An efficient material retrieval method and system for textile production

Through the combination of big data analysis, intelligent matching algorithms and virtual simulation technology, the material database is dynamically updated, which solves the problems of low selection efficiency of traditional textile material and lagging database updates, and achieves efficient and accurate material matching and database updates to adapt to market changes. Improved the efficiency and quality of textile design and production.

CN118430724BActive Publication Date: 2024-11-12NANTONG DIEXIU TEXTILE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410885245.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-03
Publication Date
2024-11-12
Estimated Expiration
2044-07-03

AI Technical Summary

Technical Problem

Traditional textile material selection methods are inefficient, difficult to accurately match material properties and textile design needs, and the material database update is lagging, making it impossible to adapt to market changes and the emergence of new materials.

Method used

A material database is established using big data analysis, a material requirement parameter model is established based on textile design requirements, an intelligent matching algorithm and virtual simulation technology is used to evaluate the applicability of materials, and an optimization mechanism based on machine learning is introduced to dynamically update the database.

Benefits of technology

It improves the efficiency and accuracy of material selection, reduces trial and error costs, accelerates the product development cycle, ensures that material selection is based on the latest data, adapts to market changes, and improves the forward-looking and innovative design.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118430724B_ABST
    Figure CN118430724B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of retrieval methods, and in particular to an efficient material retrieval method and system for textile production, comprising the following steps: S1: using big data analysis to collect textile material performance data and establish a material database; S2: establishing a material demand parameter model based on textile design requirements; S3: intelligently retrieving material data in a matching material database based on a material attribute intelligent matching algorithm; S4: using virtual simulation technology to simulate the performance of the matched materials in S3 in an actual textile process and evaluate the applicability of the materials; S5: introducing an optimization mechanism based on machine learning to continuously update and optimize the material database to adapt to changes in the textile market and the emergence of new materials. The present invention and the comprehensive method greatly improve the efficiency of material selection, reduce the trial and error cost, and accelerate the product development cycle.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of retrieval methods, and in particular to an efficient material retrieval method and system for textile production. Background Art

[0002] In the development of the textile industry, material selection has always been a core link, directly affecting product quality, performance and production costs. Traditional material selection methods mainly rely on manual experience and limited experimental data. This method is inefficient when processing a large amount of complex material data, and it is difficult to accurately capture the complex relationship between material properties and textile design requirements. With the diversification of textile design and the increase in personalized needs, traditional methods have been unable to meet the high efficiency and high precision requirements of the modern textile industry for material selection.

[0003] In addition, the rapid changes in the textile market and the continuous emergence of new materials require that the material database must be able to quickly update and optimize to adapt to market demand and technological progress. However, the existing material database management methods lack an effective mechanism to automatically identify and integrate new material information from different sources, resulting in a lag in database updates and failure to provide the latest material data to support the design and production of textiles.

[0004] Therefore, there is an urgent need for a new method that can efficiently and accurately match material properties with textile design requirements, while also having the ability to dynamically update the material database to adapt to market changes, in order to improve the efficiency and quality of textile design and production. Summary of the invention

[0005] Based on the above objectives, the present invention provides an efficient material retrieval method and system for textile production.

[0006] An efficient material retrieval method for textile production comprises the following steps:

[0007] S1: Use big data analysis to collect textile material performance data and establish a material database;

[0008] S2: Establish material demand parameter model based on textile design requirements;

[0009] S3: Based on the intelligent matching algorithm of material properties and the material requirement parameter model established in S2, intelligently retrieve the material data in the matching material database;

[0010] S4: Use virtual simulation technology to simulate the performance of the materials matched in S3 in the actual textile process and evaluate the applicability of the materials;

[0011] S5: Introduce a machine learning-based optimization mechanism to continuously update and optimize the material database to adapt to changes in the textile market and the emergence of new materials.

