A large model assisted knitted production line dynamic configuration and rapid response method

Abstract

The application provides a large model assisted knitted production line dynamic configuration and rapid response method, comprising: obtaining knitted production line historical data according to dye molecular weight, dye polarity group number and dye molecular space structure as dye property parameters, and constructing a dye property data warehouse; selecting dye molecular structure parameters, yarn surface morphology, material characteristics and friction coefficient test data according to the knitted production line historical data, and constructing a correlation prediction model of dye properties and yarn friction coefficient; according to the distribution characteristics of the dye properties and the friction coefficient, the dye formula composition is clustered and grouped, the dye combination similar to the influence on the yarn friction coefficient is determined, and a dye formula and friction coefficient mapping library is formed; obtaining knitted mechanical structure parameters, needle movement trajectory equation, fabric structure matrix and fabric density parameters, combining the dye formula and friction coefficient mapping library, and constructing a weaving process digital twin model.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a large-model-assisted dynamic configuration and rapid response method for a knitting production line. Background Art

[0002] In knitting production lines, dyes of different colors can have subtle but critical effects on the yarn's coefficient of friction. Slight changes in the yarn's coefficient of friction can lead to differences in yarn tension and speed during the weaving process, which in turn affects the uniformity, density, and texture of the fabric. Traditional production methods make it difficult to accurately predict and control the changes in the yarn's coefficient of friction caused by different dyes, often resulting in large differences between the finished product and the design expectations, affecting product quality and production efficiency. The dye formula has a complex effect on the friction behavior between the yarn and loom components. The core of this problem lies in how to accurately capture and characterize the influence of dye factors on the yarn's coefficient of friction and effectively integrate it into the weaving process model. These are key issues that need to be solved urgently. At the same time, how to balance the complexity and computational efficiency of the model while ensuring model accuracy is also an important challenge. Summary of the Invention

[0003] The present invention provides a large-model-assisted dynamic configuration and rapid response method for a knitting production line, which mainly includes:

[0004] Based on the dye molecular weight, the number of dye polar groups, and the spatial structure of the dye molecule as dye property parameters, historical data of the knitting production line is obtained and a dye property data warehouse is constructed;

[0005] Based on historical data from knitting production lines, dye molecular structure parameters, yarn surface morphology, material properties, and friction coefficient test data were selected to construct a correlation prediction model between dye properties and yarn friction coefficient.

[0006] The correlation data between dye properties and yarn friction coefficient was mined, and the correlation between dye molecular structure parameters and yarn surface morphology, material properties and friction coefficient numerical range was analyzed to obtain the distribution characteristics of dye properties and friction coefficient of yarn under different dye property combinations;

[0007] Based on the dye properties and friction coefficient distribution characteristics, the dye formula components are clustered and grouped, and the dye combinations with similar effects on the yarn friction coefficient are determined to form a dye formula and friction coefficient mapping library;

[0008] Acquire knitting machine structural parameters, needle motion trajectory equations, fabric structure matrix, and fabric density parameters, and combine them with the dye formula and friction coefficient mapping library to build a digital twin model of the weaving process;

[0009] The digital twin model was designed to set boundary conditions for dye properties, yarn characteristics, and friction coefficient. Finite element simulation was used to simulate yarn movement and fabric formation during weaving, and the effects of dye properties and yarn friction coefficient on the weaving process were analyzed.

[0010] During the weaving process, real-time data on the yarn friction coefficient of dye properties is collected, compared with the simulation results of the digital twin model, and the needle motion trajectory equation and weaving parameters are dynamically adjusted to achieve real-time optimization control of the weaving process;

[0011] When a new dye formula is used on a knitting production line, the yarn friction coefficient is predicted based on the dye formula and the friction coefficient mapping library. The weaving process is simulated and optimized using a digital twin model, providing weaving parameter setting ranges and weaving process optimization suggestions.

[0012] Continuously collect data on dye properties, yarn friction coefficient and weaving process parameters, iteratively update and adaptively adjust the digital twin model and weaving process optimization system, improve the stability of weaving quality and the process adaptability of new dye formulas, and match dye formulas with weaving parameters.

[0013] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0014] The present invention discloses a large-model-assisted dynamic configuration and rapid response method for a knitting production line. This method achieves precise matching of dye properties, yarn characteristics and weaving parameters by constructing a dye property data warehouse and a dye formula and friction coefficient mapping library, combined with a digital twin model of the weaving process. The present invention first analyzes the relationship between the dye molecular structure and the yarn friction coefficient, then uses digital twin technology to simulate the weaving process, and dynamically adjusts the weaving parameters through machine learning. When using a new dye formula, the present invention can predict the yarn friction coefficient and give weaving process optimization suggestions. Through continuous data collection and model iteration, the present invention continuously improves the stability of weaving quality and the process adaptability of new dye formulas, effectively solves the problem of the impact of changes in dye properties on knitting production, and realizes intelligent and refined control of knitting production. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The present invention is a flowchart of a large-model-assisted knitting production line dynamic configuration and rapid response method.

[0016] Figure 2 This is a schematic diagram of a large-model-assisted dynamic configuration and rapid response method for a knitting production line according to the present invention.

[0017] Figure 3 This is another schematic diagram of a large-model-assisted dynamic configuration and rapid response method for a knitting production line according to the present invention. DETAILED DESCRIPTION

[0018] To further understand the content of the present invention, the present invention is described in detail with reference to the accompanying drawings and examples. The present application is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are intended only to illustrate the relevant invention and are not intended to limit the invention. It should also be noted that, for ease of description, only portions relevant to the invention are shown in the accompanying drawings.

[0019] like Figure 1-3 In this embodiment, a large model-assisted knitting production line dynamic configuration and rapid response method may specifically include:

[0020] Step S101 , based on the dye molecular weight, the number of dye polar groups, and the dye molecular spatial structure as dye property parameters, obtain the historical data of the knitting production line and build a dye property data warehouse.

[0021] Dye molecular weight information and the number of polar groups are obtained, wherein the dye molecular weight information is obtained from the PubChem database, and the number of polar groups is calculated using the SMILES string analysis method; based on the dye molecular weight information and the number of polar groups, a three-dimensional spatial structure model of the dye molecule is constructed using the PyMOL molecular modeling tool; production batch information within the past five years is extracted from the production line history database, and the production batch information is detected and processed using the Python Pandas library to obtain a standardized data set; Apache Hive is used to build a data warehouse, and the dye property parameters and the standardized data set are associated and mapped to fact tables and dimension tables, and a multidimensional data cube is constructed using Microsoft Analysis Services; the relationship between dye properties and production parameters is analyzed using a correlation analysis method, and a dye selection prediction model is constructed using the CART decision tree algorithm; if a dye selection recommendation list needs to be generated, the association rules between dye properties and production parameters are extracted from the data warehouse using the Apriori association rule mining algorithm.

[0022] Specifically, dye molecular weight information was obtained from the PubChem database. The number of polar groups in the dyes was calculated and recorded using the SMILES string analysis method. A three-dimensional structural model of the dye molecules was constructed using the PyMOL molecular modeling tool, and the coordinate data was saved. Production batch information from the past five years was extracted from the knitting production line historical database. The data was categorized and organized according to production process parameters. Outlier detection and processing were performed using the Python Pandas library, and duplicates were removed to create a standardized dataset. A data warehouse was constructed using Apache Hive, and dye property parameters were mapped to historical production line data in fact and dimension tables. A multidimensional data cube was constructed using Microsoft Analysis Services, and index optimization was used to improve data retrieval efficiency. Correlation analysis was used to deeply analyze the relationship between dye properties and production parameters. A dye selection prediction model was constructed using the CART decision tree algorithm. A query interface for the dye property data warehouse was established for the production system to access. The Apriori association rule mining algorithm was used to extract association rules between dye properties and production parameters from the data warehouse to generate a recommended list of dye selections. During the data collection phase, dye molecular structure information was queried through the PubChem API to obtain molecular weight data. Taking azo dye as an example, its molecular weight is 327.33 g / mol. The RDKit library was used to parse the SMILES string "N=Nc1ccc(cc1)S(=O)(=O)O" to identify polar groups containing nitrogen, oxygen, and sulfur, and the number of polar groups was calculated to be five. The molecular structure file was imported using PyMOL software to generate a three-dimensional model and extract atomic coordinate data, such as [(1.23, -0.45, 0.78), (2.56, 0.89, -1.34)]. Production records from January 2019 to January 2024 were extracted from a knitting production line database, including parameters such as dye dosage, temperature, and time. Outliers were detected using the interquartilerange method in the Pandas library, for example, data with dye dosage exceeding three standard deviations were marked as outliers. Duplicate records were removed using the drop_duplicates function, resulting in a standardized dataset totaling 50,000 records. A data warehouse was created using Apache Hive, with a star schema designed. The dye batches were used as the fact table, joined to dimension tables such as dye attributes, production parameters, and time. Microsoft Analysis Services was used to construct an OLAP multidimensional data cube for sales, production, and quality, optimizing query performance through bitmap indexing. The Pearson correlation coefficient was calculated to be 0.78, indicating a strong correlation.Using the CART decision tree algorithm, with dye molecular weight, number of polar groups, and spatial structural characteristics as input variables and dyeing effect as the target variable, a prediction model was constructed with an accuracy of 85%. Finally, the Apriori algorithm was applied to mine association rules, setting a minimum support of 0.1 and a minimum confidence of 0.7. Ten valid rules, such as "high number of polar groups → high color fastness," were discovered from 50,000 records, generating a recommended list of dyes.

