Modular multi-functional integrated garment optimization method and system

Abstract

This invention relates to the field of clothing design technology, specifically to a modular, multifunctional integrated clothing optimization method and system. The method includes the following steps: acquiring body point cloud data, clothing piece attributes, and interface geometry data; constructing a human geometry and virtual clothing model; extracting areas covering the body; obtaining sweat flow and heat flux density by combining environmental temperature, humidity, and metabolic rate; constructing a humid-thermal flow field; calculating temperature and humidity normalized values; generating a comfort index; screening uncomfortable areas and replacing piece and interface parameters; and generating clothing configuration. In this invention, by integrating human geometry, physiological thermoregulation, humid-heat transfer simulation, and tactile evaluation, clothing design shifts from single-piece performance matching to overall microclimate verification. It can reveal the continuous influence of sweat, heat, and airflow between pieces and interfaces, identify areas of moisture retention, heat retention attenuation, and temperature differences, and adjust material configuration and connection forms accordingly to improve thermal and humid comfort and wearability.

Description

Technical Field

[0001] This invention relates to the field of clothing design technology, and in particular to a modular, multifunctional integrated clothing optimization method and system. Background Technology

[0002] The field of apparel design technology involves core aspects such as human body size acquisition, garment pattern construction, fabric and accessory selection, garment structural zoning, sewing and splicing methods, adaptation to wearing environment, and thermal and moisture comfort configuration. Its technical content usually revolves around the body characteristics, wearing scenarios, movement states, and functional needs of different groups of people, and designs the body, sleeves, shoulders, collar, placket, hem, and inner and outer layer structures. The overall composition of the garment is determined based on specific factors such as fabric thickness, weight, elasticity, abrasion resistance, breathability, moisture permeability, waterproof pressure, thermal insulation filling, seam type, zipper position, snap button arrangement, and Velcro fastening area.

[0003] The traditional modular multifunctional integrated clothing optimization method refers to addressing the combined needs of different wearers under specific environmental and activity conditions for functions such as warmth, windproofing, waterproofing, sweat wicking, abrasion resistance, heating, and easy replacement. This involves first collecting anthropometric data such as height, weight, chest circumference, waist circumference, shoulder width, arm length, back length, and leg length; then recording usage conditions such as temperature, humidity, wind speed, precipitation, altitude, activity duration, and exercise intensity; and finally selecting down filling sheets, fleece breathable sheets, waterproof hard shell sheets, abrasion-resistant reinforcing sheets, thin-film heating sheets, elastic adjustment sheets, and other components according to clothing areas such as chest, back, armpits, shoulders, elbows, waist, and cuffs. Windproof panels are combined based on indicators such as thermal insulation coefficient, air permeability, moisture permeability, waterproof pressure, abrasion resistance, weight per unit area, and power supply specifications of each panel. Then, zippers, snaps, Velcro, sealing strips, stitching, and binding tape are used to form a complete garment solution. Traditionally, each functional panel is matched as an independent object, selected based on the material parameters of the individual panel and the preset functions of the garment area. Connectors are mainly used as fixing, disassembly, and splicing structures. When the solution is determined, the panel position, material specifications, connection direction, overlap width, opening and closing path, and production process parameters are recorded.

[0004] Traditional optimization methods rely primarily on regional functions and individual material parameters for matching during the configuration process. Human body data, usage conditions, component performance, and connection methods are recorded in a scattered manner. It is difficult to form a continuous judgment on the heat, sweat, and airflow transfer relationships between functional components. Connectors are often limited to fixed and detachable structures. The effects of breathability barrier, moisture retention, and heat retention reduction are not included in the overall evaluation. Wearers are prone to localized dampness, temperature differences, and a significant decrease in overall comfort during high-intensity activities, resulting in configuration results that deviate from actual needs. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a modular, multifunctional integrated clothing optimization method and system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a modular multifunctional integrated clothing optimization method, comprising the following steps: S1: Obtain the target user's body point cloud data, clothing preset module attribute data, and clothing connection interface geometric structure data; construct a three-dimensional geometric model of the human body based on the body point cloud data; and generate a three-dimensional virtual clothing model based on the clothing preset module attribute data and clothing connection interface geometric structure data. S2: Based on the three-dimensional virtual clothing model, extract the local body area it covers, and combine it with preset exercise environment temperature data, preset exercise environment humidity data and exercise metabolic rate parameters to obtain local sweat flow data and local heat flux density data through the Gagge two-node physiological thermoregulation model. S3: Based on the local sweat flow rate data and local heat flux density data, establish thermodynamic boundary conditions, and combine the preset three-dimensional space air density parameters, air isobaric specific heat capacity parameters, air velocity parameters, air thermal conductivity parameters, water vapor mass fraction data and water vapor diffusion coefficient parameters to construct the distribution characteristics of the humid and hot air diffusion field, and calculate and obtain the normalized value of local skin temperature and the normalized value of local relative humidity. S4: Call the preset heat and moisture transfer simulation parameters, and calculate and obtain the local comfort index data based on the local skin temperature normalization value and the local relative humidity normalization value, combined with the preset sensory comfort neutral point temperature normalization value and the preset sensory comfort neutral point relative humidity normalization value. S5: Input the local comfort index data into the preset prediction average voting model, analyze the body sensation index of local areas, and filter the target clothing discomfort area based on the body sensation index, replace the corresponding clothing preset clothing module attribute data and clothing connection interface geometric structure data, and generate target clothing configuration information.

[0007] The present invention is improved in that the steps for obtaining the three-dimensional virtual clothing model are as follows: S111: Collect body point cloud data of the target user, extract the three-dimensional coordinate array of the body point cloud data, calculate the normal vector of the three-dimensional coordinate array, construct a spatial surface fitting equation based on the three-dimensional coordinate array and the normal vector, call the spatial surface fitting equation to connect the three-dimensional coordinate array to construct a polygonal topology network, and establish a three-dimensional geometric model of the human body. S112: Obtain the attribute data of the clothing module preset in the preset configuration file and the geometric structure data of the clothing connection interface. Based on the human body three-dimensional geometric model, extract the coordinate set of the outer boundary of the body surface. Calculate the position deviation matrix between the vertex coordinates in the geometric structure data of the clothing connection interface and the coordinate set of the outer boundary of the body surface. Call the position deviation matrix to perform mesh vertex assignment transformation on the material attribute matrix in the attribute data of the clothing module preset to obtain the spatial mapping distribution characteristics. S113: Extract the surface mesh coordinate set and the underlying material matrix based on the spatial mapping distribution features. Perform smooth interpolation continuous operation on the surface mesh coordinate set to output a smooth surface vertex sequence. Map the underlying material matrix to the smooth surface vertex sequence to construct material binding feature nodes. Perform spatial topology closure operation on the outer boundary nodes of the material binding feature nodes to generate a three-dimensional virtual clothing model.

