Railway vehicle filling dot matrix steel carbon fiber hybrid anti-climbing energy absorption optimization method and device

Abstract

The application discloses a steel-carbon fiber hybrid anti-climbing energy absorption optimization method for a track vehicle filling dot array, and comprises the following steps: establishing a finite element model of an anti-climbing energy absorption device of a track vehicle and setting design variables of a dot array energy absorption structure, generating HQLS dot array energy absorption structure samples in the range of the design variables by adopting a full factor method combined with a Latin hypercube sampling method, and generating response indexes of each sample by finite element simulation calculation; adopting a moving least square method to construct a mechanical property surrogate model, adopting a global adaptive response surface method to perform global optimization on the surrogate model, and obtaining an optimal balance solution of specific energy absorption and average crushing force based on a Pareto front, generating optimal design variables, and substituting the optimal design variables into the finite element model to verify errors; if the errors meet a preset accuracy, optimal parameters are output, otherwise, iteration is updated. The anti-climbing energy absorption optimization method can perform global optimization on the geometry and material configuration of the anti-climbing energy absorption structure of the track vehicle under the conditions of multiple parameters and multiple targets.

Description

Technical Field

[0001] This application relates to the field of energy-absorbing structure design technology for rail vehicles, specifically to an optimization method, device, and system for anti-climbing energy absorption of a steel-carbon fiber hybrid structure with a filling matrix for rail vehicles. Background Technology

[0002] With the development of the rail transit industry, the passive safety of rail vehicles during operation has received widespread attention. To ensure occupant safety and vehicle structural integrity, anti-creep energy-absorbing devices are typically installed at the vehicle ends. The core function of these devices is to absorb kinetic energy through controlled plastic deformation of the structure during a rail vehicle collision, reducing the peak impact load and thus minimizing occupant injury. However, as train speeds continue to increase, traditional anti-creep energy-absorbing structures can no longer simultaneously meet the requirements of high energy absorption efficiency and lightweight design.

[0003] Currently, most common anti-climb energy absorption devices for rail vehicles adopt the square cone type. This type of anti-climb energy absorption device relies on thin-walled buckling to absorb energy and has the advantages of simple manufacturing and low cost. However, when the speed of rail vehicles increases to the medium-high speed range (e.g., ≥17.9 km / h), the energy absorption capacity of pure metal thin-walled anti-climb structures is obviously insufficient. It is often necessary to increase the wall thickness or the overall size to increase the energy absorption. This not only leads to a significant increase in mass, but also causes problems such as an increase in the peak collision force and violent fluctuations in the energy absorption process, thereby affecting the crashworthiness and comfort of the vehicle.

[0004] To achieve a balance between lightweight design and high energy absorption performance, a hybrid metal / composite material energy-absorbing structure scheme has been proposed in recent years. This involves introducing a carbon fiber reinforced composite layer and filling it with a lattice structure inside a thin-walled metal structure. Carbon fibers have high specific strength, high specific stiffness, and excellent fatigue properties, enabling them to work in synergy with the metal layer to absorb energy during collisions. Lattice structures, due to their advantages such as high designability, periodic stability, and high energy absorption efficiency, have become a research hotspot for next-generation energy-absorbing filler materials.

[0005] However, when directly embedding lattice structures into metal / carbon fiber hybrid anti-climb devices, determining the optimal combination of parameters such as lattice wall thickness, pore size ratio, hybrid layer thickness ratio, and layup angle is a key challenge affecting energy absorption performance. Existing lattice structure designs mostly rely on single-factor experiments or empirical selection, resulting in low design efficiency and an inability to reveal the coupling effects between parameters. For energy absorption devices with multivariate coupling, limited simulation experiments alone cannot fully reflect the comprehensive impact of parameter changes on energy absorption performance (such as specific energy absorption area (SEA) and mean crushing force (MCF)).

[0006] Therefore, there is an urgent need for a steel-carbon fiber hybrid anti-climbing energy absorption optimization method for rail vehicle filling lattice that can globally optimize the geometry and material configuration of the anti-climbing energy absorption structure under multi-parameter and multi-objective conditions. Summary of the Invention

[0007] Purpose of the invention: In order to overcome the above shortcomings, the purpose of this application is to provide an optimization method and device for anti-climbing energy absorption of steel-carbon fiber hybrid filler matrix in rail vehicles.

[0008] To address the aforementioned technical problems, this application provides an optimized method for anti-climbing energy absorption of a steel-carbon fiber hybrid structure with a lattice-filled grid on rail vehicles. The method includes an energy-absorbing anti-climbing device with a lattice-filled energy-absorbing structure, comprising the following steps:

[0009] S1: Establish a finite element model of the anti-climbing energy absorption device for rail vehicles and define the design variables of the lattice energy absorption structure in the finite element model. The design variables include the steel / carbon fiber thickness ratio, carbon fiber layup angle, lattice structure wall thickness and lattice structure aperture ratio.

[0010] S2: Using the full factorial method combined with the Latin hypercube sampling method, a preset number of HQLS lattice energy-absorbing structure samples are generated within the set range of the design variables, and the response index of each sample is calculated according to the finite element simulation method. The response index includes specific energy absorption and average crushing force.

[0011] S3: A proxy model of the mechanical properties of the lattice energy-absorbing structure is constructed using the moving least squares method;

[0012] S4: The global adaptive response surface method is used to perform global optimization on the surrogate model and determine the optimal balance solution between structural specific energy absorption and average crushing force based on the Pareto front, thereby generating the optimal design variables;

[0013] S5: Substitute the optimal design variables into the finite element model to recalculate and determine whether the error is within the preset range. If yes, output the optimal design variables; otherwise, return to step S2 to regenerate the samples.

[0014] As a preferred embodiment of this application, in step S1, the finite element model of the rail vehicle anti-climb energy absorption device includes anti-climb teeth, front end plate, square cone shell, rear end plate, guide rod and anti-climb energy absorption structure;

[0015] The anti-climb energy-absorbing structure adopts an HQLS lattice energy-absorbing structure, which is formed by a number of interconnected cell units distributed along the z-axis. Each cell unit has a hexagonal base and a quadrilateral top surface, and the overall structure of the cell unit is frustum-shaped. Furthermore, the edges of each cell unit are closely connected to the edges of adjacent cell units to form a close-fitting arrangement structure, thereby increasing the energy absorption capacity of the structure by adding plastic hinges.

[0016] As a preferred embodiment of this application, in step S1, the individual cell unit of the HQLS lattice energy-absorbing structure is divided into two trapezoids, two isosceles triangles, and four ordinary triangles.

[0017] Based on the volumes of the trapezoidal, isosceles triangle, and ordinary triangle, the relative density of a single cell unit in the HQLS lattice energy-absorbing structure is calculated using the following formula:

[0018] , wherein Let the volume of the trapezoid be... The volume of a regular triangle, Let the volume of the isosceles triangle be . The height of the frustum of a single cell unit. The side length of a single cell unit.

[0019] As a preferred embodiment of this application, in step S2, the Latin hypercube sampling method calculates the number of runs ( The formula for ) is:

[0020] in, The number of design variables.

[0021] As a preferred embodiment of this application, in step S3, the fitting function approximated by the moving least squares method is:

[0022] ,in The surrogate model predicts the output value at input variable point x. For the basis function vector, These are coefficients to be determined; The transpose symbol indicates It is the inner product of the two;

[0023] Wherein, basis function vector For linear and quadratic monomials: .

