A parametric assessment system for neck injuries in zero-gravity chairs

Abstract

The present invention discloses a parametric evaluation system for neck injuries of zero-gravity seats, including an evaluation model preprocessing module, a large-scale simulation prediction module, a neck parametric injury prediction module, and an evaluation system feedback module; the evaluation model preprocessing module includes a feature recognition unit, a geometry processing unit, a true prevention unit, a model angle adjustment unit, and a grid deformation unit; the large-scale simulation prediction module includes a DOE matrix construction unit, a simulation matrix calculation unit, and a result post-processing unit; the neck parametric injury prediction module includes a neck feature extraction unit, a spatial node dimension reduction unit, a parametric regression unit, and a neck injury state prediction unit; the evaluation system feedback module includes a prediction accuracy evaluation unit, a prediction model adjustment unit, a model preprocessing accuracy evaluation unit, and a model preprocessing adjustment unit. The evaluation system of the present invention evaluates the safety of zero-gravity seats by means of a human body model, and the system structure is simple and easy to use.

Description

Technical Field

[0001] The invention relates to the technical field of safety assessment of people in a vehicle, and in particular to a parametric assessment system for neck injuries of a zero-gravity seat. Background Art

[0002] As the level of vehicle intelligence continues to increase, drivers no longer need to take on driving tasks, so smart cockpits have become the mainstream trend in the development of driverless vehicles. The biggest difference between smart cockpits and traditional cockpits is that more ergonomic zero-gravity seats have replaced traditional seats, which has significantly changed the posture of the human body on zero-gravity seats compared to traditional seats. Although this has greatly improved the comfort of the seats, the restraint system of traditional seats is not suitable for zero-gravity seats, resulting in safety hazards. However, there is currently a lack of research on human injuries caused by zero-gravity seats, so there is no restraint system suitable for zero-gravity seats. Summary of the invention

[0003] In view of this, in order to solve the above problems, the present invention proposes a parametric assessment system for human neck injuries in a zero-gravity chair.

[0004] The above technical objectives of the present invention are achieved through the following technical solutions:

[0005] A parametric evaluation system for neck injuries in zero-gravity chairs, comprising an evaluation model preprocessing module, a large-scale simulation prediction module, a neck parametric injury prediction module, and an evaluation system feedback module;

[0006] The evaluation model preprocessing module includes a feature recognition unit, a geometry processing unit, a false prevention unit, a model angle adjustment unit, and a mesh deformation unit;

[0007] The large-scale simulation prediction module includes a DOE matrix construction unit, a simulation matrix calculation unit, and a result post-processing unit;

[0008] The neck parametric injury prediction module includes a neck feature extraction unit, a spatial node dimension reduction unit, a parametric regression unit, and a neck injury state prediction unit;

[0009] The evaluation system feedback module includes a prediction accuracy evaluation unit, a prediction model adjustment unit, a model preprocessing accuracy evaluation unit, and a model preprocessing adjustment unit.

[0010] Further preferably, the following usage process is included:

[0011] Step 1, selecting volunteers to be tested for evaluating the zero-gravity seat, the zero-gravity seat to be tested and its CAE model;

[0012] Step 2, extracting the features of the volunteer to be tested and the zero-gravity chair obtained in step 1 through the feature recognition unit in the evaluation model preprocessing module, and deforming the reference human finite element model through the mesh deformation unit to obtain a human finite element model with the features of the volunteer to be tested;

[0013] Step 3, obtaining the human body geometry adapted to the zero-gravity chair and the human body finite element model adapted to the zero-gravity chair by evaluating the geometry processing unit and the model angle adjustment unit in the model preprocessing module;

[0014] Step 4, obtaining the human-chair coupling model by evaluating the preventive true unit in the model preprocessing module;

[0015] Step 5, constructing a multi-condition simulation matrix of the human-chair coupling model through the DOE matrix construction unit of the large-scale simulation prediction module, performing simulation calculations through the simulation matrix calculation unit, and obtaining simulation results through the result post-processing unit;

[0016] Step 6, extracting neck injury features from the simulation results obtained in step 5 through the neck feature extraction unit in the neck parametric injury prediction module, establishing a neck deformation state prediction model through the spatial node dimension reduction unit and the parametric regression unit, and establishing a neck injury state prediction model through the neck injury state prediction unit;

[0017] Step 7, evaluate the system through the model preprocessing accuracy evaluation unit and the prediction accuracy evaluation unit in the evaluation system feedback module. If it is lower than the predetermined threshold, feedback adjustment is performed through the model preprocessing adjustment unit and the prediction model adjustment unit respectively.

