A robust matching method for suspension and vehicle performance considering uncertainties
By establishing the ADAMS/CAR model and integrated optimization model, combined with DOE sampling and multi-objective optimization algorithms, the problems of insufficient handling stability and smoothness in suspension design were solved, a robust match between suspension and vehicle performance was achieved, and the impact of design errors on performance was reduced.
Patent Information
- Application Number
- CN202310358176.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-04-06
- Publication Date
- 2025-09-16
- Estimated Expiration
- 2043-04-06
AI Technical Summary
Existing technologies fail to simultaneously optimize the vehicle's handling stability and ride comfort in suspension design, and errors in suspension parameters lead to fluctuations in vehicle performance and a lack of robustness.
By establishing an ADAMS/CAR multi-body dynamics model, an integrated model that includes vehicle handling stability and ride comfort is constructed. ISIGHT and MATLAB are used for optimization. Combined with DOE sampling and RBF model, the NSGA-II multi-objective optimization algorithm is adopted to consider the uncertainty of suspension parameters and achieve a robust match between suspension and vehicle performance.
It effectively reduces the impact of suspension design parameter errors on vehicle performance, improves the robustness and operability of the suspension and vehicle performance, and reduces development cycle and cost.
Smart Images
Figure CN116383969B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of vehicle suspension design, and in particular relates to a robustness matching method between a suspension and a whole vehicle performance taking into account uncertain factors. Background Art
[0002] With the advancement of the global economy and technology, automobiles have become an indispensable part of modern society. Simultaneously, people's demands for overall vehicle performance are constantly increasing. As two key vehicle performance characteristics, handling stability and ride smoothness are receiving increasing attention. As a key subsystem of the automotive chassis, the suspension plays a crucial role in the vehicle's handling stability. Furthermore, the suspension cushions impact forces from the ground and reduces the vibrations caused by impact, thus affecting the vehicle's ride comfort. Therefore, when designing the suspension, both handling stability and ride smoothness should be considered comprehensively.
[0003] Since the suspension system has many design parameters, when matching the suspension with the vehicle performance, it is necessary not only to consider both of the vehicle's performance but also the actual operability.
[0004] Currently, some studies consider matching suspension performance with vehicle handling stability and ride comfort, adjusting suspension K&C characteristics. Others constrain suspension parameters within a specific range to ensure a specific vehicle performance, optimizing another performance within that range. These approaches don't directly consider both handling stability and ride comfort simultaneously, potentially resulting in suspension parameters being adjusted to achieve handling stability and ride comfort that aren't within the vehicle's performance limits.
[0005] Due to the influence of actual production and processing conditions of the suspension system, errors in suspension design parameters are inevitable. These errors will cause the vehicle's performance indicators to deviate, which may cause the designed suspension and steering system to fail to meet the vehicle's performance requirements. In suspension optimization design, the goal is generally to minimize fluctuations in vehicle performance indicators caused by uncertain factors while improving vehicle handling stability and ride comfort, thereby achieving greater vehicle performance robustness.
[0006] Therefore, in order to solve the above problems, this paper proposes a robust matching method between suspension and vehicle performance considering uncertainties. Summary of the Invention
[0007] In order to solve the above technical problems, the present invention designs a robust matching method for suspension and vehicle performance taking into account uncertain factors. On the basis of comprehensively considering the handling stability and smoothness of the vehicle under typical working conditions, the uncertainty of the suspension design parameters is also taken into account. In the optimization design of the suspension system, the present invention has strong applicability and operability.
[0008] In order to achieve the above technical effects, the present invention is implemented through the following technical solution: a method for robustly matching suspension and vehicle performance considering uncertain factors, characterized by comprising the following steps:
[0009] S1. Use multi-body dynamics software to establish an ADAMS / CAR model, determine the simulation conditions for vehicle handling stability and ride comfort, and define relevant objective evaluation indicators and data processing methods;
[0010] S2. Use ISIGHT, MATLAB, and ADAMS / CAR to build an integrated model that includes vehicle handling stability and ride comfort;
[0011] S3. Using the front and rear suspension spring stiffness curves, shock absorber damping curves, and bushing stiffness curves as research objects, sensitivity analysis was performed to identify suspension parameters that have a significant impact on various evaluation indicators of vehicle handling stability and ride comfort.
