Lightweight structure design method for automobile chassis based on topology optimization

Abstract

The application discloses a kind of based on topological optimization's automobile chassis lightweight structure design method, it is related to lightweight design technical field;Macroscopic microcosmic coupling model is constructed, and material parameter fluctuation is quantified with Monte Carlo simulation;Establish rigid-flexible coupling multibody dynamics model, simulate braking, turning and so on Working condition, use rain flow counting method to generate load spectrum;Lightweight, stiffness, modal are included in the construction of multi-objective function, and NSGA-II algorithm is solved, and manufacturing constraints such as draft slope;Divide steel, aluminum, carbon fiber material domain;Fusion bench test data correct model.The application realizes multiscale collaborative optimization, accurately maps dynamic load, balances lightweight and performance;Multi-material gradient design improves connection reliability;Feature recognition and process verification ensure manufacturability;Digital twin and robustness optimization guarantee batch consistency;Overall improve chassis design efficiency and quality, prolong service life.

Description

Technical Field

[0001] This invention relates to the field of lightweight design technology, and in particular to a method for designing lightweight automotive chassis structures based on topology optimization. Background Technology

[0002] As the core component that supports the vehicle body and transmits power, the lightweight design of the automotive chassis plays a crucial role in improving vehicle range and reducing energy consumption. Traditional chassis design relies heavily on experience-based iteration and parameter optimization, resulting in low material utilization and structural redundancy, making it difficult to achieve a balance between lightweighting and performance safety. With the increasing demand for weight reduction in new energy vehicles, simply reducing wall thickness or replacing materials can easily lead to risks such as insufficient stiffness and shortened fatigue life, failing to meet performance requirements under complex operating conditions.

[0003] Existing topology optimization techniques have significant limitations in chassis design: most employ single-scale modeling, neglecting the impact of composite material microstructure on macroscopic performance, leading to large discrepancies between optimized results and actual performance; load mapping is mostly based on static conditions, failing to fully consider the time-varying characteristics of dynamic loads and fatigue damage accumulation, making optimized structures prone to early failure during actual driving; insufficient integration of manufacturing constraints, with topology results often exhibiting poor process feasibility due to sharp corners, thin walls, and other features, requiring extensive manual correction and reducing design efficiency; lack of gradient transition design in multi-material applications, resulting in stress concentration at dissimilar material connection areas, affecting structural reliability. Furthermore, traditional methods do not adequately consider uncertainties such as material parameter fluctuations and manufacturing errors, making it difficult to guarantee the robustness of optimized solutions and consistency in mass production.

[0004] As the automotive industry moves towards intelligent and high-end development, higher demands are being placed on the precision, efficiency, and reliability of lightweight chassis design. Existing technologies are increasingly lacking in multi-scale modeling, dynamic load coupling, multi-material collaborative optimization, and integrated process-performance verification, failing to support the precise design of high-performance chassis. Therefore, developing a topology optimization method that integrates multi-scale analysis, dynamic load mapping, multi-material gradient optimization, and full-process process verification has become crucial to overcoming the technological bottlenecks in automotive chassis lightweighting. Summary of the Invention

[0005] The present invention proposes a lightweight automotive chassis structure design method based on topology optimization to solve the problems mentioned in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a lightweight automotive chassis structure design method based on topology optimization, comprising:

[0007] Collaborative modeling and parameter quantification: A macro-micro cross-scale model is constructed. At the macro level, shell elements are used to discretize the chassis assembly structure, including suspension connection points and crossbeam support areas. At the micro level, a representative volume element (RVE) for composite materials is established to describe the spatial distribution of fibers, matrix, and interface phases. The coupling of macro-micro performance is achieved through homogenization theory. The uncertainty of material parameters is quantified by using Monte Carlo simulation combined with a Kriging proxy model to construct an optimization framework with probabilistic constraints.

[0008] Load mapping and force flow analysis: A rigid-flexible coupled dynamic model was established, and full-load vertical, braking longitudinal, and turning lateral load conditions were simulated in Adams. The torque transmission path of each chassis component was calculated using Timoshenko beam theory, and the force flow density ≥100 N / mm² was extracted. 2 The load nodes are identified; the dynamic load time history curves are statistically analyzed using the rainflow counting method to generate a load spectrum, which is then converted into fatigue damage constraints using the Miner rule.

[0009] Topology optimization model construction: through lightweight coefficients Minimization is the core objective, while also considering stiffness and modal performance. The objective function weights are allocated as follows: lightweight 0.5, stiffness 0.3, and modal performance 0.2. Manufacturing constraints are introduced: draft angle of stamped parts ≥3°, minimum wall thickness 1.2-3mm; welded parts connection area size ≥10mm. Constraints are made explicit through the Heaviside projection function, and the Pareto optimal solution set is solved using the NSGA-II algorithm.

