Multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data

Abstract

This invention provides a multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles (IoV) big data, applied to the field of lightweighting technology for new energy vehicles. The method includes the following steps: generating a comprehensive weight by integrating subjective and objective weights; constructing a multi-task topology optimization objective function using a compromise programming method; performing a first-level overall frame topology optimization using the SIMP variable density method to determine the number and spatial layout of crossbeams; constructing a rigid-flexible coupling virtual prototype to extract the equivalent static load at the longitudinal / crossbeam connections, and performing a second-level cross-sectional topology optimization based on extrusion process constraints to obtain the optimal longitudinal / crossbeam cross-sectional configuration. Through this multi-level optimization architecture of "data-driven weight allocation, overall layout optimization, and local cross-sectional collaborative design," the frame lightweighting rate is increased by 15%-22%, while bending and torsional stiffness are increased by 12%-18% and 10%-15%, respectively, while ensuring the process feasibility of aluminum alloy extrusion.

Description

Technical Field

[0001] The present invention belongs to the technical field of lightweight new energy vehicles, and specifically relates to a multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data. Background Art

[0002] With global climate change and the energy crisis becoming increasingly severe, lightweight design for electric commercial vehicles has become a core issue in the industry. Aluminum alloy, with its high specific strength, excellent corrosion resistance, and recyclability, is an ideal material for achieving lightweighting. However, automotive companies currently face significant technical bottlenecks in developing all-aluminum alloy frame structures.

[0003] Existing design experience is largely limited to traditional steel frames, treating all-aluminum alloy frame design as a simple material replacement and failing to fully exploit the unique advantages of aluminum alloy in processing technology. In particular, frame cross-section design still relies on steel-based processes, failing to fully realize the potential of aluminum alloy in manufacturing. Therefore, an innovative design approach for aluminum alloy frames is urgently needed.

[0004] To determine the optimal material distribution for vehicle frame structures, topology optimization has been introduced into the field of vehicle frame design and has become a hot topic of research. Domestic and international scholars have conducted extensive research on commercial vehicle frame topology optimization, but existing technologies still face the following prominent challenges:

[0005] 1. Irrational weight distribution for multiple working conditions: Traditional methods rely on experimental design or expert experience to assign weights, which fails to fully reflect the load distribution in real user scenarios. This causes the optimization results to be out of line with actual working conditions and makes it difficult to meet complex and changing operational needs.

[0006] 2. Structural Optimization Limitations: Existing methods often focus on the topological optimization of the overall frame structure, while the design of longitudinal and transverse beam cross-sections remains based on traditional steel frame concepts, without incorporating the specific material properties of aluminum alloys for targeted optimization. Furthermore, they lack a "global-local" collaborative optimization strategy and fail to fully consider the constraints of the aluminum alloy extrusion process, resulting in the failure to fully realize the potential of lightweighting.

[0007] In summary, how to provide a multi-level topology optimization method for aluminum alloy frames that is more in line with the actual usage scenarios of electric commercial vehicles has become a key technical problem that needs to be urgently solved by those skilled in the art. Summary of the Invention

[0008] In view of the above problems in the prior art, the purpose of the present invention is to provide a multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data.

[0009] The multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data includes the following steps:

[0010] S1. Collect and analyze big data from the Internet of Vehicles, and determine objective weights based on the proportion of typical vehicle operating conditions. The typical vehicle operating conditions include bending conditions, torsion conditions, steering conditions and braking conditions;

[0011] S2. Determine subjective weights based on the analytic hierarchy process ;

[0012] S3. Objective weight , subjective weight Perform weight fusion to obtain comprehensive weight ;

[0013] S4. Construct multi-condition topology optimization objective function;

[0014] S5. Build and optimize a first-level topology optimization model to obtain an overall topology structure of the vehicle frame, wherein the overall topology structure of the vehicle frame includes the number, size, and distribution of crossbeams of the frame;

[0015] S6. Build and optimize the second-level topology optimization model to obtain the topological structure of the longitudinal / transverse beam section of the frame;

[0016] S7. Based on the overall frame topology obtained by S5 and the longitudinal / transverse beam cross-section topology obtained by S6, an optimization design is performed in combination with process constraints to obtain the final aluminum alloy frame structure.