[0024] As a preferred embodiment of this application, after constructing a proxy model of the mechanical properties of the lattice energy-absorbing structure in step S3, the method includes:

[0025] S31: Calculate the relative mean absolute error of the surrogate model:

[0026] ,in The corresponding predicted value of the response at each sampling point. For the true value of the sample point, The number of test points;

[0027] S32: Calculate the root mean square error of the surrogate model:

[0028] ;

[0029] S33: Calculate the fitting correlation coefficient of the surrogate model:

[0030] ;

[0031] S34: Achieve local approximation of the surrogate model by weighted least squares fitting and make the specific energy absorption and average crushing force of the surrogate model meet the preset accuracy requirements.

[0032] In a preferred embodiment of this application, step S4 includes:

[0033] With the primary objective of increasing the structural energy absorption ratio and the secondary objective of bringing the average crushing force close to the reference value of existing square cone energy-absorbing anti-climb devices, the following optimization objectives are established:

[0034] ,in For specific energy absorption, For lattice structure wall thickness, The aperture ratio of the lattice structure, The steel / carbon fiber thickness ratio, The angle of the carbon fiber layup. The average crushing force; These are reference values ​​for existing cone-shaped energy-absorbing anti-climb devices.

[0035] As a preferred embodiment of this application, the parameter range of the steel / carbon fiber thickness ratio is set to [0.5, 1.5]; the parameter range of the carbon fiber layup angle is set to [30, 90]; the parameter range of the lattice structure wall thickness is set to [0.18, 0.21]; and the parameter range of the lattice structure aperture ratio is set to [0.45, 0.55].

[0036] This application also provides a steel-carbon fiber hybrid anti-climb energy-absorbing device for rail vehicles with a filled dot matrix designed using the aforementioned method, characterized in that it comprises:

[0037] Front-end board;

[0038] Back-end board;

[0039] Anti-climb teeth, which are installed on the front panel;

[0040] A square-pyramidal shell, which is connected to the front end plate and the rear end plate respectively;

[0041] The guide rod connects to the rear end plate;

[0042] At least one lattice energy-absorbing structure is filled inside the square pyramidal shell. The lattice energy-absorbing structure is an HQLS lattice energy-absorbing structure, which is formed by a plurality of interconnected cell units arranged in an array along the z-axis. Each cell unit has a hexagonal base and a quadrilateral top surface. The overall structural shape of the cell unit is frustum-shaped. The edges of each cell unit are closely connected to the edges of adjacent cell units to form a close-fitting arrangement structure, thereby improving the energy absorption capacity of the structure by adding plastic hinges.

[0043] This application also provides a steel-carbon fiber hybrid anti-climb energy absorption system for rail vehicles using the method described above, comprising:

[0044] The simulation construction module is used to establish a finite element model of the anti-climbing energy absorption device for rail vehicles and define the design variables of the lattice energy absorption structure in the finite element model. The design variables include the steel / carbon fiber thickness ratio, carbon fiber layup angle, lattice structure wall thickness and lattice structure aperture ratio.

[0045] The sample generation module is used to generate a preset number of HQLS lattice energy-absorbing structure samples within the set range of the design variables using a full factorial method combined with the Latin hypercube sampling method, and to calculate and generate the response index of each sample group according to the finite element simulation method. The response index includes specific energy absorption and average crushing force.

[0046] The model building module is used to construct a proxy model of the mechanical properties of the lattice energy-absorbing structure using the moving least squares method.

[0047] The target optimization module is used to perform global optimization of the surrogate model using the global adaptive response surface method and determine the optimal balance solution between structural specific energy absorption and average crushing force based on the Pareto front, generating optimal design variables; it is also used to substitute the optimal design variables into the finite element model for recalculation and determine whether the error is within a preset range. If yes, the optimal design variables are output; otherwise, samples are regenerated.

[0048] The technical solution described in this application has the following advantages over the prior art:

[0049] 1. This application constructs a fully parameter-adjustable finite element model by setting four design variables: steel / carbon fiber thickness ratio, carbon fiber layup angle, lattice structure wall thickness, and lattice aperture ratio. This model can quantitatively describe and adjust the anti-climb energy absorption performance from three aspects: geometry, materials, and lattice topology, providing a framework for the comprehensive performance optimization of anti-climb energy absorption structures.

[0050] 2. This application combines full factorial design with Latin hypercube sampling to reasonably determine the number of calculation samples, so that each design variable is evenly distributed in the global scope, thereby effectively reducing the number of finite element simulations and improving the coverage and statistical balance of sampling. Thus, the efficiency of structural optimization is improved while ensuring the accuracy of calculation.

[0051] 3. This application constructs a proxy model of the lattice energy-absorbing structure using the moving least squares method, which can accurately approximate the nonlinear response characteristics in local regions. At the same time, the accuracy of the model is quantitatively verified by calculating RAAE, RMSE, and R² indices to ensure that the error of the proxy model is controlled within a preset threshold. This method effectively replaces a large number of finite element calculations, saving more than 70% of the calculation time.

[0052] 4. By taking the maximization of specific energy absorption and the stabilization of average crushing force as multiple optimization objectives, the GRSM algorithm was used to achieve global optimization based on the Pareto front, and the optimal combination of design variables that balances the improvement of specific energy absorption and the stability of average crushing force was obtained. As a result, the specific energy absorption of the optimized anti-climb energy absorption structure is about 35.7% higher than that of the traditional anti-climb device, and the error of the average crushing force from the reference value is only 1.2%.

[0053] 5. This application sets up a verification feedback design in the optimization process, and re-substitutes the optimal design variables into the finite element model for calculation. When the error between the prediction result and the simulation result exceeds the preset range, it automatically returns to resampling and modeling, realizing a closed-loop optimization mechanism of sampling, modeling, optimization, verification and correction, thereby ensuring the accuracy of the model and the engineering feasibility of the optimization results. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, 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 embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0055] Figure 1 This is a schematic diagram of the energy-absorbing and anti-climbing device at the front end of a rail vehicle provided in an embodiment of this application.

[0056] Figure 2 This is a schematic diagram of the dot matrix structure provided in the embodiments of this application.

[0057] Figure 3 This is a schematic diagram of the quasi-static axial compression finite element model of the HQLS lattice structure provided in the embodiments of this application.

[0058] Figure 4 This is a schematic diagram illustrating the derivation of the relative density of the HQLS lattice structure provided in the embodiments of this application.

[0059] Figure 5 This is a schematic diagram showing the convergence analysis of the HQLS lattice mesh provided in the embodiments of this application and its comparison with experimental results.

[0060] Figure 6 This is a timing diagram of the crushing of the HQLS lattice structure provided in the embodiments of this application.

[0061] Figure 7 This is a schematic diagram showing the comparison before and after the compression test of the HQLS lattice structure provided in the embodiments of this application.

[0062] Figure 8 This is a timing diagram of the quasi-static compression deformation of the HQLS lattice structure for filling content provided in the embodiments of this application.

[0063] Figure 9 This is a flowchart illustrating the optimized method for anti-climbing energy absorption using a steel-carbon fiber hybrid filler matrix for rail vehicles provided in this application embodiment.

[0064] Figure 10 This is a schematic diagram of the geometric configuration of the end-conical composite energy-absorbing structure provided in the embodiments of this application.

