Simulation design method for dynamic impact strength of luggage pull rod

Abstract

The invention is suitable for the technical field of luggage design, and particularly relates to a luggage pull rod dynamic impact strength simulation design method, which comprises the following steps: ensuring the reliability of a design result by obtaining a structure attribute set, a historical attribute set and a design attribute set of a target design luggage; modeling simulation processing is carried out based on the structure attribute set and the historical attribute set, accurate recognition of the high strain energy density area and the geometric weak position of the pull rod is achieved, a feature response set containing the geometric information and mechanical response data of the weak area is obtained, and a clear target spot is pointed out for optimization design; according to the characteristic response set and the design attribute set, variable density lattice structure generation processing is carried out on the high strain energy density area, a traditional empirical design and repeated physical test mode is replaced, the research and development period is greatly shortened, the trial and error cost is reduced, the dynamic impact strength and structural stability of the luggage pull rod are improved, and the service life of the luggage pull rod is prolonged. And reliable technical support is provided for high-performance, light-weight and low-cost design of the luggage pull rod.

Description

Technical Field

[0001] This application belongs to the field of luggage design technology, and in particular relates to a simulation design method for the dynamic impact strength of luggage handles. Background Technology

[0002] As a core functional component of luggage, the pull rod directly affects the user experience and lifespan of the luggage, and its dynamic impact strength is a key performance indicator. In daily use, the pull rod frequently faces dynamic loads such as collisions, drops, and compression, which can easily lead to problems such as tube bending and breakage, locking mechanism failure, and joint jamming.

[0003] Existing luggage handle designs largely rely on empirical selection and static mechanical testing, which has significant limitations. On the one hand, the design process lacks comprehensive coverage of dynamic impact conditions, verifying performance only through physical tests under a few fixed conditions. This makes it difficult to simulate the complex and varied impact energy, direction of impact, and point of impact in actual use, resulting in insufficient design specificity and a high risk of failure in actual use. On the other hand, traditional optimization methods often employ crude approaches such as increasing wall thickness and changing materials. While these methods can improve strength, they are often accompanied by increased weight and cost, and cannot accurately locate and strengthen weak areas of the structure. Traditional methods also have long physical testing cycles and high material costs, making it difficult to support rapid iteration of batch design schemes. This restricts the improvement of handle design quality and R&D efficiency, and fails to meet consumers' core demands for lightweight, high-strength, and long-life luggage. Summary of the Invention

[0004] This application provides a simulation design method for the dynamic impact strength of luggage handles, which can solve the problems of existing luggage handle designs relying on experience and static testing, resulting in incomplete coverage of dynamic working conditions, weak optimization targeting, difficulty in balancing lightweight and strength, and low R&D efficiency.

[0005] In a first aspect, embodiments of this application provide a simulation design method for the dynamic impact strength of a luggage handle, including: Obtain the structural attribute set, historical attribute set, and design attribute set of the target bag design; Modeling and simulation are performed based on the structural attribute set and historical attribute set. The action process of loads on the tie rod target component under different energy levels and directions is simulated through explicit dynamic analysis. The high strain energy density region and geometric weak location information of the tie rod target component are extracted to obtain the characteristic response set. Based on the characteristic response set and the design attribute set, a variable density lattice structure generation process is performed on the high strain energy density region to obtain an optimized design scheme set; the optimized design scheme set includes at least one non-uniform lattice filling topology configuration scheme for the interior or surface of the target component.

[0006] The technical solutions described in this application embodiment have at least the following technical effects: The dynamic impact strength simulation design method for luggage handles provided in this application obtains the structural attribute set, historical attribute set, and design attribute set of the target luggage design. This provides comprehensive and standardized data support for subsequent modeling, simulation, and optimization design, ensuring that the entire design process is based on accurate structural parameters, real historical failure data, and clear performance requirements, thus guaranteeing the reliability of the design results. Modeling and simulation processing based on the structural and historical attribute sets, leveraging the advantages of explicit dynamic analysis, can accurately simulate the action of loads at different energy levels and directions on the handle target component. This overcomes the limitation of traditional static testing in reproducing dynamic impact responses, enabling precise identification of high strain energy density regions and geometrically weak locations on the handle. It yields a characteristic response set containing geometric information and mechanical response data of the weak areas, providing clear guidance for optimization design. The target area is a variable density lattice structure generated based on the characteristic response set and design attribute set. Through non-uniform lattice filling topology design, mechanical properties are allocated as needed. High-density lattices are used in high-stress areas to enhance load-bearing capacity, while low-stress areas use low-density lattices to control weight, effectively balancing the dynamic impact strength and lightweight requirements of the pull rod. This avoids the weight redundancy or insufficient strength problems caused by traditional extensive optimization. The resulting optimized design scheme set provides multiple targeted topology options, constructing an intelligent design closed loop that replaces the traditional experience-based design and repeated physical testing model. This significantly shortens the R&D cycle, reduces trial-and-error costs and material consumption, and significantly improves the dynamic impact strength, structural stability, and service life of luggage pull rods, providing reliable technical support for high-performance, lightweight, and low-cost design of luggage pull rods.

[0007] In a second aspect, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method as described in any of the first aspects above.

[0008] It is understandable that the beneficial effects of the second aspect mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1This is a flowchart illustrating the simulation design method for dynamic impact strength of luggage handles according to an embodiment of this application; Figure 2 This is a schematic diagram of the impact energy of the dynamic impact strength simulation design method for luggage handles provided in an embodiment of this application; Figure 3 This is a schematic diagram of the strain energy density of the simulation design method for dynamic impact strength of luggage handles provided in an embodiment of this application; Figure 4 This is a relative density schematic diagram of the simulation design method for dynamic impact strength of luggage handles provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a dynamic impact strength simulation design system for luggage handles provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0011] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0012] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0013] Existing luggage handle designs largely rely on empirical selection and static mechanical testing, which has significant limitations. On the one hand, the design process lacks comprehensive coverage of dynamic impact conditions, verifying performance only through physical tests under a few fixed conditions. This makes it difficult to simulate the complex and varied impact energy, direction of impact, and point of impact in actual use, resulting in insufficient design specificity and a high risk of failure in actual use. On the other hand, traditional optimization methods often employ crude approaches such as increasing wall thickness and changing materials. While these methods can improve strength, they are often accompanied by increased weight and cost, and cannot accurately locate and strengthen weak areas of the structure. Traditional methods also have long physical testing cycles and high material costs, making it difficult to support rapid iteration of batch design schemes. This restricts the improvement of handle design quality and R&D efficiency, and fails to meet consumers' core demands for lightweight, high-strength, and long-life luggage.

[0014] To address the aforementioned issues, this application provides a simulation design method for the dynamic impact strength of luggage handles. This method acquires the structural attribute set, historical attribute set, and design attribute set of the target luggage design, providing comprehensive and standardized data support for subsequent modeling, simulation, and optimization design. This ensures the entire design process is based on accurate structural parameters, real historical failure data, and clear performance requirements, guaranteeing the reliability of the design results. Modeling and simulation processing based on the structural and historical attribute sets, leveraging the advantages of explicit dynamic analysis, can accurately simulate the action of loads at different energy levels and directions on the handle target component. This overcomes the limitation of traditional static testing in reproducing dynamic impact responses, enabling precise identification of high strain energy density regions and geometrically weak locations on the handle. A feature response set containing geometric information and mechanical response data of the weak regions is obtained, providing clear targets for optimization design. Based on the feature response set and design attribute set, a variable density lattice structure is generated for the high strain energy density region. Through non-uniform lattice filling topology design, mechanical properties are allocated as needed. High-stress areas utilize high-density lattice to enhance load-bearing capacity, while low-stress areas employ low-density lattice to control weight. This effectively balances the dynamic impact strength of the pull rod with the need for lightweight design, avoiding the weight redundancy or insufficient strength issues caused by traditional extensive optimization. The resulting optimized design scheme set provides multiple targeted topology configuration options, constructing an intelligent design closed loop that replaces the traditional experience-based design and repeated physical testing model. This significantly shortens the R&D cycle, reduces trial-and-error costs and material consumption, and significantly improves the dynamic impact strength, structural stability, and service life of luggage pull rods, providing reliable technical support for high-performance, lightweight, and low-cost design of luggage pull rods.

[0015] The dynamic impact strength simulation design method for luggage handles provided in this application embodiment can be applied to electronic devices. In this case, the electronic device is the executing subject of the dynamic impact strength simulation design method for luggage handles provided in this application embodiment. This application embodiment does not impose any restrictions on the specific type of electronic device.

[0016] It is understandable that electronic devices can be various intelligent devices. For example, electronic devices can be terminal devices such as laptops, ultra-mobile personal computers (UMPCs), netbooks, personal digital assistants (PDAs), and desktop computers.

[0017] To better understand the simulation design method for dynamic impact strength of luggage handles provided in this application, the specific implementation process of the simulation design method for dynamic impact strength of luggage handles provided in this application will be described below by way of example.

[0018] Figure 1 This paper presents a schematic flowchart illustrating the simulation design method for the dynamic impact strength of a luggage handle according to an embodiment of this application. The simulation design method for the dynamic impact strength of a luggage handle includes: S100: Obtain the structural attribute set, historical attribute set, and design attribute set of the target design bag.

[0019] It is understandable that the structural attribute set is the physical structure and material properties of the target luggage handle. It can cover the core structural parameters of the handle (such as the outer diameter, inner diameter, and wall thickness distribution of the handle tube, the number of handle sections and the length of each section, the gear / slot dimensions of the telescopic mechanism, the geometric parameters of the locking structure, etc.), key material parameters (such as the grade, density, elastic modulus, Poisson's ratio, dynamic yield strength, and fracture toughness of the tube material, the heat distortion temperature and impact strength of the plastic connectors, etc.), and assembly relationship parameters (such as the fit clearance of each handle section, the fastening method and preload of the connectors, and the structure and dimensions of the connection point between the handle and the luggage body). The historical attribute set is constructed based on past usage and testing data of luggage handles. It includes historical impact event records (such as energy level, load direction and location, and environmental temperature and humidity parameters at the time of impact), handle failure data (such as the location and mode of failure, such as fracture, deformation, and locking failure, and the cumulative number of uses or impacts at the time of failure), and post-failure morphology inspection data (such as the microscopic morphological characteristics of the fracture surface, dimensional changes in the deformation area, and stress trace distribution). The design attribute set consists of the performance requirements and design constraints of the handle. Performance requirements include mechanical performance indicators (such as impact strength threshold, repeated expansion and contraction fatigue life, and the load-bearing capacity of the locking structure), lightweight indicators (such as the upper limit of the overall weight of the handle), and durability indicators (such as corrosion resistance and wear resistance requirements). Design constraints include dimensional constraints (such as the total length of the handle after retraction, the maximum length after expansion, and cross-sectional size limitations), process constraints (such as machinability requirements and upper limits for material procurement and manufacturing costs), and appearance constraints (such as surface roughness, color matching, and absence of obvious protrusions or dents). The methods for obtaining the structural attribute set, historical attribute set, and design attribute set of the target bag design can include: extracting structural parameters from bag design drawings, obtaining material performance data from material supplier specifications, compiling historical failure and usage data from enterprise production and after-sales records, and clarifying design attribute requirements through market research and customer demand analysis. This ensures that the three types of attribute set data are complete, accurate, and interrelated, providing comprehensive and standardized data support for subsequent modeling, simulation, and optimization design. It also ensures that the entire design process is based on accurate structural parameters, real historical failure data, and clear performance requirements, guaranteeing the reliability of the design results.

