Suspension system dynamic model calibration method based on optimization algorithm
By optimizing algorithms and automating simulation processes, the problems of low efficiency in suspension K&C performance calibration and parameter interaction effects were solved, achieving efficient and accurate suspension model calibration, and improving the prediction accuracy and design optimization effect of vehicle development.
Patent Information
- Application Number
- CN202511660416.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-13
- Publication Date
- 2026-02-27
AI Technical Summary
Existing suspension K&C performance calibration methods are inefficient, time-consuming, and subject to parameter interaction effects, leading to repeated calibration work and failures.
A suspension system dynamic model calibration method based on optimization algorithms is adopted. Data is obtained through suspension K&C tests to establish a whole vehicle dynamic model, perform parameter sensitivity analysis, and use software such as ADAMS and MATLAB to realize an automated simulation process, establish optimization functions, and automatically iteratively adjust suspension component parameters to achieve high-precision calibration.
It improves the efficiency and accuracy of suspension K&C model calibration, shortens the development cycle, ensures the consistency between simulation and test results, outputs high-precision suspension models, and provides a reliable foundation for vehicle development.
Smart Images
Figure CN121580703A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application relates to the technical field of automobiles, in particular to a suspension system dynamics model calibration method based on an optimization algorithm. BACKGROUND
[0002] Automobile dynamics simulation mainly includes suspension K&C characteristic simulation, whole vehicle handling stability simulation, whole vehicle ride comfort simulation and the like, can be used for predicting dynamics performance, simultaneously providing load boundary for structure fatigue failure mileage prediction, guiding design optimization improvement, and plays an important role in the whole vehicle development process.
[0003] The accuracy of automobile dynamics simulation depends on the consistency of relevant parameters (including hard points, bush stiffness, suspension elastic element stiffness, damper damping, tire dynamics model and the like) of the simulation model and the sample vehicle. In order to improve the accuracy of the suspension simulation model and better serve product development, the suspension simulation model will be calibrated according to the suspension K&C test data of the sample vehicle. The suspension K&C performance test data of the sample vehicle reflects the actual manufacturing, assembly state and performance achievement of the automobile. The K&C performance calibration of the suspension model helps to improve the accuracy of the simulation model and has important significance for the whole vehicle handling performance simulation.
[0004] Suspension K (Kinematic) characteristic, namely suspension kinematics characteristic, refers to the characteristic that the wheel plane and the wheel center point produce angular displacement and linear displacement changes due to the action of the guide mechanism in the vertical reciprocating movement of the wheel. Suspension C (Compliance) characteristic, namely suspension flexible kinematics characteristic, refers to the characteristic that the wheel plane and the wheel center point produce angular displacement and linear displacement changes due to the force and torque of the ground acting on the tire. The K&C performance calibration of the suspension includes suspension K characteristic calibration (including same direction wheel jump working condition, reverse wheel jump working condition and steering working condition) and suspension C characteristic calibration (including driving force working condition, braking force working condition, same direction return torque working condition, reverse return torque working condition, same direction lateral force working condition and reverse lateral force working condition).
[0005] The K characteristic of the suspension usually shows nonlinearity, that is, the wheel toe angle and camber angle, kingpin caster angle and camber angle and the like change nonlinearly with the vertical movement of the wheel. If the dynamics simulation can accurately simulate the nonlinear K characteristic of the suspension, the target of the K&C performance calibration cannot be set only as the slope of the simulation curve being close to the test data curve, but must be set as the simulation result curve being relatively consistent with the test result curve.
[0006] The existing K&C performance calibration method mainly is that engineers manually adjust suspension related parameters (including hard points, bush stiffness, leaf spring stiffness, etc.) based on the error of test and simulation, which is low in efficiency, long in period, and the influence of the adjusted parameters on the suspension K&C performance has interactive effect, that is, different adjusted parameters have the same or different trend on the influence of a K&C performance target object, causing repeated and failed calibration work, and increasing the difficulty of calibration work. Therefore, it is necessary to establish an optimization function by means of an optimization algorithm to carry out suspension K&C calibration and output an optimal matching scheme.
[0007] In view of the above problems, the present application provides a solution. SUMMARY
[0008] In order to overcome the above-mentioned defects of the prior art, embodiments of the present application provide a suspension system dynamics model calibration method based on an optimization algorithm to solve the problems raised in the background art.