[0065] Figure 11 This is a schematic diagram of the HQLS lattice-filled steel / carbon fiber hybrid energy-absorbing cone-shaped anti-climb device provided in the embodiments of this application.

[0066] Figure 12 This is a schematic diagram of the finite element model of the HQLS lattice-filled steel / carbon fiber hybrid energy-absorbing cone-shaped anti-climb device provided in the embodiments of this application.

[0067] Figure 13 This is a schematic diagram of Latin hypercube sampling provided in an embodiment of this application.

[0068] Figure 14 This is a schematic diagram of the principal cause analysis provided in the embodiments of this application; wherein, Figure (a) shows the influence of each design variable on MCF, and Figure (b) shows the influence of each design variable on SEA.

[0069] Figure 15Figure 1 shows the MCF and SEA response surface provided in the embodiments of this application; wherein, Figure (a) shows the effect of steel / carbon fiber hybrid tube parameters on MCF, Figure (b) shows the effect of HQLS lattice structure parameters on MCF, Figure (c) shows the effect of steel / carbon fiber hybrid tube parameters on SEA, and Figure (d) shows the effect of HQLS lattice structure parameters on SEA.

[0070] Figure 16 This is a schematic diagram illustrating the operating principle and steps of GRSM provided in the embodiments of this application.

[0071] Figure 17 This is a schematic diagram of the Pareto front solution provided in the embodiments of this application.

[0072] Figure 18 This is a schematic diagram comparing the optimal solution provided in the embodiments of this application with the energy absorption of traditional energy-absorbing structures.

[0073] Figure 19 This is a schematic diagram of the module connection of the steel-carbon fiber hybrid anti-climb energy absorption optimization system provided in the embodiments of this application. Detailed Implementation

[0074] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0075] Conical energy-absorbing anti-climbing devices, as an important component of multi-stage energy absorption in trains, have received extensive research. Currently used conical energy-absorbing anti-climbing structures mainly consist of a thin-walled metal structure filled with honeycomb. The thin-walled metal structure, as a low-cost, high-strength-to-weight ratio, and high-energy-absorbing structure, has been widely used in dedicated energy-absorbing devices for railway vehicles. However, with the increasing standards for collision safety speeds, a single thin-walled structure is no longer sufficient to meet the energy consumption requirements under higher standards. Lightweight, high-strength honeycomb structures are used as the core material within the thin-walled structure. This method effectively utilizes the residual space of the thin-walled tube, and this type of conical tube is widely recognized as a lightweight and efficient energy-absorbing device with a high stroke ratio.

[0076] However, as train speeds increase, the demand for energy-absorbing capacity in energy-absorbing devices also gradually increases. This increased capacity is primarily reflected in increased geometric dimensions and mass. Considering the installation space for energy-absorbing devices and current lightweighting requirements, it is desirable for existing structures to absorb as much impact energy as possible within limited size and lightweighting standards. While meeting energy absorption requirements, the impact force of the energy-absorbing device should be as gentle as possible, minimizing or avoiding large peak values. The magnitude and fluctuation of the impact force during a collision directly affect the stability of other structures and the safety of occupants. Prolonged, excessively large impact forces or violently fluctuating impact forces pose a significant threat to occupant safety and the overall structure of the rail vehicle.

[0077] Energy-absorbing structures are mainly used for the orderly absorption of energy during train collisions, ensuring the passive safety of trains, such as... Figure 1 As shown. The mechanical characteristics of the energy-absorbing device affect parameters such as load, speed, and acceleration during a train collision, which are related to the safety of the occupants and the stability of the collision process.

[0078] Therefore, to meet design requirements and ensure energy absorption, the cone-shaped energy-absorbing structure needs to achieve the platform force specified in the design. Under current energy conservation and emission reduction demands, a lightweight design is necessary for traditional cone-shaped energy-absorbing structures. The design requirement for the cone-shaped energy-absorbing device is to generate a controllable deformation mode during a train collision, maintain a stable impact force, absorb as much collision energy as possible, and minimize the peak impact force during the collision process; a steel / carbon fiber hybrid structure filled with HQLS lattice can achieve this goal.

[0079] HQLS (hexagonal-quadrilateral porous lattice structure) is an abbreviation for hexagonal-quadrilateral lattice structure. The lattice structure used in this application is based on a secondary design of a flat-topped pyramid (HBPS). The aim is to obtain lower mass and higher specific energy absorption. The flat-topped pyramid is mainly designed based on the biomimetic principle of bamboo joints or lotus root joints. Due to its low center of mass, the flat-topped pyramid has good stability under external loads and has been extensively studied.

[0080] According to the theory of hyperfolded elements, lattice structures mainly absorb the kinetic energy of impact processes through the deformation of plastic hinges. In a flat-topped pyramid unit cell structure, there are mutual influences and constraints between each facet and cell element. Therefore, this application considers increasing the number of plastic hinges in the unit cell to improve the energy absorption capacity of the structure. Based on the flat-topped pyramid, the shape of its base is modified, and the connection between cells is ensured to be relatively tight, that is, each edge of the cell is interconnected, resulting in the following novel HQLS lattice structure.

[0081] To compare the energy absorption capacity and mass change of the new structure with the flat-topped pyramid structure, the thickness of both structures was set to 1 mm, and the length, width, and height of each individual cell were limited to 10 × 10 × 10 mm. 3 Within a cube. Furthermore, the apex shapes of several structures are all inscribed within a 5×5... Within a square. Its total height is 40 mm. The newly constructed lattice structure has a square top edge, with a side length of [missing information]. The lower base is hexagonal. The circumscribed shape of the lower base is a square with a side length of [missing value]. The height of the frustum is set to... A single cell structure is a mirror image of a frustum along the xOy plane, with a height of... The unit cell lattice structure is then arranged in a close-packed manner, and then arrayed along the z-axis to reach the required height. Each cell is 10mm high. The specific arrangement and structure are as follows: Figure 2 As shown.

[0082] This application constructs a finite element model of an L64 lattice structure and uses LS-Dyna to simulate the lattice structure. The thickness is set to 1 mm. The finite element model of the lattice structure is as follows: Figure 3 As shown. The lattice structure was placed on a rigid wall, with no constraint at the bottom. The rigid wall at the bottom was used to obtain data on the axial force during compression. The coefficient of friction between the specimen and the rigid wall was set to 0.3. A constant-speed pressure plate was applied to the top of the specimen for compression. The pressure plate used rigid shell elements, and the compression distance was 24 mm, which is the height to achieve densification of the lattice structure. Considering that the structure below the contact surface may come into contact with the pressure plate after the lattice structure deforms, a pressure plate was applied between the pressure plate and the lattice structure. The contact-automatic surface-to-surface method uses a friction coefficient of 0.3. To improve computational accuracy, the lattice structure is mass-scaled so that the ratio of its internal energy to kinetic energy is less than 5%. The applied velocity on the pressure plate is 2 m / s. Material simulation is performed using MAT24 material. This application uses Belyschko-Tsay quadrilateral shell elements with 5 integration points in the thickness direction to simulate aluminum alloys.