[0018] Further preferably, the evaluation model preprocessing module comprises the following steps:

[0019] Step 1.1, select the volunteers to be tested, the zero-gravity seat to be tested and the corresponding CAE model, identify the body surface point cloud of the volunteers to be tested through the feature recognition unit, the change angle of the leg support of the zero-gravity seat compared with the traditional seat, the change angle of the backrest of the zero-gravity seat compared with the traditional seat, and the change angle of the head support compared with the traditional seat;

[0020] Step 1.2, inputting the body surface point cloud of the volunteer to be tested into the mesh deformation unit, deforming the reference model to the body surface point cloud of the volunteer to be tested by mesh deformation technology, and obtaining the initial model of the volunteer to be tested, wherein both the reference model and the model of the volunteer to be tested are standard sitting postures corresponding to the standard seat;

[0021] Step 1.3, extract the surface mesh nodes of the initial model of the volunteer to be tested and input them into the geometry processing unit for segmentation, dividing them into four parts: calf, thigh, trunk, and head. The calf and thigh mesh nodes are adjusted at the same angle as the leg support change angle, the trunk mesh nodes are adjusted at the same angle as the seat back, and the head mesh nodes are adjusted at the same angle as the head support change angle;

[0022] Step 1.4, input the initial model of the volunteer to be tested and the adjusted mesh nodes into the model angle adjustment unit, first fix the pelvic model, and deform the trunk model to the deformed trunk mesh nodes; then fix the trunk model, and deform the head model to the deformed head mesh nodes; then fix the knee joint model, and deform the calf model to the calf mesh nodes, so as to obtain the adaptive model of the volunteer to be tested;

[0023] Step 1.5, input the adaptive model of the volunteer to be tested into the pre-simulation unit, couple it with the zero-gravity chair CAE model, and perform preventive simulation, so that the adaptive model of the volunteer to be tested fits the zero-gravity chair CAE model to obtain a human-chair coupling model;

[0024] Step 1.6, have the volunteer wear the neck pressure test device and neck protection device, lie flat on the zero-gravity chair, and wear a safety belt.

[0025] Further preferably, the large-scale simulation prediction module comprises the following steps:

[0026] Step 2.1, input the human-chair model coupling model in the evaluation model preprocessing module into the DOE matrix construction unit, where the matrix variables are collision acceleration, collision direction, leg support angle, backrest angle, head support angle, and seat belt pretension;

[0027] Step 2.2, setting the human-chair coupling model based on the DOE simulation matrix output by the DOE matrix construction unit, and inputting it into the simulation matrix calculation unit for calculation;

[0028] Step 2.3, extract the calculation results and input them into the result post-processing unit to extract the neck mesh surface nodes at the maximum displacement of the head under different working conditions, as well as the C1-C7 surface mesh nodes.

[0029] Further preferably, the neck parameterized injury prediction module comprises the following steps:

[0030] Step 3.1: extract the C1-C7 surface mesh nodes P under different working conditions. C_I , input to the neck feature extraction unit to perform spatial registration with the initial state C1-C7 surface mesh nodes, and obtain the registered C1-C7 surface mesh nodes P C, the registration process is shown as follows, P C_I Transfer to the C1-C7 surface mesh nodes in the initial state through the rotation matrix R0 and the translation matrix T0;

[0031] P C =P C_I *R0+T0

[0032] Step 3.2, calculate the maximum elongation C of C1-C7 by the following formula L , where L T is the length of the ligament after the change, L I is the initial ligament length, if C L If it exceeds 36.9%, it is considered a ligament strain;