[0009] To achieve the above object, the present application provides the following technical scheme:
[0010] A suspension system dynamics model calibration method based on an optimization algorithm, comprising the following steps:
[0011] S1: carry out a sample vehicle suspension K&C test, obtain test data of the suspension under each K&C condition, and perform polynomial fitting on the original test data; S2: establish a vehicle dynamics model based on suspension K&C assembly characteristics according to the sample vehicle parameters and the K&C performance test data obtained in step S1; S3: take the suspension K&C characteristics as variables, carry out test design (DOE) calculation of each condition of vehicle ride comfort and handling stability, and perform parameter sensitivity analysis based on the DOE data, and select the K&C characteristics sensitive to the vehicle performance as the calibration items; S4: establish a suspension rigid-flexible coupling dynamics model in ADAMS software, and write a secondary development script to realize the automation operation of model component parameter modification, each K&C condition simulation and result curve output; S5: based on the automatic simulation process of step S4, taking the model component parameters as variables, carrying out DOE calculation and sensitivity analysis of each K&C condition, and selecting the component parameters sensitive to each key K&C characteristic as the calibration variables; S6: based on the automatic simulation process of step S4, taking the coincidence of simulation and test K&C curves as the optimization target, establishing an optimization function, performing K&C model calibration solution, and outputting a high-precision suspension K&C model.
[0012] In a preferred embodiment, in the step S1, the suspension K&C working conditions include same direction wheel jump working condition, opposite direction wheel jump working condition, steering working condition, longitudinal force working condition, same direction back positive torque working condition, opposite direction back positive torque working condition, same direction side force working condition and opposite direction side force working condition; and the test data include suspension stiffness, toe angle change rate, camber angle change rate, kingpin caster angle change rate, kingpin camber angle change rate.
[0013] In a preferred embodiment, in the step S2, the sample vehicle parameters include sprung mass parameters, tire parameters, steering system parameters, powertrain and transmission system parameters, front and rear suspension parameters.
[0014] In a preferred embodiment, in the step S3, the suspension K&C characteristics of the vehicle dynamics model are modified by using MATLAB program and vscommand command, the automation operation of the simulation of the vehicle ride comfort and handling stability working condition, the output of the result curve and the post-processing of the result curve are realized.
[0015] In a preferred embodiment, in the step S4, the establishment of the suspension rigid-flexible coupling dynamics model includes: extracting the front and rear suspension hard point coordinates, bushing stiffness and related parameters such as leaf spring modeling, establishing the front and rear suspension rigid body dynamics model in ADAMS software; establishing the structural finite element model of the front and rear axle and the steering cross rod in HyperMesh software, solving the modal neutral file based on the modal synthesis method, and importing ADAMS to establish the front and rear suspension rigid-flexible coupling dynamics model.
[0016] In a preferred embodiment, in the step S4, the writing of the secondary development script includes: writing MATLAB program and CMD command script, realizing the automation operation of the modification of the part parameters such as bushing stiffness and hard point coordinates in the model, the simulation of each K&C working condition, the output of the result curve and the post-processing of the data by using ADAMS software.
[0017] In a preferred embodiment, in the step S5, the model part parameters include hard point coordinates and bushing stiffness.
[0018] In a preferred embodiment, in the step S6, the establishment of the optimization function includes: determining the suspension bushing stiffness and hard point coordinates as the optimization variables; determining the minimum sum of the Euclidean distance of the test and simulation result curves of the key K&C characteristics as the optimization target; and determining the Euclidean distance of the simulation and test of each K&C characteristic as the constraint condition, wherein the key K&C characteristics are determined by the step S3.
[0019] In a preferred embodiment, the optimization target is calculated by the following formula: ;
[0020] wherein di is the Euclidean distance of a single K&C characteristic, i = 1, 2, …, n, n is the number of K&C characteristics to be calibrated; ysj is the K&C characteristic simulation value, j = 1, 2, …, m, m is the number of points of the characteristic curve; yej is the K&C characteristic test value. The total optimization target is: .
[0021] In a preferred embodiment, in the step S6, the K&C model calibration is carried out by using an optimization algorithm including but not limited to NSGA-II, K&C working condition simulation is automatically carried out by calling ADAMS through MATLAB, the Euclidean distance of each K&C characteristic is solved, and the suspension component parameters are automatically iterated according to the set optimization target and constraint condition until the solved Euclidean distance meets the requirement.
