Design optimization method for lower control arm of armored wheel type vehicle suspension

Abstract

The invention provides a design optimization method for a suspension lower control arm of an armored wheel type vehicle, relates to the technical field of vehicle engineering, and is used for carrying out design optimization on a lower control arm which is one of main parts of a double-hinge system widely used in the armored wheel type vehicle. In the method, after finite element analysis of the lower control arm is verified, weight optimization research is carried out, verification is carried out through tests on the vehicle, in finite element model verification, strain values are collected from the lower control arm of the 8 * 8 wheeled vehicle, and similar boundary conditions are applied to a finite element model, so that the weight of the lower control arm is optimized. A method based on an agent model is adopted in optimization, a training point of the agent model is generated by using central composite design, a genetic aggregation agent modeling technology is used in ANSYS Workbench, and a result shows that the weight of a control arm can be reduced from 25.2 kg to 21.8 kg, which indicates that the weight is reduced by 13.3%.

Description

Technical Field

[0001] The invention belongs to the technical field of vehicle engineering, and more specifically, particularly relates to a design optimization method for a lower control arm of an armored wheeled vehicle suspension. Background Art

[0002] The importance of structural optimization in the defense and automotive industries has gradually increased over the past decades. The suspension system is one of the most important subsystems of an armored wheeled vehicle. The vehicle suspension system is responsible for the safety and driving comfort of the vehicle, as it carries the weight of the entire vehicle body. The main task of the suspension system is to prevent the transmission of road shocks to the vehicle chassis and other components and to ensure the stability, safety and comfort of the vehicle. The double wishbone suspension system is an independent suspension system design in wheeled vehicles, which consists of a lower control arm and an upper control arm. The overall function of the control arm is to prevent uncontrolled movement of the wheel in all road conditions by forming a connection between the suspension system and the vehicle chassis. The lower control arm is the main example problem of the present invention, which is present in every suspension unit of an armored wheeled vehicle. The manufacturing cost of the control arm accounts for a large part of the total cost of the system. The goal of the present invention is to create a finite element (FE) model of the initial design of the lower control arm of the vehicle suspension, validate the model through vehicle testing, and use a surrogate model for weight optimization.

[0003] Design optimization of vehicle suspension systems has been conducted in various prior art. Some studies are limited to numerical analysis without any experimental component, while others validate the performance of the optimized design through testing. For example, Solanki et al. performed reliability-based design optimization of the control arm of Corvette and used multiscale material models in structural analysis. Similarly, Song et al. presented a case study of surrogate-based control arm optimization using Kriging and response surface surrogate models for weight reduction. Viqaruddin and Reddy performed weight reduction and stiffness improvement of the control arm based on static analysis and torsion analysis performed using Radioss software. et al. used the butterfly optimization algorithm to optimize the optimal shape design of automobile suspension components, in which the Kriging agent model was integrated into the optimization. Karadere et al. used topology and shape optimization techniques to design lightweight automobile suspension components. Balkan et al. used the latest structural optimization technology to obtain the optimal design of the air suspension seat and simulated the ECER14 seat belt tension test on the obtained optimal design, but no physical testing was performed. Albak et al. used population-based optimization technology to improve the design of the torsion bar suspension system. In all the above studies, no structural testing was performed at the subsystem or system level.

[0004] Design optimization studies accompanied by structural testing include the following studies. Khode et al. optimized the lower control arm suspension system of a light commercial vehicle (LCV). CAD modeling was performed using CATIA software and finite element analysis was performed using ANSYS software. To identify and verify the stress values ​​of the existing LCV, a suitable fixture was designed and manufactured, and the control arm was mounted on a universal testing machine with the LCV. It is important to note that the experimental validation of this study was performed at the component level. Yende et al. used topology optimization to remove excess weight from the lower control arm of the Mac–Pherson suspension system. In this study, the finite element model of the initial design of the control arm was not validated by physical testing, but the optimized design of the control arm was directly manufactured and compared with the output of the finite element model by testing on a universal testing machine. Gutiérrez–Moizant et al. improved the design of a crank arm, where the finite element results were compared with the experimental test output. The measurements in this study were collected by using a rectangular strain gauge Rosette. All test studies were performed at the component level using a universal dynamic test bench. In all the above studies, the experimental validation structural tests performed were performed at the component level (i.e., subsystem level).