[0020] S200 is modeled and simulated based on structural attribute sets and historical attribute sets. Through explicit dynamic analysis, the action process of loads on the tie rod target component under different energy levels and directions is simulated. Information on high strain energy density regions and geometrically weak locations of the tie rod target component is extracted to obtain a characteristic response set.

[0021] It is understandable that numerical simulation methods can be used to recreate the stress response of the tie rod in actual use, thus identifying structural weaknesses. The structural attribute set provides the physical basis for the simulation, clarifying the tie rod's geometry, material properties, and assembly relationships. The historical attribute set provides the operating conditions for the simulation; by analyzing past failure cases, it clarifies the load conditions that make the tie rod prone to failure (such as the magnitude and direction of impact energy), ensuring that the simulation conditions closely match actual usage scenarios. Explicit dynamic analysis has the advantage of handling high-speed impacts and transient nonlinear problems, accurately simulating the stress wave propagation, plastic deformation, energy transfer, and accumulation processes inside the tie rod under load, avoiding the limitations of traditional static analysis in capturing transient impact responses. During the simulation, different energy levels (e.g., from 5J in a minor everyday collision to 50J in a severe drop) and multi-directional loads (e.g., axial tension / compression, radial bending, oblique impact, etc.) can be set to comprehensively cover the complex stress conditions that the tie rod may face. High strain energy density regions are where energy concentrates during impact in a tension rod. These regions are often where stress peaks occur, making them prone to plastic deformation or fracture. Geometrically weak points refer to areas with insufficient mechanical properties due to unreasonable structural design (such as excessively thin walls, sharp corner transitions, abrupt changes in cross-section, etc.). Through post-simulation data processing, information on high strain energy density regions and geometrically weak points can be extracted and integrated to form a characteristic response set. This set includes not only geometric information such as the spatial location and geometric contour of the weak points, but also dynamic mechanical response data such as the corresponding stress-time curves, cumulative strain energy data, and plastic deformation. Leveraging the advantages of explicit dynamic analysis, the action process of loads on the tension rod target component under different energy levels and directions can be accurately simulated. This overcomes the limitation of traditional static testing in reproducing dynamic impact responses, enabling precise identification of high strain energy density regions and geometrically weak points in the tension rod. The resulting characteristic response set, containing geometric information and mechanical response data of the weak points, provides clear targets for optimized design.

[0022] In one possible implementation, S200, based on modeling and simulation processing using structural attribute sets and historical attribute sets, simulates the action process of loads on the tie rod target component under different energy levels and directions through explicit dynamic analysis. It extracts information on high strain energy density regions and geometrically weak locations of the tie rod target component to obtain a characteristic response set, including: S210, perform failure mode clustering based on historical attribute sets to obtain a typical failure mode library.

[0023] It is understandable that historical attribute sets contain a large number of scattered tie rod failure case data. These failure cases may exhibit randomness and diversity due to differences in operating conditions, usage environments, and individual manufacturing deviations, making it difficult to extract universal failure patterns directly. The core purpose of failure mode clustering is to use data mining algorithms to classify these scattered failure cases according to the similarity of their failure characteristics, selecting representative typical failure modes to form a systematic failure mode library. This typical failure mode library eliminates the randomness of individual cases, focusing on the common failure patterns of tie rods under specific load conditions and structural designs, such as "bending and fracture at the root of the tie rod tube," "wear failure of the locking mechanism slot," and "deformation and jamming at the connection of multi-section tie rods." The establishment of the typical failure mode library provides a core basis for the subsequent construction of multi-condition load spectra—based on the load conditions corresponding to typical failure modes, simulation conditions can be designed specifically to ensure that the simulation can accurately reproduce key failure scenarios. It also points the way for subsequent structural optimization, enabling optimization work to focus on solving core failure problems and avoiding blind optimization.

[0024] Optionally, in step S210, failure mode clustering is performed based on the historical attribute set to obtain a typical failure mode library, including: S211. Based on the historical attribute set, an initial feature vector set describing the load, displacement, and morphological characteristics of each impact event is constructed, and the data is balanced using synthetic minority class oversampling technology to obtain a balanced feature set.

[0025] The initial feature vector set is constructed to transform unstructured data from the historical attribute set into structured data suitable for cluster analysis. Each impact event's feature vector contains three dimensions: load characteristics (e.g., impact energy, peak load, loading rate, and direction angle), displacement characteristics (e.g., maximum deformation of the tie rod after impact, displacement distribution in the deformation area, and recovery), and morphological characteristics (e.g., crack length / width at the failure site, geometric changes in the deformation area, and quantitative indicators of surface damage). Each dimension can be converted into specific numerical parameters; for example, the direction angle can be quantified as a value from 0 to 360°, and the surface damage level can be converted into a quantified value from 1 to 5, making the feature vector computable. Because failure cases in the historical attribute set may exhibit class imbalance—for example, common failure modes like "slight deformation" have a large number of samples, while extreme failure modes such as "severe fracture" and "complete failure of the locking mechanism" have a smaller number of samples—if directly used for cluster analysis, the algorithm may favor failure modes with a larger sample size, leading to the neglect of a minority of typical failure modes. Therefore, this method addresses the problem of neglecting a minority of typical failure modes. Synthetic Minority Oversampling Technique (SMOTE) interpolates new synthetic samples in the feature space of minority class samples, increasing the number of minority class samples without altering the majority class samples, thus balancing the sample numbers for each failure mode. The balanced feature set obtained after synthetic minority oversampling ensures that the clustering algorithm gives equal importance to each failure mode during training, thereby accurately identifying all typical failure modes, including extreme failure modes, and improving the completeness and representativeness of the subsequent failure mode library.

[0026] For example, the historical attribute set can be first filtered to remove impact event data with incomplete records or abnormal operating conditions (such as ambient temperatures exceeding the normal operating range of -20℃ to 60℃). Then, a 12-dimensional standardized initial feature vector set can be constructed according to three dimensions: load characteristics, displacement characteristics, and morphological characteristics. Among them, the load characteristics include impact energy (directly extracted test record value, unit J, such as 8.5J, 25.3J), peak load (extracted force sensor record value, unit N, such as 1200N, 2850N), loading rate (calculated rate of load increase from 0 to peak value, unit N / ms, such as 5.2N / ms, 18.7N / ms), and direction angle of action (with the tie rod axial direction as 0° and the radial direction as 90°). ° (quantized clockwise to 0-360°, such as 35°, 90°) 4-dimensional parameters; displacement characteristics include maximum deformation (maximum deformation measured by a 3D scanner after impact, in mm, such as 1.2mm, 5.8mm), maximum displacement of the deformation area (displacement value of the farthest point within the deformation area measured, in mm, such as 0.8mm, 4.5mm), elastic recovery (deformation recovery value after impact unloading, in mm, such as 0.9mm, 2.1mm), and residual plastic deformation (difference between maximum deformation and elastic recovery, in mm, such as 0.3mm, 3.7mm) 4-dimensional parameters; morphological characteristics include crack length L (measured crack length at the failure site, in mm, such as 0mm, 3.5m) 4-dimensional parameters. The four parameters are: crack width W (maximum crack width in mm, e.g., 0mm, 0.2mm, 0.8mm), surface damage level D (quantified into 1-5 levels, level 1 no damage, level 5 severe damage, e.g., level 1, level 3, 5), and deformation area S (projected area of ​​the deformation area measured by 3D scanning, in cm², e.g., 2.5cm², 18.3cm²). All parameters are converted into numerical values ​​to achieve calculability. To address the imbalance between the majority of samples (such as slight deformation) and the minority of samples (such as severe fracture and complete failure of the locking mechanism) in historical data, SMOTE technology is used for data balancing. First, the number of samples for each failure mode is counted, and a uniformity is set. The objective is to ensure that the difference in the number of samples among all classes does not exceed 20%. Then, the minority class samples are processed: linear interpolation is performed between the k nearest neighbors (k is 5, selected based on Euclidean distance) of each minority class sample. For example, for a minority class sample A (feature vector [8.5, 1200, 5.2, ..., 2.5]) and its nearest neighbor sample B (feature vector [10.2, 1500, 6.8, ..., 3.2]), the feature vector of the new sample C is the sum of the products of the parameters of A and the random number (0, 1), and the products of the parameters of B minus the parameters of A. That is, the values ​​corresponding to [8.5, 1200, ..., 2.5] are all superimposed with 0.A 3-fold difference in corresponding dimensions results in a balanced feature set with uniformly distributed samples across all failure modes. This ensures that subsequent clustering algorithms give equal weight to all failure modes and accurately identify typical failure modes, including extreme failure modes.

[0027] S212, based on the radial basis kernel function, calculates the sample similarity matrix of the equilibrium feature set, and applies the spectral clustering algorithm to divide it, thus obtaining a typical failure mode library.

[0028] It is understandable that the sample similarity matrix is ​​a tool for measuring the similarity between different failure case samples in a balanced feature set, and its construction quality directly affects the accuracy of the clustering results. Since it is necessary to describe the multi-dimensional and complex features of failure cases, the radial basis function (RBF) can be used. The RBF has the advantage of handling high-dimensional data and capturing nonlinear relationships between samples. The RBF calculates the kernel function value of two samples in a high-dimensional feature space. The larger the value, the more similar the failure features (such as load conditions, failure morphology, and displacement response) between the samples. For example, two samples both showing "bending fracture at the root of the tie rod tube" have similar features such as load direction (radial bending), failure location (root), and displacement characteristics (similar maximum deformation at the root), and the similarity value calculated by the RBF will be high; while the similarity value between a "bending fracture" sample and a "slot wear" sample will be extremely low. Based on the sample similarity matrix, spectral clustering algorithms can transform the clustering problem into a graph partitioning problem in graph theory. This effectively handles non-convex sample data and avoids the shortcomings of traditional clustering algorithms (such as K-Means) that are sensitive to the shape of the data distribution and prone to getting trapped in local optima. Specifically, spectral clustering first constructs a sample association graph based on the similarity matrix, then maps the samples to a low-dimensional space by calculating the eigenvalues ​​and eigenvectors of the graph's Laplacian matrix. Finally, it performs clustering in the low-dimensional space, grouping failed samples with similar features into one class. Each class corresponds to a typical failure mode, and all categories are integrated to form a typical failure mode library. Each mode in the typical failure mode library has a clear characteristic identifier (such as the corresponding load range, failure location, and failure mode).