[0008] The present invention is improved in that the step of obtaining the local heat flux density data is specifically as follows: S211: Extract three-dimensional mesh patches from the surface of the three-dimensional virtual clothing model, calculate the coordinates of the three-dimensional mesh patches and associate them with the human body spatial coordinate set, extract local areas of the body based on the human body spatial coordinate set, map the local areas of the body as thermodynamic nodes within the spatial network, and establish a partitioned thermodynamic node matrix. S212: For the partitioned thermal node matrix, call the preset exercise environment temperature data, preset exercise environment humidity data, and exercise metabolic rate parameters, and calculate the difference between the heat production flux associated with the exercise metabolic rate parameters and the heat transfer flux associated with the partitioned thermal node matrix through the preset Gagge two-node physiological thermoregulation model. Calculate the weighted sum of the heat transfer flux difference, the convective heat transfer coefficient of the preset exercise environment temperature data, and the evaporation coefficient of the preset exercise environment humidity data to obtain the heat dissipation demand data corresponding to each partitioned thermal node. S213: Call the heat dissipation demand data, extract the latent heat of vaporization demand of each of the partitioned thermal nodes, calculate the ratio of the latent heat of vaporization demand of the node to the latent heat of water vaporization constant, construct a liquid precipitation rate distribution matrix, perform time integration operation on the elements in the liquid precipitation rate distribution matrix to extract the mass flow rate scalar, perform transformation multiplication operation on the elements of the liquid precipitation rate distribution matrix to extract heat flux data, and generate local sweat flow data and local heat flux density data.

[0009] The improvement of this invention is that the process of performing time integration operation on the elements in the liquid precipitation rate distribution matrix to extract the mass flow rate scalar, and performing transformation multiplication operation on the elements of the liquid precipitation rate distribution matrix to extract heat flux data, specifically includes: Calculate the time span difference between the preset start time node and the end time node of the movement, and use it as the preset environmental time span benchmark value; An integration time window constraint is constructed based on the preset environmental time span benchmark value. The integration time window constraint is that the motion start time node is the lower limit of integration and the motion end time node is the upper limit of integration. Based on the integral time window constraint, the elements in the liquid precipitation rate distribution matrix are accumulated over time to obtain the mass flow rate scalar. Calculate the product of the liquid specific heat capacity parameter and the standard atmospheric pressure constant, and use it as the liquid enthalpy conversion coefficient; multiply the elements in the liquid precipitation rate distribution matrix with the liquid enthalpy conversion coefficient to obtain a preliminary heat flow data matrix; Calculate the ratio of surface wind speed parameter to air density parameter, and multiply this ratio by a preset mapping ratio constant as the surface convective heat transfer weight; The heat flux distribution matrix is ​​obtained by multiplying the preliminary heat flux data matrix with the surface convection heat transfer weight, and the heat flux distribution matrix is ​​used as the corresponding local heat flux density data.

[0010] The present invention is improved in that the steps for obtaining the normalized values ​​of local skin temperature and local relative humidity are specifically as follows: S311: Call the preset heat and moisture transfer simulation parameters, which include three-dimensional air density parameters, air specific heat capacity at constant pressure parameters, air velocity parameters, air thermal conductivity parameters, water vapor mass fraction data, water vapor diffusion coefficient parameters, and temperature distribution maximum limit parameters. Use the local sweat flow data and local heat flux density data to assign energy exchange flux to the mesh nodes on the surface of the three-dimensional virtual clothing model and establish thermodynamic boundary conditions. S312: Map the three-dimensional spatial air density parameters, air isobaric specific heat capacity parameters, air velocity parameters, air thermal conductivity parameters, water vapor mass fraction data, and water vapor diffusion coefficient parameters to thermodynamic boundary conditions to solve multiphase fluid dynamics transfer problems, extract the fluid momentum and heat transfer scalar sets of each voxel node in space, and use the spatial vector field distribution composed of the fluid momentum and heat transfer scalar sets of each voxel node as the distribution characteristics of the humid and hot air diffusion flow field; S313: Invoke the distribution characteristics of the humid and hot airflow diffusion field, locate the interface grid nodes of the human body geometric boundary mapping flow field, extract the absolute temperature value sequence of the interface grid nodes, generate local skin temperature data, extract the water vapor mass fraction value of the interface grid nodes, obtain the preset local environment saturated water vapor mass fraction constant, calculate the division quotient of the water vapor mass fraction value of the interface grid nodes and the local environment saturated water vapor mass fraction constant, generate the local relative humidity normalized value, call the maximum limit parameter of temperature distribution, perform a division ratio operation on the local skin temperature data and the maximum limit parameter of temperature distribution to eliminate physical dimensions, obtain the skin temperature limit ratio, extract the discrete numerical elements within the skin temperature limit ratio, linearly map the discrete numerical elements to the zero to one standard dimensionless interval, and generate the local skin temperature normalized value.

[0011] The present invention is improved in that the step of obtaining the local comfort index data is specifically as follows: S411: Based on the local area of ​​the body, call the preset sensory comfort neutral point temperature normalization value and the preset sensory comfort neutral point relative humidity normalization value, calculate the temperature deviation between the local skin temperature normalization value and the sensory comfort neutral point temperature normalization value, and calculate the humidity deviation between the local relative humidity normalization value and the sensory comfort neutral point relative humidity normalization value. S412: Extract the temperature deviation and the humidity deviation, and calculate the local comfort index data according to the preset sensory temperature weighting coefficient and the preset sensory humidity weighting coefficient.

[0012] The present invention is improved in that the formula for calculating and obtaining the local comfort index data is specifically as follows: ; in, Represents local comfort index data. This represents the preset sensory temperature weighting coefficient. Represents the normalized value of local skin temperature. The normalized value of the neutral point temperature representing sensory comfort. This represents the preset sensory humidity weighting coefficient. This represents the normalized value of local relative humidity. The normalized relative humidity value represents the sensory comfort neutral point.

[0013] The present invention is improved in that the step of obtaining the target clothing configuration information is specifically as follows: S511: Call the local comfort index data and the local body region, extract the three-dimensional network node parameters of the local body region, input the local comfort index data into the preset prediction average voting model to extract the network hidden layer weight matrix, use the network hidden layer weight matrix to perform nonlinear feature activation operation on the local comfort index data to obtain the body sensation index, set a comfort lower limit benchmark threshold for the body sensation index, compare the size relationship between the body sensation index and the comfort lower limit benchmark threshold, extract the three-dimensional network node parameters corresponding to the lower limit benchmark threshold and splice them to construct an abnormal representation patch to generate the target clothing discomfort area; S512: Call the attribute data of the replacement clothing module and the configuration data of the replacement connection interface in the preset material database. For the discomfort area of ​​the target clothing, call the attribute data of the replacement clothing module to rewrite the underlying material feature vector of the preset clothing module attribute data to obtain the adjusted module attribute data. Call the configuration data of the replacement connection interface to cover the edge stitching point coordinates of the geometric structure data of the clothing connection interface to obtain the adjusted connection structure data. Map the adjusted module attribute data to the adjusted connection structure data and perform node feature binding operation to construct a rendering parameter set to generate the target clothing configuration information.

[0014] The present invention is improved in that the method for obtaining the lower limit benchmark threshold of comfort is as follows: Subjective thermal comfort voting scores and corresponding historical objective prediction index sets of multiple groups of subjects under different environmental conditions are extracted from a pre-set sample test database. Based on the label boundaries representing comfort and discomfort in the voting scores, they are divided into different evaluation data clusters. A mapping distribution curve is constructed between the numerical sequence of the historical objective prediction index set and the frequency of occurrence of the evaluation data cluster representing discomfort corresponding to each value in the sequence. The inflection point position where the slope of the mapping distribution curve changes non-linearly is calculated. The historical objective prediction index value corresponding to the inflection point position is extracted, and this value is defined as the critical value. This critical value is then calibrated as the lower limit benchmark threshold for comfort.