[0010] Material gradient distribution optimization: The design domains of high-strength steel, aluminum alloy, and carbon fiber composite materials are divided. The mechanical properties of the mixed region are described by the material interpolation function E(ρ1, ρ2)=ρ1E1+ρ2E2+(1-ρ1-ρ2)E3. A transition zone of ≥15mm is set at the interface of dissimilar materials. The elastic modulus is smoothly transitioned according to the linear gradient formula E(x)=E1+(E2-E1)(x-x1) / L. A titanium alloy transition layer is added to the steel-aluminum connection area. The layer is formed by anodizing to prevent galvanic corrosion.

[0011] Furthermore, it also includes:

[0012] Topology result feature extraction and reconstruction: Develop a feature recognition algorithm to automatically extract features of stiffeners, hollow structures and variable cross-section regions from the optimization results, and fit the boundaries using B-spline curves; perform geometric reinforcement on the non-design domains of bolt holes and welds, and use the progressive structural optimization (ESO) algorithm to generate radial stiffeners in the connection area, with the stiffener thickness being directly proportional to the connection load;

[0013] Process feasibility verification and optimization: For the stamping process, the forming limit diagram (FLD) was generated by Dynaform simulation to ensure that the principal strain is within the safe zone; for the casting process, the filling process was simulated by ProCAST, and the gate position was optimized to make the Reynolds number ≤2000 to avoid air entrapment and shrinkage defects; orthogonal experiments were designed using the Taguchi method to analyze the influence of mold temperature and stamping speed parameters on structural performance.

[0014] Performance testing and digital twin correction steps: Build a bench test system, implant strain gauges in stress concentration areas, and conduct static stiffness and fatigue tests using a hydraulic servo testing machine; establish a digital twin model, and use a Kalman filter to fuse measured data and simulation results to correct the material's elastic modulus and boundary conditions.

[0015] Robustness optimization and lifetime prediction steps: The 6σ quality management method is adopted, design variable tolerance zones are set, and the performance fluctuation index σ is calculated using the Kriging surrogate model. F ≤3%; Based on the modified Miner's rule and the local stress-strain method, the fatigue life of the component is predicted to be ≥10. 6 The cycle involves adding process fillets in high-stress areas to reduce the stress concentration factor to 1.2.

[0016] Furthermore, in the collaborative modeling, the construction of the RVE model needs to consider the fiber arrangement. For carbon fiber composite materials, the representative volume element size is 5-10 times the fiber diameter, and the micro stress distribution is calculated by the finite element method. When macro-micro coupling occurs, the homogenization theory is used to map the micro equivalent properties to the macro element, and a correction coefficient is introduced during the mapping process to compensate for the scale effect.

[0017] Furthermore, in the load mapping, the acquisition of dynamic loads needs to cover extreme working conditions. Through flexible body processing in the dynamic model, the strain response time history of the component under dynamic loads is obtained. The load mapping algorithm needs to include a coordinate transformation module to transform the global coordinate system load of the dynamic simulation into the local coordinate system load of the topology optimization model, and balance the requirements of normal use and extreme safety through load condition weight allocation.

[0018] Furthermore, in the construction of the topology optimization model, the density variable of the variable density method needs to be sensitively filtered, and the filtering radius is set to 1.5-2 times the unit size to avoid the optimization results from being checkerboard or grid-dependent; the calculation of the lightweight coefficient in the objective function needs to be combined with the chassis wheelbase L and torsional stiffness F, where the torsional stiffness is calculated by applying a ±1° torsional angle, and the target value of the first-order modal frequency needs to be higher than the main frequency of the road excitation to avoid resonance.

[0019] Furthermore, in the optimization of the material gradient distribution, material selection needs to be based on a cost-performance trade-off model, establishing a material cost matrix and performance matrix, and determining the optimal material for each region using the analytic hierarchy process (AHP). The design of the gradient transition zone requires the use of nonlinear gradient functions and exponential functions. k is a gradient coefficient, ensuring that the difference in elastic modulus between the start and end points of the transition zone is ≤5%, thus avoiding stress concentration; the titanium alloy transition layer connecting steel and aluminum needs to undergo surface treatment to form a 5-10μm oxide film, and its corrosion resistance is verified through electrochemical testing.

[0020] Furthermore, in the topology result feature extraction and reconstruction, the feature recognition algorithm needs to include three modules: edge detection, morphological processing, and feature classification. The topology density cloud map is processed to automatically distinguish the direction of the reinforcing ribs, the shape of the hollow structure, and the thickness change rate of the variable cross section. During the reconstruction process, parametric modeling is used for complex features, and geometric tolerance analysis is used to ensure assembly compatibility with surrounding components.