[0012] S4. Using the selected suspension parameters as design variables, sampling is performed using DOE design to construct an approximate model to fit the mapping relationship between the design variables and various evaluation indicators;
[0013] S5. Taking the objective evaluation indicators of vehicle handling stability and ride comfort as the optimization objectives, the selected suspension parameters as the optimization variables, and establishing a robustness optimization model. The established optimization model is solved using a multi-objective optimization algorithm.
[0014] S6. Analyze the robustness optimization results to verify the effectiveness of the optimization and ultimately achieve a robust optimization matching design between the suspension and vehicle performance.
[0015] Furthermore, the vehicle handling stability simulation working conditions include: one or more of a fixed radius steady-state rotation and a steering wheel sinusoidal sweep simulation.
[0016] Furthermore, the specific definition of the evaluation index of vehicle handling stability is as follows:
[0017] Steady-state rotation at a constant radius: Steering wheel angle change gradient when the lateral acceleration of the steering wheel angle-lateral acceleration curve is 0.05g-0.4g; vehicle body roll angle change gradient when the lateral acceleration of the vehicle body roll angle-lateral acceleration curve is 0.05g-0.4g; steering wheel angle when the lateral acceleration of the steering wheel angle-lateral acceleration curve is 0.05g; steering wheel angle when the lateral acceleration of the steering wheel angle-lateral acceleration curve is 0.7g;
[0018] Steering wheel sinusoidal sweep: gain of the yaw rate-steering wheel angle gain curve at a frequency of 0.5 Hz, gain of the body roll angle-lateral acceleration gain curve at a frequency of 0.5 Hz, and lag time of the lateral acceleration-steering wheel angle lag time curve at a frequency of 0.5 Hz.
[0019] Furthermore, the vehicle ride comfort simulation condition is one or more of random road input and pulse road input simulation.
[0020] Furthermore, the specific definition of the evaluation index of ride comfort is as follows:
[0021] The evaluation indicators of the random road input simulation condition include the vertical acceleration and longitudinal RMS value of the front and rear seat rails. The time-domain curves of the vertical and longitudinal acceleration of the front and rear seat rails obtained from the pulse road input simulation experiment are processed using the continuous RMS method to obtain the Rmsglobal curve. The maximum value of the Rmsglobal curve is selected as the objective evaluation indicator of the pulse road input smoothness.
[0022] Furthermore, the specific method for data processing of the time domain curves of the longitudinal acceleration of the front and rear seat rails is as follows:
[0023]
[0024] v is the vehicle speed and l is the vehicle's wheelbase.
[0025] Furthermore, in S4, the sampling method using DOE design is an optimal Latin hypercube experiment.
[0026] Furthermore, in the above S4, the approximate model for fitting the mapping relationship between the design variables and the evaluation indicators is one or more of the radial basis function neural network agent model RBF model, RSM, and Kriging model.
[0027] Furthermore, in S5, the optimization algorithm is one of NSGA-II, MOPSO, NCGA, and AMGA optimization algorithms.
[0028] Furthermore, in S6, the robustness analysis of the optimization scheme adopts any one of reliability evaluation, Monte Carlo simulation, and experimental design.
