Method for determining influence of head kinematics boundary condition on head injury

Through simulation and monomeric linear regression analysis of multiple traffic accident cases, the impact of head kinematic boundary conditions on head injury was determined, the problem of low accuracy in the existing technology was solved, and the head risk prediction was improved The authenticity and diversity provide important data for VRU safety assessment and damage prediction.

CN118917127AActive Publication Date: 2024-11-08CHANGSHA UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202410842029.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-10-31
Filing Date
2024-06-27
Publication Date
2024-11-08
Estimated Expiration
2044-06-27

AI Technical Summary

Technical Problem

When determining the correlation between head kinematic boundary conditions and head injury indicators in car collision accidents, the prior art is not very accurate and cannot fully reflect the head collision response in real accidents, resulting in the real prediction of head risk. Insufficient sex and accuracy.

Method used

By obtaining accident information from multiple traffic accident cases, establishing multi-rigid body mannequin and vehicle models, conducting simulation of traffic accident process, outputting head injury indicators and extracting head kinematic boundary conditions, and using one-element linear regression analysis method to calculate each The correlation between head kinematic boundary conditions and damage indicators determines the degree of influence.

Benefits of technology

Improve the accuracy and diversity of the correlation between head kinematic boundary conditions and head injury indicators, and provide more realistic and reliable head risk prediction data, providing important for the optimization of VRU safety assessment regulations and the establishment of damage prediction equations. parameter.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917127A_ABST
    Figure CN118917127A_ABST
Patent Text Reader

Abstract

The invention discloses a method for determining the influence of a head kinematics boundary condition on head injury, and the method comprises the following steps: obtaining accident information of a plurality of traffic accident cases meeting a preset condition, and the plurality of traffic accident cases relate to the collision between a vulnerable road user VRU and a vehicle; carrying out analogue simulation on a traffic accident process according to accident information of the traffic accident case, outputting a head injury index and extracting a head kinematics boundary condition in collision; carrying out correlation analysis on the extracted head kinematics boundary conditions and the output head injury indexes which are subjected to analogue simulation on a plurality of traffic accident cases by adopting a unary linear regression analysis method, and calculating the correlation between each head kinematics boundary condition and each head injury index; and determining the influence degree of each head kinematics boundary condition on the head injury according to a calculation result. According to the invention, the accuracy and diversity of the correlation between the determined head kinematics boundary condition and the head injury index can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of traffic safety technology, and in particular to a method for determining the correlation between head kinematic boundary conditions and head injury indicators in a car collision accident. Background Art

[0002] In traffic accidents involving Vulnerable Road Users (VRUs) and vehicles, head injuries have always been the main cause of VRU deaths and injuries. However, the head impactor method used in the current pedestrian protection evaluation procedures only considers a single collision boundary and cannot fully reflect the entire head impact response process in real accidents. In addition, the adopted head injury evaluation criterion HIC only considers the linear movement of the head, which has its limitations in the risk assessment of head injuries. At the same time, evaluation procedures for the protection of other VRUs are urgently needed to be developed.

[0003] The existing determination of the correlation between head kinematic boundary conditions and head injury indicators is based on early animal and cadaver experiment data, or sports accident data. Compared with the collision load in traffic accidents, the head impact magnitude in such cases is relatively much smaller, which leads to a low frequency of head injuries in accidents, making it difficult to obtain accident data with specific injury reports, greatly affecting the authenticity of head risk prediction. In addition, the reconstruction method used in the existing determination method is relatively simple, and the input impact conditions are slightly single and ideal, which cannot fully reflect the actual load scenario of head collision, resulting in the low accuracy of the correlation between the determined head kinematic boundary conditions and head injury indicators. Summary of the invention

[0004] The main purpose of this application is to provide a method for determining the influence of head kinematic boundary conditions on head injuries, aiming to improve the accuracy and diversity of the correlation between the determined head kinematic boundary conditions and head injury indicators.

[0005] To achieve the above object, an embodiment of the present application provides a method for determining the influence of head kinematic boundary conditions on head injury, the method comprising the following steps:

[0006] S1. Obtaining accident information of multiple traffic accident cases that meet preset conditions, wherein the multiple traffic accident cases all involve collisions between vulnerable road users (VRUs) and vehicles;

[0007] S2. For each traffic accident case, simulate the traffic accident process according to the accident information of the traffic accident case, output the head injury index and extract the head kinematic boundary conditions in the collision;

[0008] Wherein, step S2 comprises:

[0009] S21. Establish a multi-rigid body human model and a multi-rigid body vehicle model according to the accident information of the traffic accident case;

[0010] S22, using multi-rigid body dynamics analysis software to simulate and calculate the traffic accident process according to the accident information, the multi-rigid body pedestrian model and the multi-rigid body vehicle model to obtain a virtually reconstructed traffic accident scene;

[0011] S23, establishing a multi-rigid body-finite element pedestrian coupling model according to the human body finite element model and the multi-rigid body pedestrian model;

[0012] S24, after replacing the multi-rigid body pedestrian model with the multi-rigid body-finite element pedestrian coupling model in the virtually reconstructed traffic accident scene, perform traffic accident process simulation calculation again, and output head injury indicators, wherein the head injury indicators include head injury criterion HIC, brain injury threshold generalized acceleration model GAMBIT, brain injury criterion BrIC, rotational injury criterion RIC, head impact power HIP, maximum principal strain MPS and cumulative strain damage measurement CSDM; and

[0013] S25, intercepting the complete collision process information between the head of the vulnerable road user VRU and the vehicle from the moment of contact to the moment of separation of the head of the VRU and the vehicle during the simulated traffic accident, and extracting the head kinematic boundary conditions in the collision from the complete collision process information, wherein the head kinematic boundary conditions include the linear velocity peak value PLV, the rotational velocity peak value PRV, the linear acceleration peak value PLA, the rotational acceleration peak value PRA and the collision duration ID of the center of mass during the collision between the head of the VRU and the vehicle;

[0014] S3. After completing the simulation of the multiple traffic accident cases, a univariate linear regression analysis method is used to perform a correlation analysis on the head kinematic boundary conditions extracted from the simulation of the multiple traffic accident cases and the output head injury indicators, and the correlation between each head kinematic boundary condition and each head injury indicator is calculated, and the degree of influence of each head kinematic boundary condition on the head injury is determined according to the calculation results.