[0029] For example, the 12-dimensional numerical parameters of the balanced feature set can be standardized first (using Z-score standardization to transform each feature dimension into a distribution with a mean of 0 and a variance of 1, eliminating the influence of dimensional differences on similarity calculation); then, the sample similarity matrix can be calculated based on the radial basis function kernel function, with the kernel function parameter σ set to 2.5 (determined through cross-validation to ensure that the similarity of similar samples is concentrated in the range of 0.7-1.0, and the similarity of dissimilar samples is below 0.3). The kernel function calculation formula is K(xi,xj)=exp(-||xi-xj||² / (2σ²)), where xi and xj are two... The feature vectors of the samples, ||xi-xj||, represent the Euclidean distance. The similarity values ​​between all pairs of samples are calculated using the kernel function formula, constructing an n×n dimensional similarity matrix (n is the total number of samples in the balanced feature set, and the matrix elements range from 0 to 1, with values ​​closer to 1 indicating higher sample similarity). Then, a spectral clustering algorithm is applied for partitioning: First, an undirected weighted graph is constructed based on the similarity matrix, with each sample as a node and the similarity value as the weight of the edges between nodes. Second, the normalized Laplacian matrix of the graph is calculated as L = ID⁻¹ / 2WD⁻¹ / 2 (where I is the identity matrix, D is the diagonal matrix, and Dii is the node i). The degree of i is the sum of the weights of all edges connected to i, and W is the similarity matrix. The third step is to solve for the eigenvalues ​​and eigenvectors of the Laplacian matrix, and select the eigenvectors corresponding to the first k smallest non-zero eigenvalues ​​(the value of k is determined by the silhouette coefficient method, iterating through the values ​​of k=2-8, and selecting the largest silhouette coefficient k=4 as the number of clusters). The fourth step is to form an n×k dimensional feature matrix from the selected k eigenvectors, normalize each row of the matrix, and then use the K-Means algorithm for clustering. Finally, four typical failure modes are obtained and integrated to form a typical failure mode library. The feature identifiers of each mode are as follows: Mode 1 (pipe root) Mode 1 (bending fracture) corresponds to a load direction angle of 80°-100°, crack length ≥8mm, and residual plastic deformation ≥5mm; Mode 2 (locking mechanism slot wear) corresponds to a peak load of 1500N-2500N, surface damage level 3-4, and deformation area of ​​5-12cm²; Mode 3 (bar joint deformation jamming) corresponds to a loading rate of 10-20N / ms, elastic recovery ≤1.5mm, and maximum displacement of deformation area ≥6mm; Mode 4 (impact indentation in the middle of the tube) corresponds to an impact energy of 15-30J, an action direction angle of 150°-210°, crack length of 0mm, and surface damage level 2-3.

[0030] S220 constructs a multi-condition load spectrum based on a typical failure mode library, and randomly generates combinations of impact energy, load application point position and impact angle that conform to the probability distribution of actual use scenarios of bags, forming a load condition sample set.

[0031] It is understandable that the typical failure mode library clarifies the core load characteristics that make the pull rod prone to failure, while the construction of the multi-condition load spectrum is to expand these core characteristics into a systematic load set covering various possible working conditions in actual use, ensuring that subsequent simulations can comprehensively and realistically simulate the stress environment of the pull rod. In actual use, the impact energy, load application point, and impact angle of the luggage pull rod are not fixed values, but exhibit certain probability distribution patterns. For example, in daily use, the probability of minor impacts (low energy) is much higher than that of severe drops (high energy), the load application point is mostly concentrated in the middle and root of the pull rod, and the impact angle is mainly radial and oblique. Therefore, the construction of the load spectrum needs to be based on probabilistic statistical methods, rather than simply listing fixed working conditions. Therefore, by randomly generating load parameter combinations that conform to this probability distribution, various complex working conditions that the pull rod may encounter in long-term use can be simulated, including common working conditions and low-probability but high-risk extreme working conditions. The resulting load case sample set is comprehensive (covering different energy levels, points of application, and angles) and representative (the sample distribution is consistent with the actual usage probability), enabling subsequent simulation analysis to fully capture the mechanical response of the tie rod under various working conditions and avoid missing key weak areas due to incomplete working condition settings.

[0032] In one possible implementation, S220, based on a typical failure mode library, performs multi-condition load spectrum construction processing, randomly generating combinations of impact energy, load application point location, and impact angle that conform to the probability distribution of actual bag usage scenarios, forming a load condition sample set, including: S221. Based on the load conditions corresponding to each failure mode in the typical failure mode library, the marginal probability density functions of impact energy, load application point location and impact angle are fitted respectively using the probability density estimation method.

[0033] It is understandable that each failure mode in the typical failure mode library corresponds to a specific load condition range. For example, "bending fracture at the root of the tie rod" often occurs under conditions of impact energy of 15-30J, a point of impact 5-10cm from the root, and an impact angle of 30-60° (radial deviation). The core function of probability density estimation methods (such as kernel density estimation and parametric distribution fitting) is to construct a mathematical function that describes the probability distribution of individual load parameters (impact energy, load application point, impact angle) based on these historical load data. Taking impact energy as an example, by statistically analyzing the impact energy data of all failure cases in the typical failure mode library, the kernel density estimation method can be used to fit the marginal probability density function of the impact energy. The marginal probability density function can clearly present the probability density of different energy values—the higher the function value corresponding to the energy value, the greater the probability of failure caused by that energy level in actual use. Similarly, the edge probability density function at the load application point can represent the probability distribution of load application at different locations on the tie rod (e.g., the probability density at the root and middle is higher than at the end), and the edge probability density function at the impact angle can represent the probability distribution of impact in different directions (e.g., the probability density of radial impact is higher than that of axial impact). The construction of the edge probability density function provides the foundation for the subsequent establishment of the joint probability distribution model, ensuring that the distribution of each load parameter conforms to the actual failure case patterns.

[0034] For example, we can first extract all historical load raw data corresponding to the four types of failure modes from the typical failure mode library, classify and organize them according to three parameters: impact energy, load application point location, and impact angle, and remove outliers (using the 3σ criterion, i.e., removing data that exceeds the parameter mean ± 3 times the standard deviation, such as the extreme value of 55J in impact energy that deviates from the overall distribution); for the impact energy parameter, we use the kernel density estimation method to fit the marginal probability density function, select a Gaussian kernel for the kernel function, and determine the bandwidth to be 1.8 through cross-validation. Based on all valid impact energy data (such as mode... The energy data collection of various modes (15-30J, Mode 1, Mode 4, etc.) was used to calculate a continuous probability density curve using the kernel density estimation formula f(x)=1 / (nh)Σexp(-(x-xi)² / (2h²)) (where n is the data volume, h is the bandwidth, and xi is the energy of a single impact). The peak of the curve corresponds to 22J, indicating that this energy level has the highest probability density of failure. The probability density values ​​in the 18-25J range are all higher than 0.08, which is the high-incidence energy range for failure. Regarding the load application point location parameters, the following was first... The tie rod was divided into three regions: root (0-5cm), middle (5-15cm), and end (15-20cm). The load frequency in each region was statistically analyzed, and a parametric distribution (Beta distribution) was fitted. The distribution parameters α=2.3 and β=1.1 were solved using the maximum likelihood estimation method. The fitted marginal probability density function showed a probability density of 0.42 in the root region, 0.48 in the middle region, and 0.10 in the end region, verifying that the middle and root regions are high-incidence areas for load application. Regarding the impact angle parameters, axial (0°±30°) and radial (90°) parameters were also analyzed. Classified by ±30° and oblique (other angles), kernel density estimation (Gaussian kernel, bandwidth 2.5) was used to fit the data based on historical angle data. The resulting probability density function showed peaks at 85° and 275°. The cumulative probability density of radial and near-radial angles (60°-120°, 240°-300°) reached 0.65, clearly indicating that radial impact was the dominant failure direction. Finally, independent marginal probability density functions of the three parameters were obtained, and each function passed the KS test (significance level 0.05) to ensure that the fitting results were not significantly different from the original data distribution.

[0035] S222, based on the edge probability density function and the predefined correlation constraints between impact energy and the location of the load application point and the impact angle, a joint probability distribution model is obtained.

[0036] It is understandable that the marginal probability density function only describes the independent distribution of a single load parameter. However, in actual use, impact energy, load application point location, and impact angle are not completely independent but rather correlated. For example, high-energy impacts (e.g., above 30J) are more likely to occur in the middle of the tie rod (rather than at the end, as the end is protected by the housing) and are mostly oblique impacts (rather than axial impacts, as axial impacts are easily buffered by the tie rod extension mechanism). Low-energy impacts (e.g., below 5J) may act randomly at various locations on the tie rod, with a more uniform directional distribution. The predefined correlation constraints are set based on actual usage logic and historical data patterns to quantify the correlation between these parameters. For example, the correlation coefficient matrix clarifies the negative correlation between impact energy and application point location (higher energy, more concentrated application point in the middle) and the positive correlation between impact energy and impact angle (higher energy, higher proportion of oblique impacts). The purpose of the joint probability distribution model is to combine the marginal distributions of the three load parameters with their correlation constraints to construct a mathematical model that can describe the joint probability of multiple parameter values. The joint probability distribution model overcomes the limitations of independent distribution of individual parameters, accurately reflects the probability of occurrence of different combinations of working conditions in actual working conditions, ensures that the subsequently generated load condition samples conform to the parameter combination rules in the actual scenario, and improves the realism of the simulation.

[0037] S223 generates initial sampling points based on the joint probability distribution model, and generates a load condition sample set by iteratively adjusting the sampling density.

[0038] It is understandable that the initial sampling points are generated based on a joint probability distribution model, using a series of load parameter combinations (impact energy, point of impact, impact angle) obtained through random sampling algorithms (such as Monte Carlo sampling and Latin hypercube sampling). Each sampling point corresponds to a specific simulation condition. The initial sampling points may have uneven distribution; for example, some high-probability combination conditions may have overly dense sampling points, while some low-probability but critical extreme conditions (such as high energy + root impact + vertical impact) may have insufficient sampling points, leading to insufficient coverage of critical conditions in subsequent simulations. The purpose of iteratively adjusting the sampling density is to optimize the sampling point distribution: by calculating the joint probability density value corresponding to each sampling point, the sampling density is appropriately reduced in high-probability condition areas (to avoid redundancy), and the sampling density is increased in low-probability but critical extreme condition areas (to ensure coverage). Simultaneously, the overall number of sampling points is ensured to meet the balance between simulation efficiency and accuracy (e.g., the total number of sampling points is controlled at 100-200 sets, ensuring comprehensiveness while avoiding excessive computation). After multiple rounds of iterative adjustments, the final load condition sample set can accurately reflect the actual probability distribution while also covering both common and extreme load conditions, ensuring that each sample is representative and providing scientific and comprehensive load condition inputs for subsequent explicit dynamic simulations.