[0083] For lattice structures, relative density is a crucial indicator of their energy absorption capacity and lightweight design. Therefore, it is necessary to derive the relationship between the geometric parameters of the lattice structure and its relative density for subsequent use. The lattice structures used above are all obtained by arraying their unit cells. Since the structures used in this application are arranged in a close-packed manner, their relative density is the volume ratio within the cube after arrangement. The number of arranged lattice structures can be calculated as an integer; for example, the lattice structure used above contains 80 identical unit cells. Therefore, to calculate the relative density of the lattice, it is necessary to first study the volume of a single cell. The unit cell of the lattice structure is a frustum mirrored along the xOy plane. Then, the unit cell lattice structures are arranged in a close-packed manner. Finally, they are arrayed along the z-axis to reach the required height.

[0084] The key geometric parameters of the lattice structure unit cell are listed in the upper figure. The upper base is a square, and its side length is set to... The lower base is hexagonal, and the circumscribed shape is a square with a side length of . The height of the frustum is set to .Depend on Figure 4 It can be seen that the unit cell structure can be divided into several different faces, mainly into three categories: two trapezoidal faces, two isosceles triangular faces, and four ordinary triangular faces. Therefore, the volume of the unit cell structure will be obtained by calculating the volume of these different faces. In the calculation process, it is first necessary to assume that the unit cell lattice structure has a uniform wall thickness. The first part is a trapezoidal surface, and the length of the upper base of the trapezoidal surface is... The length of the lower base is The height of the trapezoidal surface is a variable related to the height of the unit cell; according to the trapezoidal area formula, the relationship between the volume, thickness, and other geometric parameters of the trapezoidal surface can be obtained:

[0085] ;

[0086] For an isosceles triangle, since the circumscribed surfaces of both the top and base are squares, the height of the isosceles triangle is the same as the height of the trapezoidal surface. Therefore, the volume of the isosceles triangle can be calculated as follows:

[0087] ;

[0088] For four ordinary triangular faces, due to the symmetry of the cell, their shape and area remain consistent. To obtain the height of the triangular face... Project point C onto the plane containing point AB and record it as point D. For example... Figure 4 As shown, we need to first calculate the distance from point D to line AB. Line AB can be recorded as... The coordinates of point D are Then the distance DE from point D to line AB is:

[0089] ;

[0090] Wherein, the coordinates of point D correspond to ( ), This is the length of line CE. From this, the volume of the second surface of the lattice structure can be obtained, as follows:

[0091] ;

[0092] The formula for calculating the relative density of a unit cell in a lattice structure is:

[0093] ;

[0094] Based on the arrangement of the unit cell lattice structure and the geometric parameters used, the relative density of the lattice structure can be calculated. Here, compared with the parameter settings in the finite element method, the relative density of the lattice structure calculated by the formula for the relative density of the unit cell is 0.334872, while the relative density calculated by the model is 0.332569, with an error of 0.68%.

[0095] Therefore, a quasi-static compression test was conducted on the HQLS lattice structure. The quasi-static compression test was carried out on an MTS hydraulic testing machine, with a compression speed set to 2 mm / min and a compression amount of 24 mm, accounting for 60% of the total length. The overall structural parameters of the lattice structure are 40×40×40 mm. The force-displacement curve of the test is shown below. Figure 5 As shown.

[0096] Depend on Figure 5 It can be seen that the deformation process of the lattice structure used in this application is mainly divided into three stages: the linear elastic stage, the plateau stage, and the compaction stage; as Figure 5 The AB segment represents the linear elastic stage of the lattice structure. During the quasi-static compression process, the lattice structure does not exhibit a significant peak force, and the compression process is relatively stable. The BC segment is a plateau segment, where the energy absorption process is relatively stable, a plateau force appears, and the lattice structure begins to undergo plastic deformation. After point C, the compaction stage begins, where the axial force gradually increases, the lattice structure becomes denser, and the stage in which the lattice structure dissipates energy through plastic deformation is essentially over. In this stage, the axial force increases. Figure 6 and Figure 7The diagram illustrates the deformation of the lattice structure during compression. The middle layer gradually extends outward and bulges during compression. As the compression displacement increases, cracks appear at the cone angles of the cells. During axial compression, the units on the left and right sides of the second and third layers bulge outward, bending and compressing towards the axis of symmetry, while the middle unit only undergoes compression deformation in the axial direction. Therefore, L64 as a whole exhibits an "X"-shaped deformation. A comparison of the mechanical properties between the HQLS lattice and the flat-top pyramid lattice is shown in Table 1.

[0097] Table 1 Comparison of mechanical properties between HQLS lattice and flat-top pyramid lattice

[0098]

[0099] To balance computational accuracy and efficiency, this application conducted a mesh sensitivity analysis on lattice structures with different mesh sizes. For example... Figure 5 The figure shows the force-displacement curves under different grid sizes, including 0.5 mm, 1 mm, 1.5 mm and 2 mm grids. It can be seen from the figure that the force-displacement curves increase with the increase of grid size when the grid size is different.

[0100] This application compares the errors of various mechanical parameters of finite element models of lattice structures with those of experiments under different mesh sizes. Taking into account both the accuracy of the mechanical parameters and computational efficiency, when the mesh size is 1 mm, the errors of all mechanical parameters are less than 4%, as shown in Table 2.

[0101] Table 2 Comparison of HQLS Mesh Convergence Accuracy

[0102]

[0103] Therefore, this application selects 1 mm as the size of the lattice structure mesh. The error between the experiment and the simulation is small, and it is believed that the finite element model constructed in this application has high reliability.

[0104] Furthermore, this application also conducted a quasi-static compression test on a lattice structure with a height of 100 mm used for filling the hybrid tube, in order to compare the energy of the lattice and the hybrid tube after compression separately with that of the filled hybrid tube. The compression timing diagram of the 100 mm high lattice structure is shown below. Figure 8 As shown.

[0105] During compression, it was observed that the deformation of the lattice structure started from both ends and gradually concentrated towards the center, with the upper lattice structure being compacted first. The HQLS lattice exhibited significant outward bulging during compression, but no cracking occurred. Table 3 shows the errors between the mechanical parameters obtained from the 100mm height HQLS lattice experiment and the finite element simulation.

[0106] Table 3 Comparison of 100 mm lattice test and finite element method accuracy

[0107]

[0108] The maximum value is 3.9%, which proves that the finite element model of the lattice structure has high accuracy.

[0109] Therefore, for reference Figure 9 As shown in some embodiments of this application, this application relates to a method for optimizing the anti-climbing energy absorption of a steel-carbon fiber hybrid dot matrix in rail vehicles, comprising the following steps:

[0110] S1: Establish a finite element model of the anti-climbing energy absorption device for rail vehicles and define the design variables of the lattice energy absorption structure in the finite element model. The design variables include the steel / carbon fiber thickness ratio, carbon fiber layup angle, lattice structure wall thickness and lattice structure aperture ratio.

[0111] S2: Using a full factorial approach combined with the Latin hypercube sampling method, a preset number of HQLS lattice energy-absorbing structure samples are generated within the set range of the design variables. The response index of each sample is calculated using the finite element simulation method. The response index includes specific energy absorption and average crushing force. The set range and preset number are set and adjusted by the designer according to actual needs.

[0112] S3: A proxy model of the mechanical properties of the lattice energy-absorbing structure is constructed using the moving least squares method;

[0113] S4: The global adaptive response surface method is used to perform global optimization on the surrogate model and determine the optimal balance solution between structural specific energy absorption and average crushing force based on the Pareto front, thereby generating the optimal design variables;

[0114] S5: Substitute the optimal design variables into the finite element model to recalculate and determine whether the error is within the preset range. If yes, proceed to step S50: output the optimal design variables. If no, return to step S2 to regenerate the sample. The preset range is set and adjusted by the designer according to actual needs.