[0033] C L =(L T -L I ) / L I *100%

[0034] Step 3.3, input the extracted C1-C7 surface grid nodes under different working conditions into the spatial node dimension reduction unit, expand the spatial node coordinates into one-dimensional spatial information G, use PCA to reduce the dimension of the spatial information, and obtain the low-dimensional principal component matrix P and principal component score matrix S, as shown in the following formula;

[0035] G=SP

[0036] Step 3.4, input the principal component matrix into the parameterized regression unit, and perform regression modeling with the characteristic parameters F such as collision acceleration, collision direction, leg support angle, backrest angle, head support angle, and seat belt preload using the multivariate quadratic regression model as the benchmark model to establish a neck deformation state prediction model, as shown in the following formula, where A is the regression model matrix coefficient, and ε is the regression model offset coefficient;

[0037] S=FA+ε

[0038] Step 3.5, input the principal component matrix and the corresponding injury status into the neck injury status prediction unit, and establish a neck injury status prediction model based on the support vector machine model. The neck injury status prediction model takes the principal component matrix as input and the neck injury status as output.

[0039] Further preferably, the evaluation system feedback module comprises the following steps:

[0040] Step 4.1 Input the human-chair coupling model established in the evaluation model preprocessing module into the model preprocessing accuracy evaluation unit to obtain the head displacement and average neck pressure, and compare them with the volunteer experiment;

[0041] Step 4.2, if the error is less than 5%, the human-chair coupling model is verified; if the error is greater than 5%, the model preprocessing accuracy evaluation unit outputs a feedback signal to the model preprocessing module, and steps 1.1-1.6 are repeated;

[0042] Step 4.3, input the neck deformation state prediction model established by the parameterized regression unit into the prediction accuracy evaluation unit, input the collision acceleration, collision direction, leg support angle, backrest angle, head support angle, seat belt preload, and obtain the corresponding neck space grid node through the neck deformation state prediction model;

[0043] Step 4.4, comparing the predicted neck space grid nodes with the neck space grid nodes obtained by simulation under the same boundary conditions in the large-scale simulation prediction module, and calculating the spatial distance between nodes with the same node number;

[0044] Step 4.5, if the average distance is less than 2 mm, the test is passed; if it is greater than 2 mm, the prediction model adjustment unit outputs a feedback signal to the parameterized regression unit for remodeling.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] First, the evaluation system proposed in the present invention evaluates the safety of zero-gravity seats by means of a human body model, which can show the neck injury under different collision conditions, improving the limitations of experimental evaluation. At the same time, the system has a simple structure and is easy to use.

[0047] Second, the evaluation system proposed by the present invention can be adaptively adjusted according to different characteristics of the human body to obtain evaluation results corresponding to the characteristics of the test person. Compared with the traditional evaluation system that can only evaluate the human body with characteristic signs, the present invention can evaluate the human body with different characteristics, so that the evaluation results of the zero-gravity chair to be tested can be more generalized;

[0048] Thirdly, the model preprocessing function proposed in the present invention can perform mesh deformation and posture adjustment on the human body model, so that the human body model can adaptively adapt to zero-gravity chairs of different structures. Compared with the traditional evaluation system in which the human body model is fixed to a standard posture, the evaluation system proposed in the present invention is more flexible.

[0049] Fourthly, the parametric neck injury prediction module proposed in the present invention can perform customized predictions by inputting characteristic parameters, and the prediction results are displayed in two forms, 3D and 1D, which can intuitively reflect the human neck injury situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a framework diagram of the evaluation system of the present invention;

[0051] Figure 2 is an evaluation flow chart of the present invention;

[0052] Figure 3 It is a framework diagram of the evaluation process of the evaluation system of the present invention. DETAILED DESCRIPTION

[0053] In order to make those skilled in the art better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with embodiments. It will be understood by those skilled in the art that the following embodiments are only used to illustrate the present invention and should not be regarded as limiting the scope of the present invention.