[0017] Preferably, the specific process of step S1 is as follows:

[0018] S1.1. Collecting IoV big data to construct an analysis sample, wherein the IoV big data includes at least: time information, latitude and longitude information, road condition information, acceleration information, braking information, and steering information;

[0019] S1.2. Preprocess the collected Internet of Vehicles big data;

[0020] S1.3. Statistical analysis is performed on the pre-processed Internet of Vehicles big data to obtain the proportion of typical vehicle operating conditions and form objective weights. .

[0021] Preferably, the specific process of preprocessing the collected Internet of Vehicles big data is as follows:

[0022] S1.2.1. Convert the format of the raw data of the analysis sample;

[0023] S1.2.2. Establish a spatiotemporal correlation model based on time information and latitude and longitude information to obtain the corresponding relationship between time and vehicle position changes;

[0024] S1.2.3. Identify anomalies based on spatiotemporal correlation models and perform data cleaning on anomalies;

[0025] S1.2.4. Smooth and denoise the cleaned data.

[0026] Preferably, statistical analysis is performed on the pre-processed Internet of Vehicles big data to obtain the proportion of typical vehicle operating conditions and form an objective weight. The specific process is as follows:

[0027] S1.3.1. Operating condition feature extraction: Based on the parameters of the analysis sample, a comprehensive vector of the vehicle operating condition is constructed through principal component analysis;

[0028] S1.3.2. Operating Condition Cluster Analysis: Classify the comprehensive vector of vehicle operating conditions through the unsupervised self-learning K-means cluster analysis algorithm, construct the vehicle's typical user usage scenario and driving condition state space, determine the proportion of vehicles in different operating conditions, and set the proportion of typical vehicle conditions as the objective weight. .

[0029] Preferably, the specific process of step S2 is as follows:

[0030] S2.1. Construct a judgment matrix based on the importance scale of typical working conditions;

[0031] S2.2. Calculate the eigenvectors, weights, and eigenroots of the judgment matrix using the analytic hierarchy process.

[0032] S2.3. Calculate the random consistency ratio CR and determine whether the constructed judgment matrix meets the consistency requirements;

[0033] The calculation formula of random consistency ratio CR is as follows:

[0034]

[0035]

[0036] Where, is the maximum characteristic root, CI is the consistency index, n is the order of the judgment matrix, RI is the random consistency index;

[0037] S2.4. When the judgment matrix meets the consistency requirements, the weight value of each typical working condition is obtained to determine the subjective weight ,When the judgment matrix does not meet the consistency requirements, the judgment matrix is ​​reconstructed.

[0038] Preferably, comprehensive weight The calculation formula is as follows:

[0039]

[0040] Where, For the j The comprehensive weight of each working condition, For the j The objective weight of each working condition, No. j The subjective weight of each working condition.

[0041] Preferably, the calculation formula of the multi-condition topology optimization objective function is as follows:

[0042]

[0043] Where, is the weighted strain energy of the structure under M working conditions; For the k The comprehensive weight of each working condition; It is i The density of the units, It is i The volume of a unit; For the k The total structural flexibility of each working condition; For the k The maximum value of the total flexibility of the structure obtained by single-objective topology optimization under each working condition is: For the k The minimum value of the total flexibility of the structure obtained by the single-objective topology optimization of each working condition, p is the penalty factor, is the optimized structural volume, is the volume of the topological region, f is the volume fraction, which refers to the upper limit of the percentage of the optimized structure volume to the volume of the topological region.

[0044] Preferably, the specific process of step S5 is as follows:

[0045] S5.1. Construct an initial geometric model for topology optimization based on the original geometric features and external dimensions of the sampled vehicle frame.