[0029] The beneficial effects of the present invention are:
[0030] This method utilizes ADAMS / CAR multibody dynamics simulation technology to establish a multibody dynamics model, reducing development cycles and costs. When matching suspension with vehicle performance, this method directly targets vehicle ride comfort and handling stability evaluation indicators. Using ISIGHT, MATLAB, and ADAMS / CAR, an integrated optimization model encompassing both vehicle handling stability and ride comfort is constructed. Through DOE sampling and the establishment of an RBF model, the practical application problem is transformed into a multi-objective optimization problem. The uncertainty of the design parameters is considered in the optimization problem to prevent excessive fluctuations in vehicle performance caused by design errors. This method boasts a simple principle, strong operability, and high reliability. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0032] Figure 1 A flow chart of a method for robustly matching suspension and vehicle performance considering uncertainties provided by an embodiment of the present invention;
[0033] Figure 2 The multi-body dynamics model of the vehicle provided in the embodiment of the present invention;
[0034] Figure 3 A schematic diagram of a vehicle front suspension model provided by an embodiment of the present invention;
[0035] Figure 4 A schematic diagram of a vehicle rear suspension model provided by an embodiment of the present invention;
[0036] Figure 5 This is a diagram illustrating objective evaluation indicators of typical handling stability conditions provided by an embodiment of the present invention;
[0037] Figure 6 This is a diagram illustrating objective evaluation indicators of typical ride comfort conditions provided by an embodiment of the present invention;
[0038] Figure 7 Flowchart for constructing the RBF approximation model provided by the embodiment of the present invention;
[0039] Figure 8The embodiment of the present invention provides a robustness optimization model based on the RBF approximation model built using ISIGHT and ADAMS / CAR. DETAILED DESCRIPTION
[0040] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.
[0041] Example 1
[0042] Figure 1 A flowchart of a method for robustly matching suspension and vehicle performance considering uncertainties provided by an embodiment of the present invention. The method specifically includes the following steps:
[0043] S1. Use multi-body dynamics software to establish an ADAMS / CAR vehicle simulation model, determine the simulation conditions for vehicle handling stability and ride comfort, and define relevant objective evaluation indicators and related data processing methods.
[0044] The specific implementation method is: using multi-body dynamics software to establish a multi-body dynamics model of the vehicle. The multi-body dynamics model established in the embodiment of the present invention is as follows: Figure 2 The front suspension of the model is McPherson suspension, as shown in Figure 3 As shown, the rear suspension is a multi-link suspension. Figure 4 shown.
[0045] In the embodiment of the present invention, four typical working conditions are selected to evaluate the vehicle's handling stability and ride comfort. The typical handling stability working condition evaluation index is as follows: Figure 5 As shown in the figure, the evaluation index of typical driving smoothness conditions is as follows Figure 6 shown.
[0046] The simulation conditions for vehicle handling stability are fixed radius steady-state rotation and steering wheel sinusoidal sweep simulation.
[0047] The simulation conditions for vehicle smoothness are random road input and pulse road input simulation.
[0048] Determine the evaluation indicators for constant radius steady-state rotation simulation.
[0049] Evaluation index H1: Steering wheel angle gradient in the linear region. The slope of the steering wheel angle to lateral acceleration curve from 0.05g to 0.4g, such as Figure 3 Evaluation indicator H2: Roll angle gradient in the linear region. The slope of the body roll angle versus lateral acceleration curve from 0.05g to 0.4g.
[0050] Determine the evaluation indicators for steering wheel sinusoidal swept frequency simulation.
[0051] Evaluation index H3: Yaw rate gain relative to steering wheel angle at a frequency of 0.5 Hz. Evaluation index H4: Lag time of lateral acceleration relative to steering wheel angle at a frequency of 0.5 Hz. Evaluation index H5: Body roll angle gain relative to lateral acceleration at a frequency of 0.5 Hz.
[0052] Determine the random road surface smoothness simulation evaluation index. In the embodiment of the present invention, the harmonic superposition method is used to establish the B-level road surface. The specific establishment method is cited in the literature:
[0053] Huang Zhiqiang, Zheng Wanghui. Implementation of ADAMS 3D random road surface modeling using Matlab[J]. Modern Defense Technology, 2018, 46(03): 165-170.