[0054] Example, see Figure 1 , a parametric evaluation system for neck injury in zero-gravity chairs, including an evaluation model preprocessing module, a large-scale simulation prediction module, a neck parametric injury prediction module, and an evaluation system feedback module;

[0055] The evaluation model preprocessing module includes a feature recognition unit, a geometry processing unit, a false prevention unit, a model angle adjustment unit, and a mesh deformation unit;

[0056] The large-scale simulation prediction module includes a DOE matrix construction unit, a simulation matrix calculation unit, and a result post-processing unit;

[0057] The parametric neck injury prediction module includes a neck feature extraction unit, a spatial node dimension reduction unit, a parametric regression unit, and a neck injury state prediction unit;

[0058] The evaluation system feedback module includes a prediction accuracy evaluation unit, a prediction model adjustment unit, a model preprocessing accuracy evaluation unit, and a model preprocessing adjustment unit.

[0059] According to the above evaluation system, this embodiment provides a use process of the evaluation system, referring to Figure 2 , including the following steps:

[0060] Step 1, selecting volunteers to be tested for evaluating the zero-gravity seat, the zero-gravity seat to be tested and its CAE model;

[0061] Step 2, extracting the features of the volunteer to be tested and the zero-gravity chair obtained in step 1 through the feature recognition unit in the evaluation model preprocessing module, and deforming the reference human finite element model through the mesh deformation unit to obtain a human finite element model with the features of the volunteer to be tested;

[0062] Step 3, obtaining the human body geometry adapted to the zero-gravity chair and the human body finite element model adapted to the zero-gravity chair by evaluating the geometry processing unit and the model angle adjustment unit in the model preprocessing module;

[0063] Step 4, obtaining the human-chair coupling model by evaluating the preventive true unit in the model preprocessing module;

[0064] Step 5, constructing a multi-condition simulation matrix of the human-chair coupling model through the DOE matrix construction unit of the large-scale simulation prediction module, performing simulation calculations through the simulation matrix calculation unit, and obtaining simulation results through the result post-processing unit;

[0065] Step 6, extracting neck injury features from the simulation results obtained in step 5 through the neck feature extraction unit in the neck parametric injury prediction module, establishing a neck deformation state prediction model through the spatial node dimension reduction unit and the parametric regression unit, and establishing a neck injury state prediction model through the neck injury state prediction unit;

[0066] Step 7, evaluate the system through the model preprocessing accuracy evaluation unit and the prediction accuracy evaluation unit in the evaluation system feedback module. If it is lower than the predetermined threshold, feedback adjustment is performed through the model preprocessing adjustment unit and the prediction model adjustment unit respectively.

[0067] Among them, refer to Figure 3 , the evaluation model preprocessing module includes the following steps:

[0068] Step 1.1, select the volunteers to be tested, the zero-gravity seat to be tested and the corresponding CAE model, identify the body surface point cloud of the volunteers to be tested through the feature recognition unit, the change angle of the leg support of the zero-gravity seat compared with the traditional seat, the change angle of the backrest of the zero-gravity seat compared with the traditional seat, and the change angle of the head support compared with the traditional seat;

[0069] Step 1.2, inputting the body surface point cloud of the volunteer to be tested into the mesh deformation unit, deforming the reference model to the body surface point cloud of the volunteer to be tested by mesh deformation technology, and obtaining the initial model of the volunteer to be tested, wherein both the reference model and the model of the volunteer to be tested are in the standard sitting posture corresponding to the standard seat;

[0070] Step 1.3, extract the surface mesh nodes of the initial model of the volunteer to be tested and input them into the geometry processing unit for segmentation, dividing them into four parts: calf, thigh, trunk, and head. The calf and thigh mesh nodes are adjusted at the same angle as the leg support change angle, the trunk mesh nodes are adjusted at the same angle as the seat back, and the head mesh nodes are adjusted at the same angle as the head support change angle;

[0071] Step 1.4, input the initial model of the volunteer to be tested and the adjusted mesh nodes into the model angle adjustment unit, first fix the pelvic model, and deform the trunk model to the deformed trunk mesh nodes; then fix the trunk model, and deform the head model to the deformed head mesh nodes; then fix the knee joint model, and deform the calf model to the calf mesh nodes, so as to obtain the adaptive model of the volunteer to be tested;

[0072] Step 1.5, input the adaptive model of the volunteer to be tested into the pre-simulation unit, couple it with the zero-gravity chair CAE model, and perform preventive simulation, so that the adaptive model of the volunteer to be tested fits the zero-gravity chair CAE model to obtain a human-chair coupling model;

[0073] Step 1.6, have the volunteer wear the neck pressure test device and neck protection device, lie flat on the zero-gravity chair, and wear a safety belt.

