Seat parameter optimization method based on driver comfort

Abstract

The application provides a seat parameter optimization method based on driver comfort, which comprises the following steps: establishing a driver restraint system finite element model based on a vehicle finite element model, a dummy model and a pre-established safety belt model, an airbag model and a seat model; determining seat parameters including a seat backrest angle, a seat cushion angle, a seat leg support angle and a seat filler stiffness and establishing an orthogonal experiment table with the seat parameters as experimental factors; performing vehicle collision simulation based on the orthogonal experiment table and the driver restraint system finite element model to obtain seat comfort simulation results; constructing a response surface model corresponding to the seat parameters based on each body pressure distribution index based on the seat comfort simulation results; and iteratively solving the response surface model based on the seat comfort simulation results to obtain the optimization results of the seat parameters. The seat parameters can be optimized by adopting the application, so that the seat comfort can be improved on the basis of ensuring the safety of the driver.

Description

Technical Field

[0001] This invention relates to the field of automotive seat design technology, and in particular to a method for optimizing seat parameters based on driver comfort. Background Technology

[0002] Existing passive safety technologies still have limitations when dealing with drivers in abnormal seating positions. For driver restraint systems, seat design is one of the important factors affecting comfort. Analyzing seat comfort in pre-collision restraint systems is of great significance for improving driver comfort and providing more comprehensive safety protection.

[0003] Currently, autonomous vehicles cannot completely avoid traffic accidents, and passive safety technologies still need continuous improvement to provide more comfortable safety protection for drivers. In the future, advanced driver assistance technologies will perceive driver status, road conditions, and the level of danger more accurately, enabling earlier collision warnings and leaving more room for the design of driver restraint systems.

[0004] However, traditional restraint systems distribute the load unevenly on the driver; under certain collision conditions, the restraint system cannot effectively restrain the driver's movement, thus causing additional injury; and unreasonable seat stiffness can cause discomfort or additional injury to the driver. Therefore, existing passive safety technologies are practically unable to provide a comfortable and safe protection for the driver inside the vehicle through restraint systems alone. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a method for optimizing seat parameters based on driver comfort, so as to alleviate the above-mentioned problems existing in the related art.

[0006] This invention provides a method for optimizing seat parameters based on driver comfort, comprising: acquiring a vehicle finite element model, and configuring a dummy model and pre-established seat belt, airbag, and seat models for the vehicle finite element model to establish a driver restraint system finite element model; wherein the seat model has leg support and an adjustable mechanism; determining seat parameters including seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness, and establishing an orthogonal experimental table with the seat parameters as experimental factors; performing vehicle collision simulation based on the orthogonal experimental table and the driver restraint system finite element model to obtain seat comfort simulation results; wherein the seat comfort simulation results include output values ​​corresponding to multiple body pressure distribution indices; constructing a response surface model for each body pressure distribution index corresponding to the seat parameter based on the seat comfort simulation results, and iteratively solving the response surface model based on the seat comfort simulation results to obtain the optimized results of the seat parameters; wherein the response surface model characterizes the relationship between the seat parameters and the body pressure distribution indices.

[0007] This invention provides a method for optimizing seat parameters based on driver comfort. First, a finite element model of the vehicle is acquired, and a dummy model, along with pre-established seatbelt, airbag, and seat models, are configured within this model to create a finite element model of the driver restraint system. Then, seat parameters, including seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness, are determined, and an orthogonal experimental table is established using these parameters as experimental factors. Next, vehicle collision simulations are performed based on the orthogonal experimental table and the finite element model of the driver restraint system to obtain seat comfort simulation results. Following this, a response surface model is constructed based on the seat comfort simulation results, corresponding to each body pressure distribution index and the seat parameter. The response surface model is then iteratively solved based on the seat comfort simulation results to obtain the optimized seat parameters. Using this technique, a finite element model of the driver restraint system and an orthogonal experimental table can be established based on the vehicle finite element model and seat parameters. A response surface model is constructed through vehicle collision simulations, and then the seat parameters are optimized by iteratively solving the response surface model, thereby improving seat comfort while ensuring driver safety.

[0008] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0009] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0010] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0011] Figure 1 This is a flowchart illustrating a method for optimizing seat parameters based on driver comfort in an embodiment of the present invention.

[0012] Figure 2 This is a flowchart of the finite element method in an embodiment of the present invention;

[0013] Figure 3 This is a detailed illustration of the vehicle structure and interior in an embodiment of the present invention;

[0014] Figure 4 This is a schematic diagram of dummy positioning in an embodiment of the present invention;

[0015] Figure 5 This is a schematic diagram of dummy adjustment in an embodiment of the present invention;

[0016] Figure 6 This is a schematic diagram of the THUMS dummy pre-simulation in an embodiment of the present invention;

[0017] Figure 7 These are driver motion response diagrams from real vehicle tests and simulation tests in this embodiment of the invention;

[0018] Figure 8 This is a damage curve diagram of various parts of the driver in an embodiment of the present invention;

[0019] Figure 9 This is a diagram showing the output parameters of the seat belt in an embodiment of the present invention;

[0020] Figure 10 This is a diagram of seat contact force in an embodiment of the present invention;

[0021] Figure 11 This is a compressive stress cloud diagram of the seat filling in an embodiment of the present invention;

[0022] Figure 12 This is an example diagram illustrating the measurement of seat parameters in an embodiment of the present invention;

[0023] Figure 13 This is a diagram illustrating the optimization process of the response surface model in an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Currently, autonomous vehicles cannot completely avoid traffic accidents, and passive safety technologies still need continuous improvement to provide more comfortable safety protection for drivers. Traditional restraint systems distribute the load unevenly on the driver; under certain collision conditions, the restraint system cannot effectively restrain the driver's movement, thus causing additional injury; unreasonable seat stiffness can cause discomfort or additional injury to the driver. Therefore, existing passive safety technologies are practically unable to provide comfortable safety protection for drivers through restraint systems alone. Based on this, this invention provides a seat parameter optimization method based on driver comfort, which can alleviate the above-mentioned problems in related technologies.

[0026] See Figure 1 As shown, the seat parameter optimization method based on driver comfort may include the following steps:

[0027] Step S102: Obtain the vehicle finite element model, and configure the dummy model, as well as the pre-established seat belt model, airbag model, and seat model for the vehicle finite element model to establish the driver restraint system finite element model.

[0028] The seat model features leg support and an adjustable mechanism.

[0029] Step S104: Determine the seat parameters, including the seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness, and establish an orthogonal experimental table with the seat parameters as experimental factors.

[0030] Step S106: Based on the orthogonal experimental table and the finite element model of the driver constraint system, a vehicle collision simulation is performed to obtain the simulation results of seat comfort.

[0031] The simulation results for seat comfort can include output values ​​corresponding to multiple body pressure distribution indicators.