[0054] The vertical acceleration time history curve of the vehicle center of mass obtained by simulating the established vehicle simulation model at a common speed of 60km / h on a Class B road is calculated by the Z-direction frequency weighting function w k The filtering network of (f) obtains the weighted acceleration time history a w (t), preferably, by The vertical weighted acceleration root mean square value R1 of the vehicle's center of mass is calculated to evaluate the ride comfort on random roads. The specific calculation method is cited in the following literature:
[0055] Yu Fan, Lin Yi. Automotive System Dynamics[M]. Machinery Industry Press
[0056] Determine the pulse road surface smoothness simulation evaluation indicators. Evaluation indicator R2: The peak value of the vehicle's center of mass longitudinal acceleration Rmslocal curve. Evaluation indicator R3: The peak value of the vehicle's center of mass vertical acceleration Rmslocal curve.
[0057] S2. Use ISIGHT, MATLAB, and ADAMS / CAR to build a multi-condition simulation integrated model that includes typical vehicle handling stability and ride comfort, and simulate and calculate various evaluation indicators.
[0058] After the simulation is completed, the values of the evaluation indicators of each working condition are calculated based on the simulation data. In specific implementation, Isight is used to drive AdamsCar to perform multi-working condition simulation to obtain vehicle yaw acceleration, lateral acceleration, steering wheel angle and other parameters related to the calculation of handling stability evaluation indicators, as well as vertical and longitudinal accelerations of the vehicle center of mass, which are related to the calculation of smoothness indicators.
[0059] The obtained relevant parameters are used as input parameters in MATLAB, and a program is written using MATLAB to perform data processing on the input parameters to obtain various evaluation indicators. Preferably, the obtained various evaluation indicators are used as MATLAB output parameters.
[0060] S3. In an embodiment of the present invention, the spring stiffness, shock absorber damping, bushing stiffness in the suspension system are optimized and matched with the performance of the entire vehicle for robustness.
[0061] To improve the computational efficiency of the optimized matching, a sensitivity analysis of the suspension design parameters was conducted. Specifically, the front and rear suspension spring stiffness curves, shock absorber damping curves, and bushing stiffness curves were used as design variables, and the evaluation indicators under working conditions such as fixed-radius steady-state rotation, steering wheel sinusoidal sweep, random road driving, and bump road driving were used as response variables. The perturbation method was used to perform a sensitivity analysis to screen out the suspension parameters that have a greater impact on the evaluation indicators of typical vehicle handling stability and smoothness conditions.
[0062] S4. Suspension parameters with an influence greater than 5% are used as design variables, sampling is performed using DOE design, and the mapping relationship between the design variables and various evaluation indicators is fitted using the RBF neural network model. Figure 7 Construct a flowchart for the RBF approximation model.
[0063] The accuracy of the established model was evaluated using the coefficient of certainty.
[0064] In this implementation case, the accuracy of the approximate model is evaluated using the coefficient of certainty R2, which is expressed as:
[0065]
[0066] Among them, y i The actual value of the sample point represented by The mean of the actual values of the sample points represented by represents the predicted value of the surrogate model.
[0067] S5. Use the NSGA-II multi-objective optimization algorithm to perform deterministic multi-objective optimization on the established agent model.
[0068] Considering that the uncertainty of the optimization variables satisfies the normal distribution, the evaluation index obtained by deterministic optimization is used as the mean of robust optimization, and the NSGA-II multi-objective optimization algorithm is used to carry out robust optimization. Figure 8 shown.
[0069] S6. In an embodiment of the present invention, Monte Carlo simulation is used to analyze the robustness optimization results, verify the effectiveness of the optimization, and ultimately achieve a robust optimization matching design of the suspension and vehicle performance.