[0039] For example, based on the constructed joint probability distribution model (including the edge distribution of impact energy, load application point location, impact angle, and parameter correlation constraints), the Latin hypercube sampling algorithm can be used to generate 200 initial sampling points. The Latin hypercube sampling algorithm can ensure that the sampling values ​​of each parameter dimension uniformly cover its value range, reducing the initial distribution deviation. Then, the joint probability density value of each sampling point is calculated, and a probability density threshold of 0.03 is set (determined by statistically analyzing the density distribution of the initial sampling points). The sampling points are divided into high-probability regions (joint probability density ≥ 0.03) and low-probability regions (joint probability density < 0.03). For high-probability regions (such as combinations of impact energy 18-25J, central impact point, and radial impact), 30% of redundant sampling points are removed (70 groups are retained) to avoid wasting simulation resources. For low-probability but critical extreme condition regions, sampling is supplemented according to risk level: Level 1 extreme condition (high energy 35-45J + root impact). Twenty sets of samples were added for point + vertical impact, and 15 sets were added for each of the secondary extreme conditions (low energy 5-10J + end point + axial impact, high energy 35-45J + middle point + oblique impact) to ensure full coverage of extreme failure scenarios. At the same time, the total number of sampling points was set to a maximum of 180 sets, and was adjusted through two rounds of iteration: the first round removed redundant points in high-probability areas and added extreme condition points to 160 sets; the second round added 20 sets for density gaps that still existed after the addition (such as medium energy 25-35J + root point + oblique impact), finally forming a set of 180 load condition samples. Each sample contains specific parameter combinations (such as impact energy 22J, point of impact position 8cm, impact angle 95°, impact energy 40J, point of impact position 3cm, impact angle 0°, etc.), and passed the distribution uniformity test (quantile deviation of sampling values ​​of each parameter dimension ≤5%) to ensure that the samples not only fit the actual probability distribution, but also fully cover common and extreme conditions.

[0040] S230, based on the structural property set and load condition sample set, simulates the action process of loads on the tie rod target component under different energy levels and action directions through explicit dynamic analysis, extracts the high strain energy density region and geometric weak location information of the tie rod target component, and obtains the characteristic response set.

[0041] It is understandable that the set of structural properties (the "physical model" of the tie rod) can be combined with the set of load case samples ("stress conditions") to complete the simulation process from load condition input to response output through explicit dynamic analysis. The geometric parameters, material properties, and assembly relationships in the set of structural properties provide the foundation for building a high-precision simulation model—for example, setting the geometric boundaries and material parameters of the model based on the tube wall thickness and the material's elastic modulus, and setting the contact conditions based on the assembly gaps. Each sample in the load case sample set corresponds to a specific set of impact load parameters. By transforming these parameters into load boundary conditions in the simulation (such as applying a transient impact force of specified energy at a specific point of application and setting the impact direction vector), the actual stress process of the tie rod under that condition can be simulated. Since traditional analysis methods cannot handle high-speed impact and large deformation problems, this method uses explicit dynamic analysis to solve transient dynamic equations to accurately capture the transient responses such as stress wave propagation, plastic deformation initiation and development, strain energy accumulation, and contact collisions inside the tie rod under impact loads. After the simulation is completed, data mining is performed on the simulation results to extract information on high strain energy density regions (regions with concentrated energy and prone to failure) and geometrically weak locations (regions with insufficient mechanical properties due to unreasonable structural design). This includes the three-dimensional spatial coordinates, geometric contours, peak stress, accumulated strain energy, and plastic deformation of these regions. All information is integrated to form a characteristic response set, which serves as the direct basis for subsequent structural optimization design, clarifying where optimization is needed and the mechanical problems that need to be solved.

[0042] Optionally, S230, based on the structural property set and load case sample set, the action process of loads on the tie rod target component under different energy levels and directions is simulated through explicit dynamic analysis. Information on high strain energy density regions and geometrically weak locations of the tie rod target component is extracted to obtain a characteristic response set, including: S231 uses transient nonlinear finite element analysis based on structural property set and load condition sample set. By calculating stress wave propagation, plastic hinge formation process and damage accumulation of the target component under different representative impact conditions, transient response field data is obtained.

[0043] Transient nonlinear finite element analysis is a core technique in explicit dynamic analysis, suitable for handling complex mechanical behaviors during tie rod impacts, including material nonlinearity (such as stress-strain changes after the material enters the plastic stage), geometric nonlinearity (such as structural shape changes caused by large deformation of the tie rod), and contact nonlinearity (such as friction and collision contact between multiple tie rod sections). The structural property set provides the modeling foundation for finite element analysis. Geometric entities of the finite element model are constructed based on geometric parameters, constitutive relations of elements are defined based on material properties (such as elastic-plastic constitutive models and damage constitutive models), and contact properties between elements (such as friction coefficient and contact stiffness) are set based on assembly relationships. Representative impact conditions in the load case sample set (such as selected high-probability conditions and extreme conditions) are transformed into transient load boundary conditions of the finite element model (such as time-varying impact force and displacement constraints). During the analysis, the tie rod can be solved using discretized finite element methods (FEA) to calculate the stress, strain, displacement, and energy of each element at different time steps. This allows for the reconstruction of the stress wave propagation path within the tie rod (e.g., diffusion from the impact point to both ends), the formation process of plastic hinges (e.g., the formation of plastic hinges at the root of the tube due to bending stress concentration, leading to excessive local deformation), and the damage accumulation law (e.g., the damage evolution of the material under repeated impacts or a single high-energy impact, from microcrack initiation to macroscopic failure). Transient response field data is global mechanical response data containing the time dimension, such as three-dimensional distribution data of stress field, strain field, displacement field, and energy field changing over time. It comprehensively and meticulously reflects the mechanical behavior of the tie rod during the impact process, providing rich raw data for subsequent identification of key areas.

[0044] For example, S231, transient nonlinear finite element analysis is performed based on the structural property set and load case sample set. By calculating the stress wave propagation, plastic hinge formation process, and damage accumulation of the tie rod target component under different representative impact conditions, transient response field data is obtained, including: S2311 performs refined geometric repair and idealization on the target component of the tie rod based on the structural attribute set, and performs function-oriented region classification on the processed model. By distinguishing the mechanism regions, a computational geometric model that can be used for simulation analysis is obtained.

[0045] It is understandable that the geometric data in the structural attribute set can originate from design drawings. The original geometric model may contain some issues unsuitable for direct use in finite element analysis, such as minor geometric defects (e.g., sharp corners, burrs, redundant chamfers), non-manifold geometry (e.g., overlapping surfaces, unclosed edges), and redundant features (e.g., small bosses used for assembly markings). These issues can lead to mesh generation failure or distorted simulation results. The purpose of refined geometric repair is to correct these defects, such as removing burrs, smoothing sharp corners, closing discontinuous geometric boundaries, and deleting redundant features, thus ensuring the geometric model has integrity and smoothness. Idealization, on the other hand, simplifies complex structures without affecting the accuracy of mechanical performance simulation. For example, it simplifies the gear tooth profile of a complex locking mechanism to an equivalent cylindrical surface (if its local details have little impact on the overall impact response), and simplifies small threaded connections to rigid connections, thereby reducing the difficulty of mesh generation and computational load. Function-oriented region classification divides the geometric model into different regions based on the different functions of various parts of the tie rod, such as the tube bearing region, the locking mechanism functional region, the transition region of the connection parts, and the mechanism motion region (e.g., the sliding region between expansion joints). The purpose of distinguishing the mechanism regions is to clarify the mechanical property requirements of different regions. For example, contact friction needs to be considered in the moving region, while stress distribution needs to be the focus in the load-bearing region. By classifying these regions, targeted strategies can be adopted in subsequent mesh generation and boundary condition settings (such as refining the mesh in the load-bearing region and setting special contact properties in the moving region). The resulting computational geometry model retains the key geometric features that influence the mechanical response while removing redundant information and correcting geometric defects, thus meeting the dual requirements of finite element analysis for model accuracy and computational efficiency.

[0046] For example, the original 3D geometric model (such as a STEP format file) of the target component of the tie rod can be extracted from the structural attribute set and imported into Geomagic DesignX software for fine geometric repair. The software's built-in geometric diagnostic tools are used to identify model defects, deleting redundant features such as assembly mark bosses with a diameter ≤0.5mm and burrs with a height ≤0.3mm. Sharp corners with a radius ≤0.2mm are rounded (with a radius of 0.5mm). Discontinuous geometric boundaries are closed using the surface stitching function (gaps ≤0.1mm are directly stitched, while gaps greater than 0.1mm are repaired using the surface patching function), eliminating non-manifold geometric problems such as surface overlap and edge misalignment. Subsequently, idealization is performed to simplify the gear tooth profile of the locking mechanism into an equivalent cylindrical surface (the cylinder diameter is consistent with the gear pitch circle diameter, and the length retains the original gear axial dimension). Small threaded connections below M3 are simplified into rigid fixed surfaces (retaining the central axis and connection surface position of the threaded connection), ignoring surface scratches with a depth ≤0.2mm. The model is divided into four categories using the software's region division tool: the pipe body load-bearing area (main pipe section of the tie rod, bearing the main impact load), the locking mechanism functional area (including simplified gear and slot structures, responsible for telescopic locking), the connection transition area (the joints of each tie rod section and the connection end with the box body, where stress is prone to concentration), and the mechanism motion area (the mating surfaces between telescopic joints and the active contact surfaces of the locking mechanism, where relative motion exists). Finally, each area is labeled with attributes (e.g., the load-bearing area is labeled as "high stress concern area" and the motion area is labeled as "contact friction area"), and exported as an ANSYS-recognizable IGES format file, resulting in a computational geometric model that can be directly used for finite element simulation analysis. The geometric accuracy error of this model is ≤0.1mm, meeting the simulation analysis requirements for model integrity, smoothness, and functional specificity.

[0047] S2312, based on the computational geometry model, an initial mesh sensitivity analysis is performed to obtain the recommended mesh density for key size regions. Then, based on the recommended mesh density, a partitioned and refined hexahedral dominant mesh is applied to different functional regions in the computational geometry model to obtain a high-quality discrete model.

