score based on the normalized index score and the weight; The alloy composition schemes are ranked from high to low based on the weighted scores to screen out target alloy combinations.
7. The method of claim 6, wherein the calculating a weighted score based on the normalized indicator score and weight comprises: determining a density score, a Young's modulus score, and a tensile strength score from the normalized index score; determining a density weight, a Young's modulus weight and a tensile strength weight according to the weight; and summing the product of the density score and the density weight, the product of the Young modulus score and the Young modulus weight and the product of the tensile strength score and the tensile strength weight to obtain a weighted score.
8. An alloy material composition screening system, comprising: The data acquisition module is used for constructing an alloy component performance data set; The model construction module is used for constructing a multi-model performance prediction model, training the multi-model performance prediction model according to the alloy component performance data set and screening a performance prediction reference model; The search optimization module is used for carrying out global search according to a genetic algorithm and the performance prediction reference model to obtain an alloy component scheme and a corresponding performance prediction result; The weighting and scoring module is used for carrying out weighting and scoring sequencing according to the alloy component scheme and the performance prediction result so as to screen out target alloy combinations; And the simulation verification module is used for performing simulation verification and process analysis on the target alloy combination so as to optimize the performance prediction reference model.
9. An electronic device, comprising: One or more processors; A memory for storing one or more programs; when executed by the one or more processors, causes the one or more processors to implement the method of any of claims 1 to 7.
10. A computer readable medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of the method according to any of claims 1 to 7.

Description

Alloy material component screening method, system, equipment and medium Technical Field The invention relates to the technical field of alloy materials, in particular to a method, a system, equipment and a medium for screening alloy material components. Background In recent years, with rapid development of high-performance fields such as aerospace, automobile manufacturing, rail transit and the like, the demand for lightweight high-strength materials is increasing. Aluminum-based alloys (Al-based alloys) are a research hotspot for critical structural materials due to their low density, high specific strength, good workability, strong corrosion resistance, and the like. Traditional Al-based alloy material development mainly passes experimental repeated tests based on experience, however, due to the characteristics of complexity and nonlinearity of interaction among multiple elements, the alloy types formed by multiple alloy components often reach tens of thousands, repeated enumeration and experimental tests can cause huge time and cost loss, and the designed alloy performance is not ideal. In addition, the traditional experimental method is mostly a single-target optimization problem, and multiple targets such as low density, high strength and high modulus are difficult to be considered. Disclosure of Invention The invention mainly aims to provide a method, a system, equipment and a medium for screening alloy material components, which aim to solve at least one of the technical problems. In a first aspect, an embodiment of the present invention provides a method for screening components of an alloy material, including: Constructing an alloy component performance data set; constructing a multi-model performance prediction model, and training the multi-model performance prediction model according to the alloy component performance data set so as to screen a performance prediction reference model; Performing global search according to a genetic algorithm and the performance prediction reference model to obtain an alloy composition scheme and a corresponding performance prediction result; Weighting, scoring and sorting are carried out according to the alloy component scheme and the performance prediction result so as to screen out target alloy combinations; And performing simulation verification and process analysis on the target alloy combination to optimize the performance prediction reference model. In some embodiments, the constructing an alloy composition property dataset includes: collecting alloying element composition, mechanical property and heat treatment process data of the target alloy; Analyzing the alloying element composition, the mechanical property and the heat treatment process data to obtain element quality and an optimization target; carrying out standardization treatment on the element quality and the optimization target; And constructing an alloy component performance data set according to the standardized element quality and the optimization target. In some embodiments, the constructing a multi-model performance prediction model, training the multi-model performance prediction model from the alloy composition performance dataset to screen out a performance prediction benchmark model, comprises: constructing a multi-model performance prediction model based on a plurality of machine learning models; Training the multi-model performance prediction model according to the alloy component performance data set to obtain a trained model; Constructing an evaluation index according to the average absolute value error, the mean square error and the decision coefficient; And screening the trained model based on the evaluation index to obtain a performance prediction reference model. In some embodiments, the multi-model performance prediction model comprises a back propagation neural network, a support vector machine model and a random forest model, and the performance prediction reference model comprises a back propagation neural network. In some embodiments, the performing global search according to a genetic algorithm and the performance prediction reference model to obtain an alloy composition scheme and a corresponding performance prediction result includes: Setting a chromosome coding mode by adopting a genetic algorithm, determining a component search boundary and configuring genetic algorithm control parameters; calculating the predicted performance of each alloy individual component by taking the performance prediction reference model as an evaluation tool, and constructing a multi-objective comprehensive score function as an adaptability function of the genetic algorithm; performing selection, crossover and mutation genetic operations; and performing iterative search based on the genetic algorithm to obtain an alloy composition scheme and a corresponding performance prediction result of each individual. In some embodiments, the weighted scoring ordering according to the alloy composition sch