[0015] Preferably, the prediction model of the univariate linear regression analysis method is:

[0016] Y t =ax t +b

[0017] In the above formula, x t is the head kinematic boundary condition; Y t is the head injury index; a, b represent the parameters of the univariate linear regression equation, and a, b are obtained by the following two equations:

[0018]

[0019] In the above two formulas, n represents the number of data in the data set, x i is the value of the i-th head kinematic boundary condition in the data set, Y i is the value of the i-th head injury indicator in the data set.

[0020] Preferably, the preset conditions include:

[0021] The accident participants in the accident information involve VRU;

[0022] Accident information is sufficient to support virtual reconstruction of traffic accident scenes;

[0023] The VRU head of the accident information collided with the vehicle body.

[0024] Preferably, the multiple traffic accident cases include pedestrian-vehicle collision accident cases and two-wheeled vehicle and automobile collision accident cases. For pedestrian-vehicle collision accident cases, the multi-rigid body vehicle model includes a multi-rigid body automobile vehicle model; for two-wheeled vehicle and automobile collision accident cases, the multi-rigid body vehicle model includes a multi-rigid body automobile vehicle model and a multi-rigid body two-wheeled vehicle model.

[0025] Preferably, before performing the simulation of the traffic accident process, the method further includes:

[0026] Removing the front windshield structure of the multi-rigid body automobile vehicle model;

[0027] The coupling module in the multi-rigid body dynamics analysis software is used to connect the finite element front windshield model with the rigid bodies of the corresponding parts of the multi-rigid body vehicle model to establish a multi-rigid body-finite element vehicle coupling model;

[0028] In the virtually reconstructed traffic accident scene, the multi-rigid body vehicle model is replaced by a multi-rigid body-finite element vehicle coupling model.

[0029] Preferably, the establishing of a multi-rigid body-finite element pedestrian coupling model based on the human body finite element model and the multi-rigid body pedestrian model comprises:

[0030] Removing the head and neck structure of the multi-rigid-body pedestrian model;

[0031] The head and neck of the human finite element model are extracted, and the muscles and end nodes of the cervical vertebrae in the neck of the human finite element model, which are originally connected to the torso, are connected to the rigid bodies at the corresponding parts of the upper torso of the multi-rigid body pedestrian model after the head and neck are removed, using the coupling module in the multi-rigid body dynamics analysis software, to obtain a multi-rigid body-finite element pedestrian coupling model.

[0032] Preferably, the multi-rigid body human body model is obtained by scaling and adjusting the TNO 50 percentile adult male and 5 percentile female multi-rigid body pedestrian models using the multi-rigid body dynamics analysis software MADYMO; the human body finite element model uses the virtual human body model THUMS 50 percentile adult male finite element model.

[0033] Preferably, the calculating the correlation between each head kinematic boundary condition and each head injury index, and determining the influence of each head kinematic boundary condition on the head injury according to the calculation result, includes:

[0034] Calculate the correlation coefficient R between each head kinematic boundary condition and each head injury index to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the influence of various head kinematic boundary conditions on head injury.

[0035] Preferably, before performing correlation analysis on the extracted head kinematic boundary conditions and the output head injury indexes by simulating the plurality of traffic accident cases using a univariate linear regression analysis method, the method further includes:

[0036] The linear acceleration peak values ​​in the extracted head kinematic boundary conditions are divided into three magnitude groups: low, medium and high according to the peak value.

[0037] The peak values ​​of rotational acceleration in the extracted head kinematic boundary conditions are divided into three magnitude groups: low, medium and high according to the peak values; and

[0038] The peak value of rotation speed in the extracted head kinematic boundary conditions is divided into two magnitude groups: low and high.

[0039] The univariate linear regression analysis method is used to simulate the multiple traffic accident cases to perform correlation analysis on the extracted head kinematic boundary conditions and the output head injury indicators, including:

[0040] The correlations between the low, medium and high magnitude groups of the peak linear acceleration and each head injury index were calculated;

[0041] The correlations between the low, medium and high magnitude groups of the peak rotational acceleration and each head injury index were calculated;

[0042] The correlations between the low and high magnitude groups of peak rotational velocity and various head injury indicators were calculated.

[0043] Preferably, the calculation of the correlation between the three magnitude groups of low, medium and high of the peak linear acceleration and each head injury index includes:

[0044] The correlation coefficients R between the low, medium and high levels of the peak linear acceleration and each head injury index were calculated to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the impact of low, medium and high peak linear acceleration levels on head injury;

[0045] The correlations between the three magnitude groups of low, medium and high calculated rotational acceleration peak values ​​and various head injury indicators include:

[0046] The correlation coefficients R between the low, medium and high magnitude groups of the peak value of rotational acceleration and each head injury index were calculated to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the impact of low, medium and high peak linear acceleration levels on head injury;

[0047] The correlation between the low and high magnitude groups of the calculated rotational velocity peak and each head injury index includes:

[0048] The correlation coefficients R between the low, medium and high levels of the peak rotation speed and each head injury index were calculated to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the impact of low and high peak rotation speed groups on head injuries.

