Composite material formula process optimization method based on performance requirements

Abstract

The invention provides a composite material formula process optimization method based on a performance demand, and relates to the technical field of data processing, the method comprises the following steps: obtaining matrix material information and a performance demand, and determining a missing performance demand and a missing degree thereof; determining a reinforcing material configuration relation list; configuring formula test variables; constructing a test space, obtaining a test data set, analyzing and processing the test data set, and constructing a target data set; and optimizing the target data set to obtain a formula process optimization result. The technical problem that the performance of the composite material cannot meet the required performance due to the lack of performance fusion analysis of the reinforcing material in the prior art can be solved, after the missing performance analysis is performed on the base material, the formula processing technology of the reinforcing material is optimized and analyzed, and the performance of the composite material is improved. The technical effects that the formula process accuracy of the composite material is improved, and then the adaptation degree of the performance of the composite material and the required performance is guaranteed are achieved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method for optimizing composite material formulation processes based on performance requirements. Background Art

[0002] With the advancement of industrial technology, composite materials are increasingly being used in various fields. Optimizing the performance of composite materials is a key step in enhancing their application value. This optimization includes optimizing strength and thermal properties. The strength of composite materials can be effectively improved by increasing the volume content of reinforcements, optimizing their distribution and orientation, and performing surface treatments. However, to enhance the performance of composite materials, one or more reinforcements are often added. This can affect the performance of different materials, leading to performance deviations in the composite material.

[0003] The formulation process of reinforcing materials is an important step in improving the performance of composite materials. However, the traditional formulation optimization process lacks an integrated analysis of the performance of reinforcing materials, resulting in the performance of composite materials being difficult to meet the required performance. Summary of the Invention

[0004] The purpose of this application is to provide a composite material formulation process optimization method based on performance requirements to solve the technical problem in the prior art that the performance of the composite material is difficult to meet the required performance due to the lack of performance fusion analysis of the reinforced material.

[0005] In view of the above problems, the present application provides a composite material formulation process optimization method based on performance requirements, the method comprising: obtaining matrix material information and performance requirements, performing performance support decomposition based on the matrix material information and the performance requirements, determining missing performance requirements and their missingness, wherein the missing performance requirements are performances in which the deviation between the performance requirements and the performance support is greater than a preset value, and the missingness is the ratio of the deviation between the performance requirements and the performance support to the performance requirements; performing a reinforcement material matching analysis based on the missing performance requirements and their missingness, and determining a reinforcement material configuration relationship list; performing a relationship matching analysis based on the missing performance requirements and their missingness and the reinforcement material configuration relationship list, fusing and extracting the fusion performance relationship between the matrix material information and the reinforcement material, and configuring formulation test variables; obtaining processing parameters, configuring process test variables, constructing a test space based on the formulation test variables and the process test variables, obtaining a test data set, analyzing and processing the test data set, and constructing a target data set; setting constraints based on the performance requirements and the processing technology, optimizing the target data set, and obtaining formulation process optimization results.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The method includes obtaining matrix material information and performance requirements, performing performance support decomposition based on the matrix material information and the performance requirements, determining missing performance requirements and their missingness, wherein the missing performance requirements are performance requirements for which the deviation between the performance requirements and the performance support is greater than a preset value, and the missingness is the ratio of the deviation between the performance requirements and the performance support to the performance requirements; performing reinforcement material matching analysis based on the missing performance requirements and their missingness, and determining a reinforcement material configuration relationship list; performing relationship matching analysis based on the missing performance requirements and their missingness and the reinforcement material configuration relationship list, fusing and extracting the fusion performance relationship between the matrix material information and the reinforcement material, and configuring formula test variables; obtaining processing parameters, configuring process test variables, constructing a test space based on the formula test variables and the process test variables, obtaining a test data set, analyzing and processing the test data set, and constructing a target data set; setting constraints based on the performance requirements and the processing technology, optimizing the target data set, and obtaining a formula process optimization result. Thus, after performing missing performance analysis on the matrix material, optimizing the reinforcement material formula and processing technology, the accuracy of the composite material formula process is improved, thereby ensuring the compatibility of the composite material performance with the required performance.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, which can be implemented in accordance with the contents of the description, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without creative work.