[0012] Furthermore, the S1 specifically includes:

[0013] S11: Set data collection standards, including physical properties, chemical properties and color attributes of textile materials;

[0014] S12: Automatically collect performance data of textile materials from public scientific research literature, patents, product manuals and official websites of textile material manufacturers;

[0015] S13: Use natural language processing technology to process and analyze the collected unstructured data, extract information that matches the set data collection standards, and convert it into structured data;

[0016] S14: Data cleaning techniques are used for the collected data, including removing duplicate data, correcting erroneous data, and filling missing values;

[0017] S15: Store and manage the cleaned structured data.

[0018] Furthermore, the S2 specifically includes:

[0019] S21: Analyze the design requirements of textiles, including durability, elasticity, color fastness, chemical resistance, and environmental protection;

[0020] S22: According to the parameter quantification standard of each design requirement, the design requirements are converted into specific material parameters;

[0021] S23: Establish a parameter weight system. According to the importance of the design requirements, assign weights to each material parameter through the hierarchical analysis method to ensure that the sum of all material parameter weights is 1. Perform weight standardization. The formula is: ,in, is the material parameter The standardized weights of is the original weight, N is the total number of material parameters;

[0022] S24: Use the weighted summation method to build a material demand parameter model and calculate the comprehensive score of each material to reflect the degree to which the material meets the design requirements. The comprehensive score represents: , where S is the comprehensive score of the material, It is The standardized value of a material parameter, It is The weight of a material parameter.

[0023] Furthermore, the analytic hierarchy process in S23 specifically includes:

[0024] S231, establish a hierarchical model:

[0025] Target layer: the final goal, selecting the best textile material;

[0026] Criteria layer: standards or criteria for evaluating materials, i.e. material parameters converted into textile design requirements;

[0027] Solution level: material options;

[0028] S232, construct a judgment matrix: At the criterion level, for each pair of material parameters, compare the relative importance of achieving the target level, give a quantitative ratio, and form a judgment matrix A, where the elements in the judgment matrix are Indicates material parameters Relative to material parameters The importance of

[0029] S233, calculate weight: use the eigenvalue and eigenvector of the judgment matrix to calculate the weight. The eigenvector corresponding to the maximum eigenvalue is the relative weight of each material parameter. The calculation steps include solving the maximum eigenvalue of the judgment matrix A. and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector ;

[0030] S234, consistency test: calculate the consistency index CI and consistency ratio CR, where:

[0031] , where n is the number of criterion layers;

[0032] ,in, is the average random consistency index, which depends on the number of criterion levels. If , then the consistency of the judgment matrix is ​​acceptable and the weight result is reliable.

[0033] Furthermore, the S3 specifically includes:

[0034] S31: Taking each material parameter and corresponding weight in the material demand parameter model as input conditions, constructing query conditions of the intelligent matching algorithm;

[0035] S32: Using an intelligent matching algorithm, searching in a material database, and scoring according to the degree of matching between the material data and the input conditions;

[0036] S33: The intelligent matching algorithm uses cosine similarity to calculate the similarity between each material data in the material database and the material demand parameter model;

[0037] S34: ranking the materials in the database according to the similarity scores, and selecting the material with the highest score as the matching result;

[0038] S35: Output the matching results, including a list of materials meeting the textile design requirements and related performance data.

[0039] Furthermore, the cosine similarity is calculated as: , where A and B represent the material property vector and the material demand parameter model vector respectively, and The vectors A and B are The weight of the dimension, It is The weight of a material parameter reflects the relative importance of the material parameter in material selection, and n is the dimension of the vector, that is, the total number of material parameters considered.

[0040] Furthermore, the S4 specifically includes:

[0041] S41: Select the material that meets the requirements selected by the intelligent matching algorithm, and extract the property data of the selected material, including physical properties and chemical properties;

[0042] S42: Build a virtual model of a textile design, including the design’s structure, dimensions, and intended use information;

[0043] S43: Using ANSYS simulation software, apply the property data of the selected materials to the virtual model of textile design and set the simulation parameters, including environmental conditions and loading conditions;

[0044] S44: Run simulations to simulate the performance of materials during the textile process, including the response of materials under stress, wear, and environmental influences, and record simulation results, including deformation, strength loss, and color change;

[0045] S45: Analyze simulation results, evaluate the performance of materials in the actual textile process, and determine whether the materials meet the design requirements and intended use;

[0046] S46: If the simulation results show that the material performance is poor or cannot meet the requirements, return to S3 based on the evaluation results and re-match the material for simulation evaluation.

