Preparation optimization method and system of high memory polyester backing cloth

Abstract

The application discloses a preparation optimization method and system of high-memory polyester lining cloth, and relates to the related technical field of textile fabrics.The method comprises the following steps: receiving a product application scene of a target polyester lining cloth, performing scene adaptation analysis, and establishing a function requirement matrix; utilizing a structure response table to perform adaptation analysis on the function requirement matrix; extracting key indicators from a hierarchical partition structure combination, and establishing a multi-objective evaluation function mapped with the hierarchical partition structure; establishing a parameter optimization candidate set; performing hierarchical control on the hierarchical partition structure combination, utilizing finite element simulation to perform response matching; establishing a turn deviation according to the response matching result, and updating the parameter optimization candidate set to complete preparation optimization.The technical problems of poor preparation scientificity and production efficiency of the high-memory polyester lining cloth caused by the difficulty of the polyester lining cloth in adapting to diversified application scenes and the difficulty of balancing performance indicators in the prior art are solved, and the technical effect of improving the preparation process scientificity and production efficiency is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of textile fabrics, specifically relates to a preparation optimization method and system of high-memory polyester lining cloth. BACKGROUND

[0002] As a key component of clothing accessories, polyester lining cloth directly affects the overall quality and wearing experience of clothing. High-memory polyester lining cloth is widely used in high-end clothing, professional clothing, functional clothing, etc. due to its excellent shape retention ability, wrinkle resistance and recovery performance. However, in the existing preparation process of polyester lining cloth, there is a lack of systematic analysis and adaptation of functional requirements in different application scenarios, which cannot accurately control the performance of the lining cloth in different scenarios, resulting in a disconnection between product performance and actual demand. The structural design and performance optimization of the lining cloth lack a scientific hierarchical system and quantitative evaluation standard, making it difficult to balance and coordinate multiple properties such as memory, flexibility, and durability. In addition, the existing preparation process lacks response matching analysis of the composite structure and functional requirements of the lining cloth, making it difficult to predict product performance in advance, adjust production parameters in a timely manner, and thus unable to achieve precision and intelligentization, resulting in unstable product quality, low production efficiency, and inability to meet the urgent demand for high-quality and high-performance polyester lining cloth in the market.

[0003] Therefore, in the related art at present, there is a technical problem that polyester lining cloth is difficult to adapt to diversified application scenarios and performance indicators are difficult to balance, resulting in poor scientificity and production efficiency in the preparation of high-memory polyester lining cloth. SUMMARY

[0004] The present application provides a preparation optimization method and system of high-memory polyester lining cloth, which solves the technical problem of poor scientificity and production efficiency in the preparation of high-memory polyester lining cloth due to the difficulty of polyester lining cloth to adapt to diversified application scenarios and the difficulty of performance indicators to balance in the prior art, and achieves the technical effect of improving the scientificity and production efficiency of the preparation process.

[0005] The application provides a preparation optimization method of high-memory polyester lining cloth, which comprises the following steps: after receiving a product application scene of a target polyester lining cloth, performing scene adaptation analysis, and establishing a function demand matrix; activating a structure response table of structure partition function, performing adaptation analysis on the function demand matrix by using the structure response table, establishing a hierarchical partition structure combination, and the hierarchical partition structure combination comprises a memory support layer, a flexible buffer layer and a durable packaging layer; extracting key indicators from each hierarchical partition structure in the hierarchical partition structure combination, and performing weight compensation on the key indicator extraction result by using the function demand matrix, establishing a multi-objective evaluation function mapped with the hierarchical partition structure; after configuring a parameter space, performing hierarchical optimization of the corresponding hierarchical partition structure by using the multi-objective evaluation function, and establishing a parameter optimization candidate set; after hierarchical control of the hierarchical partition structure combination based on the parameter optimization candidate set, performing response matching of the composite structure and the function demand matrix by using finite element simulation; establishing a turning deviation according to the response matching result, updating the parameter optimization candidate set by using the turning deviation, and completing preparation optimization according to the updating result.

[0006] In possible implementation manners, the preparation optimization method of the high-memory polyester lining cloth further performs the following processing: performing multi-dimensional decomposition on the product application scene, establishing a scene dimension matrix, and the multi-dimensional decomposition comprises purpose decomposition, wearing scene decomposition, use frequency and duration decomposition, washing and caring requirement decomposition and touch demand decomposition; establishing a target function set, and calling a semantic association model to perform semantic matching analysis of each scene dimension in the scene dimension matrix and the target function set, and establishing the function demand matrix.

[0007] In possible implementation manners, the preparation optimization method of the high-memory polyester lining cloth further performs the following processing: according to the structure response table, mapping each function demand in the function demand matrix to a structure unit with the highest response value, establishing a mapping path; performing function item clustering analysis in the function demand matrix by using the mapping path, establishing a function subset through a structure response similarity index, a target function interaction index and a process integrability index, discriminating a combination relationship of the function subset in a structure hierarchy, and establishing the hierarchical partition structure combination.

[0008] In a possible implementation, the preparation optimization method of the high-memory polyester lining fabric further performs the following processing: obtaining a first multi-objective evaluation function corresponding to the memory support layer, the first multi-objective evaluation function being a double-objective function, and evaluation targets of the first multi-objective evaluation function including a memory retention target and a thermal resilience target; randomly establishing an initial parameter set, each individual in the initial parameter set representing a combination of a group of heat treatment and resilience control parameters; after calculating fitness values of each individual in the initial parameter set by using the first multi-objective evaluation function, performing individual non-dominated sorting and congestion calculation; after configuring random probabilities according to the individual non-dominated sorting result and the congestion calculation result, performing individual selection, crossover and mutation, and iteratively updating the initial parameter set to establish a parameter optimization candidate set.

[0009] In a possible implementation, the preparation optimization method of the high-memory polyester lining fabric further performs the following processing: distributing individual hierarchical selection probabilities by using the individual non-dominated sorting result; establishing individual congestion selection probabilities according to the current iteration number and the congestion calculation result; and after configuring random probabilities by using the individual hierarchical selection probabilities and the individual congestion selection probabilities, performing individual selection, crossover and mutation.

[0010] In a possible implementation, the preparation optimization method of the high-memory polyester lining fabric further performs the following processing: obtaining a second multi-objective evaluation function corresponding to the soft buffer layer, the second multi-objective evaluation function being a three-objective function, and evaluation targets of the second multi-objective evaluation function including a comfort target, a shock absorption effect target and a deformability target; performing search optimization of the soft buffer layer by using the second multi-objective evaluation function to establish a parameter optimization candidate set.

