Model simulation method and system for digital spinning sweater

Abstract

The invention relates to the technical field of sweater model simulation, in particular to a model simulation method and system for digital spinning of sweaters. The method comprises the following steps: obtaining a to-be-designed user omni-directional image, textile raw material multi-scale parameters and a textile process log; analyzing geometric morphology parameters of the user omnibearing image to be designed, performing three-dimensional geometric point cloud modeling, and constructing a user three-dimensional morphology model; performing physical characteristic parameter analysis on the multi-scale parameters of the textile raw materials, and performing deep parameter characteristic mining to obtain multi-scale state characteristics of the textile raw materials; and performing external mechanical deformation response analysis and internal fiber elastic response analysis based on the multi-scale state characteristics of the textile raw materials to generate external mechanical deformation response data of the textile raw materials and internal fiber elastic response characteristics of the textile raw materials. According to the invention, personalized and visual user try-on simulation is realized through sweater model simulation, and the production efficiency of spinning sweaters is improved.

Description

Technical Field

[0001] The invention relates to the technical field of sweater model simulation, and in particular to a model simulation method and system for digital textile sweaters. Background Art

[0002] With the development of computer simulation technology, especially with the support of 3D modeling, virtual reality and artificial intelligence algorithms, the model simulation method of digital textile sweaters has gradually become a research hotspot in the textile field. By establishing a virtual model of a sweater and conducting multi-level and multi-physical field simulations, the physical properties and wearing effects of the sweater can be effectively predicted, such as the elasticity, breathability, warmth retention and physical stability of the fabric. The widespread application of this technology can not only improve the accuracy of the design, but also enable rapid adjustments in simulations of different materials, styles and sizes, significantly shortening the product development cycle, reducing production costs, and better meeting consumers' growing demand for personalization and customization.

[0003] However, the current model simulation technology of digital textile sweaters still faces certain challenges. Traditional textile modeling methods mainly rely on simple geometric modeling and physical parameter settings. It is difficult to truly restore the microstructure and mechanical properties of sweater fiber materials, and it is even more difficult to accurately predict the complex mechanical behavior during the weaving process. This not only limits the innovation of sweater design, but also makes it difficult to support virtual try-on and production process optimization based on digital models. In addition, fabrics are affected by multiple factors such as yarn quality, weaving method, and textile machine parameters in actual production. How to achieve accurate mapping of these multiple dimensions in digital simulation is a key problem in the current technological development. Therefore, there is an urgent need for a more accurate and efficient digital textile sweater model simulation method that can fully consider the influence of material properties, weaving process and external environment, so as to achieve more realistic and reliable virtual sample development and optimization. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a model simulation method and system for a digital textile sweater to solve at least one of the above technical problems.

[0005] To achieve the above object, the present invention provides a model simulation method for a digital textile sweater, comprising the following steps: Step S1: obtaining a full-dimensional image of the user to be designed, multi-scale parameters of textile raw materials and textile process logs; analyzing the geometric morphological parameters of the full-dimensional image of the user to be designed, and performing three-dimensional geometric point cloud modeling to construct a three-dimensional morphological model of the user; Step S2: performing physical characteristic parameter analysis on multi-scale parameters of textile raw materials and performing deep parameter feature mining to obtain multi-scale state characteristics of textile raw materials; Step S3: performing external mechanical deformation response analysis and internal fiber elastic response analysis based on the multi-scale state characteristics of the textile raw material to generate external mechanical deformation response data of the textile raw material and internal fiber elastic response characteristics of the textile raw material; Step S4: Perform dynamic textile deformation mapping on the textile process log according to the external mechanical deformation response data of the textile raw materials and the internal fiber elastic response characteristics of the textile raw materials, and construct a digital twin model of the textile sweater; Step S5: Perform dye penetration interactive mining on the digital twin model of the textile sweater, and then perform dynamic color rendering to construct a dynamic rendering digital twin model; Step S6: Based on the dynamic rendering digital twin model, a dynamic try-on simulation is performed on the user's three-dimensional morphological model, and the intelligent sweater parameters are tuned to build an intelligent parameter optimized sweater model.

[0006] The present invention can accurately capture the user's body characteristics through geometric morphological parameter analysis and three-dimensional point cloud modeling of the user's omnidirectional image, provide an accurate reference for subsequent design, and construct a user's three-dimensional morphological model to help personalize clothing and improve the accuracy and conformity of the design. Through physical feature parameter analysis and deep feature mining, the characteristics of textile raw materials can be fully understood, providing a basis for subsequent simulation and design. Obtaining multi-scale state characteristics of textile raw materials is helpful in selecting suitable raw materials and improving the quality and comfort of sweaters. External mechanical deformation response analysis and internal fiber elastic response analysis can help understand the behavior of textile raw materials under different conditions, providing a basis for subsequent simulation and optimization. Generating these data is helpful to simulate the actual deformation and elastic characteristics of textiles, and improve the authenticity of model simulation. Through dynamic textile deformation mapping, the characteristics of raw materials can be mapped into the model, and a real digital twin model can be built, which helps to simulate the morphological changes of textiles. Building a digital twin model can provide a reliable reference for subsequent design and production, and improve product quality and compliance. Through dye penetration interactive mining and dynamic color rendering, different dyeing effects can be simulated to provide designers with more choices and inspiration. Building a dynamic rendering digital twin model helps to show the effects of sweaters under different color schemes and improve the diversity and innovation of the design. Through dynamic try-on simulation and intelligent parameter tuning, it can be customized according to the user's physical characteristics and needs, improving the comfort and fit of clothing. Building an intelligent parameter optimized sweater model helps to improve production efficiency and customer satisfaction, and realize customized production and services.

[0007] Preferably, step S1 comprises the following steps: Step S11: obtaining a user's omnidirectional image to be designed, multi-scale parameters of textile raw materials, and a textile process log; Step S12: reconstructing the image details of the designed omnidirectional image of the user to construct a super-resolution reconstructed image; Step S13: performing deep visual semantic segmentation on the super-resolution reconstructed image to identify the user's body part information; Step S14: Identify key points of user body parts based on the user body part information and mark the key points of the body parts; Step S15: performing a user bone structure analysis based on key points of body parts to generate a user bone structure feature; Step S16: Performing geometric morphological parameter analysis on the user's skeletal structural features to generate the user's geometric morphological parameters; performing three-dimensional geometric point cloud modeling on the user's geometric morphological parameters to construct a three-dimensional morphological model of the user.

[0008] The present invention can establish a comprehensive basic data set by acquiring all-round images, textile parameters and process logs, and provide necessary information and background for subsequent steps. The comprehensive use of these data helps to ensure the accuracy and integrity of the design process. Image detail reconstruction can improve the clarity and quality of the image, help capture the detailed features of the user's body, and provide more accurate data for subsequent analysis and modeling. Constructing super-resolution reconstructed images helps to improve the detailed display of the image and provide a better basis for subsequent body part information recognition. Through key point recognition, important feature points of the user's body can be marked, which is helpful for subsequent bone structure analysis and generation of geometric morphological parameters. Marking key points of body parts can provide more accurate data for subsequent steps and ensure the accuracy and authenticity of the model. Through user bone structure analysis, the user's bone characteristics and structure can be understood, which provides a basis for subsequent geometric morphological parameter analysis. Generating user bone structure characteristics helps to simulate the user's body shape and provide key information for subsequent model establishment. Through geometric morphological parameter analysis, the user's geometric morphological parameters can be generated, which provides a basis and basis for establishing a user's three-dimensional morphological model. Constructing a user's three-dimensional morphological model can better understand the user's body structure and provide an accurate basis for subsequent sweater design and fitting simulation.

[0009] Preferably, the specific steps of step S12 are: Perform filtering and denoising processing on the designed user's omnidirectional images one by one to obtain a filtered and denoised omnidirectional image; Perform global outlier scanning on the filtered and denoised omnidirectional image to extract image outliers; Perform abnormal outlier elimination on image outliers to obtain an outlier optimized omni-directional image; Perform image detail recognition on the outlier optimized omnidirectional image and extract image detail features; Perform magnification and distortion analysis on image detail features to obtain image magnification and distortion data; Based on the image magnification distortion data, the outlier optimized omnidirectional image is smoothed to obtain a smooth image of the detailed parts; Extract the original resolution point cloud of the smoothed image of the detail part to obtain the original image point cloud resolution; Perform deep convolution learning based on the original image point cloud resolution to obtain point cloud resolution convolution features; Based on the point cloud resolution convolution features, super-resolution reconstruction is performed on the smooth image of the detail parts to construct a super-resolution reconstructed image.

[0010] The present invention reduces noise and interference in the image, improves image quality and clarity, helps to accurately extract detailed features in the image, provides more accurate data for subsequent processing steps, finds abnormal points in the image, helps identify erroneous or abnormal data, extracts outliers to help further optimize the image, ensures the accuracy and integrity of the data, cleans up abnormal outliers in the image, reduces interference and misleading data, optimizes the full range of images, helps improve the accuracy and efficiency of subsequent processing steps, extracts detailed features in the image, helps to better understand the image content, and the identification of detailed features provides important information for subsequent analysis and processing, understands the distortion that occurs during image magnification, and helps maintain image quality. The analysis helps to maintain the authenticity and clarity of image details in subsequent processing. The smoothing process can reduce the noise and discontinuity in the image. The smoothing process helps to improve the overall quality and visualization of the image. The point cloud data of the original image is extracted to provide a basis for subsequent analysis and processing. This step helps to retain the details and accuracy of the image. Through deep learning technology, the feature information in the point cloud data is extracted. The learned convolution features can be used for subsequent model construction and analysis, and super-resolution reconstructed images are constructed to improve the clarity and detail display of the image, which helps to generate more realistic and accurate images and provide a better visual basis for the model simulation of digital textile sweaters.