[0047] Furthermore, the S5 specifically includes:

[0048] S51: Build a machine learning-based model to analyze textile market trends, user preferences, and performance data of new materials to learn and identify patterns and trends from market data;

[0049] S52: Regularly collect and organize new information on textile materials from market reports, scientific research articles, patents, and online databases, including emerging materials, updated data on material properties, and new trends in textile design;

[0050] S53: Use the established machine learning model to automatically analyze the collected new information, evaluate the importance and reliability of the new information, and determine the new information to be included in the material database;

[0051] S54: Automatically update the material database based on the analysis results of the machine learning model, including adding data of new materials, updating performance parameters of existing materials, and deleting data of outdated or no longer used materials.

[0052] Furthermore, the machine learning model adopts a random forest model, which specifically includes:

[0053] Data preparation: Collect and organize the property data of textile materials, including new materials and updated information of existing materials, as well as performance data in actual applications, and label each data point (material) with a classification label;

[0054] Model training: Use the random forest algorithm to build multiple decision trees on the training set. During the construction of each tree, a feature subset is randomly selected and the best split point is searched on the feature subset. The decision tree is built without pruning. Each tree grows randomly until it cannot be split. The training formula for each tree in the random forest is: ,in, It is trees, D is the training data set, is a randomly selected subset of all features;

[0055] Prediction and evaluation:

[0056] For classification problems, the prediction result of random forest is the voting result of the majority of trees. For regression problems, the prediction result is the average of the prediction values ​​of all trees. Cross-validation or a reserved test set is used to evaluate the accuracy and generalization ability of the model.

[0057] Applied to material database optimization: Apply the trained random forest model to the collected new material data to predict whether the new material is suitable for the current market trend or design requirements, and update the material database based on the model's prediction results, including adding new materials, updating the ratings or classifications of existing materials, and removing outdated materials or materials that no longer meet the requirements.

[0058] An efficient material retrieval system for textile production, used to implement the above-mentioned efficient material retrieval method for textile production, includes the following modules:

[0059] Data collection and database establishment module: automatically collects performance data of textile materials, including physical properties, chemical properties and environmental adaptability, through big data technology, and establishes and maintains a material database;

[0060] Demand parameter model building module: converts textile design requirements into quantifiable material demand parameters and forms a material demand parameter model to guide the subsequent material retrieval process;

[0061] Intelligent matching algorithm module: Based on the intelligent matching algorithm of material properties, the optimal material is retrieved and matched from the material database according to the established material requirement parameter model;

[0062] Virtual simulation evaluation module: Use virtual simulation technology to simulate the performance of the matched materials in the actual textile process, evaluate the applicability and performance of the materials through simulation results, and verify the selection of materials;

[0063] Database optimization and update module: Introduces a machine learning-based optimization mechanism to continuously update and optimize the material database based on market trends, user feedback, and new material information.

[0064] Beneficial effects of the present invention:

[0065] The present invention, by combining big data analysis, intelligent matching algorithms and virtual simulation technology, can quickly and accurately screen out materials that meet specific textile design requirements. This comprehensive method greatly improves the efficiency of material selection, reduces trial and error costs, and accelerates the product development cycle.

[0066] The present invention introduces an optimization mechanism based on machine learning that can continuously update and optimize the material database to adapt to the rapid changes in the textile market and the emergence of new materials. This dynamic adaptability ensures that the material selection process is always based on the latest and most accurate data, thereby improving the foresight and innovation of the design.