[0022] The technical effects and advantages of the suspension system dynamics model calibration method based on the optimization algorithm are as follows:
[0023] 1. The optimization algorithm is introduced to convert the inefficient and repeated process of manual adjustment of parameters by engineers into an automatic and intelligent calibration process. By establishing an optimization objective function and automatically iterating the suspension component parameters by using the optimization algorithm, the working efficiency of the K&C model calibration is greatly improved, and the development cycle is shortened.
[0024] 2. The parameter sensitivity analysis can accurately identify the suspension component parameters that significantly affect the vehicle performance and key K&C characteristics, avoiding the interaction effect problem caused by blind adjustment of parameters, and improving the accuracy and stability of the calibration process.
[0025] 3. The K&C curve coincidence degree of simulation and test is taken as the optimization target, and the Euclidean distance is used to quantify the curve difference, so as to ensure the accuracy of the calibration result. The finally output high-precision suspension K&C model can more accurately reflect the actual performance of the sample vehicle, and provide a reliable basis for subsequent vehicle dynamics simulation, thereby improving the prediction accuracy and design optimization effect in the vehicle development process.
[0026] 4. The cooperation of MATLAB, TruckSim and ADAMS and other software realizes the full-process automation from test data acquisition, vehicle model establishment, parameter sensitivity analysis, rigid-flexible coupling model simulation to optimization solution, significantly reduces the manual intervention, and improves the reliability and repeatability of the calibration process. BRIEF DESCRIPTION OF DRAWINGS
[0027] Figure 1 is a flow chart of a suspension system dynamics model calibration method based on an optimization algorithm;
[0028] Figure 2 ADAMS suspension rigid-flexible coupling dynamics model schematic diagram of the present application;
[0029] Figure 3 TruckSim whole vehicle dynamics model spring mass part modeling interface of the present application;
[0030] Figure 4 Suspension K&C characteristic sensitivity analysis results of the present application on the comprehensive total weighted acceleration RMS value of each measurement point of the cab under the whole vehicle random road condition;
[0031] Figure 5 Suspension K&C characteristic sensitivity analysis results of the present application on the understeering degree of the whole vehicle under the steady-state turning condition;
[0032] Figure 6 Part parameter sensitivity analysis results of the present application on the front axle bounce longitudinal displacement coefficient.
[0033] Figure 7 Part parameter sensitivity analysis results of the present application on the front axle roll-steering coefficient.
[0034] Figure 8 Part parameter sensitivity analysis results of the present application on the front axle longitudinal force toe angle coefficient.
[0035] Figure 9 Part parameter sensitivity analysis results of the present application on the front axle lateral flexibility. DETAILED DESCRIPTION
[0036] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present application.
[0037] It is assumed that DOE (Design of Experiments) refers to test design, a branch of mathematical statistics, which is a sampling algorithm that can understand the relationship between input variables and output responses according to limited test data combined with sensitivity algorithms;
[0038] Suspension K&C characteristics refer to the kinematic and compliance characteristics of the suspension. Influenced by the characteristics of the suspension's hard points and elastic components, they play a decisive role in a vehicle's handling, stability, and ride comfort. Specifically, K-characteristics, or suspension kinematics, refer to the angular and linear displacements of the wheel plane and wheel center point caused by the suspension's guiding mechanism during the wheel's reciprocating motion in the vertical direction. C-characteristics, or suspension compliance, refer to the angular and linear displacements of the wheel plane and wheel center point caused by the forces and moments exerted by the ground on the tires.
[0039] This invention provides a suspension system dynamics model calibration method based on optimization algorithms, aiming to solve the problems of low efficiency, long cycle time, and repeated calibration and failure due to parameter interaction effects in existing K&C performance calibration methods. This method achieves high-precision calibration of the suspension dynamics model through automated processes and optimization algorithms.