[0011] In a possible implementation, the preparation optimization method of the high-memory polyester lining fabric further performs the following processing: obtaining a third multi-objective evaluation function corresponding to the durable packaging layer, the third multi-objective evaluation function being a four-objective function, and evaluation targets of the third multi-objective evaluation function including an anti-aging target, a chemical resistance target, a wear resistance target and an ultraviolet resistance performance target; performing search optimization of the durable packaging layer by using the third multi-objective evaluation function to establish a parameter optimization candidate set.

[0012] In a possible implementation, the preparation optimization method of the high-memory polyester lining fabric further performs the following processing: establishing a multi-layer composite finite element simulation structure according to the hierarchical partition structure combination; inputting the parameter optimization candidate set as input data into the multi-layer composite finite element simulation structure; calling a coupling solver to perform thermal-force coupling and viscoelastic-thermal resilience response coupling to generate a simulation result; and performing response matching on a functional requirement matrix by using the simulation result.

[0013] In a possible implementation, the preparation optimization method of the high-memory polyester lining cloth further performs the following processing: establishing a local search guide strategy based on the rotation deviation; and performing local search optimization on the parameter optimization candidate set by using the local search guide strategy to update the parameter optimization candidate set.

[0014] The application also provides a preparation optimization system of a high-memory polyester lining cloth, which comprises: a scene adaptation analysis module, which is configured to perform scene adaptation analysis after receiving a product application scene of a target polyester lining cloth, and establish a function demand matrix; a matrix adaptation analysis module, which is configured to activate a structure response table of a structure partition function, perform matrix adaptation analysis on the function demand matrix by using the structure response table, and establish a hierarchical partition structure combination, wherein the hierarchical partition structure combination comprises a memory support layer, a flexible buffer layer, and a durable packaging layer; a key indicator extraction module, which is configured to extract key indicators from each hierarchical partition structure in the hierarchical partition structure combination, and establish a multi-objective evaluation function mapped with the hierarchical partition structure by using the function demand matrix to compensate the key indicator extraction result; a hierarchical optimization module, which is configured to perform hierarchical optimization on the corresponding hierarchical partition structure by using the multi-objective evaluation function after configuring a parameter space, and establish a parameter optimization candidate set; a response matching module, which is configured to perform response matching of a composite structure and the function demand matrix by using finite element simulation after performing hierarchical control on the hierarchical partition structure combination based on the parameter optimization candidate set; and a rotation deviation establishing module, which is configured to establish a rotation deviation according to the response matching result, update the parameter optimization candidate set by using the rotation deviation, and complete preparation optimization according to the update result.

[0015] The application provides a preparation optimization method and system of a high-memory polyester lining cloth, which receives a product application scene of a target polyester lining cloth, performs scene adaptation analysis, and establishes a function demand matrix; performs matrix adaptation analysis on the function demand matrix by using a structure response table; extracts key indicators from a hierarchical partition structure combination, and establishes a multi-objective evaluation function mapped with the hierarchical partition structure; establishes a parameter optimization candidate set; performs hierarchical control on the hierarchical partition structure combination, and performs response matching by using finite element simulation; establishes a rotation deviation according to the response matching result, and updates the parameter optimization candidate set to complete preparation optimization. The application solves the technical problems that the polyester lining cloth is difficult to adapt to diversified application scenes, performance indicators are difficult to balance, and the preparation of the high-memory polyester lining cloth is poor in scientific nature and production efficiency, and achieves the technical effect of improving the scientific nature and production efficiency of the preparation process. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. In the present application, a flow chart is used to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously according to needs. At the same time, other operations can be added to these processes, or one or more steps of operations can be removed from these processes.

[0017] Figure 1 A preparation optimization method flowchart of a high-memory polyester lining cloth provided by the embodiments of the present application.

[0018] Figure 2 A preparation optimization system structure schematic diagram of a high-memory polyester lining cloth provided by the embodiments of the present application.

[0019] Legend: scene adaptation analysis module 10, matrix adaptation analysis module 20, key indicator extraction module 30, hierarchical optimization module 40, response matching module 50, and revolution deviation establishment module 60. DETAILED DESCRIPTION

[0020] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the content of the specification can be implemented, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0021] In order to make the purposes, technical solutions and advantages of the present application more clear, the present application will be further described in detail below with reference to the drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by those of ordinary skill in the art without making creative labor belong to the scope of protection of the present application.

[0022] In the following description, "some embodiments" are referred to, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict, and the term "first\second" referred to is only to distinguish similar objects, and does not represent a specific order for the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0023] The embodiments of the present application provide a preparation optimization method of high-memory polyester lining cloth, as shown in the method, the method comprises the following steps: Figure 1

[0024] Step S100, after receiving the product application scene of the target polyester lining cloth, performing scene adaptation analysis, and establishing a function requirement matrix.

[0025] Preferably, the product application scene of the target polyester lining cloth is obtained, that is, the specific scene of the final use of the target polyester lining cloth, which can include but is not limited to the clothing field, such as suit lining (high stiffness and memory are required), shirt collar lining (softness and wrinkle resistance are required), wedding dress support lining (high strength and shaping are required), etc.; luggage field, such as luggage reinforcement lining (wear resistance and structural stability are required), backpack buffer lining (elasticity and shock resistance are required), etc.; special field, such as medical protector lining (air permeability and biocompatibility are required), industrial equipment protection lining (high temperature resistance or corrosion resistance is required), etc.; different scenes have significant differences in performance requirements for the target polyester lining cloth.

[0026] ​Preferably, the scene adaptation analysis is performed, i.e. the core requirements of the product application scene are disassembled, such as the lining for close-fitting underwear, which needs to consider softness, air permeability, skin-friendliness, etc.; the lining for outdoor work clothes pays more attention to wear resistance, strength, wash resistance, etc. Specifically, through questionnaire survey, interview, etc., the end user's pain points (such as "shirt collar is easy to deform") of the lining in a specific scene are collected, and the performance shortcomings (such as "the breaking strength of a certain bag lining is insufficient") of the existing lining in the same scene are disassembled, the functional properties of the target polyester lining are analyzed and determined, which may include strength, resilience, wear resistance, wrinkle resistance, softness, air permeability, fit, wash resistance, etc. Then the results of the scene adaptation analysis are converted into specific function items, and the importance of each function item is weighted according to the product application scene of the target polyester lining. For example, in sports clothing lining, "resilience" and "moisture absorption and air permeability" are important, and the weight can be set to high; while "soft touch" is relatively secondary, and the weight is set to medium; and the target level of each function item is set, such as "resilience" requires to reach the level of "strong", "moisture absorption and air permeability" reaches the level of "good", etc. Finally, a structured matrix table, i.e. a function requirement matrix, is constructed to clearly show each function item and its quantitative requirements.