[0067] The present invention, by introducing improved cosine similarity calculation, brings remarkable results in accurately matching material properties with textile design requirements. By considering the weight of each attribute, it ensures that the material selection process is more in line with actual design requirements and application scenarios, and improves the accuracy and reliability of material matching. This refined matching method not only improves the performance and quality of the final product, but also provides strong technical support for innovative design and customized production of textiles. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0069] Figure 1 A schematic diagram of a retrieval method flow chart of an embodiment of the present invention;

[0070] Figure 2 Schematic diagram of system function modules according to an embodiment of the present invention. DETAILED DESCRIPTION

[0071] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0072] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0073] like Figure 1 As shown, an efficient material retrieval method for textile production comprises the following steps:

[0074] S1: Use big data analysis to collect textile material performance data and establish a material database;

[0075] S2: Establish material demand parameter model based on textile design requirements;

[0076] S3: Based on the intelligent matching algorithm of material properties and the material requirement parameter model established in S2, intelligently retrieve the material data in the matching material database;

[0077] S4: Use virtual simulation technology to simulate the performance of the materials matched in S3 in the actual textile process and evaluate the applicability of the materials;

[0078] S5: Introduce a machine learning-based optimization mechanism to continuously update and optimize the material database to adapt to changes in the textile market and the emergence of new materials.

[0079] S1 specifically includes:

[0080] S11: Set data collection standards, including physical properties (strength, elasticity), chemical properties (corrosion resistance, heat resistance) and color properties (color fastness, color difference) of textile materials;

[0081] S12: Automatically collect performance data of textile materials from public scientific research literature, patents, product manuals and official websites of textile material manufacturers;

[0082] S13: Use natural language processing technology to process and analyze the collected unstructured data, extract information that matches the set data collection standards, and convert it into structured data;

[0083] S14: Data cleaning techniques are used for the collected data, including removing duplicate data, correcting erroneous data, and filling missing values;

[0084] S15: Store and manage the cleaned structured data.

[0085] S2 specifically includes:

[0086] S21: Analyze the design requirements of textiles, including durability, elasticity, color fastness, chemical resistance, and environmental protection;

[0087] S22: According to the parameter quantification standard of each design requirement, the design requirements are converted into specific material parameters. The specific quantification standards are as follows:

[0088] 1. Durability:

[0089] Parameter definition: Measures the service life of textiles and their ability to maintain performance under specific conditions of use.

[0090] Quantification method:

[0091] Wear testing: Conducted by standard wear testing equipment (Martindale wear tester), recording the number of cycles required to reach a specific degree of wear.

[0092] Tear Strength: A measure of a textile's ability to resist tearing when subjected to force, measured in Newtons, using standard tear testing equipment.

[0093] 2. Elasticity:

[0094] Parameter definition: Describes the ability of a textile to return to its original shape after being stretched.

[0095] Quantification method:

[0096] Stretch recovery rate: Measures the percentage of the original length that a textile can recover after being subjected to a certain stretching force and then released.

[0097] Elastic recovery time: The time required for the material to recover from its maximum stretched state to its initial state.

[0098] 3. Color speed:

[0099] Parameter definition: A measure of the ability of textiles to maintain color stability under various conditions (washing, sun exposure, friction).

[0100] Quantification method:

[0101] Washing color fastness test: According to ISO105C06 standard, the color change or fading degree is evaluated after multiple washings, and the score is scored on a grading system from 1 to 5, with 5 being the best.

[0102] Daylight color fastness test: According to ISO105B02 standard, expose to artificial light under controlled conditions, then evaluate the color change and use a graded rating system.

[0103] 4. Chemical resistance:

[0104] Parameter definition: Describes the ability of textiles to resist damage from chemical substances, including acids, alkalis, solvents, etc.

[0105] Quantification method:

[0106] Chemical corrosion testing: Evaluating changes in the quality, strength or appearance of textiles after exposing them to specific chemicals for a certain period of time.

[0107] pH Retention: Measures the change in pH of textiles after chemical treatment, compared to the original pH.

[0108] 5. Environmental protection:

[0109] Parameter definition: A measure of the environmental impact of textiles during production, use and disposal.