[0005] The present invention optimizes the design of the suspension lower control arm of the 8×8 armored wheeled vehicle, wherein the finite element model of the initial design is verified by measuring with strain gauges on a "complete vehicle in the actual field". Unlike the existing research in the prior art that conducts subsystem level testing, the present invention conducts structural testing at the system level, providing a novelty compared to the existing research. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a design optimization method for a lower control arm of an armored wheeled vehicle suspension to solve the above problems.

[0007] A design optimization method for a lower control arm of an armored wheeled vehicle suspension comprises the following steps:

[0008] S1: Define the design variables. The design variables are the length d1, thickness d2, width d3 and length d4 of the lower control arm. Determine that the lower limit of d1 is 75 mm and the upper limit is 105 mm, the lower limit of d2 is 7 mm and the upper limit is 12 mm, the lower limit of d3 is 80 mm and the upper limit is 103.9 mm, and the lower limit of d4 is 180 mm and the upper limit is 214.5 mm;

[0009] S2: The optimization goal is to minimize the weight of the control arm, and the maximum vonMises stress generated by the optimized design does not exceed the maximum vonMises stress generated in the initial control arm design;

[0010] S3: Determine the material of the lower control arm;

[0011] S4: perform finite element modeling;

[0012] S5: Conduct mesh convergence study;

[0013] S6: Validate the finite element model by conducting field tests on the actual manufactured components;

[0014] S7: Perform agent-based optimization;

[0015] S8: Optimization was performed using the multi-objective genetic algorithm (MOGA) method of ANSYS v19.2 software;

[0016] S9: Perform quasi-static load scenario analysis on the optimized lower control arm.

[0017] Preferably, the material of the lower control arm is austenitic adjustable cast iron, and its material properties are: density 7200kg·m 3 , yield strength 700MPa, Poisson's ratio 0.28, and Young's modulus 130GPa.

[0018] Preferably, during finite element modeling, the load of the analysis model is applied to the main shaft of the suspension unit, one end of the control arm is connected to the steering knuckle through a ball hinge, and the other end is connected to the vehicle body through a rotating connection.

[0019] Preferably, the mesh convergence study uses 10-node tetrahedral elements and convergence is achieved at an element size of 2.5 mm.

[0020] Preferably, the finite element model is validated by performing field tests on actual manufactured components, where a strain gauge device is mounted on the lower control arm of an 8×8 vehicle suspension, the front left axle is raised to different heights and the strain is measured.

[0021] Preferably, in the agent-based optimization, a central composite design experiment is used to generate training points of the agent model, the agent model is constructed using genetic aggregation in ANSYS, and the accuracy of the agent model is measured by the root mean square error RMSE.

[0022] Preferably, the optimization process uses the multi-objective genetic algorithm MOGA method of ANSYSv19.2 software, the initial mass of the lower control arm is 25.2 kg, and the mass after optimization is 21.8 kg.

[0023] Preferably, the optimized lower control arm is analyzed under a quasi-static load scenario, including longitudinal and lateral impact scenario analysis.

[0024] Preferably, in the longitudinal impact scenario, the force is 2500N, the loading direction is longitudinal, and it is applied to the ball hinge connection point.

[0025] Preferably, in the lateral impact scenario, the force is 3000N, the loading direction is lateral, and it is applied to the ball hinge connection point.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] 1. In the present invention, the weight of the armored wheeled vehicle is significantly reduced: by optimizing the design of the lower control arm, the mass of each control arm is reduced from 25.2 kg to 21.8 kg, and the total weight reduction of each 8×8 wheeled vehicle is about 27 kg, which helps to improve the mobility and fuel efficiency of the vehicle.

[0028] 2. In the present invention, structural strength and safety are guaranteed: while reducing weight, the maximum von Mises stress generated by the optimized design does not exceed the initial design, and under quasi-static load scenarios (such as longitudinal and lateral impacts), the maximum von Mises stress value is lower than the yield strength of the component, and the safety factor meets the requirements, ensuring the driving safety of the vehicle under complex road conditions.