[0021] Furthermore, in the process feasibility verification and optimization, the simulation of the stamping process needs to include the dynamic process of sheet metal forming. The forming force and springback amount are calculated through the simulation results, and mold compensation is used for areas with excessive springback. The simulation of the casting process needs to analyze the filling time, solidification time and temperature field distribution, and optimize the riser design to eliminate shrinkage porosity. The Taguchi test design needs to select parameters such as mold temperature, stamping speed, holding time and cooling rate, with three levels for each parameter, and determine the optimal combination through signal-to-noise ratio (S / N) analysis.

[0022] Furthermore, the lightweighting effect evaluation needs to establish a comprehensive index system, including weight reduction rate, material utilization rate, manufacturing cost change rate, and performance retention rate; the evaluation process requires vehicle-level verification, and the impact of the chassis on the vehicle's ride comfort and handling stability is calculated through ADAMS / Car simulation.

[0023] Compared with existing technologies, the beneficial effects of this invention are:

[0024] Multi-scale collaborative modeling deeply couples macroscopic structure with microscopic material properties, fully leveraging the performance potential of composite materials. It solves the performance prediction bias problem caused by traditional single-scale modeling, making the optimization results more in line with actual use needs and providing a fundamental guarantee for structural reliability.

[0025] Multi-condition load mapping and force flow analysis techniques accurately capture dynamic load characteristics. Combined with fatigue damage constraints, this ensures the durability of the optimized structure under complex driving scenarios and avoids the risk of early failure. The multi-objective topology optimization framework balances lightweighting, stiffness, and modal performance. Through explicit processing of manufacturing constraints, the topology results can meet the requirements of processes such as stamping and casting without extensive manual correction, significantly improving design efficiency.

[0026] Multi-material gradient distribution optimization achieves a smooth transition between different materials. The titanium alloy transition layer design in the dissimilar material connection area effectively reduces stress concentration and improves structural reliability and corrosion resistance. Topology result feature extraction and reconstruction technology automatically identifies key load-bearing features, and parametric modeling ensures the assembly compatibility of the structure with surrounding components, reducing subsequent adjustment costs.

[0027] Process feasibility verification and digital twin correction form a closed-loop optimization, ensuring the stability and consistency of the design scheme in mass production. Robust optimization and life prediction technologies reduce the impact of material fluctuations and manufacturing errors on performance, extending the chassis's service life. Overall, the method in this application comprehensively improves the accuracy, efficiency, and reliability of lightweight chassis design, providing a systematic solution for high-performance lightweight automotive chassis. Attached Figure Description

[0028] Figure 1 This is a schematic block diagram of the lightweight automotive chassis structure design method based on topology optimization proposed in this invention.

[0029] Figure 2 This is a comparative schematic diagram of the stress distribution in the material gradient transition zone of a lightweight automotive chassis structure design method based on topology optimization proposed in this invention.

[0030] Figure 3 This diagram illustrates the impact of manufacturing errors on performance of a lightweight automotive chassis structure design method based on topology optimization proposed in this invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0033] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection or an indirect connection through an intermediate medium; and they may refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The invention will now be described in further detail with reference to the accompanying drawings.

[0034] Reference Figures 1 to 3 A lightweight automotive chassis structure design method based on topology optimization, comprising:

[0035] Collaborative Modeling and Parameter Quantization: A macro-micro cross-scale model is constructed. At the macro level, shell elements (mesh size 5-10mm) are used to discretize the chassis assembly structure, covering key load-bearing areas such as suspension connection points and crossbeam support areas. At the micro level, representative volumetric elements (RVEs) of composite materials are established to accurately describe the spatial distribution of fibers (diameter 0.01-0.1mm), matrix, and interface phases. The coupling of macro- and micro-level performance is achieved through homogenization theory. The uncertainties of material parameters (elastic modulus coefficient of variation 5%, Poisson's ratio fluctuation ±0.02) are quantified. Monte Carlo simulation (sample size ≥1000 times) combined with a Kriging surrogate model is used to construct an optimization framework with probabilistic constraints, ensuring structural reliability ≥95%.

[0036] Load mapping and force flow analysis: A rigid-flexible coupled multibody dynamics model was established, and typical working conditions such as full-load vertical (vehicle weight × 1.2), braking longitudinal (deceleration 0.8g), and cornering lateral (centrifugal acceleration 0.6g) were simulated in Adams. The torque transmission path of each chassis component was calculated using Timoshenko beam theory, and force flow density ≥100N / mm² was extracted. 2 The key load nodes are identified. Rainflow counting is used to statistically analyze the dynamic load time history curves, generating a load spectrum. This spectrum is then converted into fatigue damage constraints using Miner's rule, ensuring that the cumulative damage value is ≤1.

[0037] Topology optimization model construction: with lightweight coefficients Minimization is the core objective, while also considering stiffness (flexibility C) and modal performance (first-order frequency f1). The objective function weights are allocated as follows: lightweight 0.5, stiffness 0.3, and modality 0.2. Manufacturing constraints are introduced: draft angle of stamped parts ≥3°, minimum wall thickness 1.2-3mm (distributed according to load-bearing requirements); welded parts connection area size ≥10mm. Constraints are made explicit using the Heaviside projection function, and the NSGA-II algorithm is used to solve for the Pareto optimal solution set (population size 100, iterations 200).