[0049] The method for determining the influence of head kinematic boundary conditions on head injury proposed in the present application obtains accident information of multiple traffic accident cases that meet preset conditions, wherein the multiple traffic accident cases all involve collisions between vulnerable road users (VRU) and vehicles, that is, real traffic accident data of collisions between vulnerable road users (VRU) and vehicles are used as samples, and the collision boundaries of human head caused by traffic accidents are more complex and diverse, and the data source is authentic and reliable, which greatly improves the authenticity of head risk prediction; for each traffic accident case, a traffic accident process is simulated according to the accident information of the traffic accident case, and head injury indicators are output and head kinematic boundary conditions in the collision are extracted. First, according to the accident information, multi-rigid body pedestrian model and multi-rigid body vehicle model, multi-rigid body dynamics analysis software is used to simulate and calculate the traffic accident process to obtain a virtually reconstructed traffic accident scene, rather than direct accident information, multi-rigid body-finite element pedestrian coupling model and multi-rigid body vehicle model, which can greatly reduce the amount of calculation of the virtually reconstructed traffic accident scene; in the virtually reconstructed traffic accident scene, the multi-rigid body pedestrian model is replaced with the multi-rigid body-finite element After the pedestrian coupling model is established, the traffic accident process is simulated and calculated again to make the output head injury index and the extracted head kinematic boundary conditions more accurate; and the univariate linear regression analysis method is used to simulate the multiple traffic accident cases, and the extracted head kinematic boundary conditions and the output head injury index are analyzed for correlation, and the correlation between each head kinematic boundary condition and each head injury index is calculated, and the influence of each head kinematic boundary condition on the head injury is determined according to the calculation results; in this way, the influence of multiple head kinematic boundary conditions on multiple head injuries can be obtained at the same time, and the accuracy of the correlation between the determined head kinematic boundary conditions and the head injury index is high and diverse, which can provide important data and method references for the optimization and updating of VRU safety assessment regulations / procedures, and provide important parameters for the establishment of injury prediction equations. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A schematic flow chart of a method for determining the influence of head kinematic boundary conditions on head injury provided in one embodiment of the present application;

[0051] Figure 2 for Figure 1 A detailed flow chart of step S2 in the method shown;

[0052] Figure 3 It is a schematic diagram of the linear regression results between the peak value of collision linear acceleration and various head injury indicators;

[0053] Figure 4This is a schematic diagram of the linear regression results between the peak value of collision rotational acceleration and various head injury indicators;

[0054] Figure 5 It is a schematic diagram of the linear regression results between the peak value of collision linear velocity and various head injury indicators;

[0055] Figure 6 This is a schematic diagram of the linear regression results between the peak collision rotation velocity and various head injury indicators;

[0056] Figure 7 This is a schematic diagram of the linear regression results between collision duration and various head injury indicators;

[0057] Figure 8 A schematic diagram showing the overall correlation coefficient comparison of head injury indicators based on head kinematics and brain tissue deformation;

[0058] Fig. 9 It is the distribution diagram of the peak value of linear acceleration after grouping;

[0059] Fig.10 is the distribution diagram of the peak value of rotational acceleration after grouping;

[0060] Fig.11 is the distribution diagram of the peak rotation speed after grouping;

[0061] Fig.12 This is a schematic diagram of the linear regression results between the peak value of linear acceleration and various head injury indicators in the low-level group;

[0062] Fig.13 This is a schematic diagram of the linear regression results between the peak linear acceleration and various head injury indicators in the middleweight group;

[0063] Fig.14 This is a schematic diagram of the linear regression results between the peak value of linear acceleration in the high-intensity group and various head injury indices;

[0064] Fig.15 This is a schematic diagram of the linear regression results between the peak value of rotational acceleration and various head injury indicators in the low-intensity group;

[0065] Fig.16 This is a schematic diagram of the linear regression results between the peak value of rotational acceleration and various head injury indicators in the middleweight group;

[0066] Fig.17 This is a schematic diagram of the linear regression results between the peak value of rotational acceleration and various head injury indicators in the high-intensity group;

[0067] Fig.18 This is a schematic diagram of the linear regression results between the peak rotational velocity and various head injury indicators in the low-intensity group;

[0068] Fig.19Schematic diagram of the linear regression results of the peak rotational velocity and various head injury indicators in the high-intensity group.

[0069] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0070] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0071] See also Figure 1 An embodiment of the present application provides a method for determining the influence of head kinematic boundary conditions on head injury, the method comprising the following steps:

[0072] S1. Obtaining accident information of multiple traffic accident cases that meet preset conditions, wherein the multiple traffic accident cases all involve collisions between vulnerable road users (VRUs) and vehicles;

[0073] S2. For each traffic accident case, simulate the traffic accident process according to the accident information of the traffic accident case, output the head injury index and extract the head kinematic boundary conditions in the collision;

[0074] S3. After completing the simulation of the multiple traffic accident cases, a univariate linear regression analysis method is used to perform a correlation analysis on the head kinematic boundary conditions extracted from the simulation of the multiple traffic accident cases and the output head injury indicators, and the correlation between each head kinematic boundary condition and each head injury indicator is calculated, and the degree of influence of each head kinematic boundary condition on the head injury is determined according to the calculation results.

[0075] In step S1, the preset conditions include:

[0076] The accident participants in the accident information involve VRU;

[0077] Accident information is sufficient to support virtual reconstruction of traffic accident scenes;

[0078] The VRU head of the accident information collided with the vehicle body.