[0011] Preferably, the specific steps of step S2 are: Step S21: performing physical characteristic parameter analysis on multi-scale parameters of textile raw materials to extract physical characteristic parameters of textile raw materials; Step S22: Calculating the fiber mechanical properties of the physical characteristic parameters of the textile raw material, thereby obtaining the fiber mechanical property data of the textile raw material; Step S23: Calculating the inter-fiber friction coefficient based on the physical characteristic parameters of the textile raw materials to obtain the inter-fiber friction parameters; Step S24: Analyze the mechanical properties of the fiber bundle based on the physical characteristic parameters of the textile raw materials to obtain the mechanical property data of the fiber bundle; Step S25: Perform deep parameter feature mining on the mechanical property data of textile raw material fibers, the friction parameters between fibers, and the mechanical property data of fiber bundles, so as to obtain multi-scale state characteristics of textile raw materials.

[0012] The present invention can deeply understand the physical properties of textile raw materials through physical characteristic parameter analysis, provide a basis for subsequent performance calculation and characteristic analysis, extract physical characteristic parameters of textile raw materials to help determine the characteristics and properties of materials, and provide accurate input data for model simulation, fiber mechanical property calculation can reveal the performance of textile raw materials under stress, provide key information for further analysis and design, and obtain fiber mechanical property data to help evaluate the strength, elasticity and other characteristics of materials, which is of great significance in model simulation, calculate the friction coefficient between fibers to help understand the interaction between fibers of textile raw materials, affecting the performance and characteristics of textiles, obtain the friction parameters between fibers to help simulate the performance of textiles under different conditions, and provide more accurate data for model simulation, through the mechanical property analysis of fiber bundles, the stress conditions and performance characteristics of fiber bundles of textile raw materials can be understood, providing an important reference for design and simulation, and obtain the mechanical property data of fiber bundles to help evaluate the performance of textiles in actual use, and provide practical information for model simulation, through deep parameter feature mining, fiber mechanical properties, friction parameters and fiber constraint mechanical properties can be integrated to obtain multi-scale state characteristics of textile raw materials, these multi-scale state characteristics can fully display the performance and characteristics of textile raw materials, and provide comprehensive data support for model simulation of digital textile sweaters.

[0013] Preferably, step S3 specifically comprises the following steps: Step S31: performing external mechanical deformation response analysis on the multi-scale state characteristics of the textile raw material to generate external mechanical deformation response data of the textile raw material; Step S32: performing internal fiber bundle discrete analysis modeling based on multi-scale parameters of the textile raw material to obtain discrete fiber particles inside the textile raw material; Step S33: simulating the external load action on the discrete fiber particles inside the textile raw material to extract simulation data of the external load action on the discrete fiber particles; Step S34: performing particle-by-particle force balance calculation on the fiber discrete particle interaction simulation data, and extracting the force balance data of each fiber discrete particle; Step S35: Performing an internal fiber elastic response analysis based on the force balance data of each discrete fiber particle, thereby obtaining the internal fiber elastic response characteristics of the textile raw material.

[0014] The present invention can understand the deformation and response characteristics of textile raw materials under external force conditions through external mechanical deformation response analysis, provide key data for model simulation, and generate external mechanical deformation response data of textile raw materials to help evaluate the deformation behavior and performance of materials, and provide a basis for design and simulation. Through internal fiber bundle discrete analysis modeling, the structure and arrangement of internal fibers of textile raw materials can be simulated, providing a basis for subsequent simulation and analysis. Obtaining internal fiber discrete particles of textile raw materials is helpful to understand the internal structure of materials and provide detailed information for model simulation. Through external load simulation, the behavior of fiber discrete particles under external load can be simulated, providing basic data for further analysis, and extracting fiber discrete particles. The simulation data of external load on particles helps to understand the response of materials under stress and provide necessary information for subsequent steps. The force balance calculation of each particle can reveal the interaction and stress between discrete fiber particles, providing a basis for simulating the internal mechanical properties of textile raw materials. Extracting the force balance data of each discrete fiber particle helps to understand the internal micromechanical behavior of the material and provide key data for model simulation. Through the analysis of internal fiber elastic response, the elastic characteristics and response behavior of the internal fibers of textile raw materials under stress can be understood, providing important information for model simulation. Obtaining the elastic response characteristics of the internal fibers of textile raw materials helps to evaluate the elastic properties and deformation characteristics of the material and provide support for design and simulation.

[0015] Preferably, the specific steps of step S31 are: Based on the multi-scale state characteristics of textile raw materials, multi-frequency pressure application simulation is performed on textile raw materials to extract the pressure simulation response state parameters of textile raw materials; Identify fiber deformation characteristics of textile raw material pressure simulation response state parameters and generate fiber deformation characteristic data; Conduct stress-strain analysis on fiber deformation characteristic data to obtain stress-strain data of textile raw materials; Conduct strain response change analysis on stress-strain data of textile raw materials to obtain strain response change curve; The external mechanical deformation response analysis is performed on the strain response change curve to generate the external mechanical deformation response data of textile raw materials.

[0016] The present invention can simulate the pressure response of textile raw materials under different frequencies through multi-frequency pressure application simulation, understand the deformation and response of materials under different frequencies, extract the pressure simulation response state parameters of textile raw materials, help evaluate the performance of materials under multi-frequency pressure, and provide a data basis for subsequent analysis, identify fiber deformation characteristics, help understand the deformation mode and degree of textile raw materials under pressure, and provide key information for subsequent analysis, generate fiber deformation characteristic data, help understand the internal structure change of materials, and provide a basis for stress-strain analysis, stress-strain analysis can help determine the stress distribution and deformation of textile raw materials under applied pressure, and provide important data for material performance evaluation, obtain stress-strain data of textile raw materials, help understand the strength and deformation characteristics of materials, and provide support for model simulation, strain response change analysis can reveal the deformation characteristics and response behavior of textile raw materials under different strains, and provide key information for further analysis, obtain strain response change curves, help understand the performance changes of materials under different strains, and provide a data basis for subsequent steps, external mechanical deformation response analysis can help understand the overall deformation characteristics of textile raw materials under different strain response changes, and provide a comprehensive understanding for model simulation, and generate external mechanical deformation response data of textile raw materials to evaluate the deformation performance of materials under external mechanical action, and provide support for design and simulation.

[0017] Preferably, the specific steps of step S4 are: Step S41: performing a multi-stage textile process analysis on the textile process log, and extracting textile process data of each stage; Step S42: Performing process timing logic analysis on the textile process data of each stage to generate textile process timing logic; Step S43: performing a textile process simulation on the textile sweater based on the textile process sequential logic to obtain a textile sweater simulation model; Step S44: Based on the external mechanical deformation response data of the textile raw materials and the internal fiber elastic response characteristics of the textile raw materials, the textile sweater simulation model is dynamically mapped to construct a digital twin model of the textile sweater.

[0018] By analyzing the textile process log, the present invention can deeply understand the manufacturing process of the digital textile sweater and extract the process data of each stage. Extracting the textile process data of each stage is helpful to understand the specific operations and parameters of each step, and provide basic data for subsequent analysis. Through the process timing logic analysis, the time sequence and logical relationship between the textile processes can be established to ensure the smooth progress of the manufacturing process. Generating the textile process timing logic is helpful to optimize the production process, improve efficiency and quality control. Through process textile simulation, the manufacturing process of the digital textile sweater can be simulated to obtain a simulation model for further analysis and optimization. Obtaining the textile sweater simulation model is helpful to understand the details and improvement points of the entire production process, improve production efficiency and product quality. Through dynamic textile deformation mapping, the external mechanical deformation response data is combined with the internal fiber elastic response characteristics to construct a digital twin model of the digital textile sweater. Constructing the digital twin model is helpful to simulate the deformation and performance of the textile sweater under different conditions, and provide an important reference for product design and optimization.

[0019] Preferably, the specific steps of step S5 are: Step S51: performing preset dyeing demand analysis on the textile process log to extract preset process dyeing demand data; Step S52: Calculate dye parameters for preset process dyeing requirement data and extract dye characteristic parameters; Step S53: performing dye penetration interaction mining on dye characteristic parameters based on multi-scale state characteristics of textile raw materials to generate textile raw material dye penetration interaction data; Step S54: performing dye diffusion evolution analysis on the textile raw material dye penetration interaction data to generate dynamic dye diffusion evolution data; Step S55: Perform dynamic color rendering on the digital twin model of the textile sweater according to the dynamic dye diffusion evolution data to construct a dynamic rendering digital twin model.