[0029] 3. Improved design efficiency and accuracy: The use of advanced technologies and methods such as finite element modeling, agent-based optimization methods, and multi-objective genetic algorithms has greatly reduced the design cycle and cost, while improving the accuracy and reliability of the design.

[0030] 4. Contribute to the overall vehicle weight reduction research: The weight reduction of the control arm provides important reference and support for the weight reduction of the entire vehicle, which helps to improve the overall performance of the vehicle.

[0031] 5. Expanded ideas and methods for design optimization: The optimization process and technical means in the present invention can provide useful reference and reference for the design optimization of similar vehicle components and promote the technological development in related fields.

[0032] 6. Enhanced vehicle adaptability: The optimized lower control arm can better adapt to different road conditions and driving conditions, and improves the reliability and stability of the vehicle in various complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is an isometric view of the initial design of the lower control arm of the present invention;

[0034] Figure 2 is the design variable of the control arm in the present invention;

[0035] Figure 3 is a schematic diagram of a global model of a suspension unit in the present invention;

[0036] Figure 4 is a schematic diagram of the control arm connection points in the present invention, wherein a) left, b) right, c) both;

[0037] Figure 5 is a schematic diagram of a finite element mesh of a control arm in the present invention;

[0038] Figure 6 is the grid convergence result in the present invention;

[0039] Figure 7 Schematic diagram of von Mises stress (MPa) distribution in the initial design convergent finite element model of the present invention;

[0040] Figure 8 is a schematic diagram of a strain gauge installed in a lower control arm of an 8×8 vehicle according to the present invention;

[0041] Fig. 9 It is a schematic diagram of strain gauge results corresponding to three different suspension heights in the present invention;

[0042] Fig.10 is an agent-based optimization flow chart of the present invention;

[0043] Fig.11 is a schematic diagram of the design stages in the present invention, wherein a) an initial design, b) an optimal design;

[0044] Fig.12 is a schematic diagram of von Mises stress in the best design of the present invention;

[0045] Fig.13 is a schematic diagram of von Mises stress (MPa) generated on the control arm due to a pothole impact scenario in the present invention;

[0046] Fig.14 It is a schematic diagram of the von Mises stress (MPa) generated on the control arm due to the lateral impact scenario in the present invention. DETAILED DESCRIPTION

[0047] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0048] The present invention provides a method for designing and optimizing the lower control arm of the suspension of an armored wheeled vehicle. The component designed and optimized by the present invention is the lower control arm in the suspension system of an 8×8 wheeled armored military vehicle, which is one of the largest components in the system. The lower control arm is connected to the frame on the vehicle chassis through rotational connections at two different points, and is connected to the lower part of the steering knuckle at the wheel end on the wheel side through a ball joint. The main task of the control arm is to prevent the wheel from uncontrolled movement other than up and down (along the vertical axis) when the vehicle enters any road condition, and to enable each suspension unit to operate independently. An isometric view of the initial design is shown in FIG. Figure 1 shown.

[0049] In the present invention, starting from an existing design, the control arm design optimization problem is formulated as the goal of seeking minimum weight without exceeding the maximum vonMises stress produced in the optimized design by the initial control arm design.

[0050] Find x={d1, d2, d3, d4} (1)

[0051] Minimize W(x) (2)

[0052] subject to σ MaxVM (x)-σ iniMaxVM ≤0 (3)

[0053] x L ≤x≤x U (4)

[0054] In the formula, d1, d2, d3 and d4 in formula (1) represent design variables, which constitute the design variable vector x. W(x) in formula (2) represents the weight of the control arm. Symbol σ MaxVM (x) represents the maximum vonMises stress on the control arm for a given design variable vector x, and the symbol σ iniMaxVM Then represents the maximum vonMises stress generated in the initial control arm design. Considering the size constraints that should not be violated on the control arm, the design variables are limited to x L and the upper limit x U The initial values ​​of the design variables and their upper and lower limits are shown in Table 1:

[0055] Table 1: Initial values ​​of design variables and their upper and lower limits