[0040] Reference Figure 1 The present invention provides a method for calibrating a suspension system dynamic model based on an optimization algorithm, comprising the following steps:
[0041] S1: Conduct K&C tests on the prototype vehicle suspension to obtain test data. This step is fundamental to the entire calibration process and aims to obtain suspension K&C performance data for the actual prototype vehicle. Specifically, it is necessary to conduct K&C tests on the prototype vehicle suspension to obtain test data under various K&C conditions. These conditions include, but are not limited to, same-direction wheel hop, opposite-direction wheel hop, steering, longitudinal force, same-direction self-centering torque, opposite-direction self-centering torque, same-direction lateral force, and opposite-direction lateral force. Under these conditions, the test data to be collected includes key parameters such as suspension stiffness, toe angle change rate, camber angle change rate, caster angle change rate, and kingpin inclination angle change rate. For subsequent calculations and analysis, polynomial fitting is performed on the original test data to calculate the Euclidean distance between the test and simulation result curves in step S6, serving as the target and constraint for K&C benchmarking. By fitting the experimental data, a smooth and continuous curve can be obtained, which more accurately reflects the nonlinear characteristics of the suspension and provides a reliable benchmark for subsequent simulation model comparison.
[0042] S2: Based on prototype vehicle parameters and K&C test data, establish a TruckSim vehicle model. After obtaining the actual K&C test data, a simulation model that reflects the overall vehicle dynamics needs to be established. This step, based on prototype vehicle parameters (including sprung mass parameters, tire parameters, steering system parameters, powertrain and transmission system parameters, front and rear suspension parameters, etc.) and the K&C performance test data obtained in step S1, establishes a vehicle dynamics model based on the suspension K&C assembly characteristics in TruckSim software, such as...Figure 3 The whole vehicle model will be used for subsequent parameter sensitivity analysis of suspension K&C assembly characteristics on vehicle handling and ride performance, providing a platform for identifying key K&C characteristics. TruckSim software can easily integrate various vehicle subsystem models and perform vehicle-level dynamics simulation.
[0043] S3: Take suspension K&C characteristics as variables, carry out DOE calculation of vehicle ride performance and handling stability in each working condition, and perform parameter sensitivity analysis based on DOE data to select K&C characteristics sensitive to vehicle performance as benchmark items. In order to understand the influence of suspension K&C characteristics on vehicle performance, systematic DOE and parameter sensitivity analysis are needed. Specifically, by writing MATLAB program and vscommand command, the automation operation of modifying suspension K&C characteristics of vehicle dynamics model, vehicle ride performance and handling stability working condition simulation, result curve output and result curve post-processing is realized. Based on the automation program, DOE calculation of vehicle ride performance and handling stability working condition is carried out with suspension K&C characteristics as variables. Through DOE calculation, the influence of different K&C characteristics combination on vehicle performance indicators (such as ride performance, handling stability) can be systematically explored. Then, based on DOE data, parameter sensitivity analysis is carried out to find out the key K&C characteristics, i.e. the K&C characteristics that have the greatest influence on vehicle performance indicators, which will be used as the target item of K&C performance benchmarking. For example, Figure 4 The sensitivity analysis results of K&C characteristics on vehicle ride performance indicators are shown, Figure 5 The sensitivity analysis results of K&C characteristics on understeering degree are shown. These analysis results help to focus on the key points of calibration work and improve efficiency.
[0044] S4: Establish suspension rigid-flexible coupling dynamics model in ADAMS software, and write secondary development script to realize automation operation of model component parameter modification, K&C working condition simulation and result curve output. In order to perform high-precision suspension K&C performance simulation, a detailed suspension model needs to be established. This step includes: first, extract front and rear suspension hard point coordinates, bushing stiffness and related parameters such as leaf spring modeling, and establish front and rear suspension rigid body dynamics model in ADAMS software. Second, establish structural finite element model of front and rear axle and steering tie rod in HyperMesh software, solve modal neutral file (*.mnf) based on modal synthesis method, and import the file into ADAMS to establish front and rear suspension rigid-flexible coupling dynamics model for suspension K&C performance simulation, such as Figure 2The rigid-flexible coupling model can more truly reflect the dynamic characteristics of the suspension under force deformation. In order to realize subsequent automatic optimization, it is also necessary to write a secondary development script. Specifically, MATLAB programs and CMD command scripts are written to realize the automatic operation of calling ADAMS software to modify the parameters of parts such as bushing stiffness and hard point coordinates in the model, simulation of various K&C working conditions, output of result curves, and data post-processing. This lays the foundation for subsequent establishment of K&C benchmarking process for front and rear suspensions, enabling the optimization algorithm to automatically adjust model parameters and perform simulation.