[0027] Further, step S100 further includes step S110 of performing multi-dimensional disassembly on the product application scene to establish a scene dimension matrix, wherein the multi-dimensional disassembly includes use purpose disassembly, wearing scene disassembly, use frequency and duration disassembly, washing and caring requirement disassembly, and touch feeling demand disassembly; and step S120 of establishing a target function set and calling a semantic association model to perform semantic matching analysis of each scene dimension in the scene dimension matrix and the target function set to establish a function requirement matrix.

[0028] Preferably, according to the purpose of use, the wearing scene, the frequency and duration of use, the cleaning and maintenance requirements, and the touch feeling demand, the product application scene is multi-dimensionally decomposed, that is, the abstract product application scene is decomposed into quantifiable and analyzable specific dimensions to form a structured scene description. Specifically, the core demands and task targets of users using the product are analyzed, for example, the purposes of using sports shoes can include "daily commuting", "professional running", "fitness training", etc., and a list of use purposes is output. The actual use environment of the product is identified, such as temperature (high / low), humidity, leisure, business, sports competition, outdoor exploration, etc., and a combination of environmental parameters is output. The intensity and duration of users using the product are quantified, including frequency and duration, and the use intensity (such as "high frequency and long duration" and "low frequency and short duration") is output. The demand and limitation of users on product cleaning and maintenance are analyzed, including whether it can be machine washed, temperature resistance (such as ≤40℃), whether it needs special care (waterproof coating maintenance), chemical resistance (such as detergent corrosion), and a list of cleaning and maintenance condition constraints is output. The subjective preference and functional demand of users on product contact feeling are analyzed, and the priority order of touch attributes (such as "softness > breathability") is output. Finally, a scene dimension matrix is established.

[0029] Preferably, the functions to be realized by the product are abstracted into traceable functional units to form a "vocabulary" of functional design, such as basic functions (shock absorption, support stability, wear resistance, and slip resistance), extended functions (breathability, sweat-wicking, antibacterial, and deodorization), and constraint functions (lightweight, machine washable), and then the semantic matching analysis of each scene dimension in the scene dimension matrix and the target function set is performed through a semantic association model (such as natural language processing and knowledge graph) to obtain the mapping relationship between the scene dimension and the functional response, to clarify the logical chain of scene demand and functional response, and then to construct a "scene-function" association matrix, as shown in Table 1, which is an example data of a functional demand matrix.

[0030] Table 1 Quantification table of functional demand of polyester lining

[0031] Functionality Weight Target level Resilience High Strong Form retention High Strong Soft touch Medium Medium Moisture absorption and air permeability Low General Wear and wash resistance High Strong

[0032] In step S200, a structure response table of structure partition function is activated, and the structure response table is used for adaptive analysis of the functional demand matrix to establish a hierarchical partition structure combination, and the hierarchical partition structure combination includes a memory support layer, a soft buffer layer, and a durable packaging layer.

[0033] Preferably, the structure response table is a pre-established correspondence table for clarifying the relationship between different lining functions and the structure elements for realizing these functions. Specifically, the content of the structure response table for structure partition function includes: high orientation, high crystalline fiber + heat treatment structure determines the resilience and shape retention of the lining; low modulus fiber, cavity structure, surface fluffing treatment determines the soft touch and moisture permeability of the lining; high density fabric + surface coating / resin crosslinking determines wear resistance and washing resistance; the structure response table for activating structure partition function refers to applying the correspondence relationship of the structure response table to the preparation optimization process of the target polyester lining, and performing adaptive analysis on the function demand matrix. Specifically, according to the function items and their weights, target levels in the function demand matrix, and by referring to the structure response table, the corresponding structure elements are determined. For example, according to the function demand matrix, the weights of “resilience” and “shape retention” are high and the target level is strong. According to the structure response table, high orientation, high crystalline fiber + heat treatment structure plays a key role in these two functions, so the corresponding structure elements are determined. At the same time, the structure elements corresponding to different functions are comprehensively considered to avoid conflicts between structure elements. For example, the structure elements corresponding to soft touch and wear resistance and washing resistance may conflict in implementation, which needs to be balanced and coordinated so that each function can meet the demand.

[0034] Preferably, based on the adaptive analysis result, high orientation, high crystalline fiber is selected to construct a memory support layer in combination with heat treatment structure to realize the high-weight resilience and shape retention functions. The memory support layer is mainly used to give the lining good resilience and shape retention ability, so that the lining can maintain a specific shape and is not easy to deform during use. A soft and cushioning layer is created by using low modulus fiber, setting a cavity structure and performing surface fluffing treatment to meet the functional requirements such as soft touch, so that the lining feels soft and comfortable when it contacts the skin, and at the same time provides a certain degree of cushioning to improve the wearing experience. According to the function requirements of wear resistance and washing resistance, a durable packaging layer is constructed by high-density fabric combined with surface coating or resin crosslinking to enhance the wear resistance of the lining, so that it can withstand multiple washings and rubbing without damage, prolonging the service life. Finally, the memory support layer, the soft and cushioning layer, and the durable packaging layer are combined to form a polyester lining hierarchical partition structure combination that meets the requirements of the application scenario, which is used to closely link the functional requirements and the structure design.

[0035] Further, step S200 further comprises step S210 of mapping each item of function demand in the function demand matrix to the structure unit with the highest response value according to the structure response table to establish a mapping path; and step S220 of performing clustering analysis of the function items in the function demand matrix by using the mapping path, establishing a function subset through a structure response similarity index, a target function interaction index, and a process integrability index, and discriminating the combination relationship of the structure hierarchy for the function subset to establish a hierarchical partition structure combination.

[0036] Preferably, the structure response table shows the association of different functions and structure elements, for each function requirement in the function requirement matrix, find the structure unit in the structure response table that can make it achieve the highest response value (i.e. the function target level that can best meet the function target level), for example, for the "resilience" function requirement, it can be known from the structure response table that the high orientation, high crystalline fiber + heat treatment structure has the greatest influence on it, and this is the highest response value structure unit corresponding to the "resilience" function requirement; after determining the highest response value structure unit corresponding to each function requirement, the corresponding relationship between the function requirement and the structure unit is established to form a mapping path, i.e. the function-structure association relationship, to clearly show the corresponding structure for realizing each function.

[0037] Preferably, the function items in the function requirement matrix are clustered and analyzed through the mapping path, specifically including establishing a function subset through a structure response similarity index, i.e. according to the mapping path, analyzing the structure units corresponding to different function items, if the structure units corresponding to multiple function items are similar, it means that these functions have similarity in structure implementation, such as "resilience" and "shape retention" are mainly determined by high orientation, high crystalline fiber + heat treatment structure, these two functions have similarity in structure response, and then the corresponding function subset is established.