[0056] Design variables Initial value [mm] Lower limit [mm] Upper limit [mm] <![CDATA[d1]]> 75 75 105 <![CDATA[d2]]> 12 7 12 <![CDATA[d3]]> 103.9 80 103.9 <![CDATA[d4]]> 214.5 180 214.5

[0057] The suspension control arm of a military vehicle is usually made of cast steel or cast iron because of its high strength. The component material used in the present invention is austenitic adjustable cast iron, and its characteristics are shown in Table 2:

[0058] Table 2: Material properties of austenitic adjustable cast iron

[0059] Material properties unit value Density (ρ) <![CDATA[kg·m -3 ]]> 7200 <![CDATA[Yield strength (σ ty )]]> MPa 700 Poisson's ratio (v) - 0.28 Young's modulus (E) GPa 130

[0060] The loads and boundary conditions for the analytical model are determined by selecting a vehicle scenario. This reflects the data obtained from a suspension unit weighing 4,000 kg in static and crash positions, with the measurements being taken when the vehicle arrives at the weighbridge.

[0061] Within the scope of the invention, the loads of the analytical model are applied on the main axes shown in the schematic diagram of the suspension unit, such as Figure 3 The control arm is connected to the steering knuckle at one end through a ball joint and to the vehicle body at the other end through a revolute connection. These boundary conditions were maintained throughout all analyses in this study and the provided analysis result images belong to the submodel created under the main model. The connection points shown as vehicle body connections in the dependent model were assigned as revolute connection conditions in the ANSYS model, while the areas shown as ball joint-steering knuckle connections were assigned as spherical connections in the main model. In addition, a similar connection is also located on the upper control arm, while the upper point of the suspension unit is assigned in the main model.

[0062] In the mesh study, 10-node tetrahedral elements were used, such as Figure 5 As shown. In order to conduct mesh convergence studies, finite element analysis was performed for different element sizes. The maximum stress coordinates obtained are (-122, -221, -17), and the variation of the maximum vonMises stress with respect to the element size is shown, as shown in Figure 6 and Table 3. It can be seen that convergence is achieved at a cell size of 2.5 mm, so this cell size will be used in subsequent optimization studies in this paper. The stress distribution of the converged mesh is shown in Figure 7 For this initial design, the mass is 25.2 kg and the maximum vonMises stress generated in the control arm is 97.80 MPa, as shown in Table 3:

[0063]

[0064]

[0065] After validating the finite element model through convergence studies, it is intended to verify the model by performing field tests on the actual manufactured component. In the present invention, the measurements are performed in the areas where the maximum vonMises stresses are predicted in the finite element model. That is, the stress distribution on the component estimated by finite elements serves as a reference for the physical tests. Figure 8 As shown, the strain gauge device is mounted on the lower control arm of the suspension of an 8×8 vehicle on a workbench.

[0066] Since the test was carried out on a vehicle, stabilizer legs were installed on the vehicle chassis to comply with safety regulations and the test continued with wheel load only. Since this reduced the load on the lower suspension arm, the readings were relatively small. The gain factor of the strain gauge used in the test was 2.125 ± 0.5%.

[0067] The front left axle was raised (by forklift) to three different heights and the strain was measured. Fig. 9As shown in Figure 2, the average strain values ​​corresponding to the three different suspension heights measured are 75, 65, and 50 microstrains, respectively. Note that the strain is measured in the y direction (ε yy )Measured (reference Figure 7 In addition, the comparison between the field test and the finite element analysis results is shown in Table 4 below:

[0068] serial number Wheel force(N) <![CDATA[Predicted ε yy (microstrain)]]> <![CDATA[ε under test yy (microstrain)]]> Percent Difference 1 890 47.8 50.0 -4.4 2 1340 63.8 65.0 -1.9 3 1610 78.9 75.0 +5.2