[0045] S5: Based on the automatic simulation process of step S4, taking the model part parameters as variables, carrying out DOE calculation and sensitivity analysis of K&C working conditions, and selecting the part parameters sensitive to the key K&C characteristics as the benchmark variables. After establishing the automatic simulation platform, it is necessary to further identify which part parameters have the greatest impact on the key K&C characteristics. Based on the automatic simulation process built in S4, taking the model part parameters (such as hard point coordinates, bushing stiffness, etc.) as variables, DOE calculation of K&C working conditions is carried out. Through DOE calculation, the influence of different part parameter combinations on K&C characteristics can be systematically studied. Then, based on the DOE data, parameter sensitivity analysis is carried out, and the part parameters sensitive to the key K&C characteristics are selected as the benchmark variables. For example, Figure 6 sensitivity results of part parameters on the front axle bounce longitudinal displacement coefficient are shown, Figure 7 sensitivity results of part parameters on the front axle roll steering coefficient are shown, Figure 8 sensitivity results of part parameters on the front axle longitudinal force toe angle coefficient are shown, Figure 9 sensitivity results of part parameters on the rear axle lateral flexibility are shown. These analysis results are helpful to determine the parameter range and priority that the optimization algorithm needs to adjust.
[0046] S6: Based on the automatic simulation process of step S4, taking the simulation and test K&C curves as the optimization target, establishing the optimization function, solving the K&C model benchmarking, and outputting the high-precision suspension K&C model. This is the core step of the entire calibration method. First, determine the three optimization elements: 1. Optimization variables: take the part parameters such as suspension bushing stiffness and hard point coordinates as optimization variables, which are the parameters sensitive to the key K&C characteristics selected in step S5. 2. Optimization target: take the minimum sum of Euclidean distance of key K&C characteristic test and simulation result curves as the optimization target. 3. Constraint condition: take the Euclidean distance of K&C characteristic simulation and test as the constraint condition, in which the key K&C characteristics are determined by step S3.
[0047] The optimization target is obtained by data processing of the test and simulation result curves through MATLAB programming. First, the Euclidean distance of the test and simulation result curves of a single K&C characteristic is calculated:
[0048] ;
[0049] In formula (1), (i=1, 2, …, n) is the Euclidean distance of a single K&C characteristic, n is the number of the target K&C characteristics, i.e. the key K&C characteristics selected in step S3, (j=1, 2, …, m) is the K&C characteristic simulation value, (j=1, 2, …, m) is the K&C characteristic test value, and m is the number of points of the characteristic curve. Then, the sum of the Euclidean distances of all key K&C characteristic simulation and test curves is calculated:
[0050] ;
[0051] According to the three optimization elements, a MATLAB program is written, combined with the ADAMS software CMD command script written in step S4, and an optimization algorithm including but not limited to NSGA-II is adopted. The optimization algorithm automatically carries out K&C working condition simulation through MATLAB calling ADAMS, solves the Euclidean distance of each K&C characteristic, and automatically iterates the suspension component parameters according to the set optimization target and constraint conditions until the solved Euclidean distance meets the requirements. Finally, the calibration scheme, i.e. the high-precision suspension K&C model parameter after optimization adjustment, is output.
[0052] The application of the suspension system dynamics model calibration method based on the optimization algorithm according to the above embodiment will be described below through a specific example. For a certain type of light truck, the suspension K&C performance is calibrated by applying the suspension system dynamics model calibration method based on the optimization algorithm provided by the present application. The vehicle adopts a monocoque suspension. First, DOE analysis is performed through TruckSim to obtain the sensitivity of each K&C characteristic to the vehicle ride comfort index (the overall total weighted acceleration RMS of each measurement point of the cab under random road surface), and the corresponding sensitivity analysis result is shown in Figure 4 or the sensitivity result is shown in Table 1, and the sensitivity result of the insufficient steering degree is shown in Table 2 and Figure 5The key K&C characteristics are selected as the benchmark items through the sensitivity analysis results, the hard point coordinates and the bush stiffness are used as the optimization variables, and the Euclidean distance between the test and simulation results curves of the key K&C characteristics is minimized as the optimization target, so that the benchmark solution is obtained, the Euclidean distance between the test and simulation results curves of the K&C characteristics of the original model and the benchmark model is shown in Table 3, and the part parameters of the front suspension in the benchmark scheme and the original scheme are compared in Table 4. It can be seen that the optimization algorithm is successfully applied to the suspension K&C curve benchmarking, and the working efficiency and the benchmarking accuracy are greatly improved, and a high-precision suspension dynamics model can be output.