[0020] The present invention can determine the dyeing requirements and parameters of the digital textile sweater through dyeing demand analysis, extract the required process dyeing data, and extract the preset dyeing demand data to help ensure that the dyeing process meets the design requirements and provide basic data for subsequent steps. The characteristic parameters of the required dyes can be determined by calculating the dye parameters, so that accurate dyeing formula and control can be performed during the dyeing process. Extracting the characteristic parameters of the dyes helps to ensure the accuracy and consistency of the dyeing process and improve the quality and stability of the dyeing effect. Through dye penetration interaction mining, the penetration and interaction of the dyes in the textile raw materials can be deeply understood, and relevant data can be generated. Generating textile raw material dye penetration interaction data helps to understand the behavior of the dye in the textile material and provide a basis for subsequent analysis and control. Through dye diffusion evolution analysis, the diffusion process of the dye in the textile raw materials can be simulated, and dynamic dye diffusion evolution data can be generated. Generating dynamic dye diffusion evolution data helps to understand the propagation mode and speed of the dye in the material, and provides support for the control and optimization of the dyeing process. Through dynamic color rendering, the digital twin model of the textile sweater can be rendered in color according to the dynamic dye diffusion evolution data to show the color changes at different stages. Constructing a dynamic rendering digital twin model helps to intuitively show the changes in the dyeing effect and provide visualization support for design and display.

[0021] Preferably, the specific steps of step S6 are: Step S61: dynamically try on and simulate the user's three-dimensional morphological model based on the dynamic rendering digital twin model, thereby obtaining a textile sweater-user morphological simulation model; Step S62: performing fit analysis on the textile sweater-user form simulation model to extract sweater fit data; Step S63: calculating the tightness of the textile sweater-user morphology simulation model to generate the tightness of the user's sweater; Step S64: performing personalized user try-on feature analysis on the sweater fit data and the tightness of the user's sweater, and extracting personalized user try-on feature parameters; Step S65: Optimizing the parameters of the smart sweater based on the personalized user try-on feature parameters to obtain the personalized sweater optimization parameters; Step S66: Perform personalized process simulation optimization on the textile sweater-user form simulation model according to the personalized sweater optimization parameters, and construct an intelligent parameter optimized sweater model.

[0022] The present invention can simulate the actual effect of a sweater on a user by dynamically trying on a user's three-dimensional morphological model, and obtain a textile sweater-user morphological simulation model. The obtained simulation model is helpful to evaluate the fitness and appearance effect of a sweater under different user body shapes, and provides a reference for personalized design. Through fit analysis, the fit of a sweater on a user can be evaluated, and fit data can be extracted. The extracted data is helpful to understand the degree of adaptation of the sweater to the user's body shape, and provides a basis for subsequent adjustment and optimization. The tightness of the user's sweater can be calculated to determine the comfort and wearing feeling of the sweater, and provide a better wearing experience for the user. The generation of tightness data is helpful to understand the comfort and adaptability of the sweater, and provides a reference for personalized design. By analyzing the personalized user try-on feature parameters, we can gain an in-depth understanding of the clothing needs and preferences of different users. The extracted feature parameters are conducive to personalized design and customization, and improve user satisfaction and wearing comfort. Through intelligent parameter tuning, the sweater design can be optimized and adjusted according to the personalized user try-on feature parameters to improve the wearing effect and comfort. The obtained personalized sweater optimization parameters help to meet the personalized needs of users and improve the market competitiveness of products. According to the personalized sweater optimization parameters, the simulation model is optimized through process simulation, and an intelligent parameter optimization sweater model that better meets user needs can be constructed. The constructed model helps to improve the personalization of sweater design and wearing experience, and enhance the market appeal of the product and user satisfaction.

[0023] In this specification, a model simulation system of a digital textile sweater is provided, which is used to execute the model simulation method of the digital textile sweater as described above, comprising: The three-dimensional morphology module is used to obtain the user's omnidirectional image to be designed, the multi-scale parameters of textile raw materials and the textile process log; analyze the geometric morphology parameters of the user's omnidirectional image to be designed, and perform three-dimensional geometric point cloud modeling to construct the user's three-dimensional morphology model; The multi-scale state module is used to analyze the physical characteristic parameters of the multi-scale parameters of textile raw materials and conduct deep parameter feature mining to obtain the multi-scale state characteristics of textile raw materials; Mechanical deformation module, used to analyze external mechanical deformation response and internal fiber elastic response based on multi-scale state characteristics of textile raw materials, so as to generate external mechanical deformation response data of textile raw materials and internal fiber elastic response characteristics of textile raw materials; The textile deformation module is used to dynamically map the textile deformation of the textile process log according to the external mechanical deformation response data of the textile raw materials and the elastic response characteristics of the internal fibers of the textile raw materials, and to build a digital twin model of the textile sweater; The dynamic color rendering module is used to interactively mine dye penetration on the digital twin model of textile sweaters, and then perform dynamic color rendering to build a dynamic rendering digital twin model; The dynamic try-on simulation module is used to perform dynamic try-on simulation on the user's three-dimensional morphological model based on the dynamic rendering digital twin model, and to optimize the parameters of the smart sweater to build an intelligent parameter-optimized sweater model.

[0024] The present invention constructs a three-dimensional morphological model of the user through geometric morphological parameter analysis and three-dimensional geometric point cloud modeling of the user's omnidirectional image, so as to better understand the user's physical characteristics and morphology and provide a basis for subsequent design and simulation. By performing physical characteristic parameter analysis and deep parameter feature mining on multi-scale parameters of textile raw materials, multi-scale state characteristics of textile raw materials are obtained, which is helpful for in-depth understanding of the characteristics and state of textile raw materials and provides more accurate data support in the design and production process. External mechanical deformation response analysis and internal fiber elastic response analysis are performed based on the multi-scale state characteristics of textile raw materials to generate deformation and elastic response data of textile raw materials, which is helpful for understanding the deformation of textile materials under external mechanical action and provides a basis for subsequent simulation and optimization. The external deformation response data and internal fiber elastic response characteristics are used to perform dynamic textile deformation mapping on the textile process log, and a digital twin model of the textile sweater is constructed. This makes it possible to more accurately simulate the performance of textiles under different deformation conditions, providing a reference for the design and production process. Through dye penetration interactive mining and dynamic color rendering, a dynamic rendering digital twin model is constructed, allowing the design team to better preview and adjust the color effects of the sweater, understand the appearance of the sweater more intuitively, and improve design efficiency and product quality. Through the dynamic rendering digital twin model, the user's three-dimensional morphological model is tried on and simulated and intelligently tuned, and an intelligent parameter-optimized sweater model is constructed, which helps personalized design and production, improves the fit and comfort of the sweater, and enhances user experience and product competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of the steps of a model simulation method for a digital textile sweater according to the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0026] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0027] The present application example provides a model simulation method and system for a digital textile sweater. The execution subject of the model simulation method and system for a digital textile sweater includes but is not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio image management system, an information management system, and a cloud data management system.

[0028] See also Figures 1 to 4 The present invention provides a model simulation method for a digital textile sweater, and the model simulation method for a digital textile sweater comprises the following steps: Step S1: obtaining a full-dimensional image of the user to be designed, multi-scale parameters of textile raw materials and textile process logs; analyzing the geometric morphological parameters of the full-dimensional image of the user to be designed, and performing three-dimensional geometric point cloud modeling to construct a three-dimensional morphological model of the user; Step S2: performing physical characteristic parameter analysis on multi-scale parameters of textile raw materials and performing deep parameter feature mining to obtain multi-scale state characteristics of textile raw materials; Step S3: performing external mechanical deformation response analysis and internal fiber elastic response analysis based on the multi-scale state characteristics of the textile raw material to generate external mechanical deformation response data of the textile raw material and internal fiber elastic response characteristics of the textile raw material; Step S4: Perform dynamic textile deformation mapping on the textile process log according to the external mechanical deformation response data of the textile raw materials and the internal fiber elastic response characteristics of the textile raw materials, and construct a digital twin model of the textile sweater; Step S5: Perform dye penetration interactive mining on the digital twin model of the textile sweater, and then perform dynamic color rendering to construct a dynamic rendering digital twin model; Step S6: Based on the dynamic rendering digital twin model, a dynamic try-on simulation is performed on the user's three-dimensional morphological model, and the intelligent sweater parameters are tuned to build an intelligent parameter optimized sweater model.

[0029] The present invention can accurately capture the user's body characteristics through geometric morphological parameter analysis and three-dimensional point cloud modeling of the user's omnidirectional image, provide an accurate reference for subsequent design, and construct a user's three-dimensional morphological model to help personalize clothing and improve the accuracy and conformity of the design. Through physical feature parameter analysis and deep feature mining, the characteristics of textile raw materials can be fully understood, providing a basis for subsequent simulation and design. Obtaining multi-scale state characteristics of textile raw materials is helpful in selecting suitable raw materials and improving the quality and comfort of sweaters. External mechanical deformation response analysis and internal fiber elastic response analysis can help understand the behavior of textile raw materials under different conditions, providing a basis for subsequent simulation and optimization. Generating these data is helpful to simulate the actual deformation and elastic characteristics of textiles, and improve the authenticity of model simulation. Through dynamic textile deformation mapping, the characteristics of raw materials can be mapped into the model, and a real digital twin model can be built, which helps to simulate the morphological changes of textiles. Building a digital twin model can provide a reliable reference for subsequent design and production, and improve product quality and compliance. Through dye penetration interactive mining and dynamic color rendering, different dyeing effects can be simulated to provide designers with more choices and inspiration. Building a dynamic rendering digital twin model helps to show the effects of sweaters under different color schemes and improve the diversity and innovation of the design. Through dynamic try-on simulation and intelligent parameter tuning, it can be customized according to the user's physical characteristics and needs, improving the comfort and fit of clothing. Building an intelligent parameter optimized sweater model helps to improve production efficiency and customer satisfaction, and realize customized production and services.