[0079] The accident information of the multiple traffic accident cases can be obtained by screening in the traffic accident database according to the preset conditions, that is, using real accident information for virtual reconstruction of traffic accident scenes. As an example, in an application example of the present embodiment, the accident information of multiple traffic accident cases comes from the In-depth Investigation of Vehicle Accident in Changsha (IVAC) and the traffic accident database of China Automotive Engineering Research Institute. Based on the preset conditions, a total of 50 typical traffic accident cases in which the head of a VRU collided with a vehicle were screened out, of which 25 were human-vehicle collision accidents and 25 were two-wheeled vehicle-car collision accidents. The basic accident information obtained based on the accident information of the 50 traffic accident cases is shown in Table 1 below.

[0080] Table 1 Basic accident information table

[0081]

[0082]

[0083] The accident information of a traffic accident case includes pedestrian information and vehicle information. The pedestrian information may include gender, age, height, weight, speed at the time of collision, etc. The vehicle information may include speed at the time of collision, brand and model, curb weight, vehicle size, speed at the time of collision, etc.

[0084] The number and range of accident information for traffic accident cases are not limited to this. The number of cases may be more than 50 or less than 50, and the range of cases may be selected from other traffic accident databases.

[0085] See also Figure 2 , step S2 comprises:

[0086] S21. Establish a multi-rigid body human model and a multi-rigid body vehicle model according to the accident information of the traffic accident case;

[0087] S22, using multi-rigid body dynamics analysis software to simulate and calculate the traffic accident process according to the accident information, the multi-rigid body pedestrian model and the multi-rigid body vehicle model to obtain a virtually reconstructed traffic accident scene;

[0088] S23, establishing a multi-rigid body-finite element pedestrian coupling model according to the human body finite element model and the multi-rigid body pedestrian model;

[0089] S24. After replacing the multi-rigid body pedestrian model with the multi-rigid body-finite element pedestrian coupling model in the virtually reconstructed traffic accident scene, the traffic accident process simulation calculation is performed again to output head injury indicators, wherein the head injury indicators include head injury criterion (HIC), generalized acceleration model for brain injury threshold (GAMBIT), brain injury criterion (BrIC), rotational injury criterion (RIC), head impact power (HIP), maximum principal strain (MPS) and cumulative strain damage measure (CSDM); and

[0090] S25. Capture the complete collision process information between the head of the vulnerable road user VRU and the vehicle from the moment of contact to the moment of separation of the head of the VRU and the vehicle during the simulated traffic accident, and extract the head kinematic boundary conditions in the collision from the complete collision process information, wherein the head kinematic boundary conditions include the peak linear velocity (Peak Liner Velocity, PLV), the peak rotational velocity (Peak Rotational Velocity, PRV), the peak linear acceleration (Peak Liner Acceleration, PLA), the peak rotational acceleration (Peak Rotational Acceleration, PRA) and the collision duration (Impact Duration, ID) of the center of mass during the collision between the VRU head and the vehicle.

[0091] In step S21, the multi-rigid body human body model is obtained by scaling and adjusting the 50th percentile adult male and 5th percentile female multi-rigid body pedestrian models of the Netherlands Organization for Applied Scientific Research (TNO) using the multi-rigid body dynamics analysis software MADYMO.

[0092] The multiple traffic accident cases include pedestrian-vehicle collision accident cases and two-wheeler-car collision accident cases. For pedestrian-vehicle collision accident cases, the multi-rigid body vehicle model includes a multi-rigid body car model; for two-wheeler-car collision accident cases, the multi-rigid body vehicle model includes a multi-rigid body car model and a multi-rigid body two-wheeler model. Both the multi-rigid body car model and the multi-rigid body two-wheeler model are multi-body modeled according to the specific vehicle model in the accident, and stiffness characteristics are given to their main structures.

[0093] In step S22, the multi-rigid body dynamics analysis software may be TNO's multi-rigid body dynamics analysis software MADYMO (MAthematical DYnamic MOdel).

[0094] In step S23, the multi-rigid body-finite element pedestrian coupling model is established according to the human body finite element model and the multi-rigid body pedestrian model, including:

[0095] Removing the head and neck structure of the multi-rigid-body pedestrian model;

[0096] The head and neck of the human finite element model are extracted, and the muscles and end nodes of the cervical vertebrae originally connected to the torso of the neck of the human finite element model are connected to the rigid bodies at the corresponding parts of the upper torso of the multi-rigid body pedestrian model after the head and neck are removed (from left to right, the left clavicle, upper torso and right clavicle) using the coupling module in the multi-rigid body dynamics analysis software to obtain a multi-rigid body-finite element pedestrian coupling model.

[0097] The human finite element model adopts the 50th percentile adult male finite element model of the virtual human model THUMS (Total Human Model for Safety) established by Toyota Central Research and Development Center. The head and neck of the human finite element model include the head and cranial brain, neck, thoracic vertebrae T1-T3, etc. The human finite element model includes key anatomical structures of the human cranial brain, such as scalp, skull, meninges, cerebrospinal fluid, brain, cerebellum, brainstem, sickle body, etc. The multi-rigid pedestrian model is composed of hinges with human joint characteristics and ellipsoids representing the geometric dimensions, mass and moment of inertia of various parts of the human body.

[0098] In step S24, the head injury indicators include the head injury criterion HIC, the generalized acceleration model of brain injury threshold GAMBIT, the brain injury criterion BrIC, the rotational injury criterion RIC, the head impact power HIP, the maximum principal strain MPS and the cumulative strain damage measurement CSDM, among which HIC, GAMBIT, BrIC, RIC and HIP are head injury indicators based on head kinematics, and MPS and CSDM are head injury indicators based on brain tissue strain.