[0110] Quantification method:

[0111] Life cycle assessment: Evaluate the environmental impact of textiles from raw material acquisition, production, use to disposal.

[0112] Recyclability index: It is assessed based on the material, structure and ease of disposal of textiles after disposal, usually expressed as a percentage.

[0113] S23: Establish a parameter weight system. According to the importance of the design requirements, assign weights to each material parameter through the hierarchical analysis method. For example, if the design focuses on durability, then parameters related to durability (such as wear test results) will be given higher weights. Ensure that the sum of all material parameter weights is 1 and perform weight standardization. The formula is: ,in, is the material parameter The standardized weights of is the original weight, N is the total number of material parameters;

[0114] S24: Use the weighted summation method to build a material demand parameter model and calculate the comprehensive score of each material to reflect the degree to which the material meets the design requirements. The comprehensive score represents: , where S is the comprehensive score of the material, It is The standardized value of a material parameter, It is The weights of the material parameters are used to convert the design requirements of the textile into a set of actionable, quantified material property requirements.

[0115] The analytic hierarchy process is a decision analysis method that breaks down the decision problem into smaller parts, analyzes and evaluates them separately, and then combines the analysis results of these parts to make the final decision. The analytic hierarchy process in S23 specifically includes:

[0116] S231, establish a hierarchical model:

[0117] Target layer: the final goal, selecting the best textile material;

[0118] Criteria layer: standards or criteria for evaluating materials, i.e. material parameters converted into textile design requirements;

[0119] Solution level: material options;

[0120] S232, construct a judgment matrix: At the criterion level, for each pair of material parameters, compare the relative importance of achieving the target level and give a quantitative ratio. The ratio is assigned according to a standard relative importance level, such as from 1 (both are equally important) to 9 (one is extremely more important than the other), forming a judgment matrix A, where the elements in the judgment matrix are Indicates material parameters Relative to material parameters The importance of

[0121] S233, calculate weight: use the eigenvalue and eigenvector of the judgment matrix to calculate the weight. The eigenvector corresponding to the maximum eigenvalue is the relative weight of each material parameter. The calculation step includes solving the maximum eigenvalue of the judgment matrix A. and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector ;

[0122] S234, consistency test: calculate the consistency index CI and consistency ratio CR, where:

[0123] , where n is the number of criterion layers;

[0124] ,in, is the average random consistency index, which depends on the number of criterion levels. If , then the consistency of the judgment matrix is ​​acceptable and the weight result is reliable.

[0125] S3 specifically includes:

[0126] S31: Taking each material parameter and corresponding weight in the material demand parameter model as input conditions, constructing query conditions of the intelligent matching algorithm;

[0127] S32: Using an intelligent matching algorithm, searching in a material database, and scoring according to the degree of matching between the material data and the input conditions;

[0128] S33: The intelligent matching algorithm uses cosine similarity to calculate the similarity between each material data in the material database and the material demand parameter model;

[0129] S34: ranking the materials in the database according to the similarity scores, and selecting the material with the highest score as the matching result;

[0130] S35: Output the matching results, including a list of materials meeting the textile design requirements and related performance data.

[0131] Cosine similarity is calculated as: , where A and B represent the material property vector and the material demand parameter model vector respectively, and The vectors A and B are The weight of the dimension, It is The weight of a material parameter reflects the relative importance of the material parameter in material selection. n is the dimension of the vector, that is, the total number of material parameters considered.

[0132] The above improves the calculation of cosine similarity by introducing weights. When calculating the dot product and the sum of the squares of vector A and vector B, the weight of each parameter is introduced. , to reflect the influence of different parameters on the final matching results, so that the special preferences of design requirements can be more accurately reflected.

[0133] The improved cosine similarity calculation process is as follows:

[0134] 1. Preparation of vectors and weights: Represent the properties and demand parameter model of each material as vectors A and B, and determine the weight vector W for each parameter. Each element in the weight vector Corresponding to the parameter the relative importance of .

[0135] 2. Data preprocessing: Standardized vectors A and B: Standardize the material property vector and the demand parameter model vector, for example, using Z-score standardization to eliminate the impact of different parameter magnitudes.