[0009] Figure 1 This is a flow chart of the composite material formulation process optimization method based on performance requirements for this application. DETAILED DESCRIPTION

[0010] This application provides a method for optimizing composite material formulations based on performance requirements, addressing the existing technical problem of composite material performance failing to meet required performance requirements due to a lack of performance fusion analysis of the reinforcement material. By analyzing the missing properties of the matrix material and optimizing the reinforcement material formulation and processing, the accuracy of the composite material formulation process is improved, thereby ensuring the compatibility of the composite material's performance with the required performance.

[0011] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0012] Please see the attached Figure 1 , this application provides a composite material formulation process optimization method based on performance requirements, the method specifically comprising the following steps: Step 1: Obtain base material information and performance requirements, perform performance support decomposition based on the base material information and the performance requirements, and determine missing performance requirements and their missingness. The missing performance requirements are performance requirements for which the deviation between the performance requirements and the performance support is greater than a preset value, and the missingness is the ratio of the deviation between the performance requirements and the performance support to the performance requirements. Specifically, composite materials refer to materials composed of two or more different types of materials. Composite materials usually consist of a matrix material and a reinforcement material. The matrix material is the continuous phase in the composite material, used to maintain the overall structural shape and evenly transfer the load to the reinforcement material. It usually has good toughness and ductility, can absorb vibration and impact energy, and can also protect the reinforcement material from direct damage from the external environment, such as polyethylene, polypropylene, etc. The reinforcement material is a dispersed phase embedded in the matrix material. Its main function is to improve the mechanical properties, thermal properties and electrical properties of the composite material, such as carbon fiber, glass fiber, aramid fiber, etc.

[0013] Matrix material information refers to the composition, structure, and properties of the matrix material used to synthesize the composite material. Performance requirements refer to the performance requirements for the composite material, such as the hardness and toughness of the composite material. These requirements must be determined based on actual conditions. Performance support decomposition is then performed. In layman's terms, the matrix material's properties in the matrix material information are compared with the performance requirements. If any of the matrix material's properties deviate significantly from the performance requirements, and these deviations exceed a preset threshold, these properties are considered missing performance requirements. Performance support is the quantitative value of any matrix material property, and missingness is the ratio of the deviation between the performance requirement and the performance support to the performance requirement. It reflects the degree of deficiency in the matrix material in a specific performance. For example, if the composite material's performance requirement for tensile strength is 100 MPa, and the actual tensile strength of the matrix material is 90 MPa, the actual tensile strength is the performance support, the missing performance requirement is 10 MPa, and the missing degree is 10%. This provides the missing performance requirements and their missing degrees, facilitating subsequent optimization of the reinforcement formulation to ensure that the composite material's properties meet the performance requirements.

[0014] Step 2: Based on the missing performance requirements and their missing degrees, perform a reinforcement material matching analysis to determine a reinforcement material configuration relationship list; Specifically, based on the missing performance requirements and the degree of missing, select reinforcing materials that can make up for the missing performance requirements and the degree of missing. For example, if the tensile strength of the matrix material is insufficient, it is necessary to find fiber-reinforced materials that can significantly improve the strength. First, professional and technical personnel in this field construct a reinforcing material database based on existing technology. The reinforcing material database contains different reinforcing materials and corresponding reinforcing properties. For example, carbon fiber is usually used to improve strength and modulus, while glass fiber may be more suitable for improving heat resistance and corrosion resistance. Further based on the missing performance requirements and their degree of missing, traversal matching is performed in the reinforcing material database to match the reinforcing properties with the missing performance requirements and the corresponding reinforcing materials. Then, according to the degree of missing and the degree of improvement of the performance by a certain number of reinforcing materials, the addition ratio of the reinforcing material is obtained, and a list of reinforcing material configuration relationships is formed with reinforcing materials, reinforcing properties and addition ratios to provide a basis for subsequent formula optimization. It should be noted that one performance in the missing performance requirements can correspond to multiple reinforcing materials.