[0030] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a model simulation method of a digital textile sweater of the present invention. In this example, the steps of the model simulation method of a digital textile sweater include: Step S1: obtaining a full-dimensional image of the user to be designed, multi-scale parameters of textile raw materials and textile process logs; analyzing the geometric morphological parameters of the full-dimensional image of the user to be designed, and performing three-dimensional geometric point cloud modeling to construct a three-dimensional morphological model of the user; In this embodiment, a high-resolution camera or a 3D scanner is used to obtain all-round images of the user to be designed. These images should cover all angles of the user, including the front, side and back. It is ensured that the images are taken under uniform lighting conditions to reduce the impact of shadows and reflections on post-processing. Relevant multi-scale parameter data of textile raw materials are collected, including fiber diameter, density, strength, elongation, etc. These data can be obtained through experimental measurement or from a material database. Textile process logs related to the sweater to be designed are collected to record information including textile process, process parameters, dyeing requirements, etc. The acquired user images are preprocessed using image processing software (such as OpenCV or MATLAB). These steps include denoising, contrast enhancement and edge detection, etc. By analyzing the image, the user's geometric parameters, such as height, shoulder width, chest circumference, waist circumference, hip circumference, etc., are extracted. This can be achieved through manual measurement or automated tools (such as deep learning models). A three-dimensional reconstruction algorithm (such as Structure from Motion (SfM) or Mult i-ViewStereo (MVS)) converts the processed images into point cloud data. These algorithms can generate dense 3D point clouds by combining the information of multiple images. Use point cloud processing software (such as CloudCompare or PCL) to filter and reduce noise on the generated point cloud, remove abnormal points, and align the point cloud to ensure the integrity and accuracy of the point cloud. Use 3D modeling software (such as Blender, MeshLab or 3dsMax) to convert the processed point cloud data into a 3D geometric model. Algorithms such as Delaunay triangulation or Poisson reconstruction can be used to create a mesh model. The generated 3D model is optimized, including simplifying the mesh, smoothing the surface and filling holes to improve the model quality and ensure its suitability for subsequent textile design and simulation. The final 3D geometric model is saved in a standard format (such as OBJ, STL or FBX) for subsequent processing and application. Verify the consistency of the generated user 3D morphology model with the actual user morphology, and adjust and optimize according to user feedback.

[0031] Step S2: performing physical characteristic parameter analysis on multi-scale parameters of textile raw materials and performing deep parameter feature mining to obtain multi-scale state characteristics of textile raw materials; In this embodiment, multi-scale parameter data of textile raw materials are collected, including but not limited to fiber diameter, density, strength, elongation, specific gravity, thermal conductivity, etc. These data can be obtained through experimental measurement, literature review or material database. The collected multi-scale parameters are organized into a structured format (such as an Excel table or database) to ensure the integrity, accuracy and consistency of the data, determine the unit and range of each parameter for subsequent analysis, and define physical characteristic parameters, including elastic modulus, tensile strength, compressive strength, stiffness, etc. These parameters are important indicators for evaluating the performance of textile raw materials. According to the characteristics of the textile raw materials, a suitable calculation method is selected. Commonly used methods include: Hooke's law: used to calculate elastic modulus, stress-strain curve analysis: used to calculate tensile strength and elongation, calculate the physical characteristic parameters according to the defined characteristic parameters and the selected calculation method, record the calculation results, and compare them with known standards to ensure the rationality and effectiveness of the results, standardize or normalize the physical characteristic parameters for subsequent analysis and mining, remove outliers and noise, ensure the clarity and accuracy of the data, and select a suitable depth Feature mining methods, such as principal component analysis (PCA): used to reduce data dimensions and extract main features, cluster analysis: used to identify patterns and groups in data, machine learning algorithms: such as random forests or support vector machines, to evaluate feature importance, apply selected feature mining methods to the sorted physical feature data, extract key deep parameter features, generate multi-scale state features of textile raw materials, including their performance under different conditions, record the mined multi-scale state features in a database or document to ensure data traceability and availability, use data visualization tools (such as Matplotlib, Tableau or PowerBI) to visualize the analysis results to help understand the relationship between the parameters, verify whether the results of deep feature mining are consistent with the actual material performance, adjust and optimize the model if necessary, apply the obtained multi-scale state features of textile raw materials to subsequent product design and process optimization to ensure that the material performance meets the design requirements, collect user feedback in actual applications to evaluate the effectiveness of the extracted features, and provide a reference for future feature extraction and analysis.

[0032] Step S3: performing external mechanical deformation response analysis and internal fiber elastic response analysis based on the multi-scale state characteristics of the textile raw material to generate external mechanical deformation response data of the textile raw material and internal fiber elastic response characteristics of the textile raw material; In this embodiment, it is ensured that the existing multi-scale state characteristic data of textile raw materials are complete, including physical properties, geometric properties and material parameters, and the experimental conditions for external mechanical deformation response analysis and internal fiber elastic response analysis are determined, including loading mode (such as tension, compression, shear), load size, temperature and humidity, etc. Finite element analysis (FEA) software (such as ANSYS, ABAQUS or COMSOL) is used to establish a mechanical model of the textile raw material. This model should take into account the anisotropy and nonlinear characteristics of the material, set appropriate boundary conditions, simulate the constraints in actual use, apply external loads to the model, record the type, size and application method (static or dynamic) of the applied load, run finite element analysis, monitor the response of the model under external mechanical action, including stress, strain, displacement and other data, generate external mechanical deformation response data of the textile raw material, record the stress and strain distribution of each node, and construct the internal fiber based on the multi-scale characteristics of the textile raw material. A discrete model of the bundle is constructed, and the discrete element method (DEM) or a fiber-based model is used to set corresponding physical parameters for each fiber particle in the model, such as elastic modulus, Poisson's ratio and strength, etc., and a suitable external load is applied to the internal fiber model to simulate the stress conditions of the fiber bundle in actual use. Numerical simulation methods (such as finite element method or molecular dynamics simulation) are used to analyze the elastic response of the internal fibers, and the stress and strain responses of each fiber are extracted. The obtained elastic response characteristic data of the internal fibers of the textile raw materials are recorded, including the stress, strain and their interaction of each fiber, and the external mechanical deformation response data and the internal fiber elastic response characteristic data are integrated to form a comprehensive data set for subsequent analysis. The response data are analyzed using data analysis software (such as MATLAB, Python or R), including statistical description and visual display, and stress-strain curves, displacement distribution diagrams and fiber response diagrams are drawn to help understand the performance of the material under external mechanical action.

[0033] Step S4: Perform dynamic textile deformation mapping on the textile process log according to the external mechanical deformation response data of the textile raw materials and the internal fiber elastic response characteristics of the textile raw materials, and construct a digital twin model of the textile sweater; In this embodiment, it is ensured that the external mechanical deformation response data and the internal fiber elastic response characteristic data of the textile raw materials are complete and in a unified format. These data should include information such as stress, strain, displacement and fiber response. Relevant textile process logs are collected to record important parameters and steps in the textile process, including weaving methods, machines used, operating conditions, etc. A dynamic mapping model is constructed based on the collected external mechanical deformation response data to determine how to combine the mechanical response with the actual textile process, and a suitable interpolation method (such as linear interpolation, spline interpolation or Kriging interpolation) is selected to process and map the mechanical response data so as to combine it with the time series data in the process log, and the external mechanical deformation response data and the internal fiber elastic response characteristic data are combined with the time series data in the process log. Each stage is mapped to generate a dynamic deformation model for each process stage. During the mapping process, the influence of environmental factors such as time, temperature, and humidity on the performance of textile materials is considered, the framework of the digital twin model is designed, and its structure, function, and required input and output parameters are determined. The model should be able to reflect the dynamic behavior of textile sweaters under different process conditions in real time. Suitable modeling tools or software (such as MATLAB Simulink, AnyLogic, Unity or other simulation software) are selected to build the digital twin model. In the selected software, the digital twin model of the textile sweater is constructed according to the dynamic mapping results and process log information, and the mechanical response data and process information are input into the model to ensure that it can update and display the performance of the sweater under different process conditions in real time.

[0034] Step S5: Perform dye penetration interactive mining on the digital twin model of the textile sweater, and then perform dynamic color rendering to construct a dynamic rendering digital twin model; In this embodiment, property data related to the dye are collected, such as the chemical composition, molecular size, solubility, diffusion coefficient of the dye and its interaction characteristics with textile raw materials, and the experimental data of the dye under different conditions are sorted out, including influencing factors such as temperature, pH value, time, etc. A dye penetration model is established, and the multi-scale characteristics of the textile raw materials and their influence on the dye penetration are considered. A mathematical model (such as Fick's law) can be used to describe the diffusion process of the dye. Numerical simulation software (such as COMSOL Multiphysics, ANSYS Fluent) or a custom algorithm is used to simulate the penetration process of the dye in the textile sweater, and data such as the penetration depth, diffusion speed and concentration change of the dye in the fiber are recorded to generate dye penetration interaction data. A color mapping model is designed based on the dye penetration interaction data to determine how to correspond the dye concentration to the color change. , you can use color space models (such as HSV or RGB) to map dye concentration to visual color, select suitable rendering tools or engines (such as Unity, UnrealEngine or Blender) for dynamic color rendering. These tools should support real-time rendering and dynamic updates, implement dynamic color rendering in the selected rendering tool, combine dye penetration interaction data, update the color performance of textile sweaters in real time, set rendering parameters such as lighting, reflection and material properties to enhance visual effects and ensure the realism of rendering results, compare rendering results with actual dyeing samples, verify the accuracy and authenticity of dynamic rendering models, collect user feedback, evaluate the effect of color rendering, and ensure that it meets design requirements, optimize the rendering model according to verification results and user feedback, adjust mapping algorithms and rendering parameters to improve rendering effects.