[0069] Since high-fidelity engineering simulation and analysis models require a long time, and these analyses need to be repeated many times during the design optimization process, surrogate models are often used to reduce computational costs. Surrogate model-based optimization studies usually include defining the design space, experimental design studies, sampling from experiments or simulations, creating surrogate models, and integrating them into approximate optimization algorithms. A typical flow chart of surrogate-based optimization studies is shown in Figure 1. Fig.10 As shown. At the beginning of the design optimization, a number of design variables that are considered important are first identified. However, since a large number of variables may increase the computational cost, less important variables can be eliminated by simplifying the design variables. This is done in the dimensionality reduction phase and can be determined by performing a sensitivity analysis. In order to evaluate different designs, after the surrogate model is generated, the main stage shows that the process of searching for the best design using an optimization algorithm (genetic algorithm, particle swarm optimization, simulated annealing, etc.) continues in the final stage. In the present invention, the multi-objective genetic algorithm (MOGA) available in ANSYS is used, which is a variant of the non-dominated sorting genetic algorithm-II (NSGA-II) based on the concept of controlled elite. MOGA optimizes the objective function through selection, crossover and mutation, striving to optimize the global optimal candidate points until they converge to the maximum allowed Pareto percentage.

[0070] Surrogate models are referred to as meta-models, response surface models, auxiliary models, surrogate models or approximate models in the literature. The most commonly used surrogate methods include response surface, Kriging, radial-based functions, nonparametric regression, artificial neural networks and a collection of surrogate models, which is called genetic aggregation in ANSYS. In the present invention, genetic aggregation in ANSYS is used.

[0071] The training points for the surrogate model are generated by using a central composite design experiment. Since the present invention has four design variables, the central composite design formula results in 2 4 +2×4+1=25 training points. For these training points, finite element analysis is performed to calculate the mass and maximum vonMises stress values. Then, the genetic aggregation agent model in ANSYS is constructed as described above.

[0072] The accuracy of the constructed proxy model is measured by the root mean square error (RMSE), which represents the root mean square of the error (residual). The test points used in this invention are independent of the training points for RMSE calculation. This is a standard and widely used error metric that can be applied to any proxy model. The calculation formula of RMSE is as follows:

[0073]

[0074] Where N is the number of test points, is the response prediction of the ith test point, y i is the actual response of the ith test point.

[0075] In the present invention, five test points were generated. The RMSE calculation results of mass and maximum vonMises stress values ​​were 0.0259 kg and 2.59 MPa, respectively. This paper believes that there is room for improvement in the RMSE of vonMises stress (the target RMSE is less than 2 MPa), so it is decided to further improve the accuracy of the surrogate model. When the test points indicate that the response surface quality is poor or can be improved, these points can be used as refinement points to improve the quality of the response surface. The refined points are solved by a new simulation, similar to the experimental design points, and a new output is obtained, and the response is used as the input for generating the surface. In addition, ANSYS Design Xplorer also allows the creation and verification of surrogate models using refined points, using the universal prediction-based surrogate modeling adaptive refinement technology (UPSMART) algorithm.

[0076] For the genetic aggregation solution, an automatic refinement option is provided. At least one result value (mass or vonMises stress value) must be selected for refinement. Here, the present invention selects vonMises stress. The refined points are then added to the training set to improve the quality of the response surface. Five new test points are generated in the second stage. The genetic aggregation agent model is rebuilt and the RMSE values ​​are re-evaluated to 0.011 kg for mass and 1.69 MPa for maximum vonMises stress. The present invention believes that the target RMSE value has been achieved, so the improvement process is terminated. Finally, the optimization problem previously stated in formula (1) is solved and the results are reported in the next section.

[0077] The training points constructed using the central composite design experiment (DoE) and the corresponding responses are listed in Table 5. The test points used to measure the accuracy of the surrogate model are shown in Table 6. These test points are then used as refinement points in the next stage. The new test points for the next stage are listed in Table 7.

