Helmet energy absorption structure design method based on damage mechanics

Abstract

The invention discloses a helmet energy absorption structure design method based on injury mechanics, which comprises the following steps: (1) performing head rotary injury mechanics analysis, performing a head injury analogue simulation experiment based on a head injury mechanics model, analyzing the stress distribution of the head in collision and the influence of the rotary acceleration on the brain injury, and determining the head rotary injury mechanics model; generating rotary damage influence evaluation; (2) selecting a helmet energy-absorbing material and evaluating the performance of a multi-layer material of an energy-absorbing layer, testing the energy absorption performance of foam materials with different densities on the basis of rotary damage influence evaluation, and generating an energy-absorbing material performance optimization result; (3) multi-level design and performance optimization of the energy absorption layer; (4) verifying the performance of the energy absorption layer structure; and (5) comprehensive optimization and design adjustment of the energy absorption layer. According to the helmet energy absorption structure design method based on damage mechanics, transmission and absorption of the rotation acceleration of the head are controlled through an optimization design method, and multi-layer and systematized structural design and optimization are conducted on a helmet energy absorption layer.

Description

Technical Field

[0001] The invention relates to a design method for a helmet energy absorption structure based on damage mechanics, and belongs to the technical field of human body safety protection in traffic safety. Background Art

[0002] As an important part of traffic safety protection equipment, the core design goal of helmets is to effectively reduce the damage that may be caused by external impacts to the head. In traffic accidents, especially collisions in high-speed sports such as motorcycles and bicycles, head protection is particularly important. Traditional helmet designs mostly focus on absorbing impact energy through linear energy-absorbing materials (such as polystyrene foam or polyurethane foam), thereby reducing direct damage to the head. However, these traditional design methods mainly focus on mitigating linear impacts and ignore the protection of rotational injuries.

[0003] Research has found that rotational acceleration of the head is one of the most serious factors in traumatic brain injury (TBI), especially in the case of oblique impact, where rotational injury has a more significant impact on the brain than linear impact. Rotational acceleration can cause diffuse axonal injury (DAI), concussion and other serious brain injuries. The energy absorption design of traditional helmets often cannot effectively suppress the transmission of this rotational acceleration, resulting in unsatisfactory protection of the brain.

[0004] Most of the current solutions are still at the material level or single structure improvement, lacking a comprehensive and systematic design method. The existing technology has not deeply explored how to control the transmission and absorption of head rotation acceleration through optimization design methods, and lacks sufficient multi-level and systematic analysis methods for the structural design of the energy absorption layer.

[0005] Head injury standards include HIC, HICrot, HIPmax, GAMBIT, and BrIC. HIC is the head injury standard. When the value is <500, it means that there is no life-threatening danger; when the value is >700, it means that there is a life-threatening danger; HICrot is the rotational head injury standard; HIPmax is the peak power of the head impact; GAMBIT is the head injury threshold; BrIC is the brain injury standard. These data are used as a quantitative evaluation value for evaluating rotational head injuries during the research process. These values ​​are positively correlated with the risk of head injury. The larger the value, the higher the risk of head injury. In addition to HIC, which is a linear injury evaluation standard, the HIPmax, GAMBIT, and BrIC evaluation standards are more inclined to rotational injury evaluation. Summary of the invention

[0006] The main technical problem solved by the present invention is to provide a helmet energy absorption structure design method based on damage mechanics. The helmet energy absorption structure design method based on damage mechanics controls the transmission and absorption of head rotational acceleration through an optimization design method, and performs multi-level and systematic structural design and optimization of the helmet energy absorption layer.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is: to provide a method for designing a helmet energy absorption structure based on damage mechanics, comprising the following steps:

[0008] (1) Mechanical analysis of head rotational injury: Based on the mechanical model of head injury, a head injury simulation experiment is conducted to analyze the force distribution of the head during a collision and the impact of rotational acceleration on brain injury, and generate a rotational injury impact assessment;

[0009] (2) Selection of helmet energy-absorbing materials and performance evaluation of multi-layer materials of the energy-absorbing layer. Based on the evaluation of the impact of rotational injuries, the energy absorption performance of foam materials of different densities is tested, the thickness, flexibility and strength parameters of the foam materials are adjusted, and the performance optimization results of the energy-absorbing materials are generated to provide a basis for the distribution and multi-layer design of the subsequent energy-absorbing layer;