[0015] Step 3: Perform relationship matching analysis based on the missing performance requirements and their missing degrees, and the list of reinforcement material configuration relationships, fuse and extract the fusion performance relationship between the matrix material information and the reinforcement material, and configure the formula test variables; Specifically, after the reinforcing material is added, it will enhance the missing performance of the matrix material, and may also have a positive or negative impact on other properties, such as reducing strength while enhancing tensile strength. Based on this, a relationship matching analysis is performed according to the missing performance requirements and their missing degrees, and the reinforcing material configuration relationship list, that is, the reinforcing material and the reinforcing material addition ratio corresponding to any missing performance and its missing degree in the missing performance requirements are extracted, and further according to the reinforcing material and the reinforcing material addition ratio, the performance impact of the reinforcing material on the matrix material is analyzed to obtain a fusion performance relationship, that is, the fusion performance relationship includes the performance impact relationship of the reinforcing material on the matrix material. According to the fusion performance relationship, the proportion of the reinforcing material is adjusted so that the reinforcing material can make up for the missing performance requirements, which will not cause other properties to not meet the performance requirements. If a certain reinforcing material has a greater impact on other properties, then it is necessary to use a variety of reinforcing materials used to improve the missing performance requirements to improve the performance. Then the types and component ratios of the multiple reinforcing materials are the formula test variables.

[0016] Step 4: Acquire processing parameters, configure process test variables, construct a test space according to the recipe test variables and the process test variables, obtain a test data set, analyze and process the test data set, and construct a target data set; Specifically, processing parameters refer to the processing parameters for the synthetic processing of the matrix material and the reinforcing material, including but not limited to temperature, pressure, time, etc. Parameters among the processing parameters that have a significant impact on the performance of the composite material, such as temperature and pressure, are used as process test variables. Specifically, the correlation between different parameters and the performance of the composite material can be analyzed through existing correlation analysis methods, and parameters with correlations greater than a preset threshold can be obtained as process test variables. A multidimensional test space is constructed using formula test variables (type of reinforcing material, component ratio, etc.) and process test variables (temperature, pressure, time, etc.). Based on the constructed test space, the historical record data set is extracted to construct a test data set. The test data set is analyzed and processed, that is, the data of different dimensions in the test data set are mutated to obtain the target data set, which provides support for subsequent formula process optimization and ensures optimization accuracy.

[0017] Step 5: Based on the performance requirements and processing technology constraints, the target data set is optimized to obtain the formulation process optimization results.

[0018] Specifically, performance requirements include material strength, modulus, heat resistance, corrosion resistance, etc., while processing technology includes temperature, pressure, time, etc. Constraints refer to the range of constraints on parameters in the processing technology, that is, the processing technology may affect the performance of the material, and it is necessary to ensure that the material performance meets the performance requirements. The parameter range of the processing technology that meets the performance requirements is set as a constraint. Based on the constraints, the target data set is screened to find the optimal formula and process parameters that meet all constraints. That is, the formula process in the target data set that meets all constraints and is closest to the required performance is selected as the formula process optimization result. The formula process optimization result is included in the parameters corresponding to the processing parameters and the configuration process test variables, thereby achieving the formula process optimization of the composite material, ensuring that the performance of the composite material meets the performance requirements, and improving the synthesis accuracy of the composite material.

[0019] Furthermore, step three of this application also includes: According to the missing performance requirements and their missingness, a priority is set according to the missingness, and the list of reinforcement material configuration relationships is traversed and matched according to the priority to determine the matching relationship of the reinforcement material; according to the fusion performance relationship between the matrix material information and the reinforcement material, a fusion adjustment coefficient is generated, and the fusion adjustment coefficient includes forward fusion adjustment and reverse fusion adjustment, and the matching relationship of the reinforcement material is adjusted using the fusion adjustment coefficient; according to the various performance indicators of the performance requirements, the matching relationship of the reinforcement material is configured and extracted to obtain the deviation performance indicator as the formula test variable.

