A method for dynamically evaluating the crosswind stability of a vehicle

Abstract

The present application relates to the field of vehicle evaluation method, specifically relates to a kind of automobile crosswind stability dynamic evaluation method, comprising: the establishment target vehicle's automobile aerodynamics model and multibody dynamics virtual prototype model;Load instantaneous measured crosswind data, and calculate automobile transient aerodynamic force and aerodynamic moment;Load road information to calculate the vehicle transient dynamics response of multibody dynamics virtual prototype model;The aerodynamic load obtained and the dynamic response parameters are bidirectional data exchange in specified time step, and the calculation condition is updated in next time step, then the updated calculation condition is given to the fluid mechanics control equation and Lagrange equation, automobile aerodynamics and automobile dynamics dynamic coupling solution is carried out, the crosswind response data of vehicle changes with time is obtained, to dynamically evaluate the crosswind stability of car.The present application can truly reflect the change of automobile aerodynamic force and body motion posture under crosswind environment, and accurately evaluate the aerodynamic stability of high-speed automobile under crosswind environment.

Description

Technical Field

[0001] This invention relates to the field of vehicle evaluation methods, specifically to a dynamic evaluation method for vehicle crosswind stability. Background Technology

[0002] Crosswind stability of automobiles involves multiple disciplines, including automotive aerodynamics and multibody dynamics, and is a crucial indicator related to vehicle safety. A systematic evaluation of vehicle crosswind stability must consider the coupled effects of aerodynamic disturbances, vehicle path, and changes in vehicle attitude. Within this coupled system, the instability caused by wind loads and the instantaneous motion of the vehicle are the most significant factors jeopardizing driving safety. The main research methods for vehicle crosswind stability include road testing, wind tunnel testing, and numerical simulation.

[0003] With the development of computer technology, numerical simulation has been widely used in the study of crosswind aerodynamic characteristics and stability of automobiles. However, there are significant limitations in the modeling of crosswind stability of automobiles under aerodynamic loads in numerical simulations. The vehicle's trajectory is preset, ignoring the influence of instantaneous changes in vehicle attitude on the surrounding turbulent structure under crosswind conditions. In the dynamic numerical calculations involving stability, aerodynamic loads are obtained in advance through quasi-static numerical simulations or wind tunnel tests, fitted into spline curves, and then applied as boundary conditions to the dynamic model for calculation. This does not consider the changes in aerodynamic forces caused by the shift of the wind pressure center during vehicle movement, which is not entirely consistent with the actual driving conditions of automobiles, resulting in a significant deviation between the evaluation of crosswind stability and the actual situation. Summary of the Invention

[0004] The present invention aims to provide a dynamic evaluation method for crosswind stability of automobiles, which dynamically evaluates the crosswind aerodynamic characteristics and lateral stability of automobiles by considering the interaction between the turbulent motion around the vehicle body and the changes in the vehicle's driving posture.

[0005] The dynamic assessment method for vehicle crosswind stability in this scheme includes:

[0006] Step 1: Establish the aerodynamic model and multibody dynamics virtual prototype model of the target vehicle;

[0007] Also includes:

[0008] Step 2: Read the measured crosswind data based on the time series and apply it as a boundary condition to the fluid domain to calculate the aerodynamic load of the car aerodynamic model under the transient flow field. The aerodynamic load includes aerodynamic force and aerodynamic torque.

[0009] Step 3, a road model is established to simulate real road excitation, and a motion simulation is performed based on the road model, differential Lagrange equations are solved, dynamic response parameters of a multi-body dynamics virtual prototype model are calculated, and a vehicle motion state is updated;

[0010] Step 4, after the flow field and the vehicle motion reach a steady state in steps 2 and 3, a coupling calculation communication mode is established, the aerodynamic load in step 2 and the dynamic response parameters in step 3 are stored in a shared folder, and bidirectional data exchange is performed at a specified time step, and then the updated calculation conditions are assigned to the fluid mechanics control equation and the Lagrange equation for solving, so as to obtain side wind response data varying with time, which represents the side wind stability of the target vehicle.

[0011] Further, in step 2, the transient flow field of the target vehicle is calculated by the URANS method, and the aerodynamic load is calculated by solving the fluid mechanics control equation. The N-S equation for time averaging is solved by introducing the turbulent fluctuation velocity. After the Reynolds average is performed on the continuity equation and the momentum equation, the RANS equation set is obtained as follows:

[0012] ;

[0013] ;

[0014] wherein, is the fluid density; is time; , represents the Reynolds average velocity, and is respectively the component of the velocity in the orthogonal coordinate , . represents the Reynolds average pressure; represents the additional stress tensor after the turbulence averaging.