[0038] Material gradient distribution optimization: Classification of high-strength steel (density 7.8 g / cm³) 3 ), aluminum alloy (2.7g / cm) 3 ), carbon fiber composite material (1.6g / cm) 3 The design domain is described by the material interpolation function E(ρ1, ρ2) = ρ1E1 + ρ2E2 + (1 - ρ1 - ρ2)E3, which describes the mechanical properties of the mixed region. A transition zone of ≥15mm is set at the interface of dissimilar materials, and a smooth transition of the elastic modulus is achieved using the linear gradient formula E(x) = E1 + (E2 - E1)(x - x1) / L. A titanium alloy transition layer (1-2mm thick, overlap length ≥20mm) is added to the steel-aluminum connection area, and an insulating layer (resistance ≥10Ω) is formed after anodizing. 6 Ω·cm 2 To prevent galvanic corrosion.

[0039] Feature extraction and reconstruction of topology results: A feature recognition algorithm was developed to automatically extract key features from the optimization results, such as stiffeners (width ≥ 5 mm), hollow structures (diameter ≥ 8 mm), and variable cross-section regions. Boundaries were fitted using B-spline curves (curvature change rate ≤ 5%). Geometric reinforcement was applied to non-design areas such as bolt holes and welds. The Progressive Structure Optimization (ESO) algorithm was used to generate radial stiffeners in the connection area. The stiffener thickness was directly proportional to the connection load (thickness increased by 0.5 mm for every 1000 N increase in load).

[0040] Process feasibility verification and optimization: For the stamping process, a forming limit diagram (FLD) was generated using Dynaform simulation to ensure that the principal strain is within the safe zone (less than 80% of the material's yield strain). For the casting process, ProCAST was used to simulate the filling process, and the gate position was optimized to ensure a Reynolds number ≤ 2000, avoiding air entrapment and shrinkage cavities. An orthogonal experiment was designed using the Taguchi method to analyze the influence of parameters such as mold temperature (±10℃) and stamping speed (±50mm / s) on structural performance, determining the optimal process window.

[0041] This invention also includes:

[0042] Performance testing and digital twin correction steps: A bench test system was constructed, and strain gauges (measurement accuracy ±1με) were implanted in stress concentration areas (stress gradient ≥50MPa / mm). Static stiffness and fatigue tests were conducted using a hydraulic servo testing machine (loading frequency 5-20Hz). A digital twin model was established, and a Kalman filter was used to fuse measured data and simulation results to correct the material's elastic modulus and boundary conditions, ensuring the simulation error was ≤5%.

[0043] Robustness optimization and life prediction steps: The 6σ quality management method is adopted, the design variable tolerance zone is set (±0.1mm), and the performance fluctuation index σ is calculated using the Kriging surrogate model. F ≤3%. Based on the modified Miner's rule and the local stress-strain method, the fatigue life of key components is predicted to be ≥10. 6 In the high-stress area (amplitude ≥ 200 MPa), process fillets (radius ≥ 2 mm) are added to reduce the stress concentration factor to below 1.2.

[0044] In this invention, the construction of the RVE model in the multi-scale collaborative modeling needs to consider the fiber arrangement (unidirectional, woven, or random distribution). For carbon fiber composite materials, the representative volume element size is 5-10 times the fiber diameter. The micro-stress distribution is calculated using the finite element method to ensure that the interfacial phase shear strength is ≥50MPa. During macro-micro coupling, the homogenization theory is used to map the micro-equivalent properties (elastic matrix, Poisson's ratio matrix) to the macro-element. A correction coefficient (0.95-1.05) is introduced during the mapping process to compensate for the scale effect. Finally, the stiffness calculation error of the macro model is controlled within ±3%.

[0045] In the multi-condition load mapping, the acquisition of dynamic loads needs to cover extreme conditions (such as emergency braking deceleration of 1.0g and extreme turning centrifugal acceleration of 0.8g). Through flexible body processing in the multibody dynamics model (setting the chassis crossbeams and longitudinal beams as flexible bodies and using modal synthesis to reduce their order), the strain response time history of the components under dynamic loads is obtained. The load mapping algorithm needs to include a coordinate transformation module to convert the global coordinate system loads of the dynamic simulation into the local coordinate system loads of the topology optimization model, with a transformation error ≤1%. Furthermore, the algorithm balances normal use and extreme safety requirements through load condition weight allocation (0.6 for normal conditions and 0.4 for extreme conditions).