[0136] 3. Calculate the weighted dot product: multiply the weight of each parameter by the corresponding attribute value, and then calculate the weighted dot product: .

[0137] 4. Calculate the norm of the weighted vector: Calculate the Euclidean norm (i.e. the length of the vector) of the weighted vectors A and B respectively: and .

[0138] 5. Calculate improved cosine similarity: Calculate improved cosine similarity using the values ​​of weighted dot product and weighted vector norm.

[0139] S4 specifically includes:

[0140] S41: Select the material that meets the requirements selected by the intelligent matching algorithm, and extract the property data of the selected material, including physical properties and chemical properties;

[0141] S42: Build a virtual model of a textile design, including the design’s structure, dimensions, and intended use information;

[0142] S43: Using ANSYS simulation software, apply the property data of the selected materials to the virtual model of textile design and set the simulation parameters, including environmental conditions and loading conditions;

[0143] S44: Run simulations to simulate the performance of materials during the textile process, including the response of materials under stress, wear, and environmental influences, and record simulation results, including deformation, strength loss, and color change;

[0144] S45: Analyze simulation results, evaluate the performance of materials in the actual textile process, and determine whether the materials meet the design requirements and intended use;

[0145] S46: If the simulation results show that the material performance is poor or cannot meet the requirements, return to S3 based on the evaluation results and re-match the material for simulation evaluation.

[0146] S5 specifically includes:

[0147] S51: Build a machine learning-based model to analyze textile market trends, user preferences, and performance data of new materials to learn and identify patterns and trends from market data;

[0148] S52: Regularly collect and organize new information on textile materials from market reports, scientific research articles, patents, and online databases, including emerging materials, updated data on material properties, and new trends in textile design;

[0149] S53: Use the established machine learning model to automatically analyze the collected new information, evaluate the importance and reliability of the new information, and determine the new information to be included in the material database;

[0150] S54: Automatically update the material database based on the analysis results of the machine learning model, including adding data of new materials, updating performance parameters of existing materials, and deleting data of outdated or no longer used materials.

[0151] The machine learning model uses a random forest model, which includes:

[0152] Data preparation: Collect and organize the attribute data of textile materials, including new materials and updated information of existing materials, as well as performance data in actual applications, and label each data point (material) with a classification label, such as "suitable for new trends", "better than average performance", etc., or a continuous value label for regression analysis, such as durability score, cost-effectiveness ratio, etc.;

[0153] Model training: Use the random forest algorithm to build multiple decision trees on the training set. During the construction of each tree, a feature subset is randomly selected and the best split point is searched on the feature subset. The decision tree is built without pruning. Each tree grows randomly until it cannot be split. The training formula for each tree in the random forest is: ,in, It is trees, D is the training data set, is a randomly selected subset of all features;

[0154] Prediction and evaluation:

[0155] For classification problems, the prediction result of random forest is the voting result of the majority of trees. For regression problems, the prediction result is the average of the prediction values ​​of all trees. Cross-validation or a reserved test set is used to evaluate the accuracy and generalization ability of the model.

[0156] Applied to material database optimization: Apply the trained random forest model to the collected new material data to predict whether the new material is suitable for the current market trend or design requirements, and update the material database based on the model's prediction results, including adding new materials, updating the ratings or classifications of existing materials, and removing outdated materials or materials that no longer meet the requirements.

[0157] like Figure 2 As shown, an efficient material retrieval system for textile production is used to implement the above-mentioned efficient material retrieval method for textile production, including the following modules:

[0158] Data collection and database establishment module: automatically collects performance data of textile materials, including physical properties, chemical properties and environmental adaptability, through big data technology, and establishes and maintains a material database;

[0159] Demand parameter model building module: converts textile design requirements into quantifiable material demand parameters and forms a material demand parameter model to guide the subsequent material retrieval process;

[0160] Intelligent matching algorithm module: Based on the intelligent matching algorithm of material properties, the optimal material is retrieved and matched from the material database according to the established material requirement parameter model;

[0161] Virtual simulation evaluation module: Use virtual simulation technology to simulate the performance of the matched materials in the actual textile process, evaluate the applicability and performance of the materials through simulation results, and verify the selection of materials;

[0162] Database optimization and update module: Introduces a machine learning-based optimization mechanism to continuously update and optimize the material database based on market trends, user feedback, and new material information.