[0015] Further, in step 2, on the premise of the RANS equation set, the eddy viscosity is introduced to simulate the turbulence dissipation based on the Boussinesq assumption, the aerodynamic force and the aerodynamic moment of the automobile aerodynamics model are solved by using the turbulence model, the turbulence model uses the RNG model, and is expressed as:

[0016] ;

[0017] ;

[0018] wherein, and are the Prandtl numbers of and , = = 0.7194; constant =1.42, =1.68.

[0019] Furthermore, in step 3, the multibody mechanics virtual prototype model is represented by several components and constraint equations describing the vehicle's motion. Each component and constraint equation can be represented in the global coordinate system as follows:

[0020] ;

[0021] in, For generalized coordinate variables, For variables The first derivative, Represents the kinetic energy of the component. The force matrix, For a certain constraint secondary Constraint equations It is a Lagrange multiplier array;

[0022] Combining the constraint equations for each component and kinematic pair, we obtain the governing equations for the dynamics of the vehicle multibody system:

[0023] ;

[0024] in, Let be the constraint equation matrix.

[0025] Furthermore, in step 4, the motion of the target vehicle in the fluid domain is realized through a custom UDF program and dynamic mesh technology.

[0026] The vehicle motion parameters obtained from the virtual prototype model calculation will be used as the velocity of the model boundary movement and rotation in the calculation of the fluid dynamics control equations.

[0027] Furthermore, in step 4, the UDF program is used to implement data read and write functions. The UDF program reads the dynamic response parameters of the car by continuously scanning the shared folder and writes the aerodynamic force obtained from the simulation at that time step into the shared folder.

[0028] Furthermore, in step 4, the aerodynamic and torque data stored in the current time step are read through Matlab / Simulink, imported into the multibody mechanics virtual prototype model, the differential Lagrange equation is solved to update the vehicle motion state, and then the calculated vehicle motion parameters of the current time step are stored and assigned to the vehicle aerodynamic physical model to prepare for the solution of the next time step, until the calculation converges and the set simulation time is reached.

[0029] Furthermore, the time step of the fluid computational domain needs to be determined by the Courant number, the minimum grid size of the computational domain, and the fluid velocity.

[0030] Compared with existing technologies, the beneficial effects of this solution are:

[0031] This scheme utilizes a dynamic coupling analysis method, which considers the information interaction between instantaneous aerodynamic loads and real-time vehicle motion response during the simulation process. Under this method, unstable flow field information can be fully captured, and the vehicle's motion under real crosswind conditions can be reproduced, greatly improving the closeness of the simulation to the real scene and enhancing the accuracy of the simulation results in representing the real situation. Attached Figure Description

[0032] Figure 1 This is a flowchart of an embodiment of the vehicle crosswind stability dynamic evaluation method of the present invention;

[0033] Figure 2 This is a schematic diagram illustrating an embodiment of the vehicle crosswind stability dynamic evaluation method of the present invention;

[0034] Figure 3 This is a schematic diagram illustrating the iterative calculation principle of an embodiment of the vehicle crosswind stability dynamic evaluation method of the present invention. Detailed Implementation

[0035] The following detailed description provides further details on specific implementation methods.

[0036] Example

[0037] Dynamic assessment methods for vehicle crosswind stability, such as Figure 1 and Figure 2 As shown, it includes:

[0038] Step 1: Establish the aerodynamic model and multibody dynamics virtual prototype model of the target vehicle. The aerodynamic model is a refined 3D geometric model created using CATIA, followed by preprocessing and mesh generation using ANSA. The multibody dynamics virtual prototype model is created using Adams / car and includes subsystems such as front and rear suspensions, tire models, and body modules. By applying step forces of different magnitudes and directions at the center of wind pressure, the robustness of the multibody dynamics virtual prototype model under aerodynamic disturbances is verified, increasing the model's realism and improving the accuracy of subsequent dynamic evaluation.

[0039] Step 2: Read time-series measured crosswind data from any location and use it as the boundary condition for solving the external flow field of the vehicle aerodynamic model in ANSYS FLUENT. Using computational fluid dynamics (CFD) methods, solve for the aerodynamic loads experienced by the vehicle aerodynamic model under crosswind conditions. The aerodynamic loads include aerodynamic forces and moments in the x, y, and z directions. Specifically:

[0040] Solving the time-averaged Navier-Stokes equations, i.e., after taking Reynolds averages of the continuity and momentum equations, yields the RANS equations:

[0041] ;

[0042] ;

[0043] in, For fluid density; For time; , Representing the Reynolds average velocity, where velocities are on orthogonal coordinates. , The amount; Indicates the Reynolds average pressure; This represents the averaged additional stress tensor after turbulence.

[0044] Under the premise of the RANS equations, eddy viscosity is introduced based on the Boussinesq assumption to simulate turbulent dissipation, and RNG is selected. Turbulence model solves for aerodynamic forces and moments of the target vehicle aerodynamic model, RNG The model is represented as:

[0045] ;

[0046] ;

[0047] in, and They are and The Prandtl number, = =0.7194; constant =1.42, =1.68.