[0046] In the construction of the multi-objective topology optimization model, the density variable of the variable density method (SIMP model) needs to be sensitively filtered. The filtering radius is set to 1.5-2 times the element size to avoid checkerboard or mesh dependence in the optimization results. The calculation of the lightweight coefficient in the objective function needs to be combined with the chassis wheelbase (L) and torsional stiffness (F). The torsional stiffness is calculated by applying a ±1° torsional angle. The target value of the first-order modal frequency needs to be higher than the main frequency of road excitation (10-20Hz) to avoid resonance. The constraint condition is handled using the Lagrange multiplier method, which transforms stress constraints (≤80% of the material yield strength) and displacement constraints (≤90% of the design threshold) into penalty terms in the objective function. The penalty coefficient increases dynamically with the number of iterations (from 100 to 1000) to enhance the constraint satisfaction.

[0047] In this invention, the multi-material gradient distribution optimization requires material selection based on a cost-performance trade-off model. This involves establishing a material cost matrix (high-strength steel ¥5 / kg, aluminum alloy ¥15 / kg, carbon fiber composite ¥80 / kg) and a performance matrix (specific strength, specific stiffness). The optimal material for each region is determined using the analytic hierarchy process (AHP). The design of the gradient transition zone requires the use of a nonlinear gradient function (such as an exponential function). k is a gradient coefficient, ensuring that the difference in elastic modulus between the start and end points of the transition zone is ≤5%, thus avoiding stress concentration. The titanium alloy transition layer for steel-aluminum connection needs to undergo surface treatment (such as plasma spraying) to form a 5-10μm oxide film, ensuring an interlayer bonding strength ≥100MPa, and verifying corrosion resistance through electrochemical testing (no rust after ≥500 hours of salt spray testing).

[0048] In the topology feature extraction and reconstruction process, the feature recognition algorithm needs to include three modules: edge detection (using the Canny operator), morphological processing (erosion and dilation operations), and feature classification (based on support vector machines). It processes the topology density cloud map (regions with a density ≥ 0.5 are defined as entities), automatically distinguishing the direction of stiffeners (error ≤ 5° along the force flow direction), the shape of hollow structures (circular, square, or irregular), and the thickness variation rate of variable cross-sections (≤ 10% / mm). During reconstruction, complex features are modeled parametrically (e.g., the cross-section of the stiffener is trapezoidal, with an upper base width of 5-8mm, a lower base width of 8-12mm, and a height of 10-15mm), and geometric tolerance analysis (position ± 0.5mm, parallelism ± 0.1mm / m) ensures assembly compatibility with surrounding components.

[0049] In the process feasibility verification and optimization, the stamping process simulation needs to include the dynamic process of sheet metal forming (considering plastic deformation, springback, and a friction coefficient of 0.1-0.2). The forming force (≤80% of the equipment's rated tonnage) and springback amount (≤1mm) are calculated based on the simulation results. For areas with excessive springback, mold compensation is used (the compensation amount is 1.2 times the springback amount). The casting process simulation needs to analyze the filling time (≤5s), solidification time (≤30s), and temperature field distribution (gradient ≤50℃ / mm). The riser design (volume of 10-15% of the casting) is optimized to eliminate shrinkage porosity. The Taguchi experimental design needs to select four key parameters (mold temperature, stamping speed, holding time, and cooling rate), with three levels for each parameter. The optimal combination is determined through signal-to-noise ratio (S / N) analysis, ensuring that the process stability index (CPK value) is ≥1.33.

[0050] In this invention, the lightweighting effect evaluation requires the establishment of a comprehensive index system, including: weight reduction rate ((original weight - optimized weight) / original weight × 100%), material utilization rate (effective load-bearing volume / design domain volume × 100%), manufacturing cost change rate ((optimized cost - original cost) / original cost × 100%), and performance retention rate (optimized performance / original performance × 100%). The evaluation process requires vehicle-level verification, using ADAMS / Car simulation to calculate the chassis's impact on vehicle ride comfort (weighted root mean square acceleration ≤ 0.3 m / s²). 2 The impact of the lightweight chassis on handling stability (steady-state steering gain ≤ 1.2° / (m / s)) is considered to ensure that the performance of the lightweight chassis meets the overall vehicle design requirements.

[0051] Specific implementation methods of lightweight automotive chassis structure design based on topology optimization.

[0052] This embodiment takes the lightweight design of the front subframe of a pure electric SUV chassis as an example to explain in detail the implementation process of the lightweight structure design method for automotive chassis based on topology optimization. The chassis components involved are made of high-strength steel (Q&P980), aluminum alloy (6082-T6), and carbon fiber composite material (T700 / epoxy).