Claims
1. A robust matching method for suspension and vehicle performance considering uncertainties, characterized in that: The steps include: S1. Use multi-body dynamics software to establish an ADAMS / CAR model, determine the simulation conditions for vehicle handling stability and ride comfort, and define relevant objective evaluation indicators and data processing methods; S2. Use ISIGHT, MATLAB, and ADAMS / CAR to build an integrated model that includes vehicle handling stability and ride comfort; S3. Using the front and rear suspension spring stiffness curves, shock absorber damping curves, and bushing stiffness curves as research objects, sensitivity analysis was performed to identify suspension parameters that have a significant impact on various evaluation indicators of vehicle handling stability and ride comfort. S4. Using the selected suspension parameters as design variables, sampling is performed using DOE design to construct an approximate model to fit the mapping relationship between the design variables and various evaluation indicators; S5. Taking the objective evaluation indicators of vehicle handling stability and ride comfort as the optimization objectives, the selected suspension parameters as the optimization variables, and establishing a robustness optimization model. The established optimization model is solved using a multi-objective optimization algorithm. S6. Analyze the robustness optimization results to verify the effectiveness of the optimization and ultimately achieve a robust optimization matching design between the suspension and vehicle performance. The specific definitions of the evaluation index of vehicle handling stability are as follows: Steady-state rotation at a constant radius: Steering wheel angle change gradient when the lateral acceleration of the steering wheel angle-lateral acceleration curve is 0.05g-0.4g; vehicle body roll angle change gradient when the lateral acceleration of the vehicle body roll angle-lateral acceleration curve is 0.05g-0.4g; steering wheel angle when the lateral acceleration of the steering wheel angle-lateral acceleration curve is 0.05g; steering wheel angle when the lateral acceleration of the steering wheel angle-lateral acceleration curve is 0.7g; Steering wheel sinusoidal sweep: gain of the yaw rate-steering wheel angle gain curve at a frequency of 0.5 Hz, gain of the body roll angle-lateral acceleration gain curve at a frequency of 0.5 Hz, and lag time of the lateral acceleration-steering wheel angle lag time curve at a frequency of 0.5 Hz; The specific definitions of the evaluation index of ride comfort are as follows: The evaluation indicators of the random road input simulation condition include the vertical acceleration and longitudinal RMS value of the front and rear seat rails. The time-domain curves of the vertical and longitudinal acceleration of the front and rear seat rails obtained from the pulse road input simulation experiment are processed using the continuous RMS method to obtain the Rmsglobal curve. The maximum value of the Rmsglobal curve is selected as the objective evaluation indicator of the pulse road input smoothness.
2. The method for robustly matching suspension and vehicle performance considering uncertainties according to claim 1, characterized in that: The vehicle handling stability simulation working conditions include: one or more of a fixed radius steady-state rotation and a steering wheel sinusoidal sweep frequency simulation.
3. The method for robustly matching suspension and vehicle performance considering uncertainties according to claim 1, characterized in that: The vehicle ride comfort simulation working condition is one or more of random road input and pulse road input simulation.
4. The method for robustly matching suspension and vehicle performance considering uncertainties according to claim 1, characterized in that: The specific method for data processing of the time domain curves of the longitudinal acceleration of the front and rear seat rails is as follows: v is the vehicle speed and l is the vehicle's wheelbase.
5. The method for robust matching of suspension and vehicle performance considering uncertainties according to claim 1, characterized in that: In the aforementioned S4, the sampling method using DOE design is the optimal Latin hypercube experiment.
6. The method for robust matching of suspension and vehicle performance considering uncertainties according to claim 1, characterized in that: In the above-mentioned S4, the approximate model for fitting the mapping relationship between the design variables and the various evaluation indicators is one or more of the radial basis function neural network agent model RBF model, RSM, and Kriging model.
7. The method for robustly matching suspension and vehicle performance considering uncertainties according to claim 1, characterized in that: In the above S5, the optimization algorithm is one of NSGA-II, MOPSO, NCGA, and AMGA optimization algorithms.
8. The method for robustly matching suspension and vehicle performance considering uncertainties according to claim 1, characterized in that: In S6, the robustness analysis of the optimization scheme adopts any one of reliability evaluation, Monte Carlo simulation, and experimental design.