[0035] Step S6: Based on the dynamic rendering digital twin model, a dynamic try-on simulation is performed on the user's three-dimensional morphological model, and the intelligent sweater parameters are tuned to build an intelligent parameter optimized sweater model.

[0036] In this embodiment, it is ensured that the user's three-dimensional morphological model and the dynamic rendering digital twin model have been constructed and can be operated in the same environment. 3D modeling software (such as Blender, Unity or Maya) is used to ensure the compatibility of the models. A virtual fitting environment is built in the selected 3D software to ensure that the lighting, material and physical properties are consistent with the actual situation to achieve a realistic visual effect. The user's three-dimensional morphological model is combined with the dynamic rendering digital twin model, and a dynamic fitting effect is achieved through a physical simulation engine (such as Unity's physics engine or Blender's physics simulation). Simulation parameters such as gravity, friction, etc. are set to realistically simulate the performance of the sweater on the user. Now, record the contact between the sweater and the user's body, including data such as fit, deformation, and range of motion, determine the optimization goals of the smart sweater, such as improving fit, increasing comfort, optimizing appearance, etc., set quantitative indicators for each goal, such as contact area, pressure distribution, tightness, etc., select a suitable optimization algorithm, such as genetic algorithm, particle swarm optimization or Bayesian optimization, which can automatically adjust parameters according to the objective function to achieve the best design, input the data of dynamic try-on simulation into the optimization algorithm, run the algorithm to identify and adjust the key parameters of the sweater (such as size, shape, material properties, etc.), and in each iteration, evaluate the model performance and record the results until the set optimization goal is reached.

[0037] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: obtaining a user's omnidirectional image to be designed, multi-scale parameters of textile raw materials, and a textile process log; Step S12: reconstructing the image details of the designed omnidirectional image of the user to construct a super-resolution reconstructed image; Step S13: performing deep visual semantic segmentation on the super-resolution reconstructed image to identify the user's body part information; Step S14: Identify key points of user body parts based on the user body part information and mark the key points of the body parts; Step S15: performing a user bone structure analysis based on key points of body parts to generate a user bone structure feature; Step S16: Performing geometric morphological parameter analysis on the user's skeletal structural features to generate the user's geometric morphological parameters; performing three-dimensional geometric point cloud modeling on the user's geometric morphological parameters to construct a three-dimensional morphological model of the user.

[0038] In this embodiment, all-round images of the user are collected, which can be achieved through multi-angle shooting (such as 360-degree photography), ensuring that the image covers all parts of the user's body, obtaining multi-scale parameters of textile raw materials, including material thickness, elasticity, texture, etc., which can be obtained through laboratory tests or information provided by suppliers, collecting textile process logs, recording key parameters and operating steps in the textile process for subsequent analysis, and organizing the collected images, parameters and log data into a designated database or file system to ensure data integrity and accessibility, and selecting a suitable super-resolution reconstruction algorithm, such as SRCNN (super-resolution convolutional neural network) and ESPCN (efficient sub-pixel convolutional neural network). etc., preprocess the designed user's omnidirectional image, including normalization, denoising and resizing, to improve the reconstruction effect, use the selected super-resolution algorithm to process the preprocessed image to generate a higher resolution reconstructed image, evaluate the quality of the reconstructed image to ensure that its details are effectively improved, select a suitable deep learning model for semantic segmentation, such as U-Net, DeepLab or MaskR-CNN. If there is no ready-made model, collect labeled data for model training, use data enhancement technology to improve the generalization ability of the model, verify the segmentation effect of the model, adjust the hyperparameters to optimize the performance, input the super-resolution reconstructed image into the trained deep learning model, and perform deep visual semantic segmentation. Segmentation, identify the user's body parts information, save the segmentation results as a binary image or label map for subsequent processing, select a suitable key point detection algorithm, such as OpenPose, PoseNet or HRNet, input the segmented image into the key point detection model, identify the user's body parts key points (such as head, shoulders, elbows, knees, etc.), record the identified key point coordinates in a data structure (such as an array or dictionary), mark the identified body parts key points on the image to verify the accuracy of the recognition, determine the model for user bone structure analysis, usually based on the human anatomy model, build the user's bone structure according to the identified body parts key points, use connecting lines or 3D models Represent the skeleton, extract the characteristics of the skeleton structure, such as joint angles, limb proportions, etc., record them in the data structure for subsequent analysis, define the types of user geometric parameters, such as body width, length, volume, etc., use the previously extracted skeleton structure characteristics to calculate the user's geometric parameters, use 3D modeling software (such as Blender, MeshLab) or point cloud processing library (such as PCL) to convert the user's geometric parameters into a 3D point cloud model, process the point cloud (such as denoising, resampling), generate the user's 3D morphological model, verify the accuracy and completeness of the generated 3D morphological model, and store the 3D model in a suitable format (such as OBJ, PLY) for subsequent visualization and analysis.

[0039] In this embodiment, the specific steps of step S12 are: Perform filtering and denoising processing on the designed user's omnidirectional images one by one to obtain a filtered and denoised omnidirectional image; Perform global outlier scanning on the filtered and denoised omnidirectional image to extract image outliers; Perform abnormal outlier elimination on image outliers to obtain an outlier optimized omni-directional image; Perform image detail recognition on the outlier optimized omnidirectional image and extract image detail features; Perform magnification and distortion analysis on image detail features to obtain image magnification and distortion data; Based on the image magnification distortion data, the outlier optimized omnidirectional image is smoothed to obtain a smooth image of the detailed parts; Extract the original resolution point cloud of the smoothed image of the detail part to obtain the original image point cloud resolution; Perform deep convolution learning based on the original image point cloud resolution to obtain point cloud resolution convolution features; Based on the point cloud resolution convolution features, super-resolution reconstruction is performed on the smooth image of the detail parts to construct a super-resolution reconstructed image.

[0040] In this embodiment, a suitable image filtering algorithm is selected, such as Gaussian filtering, median filtering or bilateral filtering, to reduce image noise. The selected filtering algorithm is applied to each designed omnidirectional image of the user, and image noise reduction processing is performed one by one to ensure that the filtering parameters (such as window size, standard deviation, etc.) are adjusted according to the image characteristics to obtain the best noise reduction effect. The omnidirectional image after filtering and noise reduction is stored in a specified data structure for subsequent processing. A suitable outlier detection algorithm is selected, such as local outlier factor (LOF), isolation forest or threshold-based method, and global outlier scanning is performed on the omnidirectional image after filtering and noise reduction to identify outliers in the image that are inconsistent with other pixels. The outliers with large value differences are identified, and the positions and features of the outliers are recorded for subsequent processing. The criteria for removing abnormal outliers, such as the outlier degree threshold or the limit on the number of outliers, are determined. According to the set criteria, outliers are removed to generate an outlier-optimized omnidirectional image, and the integrity and validity of the image after removal are ensured. The omnidirectional image after outlier optimization is stored in a data structure for use in subsequent steps. An appropriate image detail extraction algorithm is selected, such as edge detection (Canny, Sobel) or texture analysis (LBP, Gabor filtering), and detail feature extraction is performed on the omnidirectional image after outlier optimization to obtain the key detail features of the image. The extracted detail features are stored in the data structure, and a suitable magnification distortion analysis method is selected, such as peak signal-to-noise ratio (PSNR), structural similarity (SSIM), etc., to evaluate the degree of distortion when the details are magnified, and the magnification distortion analysis is performed on the extracted image detail features to generate image magnification distortion data, and the distortion data is recorded for subsequent smoothing processing. A suitable point cloud smoothing algorithm is selected, such as Gaussian smoothing, bilateral smoothing or MLS (Moving Least Squares), and point cloud smoothing processing is performed on the outlier optimized omnidirectional image based on the image magnification distortion data to obtain a smoothed image of the detail part, ensuring that the smoothing processing does not lose important details, and a suitable A suitable point cloud extraction method is used, such as point cloud reconstruction based on depth images or 3D reconstruction algorithms, to extract point clouds of original resolution from smooth images of detail parts, to ensure that the point clouds contain sufficient detail information, to store the extracted point clouds in a suitable data structure, to select a suitable super-resolution reconstruction algorithm, such as SRCNN, ESPCN or GAN (generative adversarial network), to perform super-resolution reconstruction of smooth images of detail parts based on point cloud resolution convolution features, to generate super-resolution reconstructed images, to ensure that the details and clarity of the reconstructed images are improved, to store the generated super-resolution reconstructed images in a suitable format, and to perform quality assessment to ensure that the design requirements are met.

[0041] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: performing physical characteristic parameter analysis on multi-scale parameters of textile raw materials to extract physical characteristic parameters of textile raw materials; Step S22: Calculating the fiber mechanical properties of the physical characteristic parameters of the textile raw material, thereby obtaining the fiber mechanical property data of the textile raw material; Step S23: Calculating the inter-fiber friction coefficient based on the physical characteristic parameters of the textile raw materials to obtain the inter-fiber friction parameters; Step S24: Analyze the mechanical properties of the fiber bundle based on the physical characteristic parameters of the textile raw materials to obtain the mechanical property data of the fiber bundle; Step S25: Perform deep parameter feature mining on the mechanical property data of textile raw material fibers, the friction parameters between fibers, and the mechanical property data of fiber bundles, so as to obtain multi-scale state characteristics of textile raw materials.