[0023] Step S102 , based on historical data of the knitting production line, select dye molecular structure parameters, yarn surface morphology, material properties and friction coefficient test data to construct a correlation prediction model between dye properties and yarn friction coefficient.

[0024] The molecular structural parameters of the dyes, including molecular weight, number of polar groups, and spatial configuration, were obtained. Microscopic photographs of the yarn surface were analyzed to determine fiber alignment and surface roughness, and yarn material properties and friction coefficient data were collected. The raw data was preprocessed, and any outliers detected were removed. Missing data was supplemented using a spline interpolation algorithm. Data from different sources were standardized and units converted to obtain processed data. A multidimensional data model was constructed, with dye batches and yarn batches serving as fact tables and dye molecular structural parameters, yarn surface morphology, material properties, and friction coefficient data serving as dimension tables. Time series analysis was performed on historical data to analyze the dynamic relationship between dye properties and yarn friction coefficient. Dimensionality reduction was performed, eigenvalues ​​were calculated, and principal components whose cumulative contribution reached a preset threshold were selected to obtain the reduced data. A correlation prediction model between dye properties and yarn friction coefficient was constructed, and the number of decision trees was set. Feature selection was performed using the Gini index, and model accuracy was evaluated through cross-validation to obtain optimization recommendations for dye selection and yarn friction coefficient.

[0025] Specifically, dye molecular structural parameters, including molecular weight, number of polar groups, and spatial configuration, were obtained from a molecular structure database. ImageJ software was used to analyze microscopic photographs of the yarn surface, extracting surface morphological characteristics such as fiber alignment and surface roughness. Yarn material property data was collected using a material testing instrument, and the yarn friction coefficient was measured using a friction coefficient tester. The collected raw data was preprocessed, outliers were marked and removed, and missing data was supplemented using a spline interpolation algorithm. Data from different sources were formatted and converted to the same unit. Finally, the processed data was stored in a MySQL database. A multidimensional data model was constructed using a star schema, with dye batches and yarn batches as fact tables and dye molecular structural parameters, yarn surface morphology, material properties, and friction coefficient data as dimension tables. These data were mapped into a multidimensional data cube, and time series analysis of historical data was performed to capture the dynamic changes in the relationship between dye properties and yarn friction coefficient. Principal component analysis (PCA) was used to reduce the dimensionality of multidimensional data. Eigenvalues ​​were calculated and principal components with a cumulative contribution rate of 85% were selected. A random forest algorithm was then used to construct a prediction model linking dye properties with yarn friction coefficient. 500 decision trees were set up, and the Gini index was used for feature selection. Ten-fold cross-validation was used to evaluate model accuracy, resulting in recommendations for dye selection and yarn friction coefficient optimization. Dye molecular structural parameters were obtained from the ChemSpider database, including the molecular weight of azo dyes (327.33 g / mol, five polar groups, and a planar structure). Microscopic images of the yarn surface were analyzed using ImageJ software, with a threshold of 128, and the extracted fiber alignment was 0.85 and surface roughness of 15 μm. Yarn material properties were collected using a material testing instrument, revealing a fiber length of 38 mm and a fiber fineness of 1.5 denier. A friction coefficient tester measured the yarn friction coefficient as 0.32. The raw data was preprocessed, with outliers marked and removed using the triple standard deviation method. Missing data was imputed using the cubic spline interpolation algorithm. All length units were normalized to millimeters, and all mass units were normalized to grams. The processed data was stored in a MySQL database, and four tables were created: dye batch, yarn batch, dye properties, and yarn properties. A multidimensional data model was constructed using a star schema, with dye batch and yarn batch as fact tables, and four dimension tables linked: dye molecular structure, yarn surface morphology, material properties, and friction coefficient. Time series analysis was performed on five years of historical data, and exponential smoothing was used to predict trends in the relationship between dye properties and yarn friction coefficient. In principal component analysis, eigenvalues ​​were calculated, and the top three principal components with a cumulative contribution of 85% were selected. A random forest algorithm was used to set up 500 decision trees, and feature selection was performed using the Gini index, with three features randomly selected at each node. The model was evaluated using ten-fold cross-validation, achieving an average accuracy of 87.5%.The final model outputs dye selection recommendations. For the given yarn parameters, dyes with a molecular weight in the range of 300-350 g / mol and 4-6 polar groups are recommended, and the friction coefficient is predicted to be reduced to 0.28.

[0026] Step S103, mining the correlation data between dye properties and yarn friction coefficient, analyzing the correlation between dye molecular structure parameters and yarn surface morphology, material properties and friction coefficient numerical range, and obtaining the dye properties and friction coefficient distribution characteristics of the yarn under different dye property combinations.

[0027] The invention obtains the dye molecular structure parameters, yarn surface morphology, material properties and friction coefficient data from the data warehouse; uses the Apriori algorithm to mine association rules based on the dye molecular structure parameters, yarn surface morphology, material properties and friction coefficient data, and obtains a set of strong association rules between the dye molecular structure parameters and the yarn surface morphology, material properties and friction coefficient; uses the K-means clustering algorithm to classify the dye property combinations according to the strong association rule set, and determines the statistical distribution parameters of the yarn friction coefficient under each category; if the statistical distribution parameters meet the preset conditions, uses the multivariate linear regression method to establish a mathematical model between the dye properties and the yarn friction coefficient; obtains the friction coefficient distribution characteristics of the yarn under different dye property combinations through the mathematical model; uses the box plot and kernel density estimation method to generate visualization results based on the friction coefficient distribution characteristics; judges whether the visualization results meet the preset threshold, and if so, outputs a report on the dye property and friction coefficient distribution characteristics.

[0028] Specifically, data on dye molecular structure parameters, yarn surface morphology, material properties, and friction coefficient were extracted from a data warehouse. Outliers were processed using the triple standard deviation method, and data with different dimensions were converted to a unified scale using the Z-score normalization method. Parameters such as dye molecular weight and the number of polar groups were discretized into 10 intervals with equal width. Finally, the processed dataset was divided into training and test sets with a 7:3 ratio. Association rules were mined using the Apriori algorithm, with a minimum support of 0.05 and a minimum confidence level of 0.6. Association rules between dye molecular structure parameters and yarn surface morphology, material properties, and friction coefficient were mined, resulting in a set of strong association rules. Based on the mined association rules, the K-means clustering algorithm was used to classify dye property combinations. The silhouette coefficient method was used to determine the optimal number of clusters, and statistical distribution parameters such as the mean, standard deviation, and quartile of the yarn friction coefficient were calculated for each category. The multivariate linear regression method was applied to establish a mathematical model between dye properties and yarn friction coefficient. 10-fold cross validation was used for model verification and parameter optimization. The mapping relationship between dye property combinations and yarn friction coefficient distribution characteristics was obtained. Box plots and kernel density estimation were used to visualize the friction coefficient distribution characteristics of yarns under different dye property combinations, and a report on dye property and friction coefficient distribution characteristics was output.