[0048] Using RANS equations and RNG The model solves for the aerodynamic forces and aerodynamic moments in the x, y, and z directions three times.

[0049] Step 3: Establish a road surface model to simulate real road surface excitation. Based on the road surface model, perform motion simulation using the MBD solver Adams / car to solve the differential Lagrange equations and calculate the vehicle dynamics response parameters of the multibody dynamics virtual prototype model. These dynamic response parameters include vehicle lateral velocity, lateral acceleration, yaw rate, lateral displacement, and yaw angle, and the vehicle motion state is updated. The multibody dynamics virtual prototype model is represented by several components and constraint equations describing vehicle motion. Each component and constraint equation in the global coordinate system is as follows:

[0050] ;

[0051] in, For generalized coordinate variables, for The first derivative, Represents the kinetic energy of the component. The force matrix, For a certain constraint secondary Constraint equations It is a Lagrange multiplier array;

[0052] Combining the constraint equations of each component and kinematic pair, we obtain the dynamic equations of the automobile multibody system as follows:

[0053] ;

[0054] in, Let be the constraint equation matrix.

[0055] By considering the constraints of each component and kinematic pair of the vehicle, it is possible to realize the physical simulation of the motion of the simulated vehicle model, making the simulation results closer to the actual vehicle driving process.

[0056] The dynamic response, namely vehicle lateral velocity, lateral acceleration, yaw rate, lateral displacement, and yaw angle, is obtained by solving the dynamic equations of the multibody system of the vehicle.

[0057] In steps 2 and 3, ANSYS Fluent and Adams / car first perform separate numerical simulations for a period of time. Once the flow field and vehicle motion reach a stable state, i.e., when the residual curves of the CFD simulation have good convergence and the aerodynamic and vehicle dynamic response parameter curves do not fluctuate drastically, the data transfer and coupling solution are then performed.

[0058] Step 4: The vehicle's motion in the fluid domain is achieved through a user-defined functionality (UDF) program and dynamic meshing. The UDF program primarily handles data reading and writing. It continuously scans the shared folder to read the vehicle's motion parameters (such as speed, lateral velocity, and yaw rate) and writes the aerodynamic loads (such as lateral force and yaw moment) obtained from the fluid simulation at that time step into the shared folder. The dynamic meshing allows the vehicle's aerodynamic physical model in the computational domain to move according to the lateral velocity and yaw rate read by the UDF program.

[0059] The fluid dynamics software ANSYS Fluent reads the car's motion parameters (such as speed, lateral velocity, yaw rate, etc.) through a UDF program at each time step. These motion parameters are then used as velocity boundaries for the car's aerodynamic physical model. Dynamic meshing technology is employed to specify the vehicle's physical model boundaries and update the mesh using the current time step's movement and rotation speeds to simulate the car's transient motion. Following step 2, the flow field Navier-Stokes equations and turbulence equations are solved, and the calculated aerodynamic forces and moments at the current time step are recorded and stored. Simultaneously, Matlab / Simulink scans a shared database to read the aerodynamic forces and moments data from the current time step and imports them into Adams / car. The multibody dynamics virtual prototype model is then introduced using Adams / car's Gforce element forces, and the DAE equations are solved jointly. The calculated vehicle motion parameters are written to the shared database, and the lateral velocity, lateral acceleration, yaw rate, and lateral displacement data for the next time step are updated. In the next time step, ANSYS Fluent again uses the UDF program to scan the updated vehicle motion data and solve the Navier-Stokes equations and turbulence equations. Once the calculation converges and the set simulation time is reached, the iteration loop exits. In step 4, both ANSYS Fluent and MATLAB exist in two states: a calculation state and a scanning state. The calculation state performs normal simulation calculations, while the scanning state uses a monitoring program to continuously scan and monitor the shared database, waiting for the simulation results from the other software. The simulation flow is as follows: Figure 3 As shown.

[0060] The UDF program automatically reads and calculates the parameters related to each time step, stores the parameters obtained from the simulation at each time step, and then exchanges data bidirectionally for recalculation. This enables real-time dynamic changes of the parameters assigned to the vehicle model simulation calculation, simulating the actual operating state of the vehicle and improving the accuracy of the simulation results.

[0061] In step 4, the transient time step set in ANSYS Fluent is always consistent with the time step of the Adams / car simulation. The transient time step set in ANSYS Fluent is determined by the Courant number, the minimum mesh size of the computational domain, and the fluid velocity.

[0062] In step 4, such as Figure 3 As shown, CFD simulation and MBD simulation output time-series data on vehicle aerodynamics, torque, lateral velocity, lateral acceleration, yaw rate, and lateral displacement, respectively, to dynamically evaluate the vehicle's crosswind aerodynamic characteristics and stability.