[0053] I. Multi-scale collaborative modeling and parameter quantization

[0054] A macro-micro coupled model was established: At the macro level, a 3D geometric model of the front subframe was discretized using 10mm shell elements, defining key load-bearing areas such as suspension arm connection points and shock absorber mounting seats as the core of the design domain; at the micro level, a carbon fiber composite RVE model was constructed, with a fiber diameter of 0.07mm, a volume fraction of 55%, and an interface phase thickness of 0.0035mm (5% of the fiber diameter). The equivalent elastic constants (longitudinal E1 = 160GPa, transverse E2 = 10GPa) were calculated through periodic boundary conditions. When mapping to the macro model, a correction coefficient of 0.98 was introduced to compensate for the scale effect, and the final macro stiffness error was controlled within 2.5%. Monte Carlo simulation (1000 samples) was used to quantify the material parameter fluctuations, with an elastic modulus variation coefficient of 5%. A probabilistic constraint model P(σ≤[σ])≥0.95 was constructed to ensure structural reliability.

[0055] II. Multi-condition load mapping and force flow analysis

[0056] A rigid-flexible coupled multibody dynamics model was established in Adams, with the front subframe treated as a flexible body (modal synthesis reduced to order 20). Three typical operating conditions were simulated: full-load vertical load (vehicle weight × 1.2 = 32000 N), braking longitudinal load (deceleration 0.8 g, front axle load increased by 60%), and cornering lateral load (centrifugal acceleration 0.6 g, lateral force 12000 N). The force flow transmission path was calculated using Timoshenko beam theory, and a force flow density ≥100 N / mm² was extracted. 2 Key nodes such as the lower swing arm connection point. The dynamic load time history curve is processed by the rainflow counting method (sampling frequency 100Hz), the load spectrum is generated and converted into fatigue damage constraints (damage accumulation value ≤1) through the Miner rule, the coordinate transformation error is controlled within 0.8%, and the working condition weight allocation is 0.6 for normal working condition and 0.4 for extreme working condition.

[0057] III. Construction of Multi-Objective Topology Optimization Model

[0058] The objective function is set to lightweight coefficients. Minimize the weights, assigning a lightweight weight of 0.5, a stiffness (flexibility C) weight of 0.3, and a modal weight (first-order frequency f1) weight of 0.2. Use a SIMP model to describe the material distribution, with the density variable ρ ∈ [0, 1], and the interpolation function E(ρ) = E0ρ. 3 (Penalty factor p = 3), filter radius 15mm (1.5 times unit size) to avoid checkerboard pattern. Constraints: stress ≤ 80% of material yield strength (steel 627MPa, aluminum 224MPa), torsional stiffness ≥ 18000 N·m / °, first-order frequency ≥ 25Hz (avoiding road surface excitation frequency). The NSGA-II algorithm is used for solving, with a population size of 100 and 200 iterations. Constraints are handled using the Lagrange multiplier method, with the penalty coefficient linearly increasing from 100 to 1000.

[0059] IV. Optimization of Gradient Distribution in Multiple Materials

[0060] The design domains are divided into three categories: high-strength steel for the lower control arm connection reinforcement area, aluminum alloy for the main frame, and carbon fiber composite material for lightweight protrusions. The material interpolation function is E(ρ1, ρ2) = ρ1E1 + ρ2E2 + (1 - ρ1 - ρ2)E3. A 20mm transition zone is set at the steel-aluminum interface, and an exponential gradient function is used. Achieving a smooth transition in elastic modulus (endpoint difference ≤4%). A 1.5mm thick titanium alloy transition layer is added to the steel-aluminum connection area, and an 8μm oxide film is formed by plasma spraying. The interlayer bonding strength is 110MPa, no rust is observed after 500 hours of salt spray testing, and the electrochemical resistance is ≥10 Ω·cm. 6 Ω·cm 2 .

[0061] V. Feature Extraction and Reconstruction of Topological Results

[0062] Canny edge detection and support vector machine classification were used to automatically identify reinforcing ribs (8mm width), hollow structures (12mm diameter), and variable cross-section regions in the topological density cloud map (density ≥ 0.5 is considered solid). By fitting the boundary with B-spline curves (curvature change rate ≤ 4%), radial reinforcing ribs were generated around the bolt holes (thickness increased by 0.5mm for every 1000N increase in load). The reinforcing rib cross-section was trapezoidal (top base 6mm, bottom base 10mm, height 12mm). Geometric tolerance analysis ensured positional accuracy ±0.4mm and parallelism ±0.08mm / m, meeting assembly requirements.

[0063] VI. Process Feasibility Verification and Optimization

[0064] Stamping process simulation (Dynaform): The friction coefficient of the aluminum alloy part is 0.15. The forming limit diagram shows that the principal strain is in the safe zone (75% of the yield strain), the springback is 0.8 mm, and the die compensation is 0.96 mm. Casting process simulation (ProCAST): The Reynolds number of the aluminum alloy gate is 1800, the filling time is 4.5 s, the riser volume is 12% of the casting volume, and there are no shrinkage cavities. The Taguchi orthogonal experiment selected 4 parameters and 3 levels, including die temperature (200±10℃) and stamping speed (300±50 mm / s). The signal-to-noise ratio analysis determined the optimal combination, and the CPK value reached 1.42.