[0029] 100,000 records were extracted from a data warehouse, including data on dye molecular weight (range 200-800 g / mol), number of polar groups (0-10), yarn surface roughness (1-20 μm), and friction coefficient (0.1-0.5). Outliers were identified using the 3-times standard deviation method, and 374 records were removed. After Z-score normalization, the data had a mean of 0 and a standard deviation of 1. Parameters such as dye molecular weight were divided into 10 equal-width intervals, such as the molecular weight intervals [200-260] and [260-320]. The training set contained 69,639 records, and the test set contained 29,987. The Apriori algorithm was used to mine association rules, with a minimum support of 0.05 and a minimum confidence of 0.6. 152 strong association rules were obtained, such as {molecular weight: [320-380), polar groups: [4-5)}=>{friction coefficient: [0.25-0.30)}, with a support of 0.068 and a confidence of 0.72. K-means clustering was used to classify the dye property combinations into five categories, and the optimal number of clusters (silhouette coefficient 0.68) was determined using the silhouette coefficient method. The statistical parameters of the friction coefficient for each category were calculated, such as the mean of category 1, 0.28, standard deviation 0.03, Q1 0.26, and Q3 0.30. Multiple linear regression modeling was performed, with the independent variables being the dye molecular weight, number of polar groups, etc., and the dependent variable being the friction coefficient. The average R was obtained through 10-fold cross validation. 2The model shows that for every 100 g / mol increase in molecular weight, the friction coefficient decreases by an average of 0.015, while for every additional polar group, the friction coefficient increases by an average of 0.008. A boxplot illustrates the friction coefficient distributions for the five categories, and a kernel density estimate plot indicates a high degree of overlap between the friction coefficient distributions for categories 1 and 3, suggesting the need for further subdivision. The final report contains 152 association rules, five dye property combination categories, and their corresponding friction coefficient distribution characteristics.

[0030] The correlation between the molecular structure and chromatographic characteristics of the dye and the surface roughness and linear density of the yarn was analyzed to obtain the surface morphological characteristics of the yarn under different dye properties.

[0031] The invention relates to a method for obtaining the molecular structural parameters of the dye, measuring the absorption spectrum and reflectance spectrum of the dye, and obtaining standardized chromatographic characteristic data; measuring the surface roughness and linear density of the yarn, and forming a yarn surface morphological characteristic data set; performing dimensionality reduction processing on the molecular structural parameters of the dye and the chromatographic characteristic data set, and extracting the main characteristic variables; calculating the correlation coefficient between the main characteristic variables and the yarn surface morphological characteristic data set, and screening out significantly correlated feature combinations based on the correlation coefficient; using the significantly correlated feature combinations to establish a prediction model between the dye properties and the yarn surface morphological characteristics; using the prediction model to output the surface roughness and linear density prediction values ​​of the yarn under different dye property combinations; performing cluster analysis on the predicted values ​​to obtain multiple categories; calculating the center point and standard deviation of each category, giving a characteristic description of each category, using a scatter plot to visualize the relationship between the dye properties and the yarn surface morphological characteristics, and using a heat map to display the distribution of the yarn surface morphological characteristics under different dye property combinations.

[0032] Specifically, the dye molecular structure parameters were obtained from a dye database. The dye absorption spectrum was measured using a UV-visible spectrophotometer, and the reflectance spectrum was measured using a spectrophotometer. Data preprocessing software was used to denoise and baseline correct the raw data to obtain standardized chromatographic characteristic data. Yarn surface roughness was measured using a laser roughness meter, and linear density was directly measured using an electronic yarn density meter. The measurement results were transferred to a computer via a data acquisition card to form a yarn surface morphological characteristic dataset. Principal component analysis was used to reduce the dimensionality of the dye molecular structure and chromatographic characteristic data. The cumulative explained variance ratio method was used to determine the number of principal components to retain and extract the main characteristic variables. The correlation between the dye characteristics and the yarn surface roughness and linear density values ​​was calculated using the Pearson correlation coefficient, and significantly correlated feature combinations were screened. Based on the correlation analysis results, a support vector machine regression algorithm was used to establish a prediction model linking dye properties and yarn surface morphological characteristics. A radial basis function kernel was used to optimize the model parameters using a grid search method. Cross-validation was used to evaluate the model performance, and the predicted surface roughness and linear density values ​​of the yarn were output for different dye property combinations. The prediction results were classified into multiple categories using the K-means clustering algorithm. The center point and standard deviation of each category were calculated, and a characteristic description of each category was provided. Scatter plots were used to visualize the relationship between dye properties and yarn surface morphology, and heat maps were used to display the distribution of yarn surface morphology for different dye property combinations. Molecular structural parameters of 100 dyes were extracted from a dye database, with molecular weights ranging from 200 to 800 g / mol, the number of polar groups from 0 to 10, and spatial configurations including planar and stereo structures. Absorption spectra were measured using a UV-visible spectrophotometer with a 2 nm interval in the range of 200 to 800 nm, and reflectance spectra were measured with a 10 nm interval in the range of 400 to 700 nm. Data preprocessing software used wavelet transform for denoising and polynomial fitting for baseline correction. Yarn surface roughness was measured with a laser roughness meter with a resolution of 0.1 μm and a range of 1 to 20 μm. Linear density was measured with an electronic yarn density meter with an accuracy of 0.1 tex and a range of 10 to 50 tex. The principal component analysis retained the principal components that accounted for 95% of the cumulative explained variance, usually 5-8. The Pearson correlation coefficient threshold was set to 0.6, and 30 significantly correlated feature combinations were screened. The support vector machine regression model used the RBF kernel, and the parameters C (penalty coefficient) and γ (kernel function coefficient) were optimized through grid search. The range of C was [0.1, 1, 10, 100], and the range of γ was [0.01, 0.1, 1, 10]. The model performance was evaluated by 5-fold cross validation, and the final model R 2The predicted values ​​reached 0.85. K-means clustering was used to classify the predictions into five categories, and the center point and standard deviation of each category were calculated. For example, category 1 showed a surface roughness of 12.5 ± 1.2 μm and a linear density of 30.5 ± 2.3 tex. A scatter plot shows the relationship between dye molecular weight and yarn surface roughness, while a heat map illustrates the effect of the number of polar groups and spatial configuration on linear density. Color depth indicates linear density.

[0033] Step S104 , clustering and grouping the dye formula compositions according to the dye properties and friction coefficient distribution characteristics, determining dye combinations with similar effects on the yarn friction coefficient, and forming a dye formula and friction coefficient mapping library.

[0034] Dye property data and yarn friction coefficient data were obtained, and the Z-score normalization method was used to convert data of different dimensions into a unified scale. Principal component analysis was performed on the normalized data, and the principal components with a cumulative contribution rate of 95% were determined as input features for cluster analysis. The input features were preprocessed using the DBSCAN algorithm to obtain a data set with outliers removed. The dye formula components were grouped based on the data set using the K-means clustering algorithm, and the optimal number of clusters was determined using the silhouette coefficient method. Once the optimal number of clusters was determined, the dye property parameters and the corresponding friction coefficient mean of each cluster center were calculated. Based on the clustering results, a mapping relationship between the dye formula and the friction coefficient was constructed. The CART decision tree algorithm was used to establish a prediction model from dye properties to friction coefficient, and the prediction model parameters were optimized through grid search. Based on the prediction model, a dye formula and friction coefficient mapping library was generated. The mapping library contains dye combination features and corresponding friction coefficient prediction values. The sliding window method was used to update the dye formula and friction coefficient mapping relationship in the mapping library.