[0078] Table 5 is as follows:

[0079] serial number d1[mm] d2[mm] d3[mm] d4[mm] Mass [kg] Maximum vonMises stress [MPa] 1 90.00 9.50 91.97 197.25 23.55 95.94 2 75.00 9.50 91.97 197.25 24.15 93.95 3 105.00 9.50 91.97 197.25 23.14 97.10 4 90.00 7.00 91.97 197.25 22.69 96.18 5 90.00 12.00 91.97 197.25 24.39 100.58 6 90.00 9.50 80.00 197.25 23.50 93.85 7 90.00 9.50 103.94 197.25 23.61 94.52 8 90.00 9.50 91.97 180.00 23.40 99.42 9 90.00 9.50 91.97 214.49 23.69 98.29 10 79.44 7.74 83.54 185.10 23.25 98.44 11 100.56 7.74 83.54 185.10 22.37 97.05 12 79.44 11.26 83.54 185.10 24.43 100.94 13 100.56 11.26 83.54 185.10 23.57 97.32 14 79.44 7.74 100.40 185.10 23.31 97.56 15 100.56 7.74 100.40 185.10 22.44 95.91 16 79.44 11.26 100.40 185.10 24.50 97.68 17 100.56 11.26 100.40 185.10 23.64 97.91 18 79.44 7.74 83.54 209.39 23.44 94.24 19 100.56 7.74 83.54 209.39 22.57 95.88 20 79.44 11.26 83.54 209.39 24.63 99.78 21 100.56 11.26 83.54 209.39 23.77 94.71 22 79.44 7.74 100.40 209.39 23.53 98.17 23 100.56 7.74 100.40 209.39 22.65 95.52 24 79.44 11.26 100.40 209.39 24.71 93.22 25 100.56 11.26 100.40 209.39 23.85 93.13

[0080] Table 6 is as follows:

[0081]

[0082] Table 7 is as follows:

[0083] serial number d1[mm] d2[mm] d3[mm] d4[mm] Mass [kg] Maximum vonMises stress [MPa] 1 90.00 9.50 91.97 180.00 23.40 95.88 2 90.00 9.50 103.94 197.25 23.61 94.52 3 75.00 9.50 91.97 197.25 24.15 93.95 4 75.00 9.50 91.97 214.49 23.55 95.91 5 100.00 9.50 91.97 180.00 24.02 96.70

[0084] In the optimization step, the multi-objective genetic algorithm (MOGA) method of ANSYSv19.2 software was used. The optimization results and the comparison with the initial design are shown in Table 8. The initial mass of the lower control arm was 25.2 kg, which was reduced by 13.3% to 21.8 kg. This means a reduction of 3.35 kg per component, and a total weight reduction of about 27 kg for an 8×8 wheeled vehicle.

[0085] Table 8 is as follows:

[0086] Optimization parameters Initial Value Optimization value Mass [kg] 25.20 21.85 Maximum vonMises stress [MPa] 97.80 97.76

[0087] Furthermore, the maximum vonMi ses stress of the optimized part did not increase after weight reduction and the deviation of the measured value was less than 1%, which indicated that the optimization was successful. Fig.11 The initial design and the optimal design of the control arm are shown. Fig.12 The vonMi ses stress distribution in the optimal design is shown.

[0088] For the lower control arm that has just completed weight optimization, the present invention will conduct an analysis with quasi-static critical road load scenarios (scenarios that may cause damage and injury) to examine the impact of different situations experienced during driving. Longitudinal impact describes the scenario of a vehicle falling into a deep pit while driving straight. At the same time, lateral impact refers to the scenario where the vehicle wheels hit the road surface at high speed or are hit directly on the lateral axis. Although these scenarios are considered to cause great damage to the suspension components, it is expected that in the results of these scenarios, the stress values ​​of the control arm will be lower than the yield strength value of the component. The safety factor of the calculated results is also calculated. The two different road load scenarios selected are shown in Table 9. Based on these scenarios, the maximum vonMi ses stress values ​​of the lower control arm caused by the loads applied to the wheels are respectively Fig.13 and Fig.14 Given in.