[0099] The head injury criterion HIC is the most widely used head injury indicator, but its defect is that it only calculates the linear acceleration of the head without considering the influence of rotational motion; the rotational injury criterion RIC differs from HIC in that it only uses rotational acceleration; the generalized acceleration model of brain injury threshold GAMBIT and the head impact power HIP take into account both the linear acceleration and rotational acceleration of the head; the BrIC calculation process is based on the head rotation speed and the peak value of rotational acceleration; MPS is output through the human finite element model, and CSDM refers to the volume ratio of the area in the brain tissue where the MPS exceeds a preset threshold to the entire brain tissue. In this embodiment, the thresholds are set to 0.15 and 0.25 respectively.

[0100] In step S25, the head kinematic boundary conditions include the linear velocity peak value PLV, the rotational velocity peak value PRV, the linear acceleration peak value PLA, the rotational acceleration peak value PRA and the collision duration ID of the center of mass of the VRU head during the collision with the vehicle, wherein the linear acceleration and the rotational acceleration are important kinematic parameters for the study of craniocerebral injury, and the acceleration peak value and waveform have a significant impact on head injury. It is particularly important to obtain the acceleration response of the injury process from a real head collision for the study of injury mechanism and protection. The linear velocity and the rotational velocity are also important head kinematic parameters. The rotational velocity is highly correlated with brain injury. The linear velocity has always been the most direct input condition, affecting the linear and rotational motion of the head and the injury response. The collision duration determines the overall kinematic loading during the head collision.

[0101] In step S3, the prediction model of the univariate linear regression analysis method is:

[0102] Y t =ax t +b

[0103] In the above formula, x t is the head kinematic boundary condition; Y t is the head injury index; a, b represent the parameters of the univariate linear regression equation, and a, b are obtained by the following two equations:

[0104]

[0105] In the above two formulas, n represents the number of data in the data set, x i is the value of the i-th head kinematic boundary condition in the data set, Y i is the value of the i-th head injury indicator in the data set.

[0106] Calculating the correlation between each head kinematic boundary condition and each head injury index, and determining the influence of each head kinematic boundary condition on the head injury according to the calculation result includes:

[0107] Calculate the correlation coefficient R between each head kinematic boundary condition and each head injury index to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the influence of various head kinematic boundary conditions on head injury.

[0108] The correlation coefficient R is a coefficient used to measure the degree of correlation between two variables. The value range of R is between [-1,1]. The closer the absolute value of the correlation coefficient is to 1, the stronger the correlation between the variables. If the correlation coefficient is greater than 0, it means that there is a positive correlation between the two variables; if the correlation coefficient is less than 0, it means that there is a negative correlation between the two variables.

[0109] In the application example of this embodiment, when the univariate linear regression analysis method is used to simulate 50 traffic accident cases, the extracted head kinematic boundary conditions and the output head injury indicators are analyzed for correlation. The linear regression results between the center of mass linear velocity peak value PLV, rotational velocity peak value PRV, linear acceleration peak value PLA, rotational acceleration peak value PRA, and collision duration ID during the collision between the VRU head and the vehicle and various head injury indicators during the collision are as follows: Figures 3 to 7 The linear regression results of head injury indicators based on head kinematics and brain tissue deformation are shown in Tables 2 and 3. The overall correlation coefficient comparison is shown in Tables 2 and 3. Figure 8 As shown. Among them, p is a measure of whether the two variables are related, and the coefficient of determination R 2 = 0.2 is used as the correlation observable threshold of the linear regression results. Figure 8 Null means that the linear regression result is not significant (i.e. p>0.01).

[0110] Table 2 Linear regression results of head injury indicators based on head kinematics

[0111]

[0112]

[0113] Table 3 Linear regression results of head injury indicators based on brain tissue strain

[0114]

[0115] The above linear regression results show that the correlation between head kinematic boundary conditions and head injury indicators based on head kinematics is generally higher than the correlation between head kinematic boundary conditions and head injury indicators based on brain tissue strain, while the correlation between collision duration ID and various head injury indicators is low. In the regression analysis of head injury indicators based on head kinematics, except for collision duration ID, the parameter peaks of other head kinematic boundary conditions have good correlation with most head injury indicators. In general, the correlation between acceleration-related parameters PLA and PRA and various head injury indicators is stronger than that between velocity-related parameters PLV and PRV; the correlation between linear motion and HIC and HIP is stronger than that between rotational motion and HIC and HIP, while the correlation between rotational motion and GAMBIT, BrIC and RIC is more significant than that between linear motion and GAMBIT, BrIC and RIC.

[0116] More specifically, PRA and BrIC (R 2 =0.984, p<0.01), PRA and GAMBIT (R 2 =0.975, p<0.01), PLA and HIP (R 2 =0.845, p<0.01), PLA and HIC (R 2 =0.765, p<0.01) had the most significant correlation.

[0117] In the linear regression results of head injury indicators based on brain tissue strain, although the correlation between the peak values ​​of head kinematic boundary conditions and head injury indicators was not significant, the correlation between PRV and head injury indicators was higher than that of all other parameters. 0.25 , CSDM 0.15 , MPS were 0.628, 0.398, and 0.381, respectively (p<0.01).