[0065] VII. Performance Testing and Digital Twin Correction

[0066] Bench tests were conducted with ±1με precision strain gauges implanted. A hydraulic servo motor (10Hz loading) was used to measure the static stiffness at 18500 N·m / °, and the fatigue cycle was 10. 6The model was virtually crack-free. The digital twin model fused the measured data using a Kalman filter, correcting the elastic modulus parameter (reducing the error to 3.2%), and significantly improving the consistency between simulation and experimental results.

[0067] VIII. Robustness Optimization and Lifetime Prediction

[0068] With a dimensional tolerance of ±0.1mm, the Kriging surrogate model calculates the performance fluctuation σ. F =2.8%≤3%. The fatigue life is predicted to be 1.2 million cycles using the local stress-strain method. A 3mm process fillet is added to the high-stress area (220MPa), and the stress concentration factor is reduced to 1.15.

[0069] Implementation results data:

[0070] index Traditional Design This application's method is optimized. quality 28.5kg 23.5kg Torsional stiffness 16500 N·m / ° 18500 N·m / ° First-order modal frequency 22Hz 27Hz Material utilization rate 62％ 85％ Fatigue life (number of cycles) 800,000 1.2 million

[0071] Data comparison validates the comprehensive optimization effect of this method. While reducing weight by 5 kg (17.5%), torsional stiffness increased by 12.1%, and the first-order modal frequency increased by 22.7%, achieving a win-win situation of lightweighting and performance improvement, thanks to the precise control of stiffness and modal frequency by the multi-objective optimization framework. Material utilization increased from 62% to 85% due to the precise preservation of force flow paths and reduction of structural redundancy by topology optimization. Fatigue life was extended by 50%, attributed to fatigue damage constraints introduced by dynamic load mapping and fillet optimization in high-stress areas, avoiding the risk of early failure. Overall results show that the method in this application significantly improves the level of lightweighting and material utilization efficiency while ensuring chassis safety performance, providing an efficient and reliable technical solution for automotive chassis design.

[0072] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A lightweight automotive chassis structure design method based on topology optimization, characterized in that, include: Collaborative modeling and parameter quantification: A macro-micro cross-scale model is constructed. At the macro level, shell elements are used to discretize the chassis assembly structure, including suspension connection points and crossbeam support areas. At the micro level, a representative volume element (RVE) for composite materials is established to describe the spatial distribution of fibers, matrix, and interface phases. The coupling of macro-micro performance is achieved through homogenization theory. The uncertainty of material parameters is quantified by using Monte Carlo simulation combined with a Kriging proxy model to construct an optimization framework with probabilistic constraints. Load mapping and force flow analysis: A rigid-flexible coupled dynamic model was established, and full-load vertical, braking longitudinal, and turning lateral load conditions were simulated in Adams. The torque transmission path of each chassis component was calculated using Timoshenko beam theory, and the force flow density ≥100 N / mm² was extracted. 2 The load nodes are identified; the dynamic load time history curves are statistically analyzed using the rainflow counting method to generate a load spectrum, which is then converted into fatigue damage constraints using the Miner rule. Topology optimization model construction: through lightweight coefficients Minimization is the core objective, while also considering stiffness and modal performance. The objective function weights are allocated as follows: lightweight 0.5, stiffness 0.3, and modal performance 0.

2. Manufacturing constraints are introduced: draft angle of stamped parts ≥3°, minimum wall thickness 1.2-3mm; welded parts connection area size ≥10mm. Constraints are made explicit through the Heaviside projection function, and the Pareto optimal solution set is solved using the NSGA-II algorithm. Material gradient distribution optimization: The design domains are divided into high-strength steel, aluminum alloy, and carbon fiber composite materials, and optimization is achieved through material interpolation functions. Describe the mechanical properties of the mixed region; set a transition zone of ≥15mm at the interface of dissimilar materials, according to the linear gradient formula. To achieve a smooth transition in elastic modulus, a titanium alloy transition layer is added to the steel-aluminum connection area, which is then anodized to form an insulating layer to prevent galvanic corrosion.

2. The lightweight automotive chassis structure design method based on topology optimization according to claim 1, characterized in that, Also includes: Feature extraction and reconstruction of topological results: A feature recognition algorithm was developed to automatically extract features of stiffeners, hollow structures, and variable cross-section regions from the optimization results, and the boundaries were fitted by B-spline curves. Geometric reinforcement was performed on the non-design domains of bolt holes and welds, and radial stiffeners were generated in the connection area using the progressive structural optimization (ESO) algorithm. The stiffener thickness was directly proportional to the connection load. Process feasibility verification and optimization: For the stamping process, the forming limit diagram (FLD) was generated by Dynaform simulation to ensure that the principal strain is within the safe zone; for the casting process, the filling process was simulated by ProCAST, and the gate position was optimized to make the Reynolds number ≤2000 to avoid air entrapment and shrinkage defects; orthogonal experiments were designed using the Taguchi method to analyze the influence of mold temperature and stamping speed parameters on structural performance. Performance testing and digital twin correction steps: Build a bench test system, implant strain gauges in stress concentration areas, and conduct static stiffness and fatigue tests using a hydraulic servo testing machine; A digital twin model was established, and the measured data and simulation results were fused using a Kalman filter to correct the material's elastic modulus and boundary conditions.