[0042] In this embodiment, multi-scale parameters of textile raw materials are collected, including fiber diameter, density, strength, elongation, etc., and these data are sorted to ensure their integrity and accuracy for subsequent analysis. The physical characteristic parameters that need to be analyzed, such as elastic modulus, tensile strength, specific gravity, etc. of the fiber, are determined. A suitable analysis method is selected, including statistical analysis, regression analysis or multivariate analysis method. The multi-scale parameters of the textile raw materials are analyzed using the selected method to extract the physical characteristic parameters. The extracted physical characteristic parameters are recorded in a database or table for subsequent use. The key indicators of the mechanical properties of the fiber, such as tensile strength, shear modulus, elastic modulus, etc., are determined. Suitable calculation formulas and models are selected. Based on the material mechanics theory (such as Hooke's law, stress-strain relationship), the fiber mechanical properties data of the textile raw materials are calculated according to the physical characteristic parameters and the selected formula. The calculation results are stored in a data structure to ensure that they are convenient for subsequent analysis. The calculation method of the friction coefficient between fibers is determined, which is usually based on experimental data or theoretical models. If necessary, experiments can be designed to measure the friction between fibers. Based on the test data, the friction coefficient is calculated for the physical characteristic parameters of the textile raw materials using relevant formulas (such as Coulomb's friction law), the inter-fiber friction parameters are obtained, and the mechanical properties of the fiber bundles, such as tensile strength, compressive strength and stiffness, are determined. Appropriate analysis and calculation methods, such as finite element analysis (FEA) or macroscopic mechanical models, are selected. According to the physical characteristic parameters of the fibers and the selected methods, the mechanical properties of the fiber bundles are analyzed to generate the mechanical property data of the fiber bundles. The fiber mechanical property data, inter-fiber friction parameters and fiber bundle mechanical property data of the textile raw materials are integrated into a unified data set. Appropriate deep feature mining methods, such as principal component analysis (PCA), factor analysis or machine learning algorithms (such as random forests and support vector machines) are selected. Deep parameter feature mining is performed on the integrated data set to extract the multi-scale state characteristics of the textile raw materials. The mined features are recorded and visualized to ensure that they are easy to understand and analyze. The mined features are analyzed to verify their relevance and effectiveness to the actual textile performance. The results are organized into a report or summary for subsequent application and research.

[0043] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: performing external mechanical deformation response analysis on the multi-scale state characteristics of the textile raw material to generate external mechanical deformation response data of the textile raw material; Step S32: performing internal fiber bundle discrete analysis modeling based on multi-scale parameters of the textile raw material to obtain discrete fiber particles inside the textile raw material; Step S33: simulating the external load action on the discrete fiber particles inside the textile raw material to extract simulation data of the external load action on the discrete fiber particles; Step S34: performing particle-by-particle force balance calculation on the fiber discrete particle interaction simulation data, and extracting the force balance data of each fiber discrete particle; Step S35: Performing an internal fiber elastic response analysis based on the force balance data of each discrete fiber particle, thereby obtaining the internal fiber elastic response characteristics of the textile raw material.

[0044] In this embodiment, multi-scale state characteristic data of textile raw materials, including physical characteristics and mechanical property parameters, are collected, and a suitable mechanical model (such as an elastic model and a plastic model) is constructed to simulate the deformation response of the textile raw materials under the action of external forces. A predefined external load (such as tension, compression, and shear) is applied to the model to simulate the mechanical response under actual use conditions. Finite element analysis (FEA) software (such as ANSYS and ABAQUS) is used to perform external mechanical deformation response analysis, generate external mechanical deformation response data of the textile raw materials, record the response data, including stress, strain, and deformation, integrate multi-scale parameters of the textile raw materials, pay special attention to parameters related to the internal structure, such as fiber diameter, arrangement, and relative position, select a discrete analysis modeling method, such as modeling based on discrete element method (DEM) or particle system, use modeling software (such as MATLAB or specific DEM software) to construct a discrete model of fiber bundles inside the textile raw materials, obtain discrete fiber particles inside the textile raw materials, define the type and size of the external load applied to the discrete fiber particles, for example Such as uniform load or concentrated load, apply external load in discrete model, use numerical simulation method (such as finite element analysis or particle mechanics simulation) to simulate the external load, extract the simulation data of external load on discrete fiber particles, including stress distribution and deformation inside particles, determine the method of force balance calculation, usually based on Newtonian mechanics and statics principles, perform force balance calculation on each discrete fiber particle, consider the interaction force with surrounding particles and the applied external load, calculate the force on each particle, including tension, pressure and friction, etc., extract the force balance data of each discrete fiber particle, record it in the data structure for subsequent analysis, select a suitable elastic response model, usually based on Hooke's law or other material behavior models, based on the force balance data of each discrete fiber particle, perform internal fiber elastic response analysis, calculate elastic deformation and stress response, record the elastic response characteristics of each particle, including strain and stress distribution, organize the obtained internal fiber elastic response characteristics of textile raw materials into a report, and store it in a database for subsequent research and application.

[0045] In this embodiment, the specific steps of step S31 are: Based on the multi-scale state characteristics of textile raw materials, multi-frequency pressure application simulation is performed on textile raw materials to extract the pressure simulation response state parameters of textile raw materials; Identify fiber deformation characteristics of textile raw material pressure simulation response state parameters and generate fiber deformation characteristic data; Conduct stress-strain analysis on fiber deformation characteristic data to obtain stress-strain data of textile raw materials; Conduct strain response change analysis on stress-strain data of textile raw materials to obtain strain response change curve; The external mechanical deformation response analysis is performed on the strain response change curve to generate the external mechanical deformation response data of textile raw materials.

[0046] In this embodiment, based on the multi-scale state characteristics of the textile raw materials, a corresponding three-dimensional mechanical model is established, and software such as ANSYS or ABAQUS is used to define multi-frequency pressure conditions applied to the model, including pressure waveforms of different frequencies and amplitudes (such as sine waves, pulse waves, etc.), and these pressure conditions are applied in the model to simulate the pressure effect in the actual use environment. The simulation is run to record the response of the textile raw materials under the multi-frequency pressure, including stress, strain, displacement and other data, and the pressure simulation response state parameters of the textile raw materials are extracted and stored in a database. The simulated pressure response data are preprocessed to remove noise and outliers, and a suitable image processing or machine learning algorithm is selected, such as edge detection, feature extraction algorithm (such as SIFT or SURF), to identify fiber deformation characteristics, and the selected algorithm is applied to the simulated deformation data to extract fiber deformation feature data, such as deformation amount, deformation direction, etc., record and organize the identified fiber deformation characteristics, and collect the response parameters obtained by the pressure simulation. Force and strain data, ensure the integrity of the data, select a suitable stress-strain analysis method, usually based on the theory of material mechanics, perform stress-strain analysis on the fiber deformation characteristic data, calculate the stress-strain curve, extract the corresponding data points, record the obtained stress-strain data of the textile raw materials, including elastic modulus, yield strength and other parameters, integrate the stress-strain data, prepare for strain response change analysis, select a suitable analysis method, such as trend analysis or statistical regression analysis, analyze the change of strain under different pressure conditions, generate a strain response change curve, record the strain value and corresponding pressure value of each point, and prepare for external mechanical deformation response analysis based on the previous strain response change curve. Choose to use finite element analysis or other numerical simulation methods to evaluate the external mechanical deformation response, run the analysis, generate the deformation response data of the textile raw materials under the action of external mechanics, including maximum strain, stress distribution and deformation mode, record and summarize the obtained external mechanical deformation response data to support subsequent applications or research.

[0047] In this embodiment, step S4 includes the following steps: Step S41: performing a multi-stage textile process analysis on the textile process log, and extracting textile process data of each stage; Step S42: Performing process timing logic analysis on the textile process data of each stage to generate textile process timing logic; Step S43: performing a textile process simulation on the textile sweater based on the textile process sequential logic to obtain a textile sweater simulation model; Step S44: Based on the external mechanical deformation response data of the textile raw materials and the internal fiber elastic response characteristics of the textile raw materials, the textile sweater simulation model is dynamically mapped to construct a digital twin model of the textile sweater.

[0048] In this embodiment, textile process logs are collected to ensure that the data includes operation records, timestamps and related parameters of each stage. Different textile process stages (such as spinning, weaving, dyeing, etc.) are identified based on the timestamps and operation contents in the process logs. The textile process data of each stage, including the operation type, equipment used, material characteristics and time, are extracted. The extracted data are organized into a structured format (such as a table or database) for subsequent analysis. The basic elements of the process sequential logic are defined, including the process sequence, parallel operation conditions and dependencies. Suitable logic analysis tools and methods are selected, such as Petri nets, finite state machines or flowchart tools. The extracted process data are converted into a sequential logic model. The relationship and sequence between different processes are analyzed using the selected method. The generated textile process sequential logic is recorded and ensured to be visualized for easy understanding and communication. Based on the textile process sequential logic, the input parameters of the simulation model are defined, including process time, equipment characteristics and material characteristics. Select appropriate simulation software (such as AnyLogic, Arena or MATLAB) to establish a process simulation model of textile sweaters, build a process simulation model of textile sweaters in the selected simulation software, set the timing and logical relationship of the process, run the simulation, observe various dynamic changes in the textile process, record the simulation results, collect the external mechanical deformation response data and internal fiber elastic response characteristics of textile raw materials to ensure data integrity, define a dynamic textile deformation mapping model, determine how to map the external and internal response data to the simulation model of the textile sweater, use the simulation model combined with the response data to perform dynamic textile deformation mapping, generate the deformation characteristics of the textile sweater under different mechanical conditions, record the constructed digital twin model of the textile sweater to ensure that it can reflect the dynamic changes in actual production, verify the generated digital twin model to ensure that it is consistent with the actual process, and adjust the model parameters and mapping relationship according to the verification results to improve the accuracy and reliability of the model.