[0020] Specifically, a relationship matching analysis is performed based on the missing performance requirements and their missing degrees, and the list of reinforcement material configuration relationships, and the fusion performance relationship between the matrix material information and the reinforcement material is extracted and configured. The process of formula test variables is as follows: Based on the missing performance requirements and their degree of missingness, a priority is set according to the degree of missingness, where the degree of missingness is proportional to the priority. That is, the missing performance requirements are sorted from largest to smallest in terms of degree of missingness, with the missing performance requirement with the largest degree of missingness having the highest priority. Then, a traversal matching is performed in the reinforcement material configuration relationship list according to priority. That is, the highest-priority missing performance requirement is matched first, and then the corresponding reinforcement materials and their proportions are prioritized for the higher-priority missing performance requirements in the reinforcement material configuration relationship list, which are then used as the reinforcement material matching relationship.

[0021] Furthermore, a fusion adjustment coefficient is generated based on the relationship between the matrix material information and the fusion performance of the reinforcing material. The fusion adjustment coefficient includes forward fusion adjustment and reverse fusion adjustment. Among them, the forward fusion adjustment refers to the influence between the matrix material information and the materials contained in the reinforcing material, which affects the performance of the composite material. The forward fusion adjustment means that the direction of influence of different types of materials on the performance is the same. For example, both materials will increase the tensile strength. The superposition of the degree of influence of different types of materials on the performance can be used as a fusion adjustment coefficient; the reverse fusion adjustment means that different types of materials have different directions of influence on the composite material. For example, one reinforcing material will increase the tensile performance, and the other reinforcing material will reduce the tensile performance. According to the direction of influence of different types of materials on the performance, the degree of influence of different types of materials on the performance is offset as a fusion adjustment coefficient.

[0022] Then, the matching relationship of the reinforcing materials is adjusted using the fusion adjustment coefficient, that is, the enhanced performance in the matching relationship of the reinforcing materials is adjusted and optimized using the fusion adjustment coefficient. Furthermore, the matching relationship of the reinforcing materials is configured and extracted according to the various performance indicators of the performance requirements to obtain deviation performance indicators. In simple terms, according to the various performance indicators of the performance requirements, the performance indicators corresponding to the performance requirements after adding the reinforcing materials are extracted based on the matching relationship of the reinforcing materials, such as tensile strength, toughness, etc. Therefore, the various performance indicators of the performance requirements and the performance indicators corresponding to the performance requirements after adding the reinforcing materials are compared. If the various performance indicators of the performance requirements and the performance indicators corresponding to the performance requirements after adding the reinforcing materials are consistent, they will not be used as formula test variables. If they are inconsistent, that is, the performance requirements cannot be met after adding multiple reinforcing materials, the corresponding reinforcing materials and proportions will be used as formula test variables. The formula test variables can be optimized later to ensure that the performance of the composite material meets the requirements.

[0023] Furthermore, step 4 of this application also includes: Based on the recipe test variables and the process test variables, a historical record data set is obtained; the historical record data set is traversed according to the various performance indicators of the performance requirements to obtain an original record data set; based on the original record data set, missing variable ranges and variable inflection point predictions are extracted; based on the missing variable ranges and the variable inflection points, an experimental space is set to conduct experiments to obtain an experimental data set.

[0024] Furthermore, the present application further comprises the following steps: Taking the historical record dataset as the starting point and the experimental record dataset as the mutation direction, mutation derivation is performed to obtain a mutated dataset; fitness evaluation is performed on the mutated dataset, and mutated individuals that are better than the original record dataset are selected to replace the original record dataset to obtain the target dataset.