[0035] Specifically, dye property data and corresponding yarn friction coefficient data were extracted from a data warehouse, including parameters such as molecular weight, number of polar groups, and spatial configuration for single and mixed dyes. Outliers were processed using the Python Pandas library, and the Z-score normalization method was used to convert data of different dimensions to a unified scale, resulting in a standardized dye property and friction coefficient dataset. Principal component analysis was used to reduce the dimensionality of the dye property data, extract the main characteristic variables, calculate the contribution rate of each principal component, and select the principal component with a cumulative contribution rate of 95% as the input feature for cluster analysis. The DBSCAN algorithm was used to preprocess the data to remove outliers. The K-means clustering algorithm was then used to group the dye formula components. The silhouette coefficient method was used to determine the optimal number of clusters. The dye property parameters and corresponding friction coefficient mean of each cluster center were calculated to identify dye combinations with similar effects on the yarn friction coefficient. Based on the clustering results, a mapping relationship between dye formulas and friction coefficients was constructed. A CART decision tree algorithm was used to develop a prediction model linking dye properties to friction coefficients. Grid search was used to optimize parameters, including tree depth and the minimum number of samples per leaf node. Model performance was evaluated using 10-fold cross-validation. A library of dye formula-friction coefficient mappings was generated, containing dye combination characteristics and corresponding friction coefficient predictions. This mapping relationship was updated monthly using a sliding window approach with a window size of six months. A heat map was used to visualize the relationship between dye combinations and friction coefficients, with the horizontal axis representing different dye combinations and the vertical axis representing the friction coefficient range. Color intensity indicated frequency of occurrence. 10,000 records of dye properties and friction coefficients were extracted from a data warehouse. These included single dyes and two-component mixed dyes with molecular weights ranging from 200 to 800 g / mol, the number of polar groups from 0 to 10, and spatial configurations classified as planar or cubic. Outliers were detected using the interquartilerange method in the Pandas library, and 374 records were removed. After Z-score normalization, the data had a mean of 0 and a standard deviation of 1. The results of principal component analysis showed that the cumulative contribution rate of the first five principal components reached 96.3%, which was selected as clustering input. The DBSCAN algorithm parameters were set to ε = 0.5, MinPts = 5, and 86 outliers were identified and removed. K-means clustering tried 2-10 clusters, and the optimal number of clusters was determined to be 6 by the silhouette coefficient, resulting in 6 types of dye combinations with similar effects on the friction coefficient. The CART decision tree parameter grid search range was tree depth [3, 6, 9], minimum leaf node sample number [5, 10, 20], and 10-fold cross validation average R 2The generated mapping library contains 6,000 dye formulas corresponding to the friction coefficient. The sliding window is updated monthly, processing approximately 3,000 new data points in the past six months. The heat map visualization results show that the horizontal axis has six types of dye combinations, and the vertical axis has a friction coefficient range of 0.1-0.5. The colors range from light blue to dark red, indicating the frequency from low to high. Among them, the third type of dye combination, with a molecular weight of 400-600 g / mol, 6-8 polar groups, and a stereo configuration, is highly correlated with a low friction coefficient of 0.15-0.25, and appears dark red.

[0036] Step S105: Acquire the knitting machine structure parameters, knitting needle motion trajectory equation, fabric structure matrix and fabric density parameters, combine them with the dye formula and friction coefficient mapping library, and build a digital twin model of the weaving process.

[0037] Mechanical structural parameters, including needle cylinder diameter, needle count, and coil length, were obtained from a knitting machinery database. High-precision position sensors were used to capture needle motion information, and the needle trajectory equation was fitted using cubic spline interpolation. Based on the fabric design drawings, a convolutional neural network was used for image segmentation and feature extraction to obtain a fabric weave matrix. Density parameters were obtained using a density meter and stored in a database. A random forest regression method was used to establish a correlation model between dye formula, friction coefficient, and weaving parameters. Data from a dye formula and friction coefficient mapping library was then matched with the weaving parameters to generate a comprehensive dataset. A long-short-term memory network was used to construct a digital twin model of the weaving process. The model took the mechanical parameters, trajectory equation, weave matrix, density parameters, and dye formula as inputs, and the friction coefficient as output. The parameters of the digital twin model were optimized using a backpropagation algorithm. 3D rendering technology was used to visualize the operation of the digital twin model, displaying the needle motion and fabric formation process in real time. The digital twin model was optimized using a particle swarm optimization algorithm to improve prediction accuracy.

[0038] Specifically, mechanical structural parameters, including needle cylinder diameter, needle count, and coil length, were obtained from a knitting machinery database. High-precision position sensors were used to capture needle motion information. The needle trajectory equation was fitted using cubic spline interpolation, and the mechanical parameters and trajectory equation were stored in the database. Based on the fabric design drawings, a convolutional neural network was used for image segmentation and feature extraction to construct a fabric weave matrix. Fabric density parameters were obtained using a density meter and stored in the database. A random forest regression method was used to establish a correlation model between dye formula, friction coefficient, and weaving parameters. Data from the dye formula and friction coefficient mapping library was matched with the weaving parameters to generate a comprehensive dataset. A digital twin model of the weaving process was constructed using a long-short-term memory network. The model parameters were optimized using a back-propagation algorithm, achieving a digital representation of the weaving process. Actual weaving data was used for model validation, and the model was optimized using a particle swarm optimization algorithm to improve prediction accuracy. 3D rendering technology is used to visualize the operation process of the digital twin model, showing the knitting needle movement and fabric formation process in real time.

[0039] Mechanical structure parameters were extracted from the knitting machinery database. The needle cylinder diameter was 18 inches, the number of needles was 2400, and the coil length was 2.5 mm. A high-precision position sensor collected needle motion position information at a frequency of 1 kHz, obtaining 10,000 data points. The cubic spline interpolation method was used to fit the needle motion trajectory equation, with a fitting error of less than 0.01 mm. The convolutional neural network used the ResNet-50 architecture to segment and extract features from the 1024x1024 pixel fabric design image, with an accuracy rate of 98%. The density meter measured the fabric density to be 320 g / m 2 . The random forest regression model uses 100 decision trees, and the feature importance analysis shows that the molecular weight of the dye has the greatest impact on the friction coefficient, with a contribution rate of 35%. The long short-term memory network contains 3 hidden layers, 128 neurons in each layer, and an input sequence length of 50. The parameters are adjusted by the Adam optimizer with a learning rate of 0.001. The model was verified on 10,000 actual weaving data with an average absolute error of 0.015. The particle swarm optimization algorithm uses 50 particles and iterates 100 times, reducing the model prediction error by 12%. The three-dimensional rendering technology is implemented using OpenGL to visualize the digital twin model at a speed of 60 frames per second, showing the real-time movement of 2,400 knitting needles and the fabric formation process, with a texture accuracy of 0.1mm. The entire digital twin model runs on a workstation equipped with an RTX 3080 graphics card, and the simulation time for a single weaving process is 0.5 seconds.

[0040] Step S106: Set the boundary conditions of dye properties, yarn characteristics, and friction coefficient in the digital twin model, use the finite element simulation method to simulate the yarn movement and fabric formation process during weaving, and analyze the influence of dye properties and spinning friction coefficient during weaving on the weaving process.

[0041] Dye properties, yarn characteristics, and friction coefficient data are obtained, and boundary condition ranges are set. These ranges include the upper and lower limits of dye molecular weight, the upper and lower limits of polar group number, the upper and lower limits of yarn linear density, the upper and lower limits of yarn strength, the upper and lower limits of yarn elongation, and the upper and lower limits of friction coefficient. ANSYS finite element analysis software is used to construct yarn and fabric geometric models. Once the yarn geometric model is complete, it is discretized into beam elements, with each yarn divided into a preset number of elements. Yarn material properties and contact conditions are determined based on the yarn geometric model, and the needle motion trajectory and weaving tension are set as external loads. The weaving process is simulated using an explicit dynamics solver, which determines the time step based on the Courant-Friedrichs-Lewy condition. An orthogonal experimental design approach was used to design an experimental plan, including the number of factors and levels. Simulation experiments were conducted according to the experimental plan, and the results were tested for significance using analysis of variance. ParaView scientific visualization software was used to visualize yarn motion and stress distribution, and a weighted model of influencing factors was established using grey correlation analysis. This weighted model quantitatively describes the degree of influence of each factor on the weaving process.