[0053] Table 1 Sensitivity analysis results of comprehensive total weighted acceleration root mean square value
[0054] Variable Sensitivity Rear axle roll lateral displacement coefficient -0.27 Front axle jounce longitudinal displacement coefficient 0.06 Front axle roll steer coefficient 0.02 Rear axle jounce longitudinal displacement coefficient 0.02 Rear axle roll steer coefficient 0.005 Front axle roll lateral displacement coefficient -0.001 Rear axle lateral compliance -0.0007
[0055] Table 2 Sensitivity analysis results of insufficient steering degree
[0056] Variable Sensitivity Front axle roll steer coefficient -0.13 Rear axle roll lateral displacement coefficient 0.05 Front axle lateral compliance -0.02 Rear axle jounce longitudinal displacement coefficient 0.013 Rear axle roll steer coefficient 0.009 Front axle roll righting moment steer angle coefficient -0.009 Front axle jounce steer angle coefficient 0.008 Front axle lateral force steer angle coefficient -0.008 Rear axle lateral compliance 0.007 Front axle jounce longitudinal displacement coefficient -0.006 Front axle jounce pitch coefficient 0.004 Front axle roll righting moment steer angle coefficient -0.003
[0057] Table 3 Comparison of Euclidean distances of K&C characteristic curves
[0058] K&C characteristics Euclidean distance between original scheme and experimental curve Euclidean distance between calibrated scheme and experimental curve Front axle jounce longitudinal displacement coefficient 26 5.3 Front axle roll steer coefficient 1.6 0.32 Front axle longitudinal force camber angle coefficient 2.1 0.22 Front axle lateral compliance 8.7 3.2 Rear axle jounce longitudinal displacement coefficient 18.5 5.7 Rear axle jounce pitch coefficient 0.3 0.13 Rear axle roll steer coefficient 0.009 0.005 Rear axle roll lateral displacement coefficient 0.7 0.2
[0059] Table 4 Comparison of front suspension part parameters
[0060] Component parameters Unit Original scheme Calibrated scheme Front leaf spring front end bush X direction stiffness N / mm 5000 2000 Front leaf spring front end bush Y direction stiffness N / mm 5000 2200 Front leaf spring front end bush Z direction stiffness N / mm 5000 3000 Front leaf spring and hanger bush Y direction stiffness N / mm 5000 13000 Front leaf spring and hanger bush Rx direction stiffness N·mm / deg 100000 40000 Front leaf spring hanger and frame bush Y direction stiffness N / mm 5000 26000 Front leaf spring hanger and frame bush Rx direction stiffness N·mm / deg 100000 12000 Front leaf spring axle hub and axle bush Ry direction stiffness N·mm / deg 10000000 21400000 Front axle swing arm lower point X mm -794.96 -785 Front axle swing arm lower point Z mm -67.02 -65 Front axle straight pull rod rear point Z mm -147.84 -158 Front axle kingpin upper point X mm 1.07 1.07 Front axle kingpin upper point Y mm -734.05 -720 Front axle wheel center X mm -3.6 6.4 Front axle wheel center Z mm -141.86 -81.1 Front axle leaf spring front point X mm -600 -650 Front axle leaf spring front point Z mm -85 -115 Front axle leaf spring rear point X mm 564 564 Front axle leaf spring rear point Z mm -59 -79
[0061] Although the present application has been described in detail with reference to the preferred embodiments, the present application is not limited to the preferred embodiments. Those skilled in the art can make various equivalent modifications or replacements to the embodiments of the present application without departing from the spirit and essence of the present application, and these modifications or replacements should be within the scope of the present application. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, and these changes or replacements should be within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.