[0038] Preferably, the function subset is established through a target function interaction index, i.e. evaluating the interaction relationship between different function items, some functions may promote each other, and some may restrict each other, for example, "soft touch" and "moisture permeability" are related to the properties and structure of fibers, and can cooperate with each other to improve comfort in wearing experience, and the target functions have interaction; while "wear resistance and washing resistance" and "soft touch" may have certain contradictions, because high-density fabric and surface coating for wear resistance may reduce softness, finally, the function items are clustered according to the interaction of different functions to establish the corresponding function subset.

[0039] Preferably, the function subset is established through a process integrability index, i.e. evaluating whether the processes of each function item can be integrated together, if the implementation processes of certain function items can be completed in the same production process, or can be compatible with each other, they are classified into a category, such as low modulus fiber, cavity structure, and surface fluffing treatment for realizing "soft touch" and "moisture permeability" functions, which can be realized together in the process of fiber processing and post-processing to a certain extent, and have process integrability, and then the corresponding function subset is established.

[0040] Preferably, through cluster analysis, the functional items with similar structural responses, target functional interactions and process integrability are respectively combined and classified to obtain multiple types of functional modules, i.e. to form multiple different functional subsets, so as to avoid a structural layer bearing multiple conflicting functions, wherein each type of function is responsible for a corresponding structural partition, and the hierarchical partition structure combination is finally established through discriminant combination, so that the structural design of the lining cloth is more scientific and reasonable, and the functional requirements can be better met, i.e. "resilience" and "shape retention" form a functional subset for constructing a memory support layer; "soft touch" and "moisture absorption and air permeability" form another functional subset for constructing a soft and cushioning layer; "abrasion resistance and washing resistance" form a functional subset for constructing a durable packaging layer.

[0041] In step S300, key indicators are extracted for each hierarchical partition structure in the hierarchical partition structure combination, and the weight compensation of the key indicator extraction result is performed by using the functional requirement matrix to establish a multi-objective evaluation function mapped with the hierarchical partition structure.

[0042] Preferably, key indicators are extracted for each hierarchical partition structure in the hierarchical partition structure combination, specifically, the key indicators of the memory support layer are extracted, which can include the orientation degree of fibers (high orientation degree is beneficial to resilience and shape retention), crystallinity (high crystallinity enhances stability), dimensional stability after heat treatment, etc. to accurately describe the performance characteristics of the memory support layer; the key indicators of the soft and cushioning layer are extracted, which can include the modulus of fibers (low modulus fibers are softer), the size and distribution of cavity structures (affecting air permeability), the degree of surface fluff (related to softness), etc. to reflect the ability of the soft and cushioning layer in realizing the corresponding functions; the key indicators of the durable packaging layer are extracted, which can include the density of the fabric (high-density fabric is more wear-resistant), the thickness and hardness of the surface coating (affecting wear resistance and washing resistance), the degree of resin cross-linking (related to structural stability and washing resistance), etc. to measure the performance of the durable packaging layer.

[0043] Preferably, the weight of the key indicators is compensated and adjusted according to the correlation between different functions in the function requirement matrix and the requirements of the actual application scene, so that the weight of the key indicators is more in line with the actual function requirements. For example, the weight of "soft touch" is medium, but in some close-fitting application scenarios, the key indicator weight of the soft and flexible buffer layer needs to be appropriately increased to better meet the use requirements. Then, the key indicators of each hierarchical partition structure are taken as variables, and the weight values after weight compensation are combined to construct the target evaluation function corresponding to each level, including establishing a double-objective function with memory retention target and hot rebound target as evaluation targets, establishing a three-objective function with comfort target, shock absorption effect target and deformability target as evaluation targets, establishing a four-objective function with anti-aging target, chemical resistance target, wear resistance target and ultraviolet resistance performance target as evaluation targets, that is, forming a multi-objective evaluation function, which is used to comprehensively and comprehensively evaluate the performance of each hierarchical partition structure in meeting the functional requirements, and ensure that the overall performance of the lining cloth is optimal.

[0044] Step S400, after configuring the parameter space, the multi-objective evaluation function is used for hierarchical optimization of the corresponding hierarchical partition structure, and a parameter optimization candidate set is established.

[0045] Step S400 further includes step S410 of obtaining a first multi-objective evaluation function corresponding to the memory support layer, the first multi-objective evaluation function being a double-objective function, and the evaluation targets of the first multi-objective evaluation function including memory retention target and hot rebound target; step S420 of randomly establishing an initial parameter set, each individual in the initial parameter set representing a set of parameter combinations of heat treatment and rebound control; step S430 of calculating the fitness value of each individual in the initial parameter set by using the first multi-objective evaluation function, and then performing individual non-dominated sorting and congestion calculation; step S440 of configuring a random probability according to the individual non-dominated sorting result and the congestion calculation result, and then performing individual selection, crossover and mutation, and iterative updating of the initial parameter set, and establishing a parameter optimization candidate set with the iterative updating result.

[0046] Preferably, the core function of the memory support layer is determined by two objectives of "memory retention" and "thermal resilience", and a double objective function including the memory retention objective and the thermal resilience objective is constructed as the corresponding first multi-objective evaluation function, wherein the memory retention objective measures the ability of the lining cloth to maintain the original shape after multiple deformations, which is usually related to fiber crystallinity, orientation and heat treatment process (such as high temperature setting temperature / time), and the thermal resilience objective measures the speed and degree of deformation recovery of the lining cloth after heating, which is related to the elastic recovery ability of the fiber molecular chain, the crosslinking density and other parameters; a set of individuals (such as 100 samples) are randomly generated in the parameter feasible region, and then a set of parameter combinations representing thermal treatment and resilience control are regarded as "individuals" to establish an initial parameter set, wherein the parameters may include heat treatment temperature, time, resilience control agent concentration, etc.

[0047] Preferably, according to the first multi-objective evaluation function, the two objective values of each individual are calculated, if the objective is maximization (such as memory retention rate, thermal resilience rate), the function value is directly used, if the objective is minimization (such as permanent deformation rate), the reciprocal conversion is needed; then the individual non-dominated sorting is performed, assuming that individual A dominates individual B, when and only when the two objectives of A are not worse than B, and at least one objective is better than B, the individuals in the population are layered according to the non-dominated level (the first layer is the Pareto frontier individual, the second layer is the individual dominated by the first layer, and so on), and the individuals of better level are preferentially retained for selection; then the crowding degree is calculated, which is used to measure the distribution density of individuals in the same non-dominated layer, to avoid the population converging to a local optimal solution, and the larger the crowding degree value, the "sparser" the individual around, and the better the diversity.