[0163] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0164] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An efficient material retrieval method for textile production, characterized in that: The following steps are involved: S1: Use big data analysis to collect textile material performance data and establish a material database; S2: Establish material demand parameter model based on textile design requirements; specifically include: S21: Analyze the design requirements of textiles, including durability, elasticity, color fastness, chemical resistance, and environmental protection; S22: According to the parameter quantification standard of each design requirement, the design requirements are converted into specific material parameters; S23: Establish a parameter weight system. According to the importance of the design requirements, assign weights to each material parameter through the hierarchical analysis method to ensure that the sum of all material parameter weights is 1. Perform weight standardization. The formula is: ,in, is the material parameter The standardized weights of is the original weight, is the total number of material parameters; The analytic hierarchy process specifically includes: S231, establish a hierarchical model: Target layer: the final goal, selecting the best textile material; Criteria layer: standards or criteria for evaluating materials, i.e. material parameters converted into textile design requirements; Solution level: material options; S232, construct a judgment matrix: At the criterion level, for each pair of material parameters, compare the relative importance of achieving the target level, give a quantitative ratio, and form a judgment matrix A, where the elements in the judgment matrix are Indicates material parameters Relative to material parameters The importance of S233, calculate weight: use the eigenvalue and eigenvector of the judgment matrix to calculate the weight. The eigenvector corresponding to the maximum eigenvalue is the relative weight of each material parameter. The calculation step includes solving the maximum eigenvalue of the judgment matrix A. and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector ; S234, consistency test: calculate the consistency index CI and consistency ratio CR, where: , where n is the number of criterion layers; ,in, is the average random consistency index, which depends on the number of criterion levels. If , then the consistency of the judgment matrix is ​​acceptable and the weight result is reliable; S24: Use the weighted summation method to build a material requirement parameter model and calculate the comprehensive score of each material to reflect the degree to which the material meets the design requirements. The comprehensive score represents: ,in, It is the comprehensive score of the material. It is The standardized value of a material parameter, It is The weight of each material parameter; S3: Based on the intelligent matching algorithm of material properties and the material requirement parameter model established in S2, intelligently retrieve the material data in the matching material database; S4: Use virtual simulation technology to simulate the performance of the materials matched in S3 in the actual textile process and evaluate the applicability of the materials; S5: Introduce a machine learning-based optimization mechanism to continuously update and optimize the material database to adapt to changes in the textile market and the emergence of new materials; specifically: S51: Build a machine learning-based model to analyze textile market trends, user preferences, and performance data of new materials to learn and identify patterns and trends from market data; S52: Regularly collect and organize new information on textile materials from market reports, scientific research articles, patents, and online databases, including emerging materials, updated data on material properties, and new trends in textile design; S53: Use the established machine learning model to automatically analyze the collected new information, evaluate the importance and reliability of the new information, and determine the new information to be included in the material database; S54: Automatically update the material database based on the analysis results of the machine learning model, including adding data of new materials, updating performance parameters of existing materials, and deleting data of outdated or no longer used materials.

2. The efficient material retrieval method for textile production according to claim 1, characterized in that: The S1 specifically includes: S11: Set data collection standards, including physical properties, chemical properties and color attributes of textile materials; S12: Automatically collect performance data of textile materials from public scientific research literature, patents, product manuals and official websites of textile material manufacturers; S13: Use natural language processing technology to process and analyze the collected unstructured data, extract information that matches the set data collection standards, and convert it into structured data; S14: Data cleaning techniques are used for the collected data, including removing duplicate data, correcting erroneous data, and filling missing values; S15: Store and manage the cleaned structured data.