[0042] Specifically, dye properties, yarn characteristics, and friction coefficient data were extracted from a database. Boundary condition ranges were set, including dye molecular weight, number of polar groups, yarn linear density, strength, elongation, and upper and lower limits of the friction coefficient. These parameters were then imported into the digital twin model as initial conditions. ANSYS finite element analysis software was used to construct geometric models of the yarn and fabric. The yarns were discretized into beam elements, with each yarn divided into 500 elements. Yarn material properties and contact conditions were defined, and the knitting needle motion trajectory and weaving tension were set as external loads to establish a complete finite element simulation model. An explicit dynamics solver was used to simulate the weaving process. Based on the Courant-Friedrichs-Lewy condition, a time step of 1e-6 seconds was established. The motion and deformation of the yarn under the action of the knitting needle were simulated. The yarn position, stress, and strain were calculated at each time step, and the changes in key parameters during the fabric formation process were recorded. Using orthogonal experimental design, a three-factor, five-level experimental scheme was established, and 125 simulation experiments were conducted. The simulation results were tested for significance using analysis of variance, and a quadratic polynomial response surface equation was established to determine the influence of dye properties and friction coefficient on the weaving process. ParaView scientific visualization software was used to visualize yarn motion and stress distribution, and a weighted model of influencing factors was established using grey correlation analysis to quantitatively describe the degree of influence of each factor on the weaving process. The simulation accuracy was verified by comparison with actual weaving experimental results, with an error within 5%. Data on the properties of 100 dyes were extracted from a database, with a molecular weight range of 200-800 g / mol and the number of polar groups of 1-10. Yarn properties included a linear density of 10-50 tex, a strength of 3-8 cN / dtex, and an elongation of 5-20%. The friction coefficient ranged from 0.1-0.5. A circular knitting machine model with a diameter of 18 inches and a needle count of 2400 was constructed in ANSYS. The yarn used BEAM188 beam elements, each yarn was divided into 500 elements, and the element length was 0.5 mm. The material properties were set to linear elasticity, Young's modulus 1000 MPa, and Poisson's ratio 0.3. The knitting needle motion trajectory was set to a sine curve with a frequency of 600 rpm and an amplitude of 5 mm. The explicit dynamics solver time step was 1e-6 seconds, the total simulation time was 0.1 seconds, and 10,000 frames of data were output. The orthogonal experimental design had three factors: molecular weight, number of polar groups, and friction coefficient, with 5 levels and a total of 125 groups of experiments. Variance analysis showed that molecular weight and friction coefficient had significant effects on the weaving process (P<0.01). The fit of the quadratic polynomial response surface equation R 2 = 0.95. ParaView visualization shows that the maximum yarn stress occurs at the point of contact with the needle, with a peak stress of 80 MPa. Grey correlation analysis determined the influence weights: friction coefficient 0.42, molecular weight 0.35, and number of polar groups 0.23. Compared to actual weaving experiments, the simulated predictions showed an error of 4.2% in fabric density and 3.8% in loop length, both within acceptable ranges.

[0043] The molecular size, polarity, adsorption capacity parameters of the dye and the friction coefficient of the yarn were set as boundary conditions. Finite element simulation was used to simulate the dynamic changes of yarn tension, speed and needle motion trajectory during the weaving process, and the influence of dye properties and friction coefficient on the stability of the weaving process was analyzed.

[0044] Dye molecular size, polarity, and adsorption capacity data were obtained from a pre-established dye database. The yarn friction coefficient was measured using the inclined plane method under preset temperature and humidity conditions. Based on the acquired dye data and friction coefficient, a numerical model of the weaving process was constructed. The yarn was discretized into predefined beam elements, and the equation for the needle motion trajectory was established. An explicit dynamics solver was used to simulate the weaving process. The yarn tension, velocity, and displacement were calculated at each time step, and the real-time coordinates of the needle motion trajectory were recorded. An orthogonal experimental design approach was used to establish a multi-factor, multi-level experimental scheme. Multiple simulations were conducted to calculate the fluctuation coefficients of yarn tension, velocity, and needle motion trajectory. If the fluctuation coefficient was less than a preset threshold, the weaving process was considered stable. Otherwise, a sensitivity analysis was performed to determine the influence of each parameter on stability.

[0045] Specifically, dye molecular size, polarity, and adsorption capacity data were obtained from a dye database. The yarn friction coefficient was measured using the tilted plane method. Test conditions were a temperature of 20°C, a relative humidity of 65%, and an inclination angle of 0-90°. Outliers were corrected using the triple standard deviation method, and the processed data were input into ANSYS software as boundary conditions for finite element simulation. A numerical model of the weaving process was constructed using ANSYS. The yarn was discretized into BEAM188 beam elements, the needle trajectory equation was set as a sine function, the friction contact condition between the yarn and the needle was defined, and a Maxwell-Wiechert viscoelastic yarn constitutive model was established that accounted for dye adsorption capacity. The weaving process was simulated using the central difference method in an explicit dynamics solver. Based on the Courant-Friedrichs-Lewy condition, a time step of 1e-6 seconds was set, and the total simulation time was 1 second. The yarn tension, velocity, and displacement were calculated at each time step, and the real-time coordinates of the needle trajectory were recorded. Using an orthogonal experimental design approach, a three-factor, five-level experimental scheme was established, and 125 simulations were conducted. The simulation data were processed using an autoregressive moving average model. The fluctuation coefficients of yarn tension, speed, and needle trajectory were calculated as quantitative indicators of weaving process stability. Multiple linear regression analysis was used to establish a relationship between dye properties, friction coefficient, and weaving process stability. A sensitivity analysis was conducted to determine the influence of each parameter on stability. This was then compared with actual weaving experimental data, with an error margin of less than 5%. Data on molecular size (ranging from 1-5 nm), polarity (dipole moment 0.5-4 D), and adsorption capacity (Gibbs free energy -10 to -50 kJ / mol) for 100 dyes were extracted from a dye database. The yarn friction coefficient was measured using the tilted plane method. The tilt angle was increased by 1° per second from 0° to 90°. The angle at which the yarn started to slide was recorded. Ten iterations were repeated and the average value was taken, resulting in a friction coefficient range of 0.1-0.5. A circular knitting machine model with an 18-inch diameter and 2400 needles was constructed in ANSYS. The yarn was represented by BEAM188 units, and each yarn was divided into 1000 units. The knitting needle motion trajectory was set to Asin(ωt), A=5mm, and ω=2π600rpm. The Maxwell-Wiechert model contains three Maxwell units with elastic moduli E1=1000MPa, E2=800MPa, and E3=600MPa, and viscosity coefficients η1=10Pa·s, η2=20Pa·s, and η3=30Pa·s. The dye adsorption capacity was introduced as a correction factor for the viscosity coefficient. The central difference method was used to solve the problem, with a time step of 1e-6 seconds and a total of 1,000,000 steps. The orthogonal experimental design L25(56) included six factors: dye molecular size, polarity, adsorption capacity, friction coefficient, yarn linear density, and weaving speed. The ARMA model order (2,1) is used to process time series data and calculate the tension fluctuation coefficient σ / μ, which ranges from 2% to 10%.Multiple linear regression in R. 2 =0.92. Friction coefficient and dye adsorption capacity had the greatest impact on stability, with standardized regression coefficients of 0.45 and 0.38, respectively. Sensitivity analysis showed that for every 0.1 increase in friction coefficient, tension fluctuation increased by 1.2%. The experimental verification error was 4.3%, which was within an acceptable range.

[0046] Step S107: In the weaving process, the dye property yarn friction coefficient data is collected in real time, and the simulation results of the digital twin model are compared to dynamically adjust the needle motion trajectory equation and weaving parameters to achieve real-time optimization control of the weaving process.

[0047] The invention obtains yarn friction coefficient data collected by an optical fiber sensor installed on a weaving machine; performs real-time simulation using a physics-based finite element digital twin model based on the yarn friction coefficient data and dye property data to obtain theoretical weaving parameters and needle motion trajectories; uses a root mean square error method to compare the difference between the yarn friction coefficient data and the theoretical weaving parameters, and generates a deviation report including the error size and trend; constructs a weaving parameter optimization model based on the deviation report using a deep Q learning network, and the weaving parameter optimization model is used to dynamically adjust the needle motion trajectory equation and weaving parameters; if the output result of the weaving parameter optimization model meets a preset performance indicator threshold, the optimization effect of the weaving parameter optimization model is continuously evaluated using a sliding time window method; and dynamically adjusts the reward function of the deep Q learning network based on the optimization effect, and the reward function is used to achieve continuous optimization and closed-loop control of the weaving process.

[0048] Specifically, fiber optic sensors are installed on the weaving machine to collect yarn friction coefficient data in real time, collecting data 100 times per second. Dye property data, including dye molecular size, polarity, and adsorption capacity, is regularly retrieved from a database. The collected data is filtered and normalized using a Kalman filter. A physics-based finite element digital twin model is used for real-time simulation, updating every 0.1 seconds. The processed dye property and friction coefficient data are input into the model to calculate theoretical weaving parameters and needle motion trajectories, generating simulation results. The root mean square error method is used to compare the differences between the actual collected data and the simulation results, calculate the error value and deviation direction, and generate a deviation report containing the error magnitude and trend. Based on the deviation report, a deep Q-learning network is used to construct a weaving parameter optimization model. This model dynamically adjusts the needle motion trajectory equation and weaving parameters, including needle lifting speed, yarn tension, and yarn feed speed. The optimized parameters are fed back to the weaving machine control system in real time. Key performance indicator thresholds are set, such as a yarn breakage rate of less than 0.1% and a fabric density deviation of less than 1%. The optimization results are continuously evaluated using a sliding time window method with a window size of one hour and updated every 10 minutes. According to the long-term performance, the reward function of the deep Q-learning network is dynamically adjusted to achieve continuous optimization and closed-loop control of the weaving process.