[0015] A modular, multifunctional integrated clothing optimization system, wherein the modular, multifunctional integrated clothing optimization system is used to implement the above-mentioned modular, multifunctional integrated clothing optimization method, the system comprising: The morphological modeling module acquires the target user's body point cloud data, clothing preset module attribute data, and clothing connection interface geometric structure data. It constructs a three-dimensional geometric model of the human body based on the body point cloud data and generates a three-dimensional virtual clothing model based on the clothing preset module attribute data and clothing connection interface geometric structure data. The physiological response analysis module extracts the local body area covered by the three-dimensional virtual clothing model, and combines the preset exercise environment temperature data, preset exercise environment humidity data and exercise metabolic rate parameters to obtain local sweat flow data and local heat flux density data through the Gagge two-node physiological thermoregulation model. The microenvironment coupling module establishes thermodynamic boundary conditions based on the local sweat flow rate data and local heat flux density data, and constructs the distribution characteristics of the humid and hot air diffusion field by combining preset three-dimensional spatial air density parameters, air isobaric specific heat capacity parameters, air flow velocity parameters, air thermal conductivity parameters, water vapor mass fraction data and water vapor diffusion coefficient parameters, and calculates and obtains the normalized value of local skin temperature and the normalized value of local relative humidity. The perception evaluation module calls the preset heat and humidity transfer simulation parameters, and calculates and obtains the local comfort index data based on the local skin temperature normalization value and the local relative humidity normalization value, combined with the preset sensory comfort neutral point temperature normalization value and the preset sensory comfort neutral point relative humidity normalization value. The adaptation and optimization module inputs the local comfort index data into a preset prediction average voting model, analyzes the body sensation index of local areas, and filters the target clothing discomfort areas based on the body sensation index. It then replaces the corresponding clothing preset clothing module attribute data and clothing connection interface geometric structure data to generate target clothing configuration information.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, the basic framework for linking human geometry and virtual clothing is constructed based on body point clouds, clothing piece attributes, and interface geometry. Environmental temperature and humidity conditions and exercise metabolic rate are converted into local sweat flow and heat flux density. Thermodynamic boundary conditions are linked with parameters such as air velocity, heat conduction, and water vapor diffusion to form a humid heat flow field calculation. Local skin temperature and relative humidity are converted into comfort index and included in the tactile evaluation. This can identify uncomfortable areas and simultaneously adjust piece attributes and interface morphology, so that the heat and moisture transfer path can be continuously expressed, reducing the risk of moisture accumulation and heat preservation attenuation, reducing local temperature differences and the impact of seam barriers, and improving the matching degree between configuration and actual wearing needs. Attached Figure Description

[0017] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system module diagram of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] Please see Figure 1 This invention provides a technical solution, a modular multifunctional integrated clothing optimization method, comprising the following steps: S1: Obtain the target user's body point cloud data, clothing preset module attribute data, and clothing connection interface geometric structure data; construct a three-dimensional geometric model of the human body based on the body point cloud data; and generate a three-dimensional virtual clothing model based on the clothing preset module attribute data and clothing connection interface geometric structure data. S2: Based on the three-dimensional virtual clothing model, extract the local body area it covers, and combine it with preset exercise environment temperature data, preset exercise environment humidity data and exercise metabolic rate parameters to obtain local sweat flow data and local heat flux density data through the Gagge two-node physiological thermoregulation model. S3: Establish thermodynamic boundary conditions based on local sweat flow data and local heat flux density data, and construct the distribution characteristics of the humid and hot air diffusion field by combining preset three-dimensional air density parameters, air isobaric specific heat capacity parameters, air velocity parameters, air thermal conductivity parameters, water vapor mass fraction data and water vapor diffusion coefficient parameters, and calculate and obtain the normalized value of local skin temperature and the normalized value of local relative humidity. S4: Call the preset heat and moisture transfer simulation parameters, and calculate and obtain the local comfort index data based on the local skin temperature normalization value and the local relative humidity normalization value, combined with the preset sensory comfort neutral point temperature normalization value and the preset sensory comfort neutral point relative humidity normalization value. S5: Input the local comfort index data into the preset prediction average voting model, analyze the body sensation index of local areas, and filter the target clothing discomfort area based on the body sensation index. Replace the corresponding clothing preset clothing module attribute data and clothing connection interface geometric structure data to generate target clothing configuration information.

[0020] The specific steps for obtaining a 3D virtual clothing model are as follows: S111: Collect body point cloud data of the target user, extract the three-dimensional coordinate array of the body point cloud data, calculate the normal vector of the three-dimensional coordinate array, construct a spatial surface fitting equation based on the three-dimensional coordinate array and the normal vector, call the spatial surface fitting equation to connect the three-dimensional coordinate array to construct a polygonal topology network, and establish a three-dimensional geometric model of the human body. A high-precision 3D laser scanning device is used to acquire the geometric reflection signals of the target user's body surface in a standing posture. The optical signals are then converted into body point cloud data in a discrete point set format within a 3D spatial coordinate system. This body point cloud data is imported into GeomagicStudio 3D data processing software. A median filtering denoising algorithm is used to smooth the data within the discrete point set format, removing abnormal drift points whose spatial straight-line distance deviates from the main point cloud cluster by more than ten millimeters. After denoising, the coordinate values ​​of the body point cloud data on the three orthogonal spatial axes (X-axis, Y-axis, and Z-axis) are extracted to form a 3D coordinate array. This 3D coordinate array is traversed, and principal component analysis is used to obtain the vertical direction vector of the spatial plane formed by every three adjacent coordinate points, which is defined as the normal vector. In the software interface, a least-squares fitting function is called, substituting the 3D coordinate array and the corresponding normal vector to solve for the polynomial coefficients, thereby constructing a continuous spatial surface fitting equation. Based on the spatial surface fitting equation, the Delaunay triangulation method is used to connect three-dimensional coordinate points with a straight-line distance of less than five millimeters in space to form continuous triangular polygonal mesh units. All mesh units are combined to form a polygonal topology network, and finally a full-scale three-dimensional geometric model of the human body is output for subsequent physical simulation.

[0021] S112: Obtain the attribute data of the clothing module preset in the preset configuration file and the geometric structure data of the clothing connection interface. Extract the coordinate set of the outer boundary of the body surface based on the three-dimensional geometric model of the human body. Calculate the position deviation matrix between the vertex coordinates in the geometric structure data of the clothing connection interface and the coordinate set of the outer boundary of the body surface. Call the position deviation matrix to perform mesh vertex assignment transformation on the material attribute matrix in the attribute data of the clothing module preset to obtain the spatial mapping distribution characteristics. Next, a pre-configured configuration file is read from a local MySQL database. This configuration file is pre-entered by the user through the front-end interface. The pre-configured clothing module attribute data records the thickness and density values ​​of each clothing block, while the clothing connection interface geometric structure data records the set of vector coordinates of the seam points at the edges of each clothing component. The established 3D human body geometric model is imported into the CLO3D virtual clothing modeling software, and the absolute spatial coordinates of all vertices on the outermost contour surface of the human body model are extracted to form the outer boundary coordinate set of the body surface. The Euclidean distance difference along the normal direction between the spatial coordinates of each vertex in the clothing connection interface geometric structure data and the spatial coordinates of the nearest neighbor corresponding point in the outer boundary coordinate set of the body surface is calculated, and all differences are arranged in vertex index order to form a position deviation matrix. The numerical ranges within the position deviation matrix are evaluated one by one. For values ​​with a difference greater than zero, it indicates that the clothing deviates from the human body surface at that point, creating a gap; the original value is preserved. For values ​​with a difference equal to zero, it indicates that the clothing fits the human body perfectly at that point; the original value is also preserved. For values ​​with a difference less than zero, it indicates that the clothing model has penetrated the human body; the coordinates are shifted outward along the human body normal until the difference equals zero to prevent penetration. The position deviation matrix, after the above evaluation, is then called. Using matrix multiplication, each element in the matrix is ​​multiplied by the corresponding element in the material thickness attribute matrix within the clothing module's preset attribute data. The product is assigned to the corresponding mesh vertex to complete the mesh vertex assignment transformation, generating a spatial mapping distribution characteristic reflecting the thickness deformation state of the material at various points on the human body surface.