[0062] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims
1. A method for calibrating the dynamic model of a suspension system based on an optimization algorithm, characterized in that, Includes the following steps: S1: Conduct K&C tests on the prototype vehicle suspension, obtain test data under various K&C conditions, and perform polynomial fitting on the original test data; S2: Based on the prototype vehicle parameters and the K&C performance test data obtained in step S1, establish a vehicle dynamics model based on the characteristics of the suspension K&C assembly; S3: Using suspension K&C characteristics as variables, conduct design of experiments (DOE) calculations for various conditions of vehicle ride comfort and handling stability, and perform parameter sensitivity analysis based on DOE data to select K&C characteristics that are sensitive to vehicle performance as benchmarks. S4: Establish a suspension rigid-flexible coupling dynamic model in ADAMS software, and write a secondary development script to automate the modification of model component parameters, simulation of various K&C conditions, and output of result curves. S5: Based on the automated simulation process in step S4, using the model component parameters as variables, perform DOE calculations and sensitivity analyses for each K&C condition, and select component parameters that are sensitive to each key K&C characteristic as benchmark variables. S6: Based on the automated simulation process in step S4, with the matching of the K&C curves of simulation and experiment as the optimization objective, establish an optimization function, perform K&C model benchmarking and solution, and output a high-precision suspension K&C model.
2. The method according to claim 1, characterized in that, In step S1, the suspension K&C conditions include same-direction wheel hop condition, opposite-direction wheel hop condition, steering condition, longitudinal force condition, same-direction self-centering torque condition, opposite-direction self-centering torque condition, same-direction lateral force condition, and opposite-direction lateral force condition; the test data include suspension stiffness, toe angle change rate, camber angle change rate, caster angle change rate, and kingpin inclination angle change rate.
3. The method according to claim 1, characterized in that, In step S2, the prototype vehicle parameters include sprung mass parameters, tire parameters, steering system parameters, powertrain and transmission system parameters, and front and rear suspension parameters.
4. The method according to claim 1, characterized in that, In step S3, by writing MATLAB programs and vscommand commands, the automated operation of calling TruckSim software to modify the suspension K&C characteristics of the whole vehicle dynamics model, simulate the vehicle ride comfort and handling stability conditions, output the result curves, and perform post-processing of the result curves is realized.
5. The method according to claim 1, characterized in that, In step S4, establishing the suspension rigid-flexible coupling dynamic model includes: extracting relevant parameters such as the coordinates of the hard points of the front and rear suspensions, bushing stiffness, and leaf spring modeling; establishing the rigid body dynamic model of the front and rear suspensions in ADAMS software; establishing the structural finite element model of the front and rear axles and steering tie rods in HyperMesh software; solving the modal neutral file based on the modal synthesis method; and importing it into ADAMS to establish the front and rear suspension rigid-flexible coupling dynamic model.
6. The method according to claim 5, characterized in that, In step S4, writing the secondary development script includes writing MATLAB programs and CMD command scripts to automate the operation of calling ADAMS software to modify component parameters such as bushing stiffness and hard point coordinates in the model, simulating various K&C conditions, outputting result curves, and post-processing data.
7. The method according to claim 1, characterized in that, In step S5, the parameters of the model components include hard point coordinates and bushing stiffness.
8. The method according to claim 1, characterized in that, In step S6, establishing the optimization function includes: determining component parameters such as suspension bushing stiffness and hard point coordinates as optimization variables; determining the minimum sum of Euclidean distances between the test and simulation result curves of key K&C characteristics as the optimization objective; and determining the Euclidean distance between the simulation and test of each K&C characteristic as a constraint condition, wherein the key K&C characteristics are determined by step S3.
9. The method according to claim 8, characterized in that, The optimization objective is calculated using the following formula: ; Where, di is the Euclidean distance of a single K&C characteristic, i=1,2,…,n, and n is the number of K&C characteristics to be compared; ysj is the simulated value of the K&C characteristic, j=1,2,…,m, and m is the number of points on the characteristic curve; yej is the experimental value of the K&C characteristic. The overall optimization objective is: .
10. The method according to claim 8, characterized in that, In step S6, the K&C model benchmarking solution adopts optimization algorithms including but not limited to NSGA-II. The K&C working condition simulation is automatically carried out by calling ADAMS through MATLAB to solve the Euclidean distance of each K&C characteristic. The suspension component parameters are automatically iterated according to the set optimization objectives and constraints until the solved Euclidean distance meets the requirements.