[0049] FBG fiber optic sensors with a sensitivity of 0.1 pm / με were installed on the weaving machine to collect real-time data on the yarn friction coefficient at a sampling rate of 100 Hz. The database was updated hourly with dye properties such as molecular size of 3.5 nm, polarity of 2.8D, and adsorption energy of -35 kJ / mol. The data was filtered using a fifth-order Kalman filter with a measurement noise covariance of R = 0.01 and a process noise covariance of Q = 1e-4. An ANSYS-based finite element digital twin model, consisting of 2400 BEAM188 elements, was updated every 0.1 seconds to calculate the theoretical needle motion trajectory Asin(ωt), where A = 5 mm and ω = 2π600 rpm. The root mean square error (RMSE) was used to compare the actual data with the simulated results, for example, the friction coefficient achieved an RMSE of 0.015, and a deviation report was generated. The deep Q-learning network consists of three fully connected layers, each with 128 neurons. It uses a learning rate of 0.001, a discount factor of γ = 0.99, an ε-greedy strategy of ε = 0.1, and updates the target network every 100 steps. The needle lifting speed, yarn tension, and yarn feed speed are dynamically adjusted by ±5%, ±2 cN, and ±1 m / min, respectively. Key performance indicators are set as a yarn breakage rate of <0.1% and a fabric density deviation of <1%. The sliding window is 1 hour, updated every 10 minutes, and average performance indicators are calculated. Based on long-term performance, the reward function is adjusted every 24 hours, for example, adjusting the fabric density deviation weight from 0.5 to 0.6. This closed-loop optimization process has improved the adjustment accuracy of weaving parameters by 15% and the stability of fabric quality by 8%.

[0050] Step S108: When a new dye formula is used in the knitting production line, the yarn friction coefficient is predicted based on the dye formula and the friction coefficient mapping library, and the weaving process is simulated and optimized through the digital twin model to provide weaving parameter setting ranges and weaving process optimization suggestions.

[0051] The chemical composition and property parameters of new dye formulas from the dye formula database, including molecular weight, number of polar groups, and spatial configuration, are obtained. A weighted average method is used to calculate a comprehensive impact index based on the chemical composition and property parameters. This comprehensive impact index is input into a pre-established random forest regression model, which is used to predict the range of yarn friction coefficients. The predicted friction coefficient range is input into a digital twin model, and Latin hypercube sampling is used to generate multi-tissue fabrication parameter combinations, including needle movement speed, yarn tension, and yarn feed speed. Simulation experiments are run based on these weaving parameter combinations to simulate the weaving process. The simulation results are processed using Python's pandas library to determine the weaving efficiency, fabric quality indicators, and energy consumption levels for each parameter combination. An NSGA-II multi-objective optimization algorithm was used to optimize weaving efficiency, fabric quality, and energy consumption to identify the optimal parameter combination that met production requirements. A weaving parameter setting range report and process optimization suggestions were generated based on the optimal parameter combination. The process optimization suggestions included the value ranges of the amplitude A, angular frequency ω, and phase φ in the needle motion trajectory equation S(t), a tension control curve represented by a piecewise linear function, and a yarn feed speed adjustment strategy.

[0052] Specifically, the chemical composition and property parameters of the new dye formula, including molecular weight, number of polar groups, and spatial configuration, were obtained from the dye formula database. The weighted average method was used to calculate the combined impact, taking into account the proportions of each component. A random forest regression model, based on the dye formula and friction coefficient mapping library, was used to predict the yarn friction coefficient range. Transfer learning was used to adjust the prediction model to adapt to the characteristics of the new formula. The predicted friction coefficient range was input into the digital twin model. Combined with yarn linear density, strength, and elongation, as well as fabric structure and density parameters, Latin hypercube sampling was used to generate multi-tissue fabric parameter combinations, including needle movement speed, yarn tension, and yarn feed speed. Simulation experiments were then run to simulate the weaving process. The Python pandas library was used to process the simulation results, and the weaving efficiency, fabric quality indicators, and energy consumption levels for each parameter combination were calculated. The NSGA-II algorithm was used for multi-objective optimization to select the optimal parameter combination that met production requirements. Based on the optimization results, a weaving parameter setting range report and process optimization recommendations are generated. These include the ranges of amplitude A, angular frequency ω, and phase φ in the needle trajectory equation S(t) = A*sin(ωt+φ), a tension control curve represented by a piecewise linear function, and a yarn feed speed adjustment strategy. Expected fabric performance indicators and production efficiency assessments are also provided. The new dye formula contains a molecular weight of 600 g / mol, 8 polar groups, and a planar spatial configuration, accounting for 70%. A weighted average calculation yields a combined molecular weight of 520 g / mol and 6.5 polar groups. A random forest regression model (100 decision trees, maximum depth 10) predicts a friction coefficient range of 0.25-0.35. Transfer learning, using fine-tuning of the last two layers with a learning rate of 0.001 and 50 iterations, reduces the prediction error from 8% to 3%. The digital twin model inputs yarn characteristics: linear density 30 tex, strength 4 cN / dtex, elongation 12%; and fabric parameters: plain weave with a density of 250 strands / 10 cm. Latin hypercube sampling generated 100 parameter combinations, with a needle speed range of 500-700 rpm, a tension range of 8-12 cN, and a yarn feed speed of 0.8-1.2 m / s. ANSYS simulations were performed with 1000 stitch cycles and a time step of 0.001 s. Pandas processing yielded calculated weaving efficiency of 93-98%, a coefficient of variation in fabric strength of 2.5-4%, and energy consumption of 0.15-0.18 kWh / kg. The NSGA-II algorithm, with a population size of 50 and 100 iterations, selected the 10 optimal parameter combinations. The optimization recommendations are: needle trajectory S(t) = 5 mm * sin(2π * 600t + π / 6), a tension control curve of 8 cN from 0-50% of the cycle, increasing linearly to 10 cN from 50-100% of the cycle, and a yarn feed speed of 1.0 m / s ±5% with dynamic adjustment. The expected fabric strength coefficient of variation is <3%, and production efficiency is >95%.

[0053] Step S109: Continuously collect data on dye properties, yarn friction coefficient, and weaving process parameters, iteratively update and adaptively adjust the digital twin model and weaving process optimization system to improve weaving quality stability and process adaptability of new dye formulas, and match dye formulas with weaving parameters.

[0054] The invention relates to a method for obtaining yarn friction coefficient and weaving process parameter data collected by a high-precision sensor network, wherein the high-precision sensor network is installed on a weaving production line; based on the yarn friction coefficient and weaving process parameter data, an outlier is detected by a 3-times standard deviation method, and filtering is performed by a sliding window averaging method to obtain filtered parameter data; using the filtered parameter data, a digital twin model is updated by an online stochastic gradient descent algorithm, and the model parameters of the digital twin model are dynamically adjusted; based on the digital twin model, the weaving process is adjusted in real time to determine the adjustment values ​​of the needle movement speed, yarn tension and yarn feeding speed; association data between dye formula and weaving parameters is obtained, wherein the association data includes dye properties, friction coefficient and weaving parameters; the association data is analyzed by an Apriori algorithm to obtain the intrinsic relationship between the dye properties, friction coefficient and weaving parameters; based on the intrinsic relationship, a matching rule library is generated, and the matching rule library is used to dynamically adjust the control parameter weights.