[0022] S113: Extract the surface mesh coordinate set and the underlying material matrix based on the spatial mapping distribution features. Perform smooth interpolation continuous operation on the surface mesh coordinate set to output a smooth surface vertex sequence. Map the underlying material matrix to the smooth surface vertex sequence to construct material binding feature nodes. Perform spatial topology closure operation on the outer boundary nodes of the material binding feature nodes to generate a three-dimensional virtual clothing model. In virtual clothing modeling software, the coordinates of all grid vertices on the outermost surface are extracted to form a surface grid coordinate set based on the aforementioned spatial mapping distribution characteristics. Simultaneously, the density parameters of each grid node in the inner skin layer are extracted to form a bottom-layer material matrix. A cubic spline polynomial algorithm is used to perform smooth interpolation continuous operations on the surface grid coordinate set, filling coordinate gaps between existing discrete vertices by inserting four new vertices per square centimeter, outputting a smooth surface vertex sequence. The density parameters of each node in the bottom-layer material matrix are assigned one-to-one according to spatial position to the corresponding nodes in the smooth surface vertex sequence, constructing material binding feature nodes with physical properties. The outermost adjacent boundary nodes of the material binding feature nodes are traversed, their relative spatial straight-line distance is calculated, the distance range is determined, and boundary nodes with a spatial distance of less than one millimeter are forcibly set to the same coordinate position to merge and complete the spatial topological closure operation, eliminating physical gaps in the clothing model and outputting a 3D virtual clothing model with complete physical properties and a continuous surface.

[0023] The specific steps for obtaining local heat flux density data are as follows: S211: Extract three-dimensional mesh patches from the surface of a three-dimensional virtual clothing model, calculate the coordinates of the three-dimensional mesh patches and associate them with the human body spatial coordinate set, extract local areas of the body based on the human body spatial coordinate set, map the local areas of the body into adjustable thermodynamic nodes in the spatial network, and establish a partitioned thermodynamic node matrix. For the output 3D virtual clothing model, all 3D mesh patches generated by Delaunay triangulation are extracted from its surface. The absolute coordinates of the centroid points of each 3D mesh patch in the global world coordinate system are calculated to form a mesh patch coordinate set, which is then spatially aligned with the human body spatial coordinate set of the established 3D human geometric model. The coordinate group with the Z-axis height between the xiphoid process of the sternum and the acromion and the X-axis coordinate located on the left side of the human body's central plane is extracted from the human body spatial coordinate set as the left axillary region. The spatial parameters of this left axillary region are imported into the ANSYS Workbench simulation software. The mesh vertices in this region are converted into thermal energy control points in the heat transfer module for calculating energy exchange, defined as regulating thermodynamic nodes, and arranged in a row-column grid format according to the 3D coordinate order to establish a partitioned thermodynamic node matrix.

[0024] S212: For the partitioned thermal node matrix, the preset exercise environment temperature data, preset exercise environment humidity data, and exercise metabolic rate parameters are called. The difference between the heat production flux associated with the exercise metabolic rate parameters and the heat transfer flux associated with the partitioned thermal node matrix is ​​calculated through the preset Gagge two-node physiological thermoregulation model. The weighted sum of the heat transfer flux difference and the convective heat transfer coefficient of the preset exercise environment temperature data and the evaporation coefficient of the preset exercise environment humidity data is calculated to obtain the heat dissipation demand data corresponding to each partitioned thermal node. For the aforementioned zoned thermal node matrix, digital temperature and humidity sensors deployed in the constant-temperature test chamber collect ambient dry-bulb temperature values ​​as preset exercise environment temperature data, and relative humidity percentages as preset exercise environment humidity data. A metabolic monitor attached to the user's wrist collects heart rate data, which is converted to exercise metabolic rate parameters using a standard metabolic equivalent conversion table. In the simulation software, the Gagge two-node physiological thermoregulation model is used. The difference between the exercise metabolic rate parameters and the human basal metabolic constant is calculated to obtain the metabolic heat flux as the associated heat production flux. The heat transfer flux associated with the zoned thermal node matrix is ​​obtained by summing the radiative heat loss from the human skin surface and the heat loss from respiratory airflow. The heat transfer flux difference is calculated as: associated heat production flux - associated heat transfer flux. Heat transfer parameters are set for the test environment. The convective heat transfer coefficient is set to 4.5 based on an empirical formula using a fixed wind speed within the test chamber; a higher value indicates a stronger ability of airflow to remove surface heat. The evaporation coefficient is set to 12.8 based on the standard moisture resistance parameter of the test clothing material; a higher value indicates a stronger ability of water evaporation to convert into latent heat. Then, the heat dissipation demand of each node is calculated using the formula: Heat dissipation demand data = heat transfer flux difference + preset motion environment temperature data × convective heat transfer coefficient + preset motion environment humidity data × evaporation coefficient, thus obtaining the heat dissipation demand data corresponding to each thermal node in the partition.

[0025] S213: Call the heat dissipation demand data, extract the latent heat of vaporization demand of each thermal node in the zone, calculate the ratio of the latent heat of vaporization demand of the node to the latent heat of water vaporization constant, construct the liquid precipitation rate distribution matrix, perform time integration operation on the elements in the liquid precipitation rate distribution matrix to extract the mass flow rate scalar, perform transformation multiplication operation on the elements of the liquid precipitation rate distribution matrix to extract heat flux data, and generate local sweat flow data and local heat flux density data.

[0026] The process of extracting mass flow rate scalars by performing time integration on the elements of the liquid precipitation rate distribution matrix and extracting heat flux data by performing transformation multiplication on the elements of the liquid precipitation rate distribution matrix is ​​as follows: Calculate the time span difference between the preset start time node and the end time node of the movement, and use it as the preset environmental time span benchmark value; The integration time window constraint is constructed based on the preset environmental time span benchmark value. The integration time window constraint is that the lower limit of integration is the starting time node of the movement and the upper limit of integration is the ending time node of the movement. Based on the integral time window constraint, the elements in the liquid precipitation rate distribution matrix are accumulated over time to obtain the mass flow rate scalar. Calculate the product of the liquid specific heat capacity parameter and the standard atmospheric pressure constant, and use it as the liquid enthalpy conversion coefficient; multiply the elements in the liquid precipitation rate distribution matrix with the liquid enthalpy conversion coefficient to obtain the preliminary heat flow data matrix; Calculate the ratio of surface wind speed parameter to air density parameter, and multiply this ratio by a preset mapping ratio constant as the surface convective heat transfer weight; The heat flux distribution matrix is ​​obtained by multiplying the preliminary heat flux data matrix with the surface convection heat transfer weights, and the heat flux distribution matrix is ​​used as the corresponding local heat flux density data.