[0046] S5.2. Discretize the initial geometric model using the finite element method to generate a finite element model;

[0047] S5.3. Based on the structural characteristics of the sample frame, divide the finite element model into topological areas and non-topological areas;

[0048] S5.4. Based on the load characteristics of various typical vehicle operating conditions, set corresponding boundary conditions and construct a topological model of the overall vehicle frame;

[0049] S5.5. Perform single-condition optimization on the overall topological model of the frame: Based on the SIMP variable density method, set flexibility minimization as the optimization target and obtain the maximum flexibility value of the optimization process under different typical working conditions. and minimum flexibility value ;

[0050] S5.6, multi-condition optimization of the overall topology model of the frame: the maximum flexibility value of each typical working condition obtained by S5.5 , minimum flexibility value And the comprehensive weight obtained through S3 , calculate the objective function of the multi-condition topology optimization constructed in S4, set the flexibility minimization as the optimization goal, and perform multi-condition topology optimization design on the overall topology model of the frame using the SIMP variable density method;

[0051] S5.7. Obtain the overall topological structure of the vehicle frame including the number, size, and distribution of the frame crossbars.

[0052] Preferably, the specific process of step S6 is as follows:

[0053] S6.1. Load extraction of the second-level topology optimization model: Build a rigid-flexible coupled virtual prototype model of the entire vehicle with the sample frame as the flexible body. Perform simulation analysis of typical vehicle operating conditions to obtain the equivalent static load at the connection between the frame longitudinal beam and cross beam under different typical operating conditions.

[0054] S6.2. Second-level topology optimization model construction: Comprehensively consider the envelope space of the sampled frame longitudinal / transverse beams and the dimensions of each beam in the first-level frame overall topology optimization structure, determine the topology optimization space of each frame longitudinal / transverse beam, and construct the initial geometric model for the second-level topology optimization;

[0055] Discretize the initial geometric model into a finite element model using hexahedral mesh elements;

[0056] The nodes with connection structures in the frame longitudinal / transverse beam topology optimization space are captured using RBE3 elements, and the multi-condition equivalent static loads extracted from the virtual prototype model are applied step by step to each longitudinal / transverse beam connection for loading;

[0057] The inertia release method is used to perform virtual constraint processing on the second-level topology optimization model, and then the topology optimization model of each longitudinal / transverse beam section of the frame is established;

[0058] S6.3. Perform single-condition topology optimization on the topology optimization models of each longitudinal / transverse beam section of the frame: Based on the SIMP variable density method, set flexibility minimization as the optimization goal and obtain the maximum flexibility value of the optimization process under different typical working conditions. and minimum flexibility value ;

[0059] S6.4. Perform multi-condition topology optimization on the topology optimization models of each longitudinal / transverse beam section of the frame: the maximum flexibility value of each typical working condition obtained by S6.3 , minimum flexibility value And the comprehensive weight obtained through S3 , calculate the objective function of the multi-condition topology optimization constructed in S4, set flexibility minimization as the optimization goal, and perform multi-condition topology optimization design of the topology optimization model of each longitudinal / transverse beam section of the frame using the SIMP variable density method;

[0060] S6.5 obtains the longitudinal / transverse beam cross-section topology of the frame.

[0061] Preferably, the extraction process of the equivalent static load in step S6.1 is as follows:

[0062] By compiling a Class B virtual random road and writing drive control programs for constant speed, steering, and braking conditions, a virtual test field for vehicle bending, steering, and braking conditions was built;

[0063] By programming a single-raised obstacle course, the right wheels of the target vehicle are forced to pass over the triangular obstacles in sequence, thus building a virtual test field for torsional conditions.

[0064] Through virtual simulation tests, the load-time history of the frame longitudinal / transverse beam connection was obtained, and the equivalent static load was obtained by the equivalent static method.