[0025] Furthermore, the present application further comprises the following steps: The step length of the mutation is set, and mutation is performed with the historical record dataset as the starting point and the test record dataset as the mutation direction to obtain a first mutation dataset, wherein the step length of the mutation is a positive number randomly selected from the difference between the minimum value of the variable in the historical record dataset and the minimum value of the variable in the test record dataset; a cross-derivation rule is obtained, and the historical record dataset and the first mutation dataset are subjected to co-location and co-dimensional cross-mutation according to the cross-derivation rule to obtain the mutation dataset.

[0026] Furthermore, the present application further comprises the following steps: According to the constraint strength of each performance indicator of the performance requirement, the original data set is screened from the historical record data set, and the performance derivative direction of the original data set is determined, where the performance derivative direction is the direction of change that is positively correlated with the performance indicator; based on the historical record data set, the control variable dimension is determined, and a dimension mapping matrix is ​​configured; the original data set is projected into the dimension mapping matrix, and the performance derivative direction is used as a constraint condition for the dimension variable; based on the variable dimension, a random number interval is determined, where the random number interval is from 1 to the variable dimension, and a random number function is used to generate a randomly updated dimension number, and according to the randomly updated dimension number and the performance derivative direction, the same-position and same-dimensional parameters are randomly selected from the variant data set for replacement.

[0027] Furthermore, the present application further comprises the following steps: The original record data set is extracted according to the recipe test variables and the process test variables to construct a data set of each test variable; the data set of each test variable is ordinalized according to the data size to obtain the data distribution, and the missing interval is screened according to the data distribution to obtain the missing variable range; a data distribution curve is fitted based on the data distribution, and an inflection point prediction is performed according to the data distribution curve.

[0028] Specifically, the process of constructing the test space according to the recipe test variables and the process test variables to obtain the test data set is as follows: Using the recipe test variables and the process test variables as index elements, historical production data of the composite material corresponding to the recipe test variables and the process test variables is retrieved based on existing internet big data to obtain a historical record data set, which includes historical variables corresponding to the recipe test variables and the process test variables. The historical record data set is traversed according to each performance indicator of the performance requirement, that is, data corresponding to each performance indicator of the performance requirement is extracted as the original record data set.

[0029] Furthermore, the missing variable range and variable inflection point prediction are extracted based on the original record data set, and the method is as follows: the original record data set is extracted according to the recipe test variables and the process test variables, that is, the data corresponding to each test variable in the recipe test variables and the process test variables are extracted from the original record data set as the data set of each test variable. The data set of each test variable is sorted according to the data size, that is, it is sorted in order from large to small, and the sorting result is used as the data distribution of each test variable. The missing interval is screened according to the data distribution, that is, the interval without any data distribution is extracted as the missing variable range. In simple terms, after the data set of each test variable is sorted according to the data size, the data of each test variable is not in some intervals. For example, the data set corresponding to a certain test variable contains data from 1 to 30. After sorting in sequence, it is found that there is no data distribution in the range of 10 to 15. This range is the missing variable range.

[0030] Further, a curve is drawn based on the data distribution to obtain a fitted data distribution curve. The fitted data distribution curve can represent the changing trend of the data set of each experimental variable, thereby performing inflection point prediction based on the data distribution curve. The inflection point usually refers to the local maximum or minimum point of the data distribution curve, or the point where the slope of the curve changes significantly. Inflection point identification is performed to obtain the local maximum or minimum point of the data distribution curve, or the point where the slope of the curve changes significantly as the variable inflection point, providing a basis for the subsequent construction of the experimental data set.

[0031] Then, according to the missing variable range and the variable inflection point, the test space is set for testing, that is, the endpoint value, the middle value, etc. of the missing variable range, and the variable value corresponding to the variable inflection point are selected as the test space to conduct composite material synthesis tests, and the formula test variables, the variable parameters corresponding to the process test variables and the performance of the composite material during the test are recorded, which is used as the test data set to facilitate subsequent data variation analysis, provide data support for formula process optimization, and facilitate the screening of better process parameters.