[0115] Specifically, the cone-shaped energy-absorbing anti-creep structure is typically installed at the front of the train. As part of a tiered energy absorption process, it absorbs a portion of the kinetic energy during a collision, ensuring passenger safety. Therefore, researching the application of HQLS-filled steel / carbon fiber hybrid structures in anti-creep devices is of great significance in achieving lightweight energy-absorbing structures and ensuring the passive safety of trains. The cone-shaped anti-creep device used in this application mainly includes the following parts: anti-creep teeth, front end plate, filled honeycomb, partition, metal cone tube, guide rod, and rear end plate. Their relative positions are as follows: Figure 10 As shown, the honeycomb cells are symmetrically distributed along the guide rod.

[0116] The cone-shaped energy-absorbing anti-climb device improves its stability during crushing. An arc-shaped guide groove at its front end ensures orderly deformation and reduces the initial peak force. The anti-climb device has a total length of 1072 mm, with front and rear end plate thicknesses of 6 mm and 16 mm, respectively. In traditional cone-shaped energy-absorbing anti-climb devices, the honeycomb at the very front (closest to the anti-climb device) has lower strength, as shown in honeycomb A in the diagram. This ensures the anti-climb device can smoothly trigger and generate a symmetrical plastic hinge during impact, reducing the peak force during impact. Honeycomb A has a wall thickness of 0.02 mm, and honeycomb B has a wall thickness of 0.2 mm. The front face of the anti-climb device is a rectangle of 278 mm × 184 mm, and the rear face is a rectangle of 278 mm × 242 mm, with an increasing cross-section. The outer wall taper is 1.74°. The interior of the anti-climb device is divided into 12 zones by partitions, each zone having a length as shown in the diagram. Figure 10 As indicated. Traditional cone-shaped anti-climb devices fill the spaces between the partitions with honeycomb structures. Each honeycomb segment is 90mm wide, and its length and thickness from front to back are: 150×70 mm, 170×114 mm, 190×114 mm, 190×114 mm, 206×99 mm, 210×99 mm, 210×65 mm, 210×65 mm, 210×65 mm, 226×65 mm, 230×65 mm, and 230×65 mm. The outer thin-walled tube is 2 mm thick, and the partitions are 3 mm thick.

[0117] Referring to the arrangement of traditional cone-shaped anti-climb energy-absorbing devices, the steel / carbon fiber hybrid structure filled with HQLS lattice in this application is applied to the above structure. Specifically, the operation is as follows: For the cone-shaped outer tube, ensuring the total thickness remains the same (i.e., 2mm), the pure steel tube is replaced with a steel / carbon fiber tube, with the steel tube on the outside and the carbon fiber inner tube inside. The internal partition of the cone tube is connected to the steel tube, and the material used for the outer steel tube is consistent with that of the traditional cone-shaped energy-absorbing anti-climb device, namely Q345 steel. The arc-shaped guide groove at the front end of the traditional cone-shaped energy-absorbing anti-climb device is retained. For the internal honeycomb structure, the traditional honeycomb is replaced with the aforementioned HQLS lattice structure, with the lattice unit cell size consistent with the HQLS lattice structure. During the lattice filling process, referring to the honeycomb filling method, the lattice is symmetrically arranged on both sides of the guide rod, maintaining consistent structural dimensions. The anti-climb teeth, front and rear end plates, and guide rod maintain a certain rigidity, preventing deformation during the deformation of the thin-walled hybrid tube and lattice structure. Therefore, referring to... Figure 11 As shown.

[0118] To control variables, the dimensions of the original structure were kept consistent during the replacement of the structure with a hybrid tube and lattice structure. Furthermore, the lattice structure in the first interval (i.e., the lattice structure in contact with the front end plate) was given a lower strength, meaning a smaller thickness, to ensure smooth compression deformation. The lattice thickness was consistent with the honeycomb wall thickness. Moreover, to ensure successful triggering of the cone-shaped energy-absorbing structure, the thickness of the lattice structure at the front end of the cone-shaped energy-absorbing anti-climb device was set to 0.02 mm, and the thickness of the lattice structure in subsequent stages was set to 0.2 mm. The length, width, and height of the lattice were consistent with the honeycomb structure filled in each stage of the traditional cone-shaped energy-absorbing anti-climb device, and symmetrical.

[0119] Therefore, based on the above, a finite element model of a steel / carbon fiber hybrid energy-absorbing structure filled with HQLS lattice is constructed, and its mechanical properties are compared with those of a traditional square pyramidal energy-absorbing structure to determine the range of research parameters in order to obtain the geometric configuration with optimal mechanical performance.

[0120] Specifically, the finite element model reference Figure 12 As shown, the explicit nonlinear finite element code LS-Dyna was used in this application to establish the collision analysis model of the end energy-absorbing structure. Thin-walled components, such as the outer conical square tube and the partition, were modeled using 4-node Belytschko-Tsay shell elements with two integration points on the shell thickness and one integration point on the element plane. However, components such as the front end plate, rear end plate, guide tube, and anti-climbing teeth were modeled using hexahedral solid elements due to their greater thickness and stiffness. Regarding material parameters, the outer wall of the traditional square cone energy-absorbing anti-climbing device is made of Q345 steel, consistent with the metal tube used in this paper. Therefore, the outer wall of the square cone tube and the partition both use Q345 steel material parameters. The aluminum alloy filled in the square cone energy-absorbing anti-climbing device is mainly used to improve energy absorption.

[0121] Regarding the contact setup, self-contact is applied to the entire filling tube. `contact_automatic_single_surface` applies an adhesive surface contact to the interlayer of carbon fiber and the contact surface between carbon fiber and metal tube. (contact_automatic_one_way_surface_to_surface_tiebreak); It is important to note that after the hybrid tube and the lattice structure deform, they will interact. Therefore, surface-to-surface contact needs to be applied to their contact surfaces to prevent mutual penetration. Applying contact between the rigid wall and the lattice structure, and between the carbon fiber inner tube and the lattice structure... The contact method (contact_automatic_surface_to_surface) ensures that no penetration occurs, allowing the model to calculate successfully. The friction coefficient is set to 0.2. For lattice structures, an adhesive contact needs to be established between them and the partition. The contact_tied_node_to_surface method ensures that the lattice does not slide relative to the partition during deformation. Furthermore, the guide rod needs to be connected to the front end plate, and contact points need to be established between the guide rod and the lattice, as the lattice will interact with the guide rod after deformation. To balance computational accuracy and efficiency, the mesh size for the steel / carbon fiber hybrid outer tube is 10 mm, the mesh size for the front and rear end plates and anti-climb teeth (solid elements) is 10 mm, and the mesh size for the lattice structure is 4 mm. A schematic diagram of the finite element model construction is shown below. Figure 12 As shown, an initial velocity of 17.9 km / h is applied to the rigid wall. The rigid wall can only move in the x-direction, and the movement and rotation in other directions are constrained. The rear end plate is fully constrained.

[0122] To ensure computational accuracy, the finite element model uses a fine mesh, which leads to increased computation time and decreased efficiency. Therefore, to efficiently study the impact of various structural parameters on the results, this application employs Design of Experiments (DOE) to investigate influencing factors. The advantage of DOE lies in its ability to simultaneously vary and test the effects of multiple variables and identify the interactions between multiple factors. It allows for drawing conclusions with the fewest possible experiments while maintaining accuracy, reducing costs and computation time. Using the DOE method, the extent of parameter changes on the results can be determined, as well as the range of variable variation, ensuring that the results fluctuate within an ideal range. Furthermore, DOE can construct approximate models that replace the real model.