[0032] Step S108: Based on the seat comfort simulation results, construct a response surface model for each body pressure distribution index corresponding to the seat parameters, and iteratively solve the response surface model based on the seat comfort simulation results to obtain the optimized results of the seat parameters.

[0033] Among them, the response surface model can characterize the relationship between seat parameters and body pressure distribution indicators.

[0034] This invention provides a method for optimizing seat parameters based on driver comfort. It can establish a finite element model of the driver constraint system and an orthogonal experimental table based on the vehicle finite element model and seat parameters, and construct a response surface model through vehicle collision simulation. Then, iteratively solving the response surface model can optimize the seat parameters, thereby improving seat comfort while ensuring driver safety.

[0035] As one possible implementation method, the dummy model can be the 50th percentile dummy of Hybrid III and the 50th percentile dummy of THUMS; based on this, the above-mentioned seat parameter optimization method based on driver comfort can also include: establishing a seat belt model, an airbag model and a seat model.

[0036] The Hybrid III series dummies can simulate injuries to the head, neck, chest, pelvis, thigh, and lower leg, while the THUMS series dummies can simulate fractures, ligament ruptures, brain injuries, and internal organ injuries. Due to structural and external differences between the Hybrid III and THUMS dummies, particularly in the hip, leg, and back areas, the THUMS dummies have a more human-like surface. This difference results in a more continuous contact surface when the THUMS dummies come into contact with the seat. Therefore, the Hybrid III 50th percentile dummies are primarily used to assess driver injury, while the THUMS 50th percentile dummies are mainly used to assess driver comfort.

[0037] As one possible implementation, configuring the dummy model and the pre-established seat belt model, airbag model and seat model for the vehicle finite element model in step S102 may include: dividing the vehicle finite element model into the cockpit and embedding the dummy model, seat belt model, airbag model and seat model into the cockpit.

[0038] A published finite element model of a certain type of sedan can be obtained, and its structural integrity verified. The PRIMER software can be used to create seatbelt and airbag models, as well as a seat model with leg support and an adjustable mechanism. The dummy's posture was adjusted through pre-simulation. The dummy model, seatbelt model, and seat model were then placed within a trolley model of the driver's side restraint system of the sedan. This model segmented the cockpit and impacted a rigid barrier head-on at 56 km / h. In the simulation, a high-precision THUMS dummy (i.e., the 50th percentile THUMS dummy) was used. Displacement was applied to the THUMS dummy to achieve joint rotation, thereby adjusting the THUMS dummy to the predetermined posture. The posture of the Hybrid III dummy (i.e., the 50th percentile Hybrid III dummy) can be adjusted by adjusting the joint angles of its limbs in the Ls-PrePost software. After adjusting the dummy's posture, the finite element model of the driver restraint system is obtained. The driver restraint system finite element model can be used to conduct vehicle collision simulation experiments. Then, according to the collision restraint evaluation criteria, the output values ​​of the simulation experiment are compared with the damage to various parts of the occupants in the actual vehicle test. Some parameters in the driver restraint system finite element model are also output to determine whether the initial values ​​of the driver restraint system finite element model are reasonable, thereby verifying the effectiveness of the driver restraint system finite element model.

[0039] As a crucial component of the restraint system, the seatbelt model is designed and implemented to effectively protect the driver in the event of a collision. The specific steps for establishing and simulating the seatbelt model are as follows:

[0040] Step 1.1, Modeling and Meshing: First, a finite element model of the seatbelt is created in PRIMER software. The finite element model of the seatbelt consists of 2D shell elements and 1D seatbelt elements, used to simulate the contact and interaction between the seatbelt, the dummy's body, and the slip ring. Specifically, the part of the seatbelt in contact with the dummy uses 2D shell elements, while the part of the seatbelt near the slip ring uses 1D seatbelt elements, which can accurately simulate the tensioning and sliding process of the seatbelt.

[0041] Step 1.2, Seatbelt Path and Positioning: During the modeling process, it is necessary to set the wrapping path of the seatbelt and ensure that the contact points between the seatbelt and the dummy are accurately positioned. For example, the positions of the retractor and slip rings, and the seatbelt contact points on the dummy's shoulders, chest, abdomen, and hips, all need to be set according to the actual situation.

[0042] Step 1.3, Define the physical properties of the seat belt: The material properties of the seat belt are simulated using a material model (such as the *MAT_34 model), which can accurately reflect the mechanical behavior of the fabric material. Simultaneously, the coefficient of friction of the seat belt is set to 0.15 at the slip ring to simulate the frictional force between the seat belt and the slip ring.

[0043] Step 1.4, Seatbelt Pretensioning and Force Limiting: The force limiting function of the seatbelt is implemented by setting the keyword *ELEMENT_SEATBELT_RETRACTOR, which limits the tensile force of the seatbelt; the pretensioning function of the seatbelt is implemented using the keyword *ELEMENT_SEATBELT_PRETENSIONER. By adjusting these settings, it is ensured that the seatbelt can effectively restrain the driver in the event of a collision.

[0044] Airbags are another important occupant protection device. The airbag modeling and simulation process includes the following steps:

[0045] Step 2.1, establish the initial geometric model of the airbag: First, establish the initial geometric model of the airbag in the PRIMER software, set the airbag to be circular, generate two layers of mesh, and add two exhaust holes to the airbag.

[0046] Step 2.2, Airbag Folding: The airbag is folded using the *Thin Fold method in PRIMER software. After each fold, the mesh near the fold line needs to be re-remodeled (remesh) to ensure that the geometry and structure of the folded airbag adapt to the collision conditions.

[0047] Step 2.3, Inflation Simulation and Parameter Setting: In HyperMesh, the pressure equalization method is used to simulate the airbag inflation process, and the mass flow curve of the airbag generator is defined. Simultaneously, the curve showing the change in vent area with absolute pressure is set. These parameters effectively simulate the airbag inflation and deployment process during a collision.

[0048] Airbag contact definition with other components: To ensure good contact between the airbag and the steering wheel and occupants, the keywords *CONTACT_AIRBAG_SINGLE_SURFACE and *CONTACT_AUTOMATIC_SURFACE TO SURFACE are used to define the contact between the airbag and other components. The coefficient of friction is set to 0.35 to ensure that the friction behavior between the airbag and the driver or steering wheel is correctly simulated when the airbag deploys.

[0049] For example, the step of embedding the Hybrid III 50th percentile dummy into the cockpit may include: placing the Hybrid III 50th percentile dummy in the cockpit and setting the initial sitting posture of the Hybrid III 50th percentile dummy in the cockpit according to a preset crash test specification; wherein, the initial sitting posture causes the torso of the Hybrid III 50th percentile dummy to be in contact with the seat back, while keeping the relative position of the Hybrid III 50th percentile dummy's head and the reference point unchanged.