[0065] The beneficial effects of the present invention are:

[0066] 1. Data-driven, precise weight allocation for operating conditions. By integrating a dual weighting mechanism based on cluster analysis of connected vehicle data and expert experience, this approach breaks through the limitations of traditional objective analysis or empirical weight allocation, establishing an objective weighting model based on real-world user scenarios. Compared to traditional methods, this operating condition weight allocation better aligns with real-world user scenarios and effectively resolves the mismatch between optimization results and actual load spectra.

[0067] 2. Multi-level collaborative optimization architecture innovation. A two-level topology optimization system, "overall layout-component cross-section," is established to achieve collaborative design of the frame's macroscopic layout and local structural features. First-level overall topology optimization determines the number and distribution of crossbeams. Second-level longitudinal / crossbeam cross-section optimization, combined with extrusion process constraints, increases the frame's lightweighting by 15%-20% and improves bending stiffness by 12%-18%.

[0068] 3. Deep integration of aluminum alloy process constraints. Vertical draft constraints, minimum member size constraints (≥60mm), and extrusion constraints are simultaneously introduced during the topology optimization phase, allowing the optimization results to be directly applied to the aluminum alloy extrusion forming process, reducing the number of subsequent engineering design workflows and iterations. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0070] Figure 1 This is a flow chart of the multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data of the present invention;

[0071] Figure 2 This is a flow chart of the vehicle networking big data analysis method of the present invention;

[0072] Figure 3 It is a schematic diagram of topological area division of the first-level topological optimization of the vehicle frame involved in the present invention;

[0073] Figure 4 This is the first-level topology optimization multi-working condition flexibility iteration curve of the vehicle frame involved in the present invention;

[0074] Figure 5 This is a schematic diagram of the first-level topology optimization results of the vehicle frame involved in the present invention;

[0075] Figure 6 This is a schematic diagram of the second-level topology optimization results of the vehicle frame involved in the present invention;

[0076] Figure 7 This is the vehicle frame engineering design structure provided by the present invention.

[0077] Figure numerals: 1. front crossbeam; 2. front crossbeam of power battery system; 3. crossbeam of first bracket of battery box; 4. crossbeam of second bracket of battery box; 5. rear crossbeam of power battery system; 6. saddle beam; 7. tail crossbeam. DETAILED DESCRIPTION

[0078] Example 1

[0079] like Figure 1 As shown in the figure, the multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data is based on real-world usage scenario data of electric commercial vehicles. Through a multi-level collaborative topology optimization strategy, the collaborative topology optimization of the overall frame topology structure and the longitudinal / transverse beam cross-section topology structure is achieved. The specific steps include:

[0080] S1. Statistical analysis of collected IoV big data and determination of objective weights .

[0081] like Figure 2 As shown, determine the objective weight The specific steps are as follows:

[0082] S1.1. Collect IoV big data to construct an analysis sample. IoV big data from the target vehicle is collected through the vehicle terminal as a research sample for statistical analysis. This IoV big data includes information such as vehicle chassis number, time, latitude and longitude, road conditions, altitude, speed, mileage, acceleration, braking, steering, motor speed, and battery SOC.

[0083] In this embodiment, the Internet of Vehicles big data of 200 electric tractors of two models of a commercial vehicle manufacturer in 2023 was collected as an analysis sample. The analysis sample includes at least time information, latitude and longitude information, road condition information, acceleration information, braking information, and steering information.

[0084] S1.2. Preprocess the collected Internet of Vehicles big data.

[0085] The specific process is as follows:

[0086] S1.2.1. Convert the format of the raw data of the analysis samples.

[0087] S1.2.2. Establish a spatiotemporal correlation model based on time information and latitude and longitude information to obtain the corresponding relationship between time and vehicle position changes.

[0088] S1.2.3. Identify abnormal situations based on the spatiotemporal correlation model, use the outlier detection algorithm to clean the data of abnormal situations, and eliminate invalid data introduced by factors such as equipment failure and transmission errors.