[0027] Heat dissipation demand data corresponding to each thermal node in the simulation software are extracted, and the heat value used for water vaporization is separated as the latent heat of vaporization demand. The liquid precipitation rate distribution matrix is ​​calculated as: node latent heat of vaporization demand / water vaporization latent heat constant. The time sequence from the start to the end of exercise is recorded using a sports physiology monitor, and the pre-recorded start and end times of exercise are extracted. The preset environmental time span baseline value is calculated as: exercise end time - exercise start time. Based on this baseline value, an integral time window constraint is constructed, with the exercise start time as the lower limit of integration and the exercise end time as the upper limit of integration. Based on the integral time window constraint, the elements in the liquid precipitation rate distribution matrix are accumulated over a time dimension in one-second increments to obtain the total mass loss of water within that time period. Since this accumulated value represents the overall mass loss scale within a specific integral time window, it is defined here and output as a mass flow rate scalar. The specific heat capacity parameter of room-temperature liquid water is retrieved, and the environmental standard atmospheric pressure constant is read. In the modeling of a constant-pressure thermodynamic system, the standard atmospheric pressure constant is abstracted as a dimensionless pressure correction weighting factor to participate in the calculation. The liquid enthalpy conversion coefficient is calculated using the formula: Liquid enthalpy conversion coefficient = Liquid specific heat capacity parameter × Corrected standard atmospheric pressure constant. The preliminary heat flow data matrix is ​​calculated using the formula: Preliminary heat flow data matrix = Liquid precipitation rate distribution matrix × Liquid enthalpy conversion coefficient. The surface wind speed parameter measured by the anemometer and the local air density parameter are retrieved to calculate the surface convective heat transfer weight using the formula: Surface convective heat transfer weight = Surface wind speed parameter / Air density parameter × Preset mapping ratio constant. The preset mapping ratio constant is set to 0.023 based on the empirical value of surface friction based on Reynolds number and Prandtl number. The heat flux distribution matrix is ​​calculated using the formula: Heat flux distribution matrix = preliminary heat flux data matrix × surface convection heat transfer weight. Finally, the heat flux distribution matrix is ​​recorded in the software to generate the corresponding local heat flux density data, and the mass flow rate scalar is recorded to generate local sweat flow rate data.

[0028] The specific steps for obtaining the normalized values ​​of local skin temperature and local relative humidity are as follows: S311: Call the preset heat and moisture transfer simulation parameters, which include three-dimensional air density parameters, air specific heat capacity at constant pressure parameters, air velocity parameters, air thermal conductivity parameters, water vapor mass fraction data, water vapor diffusion coefficient parameters, and temperature distribution maximum limit parameters. Use local sweat flow data and local heat flux density data to assign energy exchange flux to the surface mesh nodes of the three-dimensional virtual clothing model and establish thermodynamic boundary conditions. The following parameters were retrieved from the fluid dynamics property database: the ambient air mass constant per cubic meter as the air density parameter in three-dimensional space; the heat capacity constant under constant pressure as the specific heat capacity parameter at constant pressure; the directional air flow velocity as the air velocity parameter; the heat conduction ratio constant as the air thermal conductivity parameter; the water vapor mass ratio in the air as the water vapor mass fraction data; the diffusion rate constant of water molecules in the air medium as the water vapor diffusion coefficient parameter; and the maximum temperature that human skin can withstand as the maximum temperature distribution limit parameter. Using local sweat flow rate data and local heat flux density data, specific mass flow rate and heat flow rate values ​​were assigned to each mesh node on the outermost surface of the three-dimensional virtual clothing model in the ANSYS Fluent fluid dynamics module to generate energy exchange flux, thereby constructing the third type of convective heat transfer conditions at the nodes and establishing thermodynamic boundary conditions.

[0029] S312: The three-dimensional air density parameters, air specific heat capacity parameters at constant pressure, air velocity parameters, air thermal conductivity parameters, water vapor mass fraction data, and water vapor diffusion coefficient parameters are mapped to thermodynamic boundary conditions to solve the multiphase fluid dynamics transfer problem. The fluid momentum and heat transfer scalar sets of each voxel node in space are extracted, and the spatial vector field distribution composed of the fluid momentum and heat transfer scalar sets of each voxel node is used as the distribution characteristics of the humid and hot air diffusion flow field. The parameters of air density, specific heat capacity, velocity, thermal conductivity, water vapor mass fraction, and diffusion coefficient in three-dimensional space are substituted into the Navier-Stokes momentum equation and energy conservation heat transfer equation associated with thermodynamic boundary conditions in the ANSYS Fluent module. Iterative solutions to the nonlinear partial differential equations are then performed. Iteration stops when the residual curve decreases to 1 x 10^-4. The fluid velocity vector and heat transfer values ​​of each three-dimensional pixel node in the simulation space are extracted from the calculation results to form a fluid momentum and heat transfer scalar set. The data sequences of all volumetric pixel nodes are then summarized to form a numerical sequence of the flow field strength in three-dimensional space, outputting the distribution characteristics of the humid and hot air diffusion flow field.

[0030] S313: Invoke the distribution characteristics of the diffusion flow field of hot and humid airflow, locate the interface grid nodes of the human body geometric boundary mapping flow field, extract the absolute temperature value sequence of the interface grid nodes, generate local skin temperature data, extract the water vapor mass fraction value of the interface grid nodes, obtain the preset local environment saturated water vapor mass fraction constant, calculate the division quotient of the water vapor mass fraction value of the interface grid nodes and the local environment saturated water vapor mass fraction constant, generate the local relative humidity normalized value, call the maximum limit parameter of temperature distribution, perform a division ratio operation on the local skin temperature data and the maximum limit parameter of temperature distribution to eliminate physical dimensions, obtain the skin temperature limit ratio, extract the discrete numerical elements within the skin temperature limit ratio, linearly map the discrete numerical elements to the standard dimensionless interval of zero to one, and generate the local skin temperature normalized value.

[0031] In the simulation post-processing interface, the diffusion flow field distribution characteristics of the humid and hot airflow are invoked to locate the interface mesh nodes where the boundary of the human body's three-dimensional geometric model coincides with the contact area of ​​the external flow field region. The absolute temperature attribute values ​​attached to the interface mesh nodes are extracted to form an array sequence to generate local skin temperature data. Simultaneously, the water vapor mass percentage attribute values ​​attached to the interface mesh nodes are extracted to form the water vapor mass fraction value. The maximum water vapor mass percentage that the air can hold at the current dry-bulb temperature is read from the meteorological standard parameter table as the preset local environmental saturated water vapor mass fraction constant. The local relative humidity normalization value is calculated as follows: = water vapor mass fraction value at the interface grid nodes / local environmental saturated water vapor mass fraction constant. The maximum tolerable temperature of human skin is extracted as the maximum temperature distribution limit parameter, and the skin temperature limit ratio is calculated as: local skin temperature data / maximum temperature distribution limit parameter. Discrete numerical elements within the skin temperature limit ratio are extracted, and the interval of each element is determined. For numerical elements less than 0.8, the temperature in this area is determined to be severely below the normal physiological metabolic baseline, belonging to the hypothermia abnormality zone, and a value of 0 is directly assigned. For numerical elements greater than or equal to 0.8 and less than or equal to 1.2, it is determined to be within the normal active regulatory range of the human body, and the original value is retained. For numerical elements greater than 1.2, it is determined to be within the extreme heat accumulation risk zone, and a value of 1 is directly assigned. For the retained numerical elements, the local skin temperature normalization value is calculated as: (retained numerical element - 0.8) / (1.2 - 0.8), thus accurately converting the state of each interval into a local skin temperature normalization value within the zero-to-one standard dimensionless interval.