[0050] For example, the step of embedding the THUMS 50th percentile dummy into the cockpit may include: placing the THUMS 50th percentile dummy in the cockpit and setting the THUMS 50th percentile dummy's sitting posture in the cockpit by applying displacement to the THUMS 50th percentile dummy.

[0051] For example, the reference point can be the front door latch mounting point of the vehicle body; the initial seating position is determined by multiple positioning parameters of the 50th percentile dummy of Hybrid III; as shown in Table 1, the multiple positioning parameters may include: the distance from the dummy's head to the roof, the distance from the dummy's head to the windshield, the distance from the dummy's nose to the edge of the steering wheel, the distance from the dummy's chest to the dashboard, the distance from the dummy's chest to the center of the steering wheel, the distance from the dummy's knee joint to the reference point, the distance from the dummy's head to the reference point, and the distance from the dummy's H point to the reference point.

[0052] Table 1 Comparison of dummy positioning parameters in simulation and experiment

[0053]

[0054] In collision simulations, dummy models play a crucial role in simulating occupant reactions and injuries. The relevant implementation steps for using dummy models in collision simulations are as follows:

[0055] Step 3.1, Dummy Positioning and Seating Posture Adjustment: According to the crash test specifications, the dummy must first be accurately positioned. Using the front door latch mounting point as a reference point, adjust the relative position between the dummy's H-point and the door lock to ensure that the dummy's torso and seat back are in close contact, while also ensuring a reasonable positional relationship between the head and the door lock. The dummy's joints (such as shoulders, hips, and knees) need to be adjusted according to the predetermined seating posture parameters.

[0056] Step 3.2, Defining the Contact Between the Dummy and Vehicle Interior Components: Finite element methods are used to define the contact between the dummy and components such as the seat, steering wheel, and airbags. The contact between the dummy and vehicle interior components should simulate their motion response and potential collision damage. In the simulation, the dummy assesses injury through contact with components such as the airbag and steering wheel.

[0057] Step 3.3, Use of THUMS Dummy: Because the surface of the THUMS dummy closely matches the actual human body structure, it can be used for more detailed injury simulations, such as simulating internal injuries (e.g., fractures, internal organ damage), thereby improving simulation accuracy. The THUMS dummy requires additional displacement to adjust the joints and simulate the correct sitting posture.

[0058] In crash simulation, the entire restraint system (including seat belt model, airbag model, dummy, etc.) needs to be placed into a sled model simulating the passenger compartment for analysis. The specific steps are as follows:

[0059] Step 4.1, Model Division and Assembly: As needed, the front passenger compartment of the vehicle is divided into quarters, retaining interior components such as the steering wheel, brake pedal, driver's side dashboard, and seats. Simultaneously, seat belts, airbags, and dummy models are embedded into the passenger compartment to simulate a collision.

[0060] Step 4.2, Collision Simulation: In the simulation, by setting the simulation conditions, the collision process will automatically trigger the seat belt restraints, airbag deployment, and dummy motion response. This can effectively evaluate the safety and comfort of the restraint system in protecting the driver.

[0061] Through steps 4.1 and 4.2 above, the seat belt model, airbag model, and dummy model can be integrated into a complete crash simulation system to evaluate the effectiveness of the restraint system under crash conditions.

[0062] like Figure 2As shown, the finite element method (FEM) can be used to analyze collision problems. The implementation process of the FEM is divided into three stages: preprocessing, solution, and post-processing. In the preprocessing stage, software such as HyperMesh, PRIMER, and LS-PrePost are used to establish the finite element model, complete mesh generation, element type definition, material property settings, and boundary condition loading, ensuring model accuracy and solution effectiveness. In the solution stage, LS-DYNA is used for collision analysis, handling large displacement, large rotation, and nonlinear material problems, suitable for explicit dynamic simulation. In the post-processing stage, HyperView and HyperGraph are used to generate simulation reports for optimization analysis. Specifically, HyperView visualizes the collision process, and HyperGraph plots the result curves, supporting design optimization. Each stage is closely integrated to ensure the accuracy of the simulation model and the reliability of the results. Automobile collision analysis is a typical structural simulation problem handled by the FEM, and it is usually carried out in three stages: preprocessing, solution, and post-processing. First, in the preprocessing stage, engineers use software such as HyperMesh to mesh the model, dividing it into small finite element elements, each with a specific shape and size to ensure computational accuracy. Simultaneously, material properties (such as elastic modulus and density) are defined, assigning appropriate materials to each element, and setting loads (such as collision force and pressure) and boundary conditions (such as fixed end and symmetry conditions). For example, in car crash simulations, different materials may be used for the car body and the dummy, and the collision location and initial velocity may need to be defined. Next, in the solution stage, a suitable solver (such as LS-DYNA) is selected to dynamically solve the crash model. The solver performs iterative calculations based on the problem type (such as nonlinear, large deformation, dynamic analysis, etc.) to obtain the stress, strain, and displacement distributions at various time points. Finally, in the post-processing stage, software such as HyperView is used to extract the solution results (such as maximum deformation, stress distribution, energy loss, etc.) and generate a results report to help engineers analyze structural performance. If some components are damaged or deformed excessively during a collision, engineers can also optimize the model based on the simulation results, such as by changing materials, reinforcing certain areas, or adjusting the load distribution to improve the safety and performance of the structure.

[0063] like Figure 3 As shown, a three-dimensional data model of the entire vehicle was reconstructed using reverse engineering techniques based on a finite element model of a certain type of sedan developed by the Center for Crash Safety and Analysis (CCSA) at George Mason University in the United States. Material information for each component was also obtained to ensure the consistency of the mechanical properties of the three-dimensional data model with the actual sedan. This three-dimensional data model contains 1086 components, including the body structure, engine, chassis, and seats, with a total of 2,255,361 nodes and 2,257,280 elements, capable of supporting complex crash simulation analyses.

[0064] In crash tests, dummies serve as crucial surrogate subjects simulating occupant injury, and their initial sitting posture significantly influences the motion response and severity of damage during a collision. Therefore, the accuracy of the dummy's positioning is fundamental to ensuring precise experimental results. Figure 4 As shown, referring to the NCAP test procedures, the front door latch mounting point is used as a reference point, and the dummy's H-point (the connection point between the torso and thigh in a two-dimensional or three-dimensional human model, representing the position of the midpoint of the driver's hip joint inside the vehicle after the driver is seated) is used as the benchmark for driver positioning. The dummy positioning steps mainly include: first, confirming the relative position of the dummy's H-point and the door lock, and adjusting the dummy's pelvic angle to make the dummy's torso fit against the seat back, while ensuring that the relative position of the dummy's head and the door lock remains consistent. Through these steps, the accurate positioning of the dummy in the finite element model is ensured, thereby improving the simulation accuracy of the crash test. Figure 4 The symbols and definitions of the positioning parameters for each dummy in Table 1 are also shown.