[0049] In this embodiment, the specific steps of step S5 are: Step S51: performing preset dyeing demand analysis on the textile process log to extract preset process dyeing demand data; Step S52: Calculate dye parameters for preset process dyeing requirement data and extract dye characteristic parameters; Step S53: performing dye penetration interaction mining on dye characteristic parameters based on multi-scale state characteristics of textile raw materials to generate textile raw material dye penetration interaction data; Step S54: performing dye diffusion evolution analysis on the textile raw material dye penetration interaction data to generate dynamic dye diffusion evolution data; Step S55: Perform dynamic color rendering on the digital twin model of the textile sweater according to the dynamic dye diffusion evolution data to construct a dynamic rendering digital twin model.

[0050] In this embodiment, records related to dyeing in the textile process log are collected, including dyeing time, temperature, type of dye and other process parameters, preset dyeing requirements are identified according to the process log, such as the required color, dyeing depth and uniformity requirements, data related to the preset dyeing requirements are extracted and organized into a structured format (such as a table or database) to ensure the accuracy and traceability of the data, key characteristic parameters of the dye, such as solubility, molecular weight, pH adaptability, durability, etc., are determined, and a suitable calculation method is selected, such as statistical analysis or theoretical model based on experimental data, to calculate dye parameters for the preset process dyeing requirement data, extract dye characteristic parameters, record the calculation results, ensure that they are convenient for subsequent use, integrate multi-scale state characteristic data of textile raw materials, including fiber structure and physical properties, construct an interaction model between dyes and textile raw materials, consider processes such as penetration, adsorption and diffusion, and use numerical simulation methods (such as finite element analysis or molecular dynamics simulation) to interactively mine dye characteristic parameters and multi-scale state characteristics to generate dye penetration. Through the interactive data, record and organize the generated data for subsequent analysis, select a suitable dye diffusion model, such as Fick's diffusion law or other related dynamic models, perform dye diffusion evolution analysis on the dye penetration interactive data, simulate the diffusion process of the dye under different conditions, generate dynamic dye diffusion evolution data, record information such as diffusion rate, concentration change and uniformity, verify the rationality of the analysis results, and compare them with the actual dyeing experimental data. If inconsistencies are found, adjust the model parameters to ensure that the existing digital twin model of the textile sweater has the necessary geometric and physical properties, select a suitable dynamic color rendering algorithm, including physically based rendering (PBR) or real-time rendering technology, and perform dynamic color rendering on the digital twin model of the textile sweater according to the dynamic dye diffusion evolution data. Simulate the distribution and changes of the dye on the fiber, generate a visual rendering result with dynamic effects, which can reflect the real-time changes in the dyeing process, evaluate the rendering results to ensure that they are consistent with the actual dyeing effect, and if necessary, adjust the rendering parameters to improve accuracy.

[0051] In this embodiment, the specific steps of step S6 are: Step S61: dynamically try on and simulate the user's three-dimensional morphological model based on the dynamic rendering digital twin model, thereby obtaining a textile sweater-user morphological simulation model; Step S62: performing fit analysis on the textile sweater-user form simulation model to extract sweater fit data; Step S63: calculating the tightness of the textile sweater-user morphology simulation model to generate the tightness of the user's sweater; Step S64: performing personalized user try-on feature analysis on the sweater fit data and the tightness of the user's sweater, and extracting personalized user try-on feature parameters; Step S65: Optimizing the parameters of the smart sweater based on the personalized user try-on feature parameters to obtain the personalized sweater optimization parameters; Step S66: Perform personalized process simulation optimization on the textile sweater-user form simulation model according to the personalized sweater optimization parameters, and construct an intelligent parameter optimized sweater model.

[0052] In this embodiment, the existing dynamic rendering digital twin model and the user's three-dimensional morphological model (such as the user's body data obtained through 3D scanning) are ensured, and a fitting simulation environment is built using 3D modeling and animation software (such as Blender, Maya, or Unity). The digital twin model of the textile sweater is combined with the user's morphological model, and a dynamic fitting simulation is run to observe the performance of the sweater on the user's three-dimensional morphology. The area in contact between the sweater and the user's body during the simulation is recorded to obtain a textile sweater-user morphological simulation model, and the simulation results are saved for subsequent analysis to determine the key parameters for fit analysis. , such as contact area, gap, pressure distribution, etc., select appropriate analysis tools and methods, such as finite element analysis (FEA) or contact analysis software, perform fit analysis on the textile sweater-user morphology simulation model, extract sweater fit data, record contact conditions, pressure distribution and potential friction problems, determine key indicators of tightness, such as elastic coefficient, deformation and applied external force, etc., select appropriate calculation methods based on material mechanics or statistical analysis, calculate the tightness of the user's sweater according to the geometric and mechanical data of the textile sweater-user morphology simulation model, record the calculation results, and convert the sweater The fit data and user tightness results are integrated to form a comprehensive data set. A suitable feature extraction method, such as principal component analysis (PCA) or cluster analysis, is selected to perform personalized user fitting feature analysis on the integrated data, and personalized user fitting feature parameters are extracted, such as optimal fit and tightness preference. The smart sweater parameters that need to be tuned are determined, such as size, shape, material selection, etc. A suitable optimization algorithm, such as genetic algorithm, particle swarm optimization or other machine learning methods, is selected. Based on the personalized user fitting feature parameters, the tuning algorithm is run to intelligently optimize the sweater parameters and generate personalized sweater optimization parameters. The personalized sweater optimization parameters are used to update the textile sweater-user morphology simulation model. A suitable process simulation software (such as AnyLogic, MATLAB or other industrial simulation tools) is selected to run personalized process simulation optimization. The performance changes of the sweater and the improvement of the user fit under the new parameters are observed. The simulation results are recorded, and a smart parameter optimization sweater model is constructed to ensure that it meets the personalized needs of users. The consistency between the optimized model and the actual user fitting experience is verified. If any discrepancy is found, adjustments and optimizations are made to ensure that the final model has strong adaptability and high comfort.

[0053] In this embodiment, a digital textile sweater model simulation system is provided, which is used to execute the digital textile sweater model simulation method as described above, including: The three-dimensional morphology module is used to obtain the user's omnidirectional image to be designed, the multi-scale parameters of textile raw materials and the textile process log; analyze the geometric morphology parameters of the user's omnidirectional image to be designed, and perform three-dimensional geometric point cloud modeling to construct the user's three-dimensional morphology model; The multi-scale state module is used to analyze the physical characteristic parameters of the multi-scale parameters of textile raw materials and conduct deep parameter feature mining to obtain the multi-scale state characteristics of textile raw materials; Mechanical deformation module, used to analyze external mechanical deformation response and internal fiber elastic response based on multi-scale state characteristics of textile raw materials, so as to generate external mechanical deformation response data of textile raw materials and internal fiber elastic response characteristics of textile raw materials; The textile deformation module is used to dynamically map the textile deformation of the textile process log according to the external mechanical deformation response data of the textile raw materials and the elastic response characteristics of the internal fibers of the textile raw materials, and to build a digital twin model of the textile sweater; The dynamic color rendering module is used to interactively mine dye penetration on the digital twin model of textile sweaters, and then perform dynamic color rendering to build a dynamic rendering digital twin model; The dynamic try-on simulation module is used to perform dynamic try-on simulation on the user's three-dimensional morphological model based on the dynamic rendering digital twin model, and to optimize the parameters of the smart sweater to build an intelligent parameter-optimized sweater model.

[0054] The present invention constructs a three-dimensional morphological model of the user through geometric morphological parameter analysis and three-dimensional geometric point cloud modeling of the user's omnidirectional image, which is helpful to better understand the user's physical characteristics and morphology, and provide a basis for subsequent design and simulation. By performing physical characteristic parameter analysis and deep parameter feature mining on multi-scale parameters of textile raw materials, multi-scale state characteristics of textile raw materials are obtained, which is helpful to deeply understand the characteristics and state of textile raw materials and provide more accurate data support in the design and production process. Based on the multi-scale state characteristics of textile raw materials, external mechanical deformation response analysis and internal fiber elastic response analysis are performed to generate deformation and elastic response data of textile raw materials, which is helpful to understand the deformation of textile materials under external mechanical action, and provide a basis for subsequent simulation and optimization. The external deformation response data and internal fiber elastic response characteristics of the textile process log are used to perform dynamic textile deformation mapping and build a digital twin model of the textile sweater. This makes it possible to more accurately simulate the performance of textiles under different deformation conditions and provide a reference for the design and production process. Through dye penetration interactive mining and dynamic color rendering, a dynamic rendering digital twin model is built, so that the design team can better preview and adjust the color effect of the sweater, understand the appearance of the sweater more intuitively, and improve design efficiency and product quality. Through the dynamic rendering digital twin model, the user's three-dimensional morphological model is tried on and simulated and intelligently tuned, and a sweater model with intelligent parameter optimization is built, which is helpful for personalized design and production, improves the fit and comfort of sweaters, and enhances user experience and product competitiveness.