[0063] Existing studies on vehicle crosswind stability typically separate crosswind aerodynamic characteristics from stability, consider only unidirectional coupling, or simplify the entire vehicle into a simple model with a few degrees of freedom. The analytical method for dynamically solving vehicle aerodynamics and multibody dynamics proposed in this embodiment effectively addresses the coupling effects between the vehicle's external flow field, driving path, and body attitude changes, thus improving driving safety. Compared to existing methods, this approach saves significant manpower, financial resources, and time. Even without crosswind stability testing conditions, it realistically reproduces the vehicle's driving state in complex wind environments, quickly and effectively guiding the positive development of vehicle aerodynamics and chassis.

[0064] The above descriptions are merely embodiments of the present invention, and common knowledge regarding specific structures and characteristics is not elaborated upon here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the structure of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A dynamic evaluation method for vehicle crosswind stability, comprising: Step 1: Establish the aerodynamic model and multibody dynamics virtual prototype model of the target vehicle; Its characteristic is that it further includes: Step 2: Read the measured crosswind data based on the time series and apply it as a boundary condition to the fluid domain to calculate the aerodynamic load of the car aerodynamic model under the transient flow field. The aerodynamic load includes aerodynamic force and aerodynamic torque. Step 3: Establish a road surface model to simulate real road surface excitation, perform motion simulation based on the road surface model, solve the differential Lagrange equation, calculate the dynamic response parameters of the multibody dynamics virtual prototype model, and update the vehicle motion state. Step 4: After the flow field and vehicle motion reach a stable state in steps 2 and 3, a coupled computational communication mode is established. The aerodynamic loads in step 2 and the dynamic response parameters in step 3 are stored in a shared folder, and bidirectional data exchange is performed at a specified time step. The updated calculation conditions are then assigned to the fluid dynamics control equations and the Lagrange equations for solving, and the crosswind response data that characterizes the crosswind stability of the target vehicle as a function of time is obtained. The motion of the target vehicle in the fluid domain is realized through a custom UDF program and dynamic mesh technology; the vehicle motion parameters obtained by multibody mechanics virtual prototype model calculation will be used as the velocity of model boundary movement and rotation to participate in the calculation of fluid dynamics control equations; The UDF program is used to implement data read and write functions. The UDF program reads the dynamic response parameters of the car by continuously scanning the shared folder and writes the aerodynamic force obtained by the simulation at that time step into the shared folder.

2. The method for dynamic evaluation of crosswind stability of a vehicle according to claim 1, characterized in that: In step 2, the aerodynamic load is calculated and solved using the hydrodynamic control equations. The averaged NS equations are introduced when solving for turbulent fluctuating velocities. After Reynolds averaging of the continuity and momentum equations, the RANS equation set is obtained as follows: ； ； in, For fluid density; For time; , Representing the Reynolds average velocity, where velocities are on orthogonal coordinates. , The amount; Indicates the Reynolds average pressure; This represents the averaged additional stress tensor after turbulence.

3. The method for dynamic evaluation of crosswind stability of a vehicle according to claim 2, characterized in that: In step 2, under the premise of the RANS equations, eddy viscosity is introduced based on the Boussinesq assumption to simulate turbulent dissipation. The aerodynamic forces and moments of the vehicle's aerodynamic model are solved using a turbulence model. The turbulence model utilizes RNG. The model is represented as: ； ； in, and They are and The Prandtl number, = =0.7194; constant =1.42, =1.

68.

4. The method for dynamic evaluation of crosswind stability of a vehicle according to claim 1, characterized in that: In step 3, the multibody mechanics virtual prototype model is represented by several components and constraint equations describing the vehicle's motion. Each component and constraint equation can be represented in the global coordinate system as follows: ； in, For generalized coordinate variables, For variables The first derivative, Represents the kinetic energy of the component. The force matrix, For a certain constraint secondary Constraint equations It is a Lagrange multiplier array; Combining the constraint equations for each component and kinematic pair, we obtain the governing equations for the dynamics of the vehicle multibody system: ； in, Let be the constraint equation matrix.

5. The method for dynamic evaluation of crosswind stability of a vehicle according to claim 1, characterized in that: In step 4, the aerodynamic and torque data stored in the current time step are read through Matlab / Simulink, imported into the multibody mechanics virtual prototype model, the differential Lagrange equation is solved to update the vehicle motion state, and then the calculated vehicle motion parameters of the current time step are stored and assigned to the vehicle aerodynamic physical model to prepare for the solution of the next time step, until the calculation converges and the set simulation time is reached.

6. The method for dynamic evaluation of crosswind stability of a vehicle according to claim 5, characterized in that: The time step is determined by the Courant number, the minimum grid size of the computational domain, and the fluid velocity.

Images