[0065] like Figure 5 As shown, the Hybrid III dummy positioning can be adjusted by referring to the dummy positioning parameters in the NCAP test procedure to adjust the dummy's sitting posture. Figure 5 The relative distances between the driver's body surface positioning point and the vehicle body and interior positioning points in the simulation model are shown. In crash tests, the dummy's initial seating posture significantly influences its motion response and final injury during the collision. Therefore, the dummy's positioning accuracy is fundamental to ensuring the accuracy of experimental test results. The driver's positioning process can be simulated in a finite element model using the dummy positioning parameters in the NCAP test procedures. First, using the front door latch mounting point as the vehicle body reference point and the dummy's H-point as the driver reference point, the relative position between the H-point and the door lock is confirmed. By adjusting the dummy's pelvic angle, the dummy's upper body is aligned with the seat back, while ensuring the dummy's head remains in the same relative position to the door lock. In the Ls-Prepost software, the angles of the dummy's various limb joints can be adjusted to achieve predetermined positions, including the shoulders, arms, wrists, thighs, knees, calves, and ankles. Finally, the relative distances between the dummy positioning points in the simulation model and the positioning points of the vehicle body and interior are shown in Table 1. The results show that the deviation between the dummy positioning parameters in the simulation and the dummy positioning parameters in the experiment does not exceed 5%, which verifies the consistency between the driver's sitting posture in the simulation and the driver's sitting posture in the experiment.

[0066] like Figure 6As shown, the THUMS dummy pre-simulation process mainly involves using CT scan data to generate a digital full-body model and constructing a finite element mesh based on this full-body model. The specific steps of the THUMS dummy pre-simulation can include: (1) CT scan data acquisition: First, a high-resolution CT scan device is used to scan the human body to obtain accurate scan data. This scan data includes three-dimensional images of various parts of the human body, which can accurately display the shape and position of bones, muscles, internal organs, etc. (2) Digital model construction: The scan data obtained from the CT scan is processed and reconstructed using medical image processing software to form a complete digital human body model. The digital human body model includes the geometry, tissue distribution, and organ structure of the human body. (3) Finite element mesh generation: Finite element analysis software (such as HyperMesh, Abaqus, etc.) is used to convert the digital human body model into a finite element mesh. The finite element mesh generation process includes dividing the entire three-dimensional model into many small units with specific physical properties. These units can simulate the physical characteristics of various parts of the human body, such as the stiffness of bones and the flexibility of tissues. Using finite element meshes, the THUMS dummy can perform detailed mechanical analyses of different parts of the human body (such as fractures, brain injuries, and internal organ damage) in collision simulations. These analyses not only accurately simulate the motion of the human body during a collision but also delve into the tissue level to precisely simulate various types of injuries. Therefore, this pre-simulation process of the THUMS dummy combines high-resolution CT scans and digital modeling to ensure that the THUMS dummy can realistically reproduce human structure and response in simulations, thus providing more accurate collision simulations and damage predictions than traditional physical dummies.

[0067] However, the THUMS dummy lacks internal mechanical structures and cannot simulate joint rotation through hinged mechanisms like the Hybrid III dummy. To address this issue, additional displacement can be applied to adjust the THUMS dummy to a predetermined sitting posture. Specifically, such as... Figure 6 As shown, the parts of the dummy that do not need adjustment can be fixed on a rigid body support, while the joints of the dummy that need to rotate are connected through 1D units and displacement is applied to achieve the rotation of the joints by utilizing the dummy's own contact behavior.

[0068] The rigid support, composed of rigid elements (such as beam or rigid body elements), is used to fix and support the dummy model. It can be defined and connected to the dummy model using finite element software (such as HyperMesh), restricting the relative motion between the support and the dummy. 1D elements, typically rod or spring elements, are used to connect joints that need to rotate. 1D elements can be placed at joint locations (such as the knee or hip joints) by defining appropriate connection points and orientations to effectively simulate joint rotation or relative displacement. When constructing 1D elements, ensure that the orientation of the 1D elements is consistent with the direction of joint rotation, and set appropriate element types (such as springs or rods) to simulate the elasticity and motion of the joint. When applying displacement to the THUMS dummy model, the magnitude of the displacement needs to be determined based on the predetermined sitting posture, usually selected based on the simulation purpose or experimental data; for example, the knee joint displacement might be 5°, 10°, or greater, adjusted according to the sitting angle. Displacement is usually applied by applying boundary conditions or loads, which can be achieved in finite element software through "displacement control" or "kinematic control." The displacement application process involves gradually controlling the magnitude of the displacement to bring the joint to the target angle. During the calculation, a solver (such as LS-DYNA) is used for nonlinear static or dynamic analysis. Displacement application at the dummy joints is typically performed on a time-step basis. In each calculation step, the applied displacement guides the rotation of the dummy model's joints. For example, for the knee joint, assuming an applied displacement of 10mm or 5°, the force required to achieve this displacement is calculated based on the model's stiffness, mass distribution, and external loads, ensuring the dummy model can correctly simulate joint rotation. The simulation of the dummy's own contact behavior can be achieved by defining contact surfaces (such as soft contact surfaces at the joints). Specifically, contact elements can be defined during the calculation to capture the interaction between contact forces and contact points. Appropriate contact algorithms (such as the penalty function method or the Lagrange multiplier method) can be used to simulate the contact behavior at the joints, ensuring that the joints rotate naturally when displacement is applied.

[0069] To reduce the computational burden of the finite element model, the complete vehicle model can be segmented to retain the cockpit. Interior components such as the steering wheel, brake pedal, driver's side dashboard, and seats are preserved within the vehicle model. Then, seatbelt models, airbag models, and dummy models are placed inside the cockpit to perform vehicle collision simulations and evaluate the safety of the restraint system. After configuring the seatbelt, airbag, and dummy models, a vehicle collision simulation is conducted. During the simulation, a collision event is simulated, and the forces exerted on the occupants, the degree of injury, and the protective effectiveness of the seatbelts and airbags are observed.

[0070] like Figure 7As shown, to further evaluate the model's collision performance, the NCAP frontal collision test procedure was used for simulation, and the results were compared with real vehicle test data. Figure 7 The simulation results illustrate the driver's motion response at different times during real-vehicle and simulation tests. In the simulation, the vehicle was designed to collide head-on with a rigid barrier at a certain speed (56 km / h), with the accelerometers installed in the same positions as in the NCAP test. The simulation results show that the deformation of the vehicle's front end in the simulation is highly consistent with the deformation in the real-vehicle test, verifying the high accuracy of the established finite element model of the driver constraint system in predicting collision behavior.