[0055] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0056] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A model simulation method for a digital textile sweater, characterized in that: The following steps are involved: Step S1: obtaining a full-dimensional image of the user to be designed, multi-scale parameters of textile raw materials and textile process logs; analyzing the geometric morphological parameters of the full-dimensional image of the user to be designed, and performing three-dimensional geometric point cloud modeling to construct a three-dimensional morphological model of the user; Step S2: performing physical characteristic parameter analysis on multi-scale parameters of textile raw materials and performing deep parameter feature mining to obtain multi-scale state characteristics of textile raw materials; Step S3: performing external mechanical deformation response analysis and internal fiber elastic response analysis based on the multi-scale state characteristics of the textile raw material to generate external mechanical deformation response data of the textile raw material and internal fiber elastic response characteristics of the textile raw material; Step S4: Perform dynamic textile deformation mapping on the textile process log according to the external mechanical deformation response data of the textile raw materials and the internal fiber elastic response characteristics of the textile raw materials, and construct a digital twin model of the textile sweater; Step S5: Perform dye penetration interactive mining on the digital twin model of the textile sweater, and then perform dynamic color rendering to construct a dynamic rendering digital twin model; Step S6: Based on the dynamic rendering digital twin model, a dynamic try-on simulation is performed on the user's three-dimensional morphological model, and the intelligent sweater parameters are tuned to build an intelligent parameter optimized sweater model.

2. The model simulation method of digital textile sweater according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: obtaining a user's omnidirectional image to be designed, multi-scale parameters of textile raw materials, and a textile process log; Step S12: reconstructing the image details of the designed omnidirectional image of the user to construct a super-resolution reconstructed image; Step S13: performing deep visual semantic segmentation on the super-resolution reconstructed image to identify the user's body part information; Step S14: Identify key points of user body parts based on the user body part information and mark the key points of the body parts; Step S15: performing a user bone structure analysis based on key points of body parts to generate a user bone structure feature; Step S16: performing geometric morphological parameter analysis on the user's skeletal structural features to generate user geometric morphological parameters; The user's geometric parameters are used to perform three-dimensional geometric point cloud modeling to construct the user's three-dimensional shape model.

3. The model simulation method of digital textile sweater according to claim 2, characterized in that: The specific steps of step S12 are: Perform filtering and denoising processing on the designed user's omnidirectional images one by one to obtain a filtered and denoised omnidirectional image; Perform global outlier scanning on the filtered and denoised omnidirectional image to extract image outliers; Perform abnormal outlier elimination on image outliers to obtain an outlier optimized omni-directional image; Perform image detail recognition on the outlier optimized omnidirectional image and extract image detail features; Perform magnification and distortion analysis on image detail features to obtain image magnification and distortion data; Based on the image magnification distortion data, the outlier optimized omnidirectional image is smoothed to obtain a smooth image of the detailed parts; Extract the original resolution point cloud of the smoothed image of the detail part to obtain the original image point cloud resolution; Perform deep convolution learning based on the original image point cloud resolution to obtain point cloud resolution convolution features; Based on the point cloud resolution convolution features, super-resolution reconstruction is performed on the smooth image of the detail parts to construct a super-resolution reconstructed image.

4. The model simulation method of digital textile sweater according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: performing physical characteristic parameter analysis on multi-scale parameters of textile raw materials to extract physical characteristic parameters of textile raw materials; Step S22: Calculating the fiber mechanical properties of the physical characteristic parameters of the textile raw material, thereby obtaining the fiber mechanical property data of the textile raw material; Step S23: Calculating the inter-fiber friction coefficient based on the physical characteristic parameters of the textile raw materials to obtain the inter-fiber friction parameters; Step S24: Analyze the mechanical properties of the fiber bundle based on the physical characteristic parameters of the textile raw materials to obtain the mechanical property data of the fiber bundle; Step S25: Perform deep parameter feature mining on the mechanical property data of textile raw material fibers, the friction parameters between fibers, and the mechanical property data of fiber bundles, so as to obtain multi-scale state characteristics of textile raw materials.

5. The model simulation method of digital textile sweater according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: performing external mechanical deformation response analysis on the multi-scale state characteristics of the textile raw material to generate external mechanical deformation response data of the textile raw material; Step S32: performing internal fiber bundle discrete analysis modeling based on multi-scale parameters of the textile raw material to obtain discrete fiber particles inside the textile raw material; Step S33: simulating the external load action on the discrete fiber particles inside the textile raw material to extract simulation data of the external load action on the discrete fiber particles; Step S34: performing particle-by-particle force balance calculation on the fiber discrete particle interaction simulation data, and extracting the force balance data of each fiber discrete particle; Step S35: Performing an internal fiber elastic response analysis based on the force balance data of each discrete fiber particle, thereby obtaining the internal fiber elastic response characteristics of the textile raw material.

6. The model simulation method of digital textile sweater according to claim 5, characterized in that: The specific steps of step S31 are: Based on the multi-scale state characteristics of textile raw materials, multi-frequency pressure application simulation is performed on textile raw materials to extract the pressure simulation response state parameters of textile raw materials; Identify fiber deformation characteristics of textile raw material pressure simulation response state parameters and generate fiber deformation characteristic data; Conduct stress-strain analysis on fiber deformation characteristic data to obtain stress-strain data of textile raw materials; Conduct strain response change analysis on stress-strain data of textile raw materials to obtain strain response change curve; The external mechanical deformation response analysis is performed on the strain response change curve to generate the external mechanical deformation response data of textile raw materials.

7. The model simulation method of digital textile sweater according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: performing a multi-stage textile process analysis on the textile process log, and extracting textile process data of each stage; Step S42: Performing process timing logic analysis on the textile process data of each stage to generate textile process timing logic; Step S43: performing a textile process simulation on the textile sweater based on the textile process sequential logic to obtain a textile sweater simulation model; Step S44: Based on the external mechanical deformation response data of the textile raw materials and the internal fiber elastic response characteristics of the textile raw materials, the textile sweater simulation model is dynamically mapped to construct a digital twin model of the textile sweater.

8. The model simulation method of digital textile sweater according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: performing preset dyeing demand analysis on the textile process log to extract preset process dyeing demand data; Step S52: Calculate dye parameters for preset process dyeing requirement data and extract dye characteristic parameters; Step S53: performing dye penetration interaction mining on dye characteristic parameters based on multi-scale state characteristics of textile raw materials to generate textile raw material dye penetration interaction data; Step S54: performing dye diffusion evolution analysis on the textile raw material dye penetration interaction data to generate dynamic dye diffusion evolution data; Step S55: Perform dynamic color rendering on the digital twin model of the textile sweater according to the dynamic dye diffusion evolution data to construct a dynamic rendering digital twin model.

9. The model simulation method of digital textile sweater according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: dynamically try on and simulate the user's three-dimensional morphological model based on the dynamic rendering digital twin model, thereby obtaining a textile sweater-user morphological simulation model; Step S62: performing fit analysis on the textile sweater-user morphology simulation model to extract sweater fit data; Step S63: calculating the tightness of the textile sweater-user morphology simulation model to generate the tightness of the user's sweater; Step S64: performing personalized user try-on feature analysis on the sweater fit data and the tightness of the user's sweater, and extracting personalized user try-on feature parameters; Step S65: Optimizing the parameters of the smart sweater based on the personalized user try-on feature parameters to obtain the personalized sweater optimization parameters; Step S66: Perform personalized process simulation optimization on the textile sweater-user form simulation model according to the personalized sweater optimization parameters, and construct an intelligent parameter optimized sweater model.

10. A digital textile sweater model simulation system, characterized in that: The method for executing the model simulation of the digital textile sweater according to claim 1 comprises: The three-dimensional morphology module is used to obtain the user's omnidirectional image to be designed, the multi-scale parameters of textile raw materials and the textile process log; analyze the geometric morphology parameters of the user's omnidirectional image to be designed, and perform three-dimensional geometric point cloud modeling to construct the user's three-dimensional morphology model; The multi-scale state module is used to analyze the physical characteristic parameters of the multi-scale parameters of textile raw materials and conduct deep parameter feature mining to obtain the multi-scale state characteristics of textile raw materials; Mechanical deformation module, used to analyze external mechanical deformation response and internal fiber elastic response based on multi-scale state characteristics of textile raw materials, so as to generate external mechanical deformation response data of textile raw materials and internal fiber elastic response characteristics of textile raw materials; The textile deformation module is used to dynamically map the textile deformation of the textile process log according to the external mechanical deformation response data of the textile raw materials and the elastic response characteristics of the internal fibers of the textile raw materials, and to build a digital twin model of the textile sweater; The dynamic color rendering module is used to interactively mine dye penetration on the digital twin model of textile sweaters, and then perform dynamic color rendering to build a dynamic rendering digital twin model; The dynamic try-on simulation module is used to perform dynamic try-on simulation on the user's three-dimensional morphological model based on the dynamic rendering digital twin model, and to optimize the parameters of the smart sweater to build an intelligent parameter-optimized sweater model.

Images