[0074] Among them, refer to Figure 3 ,The large-scale simulation prediction module includes the following steps:

[0075] Step 2.1, input the human-chair model coupling model in the evaluation model preprocessing module into the DOE matrix construction unit, where the matrix variables are collision acceleration, collision direction, leg support angle, backrest angle, head support angle, and seat belt pretension;

[0076] Step 2.2, setting the human-chair coupling model based on the DOE simulation matrix output by the DOE matrix construction unit, and inputting it into the simulation matrix calculation unit for calculation;

[0077] Step 2.3, extract the calculation results and input them into the result post-processing unit to extract the neck mesh surface nodes at the maximum displacement of the head under different working conditions, as well as the C1-C7 surface mesh nodes.

[0078] Among them, refer to Figure 3 ,The neck parametric injury prediction module includes the following steps:

[0079] Step 3.1: extract the C1-C7 surface mesh nodes P under different working conditions. C_I , input to the neck feature extraction unit to perform spatial registration with the initial state C1-C7 surface mesh nodes, and obtain the registered C1-C7 surface mesh nodes P C , the registration process is shown as follows, P C_I Transfer to the C1-C7 surface mesh nodes in the initial state through the rotation matrix R0 and the translation matrix T0;

[0080] P C =P C_I *R0+T0

[0081] Step 3.2, calculate the maximum elongation C of C1-C7 by the following formula L , where L T is the length of the ligament after the change, L I is the initial ligament length, if C L If it exceeds 36.9%, it is considered a ligament strain;

[0082] C L =(L T -L I ) / L I *100%

[0083] Step 3.3, input the extracted C1-C7 surface grid nodes under different working conditions into the spatial node dimension reduction unit, expand the spatial node coordinates into one-dimensional spatial information G, use PCA to reduce the dimension of the spatial information, and obtain the low-dimensional principal component matrix P and principal component score matrix S, as shown in the following formula;

[0084] G=SP

[0085] Step 3.4, input the principal component matrix into the parameterized regression unit, and perform regression modeling with the characteristic parameters F such as collision acceleration, collision direction, leg support angle, backrest angle, head support angle, and seat belt preload using the multivariate quadratic regression model as the benchmark model to establish a neck deformation state prediction model, as shown in the following formula, where A is the regression model matrix coefficient, and ε is the regression model offset coefficient;

[0086] S=FA+ε

[0087] Step 3.5, input the principal component matrix and the corresponding injury status into the neck injury status prediction unit, and establish a neck injury status prediction model based on the support vector machine. The neck injury status prediction model takes the principal component matrix as input and the neck injury status as output.

[0088] Among them, refer to Figure 3 ,The evaluation system feedback module includes the following steps:

[0089] Step 4.1 Input the human-chair coupling model established in the evaluation model preprocessing module into the model preprocessing accuracy evaluation unit to obtain the head displacement and average neck pressure, and compare them with the volunteer experiment;

[0090] Step 4.2, if the error is less than 5%, the human-chair coupling model is verified; if the error is greater than 5%, the model preprocessing accuracy evaluation unit outputs a feedback signal to the model preprocessing module, and steps 1.1-1.6 are repeated;

[0091] Step 4.3, input the neck deformation state prediction model established by the parameterized regression unit into the prediction accuracy evaluation unit, input the collision acceleration, collision direction, leg support angle, backrest angle, head support angle, seat belt preload, and obtain the corresponding neck space grid node through the neck deformation state prediction model;

[0092] Step 4.4, comparing the predicted neck space grid nodes with the neck space grid nodes obtained by simulation under the same boundary conditions in the large-scale simulation prediction module, and calculating the spatial distance between nodes with the same node number;

[0093] Step 4.5, if the average distance is less than 2 mm, the test is passed; if it is greater than 2 mm, the prediction model adjustment unit outputs a feedback signal to the parameterized regression unit for remodeling.