[0055] Specifically, a high-precision sensor network is installed on the weaving production line to collect real-time data on yarn friction coefficient and weaving process parameters. Dye property data, including dye molecular structure, polarity, yarn surface friction coefficient, needle motion trajectory, yarn tension, and yarn feed speed, is collected for each batch. Preliminary processing is performed using edge computing devices, with outliers detected using the triple standard deviation method and filtering performed using a sliding window averaging method. An online stochastic gradient descent algorithm is used to update the digital twin model, dynamically adjusting model parameters and optimizing the mapping between dye properties and friction coefficient. A model predictive control algorithm is used to adjust the weaving process in real time. Based on actual production data and quality inspection results, needle motion speed, yarn tension, and yarn feed speed are dynamically optimized to achieve closed-loop control of the weaving process. A database associating dye recipes with weaving parameters is constructed, and the Apriori algorithm is used to analyze the intrinsic relationships between dye properties, friction coefficient, and weaving parameters, generating a matching rule base. The mining results are applied to the model predictive control algorithm to dynamically adjust the control parameter weights. The overall system performance is regularly evaluated, including model prediction accuracy, fabric strength coefficient of variation, and process adaptability scores for new dye formulations. Performance reports are generated to provide a basis for further optimization. One hundred high-precision fiber optic sensors with a sampling rate of 1000 Hz are installed on the weaving production line to collect yarn friction coefficient and weaving parameters in real time. Dye property data is collected approximately every eight hours for each batch, including molecular weight (ranging from 200 to 800 g / mol), polarity (0-5D), and spatial configuration. Edge computing devices use the 3x standard deviation method to detect outliers, with a rejection rate of approximately 0.5%, and filter using a 60-second sliding window average. An online stochastic gradient descent algorithm updates the digital twin model hourly with a learning rate of 0.01 and a mini-batch size of 128. Model prediction accuracy has increased from an initial 85% to 95%. The model predictive control algorithm uses a 15-minute prediction horizon and a 5-second control interval. The optimization objective functions include the fabric strength coefficient of variation, with a weight of 0.6, and energy consumption, with a weight of 0.4. The Apriori algorithm, with a minimum support of 0.05 and a minimum confidence of 0.7, identified 500 valid rules from 100,000 production records. These rules were converted into control weights, such as a 5% increase in yarn tension weight for every 1D increase in dye polarity. The system generates weekly performance reports, including the root mean square error of model predictions, which has been reduced to 0.02; the coefficient of variation in fabric strength, which has dropped from 4% to 2.5%; and the process adaptability score for new dye formulations, which has increased from 70 to 85 out of 100. Through continuous optimization, weaving quality stability has improved by 20%, and the adaptation cycle to new dye formulations has been shortened by 40%.

[0056] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A large model-assisted knitting production line dynamic configuration and rapid response method, characterized in that: The method comprises: Based on the dye molecular weight, the number of dye polar groups, and the spatial structure of the dye molecule as dye property parameters, historical data of the knitting production line is obtained and a dye property data warehouse is constructed; Based on historical data from knitting production lines, dye molecular structure parameters, yarn surface morphology, material properties, and friction coefficient test data were selected to construct a correlation prediction model between dye properties and yarn friction coefficient. The correlation data between dye properties and yarn friction coefficient was mined, and the correlation between dye molecular structure parameters and yarn surface morphology, material properties and friction coefficient numerical range was analyzed to obtain the distribution characteristics of dye properties and friction coefficient of yarn under different dye property combinations; Based on the dye properties and friction coefficient distribution characteristics, the dye formula components are clustered and grouped, and the dye combinations with similar effects on the yarn friction coefficient are determined to form a dye formula and friction coefficient mapping library; Acquire knitting machine structural parameters, needle motion trajectory equations, fabric structure matrix, and fabric density parameters, and combine them with the dye formula and friction coefficient mapping library to build a digital twin model of the weaving process; The digital twin model was designed to set boundary conditions for dye properties, yarn characteristics, and friction coefficient. Finite element simulation was used to simulate yarn movement and fabric formation during weaving, and the effects of dye properties and yarn friction coefficient on the weaving process were analyzed. During the weaving process, real-time data on the yarn friction coefficient of dye properties is collected, compared with the simulation results of the digital twin model, and the needle motion trajectory equation and weaving parameters are dynamically adjusted to achieve real-time optimization control of the weaving process; When a new dye formula is used on a knitting production line, the yarn friction coefficient is predicted based on the dye formula and the friction coefficient mapping library. The weaving process is simulated and optimized using a digital twin model, providing weaving parameter setting ranges and weaving process optimization suggestions. Continuously collect data on dye properties, yarn friction coefficient and weaving process parameters, iteratively update and adaptively adjust the digital twin model and weaving process optimization system, improve the stability of weaving quality and the process adaptability of new dye formulas, and match dye formulas with weaving parameters.

2. A large model-assisted knitting production line dynamic configuration and rapid response method according to claim 1, characterized in that: The method comprises: obtaining historical data of a knitting production line based on the dye molecular weight, the number of polar groups of the dye, and the spatial structure of the dye molecule as dye property parameters, and constructing a dye property data warehouse, including: Obtaining dye molecular weight information and the number of polar groups, wherein the dye molecular weight information is obtained from the PubChem database, and the number of polar groups is calculated using a SMILES string analysis method; Based on the molecular weight information of the dye and the number of polar groups, a three-dimensional spatial structure model of the dye molecule is constructed using the PyMOL molecular modeling tool; Extracting production batch information from the past five years from the production line history database, using Python's Pandas library to perform outlier detection and processing on the production batch information to obtain a standardized data set; Apache Hive was used to build a data warehouse, and dye property parameters and standardized data sets were mapped to fact tables and dimension tables. Multidimensional data cubes were constructed using Microsoft Analysis Services. The relationship between dye properties and production parameters was analyzed using correlation analysis methods, and a dye selection prediction model was constructed using the CART decision tree algorithm; If it is necessary to generate a dye selection recommendation list, the Apriori association rule mining algorithm is used to extract the association rules between dye properties and production parameters from the data warehouse.

3. A large model-assisted knitting production line dynamic configuration and rapid response method according to claim 1, characterized in that: The method comprises the following steps: selecting dye molecular structure parameters, yarn surface morphology, material properties and friction coefficient test data based on historical data of the knitting production line to construct a correlation prediction model between dye properties and yarn friction coefficient; and Obtaining dye molecular structure parameters, wherein the dye molecular structure parameters include molecular weight, number of polar groups and spatial configuration; Analyze microscopic photos of the yarn surface to obtain fiber alignment and surface roughness, and collect yarn material property data and friction coefficient; The raw data is preprocessed. If outliers are detected, they are removed. The spline interpolation algorithm is used to fill in the missing data. The data from different sources are formatted and converted to obtain the processed data. Construct a multidimensional data model, with dye batches and yarn batches as fact tables, and dye molecular structure parameters, yarn surface morphology, material properties, and friction coefficient data as dimension tables; Time series analysis of historical data was performed to obtain the dynamic changes in the relationship between dye properties and yarn friction coefficient; Perform dimensionality reduction processing, calculate eigenvalues, select the principal component whose cumulative contribution rate reaches the preset threshold, and obtain the reduced dimensionality data; Construct a correlation prediction model between dye properties and yarn friction coefficient and set the number of decision trees; The Gini index was used for feature selection, and the model accuracy was evaluated through cross-validation to obtain optimization suggestions for dye selection and yarn friction coefficient.

4. A large model-assisted knitting production line dynamic configuration and rapid response method according to claim 1, characterized in that: The data on the correlation between dye properties and yarn friction coefficients is mined, and the correlation between dye molecular structure parameters and yarn surface morphology, material properties and friction coefficient numerical ranges is analyzed to obtain the distribution characteristics of dye properties and friction coefficients of yarns under different dye property combinations, including: Obtain dye molecular structure parameters, yarn surface morphology, material properties and friction coefficient data from the data warehouse; Based on the data of dye molecular structure parameters, yarn surface morphology, material properties and friction coefficient, the Apriori algorithm was used to mine association rules, and a set of strong association rules between dye molecular structure parameters and yarn surface morphology, material properties and friction coefficient was obtained. Based on the strong association rule set, the K-means clustering algorithm is used to classify the dye property combinations and determine the statistical distribution parameters of the yarn friction coefficient under each category; If the statistical distribution parameters meet the preset conditions, a multivariate linear regression method is applied to establish a mathematical model between the dye properties and the yarn friction coefficient; Through mathematical models, the friction coefficient distribution characteristics of yarns under different dye property combinations are obtained; According to the distribution characteristics of friction coefficient, box plot and kernel density estimation method are used to generate visualization results; Determine whether the visualization results meet the preset threshold. If so, output a report on the dye properties and friction coefficient distribution characteristics. It also includes: analyzing the correlation between the dye molecular structure, chromatographic characteristics and the yarn surface roughness and linear density value, and obtaining the surface morphological characteristics of the yarn under different dye properties.