[0071] When simulating a frontal crash test using the NCAP procedure, the simulation input data includes initial conditions, crash parameters, airbag parameters, and time step. Initial conditions include the initial moment of the collision (t = 0 ms). At this initial moment, the vehicle contacts the rigid barrier, and the dummy and airbags are in a static state before the collision. Crash parameters can include the collision velocity (e.g., vehicle collision velocity, collision angle) and relevant characteristics of the rigid barrier (e.g., the elasticity, shape, and size of the rigid barrier). Airbag parameters can include airbag deployment time, inflation pressure, and inflation rate. The simulation time step needs to be set small enough to ensure that the simulation accurately tracks the dummy's motion response at each moment. During the simulation, the dummy's motion response is affected by various mechanical constraints during the collision, specifically at the following time points:

[0072] t = 0 ms: The initial moment of the collision, when the dummy and vehicle in the simulation model come into contact with the barrier and begin to accelerate.

[0073] t = 22 ms: The airbag begins to deploy from the airbag chamber; at this point, the dummy's head shows no significant displacement. The simulation will model the airbag inflation process and record the force exerted on the dummy's head by changes in air pressure.

[0074] t = 64ms: The driver's head makes contact with the airbag. At this moment, the simulation calculates the impact force between the head and the airbag and updates the kinematic data of the dummy's head, such as acceleration and angular velocity.

[0075] t = 110ms: The airbag deflates, and the steering column collapses. At this point, the simulation system begins to calculate the reaction force of the steering wheel deformation on the dummy, while also considering the dynamic effects of the airbag deflation process on the dummy.

[0076] Simulation output data includes motion response data, damage values, and time history graphs. Motion response data primarily includes changes in physical quantities such as acceleration, displacement, and velocity of the dummy's head, chest, and lower limbs. Analyzing this data allows us to determine whether the dynamic response in the simulation is consistent with that in an actual crash test. Damage values ​​are calculated based on the stress and strain of various parts of the dummy and can include chest compression and head impact force. Damage values ​​help determine the severity of injury to the dummy during a collision and can be compared with damage data from actual tests. The time history graphs in the simulation allow us to analyze the motion changes of various parts, such as head acceleration and chest displacement.

[0077] like Figure 8 As shown, the damage curves for various parts of the occupant's body were output in accordance with the NCAP guidelines. Figure 8 (a) shows the head acceleration curve. Figure 8 (b) shows the chest acceleration curve. Figure 8 (c) shows the neck shear force curve. Figure 8 (d) in the figure shows the neck tension curve. Figure 8 (e) in the figure shows the neck bending moment curve. Figure 8 Figure (f) shows the axial force curve of the thigh. Based on... Figure 8 The following conclusions can be drawn: The peak phase of the head acceleration curve in the simulation test is consistent with that in the actual vehicle test, but the peak value of the head acceleration curve in the simulation test is slightly higher than that in the actual vehicle test; The growth trend of the chest acceleration curve in the simulation test and the actual vehicle test is similar, but the peak value of the chest acceleration curve in the simulation test is slightly lower than that in the actual vehicle test; The peak values ​​and trends of the neck injury curves (including neck shear force curve, neck tension curve, and neck bending moment curve) in the simulation test and the actual vehicle test are similar, but the phase of the corresponding peak value of the neck injury curve in the simulation test is later than that in the actual vehicle test; The peak values ​​and trends of the thigh axial force curve in the simulation test and the actual vehicle test are similar, but the phase difference of the peak value of the thigh axial force curve in the simulation test and the actual vehicle test is large, which may be related to the initial position of the thigh.

[0078] In addition, the finite element model of the driver restraint system also needs to output some parameters of the restraint system to check whether the initial values ​​of the restraint system parameters in the finite element model of the driver restraint system are reasonable. For example Figure 9 As shown, the finite element model of the driver restraint system outputs the seatbelt retractor webbing in / out curve, the seatbelt force curve at the retractor exit, and the seatbelt lap belt force curve. Based on... Figure 9It can be seen that the variation trends of the seat belt retractor webbing inlet / outlet curve and the seat belt force curve at the retractor outlet in the simulation test are consistent with the actual situation. The peak values ​​and trends of the seat belt lap belt force curves in the simulation test and the actual vehicle test are similar, indicating that the initial values ​​of the seat belt parameters in the finite element model of the driver restraint system are relatively reasonable.

[0079] like Figure 10 As shown, the effectiveness of the driver restraint system finite element model can be judged by outputting the contact force generated between the driver and the seat on the contact surface. These contact forces include: the contact force between the driver's buttocks and the seat cushion, the contact force between the driver's back and the seat backrest, and the contact force between the driver's legs and the seat leg support. Figure 10 The curves showing the change of the magnitude of each of the three contact forces over time in the simulation experiment are shown.

[0080] like Figure 11 As shown, the effectiveness of the driver restraint system finite element model can be determined by outputting the compressive stress cloud diagram of the seat padding from the driver restraint system finite element model. Figure 11 (a) in the figure is a compressive stress cloud diagram of the seat back, which is the part of the seat that supports the back. Figure 11 (b) in the figure is a compressive stress cloud diagram of the seat cushion, which supports the buttocks and thighs; Figure 11 (c) in the diagram shows the compressive stress contour of the seat leg support, which supports the lower leg or foot. Based on... Figure 11 As can be seen from the stress distribution, the compressive stress of the seat filling is symmetrically distributed from left to right, and the compressive stress distribution of the seat filling shows a distribution pattern of diffusion from the center to the outside. This stress distribution is reasonable.

[0081] Overall, the trends and peak values ​​of the driver's head injury curve, chest injury curve, neck injury curve, and thigh injury curve were basically consistent in both the simulated crash test and the real vehicle crash test. The parameters output by the restraint system were normal, and the contact force generated by the driver on the seat and the compressive stress generated by the seat padding were reasonable, indicating that the established finite element model of the driver restraint system was effective.

[0082] like Figure 12As shown, seat parameters can include the backrest angle (θ1), seat cushion angle (θ2), and leg support angle (θ3), as well as the stiffness (K) of the seat padding. During the driver's seating process, the backrest, seat cushion, and leg support each generate different body pressures. Body pressure distribution indices mainly include maximum pressure (P), maximum pressure gradient (PG), average pressure (PA), pressure distribution uniformity (SPD), and contact area (A). These indices help evaluate the comfort and support of the seat. To facilitate data analysis and result interpretation, these seat parameters and body pressure distribution indices can be used for quantitative evaluation of seat performance. Specifically, orthogonal experiments can be designed using seat parameters as factors to obtain reliable data, and response surface models for each body pressure index are constructed using multinomial regression. Subsequently, the seat parameters are optimized based on the response surface models.