[0123] Existing research shows that the thickness ratio, carbon fiber layup angle, lattice structure wall thickness, and lattice structure aperture ratio of the steel / carbon fiber hybrid tube all affect the energy absorption of the structure. With increasing steel / carbon fiber thickness ratio, both the energy absorption and mean force of the hybrid tube show a linear increasing trend. Considering that the energy absorption and mean force are relatively large when the thickness ratio is 1, and that it may interact with the lattice structure, the parameter range for the steel / carbon fiber thickness ratio is set to [0.5, 1.5]. When the carbon fiber angle is 30° or higher, both the specific energy absorption (SEA) and mean crushing force (MCF) show a regular increasing trend. Therefore, the carbon fiber layup angle range is set to [30, 90], and the layup method is [0 / HA [° staggered layup. The specific thickness ratio, carbon fiber layup angle, lattice structure wall thickness, and lattice structure aperture ratio of the steel / carbon fiber hybrid tube are shown in Table 4.]

[0124] Table 4 Parameter Range

[0125]

[0126] Compared to traditional DOE methods, Latin hypercube sampling (LHD), as a statistical method, provides a good way to search for global parameters. It is used to generate quasi-random parameter values ​​from multidimensional distributions in programming, achieving a fair distribution among input variables and reducing the number of iterations during computation. LHD is used to generate a distribution of a reasonable set of parameter values ​​from a multidimensional distribution. A Latin hypercube is a square grid including the sample locations if and only if there is only one sample per row and column. The Latin hypercube is a generalization of this concept across any number of dimensions. (See reference...) Figure 13 As shown, each sample is a unique sample in each axis-aligned hyperplane, including itself. For developing first- and second-order responses of regression models, LHD can fit with minimal data sampling.

[0127] When sampling a design space with N variables, the range of each variable is divided into M intervals of equal probability, and then M sample points are placed to meet the requirements of a Latin hypercube. Therefore, all trials have a unique level for each input variable, and the number of sample points M is independent of the number of input variables. However, it is important to note that the number of intervals M for each variable should be consistent, and the sampling method used in the Latin hypercube is sampling without replacement. Furthermore, the number of stratifications performed on the sequence is equal to the number of iterations performed on the selected samples. In this sampling method, independence between variables should be maintained by randomly selecting input parameters from the distribution. Once a variable is selected from an interval, that interval will not be used again. This method avoids unnecessary correlations between parameters.

[0128] To fit a high-quality function using the results of the DOE, it is necessary to calculate the minimum number of runs. Assume the input variables are within the commonly used range. Within this range, most output responses approximate a second-order polynomial, requiring (N+1)(N+2) / 2 runs to fit the second-order polynomial. It is recommended to increase the number of runs by 10% during the calculation process to provide redundancy and make post-processing more reliable. Therefore, the number of calculation runs ( The formula is as follows:

[0129] ,in This represents the number of design variables. Since this application has four design variables, the LHD method requires 17 calculations according to the above formula. Table 5 shows the design matrix generated using the LHD method and the specific values ​​of each variable:

[0130] Table 5 LHD Parameter Settings

[0131]

[0132] Principal causal analysis is mainly used to analyze the influence of each variable on the result. Generally, when studying the influence of one variable, the influence of other variables is ignored. This application considers two parameters, specific energy absorption and peak force, as the standard for examining the degree of influence of variables on the result.

[0133] Linear effects were calculated using a linear regression model, where the normalized input variables ranged from -1 to 1. Input variables Output response The linear effect value is calculated as follows: ,in The coefficients of the regression model Twice that of the parameter; the slope of a linear regression model represents the degree of influence of a parameter on the result. If the slope is 0, it proves that the parameter has no effect on the result; the larger the slope, the greater the influence of the parameter on the result. A positive slope indicates that the change in the parameter is positively correlated with the result, and a negative slope indicates that the change in the parameter is negatively correlated with the result. Therefore, refer to Figure 14 As shown, Figure 14 The effects of several variables considered in this application on the energy absorption and peak force results are illustrated. Among them, for MCF, lattice wall thickness has the greatest impact, followed by lattice aperture ratio (LR), steel / carbon fiber thickness ratio (HR), and carbon fiber angle (HA). Furthermore, the steel / carbon fiber thickness ratio (HR) is negatively correlated with MCF.

[0134] This application employs Moving Least Squares (MLSM) for response surface fitting. MLSM is a data fitting method that constructs a continuous function from a set of unordered sample points by establishing a weighted least squares model. Its characteristic is that the weights associated with the sample points are variable; the weights are functions of the normalized distance from the sample point to the corresponding point in the fitted function, and the weights decrease as the sample point moves further away from its corresponding point. The difference between LSM (Least Squares) and MLSM (Moving Least Squares) is:

[0135] (1) The LSM method can derive the equation, while the MLSM method cannot provide an equation because the weights are not constant; the fitting function of MLSM is not a simple polynomial, but is composed of a coefficient vector. and basis functions Composition, in which and All Functions of coordinates A vector representing random variables;

[0136] (2) The concept of tight support is introduced in MLSM, that is, the fitted function is considered to be Only subject to The influence of points within the influence region is considered, while the influence of points outside the influence region on the fitting results is ignored. The weighting function... Defined within the influence domain, it reflects the contribution of different samples to the fitted function; the magnitude of the contribution value depends on the sampling point and the influence domain. The distance between points.

[0137] Therefore, the main difference between the MLSM method and the traditional LSM method is that the weights in the LSM method remain constant. When using higher-order basis functions, the MLSM method can improve the smoothness of the fitted surface and easily increase fitting accuracy by employing different weight functions.

[0138] In the MLSM method, the approximate fitting function can be written as: ,in, The surrogate model predicts the output value at input variable point x. It is a basis function vector. The transpose symbol indicates It is the inner product of the two; These are undetermined coefficients, depending on... coordinates, and Not a constant, if If the basis functions are constant, the MLSM method becomes the LSM method. The overall accuracy of the LSM method's approximation largely depends on the choice of basis functions, which should be as similar as possible to the actual function. However, the MLSM method effectively reduces approximation error and the dependence of the fitted function on the type of basis functions by establishing a local approximation around each point using weight functions. In the absence of specific knowledge about the behavior of the solution, the basis functions... A common choice is linear and quadratic monomials:

[0139] ;

[0140] The accuracy of MLSM also depends on the weight function w(x). The weight function should prioritize sampling points close to the interpolation point and ignore points outside the influence domain. Therefore, the weight function can be expressed as:

[0141]

[0142] Among them, It is the normalized distance; It is the size of the influence domain, in order to avoid singularities. The value should be large enough to ensure sufficient sampling points within the influence domain. Therefore, in this application, The value is the sum of the (2n+1)th local sampling point and... The weight is set to twice the maximum distance between the interpolation points to ensure a sufficient number of adjacent sampling points within the domain. Therefore, when the interpolation point coincides with a sampling point, the weight is 1.

[0143] In summary, to obtain the accuracy of the approximate model, this application introduces two parameters: relative mean absolute error (RME). RAAE ), root mean square error ( ) and the correlation coefficient of the fit ( The accuracy of the fitted model is measured; the expressions for the two parameters are shown in the following formula:

[0144] , , .