[0032] The process of analyzing and processing the experimental data set and constructing the target data set is as follows: First, a mutation is performed using the historical record dataset as a starting point and the test record dataset as a mutation direction to obtain a mutated dataset. The specific process is as follows: a variable step length is set. The variable step length refers to the step length for updating parameters in the historical record dataset, wherein the variable step length is a positive number randomly selected from the difference between the minimum value of the variable in the historical record dataset and the minimum value of the variable in the test record dataset. Mutation is performed using the historical record dataset as a starting point and the test record dataset as a mutation direction. That is, the historical record dataset is updated according to the variable step length according to the direction of data change in the test record dataset, and the updated historical record dataset is used as the first mutated dataset.

[0033] Further, obtain the cross-derivation rules as follows: First, based on the performance requirements, data that meet the strength limits of various performance indicators are screened out from the historical record data set. For example, if the performance requirements include tensile strength, heat resistance, etc., the strength limits of the performance indicators are the range of constraints on these performance thresholds. Only data that meet the strength limits of the performance indicators will be selected as the original data set. The filtered original data set is analyzed to determine the performance derivative direction. The performance derivative direction refers to the direction of change that is positively correlated with the performance indicator, that is, the performance improves as the direction changes. This can be achieved through existing statistical analysis and regression analysis. Based on the historical record data set, the variables that affect the performance are determined. These variables are parameters that can be adjusted to affect the performance, such as the composition of the reinforcing material, processing temperature, pressure, etc. Each parameter is used as a control variable dimension. According to the control variable dimension, a dimension mapping matrix is ​​constructed. The elements of each row in the dimension mapping matrix refer to the parameters corresponding to a dimension, and the elements of each column refer to the control variable dimension.

[0034] The filtered original data set is projected into a dimension mapping matrix. Each row of the dimension mapping matrix represents the parameters belonging to the same control variable dimension in the original data set, and each column represents the control variable dimension. The performance derivative direction is used as a constraint condition to determine a random number interval, usually from 1 to the control variable dimension. This interval will be used to generate a random update dimension number. A random number function, such as a pseudo-random number generator, is used to generate a random number within the random number interval as the random update dimension number. Based on the randomly generated update dimension number and performance derivative direction, parameter values ​​belonging to the same position and the same control variable dimension are randomly selected from the variant data set for replacement to achieve the mining of positional formula processes, facilitate the improvement of the comprehensiveness of optimization, and thus ensure the adaptability of the formula process to the required performance of the composite material.

[0035] The historical record dataset and the first variant dataset are subjected to co-location and co-dimensional crossover mutation according to the crossover derivation rule to obtain the variant dataset.

[0036] Finally, according to the cross-derivation rule, the historical record dataset and the first variant dataset are subjected to co-location and co-dimensional cross-mutation. This involves randomly selecting parameter values ​​from the first variant dataset that are in the same position and control variable dimension as the historical record dataset, replacing the data in the historical record dataset to obtain the variant dataset. Furthermore, the variant dataset is subjected to fitness evaluation. This involves obtaining performance indicators corresponding to the required performance for the variant dataset, and selecting variant individuals with performance indicators superior to those of the original record dataset to replace the original record dataset. These variant individuals contain parameters corresponding to multiple control variable dimensions, serving as the target dataset. This helps find a more optimal parameter combination within the control variable dimension to improve the performance of the composite material and ensure compatibility with the performance requirements.

[0037] Furthermore, step five of this application also includes: Collect a set of abnormal cases, cluster them according to performance abnormalities based on the abnormal case set, and obtain multiple performance abnormality clusters; perform parameter probability and abnormal relationship fitting on the multiple performance abnormality clusters to obtain abnormal parameter information; extract environmental parameters and equipment control parameter constraints of the processing technology to obtain process constraint parameters and constraint ranges; use the abnormal parameter information to perform correlation analysis with the process constraint parameters and constraint ranges to obtain the constraint conditions.