[0083] As shown in Table 2, a unified format can be used to represent the seat parameters and the body pressure distribution index of each part of the seat. This representation method facilitates the reading of subsequent orthogonal experimental results.

[0084] Table 2 Seat Parameters and Body Pressure Variables

[0085]

[0086] Table 3 shows that the experimental factors selected for the orthogonal experiment are the backrest angle (θ1), seat cushion angle (θ2), leg support angle (θ3), and stiffness (K) of the seat filling, and each of these four experimental factors is set with five levels.

[0087] Table 3. Factors and Levels of Orthogonal Experiment

[0088]

[0089] Table 4 shows L 25 (6 5 An orthogonal array is used, since there are four input variables (i.e., the backrest angle, seat cushion angle, leg support angle, and seat padding stiffness), based on the relationship between the number of basis functions N and the number of input variables n. It is known that at least 15 basis functions are needed to construct the response surface. To improve the representativeness of the data, the number of experiments required can be increased to 1.5 times the number of basis functions, meaning that at least 23 sample points are needed. SPSS analysis shows that L 25 (6 5 The orthogonal array perfectly meets the experimental requirements. Based on the selection principles of orthogonal arrays, an orthogonal array can be designed using the factors and levels listed in Table 3.

[0090] Table 5 shows the specific experimental arrangement table for the seat parameters (i.e., the orthogonal experimental table). As shown in Table 3, the orthogonal experiment obtained 25 sample points. These sample points include 4 input variables and 5 response values. The 4 factors and 5 levels in Table 3 were taken, and the variable parameters shown in Table 2 were input. The 4 input variables represent that the seat has 4 different adjustable parameters.

[0091] Table 4. Six-Factor Five-Level Orthogonal Array

[0092]

[0093] Table 5 Experimental Arrangement Table

[0094]

[0095] As one possible implementation, the orthogonal experimental table may include multiple input values ​​corresponding to the seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness; based on this, the above step S106 (i.e., performing vehicle collision simulation based on the orthogonal experimental table and the finite element model of the driver constraint system to obtain seat comfort simulation results) may include:

[0096] Step A1: Based on the input values, establish multiple input parameter groups for the seat model in the finite element model of the driver restraint system, and set the input data for the finite element model of the driver restraint system; the input data includes the initial time of the vehicle collision, vehicle collision parameters, airbag parameters, and simulation time step.

[0097] Step A2: Using the finite element model of the driver restraint system, vehicle collision simulation is performed according to each input parameter group using its input data to obtain the simulation results corresponding to each input parameter group; wherein, the simulation results include the output value corresponding to each body pressure distribution index and the corresponding input parameter group.

[0098] Continuing from the previous example, the 25 sets of sample points generated by the experimental arrangement table shown in Table 5 can be used as 25 sets of input parameters. After setting the input data of the driver restraint system finite element model, these 25 sets of input parameters can be input into the driver restraint system finite element model to perform vehicle collision simulation. The results of these 25 vehicle collision simulations can be obtained by performing 25 vehicle collision simulations through the driver restraint system finite element model.

[0099] Table 6 shows the output variables representing the body pressure distribution of the seat back, seat cushion, and leg rest obtained by inputting different variable parameters (i.e., 25 different sets of input parameters) into the finite element model of the driver restraint system.

[0100] Table 6. Results of Comfort Simulation

[0101]

[0102] As a possible implementation, before step S106 (i.e., performing vehicle collision simulation based on orthogonal experimental tables and driver constraint system finite element model to obtain seat comfort simulation results), the above-mentioned seat parameter optimization method based on driver comfort may further include: normalizing the input values ​​corresponding to the seat back angle, seat cushion angle, seat leg support angle and seat filling stiffness, respectively.

[0103] For example, each input parameter group may include the normalized values ​​of the seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness (such as the values ​​obtained after normalization of the values ​​in the 25 sample points shown in Table 5), or each input parameter group may include the input values ​​of the seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness (such as the values ​​in the 25 sample points shown in Table 5).

[0104] The dimensionless normalization method described above (i.e., normalizing the data for each seat parameter in Table 5 using the normalization method) does not change the data distribution pattern; it only scales the values, compressing all data into the range of 0 to 1 while ensuring the consistency of data characteristics. The normalization expression can be: C represents the normalized value, and x represents the original data value before normalization. min x is the minimum value of the original data. max This represents the maximum value of the original data.

[0105] As one possible implementation, the step S108 above, which constructs a response surface model for each body pressure distribution index corresponding to the seat parameters based on the seat comfort simulation results, may include: for each body pressure distribution index, based on the input parameter set and output value corresponding to the body pressure distribution index, using a multinomial regression method to fit the response surface function expression corresponding to the body pressure distribution index, and using the response surface function expression as the response surface model corresponding to the body pressure distribution index; wherein, the response surface function expression uses the seat parameters as independent variables and the body pressure distribution index as dependent variables.

[0106] The response surface function expressions corresponding to each body pressure distribution index can be fitted using the corresponding data in Tables 5 and 6 through multinomial regression.

[0107] For example, the response surface function expression of the contact area can be:

[0108]

[0109] The response surface function expression for maximum pressure can be:

[0110]

[0111] The response surface function expression for mean pressure can be:

[0112]

[0113] The response surface function expression for the maximum pressure gradient exponent can be:

[0114]

[0115] The response surface function expression for pressure distribution uniformity can be:

[0116]

[0117] Based on the five response surface function expressions above, it can be seen that each adjustment parameter of the seat has a different impact on the evaluation index (i.e., the body pressure distribution index), and the interaction between different seat adjustment parameters also affects the final evaluation index. The response surface function expressions actually represent the relationship between the body pressure distribution index and the seat parameters, which provides the foundation for subsequent algorithm optimization.

[0118] To determine the effectiveness of the response surface model, after normalizing the data in Tables 5 and 6 and fitting the response surface function expressions corresponding to each body pressure distribution index (i.e., the five response surface function expressions mentioned above) using multinomial regression, the coefficient of determination and adjusted coefficient of determination for each body pressure distribution index's response surface model can be calculated. Table 7 shows the coefficients of determination R in each response surface model after normalization. 2 and the adjusted coefficient of determination R Adj 2 Coefficient of determination R 2 and the adjusted coefficient of determination R Adj 2 The closer the value is to 1, the better the model fit. According to Table 7, among all response surface models, the response surface models for maximum pressure and pressure distribution uniformity performed poorly, which may be due to the instability of the response value of one of the outputs. Overall, the coefficient of determination of all response surface models exceeded 0.8, indicating that all response surface models met the requirements of response surface analysis and proved that the established response surface models are effective.

[0119] Table 7. Example table of coefficient of determination and adjusted coefficient of determination.