[0118] In the above-mentioned method for determining the influence of head kinematic boundary conditions on head injury of the present application, by obtaining accident information of multiple traffic accident cases that meet the preset conditions, the multiple traffic accident cases all involve collisions between vulnerable road users VRU and vehicles, that is, using real traffic accident data of collisions between vulnerable road users (VRU) and vehicles as samples, the collision boundaries of human head caused by traffic accidents are more complex and diverse, and the data source is authentic and reliable, which greatly improves the authenticity of head risk prediction; for each traffic accident case, according to the accident information of the traffic accident case, the traffic accident process is simulated, the head injury index is output and the head kinematic boundary conditions in the collision are extracted, first, according to the accident information, multi-rigid body pedestrian model and multi-rigid body vehicle model, multi-rigid body dynamics analysis software is used to simulate and calculate the traffic accident process to obtain a virtually reconstructed traffic accident scene, rather than direct accident information, multi-rigid body-finite element pedestrian coupling model and multi-rigid body vehicle model, which can greatly reduce the amount of calculation of the virtually reconstructed traffic accident scene; in the virtually reconstructed traffic accident scene, the multi-rigid body pedestrian model is replaced with the multi-rigid body-finite element After the pedestrian coupling model is established, the traffic accident process is simulated and calculated again to make the output head injury index and the extracted head kinematic boundary conditions more accurate; and the univariate linear regression analysis method is used to simulate the multiple traffic accident cases, and the extracted head kinematic boundary conditions and the output head injury index are analyzed for correlation, and the correlation between each head kinematic boundary condition and each head injury index is calculated, and the influence of each head kinematic boundary condition on the head injury is determined according to the calculation results; in this way, the influence of multiple head kinematic boundary conditions on multiple head injuries can be obtained at the same time, and the accuracy of the correlation between the determined head kinematic boundary conditions and the head injury index is high and diverse, which can provide important data and method references for the optimization and updating of VRU safety assessment regulations / procedures, and provide important parameters for the establishment of injury prediction equations.

[0119] Considering that the VRU head in traffic accident cases has extremely complex collision injury mechanisms due to highly diverse collision loads, if the same accident case only considers the influence of a single head kinematics (linear or rotation) without specifically and comprehensively considering the unique head collision loads in traffic accident cases, this may lead to the output of head injury indicators showing "high input, low injury" or "low input, high injury" in some dimensions.

[14] As a further improvement, before performing correlation analysis on the extracted head kinematic boundary conditions and the output head injury indexes by simulating the plurality of traffic accident cases using a univariate linear regression analysis method, the method further includes:

[0120] The linear acceleration peak values ​​in the extracted head kinematic boundary conditions are divided into three magnitude groups: low, medium and high according to the peak value.

[0121] The peak values ​​of rotational acceleration in the extracted head kinematic boundary conditions are divided into three magnitude groups: low, medium and high according to the peak values; and

[0122] The rotation velocity peaks in the extracted head kinematic boundary conditions are divided into two magnitude groups: low and high according to the peak values.

[0123] In the application example of this embodiment, for 50 traffic accident cases, the distribution of the linear acceleration peak value after grouping, the distribution of the rotation acceleration peak value after grouping, and the distribution of the rotation speed peak value after grouping are respectively as follows: Figures 9 to 11 shown.

[0124] In the case of magnitude grouping, in step S3, the correlation analysis of the head kinematic boundary conditions extracted by simulating the multiple traffic accident cases and the output head injury index by using a univariate linear regression analysis method includes:

[0125] The correlations between the low, medium and high magnitude groups of the peak linear acceleration and each head injury index were calculated;

[0126] The correlations between the low, medium and high magnitude groups of the peak rotational acceleration and each head injury index were calculated;

[0127] The correlations between the low and high magnitude groups of peak rotational velocity and various head injury indicators were calculated.

[0128] The correlations between the three magnitude groups of low, medium and high calculated linear acceleration peak values ​​and various head injury indicators include:

[0129] The correlation coefficients R between the low, medium and high levels of the peak linear acceleration and each head injury index were calculated to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the impact of low, medium and high peak values ​​of rotational acceleration on head injury;

[0130] The correlations between the three magnitude groups of low, medium and high calculated rotational acceleration peak values ​​and various head injury indicators include:

[0131] The correlation coefficients R between the low, medium and high magnitude groups of the peak value of rotational acceleration and each head injury index were calculated to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the impact of low, medium and high peak linear acceleration levels on head injury;

[0132] The correlation between the low and high magnitude groups of the calculated rotational velocity peak and each head injury index includes:

[0133] The correlation coefficients R between the low, medium and high levels of the peak rotation speed and each head injury index were calculated to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the impact of low and high peak rotation speed groups on head injuries.

[0134] In the application example of this embodiment, for 50 traffic accident cases, a univariate linear regression analysis method is used to perform correlation analysis on the grouped rotation acceleration peak value PRA, the grouped linear acceleration peak value PLA, the grouped rotation velocity peak value PRV and the output head injury index. The linear regression results of the low-level group linear acceleration peak value and each head injury index are as follows: Fig.12 The linear regression results of the peak linear acceleration of the middleweight group and various head injury indicators are shown in Fig.13 The linear regression results of the peak value of high collision linear acceleration and various head injury indicators are shown in Fig.14 The linear regression results of the peak value of low-impact rotational acceleration and various head injury indicators are shown in Fig.15 The linear regression results of the peak value of rotational acceleration in a mid-impact collision and various head injury indicators are shown in Fig.16 The linear regression results of the peak value of high collision rotational acceleration and various head injury indicators are shown in Fig.17 The linear regression results of the peak rotational velocity in low collisions and various head injury indicators are shown in Fig.18 The linear regression results of the peak rotational velocity of high impact and various head injury indicators are shown in Fig.19 Among them, the linear regression results of head injury indicators based on head kinematics and brain tissue deformation at different magnitudes are shown in Tables 4 and 5.