[0048] Meshing is an essential step in finite element analysis, and mesh quality (such as element shape and density) directly impacts the accuracy and computational efficiency of simulation results. A mesh that is too sparse will lead to distorted stress calculations and an inability to capture local stress concentrations; a mesh that is too dense will significantly increase computational load and prolong simulation time. The purpose of initial mesh sensitivity analysis is to test simulation results (such as stress peak values ​​at critical locations) under different mesh densities to determine the sensitivity of simulation results to changes in mesh density: if a 10% increase in mesh density results in a stress peak change of less than 5%, then the mesh density meets the accuracy requirements; if the change is greater than 5%, further mesh refinement is needed. Through this analysis, the recommended mesh density is determined for critical dimensional regions (such as the root of the pipe, the locking mechanism slot, and stress concentration areas at connections). These regions are sensitive to mesh density and require a denser mesh (e.g., element size of 1 mm); non-critical regions (such as the middle of the pipe away from connection points) are not sensitive to mesh density and can use a sparser mesh (e.g., element size of 3 mm). The partitioned, high-density hexahedral-dominated meshing strategy prioritizes hexahedral elements (which offer higher computational accuracy and stability compared to tetrahedral elements) and divides the mesh according to the mesh density requirements of different functional regions: load-bearing and stress concentration regions are meshed with a recommended dense mesh, motion and transition regions with a medium-density mesh, and non-critical load-bearing regions with a sparse mesh. Simultaneously, a gradual transition is used in the transition regions between different mesh densities to avoid computational errors caused by abrupt changes in mesh size. The resulting high-quality discrete model can maximize the control of computational load while ensuring simulation accuracy (especially in critical regions), achieving a balance between accuracy and efficiency.

[0049] For example, please refer to Figure 2The computational geometry model can be imported into ANSYS Meshing software. First, select three key dimension regions: the pipe root, the locking mechanism slot, and the transition area of ​​the connection part, and conduct initial mesh sensitivity analysis. Set four mesh size schemes (0.8mm, 1.0mm, 1.2mm, and 1.5mm) and simulate the same typical impact condition (impact energy 22J, central impact point, radial impact). Extract the peak stress in the key regions under each scheme: the peak stress is 520MPa for the 0.8mm scheme, 515MPa for the 1.0mm scheme, 498MPa for the 1.2mm scheme, and 472MPa for the 1.5mm scheme. Calculate the stress change rate between adjacent schemes. The difference between the 1.0mm and 0.8mm schemes is only 0.97% (<5%), the difference between the 1.2mm and 1.0mm schemes is 3.3% (<5%), and the difference between the 1.5mm and 1.2mm schemes is 5.2% (>5%). Determine that the recommended mesh density for the key dimension regions is 1.0mm (element size 1.0mm). For non-critical load-bearing areas (the middle part of the pipe body far from the connection point), a mesh size of 3.0 mm is recommended, while for moving and transitional areas, a mesh size of 2.0 mm is recommended. A hexahedral-dominant meshing strategy is adopted. For critical areas such as the load-bearing area of ​​the pipe body and the locking mechanism slot, swept meshing technology is used to generate all hexahedral elements. For transitional areas at the connection points, mapped meshing technology is used to ensure element quality. For the moving areas of the mechanism, hybrid meshing technology (main hexahedral + local tetrahedral transition) is used. A size gradient factor of 1.2 is set in the transition areas of different mesh densities (the element size change between adjacent areas does not exceed 20%) to avoid abrupt mesh changes. After meshing, the model is verified by a software quality check tool, requiring element distortion rate ≤0.3, aspect ratio ≤5, and warpage ≤15°. The final discrete model contains approximately 120,000 hexahedral elements and 30,000 tetrahedral elements (for transition). The element quality pass rate in critical areas is ≥95%, the overall computational load is controllable, and the simulation accuracy of stress concentration areas is ensured.

[0050] S2313, based on the load condition sample set and high-quality discrete model, performs explicit dynamic solution calculations to obtain a database of time series simulation results of the tie rod target component under different representative impact conditions.

[0051] It is understandable that explicit dynamics solution calculation is the process of solving transient dynamic control equations using numerical algorithms based on a high-quality discrete model and a set of load case samples. Before solving, each load case parameter (impact energy, point of application, impact angle) in the load case samples needs to be transformed into boundary conditions and load application methods for the discrete model. For example, applying a time-varying impact load (calculated based on the impact energy) at a specified node (point of application), setting fixed constraints (such as displacement constraints at the connection end between the tie rod and the box), and defining contact parameters (such as the friction coefficient and collision recovery coefficient between expansion joints). The explicit dynamics algorithm uses the central difference method for time integration, which can efficiently handle transient responses under high-speed impacts. By iteratively calculating the stress, strain, displacement, velocity, energy, and other physical quantities of each finite element in each time step (such as 1e-6 seconds), it completely reconstructs the entire impact process from load application, stress wave propagation, plastic deformation to the end of the impact. Since the load case sample set contains multiple representative load cases, each load case needs to be solved and calculated separately. The simulation results for each load case are saved in time series (e.g., from 0 to 5 ms, with a set of data stored every 0.1 ms) to form a time series simulation result database. The time series simulation result database contains transient mechanical response data of the entire tie rod under each load case, such as the stress peak at the root of the tube at 2 ms, the amount of plastic deformation at 3 ms, and the cumulative strain energy curve during the entire impact process under a certain load case, providing comprehensive raw data support for subsequent field data extraction and fusion.

[0052] S2314 extracts and fuses field data from the results of each representative impact condition in the time series simulation results database to obtain transient response field data.

[0053] It is understandable that the data in the time series simulation results database is discrete data stored according to working conditions and time steps. Each data point only reflects the local mechanical state of a specific unit at a specific time, making it difficult to intuitively present the overall transient response law of the tie rod. The core of field data extraction and fusion is to systematically process these discrete data and transform them into continuous, global transient response field data. The extraction process includes: for each working condition, extracting the global distribution data of key mechanical fields (such as stress field, strain field, displacement field, strain energy density field) from the time series data at different characteristic time points (such as the peak load time, the maximum deformation time, and the impact end time); the fusion process involves associating and integrating the field data at different time points under the same working condition to form dynamic field data that evolves over time (such as the dynamic process of the stress field spreading from the impact point to the surrounding area). At the same time, the field data under different working conditions are classified and organized, and the working condition parameters (such as energy level and direction of action) corresponding to each field data are labeled to obtain transient response field data, which intuitively and comprehensively presents the transient mechanical behavior of the tie rod under different working conditions. For example, under high-energy oblique impact, the stress wave propagates from the middle of the tube to the root, and a stress peak appears at the root, forming a plastic hinge; under low-energy axial impact, the stress distribution is uniform, and only elastic deformation occurs.

[0054] S232 performs key region identification processing based on transient response field data, identifying regions where the peak equivalent stress continuously exceeds the material's dynamic yield strength and regions with high plastic strain gradients throughout the entire impact process, generating a region identification map.

[0055] It is understandable that critical area identification is based on transient response field data to screen out high-risk areas where the tie rod is prone to failure during impact. High-risk areas include two types: the first type is areas where the peak equivalent stress continuously exceeds the material's dynamic yield strength. The dynamic yield strength is the critical stress value for plastic deformation. If the peak equivalent stress in a certain area continuously exceeds this value during impact, it indicates that the area has entered the plastic deformation stage. Long-term exposure to such loads can easily lead to fatigue fracture or cumulative plastic deformation failure. The second type is areas with high plastic strain gradients. The plastic strain gradient reflects the rate of change of plastic deformation in space. High gradient areas mean severe local deformation and uneven deformation distribution, which easily leads to stress concentration, thereby triggering crack initiation and propagation (e.g., high plastic strain gradients may appear at tie rod corner transitions or abrupt changes in wall thickness). During the identification process, threshold judgments and gradient calculations can be performed on the stress field and strain field in the transient response field data. For example, the stress threshold can be set to the material's dynamic yield strength, and the set of elements continuously exceeding this threshold can be screened; the spatial gradient of plastic strain can be calculated, and the set of elements with gradient values ​​exceeding a preset threshold can be screened. The spatial locations and geometric extents of these two types of regions are marked on the 3D model of the tie rod, generating a region identification map. This map visually displays the high-risk, vulnerable areas of the tie rod, providing a clear spatial basis for subsequent geometric subdomain segmentation and feature parameter extraction.

[0056] S233, based on the preset material failure criteria and region identification map, the target component of the tie rod is segmented into geometric subdomains and its feature parameters are extracted to obtain a feature response set; wherein, the feature response set contains the geometric contour, spatial location information and corresponding dynamic mechanical response spectrum data of at least one high strain energy density region.

[0057] Understandably, the preset material failure criteria are the basis for determining whether the tie rod material has failed. These criteria need to be selected based on the tie rod material type and failure mode. For example, for metal tubes, the fracture toughness criterion is used (failure is determined when the stress intensity factor at the crack tip exceeds the material's fracture toughness); for plastic connectors, the plastic strain failure criterion is used (failure is determined when the plastic strain exceeds the material's ultimate plastic strain). Combined with the region identification map, the material failure criteria can further verify whether the identified high-risk areas meet the failure conditions, ensuring that these areas are truly weak points that need optimization. Geometric subdomain segmentation involves separating the high-risk areas (such as high strain energy density areas) marked in the region identification map from the overall tie rod model, forming independent geometric subdomains. Each subdomain corresponds to a weak area. The segmentation process must preserve the complete geometric contour (such as boundary curves and surface shapes) and spatial location information (such as coordinates relative to the overall tie rod and relative relationships with other components) of the subdomain. Feature parameter extraction involves extracting key parameters reflecting the mechanical properties of each geometric subdomain from the transient response field data. These parameters include dynamic mechanical response spectrum data (such as stress-time curves, strain energy accumulation-time curves, and plastic deformation-time curves) and static feature parameters (such as the subdomain's maximum wall thickness, minimum wall thickness, average peak stress, and maximum strain energy density). The geometric information (contour, location) and corresponding mechanical parameters of each geometric subdomain are integrated to form a feature response set. This feature response set clearly identifies the specific object of optimization (which location and shape of the weak area) and provides the mechanical problems that need to be addressed (such as excessively high peak stress or excessively rapid strain energy accumulation in the region), providing accurate and comprehensive input data for subsequent lattice structure optimization design.

[0058] For example, please refer to Figure 3Pre-defined material failure criteria can be established. For metal pipes, the fracture toughness criterion is used (material fracture toughness KIC = 50 MPa・m^(1 / 2), failure is determined when the stress intensity factor K at the crack tip ≥ KIC). For plastic connectors, the plastic strain failure criterion is used (material ultimate plastic strain εlim = 0.05, failure is determined when the plastic strain ε ≥ εlim). Region identification maps and transient response field data are imported and processed using ANSYS. Mechanical software's region filtering function, combined with failure criteria, screened high-risk areas: the stress intensity factor K of the metal tube root region was 58 MPa・m^(1 / 2) (≥KIC), and the plastic strain ε of the locking mechanism's plastic slot was 0.07 (≥εlim). These two areas were identified as high strain energy density areas requiring optimization. Geometric subdomain segmentation was performed using the software's subdomain extraction tool, based on the boundaries of the region identification map. The high-risk area at the tube root (a cylindrical segment with dimensions of φ20mm×30mm) and the locking mechanism slot area (a cuboid area with dimensions of 15mm×8mm×5mm) were separated from the overall model, preserving the complete geometric contours (cylindrical surface + end face at the tube root, cuboid side surface + bottom face at the slot). Spatial position information was recorded (the center coordinates of the tube root subdomain are (120,0,30) mm, and the center coordinates of the slot subdomain are (85,5,15) mm, both based on the bottom of the tie rod). A coordinate system is established with the endpoint as the origin. For each subdomain, dynamic mechanical response spectrum data (peak value of stress-time curve at the root of the pipe: 580 MPa; final value of cumulative strain energy-time curve: 42 J; peak value of plastic deformation-time curve at the slot: 0.8 mm) and static characteristic parameters (maximum wall thickness of 3 mm, minimum wall thickness of 2.5 mm, average peak stress of 520 MPa, maximum strain energy density of 1.8 J / mm³ at the root of the pipe; maximum wall thickness of 2 mm, minimum wall thickness of 1.5 mm, average peak stress of 350 MPa, maximum strain energy density of 1.2 J / mm³) of the slot are classified and integrated with the geometric information (profile shape, spatial coordinates) and mechanical parameters (dynamic response spectrum, static characteristic values) of the two subdomains to form a characteristic response set, and to identify the optimization target point for each weak area (the peak stress at the root of the pipe needs to be reduced to below 500 MPa, and the plastic strain at the slot needs to be reduced to below 0.05).