[0145] in and The corresponding predicted value and average value of the response for each sampling point, It is the number of test points. This represents the true value of the response at the sampled point. Typically, when... RAAE The smaller the value or R 2 A higher value indicates a higher accuracy of the surrogate model. The smaller the value, the more accurate the model. Table 6 lists the accuracy of the response surface models fitted by the MLSM method, and lists the values ​​for each. and The accuracy of the fitted models for MCF and SEA shows that the fitted models are highly accurate, and these models can be used as the basis for subsequent optimization calculations. Using the fitted models can significantly reduce computation time and improve the efficiency of optimization calculations. The error analysis of the surrogate models is shown in Table 6.

[0146] Table 6. Proxy Model Error

[0147]

[0148] As can be seen from the table, the surrogate model has high accuracy, proving that the surrogate model constructed in this application can be used for design optimization.

[0149] refer to Figure 15 It can be seen that for MCF, the lattice structure parameters have a greater impact, while the structural parameters of the steel / carbon fiber hybrid tube have a slightly smaller impact, with no significant variation. From Figure 15As shown in (b), there is an interaction between the lattice aperture ratio and the lattice wall thickness on the MCF. As the lattice wall thickness increases, the influence of the lattice aperture ratio on the MCF also gradually increases. These two parameters have a significant impact on the relative density of the lattice structure, resulting in increased lattice structure strength and an increasing MCF. For the SEA, the steel / carbon fiber hybrid tube has a greater impact, showing a trend of first increasing and then decreasing with the increase of the steel / carbon fiber thickness ratio. Compared to the carbon fiber angle, the steel / carbon fiber thickness ratio has a more significant impact, proving that the steel / carbon fiber hybrid outer wall plays a dominant role in the energy absorption process. Before its thickness ratio is below 1, the SEA of the anti-climb device shows a monotonically increasing trend. After it is above 1, the energy absorption ratio of the lattice decreases, while the energy absorption ratio of the metal outer tube increases, leading to an increase in mass and a decreasing trend in SEA. Compared with the effect of steel / carbon fiber hybrid tube alone, the SEA of the filled tube shows a trend of first increasing and then decreasing with the increase of the thickness ratio, which proves that filling the tube with a lattice structure will affect the tube's energy absorption. It is of great significance to study the matching relationship between the two and seek the optimal solution.

[0150] To achieve the requirements of lightweight design and meet the energy absorption needs of the anti-climb device, this application proposes an optimization objective: to obtain the maximum specific energy absorption while achieving an average crushing force that is closest to that of a traditional square cone-shaped energy-absorbing anti-climb device. The specific settings are as follows:

[0151] ,in For specific energy absorption, For lattice structure wall thickness, The aperture ratio of the lattice structure, The steel / carbon fiber thickness ratio, The angle of the carbon fiber layup. The average crushing force; These are reference values ​​for existing cone-shaped energy-absorbing anti-climb devices.

[0152] To obtain the optimal parameter configuration for the HQLS-filled steel / carbon fiber hybrid energy-absorbing anti-climb structure, the aforementioned design variables need to be optimized. This application employs the existing GRSM (Global Response Surface Methodology) algorithm for multi-objective optimization analysis. The GRSM algorithm has global search capabilities and has been widely applied in multi-objective structural optimization. GRSM utilizes response surfaces for computational optimization, which is an effective optimization method. This application can use the fitted response surface for optimization analysis. Its operational principle diagram is shown below. Figure 16 As shown.

[0153] Therefore, based on the boundary conditions set above, the Pareto front of the optimal solution obtained is as follows: Figure 17As shown. It can be observed that the HQLS lattice-filled steel / carbon fiber pyramidal energy-absorbing anti-climb device within the parameter range has a higher specific energy absorption than the traditional structure, meeting the optimization goal of improving specific energy absorption. Considering the need to simultaneously meet... MCF The energy absorption target most similar to traditional energy absorption structures is selected in this paper. MCF The set of parameters with the lowest error among the given parameters is taken as the optimal solution. The specific parameters are: =0.1874 mm, LR =0.4632, HR =0.8905, HA =60.8848°.

[0154] After obtaining the optimal geometric parameters, these parameters were substituted into the finite element model for calculation. The calculation results are compared and presented in Table 7:

[0155] Table 7 Comparison of Optimal Desorption Energy Parameters

[0156]

[0157] Comparing the mechanical parameters of the new anti-climb device with those of the traditional anti-climb device, among which MCF The error between them is 1.24%. EA The error between them is 0.79%. Based on this, SEA Compared to traditional structures, this represents a 35.74% improvement, effectively achieving the goal of lightweight design. (Reference) Figure 18 As shown, Figure 18 The energy absorption and force-displacement curves during the deformation process were compared. The energy absorption trends of both were similar, and the force-displacement curves showed a good match, proving that under these parameter settings, the new anti-climb device can meet the energy absorption design requirements and achieve the lightweight design requirements. SEA This represents a 35.74% improvement over the traditional structure. Finally, a comparison of the results from the finite element model and the response surface optimization reveals that both... SEA The error between them is 1.77%. MCF The error is 1.03%.

[0158] In some embodiments of this application, this application also relates to a steel-carbon fiber hybrid anti-climb energy absorption device for a rail vehicle filled with a dot matrix designed using the above method, comprising:

[0159] Front-end board;

[0160] Back-end board;

[0161] Anti-climb teeth, which are installed on the front panel;

[0162] A square-pyramidal shell, which is connected to the front end plate and the rear end plate respectively;

[0163] The guide rod connects to the rear end plate;

[0164] At least one lattice energy-absorbing structure is filled inside the square pyramidal shell. The lattice energy-absorbing structure is an HQLS lattice energy-absorbing structure, which is formed by a plurality of interconnected cell units arranged in an array along the z-axis. Each cell unit has a hexagonal base and a quadrilateral top surface. The overall structural shape of the cell unit is frustum-shaped. The edges of each cell unit are closely connected to the edges of adjacent cell units to form a close-fitting arrangement structure, thereby improving the energy absorption capacity of the structure by adding plastic hinges.

[0165] refer to Figure 19 As shown in some embodiments of this application, this application also relates to an optimization system for a steel-carbon fiber hybrid anti-climb energy-absorbing structure of a rail vehicle filled with a dot matrix using the above-described method, including the above-described anti-climb energy-absorbing device, and further comprising:

[0166] Simulation construction module 201 is used to establish a finite element model of the anti-climbing energy absorption device for rail vehicles and define the design variables of the lattice energy absorption structure in the finite element model. The design variables include the steel / carbon fiber thickness ratio, carbon fiber layup angle, lattice structure wall thickness and lattice structure aperture ratio.

[0167] The sample generation module 202 is used to generate a preset number of HQLS lattice energy-absorbing structure samples within the set range of the design variables using a full factorial method combined with the Latin hypercube sampling method, and to calculate and generate the response index of each group of samples according to the finite element simulation method. The response index includes specific energy absorption and average crushing force.

[0168] Model building module 203 is used to construct a proxy model of the mechanical properties of the lattice energy-absorbing structure using the moving least squares method;

[0169] The target optimization module 204 is used to perform global optimization of the surrogate model using the global adaptive response surface method and determine the optimal balance solution between structural specific energy absorption and average crushing force based on the Pareto front, thereby generating the optimal design variables; it is also used to substitute the optimal design variables into the finite element model for recalculation and determine whether the error is within a preset range. If yes, the optimal design variables are output; otherwise, samples are regenerated.