5. The large model-assisted knitting production line dynamic configuration and rapid response method according to claim 1, characterized in that: The method clusters and groups the dye formula compositions according to the dye properties and the friction coefficient distribution characteristics, determines the dye combinations with similar effects on the yarn friction coefficient, and forms a dye formula and friction coefficient mapping library, including: Dye property data and yarn friction coefficient data were obtained, and the Z-score normalization method was used to convert the data of different dimensions into a unified scale; Principal component analysis was performed on the standardized data to determine the principal component with a cumulative contribution rate of 95% as the input feature for cluster analysis; The DBSCAN algorithm is used to preprocess the input features to obtain a data set with outliers removed; The dye formula components were grouped using the K-means clustering algorithm based on the data set, and the optimal number of clusters was determined using the silhouette coefficient method. If the optimal number of clusters is determined, the dye property parameters and the corresponding friction coefficient mean of each cluster center are calculated; Based on the clustering results, a mapping relationship between dye formula and friction coefficient is constructed; The CART decision tree algorithm was used to establish a prediction model from dye properties to friction coefficient, and the prediction model parameters were optimized through grid search. Generate a dye formula and friction coefficient mapping library based on the prediction model, the mapping library including dye combination characteristics and corresponding friction coefficient prediction values; The sliding window method is used to update the mapping relationship between dye formula and friction coefficient in the mapping library.

6. A large model-assisted knitting production line dynamic configuration and rapid response method according to claim 1, characterized in that: The acquisition of knitting machine structure parameters, needle motion trajectory equations, fabric structure matrix and fabric density parameters, combined with dye formula and friction coefficient mapping library, to construct a digital twin model of the weaving process includes: Acquire mechanical structure parameters from a knitting machine database, wherein the mechanical structure parameters include a needle cylinder diameter, a needle number, and a coil length; A high-precision position sensor is used to collect the knitting needle movement position information, and the knitting needle movement trajectory equation is obtained by fitting the position information through the cubic spline interpolation method; According to the fabric design drawings, convolutional neural networks are used to perform image segmentation and feature extraction to obtain the fabric structure matrix; Obtain fabric density parameters through a density measuring instrument, and store the fabric structure matrix and density parameters in a database; The association model between dye formula, friction coefficient and weaving parameters was established using random forest regression method. Matching dye recipes with data from the friction coefficient mapping library and weaving parameters to obtain a comprehensive data set; A long short-term memory network is used to construct a digital twin model of the weaving process. The digital twin model takes mechanical parameters, trajectory equations, tissue structure matrix, density parameters, and dye formula as inputs, and friction coefficient as output. Optimize the parameters of the digital twin model through the back-propagation algorithm; Use 3D rendering technology to visualize the operation process of the digital twin model, showing the knitting needle movement and fabric formation process in real time; The digital twin model is optimized through the particle swarm optimization algorithm to improve the prediction accuracy.

7. The large model-assisted knitting production line dynamic configuration and rapid response method according to claim 1, characterized in that: The dye properties, yarn characteristics, and friction coefficient boundary conditions are set in the digital twin model, and the finite element simulation method is used to simulate the yarn movement and fabric formation process during the weaving process, and the influence of the dye properties and the spinning friction coefficient during weaving on the weaving process is analyzed, including: Acquire dye properties, yarn characteristics, and friction coefficient data, and set boundary condition ranges, where the boundary condition ranges include an upper limit value of dye molecular weight, a lower limit value of dye molecular weight, an upper limit value of polar group number, a lower limit value of polar group number, an upper limit value of yarn linear density, a lower limit value of yarn linear density, an upper limit value of yarn strength, a lower limit value of yarn strength, an upper limit value of yarn elongation, a lower limit value of yarn elongation, and an upper limit value and a lower limit value of friction coefficient; The yarn geometric model and the fabric geometric model are constructed using ANSYS finite element analysis software. Once the yarn geometric model is constructed, the yarn geometric model is discretized into beam elements, where each yarn is divided into a preset number of elements. The yarn material properties and contact conditions are determined according to the yarn geometric model, and the knitting needle motion trajectory and weaving tension are set as external loads; The weaving process is simulated using an explicit dynamics solver, which determines the time step based on the Courant-Friedrichs-Lewy condition. An orthogonal experimental design method is used to set up an experimental scheme, wherein the experimental scheme includes the number of factors and the number of levels, a simulation experiment is performed according to the experimental scheme, and a significance test of the simulation results is performed using variance analysis; ParaView scientific visualization software was used to display yarn motion trajectory and stress distribution, and a weight model of influencing factors was established through grey correlation analysis. The weight model of influencing factors was used to quantitatively describe the influence of each factor on the weaving process. It also includes: setting the dye molecular size, polarity, adsorption capacity parameters and yarn friction coefficient as boundary conditions, using finite element simulation to simulate the dynamic changes of yarn tension, speed and needle motion trajectory during weaving, and analyzing the influence of dye properties and friction coefficient on the stability of the weaving process.

8. A large model-assisted knitting production line dynamic configuration and rapid response method according to claim 7, characterized in that: The method sets the dye molecular size, polarity, adsorption capacity parameters and yarn friction coefficient as boundary conditions, uses finite element simulation to simulate the dynamic changes of yarn tension, speed and needle motion trajectory during weaving, and analyzes the influence of dye properties and friction coefficient on the stability of the weaving process, including: Acquiring dye molecular size, polarity, and adsorption capacity data from a pre-established dye database; measuring the yarn friction coefficient using an inclined plane method, with the measurement performed under preset temperature and humidity conditions; Based on the acquired dye data and friction coefficient, a numerical model of the weaving process is constructed. The yarn is discretized into predefined beam elements, and the knitting needle motion trajectory equation is set. The weaving process is simulated using an explicit dynamics solver; the yarn tension, velocity, and displacement are calculated at each time step, and the real-time coordinates of the needle motion trajectory are recorded; Through the orthogonal experimental design method, a multi-factor and multi-level experimental plan was set up; multiple simulation experiments were carried out to calculate the fluctuation coefficient of yarn tension, speed and knitting needle motion trajectory; If the fluctuation coefficient is less than the preset threshold, the weaving process is judged to be stable; otherwise, a sensitivity analysis is performed to determine the influence of each parameter on the stability.

9. A large model-assisted knitting production line dynamic configuration and rapid response method according to claim 1, characterized in that: The method collects dye property yarn friction coefficient data in real time during the weaving process, compares the simulation results of the digital twin model, and dynamically adjusts the needle motion trajectory equation and weaving parameters to achieve real-time optimization control of the weaving process, including: Obtaining yarn friction coefficient data collected by an optical fiber sensor, wherein the optical fiber sensor is installed on a weaving machine; Based on yarn friction coefficient data and dye property data, a physics-based finite element digital twin model is used for real-time simulation to obtain theoretical weaving parameters and needle motion trajectories. The root mean square error method is used to compare the differences between the yarn friction coefficient data and the theoretical weaving parameters, and a deviation report containing the error size and trend is generated; Based on the deviation report, a weaving parameter optimization model is constructed using a deep Q learning network, and the weaving parameter optimization model is used to dynamically adjust the needle motion trajectory equation and weaving parameters; If the output result of the weaving parameter optimization model meets the preset performance index threshold, the optimization effect of the weaving parameter optimization model is continuously evaluated using the sliding time window method; According to the optimization effect, the reward function of the deep Q learning network is dynamically adjusted, and the reward function is used to achieve continuous optimization and closed-loop control of the weaving process.

10. The large model-assisted knitting production line dynamic configuration and rapid response method according to claim 1, characterized in that: When a new dye formula is used in the knitting production line, the yarn friction coefficient is predicted based on the dye formula and the friction coefficient mapping library, and the weaving process is simulated and optimized through the digital twin model to provide weaving parameter setting ranges and weaving process optimization suggestions, including: Obtaining the chemical composition and property parameters of a new dye formula from a dye formula database, wherein the property parameters include molecular weight, number of polar groups, and spatial configuration; Based on the chemical composition and property parameters, the comprehensive impact index is calculated using the weighted average method; Inputting the comprehensive impact index into a pre-established random forest regression model, the random forest regression model is used to predict the range of yarn friction coefficient; The predicted friction coefficient range is input into the digital twin model, and Latin hypercube sampling is used to generate a multi-tissue fabric parameter combination, which includes needle movement speed, yarn tension, and yarn feed speed. Run simulation experiments based on the weaving parameter combinations to simulate the weaving process; The simulation results were processed using Python's pandas library to obtain the weaving efficiency, fabric quality index, and energy consumption level for each parameter combination. NSGA-II multi-objective optimization was performed on weaving efficiency, fabric quality indicators, and energy consumption levels to select the optimal parameter combination that meets production requirements; A weaving parameter setting range report and process optimization suggestions are generated based on the optimal parameter combination. The process optimization suggestions include the value ranges of amplitude A, angular frequency ω and phase φ in the needle motion trajectory equation S(t), the tension control curve represented by a piecewise linear function, and the yarn feeding speed adjustment strategy.

Images