[0135] Table 4 Linear regression results of head injury indicators based on head kinematics at different levels

[0136]

[0137]

[0138] Table 5 Linear regression results of head injury indicators based on brain tissue strain at different magnitudes

[0139]

[0140]

[0141] From the analysis results of head injury indicators based on head kinematics at different levels in Table 4, it can be seen that GAMBIT has a good performance in the low-level group (R 2 =0.711, p<0.01) and the high-dose group (R 2 =0.703, p<0.01), which showed a better correlation than the middleweight group; HIP showed a higher correlation in the middleweight group (R 2 =0.793, p<0.01; HIC was significantly correlated with LA in the middle and high weight groups, and the correlation was higher in the high weight group (R 2 =0.761,p<0.01).

[0142] Among the groups of rotational acceleration RA, the high-level group showed no significant correlation with the injury indexes; GAMBIT and BrIC showed high correlation in the low and medium-level RA groups (R 2 >0.82, p<0.01); RIC showed correlation only in the low-grade RA group, while HIP showed lower correlation only in the medium-grade RA group.

[0143] In the rotation speed RV level grouping, the correlation between each injury index and it is weak, and RIC is in the high level group (R 2 =0.4732, p<0.01) showed a higher 2 =0.2556, p<0.01) had a better correlation; GAMBIT and BrIC were significantly correlated only in the high-level group, while BrIC had a relatively high correlation with the RV high-level group (R 2 =0.4695,p<0.01).

[0144] When analyzing the injury index analysis results based on brain tissue strain (Table 5), it was found that the linear correlation between the linear acceleration LA, rotational acceleration RA, rotational velocity RV and head injury index was relatively limited. 0.15 Compared with its low level (R 2 =0.672, p<0.01) and high level (R 2 =0.776, p<0.01) were significantly correlated; for rotational acceleration RA, only CSDM 0.15 It has a weak correlation with its low-level group (R 2 =0.4345, p<0.01). For the rotation speed RV groups of different magnitudes, MPS, CSDM 0.15 , CSDM 0.25 The correlation is relatively significant, especially in CSDM 0.15 (R 2=0.4579, p<0.01) and CSDM 0.25 (R 2 =0.472, p<0.01) and low magnitude. It can be seen that, in general, the injury index based on brain tissue strain is more sensitive to rotational velocity RV than to linear acceleration LA and rotational acceleration RA.

[0145] The determination coefficient R was obtained by calculating the correlation coefficient R between the low, medium and high magnitude groups of the peak value of rotational acceleration and each head injury index. 2 , according to the coefficient of determination R 2 To determine the impact of low, medium and high levels of linear acceleration peak on head injuries, it can further provide important data and method references for the optimization and updating of VRU safety assessment regulations / procedures, as well as provide important parameters for the establishment of injury prediction equations.

[0146] As a further improvement, the determination method in the embodiment of the present application further includes:

[0147] Removing the front windshield structure of the multi-rigid body automobile vehicle model;

[0148] The coupling module in the multi-rigid body dynamics analysis software is used to connect the finite element front windshield model with the rigid bodies of the corresponding parts of the multi-rigid body vehicle model to establish a multi-rigid body-finite element vehicle coupling model;

[0149] In the virtually reconstructed traffic accident scene, the multi-rigid body vehicle model is replaced by a multi-rigid body-finite element vehicle coupling model.

[0150] By replacing the multi-rigid body vehicle model with a multi-rigid body-finite element vehicle coupling model in the virtually reconstructed traffic accident scene, the head injury index output in step S24 and the head kinematic boundary conditions extracted in step S25 can be made more accurate. In this way, the degree of influence of each head kinematic boundary condition on the head injury finally determined by the determination method can be made more accurate.

[0151] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for determining the influence of head kinematic boundary conditions on head injury, characterized in that: The method comprises the following steps: S1. Obtaining accident information of multiple traffic accident cases that meet preset conditions, wherein the multiple traffic accident cases all involve collisions between vulnerable road users (VRUs) and vehicles; S2. For each traffic accident case, simulate the traffic accident process according to the accident information of the traffic accident case, output the head injury index and extract the head kinematic boundary conditions in the collision; Wherein, step S2 comprises: S21. Establish a multi-rigid body human model and a multi-rigid body vehicle model according to the accident information of the traffic accident case; S22, using multi-rigid body dynamics analysis software to simulate and calculate the traffic accident process according to the accident information, the multi-rigid body pedestrian model and the multi-rigid body vehicle model to obtain a virtually reconstructed traffic accident scene; S23, establishing a multi-rigid body-finite element pedestrian coupling model according to the human body finite element model and the multi-rigid body pedestrian model; S24, after replacing the multi-rigid body pedestrian model with the multi-rigid body-finite element pedestrian coupling model in the virtually reconstructed traffic accident scene, perform traffic accident process simulation calculation again, and output head injury indicators, wherein the head injury indicators include head injury criterion HIC, brain injury threshold generalized acceleration model GAMBIT, brain injury criterion BrIC, rotational injury criterion RIC, head impact power HIP, maximum principal strain MPS and cumulative strain damage measurement CSDM; and S25, intercepting the complete collision process information between the head of the vulnerable road user VRU and the vehicle from the moment of contact to the moment of separation of the head of the VRU and the vehicle during the simulated traffic accident, and extracting the head kinematic boundary conditions in the collision from the complete collision process information, wherein the head kinematic boundary conditions include the linear velocity peak value PLV, the rotational velocity peak value PRV, the linear acceleration peak value PLA, the rotational acceleration peak value PRA and the collision duration ID of the center of mass during the collision between the head of the VRU and the vehicle; S3. After completing the simulation of the multiple traffic accident cases, a univariate linear regression analysis method is used to perform a correlation analysis on the head kinematic boundary conditions extracted from the simulation of the multiple traffic accident cases and the output head injury indicators, and the correlation between each head kinematic boundary condition and each head injury indicator is calculated, and the degree of influence of each head kinematic boundary condition on the head injury is determined according to the calculation results.