[0032] The specific steps for obtaining local comfort index data are as follows: S411: Based on a local area of ​​the body, call the preset sensory comfort neutral point temperature normalization value and the preset sensory comfort neutral point relative humidity normalization value, calculate the temperature deviation between the local skin temperature normalization value and the sensory comfort neutral point temperature normalization value, and calculate the humidity deviation between the local relative humidity normalization value and the sensory comfort neutral point relative humidity normalization value. Based on the localized area under the left armpit, the dimensionless standard skin temperature value for a person in a thermally neutral state was obtained from the international ASHRAE 55 thermal comfort benchmark database and set as 0.5 as the preset sensory comfort neutral point temperature normalization value. The dimensionless standard skin relative humidity value was also obtained and set as 0.4 as the preset sensory comfort neutral point relative humidity normalization value. The temperature deviation was calculated as: local skin temperature normalization value - preset sensory comfort neutral point temperature normalization value; the humidity deviation was calculated as: local relative humidity normalization value - preset sensory comfort neutral point relative humidity normalization value.

[0033] S412: Extract temperature and humidity deviations, and calculate local comfort index data based on preset sensory temperature and humidity weighting coefficients.

[0034] The specific formula for calculating and obtaining the local comfort index data is as follows: ; in, Represents local comfort index data. This represents the preset sensory temperature weighting coefficient. Represents the normalized value of local skin temperature. The normalized value of the neutral point temperature representing sensory comfort. This represents the preset sensory humidity weighting coefficient. This represents the normalized value of local relative humidity. The normalized relative humidity value represents the sensory comfort neutral point. Temperature and humidity deviations were extracted. Based on the objective distribution characteristic in human physiology that the density of temperature-sensing neurons is much higher than that of humidity sensors, the preset sensory temperature weighting coefficient, which determines the degree of temperature influence, was set to 0.65. A larger coefficient indicates that this indicator plays a dominant role in influencing the final sensory score. The preset sensory humidity weighting coefficient, which determines the degree of humidity influence, was set to 0.35. The comfort level was assessed by integrating the weights, and the calculation formula was: Local comfort index data = Preset sensory temperature weighting coefficient × Temperature deviation + Preset sensory humidity weighting coefficient × Humidity deviation, thus obtaining the local comfort index data.

[0035] The specific steps for obtaining the target clothing configuration information are as follows: S511: Call local comfort index data and local body regions, extract the three-dimensional network node parameters of local body regions, input the local comfort index data into the preset prediction average voting model to extract the network hidden layer weight matrix, use the network hidden layer weight matrix to perform nonlinear feature activation operation on the local comfort index data to obtain the body state index, set a comfort lower limit benchmark threshold for the body state index, compare the size relationship between the body state index and the comfort lower limit benchmark threshold, extract the three-dimensional network node parameters corresponding to the lower limit benchmark threshold and splice them to construct an abnormal representation patch to generate the target clothing discomfort area; The method for obtaining the lower limit benchmark threshold for comfort is as follows: Subjective thermal comfort voting scores and corresponding historical objective prediction index sets of multiple groups of subjects under different environmental conditions are extracted from a pre-set sample test database. Based on the label boundaries representing comfort and discomfort in the voting scores, they are divided into different evaluation data clusters. A mapping distribution curve is constructed between the numerical sequence of the historical objective prediction index set and the frequency of occurrence of the evaluation data cluster representing discomfort corresponding to each value in the sequence. The inflection point position where the slope of the mapping distribution curve changes non-linearly is calculated. The historical objective prediction index value corresponding to the inflection point position is extracted, and this value is defined as the critical value. The critical value is then calibrated as the lower limit benchmark threshold for comfort. Local comfort index data and a local body region under the left armpit were read. The 3D coordinates and physical attributes of each grid node within the local body region under the left armpit were extracted to form the 3D network node parameters. A fully connected backpropagation (BP) neural network was constructed using the TensorFlow framework in the Python environment as the preset prediction average voting model. The network structure includes one input layer, two hidden layers, and one output layer. The number of neurons in the input layer is aligned with the feature dimensions of the local comfort index data. Each of the two hidden layers has 64 neurons. Signals are transmitted between layers using a fully connected approach, and the hidden layers use a rectified linear unit activation function to process the data. The local comfort index data was input into this preset prediction average voting model, and the weight matrix of the connection between the hidden layer and the input layer was extracted as the network hidden layer weight matrix. The dot product of the network hidden layer weight matrix and each element of the local comfort index data was calculated and summed to obtain the accumulated value. The rectified linear unit activation function was used to perform a nonlinear mapping process to eliminate negative values ​​on the accumulated value to obtain the somatosensory state index. To obtain the lower limit of comfort baseline threshold, subjective thermal comfort seven-level voting scores of fifty subjects under five different ambient temperatures were exported from a historical environmental testing database, along with a set of objective predictive indicators of the physical environment recorded simultaneously during the tests. Subject data with scores between -1 and 1 were classified into comfort evaluation data clusters, while data with scores between -3, -2, and 2 and 3 were classified into discomfort evaluation data clusters. The frequency distribution was constructed by statistically analyzing the percentage of occurrences of discomfort evaluation data clusters at each value point within the objective predictive indicator set. A smooth mapping distribution curve was plotted with the indicator values ​​on the horizontal axis and the frequency distribution on the vertical axis. The second derivative of each point on this mapping distribution curve was calculated, and the inflection point where the derivative value equals zero and the slope undergoes a non-linear abrupt change was identified. The objective indicator value of 0.65 corresponding to this inflection point was defined as the critical value, and 0.65 was calibrated as the lower limit of comfort baseline threshold. The relationship between the somatosensory state index and the lower limit of comfort threshold is determined. For nodes with a somatosensory state index greater than or equal to 0.65, they are considered to be in an acceptable state and are left as is. For nodes with a somatosensory state index less than 0.65, they are considered to be in an uncomfortable state, and their corresponding 3D network node parameters are extracted. The Delaunay triangular patch generation algorithm is used to connect and stitch the extracted 3D network node parameters to construct an anomaly representation patch to generate the discomfort area under the left armpit of the target garment.

[0036] S512: Call the attribute data of the replacement clothing module and the configuration data of the replacement connection interface in the preset material database. For the discomfort area of ​​the target clothing, call the attribute data of the replacement clothing module to rewrite the underlying material feature vector of the clothing preset clothing module attribute data to obtain the adjusted module attribute data. Call the configuration data of the replacement connection interface to cover the edge stitching point coordinates of the clothing connection interface geometric structure data to obtain the adjusted connection structure data. Map the adjusted module attribute data to the adjusted connection structure data, perform node feature binding operation to construct the rendering parameter set, and generate the target clothing configuration information. In the virtual clothing modeling software, the material parameters of highly breathable mesh fabric are retrieved from the local server's material parameter configuration database as the attribute data for the replacement garment module. The coordinates of seam points with lower density and increased spacing are retrieved as the configuration data for the replacement connection interface. For the discomfort area under the left armpit of the generated target garment, the yarn count feature vector of the original high-density cotton material is removed, and the high breathability feature vector corresponding to the attribute data of the replacement garment module is filled into the original address area to complete the rewriting operation and obtain the adjusted module attribute data. The replacement connection interface configuration data is called to overwrite the original dense edge seam point coordinates of the discomfort area of ​​the target garment, performing a coordinate replacement and erasure operation to generate adjusted connection structure data with larger seam spacing. The material features in the adjusted module attribute data are assigned one by one to the vertex coordinate positions in the adjusted connection structure data, performing node feature binding operations. The material properties and geometric coordinate information are summarized to form a rendering parameter set for 3D graphics engine rendering, outputting the target garment configuration information for direct pattern making.