[0038] Specifically, based on the performance requirements and the constraints set for the processing technology, the target data set is optimized to obtain the formulation process optimization results. The method for obtaining the constraints is as follows: First, we need to collect cases of performance anomalies during the processing process, including scrap, rework, and defective products reported by customers. The anomaly case set includes abnormal performance and corresponding process parameters. We use existing statistical or machine learning methods such as K-means clustering, hierarchical clustering, and DBSCAN to perform cluster analysis on the anomaly case set, grouping similar anomalies together to form multiple performance anomaly clusters.

[0039] Each performance anomaly cluster is thoroughly analyzed to fit parameter probabilities and anomaly relationships. Parameter probabilities represent the distribution probability of process parameters within the anomaly cluster and can be obtained using existing statistical learning or machine learning methods. Anomaly relationships refer to the mutual influence between parameters and can be obtained using existing correlation analysis methods. Parameters with high probabilities and those that influence other parameters are selected as anomaly parameter information. Furthermore, the processing technology is analyzed to identify environmental parameters (such as temperature, humidity, and air pressure) and equipment control parameters (such as speed, pressure, and power) that have a direct impact on composite material performance as process constraints. Simultaneously, based on historical experience, professionals in the field determine the constraint ranges for these parameters—i.e., the minimum and maximum values ​​they should maintain under normal production conditions—as the constraint ranges. Correlation analysis is performed between the anomaly parameter information and the process constraints and their ranges, using existing data mining techniques such as association rule learning and mutual information calculation. This identifies which parameter anomalies are associated with violations of the process constraints and their ranges. The violated process constraints and their ranges are then determined as the constraints. Subsequently, formulation and process optimization is performed within the constraints to reduce the risk of anomalies after composite material production and ensure that the composite material's performance matches the required performance.

[0040] In summary, the composite material formulation process optimization method based on performance requirements provided in this application has the following technical effects: Obtaining matrix material information and performance requirements, performing performance support decomposition based on the matrix material information and the performance requirements, determining missing performance requirements and their missingness, wherein the missing performance requirements are performance requirements with a deviation value between the performance requirements and the performance support value greater than a preset value, and the missingness is the ratio of the deviation value between the performance requirements and the performance support value to the performance requirements; performing reinforcement material matching analysis based on the missing performance requirements and their missingness, and determining a reinforcement material configuration relationship list; performing relationship matching analysis based on the missing performance requirements and their missingness and the reinforcement material configuration relationship list, fusing and extracting the fusion performance relationship between the matrix material information and the reinforcement material, and configuring formula test variables; obtaining processing parameters, configuring process test variables, constructing a test space based on the formula test variables and the process test variables, obtaining a test data set, analyzing and processing the test data set, and constructing a target data set; setting constraints based on the performance requirements and the processing technology, optimizing the target data set, and obtaining a formula process optimization result. Thus, after performing missing performance analysis on the matrix material, optimizing the formula processing technology of the reinforcement material, the accuracy of the formula process of the composite material is improved, and the compatibility of the performance of the composite material with the required performance is ensured.

[0041] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0042] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A composite material formulation process optimization method based on performance requirements, characterized in that: include: Obtaining base material information and performance requirements, performing performance support decomposition based on the base material information and the performance requirements, and determining missing performance requirements and their missingness, wherein the missing performance requirements are performance requirements for which the deviation between the performance requirements and the performance support is greater than a preset value, and the missingness is the ratio of the deviation between the performance requirements and the performance support to the performance requirements; Based on the missing performance requirements and their missing degrees, performing a reinforcement material matching analysis to determine a reinforcement material configuration relationship list; Perform relationship matching analysis based on the missing performance requirements and their missing degrees and the reinforcement material configuration relationship list, fuse and extract the fusion performance relationship between the matrix material information and the reinforcement material, and configure the formula test variables; Acquire processing parameters, configure process test variables, construct a test space according to the recipe test variables and the process test variables, obtain a test data set, analyze and process the test data set, and construct a target data set; Based on the performance requirements and processing technology constraints, the target data set is optimized to obtain the recipe process optimization results.