[0120]

[0121] In general, dimensionless processing (i.e., normalization) of the data is essential, as it solves the problem of the difficulty in comprehensively analyzing the numerical values ​​of various indicators. The dimensionless processing method using normalization satisfies both the requirement of simplicity in the data processing process and the requirement of reliable data results. The effectiveness of the established response surface model is verified based on the normalized results.

[0122] As one possible implementation, step S108 above, which iteratively solves the response surface model based on the seat comfort simulation results to obtain the optimized seat parameters, may include: performing a correlation analysis on the seat comfort simulation results and determining the target body pressure distribution index based on the correlation analysis results; determining the target response surface model corresponding to the target body pressure distribution index from the response surface model, and iteratively solving the target response surface model using a preset genetic algorithm to obtain the optimized results.

[0123] For example, the target body pressure distribution index may include contact area and maximum pressure; based on this, the above-mentioned step of iteratively solving the target response surface model using a preset genetic algorithm to obtain the optimization result may include: taking maximizing the contact area as the objective and setting the constraints of the seat parameters, iteratively solving the first target response surface model corresponding to the contact area using a preset genetic algorithm to obtain the first optimization result of the seat parameters; taking minimizing the maximum pressure as the objective and setting the constraints of the seat parameters, iteratively solving the second target response surface model corresponding to the maximum pressure using a preset genetic algorithm to obtain the second optimization result of the seat parameters; and determining the optimization result based on the first optimization result and the second optimization result.

[0124] In practical applications, the steps for determining the optimization result based on the first optimization result and the second optimization result may include: if each input parameter group includes the normalized values ​​corresponding to the seat back angle, seat cushion angle, seat leg support angle, and seat filling stiffness, then the first optimization result and the second optimization result are denormalized, and the denormalized results of the first optimization result and the second optimization result are taken as the optimization result; if each input parameter group includes the input values ​​corresponding to the seat back angle, seat cushion angle, seat leg support angle, and seat filling stiffness, then the first optimization result and the second optimization result are taken as the optimization result.

[0125] Continuing from the previous example, as shown in Table 8, the Pearson correlation coefficient table can be used to represent the strength of the correlation. By studying the correlation between these five response values, the relationship between one response value and other response values ​​can be found. Thus, the effect of optimizing the overall response value can be achieved by optimizing only one response value.

[0126] Table 8 Pearson Correlation Coefficients

[0127]

[0128] Table 8 shows that: contact area has a significant negative correlation with average pressure; contact area has a slight correlation with maximum pressure, maximum pressure gradient, and pressure distribution uniformity; and maximum pressure has a significant positive correlation with maximum pressure gradient and pressure distribution uniformity. Following the selection principle for influencing factors (i.e., selecting significantly correlated and less correlated factors from the five factors of contact area, average pressure, maximum pressure, maximum pressure gradient, and pressure distribution uniformity), the three evaluation indicators of average pressure (significantly correlated with contact area), maximum pressure gradient (significantly correlated with maximum pressure but slightly correlated with contact area), and pressure distribution uniformity (significantly correlated with maximum pressure but slightly correlated with contact area) can be removed. Contact area and maximum pressure can then be selected to establish the optimization objective function. For the contact area evaluation indicator, a higher value means more sufficient contact between the driver and the seat support surface. From the correlation between contact area and average pressure, it can be seen that an increased contact area results in lower average pressure. Therefore, the contact area can be set to its maximum value (using A). max The maximum pressure (denoted by P) is used as the optimization objective; an increase in the value of the evaluation index of maximum pressure indicates uneven pressure distribution on the seat support surface, which explains the significant positive correlation between maximum pressure, maximum pressure gradient, and pressure distribution uniformity. Therefore, the maximum pressure can be taken as the minimum value (denoted by P). min The optimization objective is represented by (θ1, θ2, θ3, and K). To ensure the convergence and reliability of the calculation results, after determining the optimization objective, constraints are set on the four independent variables: seat back angle (θ1), seat cushion angle (θ2), seat leg support angle (θ3), and seat padding stiffness (K), to obtain the following two response surface models:

[0129] A max =f(θ1,θ2,θ3,K)

[0130] P min =f(θ1,θ2,θ3,K)

[0131] A max The goal of optimization is to maximize the contact area, P min The objective of the optimization is to minimize the maximum pressure; the constraints for the two response surface models are: 20°≤θ1≤40°; 10°≤θ2≤20°; 50°≤θ3≤70°; 0.7≤K≤1.3.

[0132] like Figure 13As shown, after establishing the response surface models corresponding to the contact area and maximum pressure, the population size is set to 20, the number of generations to 50, and the crossover probability to 0.8. The NSGA-II genetic algorithm is used to solve the response surface models corresponding to the contact area and maximum pressure. After approximately 200 iterations, a feasible optimal solution appears. According to the characteristics of the genetic algorithm, this trend continues until the end of the iteration. Finally, the optimal solutions for both response surface models are obtained after 1001 iterations each. Because the data was previously normalized, therefore... Figure 13 The values ​​on the vertical axis do not represent the true values ​​of the corresponding indicators.

[0133] In general, response surface models corresponding to the contact area and maximum pressure can be established based on the Pearson correlation coefficient table, and the NSGA-II genetic algorithm can be used to iteratively solve the response surface models corresponding to the contact area and maximum pressure. Finally, the optimal solution is obtained after multiple iterations.

[0134] As shown in Table 9, the optimal solution obtained from the response surface model corresponding to the contact area is represented as optimal parameter 1, and the optimal solution obtained from the response surface model corresponding to the maximum pressure is represented as optimal parameter 2. The optimized parameter results are compared with the initial parameters of the constraint system. According to Table 9, after optimization, the seat adjustment angle becomes larger, and the seat padding stiffness becomes smaller.

[0135] Table 9 Comparison of Optimized Parameters and Initial Parameters

[0136]

[0137] Continuing the previous example, to verify the effectiveness of the seat parameter optimization method based on driver comfort, as shown in Table 10, the optimal parameters 1 and 2 shown in Table 9 can be substituted into the finite element model of the driver constraint system for simulation. The simulation results of the driver constraint system finite element model are then compared with the optimization results of the response surface model (i.e., the optimization objective function value corresponding to the optimal solution of the response surface model; specifically, the contact area value for optimal parameter 1 and the maximum pressure value for optimal parameter 2). Table 10 shows that the deviations between the simulation results and the optimization results of the two response surface models are 5.7% and 10.3%, respectively, both less than 15%.

[0138] Table 10 Comparison of Simulation Results and Response Surface Model Results

[0139]

[0140] Overall, the comparison between the two optimal parameters and the initial parameters of the constraint system shows that the optimized seat adjustment angle is larger and the optimized seat padding stiffness is smaller, which means that the driver will have a more comfortable experience in a reclined posture and a softer seat. The two optimal parameters were also simulated in the finite element model, which proved that the obtained response surface model is relatively accurate and that iterative solution of the response surface model achieved a certain optimization effect.