[0094] The above embodiments are merely explanations of the present invention and are not limitations of the present invention. After reading this specification, those skilled in the art may make modifications to the embodiments without any creative contribution as needed. However, such modifications are protected by the patent law as long as they are within the scope of the claims of the present invention.

Claims

1. A parametric assessment system for neck injury in zero-gravity chairs, characterized in that: It includes evaluation model preprocessing module, large-scale simulation prediction module, neck parameterized injury prediction module, and evaluation system feedback module; The evaluation model preprocessing module includes a feature recognition unit, a geometry processing unit, a pre-simulation unit, a model angle adjustment unit, and a mesh deformation unit; The large-scale simulation prediction module includes a DOE matrix construction unit, a simulation matrix calculation unit, and a result post-processing unit; The neck parametric injury prediction module includes a neck feature extraction unit, a spatial node dimension reduction unit, a parametric regression unit, and a neck injury state prediction unit; The evaluation system feedback module includes a prediction accuracy evaluation unit, a prediction model adjustment unit, a model preprocessing accuracy evaluation unit, and a model preprocessing adjustment unit; The system is used to perform the following usage processes: Step 1, selecting volunteers to be tested for evaluating the zero-gravity seat, the zero-gravity seat to be tested and its CAE model; Step 2, extracting the features of the volunteer to be tested and the zero-gravity chair obtained in step 1 through the feature recognition unit in the evaluation model preprocessing module, and deforming the reference human finite element model through the mesh deformation unit to obtain a human finite element model with the features of the volunteer to be tested; Step 3, obtaining the human body geometry adapted to the zero-gravity chair and the human body finite element model adapted to the zero-gravity chair by evaluating the geometry processing unit and the model angle adjustment unit in the model preprocessing module; Step 4, obtaining a human-chair coupling model by evaluating a pre-simulation unit in a model pre-processing module; Step 5, constructing a multi-condition simulation matrix of the human-chair coupling model through the DOE matrix construction unit of the large-scale simulation prediction module, performing simulation calculations through the simulation matrix calculation unit, and obtaining simulation results through the result post-processing unit; Step 6, extracting neck injury features from the simulation results obtained in step 5 through the neck feature extraction unit in the neck parametric injury prediction module, establishing a neck deformation state prediction model through the spatial node dimension reduction unit and the parametric regression unit, and establishing a neck injury state prediction model through the neck injury state prediction unit; Step 7, evaluating the system through the model preprocessing accuracy evaluation unit and the prediction accuracy evaluation unit in the evaluation system feedback module, and if it is lower than the predetermined threshold, feedback adjustment is performed through the model preprocessing adjustment unit and the prediction model adjustment unit respectively; The evaluation model preprocessing module is used to perform the following steps: Step 1.1, select the volunteers to be tested, the zero-gravity seat to be tested and the corresponding CAE model, identify the body surface point cloud of the volunteers to be tested through the feature recognition unit, the change angle of the leg support of the zero-gravity seat compared with the traditional seat, the change angle of the backrest of the zero-gravity seat compared with the traditional seat, and the change angle of the head support compared with the traditional seat; Step 1.2, inputting the body surface point cloud of the volunteer to be tested into the mesh deformation unit, deforming the reference human finite element model into the body surface point cloud of the volunteer to be tested by mesh deformation technology, and obtaining the initial model of the volunteer to be tested, wherein both the reference human finite element model and the initial model of the volunteer to be tested are standard sitting postures corresponding to the standard seat; Step 1.3, extract the surface mesh nodes of the initial model of the volunteer to be tested and input them into the geometry processing unit for segmentation, dividing them into four parts: calf, thigh, trunk, and head. The calf and thigh mesh nodes are adjusted at the same angle as the leg support change angle, the trunk mesh nodes are adjusted at the same angle as the seat back, and the head mesh nodes are adjusted at the same angle as the head support change angle; Step 1.4, input the initial model of the volunteer to be tested and the adjusted mesh nodes into the model angle adjustment unit, first fix the pelvic model, and deform the torso model towards the deformed torso mesh nodes; Then fix the torso model and deform the head model towards the deformed head mesh node; Then, the knee joint model is fixed, and the calf model is deformed toward the calf mesh node, thereby obtaining an adaptive model of the volunteer to be tested; Step 1.5, input the adaptive model of the volunteer to be tested into the pre-simulation unit, couple it with the zero-gravity chair CAE model, and perform pre-simulation to make the adaptive model of the volunteer to be tested fit with the zero-gravity chair CAE model to obtain a human-chair coupling model.