2. The method according to claim 1, characterized in that The prediction model of the univariate linear regression analysis method is: Y=axt+b In the above formula, x t is the head kinematic boundary condition; Y t is the head injury index; a, b represent the parameters of the univariate linear regression equation, and a, b are obtained by the following two equations: In the above two formulas, n represents the number of data in the data set, x i is the value of the i-th head kinematic boundary condition in the data set, Y i is the value of the i-th head injury indicator in the data set.

3. The method according to claim 1, characterized in that The preset conditions include: The accident participants in the accident information involve VRU; Accident information is sufficient to support virtual reconstruction of traffic accident scenes; The VRU head of the accident information collided with the vehicle body.

4. The method according to claim 1, characterized in that The multiple traffic accident cases include pedestrian-vehicle collision accident cases and two-wheeled vehicle-car collision accident cases. For pedestrian-vehicle collision accident cases, the multi-rigid body vehicle model includes a multi-rigid body car vehicle model; for two-wheeled vehicle-car collision accident cases, the multi-rigid body vehicle model includes a multi-rigid body car vehicle model and a multi-rigid body two-wheeled vehicle model.

5. The method according to claim 1, characterized in that Before the simulation of the traffic accident process, it also includes: Removing the front windshield structure of the multi-rigid body automobile vehicle model; The coupling module in the multi-rigid body dynamics analysis software is used to connect the finite element front windshield model with the rigid bodies of the corresponding parts of the multi-rigid body vehicle model to establish a multi-rigid body-finite element vehicle coupling model; In the virtually reconstructed traffic accident scene, the multi-rigid body vehicle model is replaced by a multi-rigid body-finite element vehicle coupling model.

6. The method according to claim 1, characterized in that The multi-rigid body-finite element pedestrian coupling model is established according to the human body finite element model and the multi-rigid body pedestrian model, including: Removing the head and neck structure of the multi-rigid-body pedestrian model; The head and neck of the human finite element model are extracted, and the muscles and end nodes of the cervical vertebrae in the neck of the human finite element model, which are originally connected to the torso, are connected to the rigid bodies at the corresponding parts of the upper torso of the multi-rigid body pedestrian model after the head and neck are removed, using the coupling module in the multi-rigid body dynamics analysis software, to obtain a multi-rigid body-finite element pedestrian coupling model.

7. The method according to claim 6, characterized in that The multi-rigid body human body model is obtained by scaling and adjusting the TNO 50 percentile adult male and 5 percentile female multi-rigid body pedestrian models using the multi-rigid body dynamics analysis software MADYMO; the human body finite element model uses the virtual human body model THUMS 50 percentile adult male finite element model.

8. The method according to claim 1, characterized in that The calculating of the correlation between each head kinematic boundary condition and each head injury index, and determining the influence of each head kinematic boundary condition on the head injury according to the calculation result, includes: Calculate the correlation coefficient R between each head kinematic boundary condition and each head injury index to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the influence of various head kinematic boundary conditions on head injury.

9. The method according to claim 1, characterized in that Before performing correlation analysis on the extracted head kinematic boundary conditions and the output head injury indexes by simulating the plurality of traffic accident cases using a univariate linear regression analysis method, the method further includes: The linear acceleration peak values ​​in the extracted head kinematic boundary conditions are divided into three magnitude groups: low, medium and high according to the peak value. The peak values ​​of rotational acceleration in the extracted head kinematic boundary conditions are divided into three magnitude groups: low, medium and high according to the peak values; and The peak value of rotation speed in the extracted head kinematic boundary conditions is divided into two magnitude groups: low and high. The univariate linear regression analysis method is used to simulate the multiple traffic accident cases to perform correlation analysis on the extracted head kinematic boundary conditions and the output head injury indicators, including: The correlations between the low, medium and high magnitude groups of the peak linear acceleration and each head injury index were calculated; The correlations between the low, medium and high magnitude groups of the peak rotational acceleration and each head injury index were calculated; The correlations between the low and high magnitude groups of peak rotational velocity and various head injury indicators were calculated.

10. The method according to claim 9, characterized in that The correlation between the three magnitude groups of low, medium and high calculated linear acceleration peak values ​​and each head injury index includes: The correlation coefficients R between the low, medium and high levels of the peak linear acceleration and each head injury index were calculated to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the impact of low, medium and high peak linear acceleration levels on head injury; The correlations between the three magnitude groups of low, medium and high calculated rotational acceleration peak values ​​and various head injury indicators include: The correlation coefficients R between the low, medium and high magnitude groups of the peak value of rotational acceleration and each head injury index were calculated to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the impact of low, medium and high peak linear acceleration levels on head injury; The correlation between the low and high magnitude groups of the calculated rotational velocity peak and each head injury index includes: The correlation coefficients R between the low, medium and high levels of the peak rotation speed and each head injury index were calculated to obtain the determination coefficient R 2 , according to the coefficient of determination R 2 To determine the impact of low and high peak rotation speed groups on head injuries.

Citation Information

Patent Citations

  • Field rapid evaluation method for pedestrian head closed injuries based on improved evaluation index

    CN105447234A

  • Vehicle accident mitigation method, system and device based on data driving and medium

    CN115345057A

  • Dimensional reduction of finite element analysis solution for rapid emulation

    EP3486814A1

  • Systems, devices, and methods for assessment of brain injury

    WO2020123875A1

  • Collision severity prediction method for passenger injury risk

    WO2021072923A1