[0048] Preferably, according to the individual non-dominated sorting result and the crowding degree calculation result, a random probability is configured, that is, according to the non-dominated sorting result and the crowding degree, a selection probability is assigned to each individual (the higher the level and the larger the crowding degree, the higher the probability of being selected), and the parent individuals are screened out, and then individual crossover and mutation are performed to update the initial parameter set, including randomly exchanging part of the parameters (such as heat treatment temperature and time) of the selected parent individuals to generate child individuals, and randomly adjusting some parameters of the individuals by a small amplitude (such as ±5% temperature fluctuation) to avoid premature convergence of the algorithm; the parent and child individuals are combined, and the non-dominated sorting and crowding degree calculation are performed again, and the top N excellent individuals (such as 80) are retained as the next generation population, and the iteration is performed until the termination condition (such as the upper limit of the number of iterations, the convergence of the Pareto frontier) is met; then all the non-dominated individuals in the last generation population are used to construct a parameter optimization candidate set (i.e. the Pareto optimal solution set), and each candidate parameter combination corresponds to the optimal trade-off solution of the memory support layer under the "memory retention" and "thermal resilience" objectives, and the most suitable parameter combination can be selected according to the actual production demand (such as cost, efficiency).

[0049] Further, the step S440 further comprises a step S441 of assigning individual hierarchical selection probability by using the individual non-dominated sorting result; a step S442 of establishing individual crowding selection probability according to the current iteration number and the crowding degree calculation result; and a step S443 of performing individual selection, crossover and mutation after configuring random probability by using the individual hierarchical selection probability and the individual crowding selection probability.

[0050] Preferably, all individuals in the population are non-dominated sorted to divide the individuals into different levels, wherein the individuals in the first level (Pareto front solution) are not dominated by any other individual, the individuals in the second level are dominated by the individuals in the first level but not dominated by other individuals in the same level, and so on, and the higher the level, the worse the comprehensive performance of the individual; the individuals in the better level (such as the first level) have a higher probability of being selected as the parent generation to guide the population to converge to the Pareto front, and then a selection probability is assigned to each level, and the selection probability is inversely proportional to the level, that is, the higher the level (the larger the numerical value), the lower the selection probability.

[0051] Preferably, the individual crowding selection probability is established according to the current iteration number and the crowding degree calculation result, specifically, the individuals with large crowding degree are selected within the same level to avoid the population converging to a local dense area and maintain the diversity of solutions, for example, the individuals in the same level are sorted according to the crowding degree from high to low, and the crowding selection probability is calculated. Finally, the hierarchical selection probability and the crowding selection probability are combined to obtain the final selection probability of each individual, such as hierarchical selection + crowding sorting, that is, the individuals are first screened according to the level, and then the individuals in the same level are sorted according to the crowding degree to assign the probability; then the selected parent individuals are exchanged by single-point crossover or uniform crossover to generate child individuals; the parameters of the individuals are slightly randomly disturbed (such as ±3% fluctuation of heat treatment time) to introduce new solution space exploration; in each generation, the individuals that are both optimal and diverse are retained to make the iteration process consider both optimization efficiency and comprehensiveness.

[0052] Further, the step S400 further comprises a step S450 of obtaining a second multi-objective evaluation function corresponding to the compliant buffer layer, the second multi-objective evaluation function being a three-objective function, and the evaluation targets of the second multi-objective evaluation function including comfort target, shock absorption effect target and deformability target; and a step S460 of searching and optimizing the compliant buffer layer by using the second multi-objective evaluation function to establish a parameter optimization candidate set.

[0053] Preferably, a three-objective evaluation function (second multi-objective evaluation function) is used to evaluate the comfort, shock absorption effect and deformability of the soft buffer layer, and a set of optimal design parameters (such as material elastic modulus, buffer layer thickness, porosity, etc.) is determined through search optimization to make the three objectives reach a relatively optimal level. Specifically, the comfort objective (such as tactile comfort score) is usually maximized, the shock absorption effect objective (such as vibration attenuation rate) is usually maximized, and the deformability objective (satisfying a specific deformation range, such as compression rate between 20% and 50%) is usually interval optimized. Then, according to the design requirements, weights are assigned to each objective, a set of initial parameter individuals is randomly generated within the parameter feasible region (such as material elastic modulus 10≤E≤100MPa, thickness 5≤t≤30mm) to form an initial population. Each individual of the initial population is substituted into the second multi-objective evaluation function, and the objective values are converted into fitness values according to the objective type (maximization / interval optimization). Iteration is performed through selection, crossover and mutation until the termination condition (such as the number of iterations reaching a set value, population convergence) is met. The retained high-performance individuals (such as non-dominated solutions in the last generation population) are used as the parameter optimization candidate set.

[0054] Further, step S400 further includes step S470 of obtaining a third multi-objective evaluation function corresponding to the durable packaging layer. The third multi-objective evaluation function is a four-objective function, and the evaluation objectives of the third multi-objective evaluation function include an anti-aging objective, a chemical resistance objective, a wear resistance objective and an ultraviolet resistance performance objective. Step S480 uses the third multi-objective evaluation function to perform search optimization of the durable packaging layer to establish a parameter optimization candidate set.

[0055] Preferably, a four-objective evaluation function (third multi-objective evaluation function) is used to evaluate the anti-aging, chemical resistance, wear resistance and ultraviolet resistance performance of the durable packaging layer. The anti-aging objective is used to measure the ability of the material to resist performance degradation over a long period of use (such as strength retention rate over the life cycle), the chemical resistance objective is used to evaluate the ability to resist corrosion by chemicals such as acid, alkali and solvent (such as weight change rate after immersion), the wear resistance objective is used to represent the ability to resist friction loss (such as thickness loss after standard friction test), and the ultraviolet resistance performance objective is used to measure the ability to resist degradation or discoloration caused by ultraviolet radiation (such as color difference or strength retention rate after ultraviolet aging test). The parameters to be optimized corresponding to the objective evaluation function include material composition ratio, processing temperature, coating thickness, etc. A set of initial parameter combinations is randomly generated to construct an initial population. For each individual, the third multi-objective evaluation function is used to calculate its performance values in the four objectives. Since multiple objectives may conflict with each other (such as improving anti-aging may reduce chemical resistance), the non-dominated sorting is used to determine the superiority-inferiority relationship of the individuals. Then, through iterative optimization (such as selection, crossover and mutation operations), individuals that perform better in multiple objectives are gradually screened out to form a parameter optimization candidate set.

[0056] Step S500, based on the parameter optimization candidate set, the hierarchical partition structure combination hierarchical control, using finite element simulation for composite structure and function demand matrix response matching.

[0057] Step S500 further includes step S510, according to the hierarchical partition structure combination to establish a multi-layer composite finite element simulation structure; step S520, the parameter optimization candidate set as input data, input the multi-layer composite finite element simulation structure; step S530, calling coupling solver for thermal-mechanical coupling, viscoelastic-thermal response coupling, generating simulation results; step S540, using the simulation results for response matching of functional demand matrix.