3. The efficient material retrieval method for textile production according to claim 1, characterized in that: The S3 specifically includes: S31: Taking each material parameter and corresponding weight in the material demand parameter model as input conditions, constructing query conditions of the intelligent matching algorithm; S32: Using an intelligent matching algorithm, searching in a material database, and scoring according to the degree of matching between the material data and the input conditions; S33: The intelligent matching algorithm uses cosine similarity to calculate the similarity between each material data in the material database and the material demand parameter model; S34: ranking the materials in the database according to the similarity scores, and selecting the material with the highest score as the matching result; S35: Output the matching results, including a list of materials meeting the textile design requirements and related performance data.

4. The efficient material retrieval method for textile production according to claim 3, characterized in that: The cosine similarity is calculated as: , where A and B represent the material property vector and the material demand parameter model vector respectively, and The vectors A and B are The weight of the dimension, It is The weight of a material parameter reflects the relative importance of the material parameter in material selection, and n is the dimension of the vector, that is, the total number of material parameters considered.

5. The efficient material retrieval method for textile production according to claim 1, characterized in that: The S4 specifically includes: S41: Select the material that meets the requirements selected by the intelligent matching algorithm, and extract the property data of the selected material, including physical properties and chemical properties; S42: Build a virtual model of a textile design, including the design’s structure, dimensions, and intended use information; S43: Using ANSYS simulation software, apply the property data of the selected materials to the virtual model of textile design and set the simulation parameters, including environmental conditions and loading conditions; S44: Run simulations to simulate the performance of materials during the textile process, including the response of materials under stress, wear, and environmental influences, and record simulation results, including deformation, strength loss, and color change; S45: Analyze simulation results, evaluate the performance of materials in the actual textile process, and determine whether the materials meet the design requirements and intended use; S46: If the simulation results show that the material performance is poor or cannot meet the requirements, return to S3 based on the evaluation results and re-match the material for simulation evaluation.

6. The efficient material retrieval method for textile production according to claim 1, characterized in that: The machine learning model adopts a random forest model, which specifically includes: Data preparation: Collect and organize the property data of textile materials, including new materials and updated information of existing materials, as well as performance data in actual applications, and label each data point with a classification label; Model training: Use the random forest algorithm to build multiple decision trees on the training set. During the construction of each tree, a feature subset is randomly selected and the best split point is searched on the feature subset. The decision tree is built without pruning. Each tree grows randomly until it cannot be split. The training formula for each tree in the random forest is: ,in, It is A tree, is the training dataset, is a randomly selected subset of all features; Prediction and evaluation: For classification problems, the prediction result of random forest is the voting result of the majority of trees. For regression problems, the prediction result is the average of the prediction values ​​of all trees. Cross-validation or a reserved test set is used to evaluate the accuracy and generalization ability of the model. Applied to material database optimization: Apply the trained random forest model to the collected new material data to predict whether the new material is suitable for the current market trend or design requirements, and update the material database based on the model's prediction results, including adding new materials, updating the ratings or classifications of existing materials, and removing outdated materials or materials that no longer meet the requirements.

7. An efficient material retrieval system for textile production, used to implement an efficient material retrieval method for textile production as claimed in any one of claims 1 to 6, characterized in that: Includes the following modules: Data collection and database establishment module: automatically collects performance data of textile materials, including physical properties, chemical properties and environmental adaptability, through big data technology, and establishes and maintains a material database; Demand parameter model building module: converts textile design requirements into quantifiable material demand parameters and forms a material demand parameter model to guide the subsequent material retrieval process; Intelligent matching algorithm module: Based on the intelligent matching algorithm of material properties, the optimal material is retrieved and matched from the material database according to the established material requirement parameter model; Virtual simulation evaluation module: Use virtual simulation technology to simulate the performance of the matched materials in the actual textile process, evaluate the applicability and performance of the materials through simulation results, and verify the selection of materials; Database optimization and update module: Introduces a machine learning-based optimization mechanism to continuously update and optimize the material database based on market trends, user feedback, and new material information.