[0037] Please see Figure 2 A modular multifunctional integrated clothing optimization system is provided to implement the aforementioned modular multifunctional integrated clothing optimization method. The system includes: The morphological modeling module acquires the target user's body point cloud data, clothing preset module attribute data, and clothing connection interface geometric structure data. It constructs a three-dimensional geometric model of the human body based on the body point cloud data and generates a three-dimensional virtual clothing model based on the clothing preset module attribute data and clothing connection interface geometric structure data. The physiological response analysis module extracts the local body area covered by the three-dimensional virtual clothing model, and combines the preset exercise environment temperature data, preset exercise environment humidity data and exercise metabolic rate parameters to obtain local sweat flow data and local heat flux density data through the Gagge two-node physiological thermoregulation model. The microenvironment coupling module establishes thermodynamic boundary conditions based on local sweat flow data and local heat flux density data. It also constructs the distribution characteristics of the humid and hot air diffusion field by combining preset three-dimensional air density parameters, air isobaric specific heat capacity parameters, air velocity parameters, air thermal conductivity parameters, water vapor mass fraction data, and water vapor diffusion coefficient parameters. It calculates and obtains the normalized values ​​of local skin temperature and local relative humidity. The perception evaluation module calls the preset heat and humidity transfer simulation parameters, and calculates and obtains the local comfort index data based on the local skin temperature normalization value and the local relative humidity normalization value, combined with the preset sensory comfort neutral point temperature normalization value and the preset sensory comfort neutral point relative humidity normalization value. The adaptation and optimization module inputs local comfort index data into a preset prediction average voting model, analyzes the body's sensation index in local areas, and filters the target clothing discomfort areas based on the sensation index. It then replaces the corresponding clothing module attribute data and clothing connection interface geometric structure data to generate the target clothing configuration information.

[0038] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A modular, multifunctional integrated clothing optimization method, characterized in that, Includes the following steps: S1: Obtain the target user's body point cloud data, clothing preset module attribute data, and clothing connection interface geometric structure data; construct a three-dimensional geometric model of the human body based on the body point cloud data; and generate a three-dimensional virtual clothing model based on the clothing preset module attribute data and clothing connection interface geometric structure data. S2: Based on the three-dimensional virtual clothing model, extract the local body area it covers, and combine it with preset exercise environment temperature data, preset exercise environment humidity data and exercise metabolic rate parameters to obtain local sweat flow data and local heat flux density data through the Gagge two-node physiological thermoregulation model. S3: Based on the local sweat flow rate data and local heat flux density data, establish thermodynamic boundary conditions, and combine the preset three-dimensional space air density parameters, air isobaric specific heat capacity parameters, air velocity parameters, air thermal conductivity parameters, water vapor mass fraction data and water vapor diffusion coefficient parameters to construct the distribution characteristics of the humid and hot air diffusion field, and calculate and obtain the normalized value of local skin temperature and the normalized value of local relative humidity. S4: Call the preset heat and moisture transfer simulation parameters, and calculate and obtain the local comfort index data based on the local skin temperature normalization value and the local relative humidity normalization value, combined with the preset sensory comfort neutral point temperature normalization value and the preset sensory comfort neutral point relative humidity normalization value. S5: Input the local comfort index data into the preset prediction average voting model, analyze the body sensation index of local areas, and filter the target clothing discomfort area based on the body sensation index, replace the corresponding clothing preset clothing module attribute data and clothing connection interface geometric structure data, and generate target clothing configuration information.

2. The modular multifunctional integrated clothing optimization method according to claim 1, characterized in that, The specific steps for obtaining the 3D virtual clothing model are as follows: S111: Collect body point cloud data of the target user, extract the three-dimensional coordinate array of the body point cloud data, calculate the normal vector of the three-dimensional coordinate array, construct a spatial surface fitting equation based on the three-dimensional coordinate array and the normal vector, call the spatial surface fitting equation to connect the three-dimensional coordinate array to construct a polygonal topology network, and establish a three-dimensional geometric model of the human body. S112: Obtain the attribute data of the clothing module preset in the preset configuration file and the geometric structure data of the clothing connection interface. Based on the human body three-dimensional geometric model, extract the coordinate set of the outer boundary of the body surface. Calculate the position deviation matrix between the vertex coordinates in the geometric structure data of the clothing connection interface and the coordinate set of the outer boundary of the body surface. Call the position deviation matrix to perform mesh vertex assignment transformation on the material attribute matrix in the attribute data of the clothing module preset to obtain the spatial mapping distribution characteristics. S113: Extract the surface mesh coordinate set and the underlying material matrix based on the spatial mapping distribution features. Perform smooth interpolation continuous operation on the surface mesh coordinate set to output a smooth surface vertex sequence. Map the underlying material matrix to the smooth surface vertex sequence to construct material binding feature nodes. Perform spatial topology closure operation on the outer boundary nodes of the material binding feature nodes to generate a three-dimensional virtual clothing model.

3. The modular multifunctional integrated clothing optimization method according to claim 1, characterized in that, The specific steps for obtaining the local heat flux density data are as follows: S211: Extract three-dimensional mesh patches from the surface of the three-dimensional virtual clothing model, calculate the coordinates of the three-dimensional mesh patches and associate them with the human body spatial coordinate set, extract local areas of the body based on the human body spatial coordinate set, map the local areas of the body as thermodynamic nodes within the spatial network, and establish a partitioned thermodynamic node matrix. S212: For the partitioned thermal node matrix, call the preset exercise environment temperature data, preset exercise environment humidity data, and exercise metabolic rate parameters, and calculate the difference between the heat production flux associated with the exercise metabolic rate parameters and the heat transfer flux associated with the partitioned thermal node matrix through the preset Gagge two-node physiological thermoregulation model. Calculate the weighted sum of the heat transfer flux difference, the convective heat transfer coefficient of the preset exercise environment temperature data, and the evaporation coefficient of the preset exercise environment humidity data to obtain the heat dissipation demand data corresponding to each partitioned thermal node. S213: Call the heat dissipation demand data, extract the latent heat of vaporization demand of each of the partitioned thermal nodes, calculate the ratio of the latent heat of vaporization demand of the node to the latent heat of water vaporization constant, construct a liquid precipitation rate distribution matrix, perform time integration operation on the elements in the liquid precipitation rate distribution matrix to extract the mass flow rate scalar, perform transformation multiplication operation on the elements of the liquid precipitation rate distribution matrix to extract heat flux data, and generate local sweat flow data and local heat flux density data.

4. The modular multifunctional integrated clothing optimization method according to claim 3, characterized in that, The process of performing time integration on the elements of the liquid precipitation rate distribution matrix to extract the mass flow rate scalar, and performing transformation multiplication on the elements of the liquid precipitation rate distribution matrix to extract heat flux data, specifically involves: Calculate the time span difference between the preset start time node and the end time node of the movement, and use it as the preset environmental time span benchmark value; An integration time window constraint is constructed based on the preset environmental time span benchmark value. The integration time window constraint is that the motion start time node is the lower limit of integration and the motion end time node is the upper limit of integration. Based on the integral time window constraint, the elements in the liquid precipitation rate distribution matrix are accumulated over time to obtain the mass flow rate scalar. Calculate the product of the liquid specific heat capacity parameter and the standard atmospheric pressure constant, and use it as the liquid enthalpy conversion coefficient; multiply the elements in the liquid precipitation rate distribution matrix with the liquid enthalpy conversion coefficient to obtain a preliminary heat flow data matrix; Calculate the ratio of surface wind speed parameter to air density parameter, and multiply this ratio by a preset mapping ratio constant as the surface convective heat transfer weight; The heat flux distribution matrix is ​​obtained by multiplying the preliminary heat flux data matrix with the surface convection heat transfer weight, and the heat flux distribution matrix is ​​used as the corresponding local heat flux density data.