[0089] The results obtained from these scenarios and the safety factors calculated based on these results are shown in Table 10. According to the results, it can be observed that the maximum vonMi ses stress value is lower than the yield strength of the component, a successful result is obtained, and the lateral impact scenario has a greater impact on the control arm than other scenarios. The reason is that the force applied in the Z direction is mainly buffered by the suspension shock absorber and reflects less load on the control arm, while the lateral load is observed to apply more load to the control arm. Table 10 is as follows:

[0090] Road load conditions Maximum vonMises stress [MPa] Safety Factor Case 1 125.3 5.6 Case 2 75.2 9.3

[0091] In the present invention, the lower control arm of the double wishbone suspension system of the 8×8 armored wheeled vehicle is optimized. The mass of the lower control arm is reduced by 13.3% from the initial value of 25.2kg to 21.85kg. For one 8×8 wheeled vehicle, there are 8 lower control arms. This means that the total weight reduction per vehicle is about 27kg. The weight of the vehicle is the most critical issue in high-mobility land vehicles, especially those with swimming requirements. The output of this invention also contributes to the weight reduction research of the entire vehicle.

[0092] The optimized model was analyzed under two quasi-static loading scenarios (off-design conditions) and successfully kept the maximum vonMi ses stress below the yield strength of the component. The safety factors of the components were calculated to be 1.46 and 1.08.

[0093] The content discussed in this invention can be extended to the following studies:

[0094] 1. In order to validate the optimized component in practice, the component can be manufactured and life tested in an independent suspension test setup. This process is particularly important before mass production of safety-critical components.

[0095] 2. Extend off-design condition analysis by adding different scenarios individually, such as destructive road loads when turning.

[0096] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A design optimization method for a lower control arm of an armored wheeled vehicle suspension, characterized in that: The following steps are involved: S1: Define the design variables. The design variables are the length d1, thickness d2, width d3 and length d4 of the lower control arm. Determine that the lower limit of d1 is 75 mm and the upper limit is 105 mm, the lower limit of d2 is 7 mm and the upper limit is 12 mm, the lower limit of d3 is 80 mm and the upper limit is 103.9 mm, and the lower limit of d4 is 180 mm and the upper limit is 214.5 mm; S2: The optimization goal is to minimize the weight of the control arm, and the maximum vonMises stress generated by the optimized design does not exceed the maximum vonMises stress generated in the initial control arm design; S3: Determine the material of the lower control arm; S4: perform finite element modeling; S5: Conduct mesh convergence study; S6: Validate the finite element model by conducting field tests on the actual manufactured components; S7: Perform agent-based optimization; S8: Optimization was performed using the multi-objective genetic algorithm (MOGA) method of ANSYS v19.2 software; S9: Perform quasi-static load scenario analysis on the optimized lower control arm.

2. The method according to claim 1, characterized in that The material of the lower control arm is austenitic adjustable cast iron, and its material properties are: density 7200kg·m 3 , yield strength 700MPa, Poisson's ratio 0.28, and Young's modulus 130GPa.

3. The method according to claim 1, characterized in that: During finite element modeling, the load of the analysis model is applied to the main axis of the suspension unit. One end of the control arm is connected to the steering knuckle through a ball hinge, and the other end is connected to the vehicle body through a rotating connection.

4. The method according to claim 1, characterized in that The mesh convergence study used 10-node tetrahedral elements and achieved convergence at an element size of 2.5 mm.

5. The method according to claim 1, characterized in that: The finite element model was validated by field testing on an actual manufactured component, where a strain gauge device was mounted on the lower control arm of an 8×8 vehicle suspension, the front left axle was raised to different heights and the strain was measured.

6. The method according to claim 1, characterized in that In agent-based optimization, central composite design experiments are used to generate training points for the agent model, genetic aggregation in ANSYS is used to build the agent model, and the root mean square error (RMSE) is used to measure the accuracy of the agent model.

7. The method according to claim 1, characterized in that The optimization process uses the multi-objective genetic algorithm MOGA method of ANSYSv19.2 software. The initial mass of the lower control arm is 25.2 kg, and the mass after optimization is 21.8 kg.

8. The method according to claim 1, characterized in that The optimized lower control arm is analyzed under quasi-static load scenarios, including longitudinal and lateral impact scenarios.

9. The method according to claim 8, characterized in that In the longitudinal impact scenario, the force is 2500N, the loading direction is longitudinal, and it is applied to the ball hinge connection point.

10. The method according to claim 8, characterized in that In the lateral impact scenario, the force is 3000N, the loading direction is lateral, and it is applied to the ball hinge connection point.

Images