[0170] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0171] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for optimizing the anti-climbing energy absorption of a steel-carbon fiber hybrid matrix in a rail vehicle, characterized in that, The method includes an energy-absorbing anti-climb device with a dot matrix energy-absorbing structure, and comprises the following steps: S1: Establish a finite element model of the anti-climbing energy absorption device for rail vehicles and define the design variables of the lattice energy absorption structure in the finite element model. The design variables include the steel / carbon fiber thickness ratio, carbon fiber layup angle, lattice structure wall thickness and lattice structure aperture ratio. S2: Using the full factorial method combined with the Latin hypercube sampling method, a preset number of HQLS lattice energy-absorbing structure samples are generated within the set range of the design variables, and the response index of each sample is calculated according to the finite element simulation method. The response index includes specific energy absorption and average crushing force. S3: A proxy model of the mechanical properties of the lattice energy-absorbing structure is constructed using the moving least squares method; S4: The global adaptive response surface method is used to perform global optimization on the surrogate model and determine the optimal balance solution between structural specific energy absorption and average crushing force based on the Pareto front, thereby generating the optimal design variables; With the primary objective of increasing the structural energy absorption ratio and the secondary objective of bringing the average crushing force close to the reference value of existing square cone energy-absorbing anti-climb devices, the following optimization objectives are established: ， in For specific energy absorption, For lattice structure wall thickness, The aperture ratio of the lattice structure, The steel / carbon fiber thickness ratio, The angle of the carbon fiber layup. The average crushing force; These are reference values ​​for existing cone-shaped energy-absorbing anti-climb devices; S5: Substitute the optimal design variables into the finite element model to recalculate and determine whether the error is within the preset range. If yes, output the optimal design variables; otherwise, return to step S2 to regenerate the samples.

2. The method for optimizing the anti-climbing energy absorption of a steel-carbon fiber hybrid matrix for filling a railway vehicle according to claim 1, characterized in that, In step S1, the finite element model of the anti-climb energy absorption device for rail vehicles includes anti-climb teeth, front end plate, square pyramidal shell, rear end plate, guide rod, and anti-climb energy absorption structure; The anti-climb energy-absorbing structure adopts an HQLS lattice energy-absorbing structure, which is formed by a number of interconnected cell units arranged in an array along the z-axis. Each cell unit has a hexagonal base and a quadrilateral top surface, and the overall structure of the cell unit is frustum-shaped. Furthermore, the edges of each cell unit are closely connected to the edges of adjacent cell units to form a close-fitting arrangement structure, thereby improving the energy absorption capacity of the structure by adding plastic hinges.

3. The method for optimizing the anti-climbing energy absorption of a steel-carbon fiber hybrid matrix for filling a railway vehicle according to claim 2, characterized in that, In step S1, the individual cell unit of the HQLS lattice energy-absorbing structure is divided into two trapezoids, two isosceles triangles, and four ordinary triangles; Based on the volumes of the trapezoidal, isosceles triangle, and ordinary triangle, the relative density of a single cell unit in the HQLS lattice energy-absorbing structure is calculated using the following formula: ， The above Let the volume of the trapezoid be... The volume of a regular triangle, Let the volume of the isosceles triangle be . The height of the frustum of a single cell unit. The side length of a single cell unit.

4. The method for optimizing the anti-climbing energy absorption of a steel-carbon fiber hybrid matrix for filling a railway vehicle according to claim 1 or 2, characterized in that, In step S2, the Latin hypercube sampling method calculates the number of runs. The formula is: in, The number of design variables.

5. The method for optimizing the anti-climbing energy absorption of a steel-carbon fiber hybrid matrix for filling a railway vehicle according to claim 4, characterized in that, In step S3, the approximate fitting function obtained by the moving least squares method is: ,in The surrogate model predicts the output value at input variable point x. For the basis function vector, These are coefficients to be determined; The transpose symbol indicates It is the inner product of the two; Wherein, basis function vector For linear and quadratic monomials: .

6. The method for optimizing the anti-climbing energy absorption of a steel-carbon fiber hybrid matrix for filling a railway vehicle according to claim 5, characterized in that, After constructing a proxy model of the mechanical properties of the lattice energy-absorbing structure in step S3, the method includes: S31: Calculate the relative mean absolute error of the surrogate model: ,in The corresponding predicted value of the response at each sampling point. For the true value of the sample point, The number of test points; S32: Calculate the root mean square error of the surrogate model: ； S33: Calculate the fitting correlation coefficient of the surrogate model: ； S34: Achieve local approximation of the surrogate model by weighted least squares fitting and make the specific energy absorption and average crushing force of the surrogate model meet the preset accuracy requirements.

7. The method for optimizing the anti-climbing energy absorption of a steel-carbon fiber hybrid matrix for filling a railway vehicle according to claim 1, characterized in that, The parameter range for the steel / carbon fiber thickness ratio is set to [0.5, 1.5]; the parameter range for the carbon fiber layup angle is set to [30, 90]; the parameter range for the lattice structure wall thickness is set to [0.18, 0.21]; and the parameter range for the lattice structure aperture ratio is set to [0.45, 0.55].

8. A steel-carbon fiber hybrid anti-climb energy-absorbing device for a rail vehicle filled with a dot matrix, designed using the method described in any one of claims 1-7, characterized in that, include: Front-end board; Back-end board; Anti-climb teeth, which are installed on the front panel; A square-pyramidal shell, which is connected to the front end plate and the rear end plate respectively; The guide rod connects to the rear end plate; At least one lattice energy-absorbing structure is filled inside the square pyramidal shell. The lattice energy-absorbing structure is an HQLS lattice energy-absorbing structure, which is formed by a plurality of interconnected cell units arranged in an array along the z-axis. Each cell unit has a hexagonal base and a quadrilateral top surface. The overall structural shape of the cell unit is frustum-shaped. Furthermore, the edges of each cell unit are closely connected to the edges of adjacent cell units to form a close-fitting arrangement structure, thereby improving the energy absorption capacity of the structure by increasing plastic hinges.

9. A steel-carbon fiber hybrid anti-climb energy absorption system for a rail vehicle filled with a dot matrix using the method of any one of claims 1-7, characterized in that, include: The simulation construction module is used to establish a finite element model of the anti-climbing energy absorption device for rail vehicles and define the design variables of the lattice energy absorption structure in the finite element model. The design variables include the steel / carbon fiber thickness ratio, carbon fiber layup angle, lattice structure wall thickness and lattice structure aperture ratio. The sample generation module is used to generate a preset number of HQLS lattice energy-absorbing structure samples within the set range of the design variables using a full factorial method combined with the Latin hypercube sampling method, and to calculate and generate the response index of each sample group according to the finite element simulation method. The response index includes specific energy absorption and average crushing force. The model building module is used to construct a proxy model of the mechanical properties of the lattice energy-absorbing structure using the moving least squares method. The target optimization module is used to perform global optimization of the surrogate model using the global adaptive response surface method and determine the optimal balance solution between structural specific energy absorption and average crushing force based on the Pareto front, generating optimal design variables; it is also used to substitute the optimal design variables into the finite element model for recalculation and determine whether the error is within a preset range. If yes, the optimal design variables are output; otherwise, samples are regenerated.

Images