[0058] Preferably, the memory support layer, soft buffer layer, durable packaging layer and other multi-layer structure are combined, specifically, according to the physical size (such as thickness, shape) and spatial position of each layer material, the three-dimensional geometric model of each layer is established, and then the thermal-mechanical coupling properties (such as elastic modulus with temperature change, thermal expansion coefficient) of the memory support layer, the viscoelastic parameters (such as relaxation modulus) are defined; the nonlinear elastic properties (such as super-elastic constitutive model) of the soft buffer layer, the shock absorption related parameters (such as damping coefficient), the deformability parameters (such as Poisson's ratio) are defined; the aging related properties (such as aging model parameters) of the durable packaging layer, the chemical resistance parameters (such as diffusion coefficient), the wear resistance parameters (such as wear index), the ultraviolet resistance parameters (such as ultraviolet absorption coefficient) are defined; and then a complete composite structure model is formed, and the finite element mesh is divided for each layer, and the grid is encrypted in the key area (such as the interface between layers, stress concentration) to improve the calculation accuracy.

[0059] Preferably, the parameter optimization candidate set is input as input data, that is, each individual (parameter combination) in the candidate set is converted into input parameters recognizable by the finite element model, for example, the memory support layer is converted into heat treatment temperature, cooling rate, rebound control pressure, etc., the soft buffer layer is converted into material formula proportion, foaming density, structure porosity, etc., the durable packaging layer is converted into coating thickness, crosslinking agent content, ultraviolet protection agent concentration, etc., and then input into the multi-layer composite finite element simulation structure, call the coupling solver for thermal-mechanical coupling, viscoelastic-thermal rebound response coupling, including calculating the influence of temperature field on the mechanical properties of materials (such as the softening of materials at high temperature leading to the decrease of elastic modulus), solving the thermal stress distribution; simulate the viscoelastic behavior (such as creep, stress relaxation) of the memory support layer in the loading-unloading process, and couple the temperature field to calculate the thermal rebound deformation, solve the rebound displacement combined with the temperature history, and finally generate the simulation results, as shown in Table 2, the output simulation result example data:

[0060] Table 2 Simulation result key output data

[0061] Hierarchy Core output parameters Memory support layer Thermal resilience, memory retention, stress relaxation, temperature-strain curve Compliant cushioning layer Impact absorption energy, contact stress distribution, compression permanent set, recoverable deformation Durable packaging layer Elastic modulus attenuation rate after aging, chemical corrosion depth, wear amount, color difference after ultraviolet irradiation

[0062] Preferably, the simulation results are used to respond to the matching of the functional requirement matrix, that is, the indicators in the simulation results (such as thermal resilience, impact absorption energy) are compared with the corresponding items in the functional requirement matrix, if the simulation indicators meet the demand threshold (such as thermal resilience ≥ 85%), it is marked as "matching success", if not, the deviation value is recorded (such as thermal resilience is only 80%, deviation -5%), then the multi-level indicators are weighted and summed, the overall function matching degree is evaluated (such as memory support layer weight 40%, soft buffer layer 30%, durable packaging layer 30%), the parameter combination that meets the multi-objective demand at the same time (such as the candidate set individual that takes into account high resilience, high impact absorption energy and ultraviolet resistance) is screened out, so as to realize the multi-objective optimization of the target polyester lining composite material structure.

[0063] Step S600, according to the response matching result, the turning deviation is established, the parameter optimization candidate set is updated by using the turning deviation, and the preparation optimization is completed according to the update result.

[0064] Step S600 further includes step S610 of establishing a local search guide strategy based on the turning deviation; and step S620 of using the local search guide strategy to perform local search optimization of the parameter optimization candidate set, so as to update the parameter optimization candidate set.

[0065] Preferably, according to the response matching result, the turning deviation is established, that is, by comparing the difference between the simulation result and the actual functional requirement, the "mismatching degree" of the parameter configuration is quantified, the deviation index guiding optimization is formed, specifically, the design target (such as comfort, anti-aging, etc.) is converted into measurable indicators (such as stress distribution, deformation rate, life cycle, etc.), then for each evaluation target (such as sub-target in three / four target function), the absolute error or relative error between the simulation result and the target value is calculated, and the turning deviation is formed. Then the turning deviation is used to update the parameter optimization candidate set, that is, through sensitivity analysis (such as gradient descent, variance analysis), the key parameters that have the greatest impact on the deviation are identified, which may be material thickness, elastic modulus, formula proportion, etc., then the parameters are adjusted in the direction of decreasing deviation (such as reducing the related parameters when the simulation value is higher than the target value; otherwise, increase), or set the search step (such as ± 10%) in the current parameter neighborhood, avoid jumping out of the feasible solution region, and use local optimization algorithm (such as gradient descent, pattern search, etc.) to iteratively update the parameters under the guidance of the deviation.

[0066] Preferably, fine search is performed for high deviation areas to generate better parameter combinations. Specifically, parameter combinations with deviations exceeding a threshold are found from the candidate set as the starting point of local search. For each high deviation parameter combination, a number of neighborhood parameters are generated according to the guidance strategy (e.g., adjusting 1 to 2 key parameters), forming new candidate solutions. The newly generated parameter combinations are added to the candidate set, and the old solutions with larger deviations are removed to maintain the size of the candidate set stable. Finally, the preparation optimization is completed according to the updated results, that is, the parameter combinations after simulation-deviation correction-research are converted into actual preparation schemes, and the production process (such as material ratio, molding temperature, machining precision, etc.) is adjusted according to the optimized parameters, and finally the preparation optimization of high memory polyester backing cloth is completed, and the production efficiency is ensured.

[0067] In the foregoing, reference is made to Figure 1 A high memory polyester backing cloth preparation optimization method according to an embodiment of the application is described in detail. Next, a high memory polyester backing cloth preparation optimization system according to an embodiment of the application will be described with reference to Figure 2 A high memory polyester backing cloth preparation optimization system according to an embodiment of the application will be described with reference to

[0068] A high memory polyester backing cloth preparation optimization system according to an embodiment of the application is used to solve the technical problems of difficulty in adapting polyester backing cloth to diversified application scenarios and difficulty in balancing performance indicators in the prior art, which leads to poor scientificity and production efficiency of high memory polyester backing cloth preparation. It achieves the technical effect of improving the scientificity and production efficiency of the preparation process. As shown in Figure 2 A high memory polyester backing cloth preparation optimization system includes a scene adaptation analysis module 10, a matrix adaptation analysis module 20, a key indicator extraction module 30, a hierarchical optimization module 40, a response matching module 50, and a turning deviation establishment module 60.