2. A parametric assessment system for neck injury in zero-gravity chairs according to claim 1, characterized in that: The large-scale simulation prediction module is used to perform the following steps: Step 2.1, input the human-chair model coupling model in the evaluation model preprocessing module into the DOE matrix construction unit, where the matrix variables are collision acceleration, collision direction, leg support angle, backrest angle, head support angle, and seat belt pretension; Step 2.2, setting the human-chair coupling model based on the DOE simulation matrix output by the DOE matrix construction unit, and inputting it into the simulation matrix calculation unit for calculation; Step 2.3, extract the calculation results and input them into the result post-processing unit to extract the neck mesh surface nodes at the maximum displacement of the head under different working conditions, as well as the C1-C7 surface mesh nodes.

3. The parametric evaluation system for neck injury in zero-gravity chairs according to claim 1, characterized in that: The neck parameterized damage prediction module is used to perform the following steps: Step 3.1, extract the C1-C7 surface mesh nodes under different working conditions Input to the neck feature extraction unit to perform spatial registration with the initial state C1-C7 surface mesh nodes to obtain the registered C1-C7 surface mesh nodes , the registration process is shown as follows: Through the rotation matrix With the translation matrix , transfer to the C1-C7 surface mesh nodes of the initial state; Step 3.2, calculate the maximum elongation of C1-C7 by the following formula ,in is the length of the ligament after the change, is the initial ligament length, if If it exceeds 36.9%, it is considered a ligament strain; Step 3.3, input the extracted C1-C7 surface grid nodes under different working conditions into the spatial node dimension reduction unit, expand the spatial node coordinates into one-dimensional spatial information G, use PCA to reduce the dimension of the spatial information, and obtain the low-dimensional principal component matrix P and principal component score matrix S, as shown in the following formula; Step 3.4, input the principal component matrix into the parameterized regression unit, and perform regression modeling with the collision acceleration, collision direction, leg support angle, backrest angle, head support angle, and seat belt preload using the multivariate quadratic regression model as the benchmark human finite element model to establish a neck deformation state prediction model, as shown in the following formula, where A is the regression model matrix coefficient, is the regression model offset coefficient; Step 3.5, input the principal component matrix and the corresponding injury status into the neck injury status prediction unit, and establish a neck injury status prediction model based on the support vector machine model. The neck injury status prediction model takes the principal component matrix as input and the neck injury status as output.

4. The parametric evaluation system for neck injury in zero-gravity chairs according to claim 1, characterized in that: The evaluation system feedback module is used to perform the following steps: Step 4.1 Input the human-chair coupling model established in the evaluation model preprocessing module into the model preprocessing accuracy evaluation unit to obtain the head displacement and average neck pressure, and compare them with the volunteer experiment; Step 4.2, if the error is less than 5%, the human-chair coupling model is verified. If the error is higher than 5%, the model preprocessing accuracy evaluation unit outputs a feedback signal to the model preprocessing module, and steps 1.1-1.6 are repeated; Step 4.3, input the neck deformation state prediction model established by the parameterized regression unit into the prediction accuracy evaluation unit, input the collision acceleration, collision direction, leg support angle, backrest angle, head support angle, seat belt preload, and obtain the corresponding neck space grid node through the neck deformation state prediction model; Step 4.4, comparing the predicted neck space grid nodes with the neck space grid nodes obtained by simulation under the same boundary conditions in the large-scale simulation prediction module, and calculating the spatial distance between nodes with the same node number; Step 4.5, if the average distance is less than 2 mm, the test is passed; if it is greater than 2 mm, the prediction model adjustment unit outputs a feedback signal to the parameterized regression unit for remodeling.

Images