[0010] (3) Design the multi-level structure and performance optimization of the energy absorbing layer based on the performance optimization results of the energy absorbing material;

[0011] (4) Performance verification of the energy absorption layer structure: verify the response performance of the energy absorption layer under various impact modes through collision simulation experiments, and generate simulation verification results of the energy absorption layer performance;

[0012] (5) Comprehensive optimization and design adjustment of the energy absorption layer. Based on the simulation verification results of the multi-layer energy absorption structure and performance of the energy absorption layer obtained in the above steps, data on different material configurations and the effect of rotational acceleration reduction are collected, and these data are input into the nonlinear autoencoder for learning and training. The relationship between these helmet material parameters and head injuries is analyzed in combination with HIC, HICrot, HIPmax, GAMBIT, and BrIC standards. Then, the helmet material parameters are adjusted to optimize the material distribution, thickness, and stiffness of the multi-layer energy absorption structure of the helmet energy absorption layer, and a new helmet energy absorption structure configuration is generated. The performance of the new helmet energy absorption structure is verified again using simulated collision, and the design parameters in the multi-layer energy absorption structure of the energy absorption layer are fine-tuned according to the verification results. Finally, the optimized helmet energy absorption layer structure design scheme is output to ensure that the design meets the performance requirements and that the optimized structure has good energy absorption effect and head protection performance. Learning and training through the nonlinear autoencoder can make the test data closer to reality and make the design of the helmet energy absorption layer structure more reasonable.

[0013] Preferably, when performing the head injury simulation experiment in step (1), the head rotational acceleration, linear acceleration, and impact energy parameters are collected to analyze the force distribution of the head in the collision and the effect of rotational acceleration on brain injury, analyze whether the helmet energy absorption layer can effectively reduce the transmission of rotational acceleration, the performance of the energy absorption layer in mitigating linear acceleration, and whether the energy absorption layer design can evenly disperse the impact force to avoid stress concentration, and identify the key areas of the helmet energy absorption layer design.

[0014] Preferably, the mechanical analysis of rotational injury of the head in step (1) includes physically modeling and calculating the stress conditions of the head under different collision conditions, and evaluating the risk of brain injury caused by the rotational acceleration generated by the head during the collision in combination with the anatomical characteristics of the head.

[0015] Preferably, step (2) also includes evaluating the response of the material under different impact conditions, the response including the compression response of the material, i.e. the deformation behavior of the foam material under different impact forces, reflecting its energy absorption efficiency; the elastic recovery of the material, i.e. the shape recovery characteristics of the foam material after impact, reflecting the performance of the material that can be used multiple times; and the shear response, i.e. the ability of the foam material to withstand shear stress in rotational impact.

[0016] Preferably, the "testing the energy absorption performance of foam materials of different densities" in step (2) is performed by performing impact tests on high-density foam and low-density foam to evaluate the energy absorption characteristics of different materials under various impact conditions, including the flexibility, compressive strength and energy attenuation capacity of the materials, and adjusting and optimizing the material density and thickness configuration based on the experimental results.

[0017] Preferably, the performance optimization described in step (3) includes compression response optimization, elastic recovery optimization and shear response optimization of the energy absorption layer. By combining high-density and low-density foam materials, adjusting the hierarchical distribution of foams of different densities, stiffness transition and thickness change between foams, and based on the effect of different material distributions on the attenuation of rotational acceleration, the energy absorption layer multi-layer configuration optimization results and the cubic energy absorption structure are generated to achieve comprehensive energy absorption performance optimization under multi-directional impact conditions.

[0018] Preferably, step (4) also includes combining head injury standards such as HIC, HICrot, HIPmax, GAMBIT, and BrIC to evaluate the rotational acceleration mitigation effect of the energy absorbing layer, and further adjusting the material distribution and local structure optimization of the energy absorbing structure according to the performance simulation verification results of the energy absorbing layer.

[0019] Preferably, in step (4), the "verification of the response performance of the energy absorbing layer under various impact modes by collision simulation" is based on a simulation tool to simulate the dynamic response of the energy absorbing layer under impacts of different angles and intensities, and to evaluate the control effect of the energy absorbing layer on the rotational acceleration by head injury standards such as HIC, HICrot, HIPmax, GAMBIT, and BrIC, so as to further optimize the material distribution and local performance. The formula of the head injury standard is as follows:

[0020]

[0021] Where a(t) is the linear acceleration of the head, α(t) is the rotational acceleration of the head, and t 1 is the collision start time, t 2 is the collision end time, t 1 t 2 The time interval is 36ms

[0022] a cr =250g, α cr =25000rad / s -2

[0023] ω cr =46.41rad / s, α cr =39774.87rad / s －２ .