[0059] S300, based on the characteristic response set and the design attribute set, perform variable density lattice structure generation processing on the high strain energy density region to obtain an optimized design scheme set; the optimized design scheme set includes at least one non-uniform lattice filling topology configuration scheme for the interior or surface of the target component.

[0060] It is understandable that targeted structural optimization can be performed based on the previously identified weak areas (characteristic response set) and clear design requirements (design attribute set). High strain energy density areas are weak points in the tie rod. Traditional optimization methods (such as increasing wall thickness or changing materials) may lead to increased weight and cost, and the optimization effect is limited. Therefore, this application proposes optimization through a variable density lattice structure. The variable density lattice structure is a novel lightweight porous structure. Its advantage lies in its ability to achieve "on-demand allocation of mechanical properties" by adjusting the density distribution of lattice units. High-density lattices are used in areas of high stress concentration and high energy demand (ensuring strength and energy absorption capacity), while low-density lattices are used in areas of lower stress (achieving lightweighting), balancing strength, stiffness, and lightweight requirements. The design attribute set provides constraints and performance targets for the generation of the lattice structure, such as lightweight requirements (the overall mass after lattice structure filling should not exceed 1.2 times that of the original structure), mechanical performance requirements (the stress peak in high strain energy density areas should be reduced by more than 30% after optimization), and process constraints (the minimum size of the lattice unit should not be less than 0.5 mm to ensure machinability). The purpose of variable density lattice structure generation is to generate a non-uniform lattice topology in the geometric space of high strain energy density regions based on dynamic mechanical response data (such as stress distribution and strain energy density distribution) from the characteristic response set. Lattice elements are denser at locations with high stress and high strain energy density (high density), and sparser at locations with low stress and low strain energy density (low density). By generating multiple topology schemes with different parameter combinations (such as lattice element type, density gradient, and filling range), an optimized design scheme set is formed. Each scheme must meet the requirements of the design attribute set, constructing an intelligent design closed loop. This replaces the traditional experience-based design and repeated physical testing model, significantly shortening the R&D cycle, reducing trial-and-error costs and material consumption, and significantly improving the dynamic impact strength, structural stability, and service life of luggage handles. This provides reliable technical support for the high-performance, lightweight, and low-cost design of luggage handles.

[0061] In one possible implementation, the characteristic response set includes the geometric contour, spatial location information, and corresponding dynamic mechanical response spectrum data of at least one high strain energy density region; S300, the high strain energy density region is subjected to variable density lattice structure generation processing based on the characteristic response set and the design attribute set to obtain an optimized design scheme set, including: S310, based on the high strain energy density region and design properties, determines the three-dimensional voxelized design space of the lattice structure and the target relative density range, and performs parametric voxel discretization on the three-dimensional voxelized design space to obtain the three-dimensional voxelized design space.

[0062] It is understandable that the geometric contour and spatial location of the high strain energy density region are the direct basis for determining the design space of the lattice structure. The three-dimensional voxelized design space refers to the minimum three-dimensional spatial range that includes this high strain energy density region. It must completely cover the weak area, but avoid excessive expansion of the range, which would lead to unnecessary material waste and increased computation (e.g., the boundary of the design space extends 1-2mm beyond the contour of the weak area to ensure that the lattice structure can completely wrap around and reinforce the weak area). The performance requirements and process constraints of the design attribute set determine the target relative density range: relative density refers to the ratio of the actual volume of the lattice structure to the volume of the design space. The target relative density range needs to balance mechanical performance and lightweight requirements. For example, according to design requirements, if it is necessary to improve strength and control weight, the target relative density range can be set to 0.3-0.6 (the higher the relative density, the stronger the strength, but the greater the weight); if the process limits the minimum lattice unit size to 0.5mm, then it is necessary to ensure that the lattice unit size within this relative density range meets the processing requirements. Parametric voxel discretization divides a continuous 3D design space into a series of regular, uniformly sized cubic voxels (e.g., voxel size of 0.1 mm), with each voxel corresponding to an independently controllable design unit. Through voxel discretization, the complex 3D design space is transformed into a quantifiable and computable set of discrete units. Subsequently, the density distribution of the lattice structure can be controlled by adjusting the "presence" or "density" of each voxel, providing a basic computational unit framework for generating variable-density lattice structures.

[0063] For example, the geometric contours and spatial location information of two high strain energy density regions in the feature response set can be extracted first. For the root subdomain of the tube body (φ20mm×30mm cylindrical segment, center coordinates (120,0,30)mm), the three-dimensional voxel design space is set as a cuboid region containing the cylindrical segment and extending outward by 1.5mm, with dimensions of 23mm×23mm×33mm (length×width×height, extending 1.5mm along the cylindrical axis and radially). For the locking mechanism slot subdomain (15mm×8mm×5mm cuboid, center coordinates (85,5,15)mm), the design space is set as a cuboid region with a boundary extension of 1mm, with dimensions of 17mm×10mm×7mm, ensuring complete coverage of the weak area without excessive redundancy. Combined with the design attribute set requirements (mechanical performance needs to be improved by 30%, overall mass increase not exceeding 12%, minimum unit size of 3D printing process 0.4mm), the target relative density range is determined to be 0. The design space is 0.35-0.55, with the root subdomain (metal material) taking 0.4-0.55 (requiring higher strength) and the slot subdomain (plastic material) taking 0.35-0.45 (balancing strength and lightweight). A Python script is used for parametric voxel discretization, with a uniform voxel size of 0.2mm (meeting minimum process size requirements and balancing computational efficiency). A spatial meshing algorithm divides the two design spaces into 115×115×165 (root of the tube) and 85×50×35 (slot) cubic voxels, respectively. Each voxel is assigned a unique coordinate identifier (e.g., root of the tube voxel coordinates (x, y, z), x∈0-114, y∈0-114, z∈0-164). It is also recorded whether each voxel belongs to the high strain energy density core region. This results in two independent three-dimensional voxelized design spaces, each presented as a discrete voxel set, which can be directly used for subsequent density distribution control and topology generation of lattice elements.

[0064] S320 maps the dynamic mechanical response spectrum data in the characteristic response set to a three-dimensional voxelized design space, grows a lattice unit network structure in the three-dimensional voxelized design space, and forms an initial lattice topology.

[0065] It is understandable that dynamic mechanical response spectrum data (such as stress distribution and strain energy density distribution) is the core data reflecting the mechanical requirements within high strain energy density regions. Mapping it to a three-dimensional voxel-based design space essentially assigns a mechanical requirement weight to each voxel. For example, if a voxel's location experiences high peak stress and high strain energy density during impact, its mechanical requirement weight is high; conversely, it is low. The mapping process is implemented through interpolation algorithms, transforming discrete mechanical response data into a continuous spatial distribution field (such as stress distribution field and strain energy density distribution field), ensuring that each voxel has a corresponding quantitative index of mechanical requirements. The growth of the lattice unit network structure is a "growth on demand" process based on this spatial distribution field: guided by the mechanical requirement weight, lattice units are preferentially grown in regions with high weight (high stress, high strain energy density regions), and the unit density is high (the proportion of voxels occupied by lattice units is high); in regions with low weight (low stress, low strain energy density regions), growth is delayed or low-density lattice units are grown (the proportion of voxels occupied is low). The lattice elements can be selected from common topological types (such as tetrahedral elements, octahedral elements, and rhombic dodecahedral elements). During the growth process, it is necessary to ensure the connectivity between elements (avoiding isolated elements) and structural integrity (ensuring that the lattice structure can form a continuous load-bearing network). Through this growth method, the initial lattice topology naturally exhibits a non-uniform density distribution that matches the mechanical response distribution, achieving the optimization goal of "where the mechanical demand is high, the structure is denser."

[0066] Optionally, S320 maps the dynamic mechanical response spectrum data in the characteristic response set to a three-dimensional voxelized design space, grows a lattice unit network structure in the three-dimensional voxelized design space, and forms an initial lattice topology, including: S321 maps the dynamic mechanical response spectrum data in the characteristic response set to the corresponding position in the three-dimensional voxelized design space, generating a spatial distribution field that reflects the intensity of the mechanical response at each position.

[0067] It is understandable that dynamic mechanical response spectrum data (such as stress-time curves and cumulative strain energy data) contains information on the mechanical behavior of different locations within a high strain energy density region during the impact process. However, this data is extracted from finite element units based on previous simulations, and its spatial distribution does not perfectly match the voxel mesh in the three-dimensional voxelized design space. Therefore, it is necessary to map the dynamic mechanical response spectrum data to the corresponding locations in the three-dimensional voxelized design space. The mapping process can be achieved by using spatial interpolation algorithms (such as Kriging interpolation and linear interpolation) to extend the mechanical response data (such as maximum stress value and cumulative strain energy density) on the finite element units to each voxel in the three-dimensional voxelized design space, thus realizing the spatial continuity of the mechanical response data. For example, if the maximum stress values ​​of two adjacent finite element units are 500 MPa and 300 MPa, respectively, the maximum stress value of each voxel located between these two units (between 300 and 500 MPa) can be calculated using an interpolation algorithm. The generated spatial distribution field is a three-dimensional continuous field, with each voxel having a corresponding mechanical response intensity value (such as stress intensity value and strain energy density intensity value). This three-dimensional continuous field can intuitively and accurately reflect the differences in mechanical requirements at various locations within the design space. Locations with higher intensity values ​​require stronger structural support to resist impact and disperse stress; locations with lower intensity values ​​have lower structural strength requirements and can achieve lightweighting through low-density structures. The generation of the spatial distribution field provides a clear signal for the subsequent non-uniform growth of lattice structures.

[0068] S322, based on the spatial distribution field, determines the non-uniform spatial distribution of the initial growth points in the three-dimensional voxel design space, and performs non-uniform spatial expansion according to the intensity value of the initial growth points in the spatial distribution field to form the initial lattice topology.