2. The method according to claim 1, wherein Perform relationship matching analysis based on the missing performance requirements and their missing degrees and the reinforcement material configuration relationship list, fuse and extract the fusion performance relationship between the matrix material information and the reinforcement material, and configure the formula test variables, including: According to the missing performance requirements and their missing degrees, priorities are set according to the missing degrees, and matching is performed with the list of reinforcement material configuration relationships according to the priorities to determine matching relationships of the reinforcement materials; Generate a fusion adjustment coefficient based on the relationship between the matrix material information and the fusion performance of the reinforcement material, wherein the fusion adjustment coefficient includes a forward fusion adjustment and a reverse fusion adjustment, and adjust the matching relationship of the reinforcement material using the fusion adjustment coefficient; The matching relationship of the reinforcing material is configured and extracted according to each performance index of the performance requirement to obtain a deviation performance index as the formula test variable.

3. The method according to claim 2, wherein Constructing a test space according to the recipe test variables and the process test variables to obtain a test data set includes: Based on the recipe test variables and the process test variables, obtaining a historical record data set; Traversing the historical record data set according to each performance indicator of the performance requirement to obtain an original record data set; Extract missing variable ranges and variable inflection point predictions based on the original record data set; According to the missing variable range and the variable inflection point, an experimental space is set to conduct an experiment to obtain an experimental data set.

4. The method according to claim 3, wherein The analyzing and processing of the test data set to construct the target data set includes: Taking the historical record dataset as a starting point and the test record dataset as a mutation direction, performing mutation derivation to obtain a mutation dataset; The fitness of the variant data set is evaluated, and variant individuals that are better than the original record data set are selected to replace the original record data set to obtain the target data set.

5. The method according to claim 4, wherein Taking the historical record dataset as a starting point and the test record dataset as a mutation direction, mutation derivation is performed to obtain a mutation dataset, including: Setting a mutation step length, taking the historical record dataset as a starting point and the test record dataset as a mutation direction to perform mutation, to obtain a first mutation dataset, wherein the mutation step length is a positive number randomly selected from the difference between the minimum value of the variable in the historical record dataset and the minimum value of the variable in the test record dataset; A cross-derivation rule is obtained, and the historical record dataset and the first variant dataset are subjected to co-location and co-dimensional cross-mutation according to the cross-derivation rule to obtain the variant dataset.

6. The method according to claim 5, wherein The obtaining of cross-derivation rules includes: Filtering an original data set from the historical record data set according to the constraint strength of each performance indicator of the performance requirement, and determining a performance derivative direction for the original data set, wherein the performance derivative direction is a change direction that is positively correlated with the performance indicator; Based on the historical record data set, determining the control variable dimension and configuring the dimension mapping matrix; Projecting the original data set into the dimension mapping matrix, and using the performance derivative direction as a constraint condition of the dimension variable; Based on the variable dimension, a random number interval is determined, where the random number interval is from 1 to the variable dimension. A random number function is used to generate a random update dimension number. According to the random update dimension number and the performance derivative direction, the same-position and same-dimensional parameters are randomly selected from the variant data set for replacement.

7. The method according to claim 3, wherein Extracting missing variable ranges and variable inflection point predictions based on the original record data set includes: Extracting the original record data set according to the recipe test variables and the process test variables to construct a data set for each test variable; The data sets of the test variables are sorted according to the data size to obtain the data distribution, and the missing intervals are screened according to the data distribution to obtain the missing variable range; A data distribution curve is fitted based on the data distribution, and an inflection point prediction is performed according to the data distribution curve.

8. The method according to claim 1, wherein Based on the performance requirements and processing technology constraints, the target data set is optimized to obtain the formulation process optimization results, including: Collecting an abnormal case set, clustering the abnormal case set according to performance anomalies to obtain multiple performance anomaly clusters; Perform parameter probability and abnormal relationship fitting on the multiple performance abnormality clusters to obtain abnormal parameter information; Extracting environmental parameters and equipment control parameter constraints of the processing technology to obtain process constraint parameters and constraint ranges; The constraint conditions are obtained by performing a correlation analysis with the abnormal parameter information and the process constraint parameters and constraint ranges.

Images