[0024] Preferably, the "nonlinear autoencoder learning and training" in step (5) is to extract meaningful feature representations from complex input data through unsupervised learning. The nonlinear autoencoder can effectively process data with high dimensions and highly nonlinear relationships through its encoder and decoder structure, and compress it into a low-dimensional latent space, thereby performing efficient design optimization, wherein the encoder is used to convert high-dimensional input data x∈R d (e.g. design parameters of the energy absorbing layer) are mapped to a low-dimensional latent space z∈R k , where k < d, this mapping is achieved through a neural network, usually using a nonlinear activation function;

[0025] The output of the encoder (latent space representation) can be expressed as:

[0026] z=f θ (x)

[0027] Where: f θ (·) is the mapping function of the encoder, θ is the weight parameter of the neural network, z∈R kIt is a vector in the latent space, representing the low-dimensional features of the input data. For a given input data x, the encoder maps it to the low-dimensional latent space z through a nonlinear activation function;

[0028] The decoder converts the representation z of the latent space back to a high-dimensional space and reconstructs the original input data x. The decoder implements this mapping process through another neural network.

[0029] The output of the decoder can be expressed as:

[0030]

[0031] g φ (·) is the mapping function of the decoder, φ is the weight parameter of the decoder, is the reconstructed data output by the decoder;

[0032] The goal of the autoencoder is to minimize the error between the input data and the reconstructed data, which is usually measured by the mean square error (MSE). It can be expressed as:

[0033]

[0034] Where n is the number of samples, x ｉ is the i-th sample, is the reconstruction result of the i-th sample. The training goal of the autoencoder is to minimize the reconstruction error, that is, by continuously adjusting the network weight parameters θ and φ, the reconstructed data is as close as possible to the original input data. The optimization problem can be expressed as:

[0035]

[0036] When optimizing the energy-absorbing structure, the latent space learned by the autoencoder can be used to optimize the design parameters. For example, certain variables in the latent space (such as material thickness, stiffness, etc.) are adjusted to optimize the performance of the energy-absorbing layer. The goal of the design optimization is to minimize the reconstruction error while maximizing the rotation acceleration mitigation effect. The objective function can be expressed as:

[0037]

[0038] where P(z) is the regularization term on the latent space variable z (for example, optimizing the effect of rotational acceleration mitigation), and λ is the regularization coefficient used to balance the trade-off between error minimization and performance optimization.

[0039] When the optimization results are verified by simulation, the optimized energy-absorbing structure can be evaluated by head injury standards such as HIC, HICrot, HIPmax, GAMBIT, and BrIC. The optimization goal is to adjust the design in the potential space to maximize the rotational acceleration mitigation effect and energy absorption capacity obtained by simulation calculation. The goal of the optimization process is to achieve the following formula expression:

[0040]

[0041] in It is a comprehensive performance evaluation function that combines factors such as design safety, energy absorption effect and rotational acceleration.

[0042] The beneficial effects of the present invention are as follows: 1. The present invention can effectively reduce the transmission of rotational acceleration by designing a multi-level energy absorption structure and optimizing the material distribution of high-density and low-density foams compared with a traditional single-structure helmet. The method achieves energy absorption of multi-directional impacts by adjusting the foam density gradient, flexibility and hierarchical configuration, while reducing the probability of rotational injuries and significantly improving the safety performance of the helmet;

[0043] 2. The present invention adopts dynamic response analysis of composite foam in the design of energy absorption layer, and adjusts material layout and structural configuration through collision simulation, so that it can show stable energy absorption effect under impacts of different angles and intensities. Compared with the helmet designed for a single impact condition in the prior art, the energy absorption layer of the present invention has stronger adaptability and can better cope with complex collision environments;

[0044] 3. The present invention combines head injury standards such as HIC, HICrot, HIPmax, GAMBIT, and BrIC, and comprehensively evaluates the design effect through simulation and testing. Compared with traditional helmets that rely only on a single collision test, the verification method is more scientific and systematic. Through standardized evaluation, it ensures that the energy-absorbing layer effectively suppresses rotational acceleration and meets the head protection needs in complex collision scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work, among which:

[0046] Figure 1 It is a flowchart of the steps of a method for designing a helmet energy absorption structure based on damage mechanics of the present invention;

[0047] Figure 2is a schematic diagram of a helmet structure in an embodiment of a method for designing a helmet energy absorption structure based on damage mechanics of the present invention;

[0048] Figure 3 It is a schematic diagram of a foam material during a head drop test in a method for designing a helmet energy absorption structure based on damage mechanics of the present invention. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] A design method for helmet energy absorption structure based on damage mechanics, please refer to the attached Figure 1 To Attachment Figure 3 , including the following steps:

[0051] Step 1: Conduct a simulated head drop collision experiment to analyze the force pattern of the head;

[0052] The experiment uses a standardized dummy head model (such as the THUMS dummy model) to simulate the stress on the head in a collision, sets the drop height and angle (usually 1.5 meters, 45° or 90°), and uses an accelerometer to measure the rotational acceleration and linear acceleration of the head in real time. During the simulation experiment, a high-speed data acquisition system is used to record the acceleration changes, and the data is analyzed in combination with injury assessment standards (such as HIC, HICrot, HIPmax, GAMBIT, BrIC, etc.) to generate a rotational injury impact assessment, provide a basis for the design of the energy absorption layer, and optimize the protection area and energy absorption performance of the helmet to ensure its effectiveness in actual collisions. The energy absorption layer described in this patent refers to Figure 2 The energy absorbing buffer layer 2 in the.

[0053] Figure 2 It is a schematic diagram of a helmet structure of a helmet energy absorption structure design method based on damage mechanics of the present invention, wherein the helmet comprises a hard shell 1, an energy absorption buffer layer 2, a comfortable liner 3 and a mask 4;

[0054] Step 2: Test the energy absorption performance of foams of different densities and select materials with appropriate density and thickness to ensure effective reduction of head injuries;

[0055] First, select EPS foam materials of different densities (such as EPS40, EPS60, EPS80, etc.), and use standard compression tests and impact tests to determine the energy absorption performance of each material under different collision conditions. In the experiment, by applying impact forces of different intensities and frequencies, the impact energy absorption effect of the foam material during compression and bending is recorded. Based on the test results, the density and thickness configuration of the foam are optimized to ensure that the rotational acceleration can be absorbed to the greatest extent and the linear impact can be mitigated, and finally the best material solution is selected. In addition, combined with the evaluation of rotational acceleration, the layout of the foam material is optimized to improve the overall protective performance of the energy absorption layer. Please refer to the following examples. Figure 3 As shown, specific test experiments were conducted on the performance of the foam material to test its compression response, elastic recovery, and shear response.

[0056] Step 3: Design the composite foam structure and optimize the attenuation effect of rotational acceleration by adjusting the density gradient and foam distribution;

[0057] Taking the cylinder / matrix structure as an example, first select high-density foam (such as EPS120) as the cylindrical structure and low-density foam (such as EPS40 or EPS60) as the matrix material. The cylindrical foam is placed outside the energy-absorbing layer, and the matrix foam is located inside. By adjusting the number and size of the cylinders and the flexibility of the matrix foam, a composite foam structure with gradient density is formed.

[0058] Step 4: Evaluate the performance of the energy-absorbing layer through simulation and head injury standards, and optimize the design to improve safety.

[0059] First, finite element analysis (FEA) is used to simulate different impact types (such as frontal, side, and oblique), and then the performance of the energy absorption layer is evaluated through standards such as HIC, HICrot, HIPmax, GAMBIT, and BrIC to ensure its protective effect on the head under various collision situations. Finally, based on the simulation results and the evaluation of the damage standard, the design of the energy absorption layer is further optimized to ensure the safety of the helmet.