[0069] It can be understood that initial growth points are the "seeds" of the lattice unit network structure, and their spatial distribution directly determines the density distribution trend of the subsequent lattice structure. The non-uniform distribution of initial growth points is determined based on the spatial distribution field. The principle is that the higher the mechanical response intensity, the greater the density of initial growth points. For example, in the top 30% of the spatial distribution field intensity value (high mechanical demand region), one initial growth point is set for every 10 voxels; in the bottom 30% of the intensity value (low mechanical demand region), one initial growth point is set for every 50 voxels. This non-uniform distribution ensures that a denser lattice structure can be generated in the high mechanical demand region, while a sparser lattice structure is generated in the low mechanical demand region, thus conforming to the mechanical demand distribution from the source. Spatial expansion of the initial growth point refers to the process by which lattice units extend and connect to surrounding voxels from the growth point as the center. The expansion speed and range are determined by the intensity value of the spatial distribution field at the location of the growth point: growth points with high intensity values ​​expand quickly (occupying more adjacent voxels in one iteration), expand over a wider range (forming larger unit clusters), and have tighter connections between units (high lattice density); growth points with low intensity values ​​expand slowly, expand over a smaller range, and have relatively loose unit connections (low lattice density). During the expansion process, collision detection algorithms are needed to prevent lattice units generated from different growth points from penetrating each other, while ensuring effective connections between units (e.g., the centroid distance between adjacent units does not exceed 1.5 times the unit's side length). Through a non-uniform growth process, the final initial lattice topology can exhibit a density gradient consistent with the spatial distribution field, achieving a precise match between mechanical properties and structural distribution.

[0070] For example, S322, determining the non-uniform spatial distribution of initial growth points in the three-dimensional voxelization design space based on the spatial distribution field, and performing non-uniform spatial expansion according to the intensity value of the initial growth points in the spatial distribution field to form an initial lattice topology, including: S3221, based on the intensity value distribution of the spatial distribution field, an adaptive point placement strategy is adopted to determine the position and density of the initial growth points in the three-dimensional voxelized design space; wherein, the higher the intensity value of the spatial distribution field, the greater the spatial distribution density of the initial growth points.

[0071] It is understandable that the adaptive point placement strategy is relative to the uniform point placement strategy. Its purpose is to dynamically adjust the distribution density of the initial growth points according to the changes in the intensity value of the spatial distribution field, so as to avoid the problems of insufficient growth points in high mechanical demand areas and redundant growth points in low mechanical demand areas caused by uniform point placement.

[0072] For example, please refer to Figure 4The 3D voxel design space can be divided into multiple intensity ranges (e.g., high, medium, and low) based on the intensity value of the spatial distribution field. Each intensity range corresponds to a growth point density threshold. For example, the growth point density threshold for the high intensity range (intensity value ≥ 70% of the maximum value) is set to 10 points / cm³, for the medium intensity range (30% of the maximum value ≤ intensity value < 70% of the maximum value) it is set to 5 points / cm³, and for the low intensity range (intensity value < 30% of the maximum value) it is set to 1 point / cm³. Within each intensity range, initial growth points are generated using a random sampling algorithm (e.g., Poisson disk sampling, ensuring the minimum distance between growth points is not less than a preset value to avoid excessive density). The number of samples is determined by the volume of the range and the corresponding density threshold. For example, if the volume of the high intensity range is 2cm³ and the density threshold is 10 points / cm³, then 20 initial growth points need to be generated; if the volume of the low intensity range is 5cm³ and the density threshold is 1 point / cm³, then 5 initial growth points need to be generated. Through an adaptive dot distribution strategy, the initial growth points can be concentrated in areas with high mechanical requirements and sparsely distributed in areas with low mechanical requirements, laying the foundation for the non-uniform growth of the subsequent lattice structure and ensuring that the density distribution of the lattice structure can accurately match the distribution of mechanical requirements.

[0073] S3222, with each initial growth point as the center, grows according to the spatial distribution field intensity value of the initial growth point location to obtain a series of Voronoi cells with non-uniform size and shape. Connect the centroids of adjacent Voronoi cells after growth to obtain the initial lattice topology.

[0074] As can be understood, a Voronoi cell is a geometric partitioning method based on a set of spatial points. Centered on each initial growth point, it divides the three-dimensional space into non-overlapping cells. The distance from any point within a cell to its corresponding growth point is less than the distance to any other growth point. The growth of Voronoi cells is controlled based on the intensity value of the spatial distribution field, aiming to adjust the size and shape of the cells: the higher the intensity value at the growth point, the larger the cell's growth radius (larger cell volume), and the closer the cell shape is to a regular polyhedron (ensuring structural strength); the lower the intensity value, the smaller the cell's growth radius (smaller cell volume), and the more irregular the shape can be (reducing material usage). For example, a Voronoi cell generated by a growth point in a high-intensity region has a radius of 0.5 mm and a shape close to a regular octahedron; a cell generated by a growth point in a low-intensity region has a radius of 0.2 mm and an irregular polyhedron shape, allowing the distribution of Voronoi cells to naturally exhibit a density gradient consistent with the spatial distribution field. High-intensity regions have large cell volumes and dense distribution (many cells per unit space), while low-intensity regions have small cell volumes and sparse distribution. After cell growth, the centroids of adjacent Voronoi cells are connected by rods to form a spatial lattice network structure: cell centroids serve as lattice nodes, and connecting rods serve as lattice units. Regions with high cell density correspond to dense lattice units (high relative density), while regions with low cell density correspond to sparse lattice units (low relative density). The resulting initial lattice topology not only has a non-uniform density distribution but also possesses good structural continuity and mechanical load-bearing capacity, enabling the optimization goal of "on-demand allocation."

[0075] S330 performs equivalent mechanical performance analysis and rapid impact response evaluation on the initial lattice topology configuration. When the lattice topology configuration meets the performance requirements of the design attribute set, it is included in the optimized design scheme set.

[0076] It is understandable that the initial lattice topology is generated based on mechanical response data, but there may be unconsidered factors (such as the connection strength of lattice elements and the dynamic stability of the overall structure). Therefore, performance analysis and evaluation are needed to verify whether it meets the design requirements. Equivalent mechanical performance analysis calculates key mechanical parameters of the lattice structure, such as the equivalent elastic modulus, equivalent yield strength, and equivalent fracture toughness, using a simplified model. These parameters reflect the overall load-bearing capacity and energy absorption capacity of the lattice structure and are core indicators for rapid evaluation. Rapid impact response evaluation, based on the previously established load case sample set, uses a simplified explicit dynamic model (such as equating the lattice structure to a continuous medium model or using a coarse-grid discretized model) to quickly calculate response indicators such as peak stress, maximum deformation, and strain energy absorption under key impact conditions, avoiding excessive computation time caused by using a refined model for full-process simulation. The performance requirements in the design attribute set are the evaluation criteria. For example, the equivalent elastic modulus must be no less than 1.1 times that of the original structure (to ensure increased stiffness), the maximum stress peak during impact must be less than 80% of the material's dynamic yield strength (to avoid plastic deformation), and the total mass of the structure must not exceed 1.2 times that of the original structure (to meet lightweight requirements). If the equivalent mechanical parameters and impact response indices of the initial lattice topology configuration meet these requirements, it indicates that the configuration has achieved a balance between mechanical performance and lightweighting, and has practical application value, thus being included in the optimized design scheme set; if not, it needs to enter the subsequent iterative adjustment stage.

[0077] For example, the initial lattice topology of the pipe root and the locking mechanism slot can be analyzed for equivalent mechanical properties. A simplified model is constructed using homogenization theory combined with the finite element method. The lattice configuration of the pipe root (metallic material, relative density 0.45) yields an equivalent elastic modulus of 28 GPa, an equivalent yield strength of 320 MPa, and an equivalent fracture toughness of 48 MPa·m^(1 / 2). The lattice configuration of the slot (plastic material, relative density 0.4) yields an equivalent elastic modulus of 3.2 GPa, an equivalent yield strength of 45 MPa, and an equivalent fracture toughness of 3.5 MPa·m^(1 / 2). Subsequently, a rapid impact response assessment is performed. Three key working conditions are selected (high energy 40 J + root impact point + vertical impact, medium energy 22 J + middle impact point + radial impact, low energy 8 J + end impact point + axial impact). A coarse-mesh discretization model (element size 0.8 mm, which is 4 times larger than the mesh size of the refined model) is used for explicit dynamic simulation to extract response indices. The root lattice configuration of the tube body exhibits a peak stress of 490 MPa under high-energy conditions (lower than 80% of the dynamic yield strength of metal (610 MPa), i.e., 488 MPa, close to the threshold), a maximum deformation of 4.2 mm, and a strain energy absorption of 38 J; the slot lattice configuration exhibits a plastic strain of 0.042 mm under medium-energy conditions (lower than the ultimate plastic strain of plastic (0.05 mm), a maximum deformation of 1.1 mm, and a strain energy absorption of 5.8 J; simultaneously, the structural mass is calculated, and the mass of the root lattice configuration is greater than that of the original... The structural mass increased by 9.2%, and the slot lattice configuration increased by 7.8%, both of which did not exceed the upper limit of 12% mass increase required by the design attribute set. Compared with the design requirements (equivalent elastic modulus not less than 1.1 times that of the original structure, peak stress / plastic strain less than the failure threshold, mass increase ≤ 12%), the initial lattice configurations of the tube root and the slot both meet all performance requirements. Both were included in the optimized design scheme set, and the equivalent mechanical parameters, impact response indicators and mass data of each configuration were recorded simultaneously to form a scheme ledger.

[0078] In one possible implementation, S300, based on the characteristic response set and the design attribute set, performs variable density lattice structure generation processing on the high strain energy density region to obtain an optimized design scheme set, and also includes: S340, when the lattice topology does not meet the performance requirements of the design attribute set, adjust the growth parameters of the growing lattice unit network structure and iterate again until the preset conditions are met; wherein, the preset conditions include the lattice topology generated by the iteration meeting the performance requirements of the design attribute set or the iteration number reaching the iteration limit.

[0079] It is understandable that when the lattice topology does not meet the performance requirements of the design attribute set, the reasons for the initial lattice topology failing to meet the design requirements may be varied. For example, insufficient lattice density in high-mechanical-demand areas may lead to excessive stress peaks, or excessive overall lattice density may cause the weight to exceed the limit. In such cases, optimization is required by adjusting growth parameters. Adjustable growth parameters include: the initial growth point density threshold (e.g., increasing the growth point density in high-strength areas to improve the lattice unit density in those areas), the growth radius coefficient of the Voronoi cell (e.g., increasing the cell growth radius in high-strength areas to increase unit volume and connection strength), the diameter of the lattice unit members (e.g., thickening the member diameter in high-stress areas to improve local load-bearing capacity), and the density gradient change rate (e.g., slowing down the rate of density change from high to low to avoid stress concentration caused by abrupt structural changes). After each adjustment of growth parameters, a new lattice topology is generated, and performance analysis and evaluation are performed until the preset conditions are met, forming an iterative cycle. Preset conditions include that the iteratively generated lattice topology meets the performance requirements of the design attribute set or the number of iterations reaches the iteration limit (e.g., setting the iteration limit to 5 times) to avoid inefficiency caused by infinite iteration. If the design requirements still cannot be met after multiple iterations, it may be necessary to re-examine the rationality of the design attribute set (e.g., whether excessive pursuit of lightweighting has sacrificed necessary mechanical performance) or the accuracy of the characteristic response set (e.g., whether key mechanical response data has been omitted). Through this iterative adjustment mechanism, the performance of the lattice topology can be continuously optimized, ensuring that each scheme included in the final optimized design scheme set meets the design requirements while also taking optimization efficiency into account.