5. The modular multifunctional integrated clothing optimization method according to claim 1, characterized in that, The specific steps for obtaining the normalized values ​​of local skin temperature and local relative humidity are as follows: S311: Call the preset heat and moisture transfer simulation parameters, which include three-dimensional air density parameters, air specific heat capacity at constant pressure parameters, air velocity parameters, air thermal conductivity parameters, water vapor mass fraction data, water vapor diffusion coefficient parameters, and temperature distribution maximum limit parameters. Use the local sweat flow data and local heat flux density data to assign energy exchange flux to the mesh nodes on the surface of the three-dimensional virtual clothing model and establish thermodynamic boundary conditions. S312: Map the three-dimensional spatial air density parameters, air isobaric specific heat capacity parameters, air velocity parameters, air thermal conductivity parameters, water vapor mass fraction data, and water vapor diffusion coefficient parameters to thermodynamic boundary conditions to solve multiphase fluid dynamics transfer problems, extract the fluid momentum and heat transfer scalar sets of each voxel node in space, and use the spatial vector field distribution composed of the fluid momentum and heat transfer scalar sets of each voxel node as the distribution characteristics of the humid and hot air diffusion flow field; S313: Invoke the distribution characteristics of the humid and hot airflow diffusion field, locate the interface grid nodes of the human body geometric boundary mapping flow field, extract the absolute temperature value sequence of the interface grid nodes, generate local skin temperature data, extract the water vapor mass fraction value of the interface grid nodes, obtain the preset local environment saturated water vapor mass fraction constant, calculate the division quotient of the water vapor mass fraction value of the interface grid nodes and the local environment saturated water vapor mass fraction constant, generate the local relative humidity normalized value, call the maximum limit parameter of temperature distribution, perform a division ratio operation on the local skin temperature data and the maximum limit parameter of temperature distribution to eliminate physical dimensions, obtain the skin temperature limit ratio, extract the discrete numerical elements within the skin temperature limit ratio, linearly map the discrete numerical elements to the zero to one standard dimensionless interval, and generate the local skin temperature normalized value.

6. The modular multifunctional integrated clothing optimization method according to claim 1, characterized in that, The specific steps for obtaining the local comfort index data are as follows: S411: Based on the local area of ​​the body, call the preset sensory comfort neutral point temperature normalization value and the preset sensory comfort neutral point relative humidity normalization value, calculate the temperature deviation between the local skin temperature normalization value and the sensory comfort neutral point temperature normalization value, and calculate the humidity deviation between the local relative humidity normalization value and the sensory comfort neutral point relative humidity normalization value. S412: Extract the temperature deviation and the humidity deviation, and calculate the local comfort index data according to the preset sensory temperature weighting coefficient and the preset sensory humidity weighting coefficient.

7. The modular multifunctional integrated clothing optimization method according to claim 6, characterized in that, The specific formula for calculating and obtaining the local comfort index data is as follows: ； in, Represents local comfort index data. This represents the preset sensory temperature weighting coefficient. Represents the normalized value of local skin temperature. The normalized value of the neutral point temperature representing sensory comfort. This represents the preset sensory humidity weighting coefficient. This represents the normalized value of local relative humidity. The normalized relative humidity value represents the sensory comfort neutral point.

8. The modular multifunctional integrated clothing optimization method according to claim 1, characterized in that, The specific steps for obtaining the target clothing configuration information are as follows: S511: Call the local comfort index data and the local body region, extract the three-dimensional network node parameters of the local body region, input the local comfort index data into the preset prediction average voting model to extract the network hidden layer weight matrix, use the network hidden layer weight matrix to perform nonlinear feature activation operation on the local comfort index data to obtain the body sensation index, set a comfort lower limit benchmark threshold for the body sensation index, compare the size relationship between the body sensation index and the comfort lower limit benchmark threshold, extract the three-dimensional network node parameters corresponding to the lower limit benchmark threshold and splice them to construct an abnormal representation patch to generate the target clothing discomfort area; S512: Call the attribute data of the replacement clothing module and the configuration data of the replacement connection interface in the preset material database. For the discomfort area of ​​the target clothing, call the attribute data of the replacement clothing module to rewrite the underlying material feature vector of the preset clothing module attribute data to obtain the adjusted module attribute data. Call the configuration data of the replacement connection interface to cover the edge stitching point coordinates of the geometric structure data of the clothing connection interface to obtain the adjusted connection structure data. Map the adjusted module attribute data to the adjusted connection structure data and perform node feature binding operation to construct a rendering parameter set to generate the target clothing configuration information.

9. The modular multifunctional integrated clothing optimization method according to claim 8, characterized in that, The lower limit of the comfort baseline threshold is obtained as follows: Subjective thermal comfort voting scores and corresponding historical objective prediction index sets of multiple groups of subjects under different environmental conditions are extracted from a pre-set sample test database. Based on the label boundaries representing comfort and discomfort in the voting scores, they are divided into different evaluation data clusters. A mapping distribution curve is constructed between the numerical sequence of the historical objective prediction index set and the frequency of occurrence of the evaluation data cluster representing discomfort corresponding to each value in the sequence. The inflection point position where the slope of the mapping distribution curve changes non-linearly is calculated. The historical objective prediction index value corresponding to the inflection point position is extracted, and this value is defined as the critical value. This critical value is then calibrated as the lower limit benchmark threshold for comfort.

10. A modular, multifunctional integrated clothing optimization system, characterized in that, The system is used to implement the modular multifunctional integrated clothing optimization method as described in any one of claims 1-9, the system comprising: The morphological modeling module acquires the target user's body point cloud data, clothing preset module attribute data, and clothing connection interface geometric structure data. It constructs a three-dimensional geometric model of the human body based on the body point cloud data and generates a three-dimensional virtual clothing model based on the clothing preset module attribute data and clothing connection interface geometric structure data. The physiological response analysis module extracts the local body area covered by the three-dimensional virtual clothing model, and combines the preset exercise environment temperature data, preset exercise environment humidity data and exercise metabolic rate parameters to obtain local sweat flow data and local heat flux density data through the Gagge two-node physiological thermoregulation model. The microenvironment coupling module establishes thermodynamic boundary conditions based on the local sweat flow rate data and local heat flux density data, and constructs the distribution characteristics of the humid and hot air diffusion field by combining preset three-dimensional spatial air density parameters, air isobaric specific heat capacity parameters, air flow velocity parameters, air thermal conductivity parameters, water vapor mass fraction data and water vapor diffusion coefficient parameters, and calculates and obtains the normalized value of local skin temperature and the normalized value of local relative humidity. The perception evaluation module calls the preset heat and humidity transfer simulation parameters, and calculates and obtains the local comfort index data based on the local skin temperature normalization value and the local relative humidity normalization value, combined with the preset sensory comfort neutral point temperature normalization value and the preset sensory comfort neutral point relative humidity normalization value. The adaptation and optimization module inputs the local comfort index data into a preset prediction average voting model, analyzes the body sensation index of local areas, and filters the target clothing discomfort areas based on the body sensation index. It then replaces the corresponding clothing preset clothing module attribute data and clothing connection interface geometric structure data to generate target clothing configuration information.

Images