[0060] Step 5: Comprehensive optimization and design adjustment of the energy absorption layer based on nonlinear autoencoder

[0061] Based on the simulation verification results of the cubic energy absorption structure and energy absorption layer performance obtained in the first four steps, data on different material configurations (such as thickness, stiffness, density, etc.) and the effect of rotational acceleration mitigation are collected, and nonlinear autoencoders are used for training to compress high-dimensional design parameters into low-dimensional latent space. By adjusting the design parameters in the latent space, the material distribution, thickness and stiffness of the cubic structure are optimized to achieve the best energy absorption performance and rotational acceleration mitigation effect. New energy absorption structure configurations are generated by rapid optimization in low-dimensional latent space. Further simulation verification is used to evaluate the optimization results in combination with HIC, HICrot, HIPmax, GAMBIT, BrIC and other standards, and the design parameters in the latent space are fine-tuned according to the evaluation feedback. Finally, the optimized energy absorption layer design scheme is output to ensure that the design meets the performance requirements and the optimized structure has good energy absorption effect and head protection performance.

[0062] Table 1 Energy absorption structure performance test results:

[0063]

[0064] Through these evaluation indicators in Table 1, it can be seen that the second foam material structure EPS40 / 1 has a more obvious effect in reducing rotational injuries while maintaining the original linear protection. HIC is a linear injury evaluation standard, and the following four evaluation standards are more inclined to rotational injury evaluation.

[0065] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for designing a helmet energy absorption structure based on damage mechanics, characterized in that: The steps include: (1) Mechanical analysis of head rotational injury: Based on the mechanical model of head injury, a head injury simulation experiment is conducted to analyze the force distribution of the head during a collision and the impact of rotational acceleration on brain injury, and generate a rotational injury impact assessment; (2) Selection of helmet energy-absorbing materials and performance evaluation of multi-layer materials of the energy-absorbing layer. Based on the evaluation of the impact of rotational injuries, the energy absorption performance of foam materials of different densities is tested, and the thickness, flexibility and strength parameters of the foam materials are adjusted to generate the performance optimization results of the energy-absorbing materials. (3) Design the multi-level structure and performance optimization of the energy absorbing layer based on the performance optimization results of the energy absorbing material; (4) Performance verification of the energy absorption layer structure: verify the response performance of the energy absorption layer under various impact modes through collision simulation experiments, and generate simulation verification results of the energy absorption layer performance; (5) Comprehensive optimization and design adjustment of the energy absorption layer. Based on the simulation verification results of the multi-layer energy absorption structure and performance of the energy absorption layer obtained in the above steps, data on different material configurations and the effect of rotational acceleration reduction are collected and input into the nonlinear autoencoder for learning and training. The relationship between these helmet material parameters and head injuries is analyzed in combination with HIC, HICrot, HIPmax, GAMBIT, and BrIC standards. Then, the helmet material parameters are adjusted to optimize the material distribution, thickness, and stiffness of the multi-layer energy absorption structure of the helmet energy absorption layer. A new helmet energy absorption structure configuration is generated. The performance of the new helmet energy absorption structure is verified again using simulated collision. Based on the verification results, the design parameters in the multi-layer energy absorption structure of the energy absorption layer are fine-tuned.

2. The method for designing a helmet energy absorption structure based on damage mechanics according to claim 1, characterized in that: When performing the head injury simulation experiment in step (1), the head rotational acceleration, linear acceleration, and impact energy parameters are collected to analyze the force distribution of the head in the collision and the impact of rotational acceleration on brain injury, analyze whether the helmet energy absorption layer can effectively reduce the transmission of rotational acceleration, the performance of the energy absorption layer in mitigating linear acceleration, and whether the energy absorption layer design can evenly disperse the impact force to avoid stress concentration, and identify the key areas of the helmet energy absorption layer design.

3. The method for designing a helmet energy absorption structure based on damage mechanics according to claim 1, characterized in that: The mechanical analysis of rotational head injury in step (1) includes physical modeling and calculation of the stress conditions of the head under different collision conditions, and combining the anatomical characteristics of the head to evaluate the risk of brain injury caused by the rotational acceleration generated by the head during the collision.

4. The method for designing a helmet energy absorption structure based on damage mechanics according to claim 2, characterized in that: Step (2) also includes evaluating the response of the material under different impact conditions, the response including the compression response of the material, i.e. the deformation behavior of the foam material under different impact forces, reflecting its energy absorption efficiency; the elastic recovery of the material, i.e. the shape recovery characteristics of the foam material after impact, reflecting the performance of the material that can be used multiple times; and the shear response, i.e. the ability of the foam material to withstand shear stress in rotational impact.