[0069] The scene adaptation analysis module 10 is configured to perform scene adaptation analysis after receiving a product application scene of a target polyester lining cloth, and establish a function demand matrix; the matrix adaptation analysis module 20 is configured to activate a structure response table of a structure partition function, perform adaptation analysis on the function demand matrix by using the structure response table, establish a hierarchical partition structure combination, and the hierarchical partition structure combination includes a memory support layer, a flexible buffer layer and a durable packaging layer; the key index extraction module 30 is configured to extract key indexes of each hierarchical partition structure in the hierarchical partition structure combination, and compensate the key index extraction result by using the function demand matrix, and establish a multi-objective evaluation function mapped with the hierarchical partition structure; the hierarchical optimization module 40 is configured to perform hierarchical optimization of the corresponding hierarchical partition structure by using the multi-objective evaluation function after configuring a parameter space, and establish a parameter optimization candidate set; the response matching module 50 is configured to perform response matching of the composite structure and the function demand matrix by using finite element simulation based on hierarchical control of the hierarchical partition structure combination by using the parameter optimization candidate set; and the turning deviation establishing module 60 is configured to establish a turning deviation according to the response matching result, update the parameter optimization candidate set by using the turning deviation, and complete preparation optimization according to the update result.

[0070] In the following, the specific configuration of the scene adaptation analysis module 10 will be described in detail. The scene adaptation analysis module 10 further includes: multi-dimensional deconstruction of the product application scene to establish a scene dimension matrix, and the multi-dimensional deconstruction includes use purpose deconstruction, wearing scene deconstruction, use frequency and duration deconstruction, washing and protection requirement deconstruction, and touch demand deconstruction; establishing a target function set, and calling a semantic association model to perform semantic matching analysis of each scene dimension in the scene dimension matrix and the target function set, and establishing a function demand matrix.

[0071] In the following, the specific configuration of the matrix adaptation analysis module 20 will be described in detail. The matrix adaptation analysis module 20 further includes: mapping each function demand in the function demand matrix to a structure unit with the highest response value according to the structure response table, establishing a mapping path; performing function item clustering analysis in the function demand matrix by using the mapping path, establishing a function subset through a structure response similarity index, a target function interaction index and a process integrability index, and establishing a hierarchical partition structure combination by discriminating the combination relationship of the function subset in the structure hierarchy.

[0072] In the following, the specific configuration of the hierarchical optimization module 40 will be described in detail. The hierarchical optimization module 40 further comprises: obtaining a first multi-objective evaluation function corresponding to the memory support layer, the first multi-objective evaluation function being a double-objective function, and the evaluation objectives of the first multi-objective evaluation function including a memory retention objective and a thermal springback objective; randomly establishing an initial parameter set, each individual in the initial parameter set representing a combination of parameters for heat treatment and springback control; after calculating the fitness value of each individual in the initial parameter set using the first multi-objective evaluation function, performing individual non-dominated sorting and congestion calculation; after configuring a random probability according to the individual non-dominated sorting result and the congestion calculation result, performing individual selection, crossover and mutation, and iteratively updating the initial parameter set to establish a parameter optimization candidate set.

[0073] In the following, the specific configuration of the hierarchical optimization module 40 will be described in detail. The hierarchical optimization module 40 further comprises: assigning an individual hierarchical selection probability using the individual non-dominated sorting result; establishing an individual congestion selection probability according to the current iteration number and the congestion calculation result; after configuring a random probability using the individual hierarchical selection probability and the individual congestion selection probability, performing individual selection, crossover and mutation.

[0074] In the following, the specific configuration of the hierarchical optimization module 40 will be described in detail. The hierarchical optimization module 40 further comprises: obtaining a second multi-objective evaluation function corresponding to the compliant buffer layer, the second multi-objective evaluation function being a three-objective function, and the evaluation objectives of the second multi-objective evaluation function including a comfort objective, a shock absorption effect objective, and a deformability objective; performing search optimization of the compliant buffer layer using the second multi-objective evaluation function to establish a parameter optimization candidate set.

[0075] In the following, the specific configuration of the hierarchical optimization module 40 will be described in detail. The hierarchical optimization module 40 further comprises: obtaining a third multi-objective evaluation function corresponding to the durable packaging layer, the third multi-objective evaluation function being a four-objective function, and the evaluation objectives of the third multi-objective evaluation function including an anti-aging objective, a chemical resistance objective, a wear resistance objective, and an ultraviolet resistance performance objective; performing search optimization of the durable packaging layer using the third multi-objective evaluation function to establish a parameter optimization candidate set.

[0076] In the following, the specific configuration of the response matching module 50 will be described in detail. The response matching module 50 further comprises: establishing a multi-layer composite finite element simulation structure according to the hierarchical partition structure combination; inputting the parameter optimization candidate set as input data into the multi-layer composite finite element simulation structure; calling a coupling solver to perform thermal-mechanical coupling and viscoelastic-thermal springback response coupling to generate a simulation result; and using the simulation result to perform response matching on the functional requirement matrix.

[0077] The specific configuration of the rotation deviation establishing module 60 will be described in detail below. The rotation deviation establishing module 60 further comprises: establishing a local search guide strategy based on the rotation deviation; and performing local search optimization on the parameter optimization candidate set by using the local search guide strategy to update the parameter optimization candidate set.

[0078] The preparation optimization system of the high-memory polyester lining cloth provided in the embodiments of the application can execute the preparation optimization method of the high-memory polyester lining cloth provided in any of the embodiments of the application, has the function modules and beneficial effects corresponding to the execution method.

[0079] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server, the various units and modules included are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for the convenience of mutual differentiation, and do not serve to limit the protection scope of the present application.

[0080] The specific embodiments described above do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for optimizing the preparation of a high-memory polyester base cloth, characterized in that, The method comprises: After receiving the product application scene of the target polyester lining cloth, scene adaptation analysis is performed, a functional requirement matrix is established, including: Multi-dimensional deconstruction is performed on the product application scene, a scene dimension matrix is established, and the multi-dimensional deconstruction includes use purpose deconstruction, wearing scene deconstruction, use frequency and duration deconstruction, washing and protection requirement deconstruction, and touch demand deconstruction; A target function set is established, and semantic matching analysis of each scene dimension in the scene dimension matrix and the target function set is performed by calling a semantic association model to establish a functional requirement matrix; A structure response table of structure partition function is activated, the functional requirement matrix is adapted by using the structure response table, a hierarchical partition structure combination is established, and the hierarchical partition structure combination includes a memory support layer, a soft buffer layer, and a durable packaging layer; The structure response table is a pre-established correspondence table used to clearly show the relationship between different lining cloth functions and structure elements for realizing the functions; the content of the structure response table of the structure partition function includes that high orientation, high crystalline fiber and heat treatment structure determine the resilience and shape retention of the lining cloth, low modulus fiber, cavity structure and surface fluffing treatment determine the soft touch and moisture permeability of the lining cloth, and high density fabric and surface coating / resin crosslinking determine wear resistance and washing resistance; Key indicators are extracted from each hierarchical partition structure in the hierarchical partition structure combination, and the key indicator extraction result is compensated by using the weight of the functional requirement matrix to establish a multi-objective evaluation function mapped with the hierarchical partition structure; The key indicators of each hierarchical partition structure are used as variables, and the weight value after weight compensation is combined to construct a target evaluation function corresponding to each hierarchy, including: a double-objective function with memory retention target and heat resilience target, a three-objective function with comfort target, shock absorption effect target and deformability target, and a four-objective function with anti-aging target, chemical resistance target, wear resistance target and anti-ultraviolet performance target, i.e. a multi-objective evaluation function is formed, which is used to comprehensively and comprehensively evaluate the performance of each hierarchical partition structure in meeting the functional requirements, and ensure that the overall performance of the lining cloth is optimal; After the parameter space is configured, the multi-objective evaluation function is used to perform hierarchical optimization on the corresponding hierarchical partition structure to establish a parameter optimization candidate set; Based on the hierarchical control of the hierarchical partition structure combination based on the parameter optimization candidate set, finite element simulation is used to perform response matching of the composite structure and the functional requirement matrix; According to the response matching result, a turning deviation is established, the parameter optimization candidate set is updated by using the turning deviation, and the preparation optimization is completed according to the update result.