[0141] In summary, using the aforementioned seat parameter optimization method based on driver comfort to improve seat comfort is a feasible solution. Furthermore, this method provides important guidance for exploring the driver's seating posture before a collision.

[0142] Unless otherwise specifically stated, the relative steps, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0143] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0144] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0145] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for optimizing seat parameters based on driver comfort, characterized in that, include: A finite element model of a vehicle is obtained, and a dummy model, a pre-established seat belt model, an airbag model, and a seat model are configured for the vehicle finite element model to establish a driver restraint system finite element model; wherein, the seat model has leg support and an adjustable mechanism; The seat parameters were determined, including the seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness, and an orthogonal experimental table was established with the seat parameters as experimental factors. Based on the orthogonal experimental table and the finite element model of the driver constraint system, a vehicle collision simulation was performed to obtain the seat comfort simulation results; wherein, the seat comfort simulation results include the output values ​​corresponding to multiple body pressure distribution indicators; Based on the seat comfort simulation results, a response surface model is constructed for each body pressure distribution index corresponding to the seat parameter. The response surface model is then iteratively solved based on the seat comfort simulation results to obtain the optimized results of the seat parameter. The response surface model characterizes the relationship between the seat parameter and the body pressure distribution index. The orthogonal experimental table includes multiple input values ​​corresponding to the seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness. Based on the orthogonal experimental table and the finite element model of the driver constraint system, vehicle collision simulation is performed to obtain seat comfort simulation results, including: Based on the input values, multiple input parameter groups are established for the seat model in the finite element model of the driver restraint system, and the input data of the finite element model of the driver restraint system is set; wherein, the input data includes the initial time of the vehicle collision, vehicle collision parameters, airbag parameters, and simulation time step; The finite element model of the driver restraint system is used to perform vehicle collision simulations according to each input parameter group using the input data, and the simulation results corresponding to each input parameter group are obtained. The simulation results include the output values ​​corresponding to each body pressure distribution index and the corresponding input parameter group.

2. The seat parameter optimization method based on driver comfort according to claim 1, characterized in that, Based on the simulation results of seat comfort, the response surface model is iteratively solved to obtain the optimized results of the seat parameters, including: A correlation analysis was performed on the simulation results of the seat comfort, and the target body pressure distribution index was determined based on the correlation analysis results; The target response surface model corresponding to the target body pressure distribution index is determined from the response surface model, and the target response surface model is iteratively solved using a preset genetic algorithm to obtain the optimization result.

3. The seat parameter optimization method based on driver comfort according to claim 2, characterized in that, The target body pressure distribution indicators include contact area and maximum pressure; The target response surface model is iteratively solved using a pre-defined genetic algorithm to obtain the optimization results, including: With the goal of maximizing the contact area and the constraints of the seat parameters set, a preset genetic algorithm is used to iteratively solve the first target response surface model corresponding to the contact area to obtain the first optimization result of the seat parameters; With the goal of minimizing the maximum pressure and the constraints of the seat parameters set, a preset genetic algorithm is used to iteratively solve the second target response surface model corresponding to the maximum pressure, and the second optimization result of the seat parameters is obtained. The optimization result is determined based on the first optimization result and the second optimization result.

4. The seat parameter optimization method based on driver comfort according to claim 3, characterized in that, Based on the simulation results of the seat comfort, a response surface model is constructed for each body pressure distribution index corresponding to the seat parameters, including: For each body pressure distribution index, based on the input parameter set and output value corresponding to the body pressure distribution index, a multinomial regression method is used to fit the response surface function expression corresponding to the body pressure distribution index, and the response surface function expression is used as the response surface model corresponding to the body pressure distribution index; wherein, the response surface function expression uses the seat parameters as independent variables and the body pressure distribution index as dependent variables.

5. The seat parameter optimization method based on driver comfort according to claim 1, characterized in that, Configure the dummy model and pre-established seat belt, airbag, and seat models for the vehicle finite element model, including: The vehicle finite element model is segmented to create a cockpit, and the dummy model, seat belt model, airbag model, and seat model are embedded into the cockpit.

6. The seat parameter optimization method based on driver comfort according to claim 1, characterized in that, The dummy models used are the 50th percentile Hybrid III dummy and the 50th percentile THUMS dummy; the method further includes: establishing the seat belt model, the airbag model, and the seat model; Embedding the 50th percentile Hybrid III dummy into the cockpit includes: placing the 50th percentile Hybrid III dummy in the cockpit and setting the initial sitting posture of the 50th percentile Hybrid III dummy in the cockpit according to a preset crash test specification; wherein, the initial sitting posture causes the torso of the 50th percentile Hybrid III dummy to be in contact with the seat back, while keeping the relative position of the 50th percentile Hybrid III dummy's head and the reference point unchanged; Embedding the THUMS 50th percentile dummy into the cockpit includes: placing the THUMS 50th percentile dummy in the cockpit and setting the THUMS 50th percentile dummy's sitting posture in the cockpit by applying displacement to the THUMS 50th percentile dummy.

7. The seat parameter optimization method based on driver comfort according to claim 6, characterized in that, The reference point is the front door latch mounting point of the vehicle body; the initial seating posture is determined by multiple positioning parameters of the 50th percentile dummy of Hybrid III; the multiple positioning parameters include: the distance from the dummy's head to the roof, the distance from the dummy's head to the windshield, the distance from the dummy's nose to the edge of the steering wheel, the distance from the dummy's chest to the dashboard, the distance from the dummy's chest to the center of the steering wheel, the distance from the dummy's knee joint to the reference point, the distance from the dummy's head to the reference point, and the distance from the dummy's H point to the reference point.

8. The method for optimizing seat parameters based on driver comfort according to claim 3, characterized in that, Before obtaining the seat comfort simulation results by performing vehicle collision simulation based on the orthogonal experimental table and the finite element model of the driver constraint system, the method further includes: normalizing the input values ​​corresponding to the seat back angle, seat cushion angle, seat leg support angle and seat filling stiffness respectively. Each set of input parameters includes normalized values ​​for the seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness, or each set of input parameters includes input values ​​for the seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness.

9. The method for optimizing seat parameters based on driver comfort according to claim 8, characterized in that, Determining the optimization result based on the first optimization result and the second optimization result includes: If each input parameter group includes the normalized values ​​corresponding to the seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness, then the first optimization result and the second optimization result are denormalized, and the denormalized results of the first optimization result and the second optimization result are taken as the optimization result. If each input parameter group includes input values ​​corresponding to the seat back angle, seat cushion angle, seat leg support angle, and seat padding stiffness, then the first optimization result and the second optimization result are taken as the optimization result.

Images