5. The method for designing a helmet energy absorption structure based on damage mechanics according to claim 4, characterized in that: The "testing the energy absorption performance of foam materials of different densities" described in step (2) is to conduct impact tests on high-density foam and low-density foam to evaluate the energy absorption characteristics of different materials under various impact conditions, including the flexibility, compressive strength and energy attenuation capacity of the materials, and adjust and optimize the material density and thickness configuration based on the experimental results.

6. The method for designing a helmet energy absorption structure based on damage mechanics according to claim 1, characterized in that: The performance optimization described in step (3) includes compression response optimization, elastic recovery optimization and shear response optimization of the energy absorbing layer. By combining high-density and low-density foam materials, adjusting the hierarchical distribution of foams of different densities, the stiffness transition and thickness change between foams, and based on the effect of different material distributions on the attenuation of rotational acceleration, the energy absorbing layer multi-layer configuration optimization results and the cubic energy absorbing structure are generated.

7. The method for designing a helmet energy absorption structure based on damage mechanics according to claim 1, characterized in that: Step (4) also includes combining head injury standards such as HIC, HICrot, HIPmax, GAMBIT, and BrIC to evaluate the rotational acceleration mitigation effect of the energy absorbing layer, and further adjusting the material distribution and local structure optimization of the energy absorbing structure based on the performance simulation verification results of the energy absorbing layer.

8. The method for designing a helmet energy absorption structure based on damage mechanics according to claim 1, characterized in that: In step (4), the "verification of the response performance of the energy absorbing layer under various impact modes by collision simulation" is based on simulation tools to simulate the dynamic response of the energy absorbing layer under impacts of different angles and intensities, and to evaluate the control effect of the energy absorbing layer on rotational acceleration by head injury standards such as HIC, HICrot, HIPmax, GAMBIT, and BrIC, so as to further optimize the material distribution and local performance. The formula of the head injury standard is as follows: Where a(t) is the linear acceleration of the head, α(t) is the rotational acceleration of the head, t1 is the start time of the collision, t2 is the end time of the collision, and the time interval t1t2 is 36ms a cr =250g,a cr =25000rad / s -2 oh cr =46.41rad / s, a cr =39774.87rad / s -2 。 9. The method for designing a helmet energy absorption structure based on damage mechanics according to claim 1, characterized in that: The "nonlinear autoencoder learning and training" described in step (5) extracts meaningful feature representations from input data through unsupervised learning. The nonlinear autoencoder processes data with high-dimensional and highly nonlinear relationships through its encoder and decoder structures and compresses it into a low-dimensional latent space. The encoder is used to map the high-dimensional input data x ∈ R d to the low-dimensional latent space z ∈ R k , where k < d, and this mapping is achieved through a neural network using a nonlinear activation function; The output of the encoder is represented as: z=f θ (x) Where: f θ (·) is the mapping function of the encoder, θ is the weight parameter of the neural network, z∈R k It is a vector in the latent space, representing the low-dimensional features of the input data. For a given input data x, the encoder maps it to the low-dimensional latent space z through a nonlinear activation function; The decoder converts the representation z of the latent space back to a high-dimensional space and reconstructs the original input data x. The decoder implements this mapping process through another neural network. The output of the decoder is represented as: g φ (·) is the mapping function of the decoder, φ is the weight parameter of the decoder, is the reconstructed data output by the decoder; The goal of the autoencoder is to minimize the error between the input data and the reconstructed data, which is measured by the mean square error. It is expressed as: Where n is the number of samples, x i is the i-th sample, is the reconstruction result of the i-th sample. The training goal of the autoencoder is to minimize the reconstruction error, that is, by continuously adjusting the network weight parameters θ and φ, the reconstructed data is close to the original input data. The optimization problem is expressed as: When optimizing the energy absorption structure, the latent space learned by the autoencoder is used to optimize the design parameters. The goal of the design optimization is to minimize the reconstruction error while maximizing the rotation acceleration mitigation effect. The objective function is expressed as: Where P(z) is the regularization term for the latent space variable z, and λ is the regularization coefficient, which is used to balance the trade-off between error minimization and performance optimization. When verifying the optimization results through simulation, the optimized energy-absorbing structure is evaluated for performance through head injury standards such as HIC, HICrot, HIPmax, GAMBIT, and BrIC. The optimization goal is to adjust the design in the potential space to maximize the rotational acceleration mitigation effect and energy absorption capacity obtained through simulation calculation. The goal of the optimization process is to achieve the following expression:

Images