2. The optimized method for preparing a high-memory polyester interlining as described in claim 1, characterized in that, The functional requirement matrix is adapted by using the structure response table, and the hierarchical partition structure combination is established, including: According to the structure response table, each item of functional requirement in the functional requirement matrix is mapped to a structure unit with the highest response value to establish a mapping path; Performing function item clustering analysis in the function requirement matrix by using the mapping path, establishing a function subset by a structure response similarity index, a target function interaction index, and a process integrability index, performing structure level combination relationship discrimination on the function subset, and establishing a hierarchical partition structure combination.

3. The optimized method for preparing a high-memory polyester interlining as described in claim 1, characterized in that, After the parameter space is configured, performing hierarchical optimization on the corresponding hierarchical partition structure by using the multi-objective evaluation function, and establishing a parameter optimization candidate set, including: Obtaining a first multi-objective evaluation function corresponding to the memory support layer, the first multi-objective evaluation function being a double-objective function, and evaluation targets of the first multi-objective evaluation function including a memory retention target and a thermal springback target; Randomly establishing an initial parameter set, each individual in the initial parameter set representing a combination of a group of heat treatment and springback control parameters; After calculating the fitness value of each individual in the initial parameter set by using the first multi-objective evaluation function, performing individual non-dominated sorting and congestion calculation; After configuring a random probability according to the individual non-dominated sorting result and the congestion calculation result, performing individual selection, crossover and mutation, and iteratively updating the initial parameter set to establish a parameter optimization candidate set.

4. The optimized method for preparing a high-memory polyester interlining as described in claim 3, characterized in that, After configuring a random probability according to the individual non-dominated sorting result and the congestion calculation result, performing individual selection, crossover and mutation, including: Distributing individual hierarchical selection probability by using the individual non-dominated sorting result; Establishing individual congestion selection probability according to the current iteration number and the congestion calculation result; After configuring a random probability by using the individual hierarchical selection probability and the individual congestion selection probability, performing individual selection, crossover and mutation.

5. The optimized method for preparing a high-memory polyester interlining as described in claim 1, characterized in that, After the parameter space is configured, performing hierarchical optimization on the corresponding hierarchical partition structure by using the multi-objective evaluation function, and establishing a parameter optimization candidate set, further including: Obtaining a second multi-objective evaluation function corresponding to the compliant buffer layer, the second multi-objective evaluation function being a three-objective function, and evaluation targets of the second multi-objective evaluation function including a comfort target, a shock absorption effect target, and a deformability target; Performing search optimization on the compliant buffer layer by using the second multi-objective evaluation function to establish a parameter optimization candidate set.

6. The optimized method for preparing a high-memory polyester interlining as described in claim 1, characterized in that, After the parameter space is configured, performing hierarchical optimization on the corresponding hierarchical partition structure by using the multi-objective evaluation function, and establishing a parameter optimization candidate set, further including: Obtaining a third multi-objective evaluation function corresponding to the durable packaging layer, the third multi-objective evaluation function being a four-objective function, and evaluation targets of the third multi-objective evaluation function including an anti-aging target, a chemical resistance target, a wear resistance target, and an ultraviolet resistance performance target; Performing search optimization on the durable packaging layer by using the third multi-objective evaluation function to establish a parameter optimization candidate set.

7. The optimized method for preparing a high-memory polyester interlining as described in claim 1, characterized in that, After the hierarchical partition structure combination is controlled based on the parameter optimization candidate set, performing response matching of the composite structure and the function requirement matrix by using finite element simulation, including: Establishing a multi-layer composite finite element simulation structure according to the hierarchical partition structure combination; Inputting the parameter optimization candidate set as input data into the multi-layer composite finite element simulation structure; Calling a coupling solver to perform thermal-mechanical coupling and viscoelastic-thermal springback response coupling to generate a simulation result; The simulation result is used for response matching of the function demand matrix.

8. The optimized method for preparing a high-memory polyester interlining as described in claim 1, characterized in that, The parameter optimization candidate set is updated by using the rotation deviation. A local search guide strategy is established based on the rotation deviation. The parameter optimization candidate set is locally searched and optimized by using the local search guide strategy, so as to update the parameter optimization candidate set.

9. A system for optimizing the preparation of a high memory polyester scrim, characterized in that, The system is used for implementing the preparation optimization method of the high-memory polyester lining cloth according to any one of claims 1 to 8, and the system comprises: A scene adaptation analysis module is configured to perform scene adaptation analysis after receiving a product application scene of a target polyester lining cloth, and to establish a function demand matrix. A matrix adaptation analysis module is configured to activate a structure response table of a structure partition function, to perform matrix adaptation analysis on the function demand matrix by using the structure response table, to establish a hierarchical partition structure combination, and to include a memory support layer, a flexible buffer layer and a durable packaging layer in the hierarchical partition structure combination. A key index extraction module is configured to extract key indexes of each hierarchical partition structure in the hierarchical partition structure combination, to perform weight compensation on the key index extraction result by using the function demand matrix, and to establish a multi-objective evaluation function mapped with the hierarchical partition structure. A hierarchical optimization module is configured to perform hierarchical optimization of the corresponding hierarchical partition structure by using the multi-objective evaluation function after configuring a parameter space, and to establish a parameter optimization candidate set. A response matching module is configured to perform response matching of a composite structure and the function demand matrix by using finite element simulation based on hierarchical control of the hierarchical partition structure combination by using the parameter optimization candidate set. A rotation deviation establishment module is configured to establish a rotation deviation according to a response matching result, to update the parameter optimization candidate set by using the rotation deviation, and to complete preparation optimization according to an update result.

Images