[0080] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0081] Corresponding to the dynamic impact strength simulation design method for luggage handles in the above embodiments, this application also provides a dynamic impact strength simulation design system for luggage handles, wherein each unit of the system can realize each step of the dynamic impact strength simulation design method for luggage handles. Figure 5 The diagram shows the structural block diagram of the dynamic impact strength simulation design system for luggage handles provided in the embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0082] Reference Figure 5 The dynamic impact strength simulation design system for luggage handles includes: The acquisition unit is used to acquire the structural attribute set, historical attribute set, and design attribute set of the target design bag; The response unit is used to perform modeling and simulation processing based on the structural attribute set and the historical attribute set. It simulates the action process of loads on the tie rod target component under different energy levels and action directions through explicit dynamic analysis, extracts the high strain energy density region and geometric weak location information of the tie rod target component, and obtains the characteristic response set. The result unit is used to perform variable density lattice structure generation processing on the high strain energy density region according to the characteristic response set and the design attribute set to obtain an optimized design scheme set; the optimized design scheme set includes at least one non-uniform lattice filling topology configuration scheme for the interior or surface of the target component.

[0083] It should be noted that the information interaction and execution process between the above systems / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0084] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit module can exist physically separately, or two or more unit modules can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0085] This application also provides an electronic device. Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 6 As shown, the electronic device 6 of this embodiment includes: at least one processor 60 ( Figure 6 Only one is shown in the image), at least one memory 61 ( Figure 6 (Only one is shown in the image) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, it causes the electronic device 6 to implement the steps in any of the above embodiments of the dynamic impact strength simulation design method for luggage handles, or causes the electronic device 6 to implement the functions of each unit in the above embodiments of the system.

[0086] For example, the computer program 62 may be divided into one or more units, which are stored in the memory 61 and executed by the processor 60 to complete this application. The one or more units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 62 in the electronic device 6.

[0087] Electronic device 6 can be a computing device or terminal device such as a mobile phone, tablet computer, desktop computer, laptop, handheld computer, and cloud server. This electronic device may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will understand that... Figure 6 This is merely an example of electronic device 6 and does not constitute a limitation on electronic device 6. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0088] The processor 60 can be a Central Processing Unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0089] In some embodiments, the memory 61 may be an internal storage unit of the electronic device 6, such as a hard disk or memory of the electronic device 6. In other embodiments, the memory 61 may be an external storage device of the electronic device 6, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 6. Furthermore, the memory 61 may include both internal and external storage units of the electronic device 6. The memory 61 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 61 can also be used to temporarily store data that has been output or will be output.

[0090] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0091] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0092] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A simulation design method for the dynamic impact strength of luggage trolley rods, characterized in that, The method includes: Obtain the structural attribute set, historical attribute set, and design attribute set of the target bag design; Modeling and simulation are performed based on the structural attribute set and historical attribute set. The action process of loads on the tie rod target component under different energy levels and directions is simulated through explicit dynamic analysis. The high strain energy density region and geometric weak location information of the tie rod target component are extracted to obtain the characteristic response set. Based on the characteristic response set and the design attribute set, a variable density lattice structure generation process is performed on the high strain energy density region to obtain an optimized design scheme set; the optimized design scheme set includes at least one non-uniform lattice filling topology configuration scheme for the interior or surface of the target component.

2. The method as described in claim 1, characterized in that, Based on modeling and simulation processing using the aforementioned structural attribute set and historical attribute set, the action process of loads on the tie rod target component under different energy levels and directions is simulated through explicit dynamic analysis. Information on high strain energy density regions and geometrically weak locations of the tie rod target component is extracted to obtain a characteristic response set, including: Failure mode clustering is performed based on the historical attribute set to obtain a typical failure mode library; Based on the typical failure mode library, a multi-condition load spectrum is constructed and processed to randomly generate a combination of impact energy, load application point position and impact angle that conforms to the probability distribution of actual use scenarios of bags and luggage, forming a load condition sample set. Based on the structural property set and the load condition sample set, the action process of loads on the tie rod target component under different energy levels and directions is simulated through explicit dynamic analysis. The high strain energy density region and geometric weak location information of the tie rod target component are extracted to obtain the characteristic response set.

3. The method as described in claim 2, characterized in that, The failure mode clustering process based on the historical attribute set to obtain a typical failure mode library includes: Based on the historical attribute set, an initial feature vector set describing the load, displacement, and morphological characteristics of each impact event is constructed, and the data is balanced using synthetic minority class oversampling technology to obtain a balanced feature set. The sample similarity matrix of the balanced feature set is calculated based on the radial basis kernel function, and then partitioned using the spectral clustering algorithm to obtain a library of typical failure modes.

4. The method as described in claim 2, characterized in that, The process involves constructing a multi-condition load spectrum based on the typical failure mode library, randomly generating combinations of impact energy, load application point location, and impact angle that conform to the probability distribution of actual bag usage scenarios, forming a load condition sample set, including: Based on the load conditions corresponding to each failure mode in the typical failure mode library, the marginal probability density functions of impact energy, load application point location and impact angle are respectively fitted using the probability density estimation method. Based on the edge probability density function and the predefined correlation constraints between impact energy and the position of the load application point and the impact angle, a joint probability distribution model is obtained; Initial sampling points are generated based on the joint probability distribution model, and the sampling density is adjusted iteratively to generate the load condition sample set.

5. The method as described in claim 2, characterized in that, The process of simulating the action of loads on the tie rod target component under different energy levels and directions through explicit dynamic analysis based on the structural property set and the load condition sample set is used to extract information on the high strain energy density region and geometrically weak location of the tie rod target component, resulting in a characteristic response set, including: Transient nonlinear finite element analysis is performed based on the structural property set and the load condition sample set. By calculating the stress wave propagation, plastic hinge formation process and damage accumulation of the tie rod target component under different representative impact conditions, transient response field data is obtained. Based on the transient response field data, key region identification processing is performed to identify regions where the peak equivalent stress continuously exceeds the dynamic yield strength of the material throughout the entire impact process, as well as regions with high plastic strain gradients, and generate region identification maps. Based on the preset material failure criteria and the region identification map, the target component of the tie rod is segmented into geometric subdomains and its feature parameters are extracted to obtain the feature response set; wherein, the feature response set includes the geometric contour, spatial location information and corresponding dynamic mechanical response spectrum data of at least one of the high strain energy density regions.

6. The method as described in claim 5, characterized in that, The transient nonlinear finite element analysis based on the structural property set and the load condition sample set is performed to obtain transient response field data by calculating the stress wave propagation, plastic hinge formation process, and damage accumulation of the tie rod target component under different representative impact conditions. This includes: Based on the structural attribute set, the tie rod target component is subjected to refined geometric repair and idealization, and the processed model is classified into functional regions. By distinguishing the mechanism regions, a computational geometric model that can be used for simulation analysis is obtained. Based on the computational geometry model, an initial mesh sensitivity analysis is performed to obtain the recommended mesh density for key size regions. Then, based on the recommended mesh density, a partitioned and refined hexahedral dominant mesh is applied to different functional regions in the computational geometry model to obtain a high-quality discrete model. Based on the load condition sample set and the high-quality discrete model, explicit dynamic solutions are performed to obtain a database of time series simulation results of the tie rod target component under different representative impact conditions. The results of each representative impact condition in the time series simulation results database are used to extract and fuse field data to obtain transient response field data.

7. The method as described in claim 1, characterized in that, The feature response set includes the geometric contour, spatial location information and corresponding dynamic mechanical response spectrum data of at least one of the high strain energy density regions. The process of generating a variable density lattice structure for the high strain energy density region based on the characteristic response set and the design attribute set, to obtain an optimized design scheme set, includes: Based on the high strain energy density region and the design properties, the three-dimensional voxelized design space of the lattice structure and the target relative density range are determined, and the three-dimensional voxelized design space is parametrically discretized to obtain the three-dimensional voxelized design space. The dynamic mechanical response spectrum data in the feature response set is mapped to the three-dimensional voxelized design space, and a lattice unit network structure is grown in the three-dimensional voxelized design space to form an initial lattice topology. The initial lattice topology is subjected to equivalent mechanical performance analysis and rapid shock response evaluation. When the lattice topology meets the performance requirements of the design attribute set, the lattice topology is included in the optimized design scheme set.

8. The method as described in claim 7, characterized in that, The step of generating a variable density lattice structure for the high strain energy density region based on the characteristic response set and the design attribute set to obtain an optimized design scheme set further includes: When the lattice topology configuration does not meet the performance requirements of the design attribute set, the growth parameters of the grown lattice unit network structure are adjusted and iterated again until the preset conditions are met; wherein, the preset conditions include the lattice topology configuration generated iteratively meeting the performance requirements of the design attribute set or the number of iterations reaching the iteration limit.

9. The method as described in claim 7, characterized in that, The step of mapping the dynamic mechanical response spectrum data in the feature response set to the three-dimensional voxelized design space, and growing a lattice unit network structure in the three-dimensional voxelized design space to form an initial lattice topology includes: The dynamic mechanical response spectrum data in the feature response set is mapped to the corresponding position in the three-dimensional voxelized design space to generate a spatial distribution field that reflects the intensity of mechanical response at each position. Based on the spatial distribution field, the non-uniform spatial distribution of the initial growth points is determined in the three-dimensional voxelization design space. Based on the intensity value of the initial growth points in the spatial distribution field, non-uniform spatial expansion is performed to form the initial lattice topology.

10. The method as described in claim 9, characterized in that, The process of determining the non-uniform spatial distribution of initial growth points within the three-dimensional voxelization design space based on the spatial distribution field, and performing non-uniform spatial expansion according to the intensity value of the initial growth points in the spatial distribution field to form an initial lattice topology includes: Based on the intensity value distribution of the spatial distribution field, an adaptive point placement strategy is adopted to determine the location and density of the initial growth points in the three-dimensional voxelized design space; wherein, the higher the intensity value of the spatial distribution field, the greater the spatial distribution density of the initial growth points. Centered on each initial growth point, growth is performed based on the intensity value of the spatial distribution field at the location of the initial growth point to obtain a series of Voronoi cells with non-uniform size and shape. The centroids of adjacent Voronoi cells after growth are connected to obtain the initial lattice topology.

Images