3. The lightweight automotive chassis structure design method based on topology optimization according to claim 1, characterized in that, Also includes: Robustness optimization and life prediction steps: The 6σ quality management method is adopted, design variable tolerance bands are set, and performance fluctuation indices are calculated using the Kriging surrogate model. Based on the modified Miner's rule and the local stress-strain method, the fatigue life of the component is predicted to be ≥10. 6 The cycle involves adding process fillets in high-stress areas to reduce the stress concentration factor to 1.

2.

4. The lightweight automotive chassis structure design method based on topology optimization according to claim 1, characterized in that, In the collaborative modeling, the construction of the RVE model needs to consider the fiber arrangement. For carbon fiber composite materials, the representative volume element size is 5-10 times the fiber diameter, and the micro stress distribution is calculated by the finite element method. When macro-micro coupling occurs, the homogenization theory is used to map the micro equivalent properties to the macro element. In the mapping process, a correction coefficient is introduced to compensate for the scale effect.

5. The lightweight automotive chassis structure design method based on topology optimization according to claim 1, characterized in that, In the load mapping, the acquisition of dynamic loads needs to cover extreme working conditions. The strain response time history of the component under dynamic loads is obtained through flexible body processing in the dynamic model. The load mapping algorithm needs to include a coordinate transformation module to transform the global coordinate system load of the dynamic simulation into the local coordinate system load of the topology optimization model, and balance the requirements of normal use and extreme safety through load condition weight allocation.

6. The lightweight automotive chassis structure design method based on topology optimization according to claim 1, characterized in that, In the construction of the topology optimization model, the density variable of the variable density method needs to be sensitively filtered. The filtering radius is set to 1.5-2 times the unit size to avoid the optimization results from being chessboard-like or grid-dependent. The calculation of the lightweight coefficient in the objective function needs to be combined with the chassis wheelbase L and torsional stiffness F. The torsional stiffness is calculated by applying a ±1° torsional angle. The target value of the first-order modal frequency needs to be higher than the main frequency of the road excitation to avoid resonance.

7. The lightweight automotive chassis structure design method based on topology optimization according to claim 1, characterized in that, In the optimization of the material gradient distribution, material selection needs to be based on a cost-performance trade-off model, establishing a material cost matrix and performance matrix, and determining the optimal material for each region using the analytic hierarchy process (AHP). The design of the gradient transition zone requires the use of nonlinear gradient functions and exponential functions. k is a gradient coefficient, which ensures that the difference between the elastic modulus at the beginning and end of the transition zone is ≤5% to avoid stress concentration; the titanium alloy transition layer connecting steel and aluminum needs to be surface treated to form a 5-10μm oxide film, and its corrosion resistance is verified by electrochemical testing.

8. The lightweight automotive chassis structure design method based on topology optimization according to claim 2, characterized in that, In the topology result feature extraction and reconstruction, the feature recognition algorithm needs to include three modules: edge detection, morphological processing, and feature classification. The topology density cloud map is processed to automatically distinguish the direction of the reinforcing ribs, the shape of the hollow structure, and the thickness change rate of the variable cross section. During the reconstruction process, parametric modeling is used for complex features, and geometric tolerance analysis is used to ensure assembly compatibility with surrounding components.

9. The lightweight automotive chassis structure design method based on topology optimization according to claim 2, characterized in that, In the process feasibility verification and optimization, the simulation of the stamping process needs to include the dynamic process of sheet metal forming. The forming force and springback amount are calculated through the simulation results, and mold compensation is used for areas with excessive springback. The simulation of the casting process needs to analyze the filling time, solidification time and temperature field distribution, and optimize the riser design to eliminate shrinkage porosity. The Taguchi experimental design needs to select parameters such as mold temperature, stamping speed, holding time and cooling rate, with three levels for each parameter, and determine the optimal combination through signal-to-noise ratio (S / N) analysis.

10. The lightweight automotive chassis structure design method based on topology optimization according to claim 1, characterized in that, The evaluation of the lightweighting effect needs to establish a comprehensive index system, including weight reduction rate, material utilization rate, manufacturing cost change rate, and performance retention rate; the evaluation process requires vehicle-level verification, and the impact of the chassis on the ride comfort and handling stability of the vehicle is calculated through ADAMS / Car simulation.

Images