[0089] S1.2.4. Smooth and denoise the cleaned data using the SG filter. The SG filter can effectively remove noise interference while retaining the main features of the data, which is conducive to further improving data quality.

[0090] S1.3. Statistical analysis is performed on the pre-processed Internet of Vehicles big data to obtain typical vehicle operation conditions and their proportions, and form objective weights. Typical vehicle operating conditions include bending, twisting, steering and braking.

[0091] The specific process is as follows:

[0092] S1.3.1. Operating condition feature extraction: Based on the user operating condition characterization parameters in the analysis sample, the principal component analysis method is used to construct a comprehensive vector of the vehicle operating condition.

[0093] S1.3.2. Operating Condition Cluster Analysis: Classify the comprehensive vector of vehicle operating conditions through the unsupervised self-learning K-means cluster analysis algorithm, construct the vehicle's typical user usage scenarios and driving condition state space, determine the proportion of vehicles in different operating conditions, and set the typical condition proportion as the objective weight for multi-condition topology optimization .

[0094] The distribution of typical working conditions of the target vehicle in this implementation is as follows: bending condition accounts for 71.5%, torsion condition accounts for 6.1%, steering condition accounts for 12.8%, and braking condition accounts for 9.6%. Therefore, the objective weights for subsequent multi-working condition topology optimization are obtained. =(0.71, 0.06, 0.13, 0.10).

[0095] S2. Obtain the subjective weight of engineering experts based on the analytic hierarchy process .

[0096] The specific process is as follows:

[0097] S2.1. Experienced engineering experts conduct subjective evaluations of the importance of each vehicle operating condition: A judgment matrix is ​​constructed based on the importance scale of typical operating conditions. The importance scale is shown in Table 1, and the judgment matrix is ​​shown in Table 2.

[0098] Table 1 Importance scale

[0099]

[0100] Table 2 Judgment Matrix

[0101]

[0102] S2.2. Calculate the eigenvectors, eigenvector weights, and eigenroots of each operating condition in the judgment matrix using the AHP (Analytical Hierarchy Process). The results of the AHP (Analytical Hierarchy Process) analysis in this embodiment are shown in Table 3.

[0103] Table 3 AHP hierarchical analysis results

[0104]

[0105] S2.3, according to the maximum characteristic root And the random consistency index RI, calculate the consistency index CI and random consistency ratio CR, the calculation formula is as follows:

[0106]

[0107]

[0108] Specifically, the random consistency index RI and the judgment matrix order n The relationship between them is shown in Table 4:

[0109] Table 4 Random consistency index RI and judgment matrix order n relationship

[0110]

[0111] If the random consistency ratio CR value is less than 0.1, it is considered that the judgment matrix meets the consistency requirements. If the random consistency ratio CR value is greater than 0.1, it is considered that the judgment matrix does not meet the consistency requirements and the judgment matrix needs to be reconstructed by experts.

[0112] S2.4. Determine subjective weights based on the results of AHP , subjective weight is the matrix of weight values ​​of the characteristic vectors of typical working conditions of each vehicle.

[0113] The order of the judgment matrix in this embodiment n The calculated random consistency ratio CR value is less than 0.1, and it is considered that the judgment matrix has satisfactory consistency, so the subjective weight is obtained. =(0.12, 0.53, 0.29, 0.06).

[0114] S3. Objective weight , subjective weight Perform weight fusion to obtain comprehensive weight .

[0115] The objective weights obtained from the analysis of Internet of Vehicles big data are calculated using the square root average method. and subjective weights based on analytic hierarchy process Fusion is performed to obtain a comprehensive weight matrix , improve the robustness of the weight coefficient. The calculation formula is as follows:

[0116]

[0117] Where, For the j The comprehensive weight of each working condition, For the j The objective weight of each working condition, No. j The subjective weight of each working condition.

[0118] S4. Construct the multi-condition topology optimization objective function.

[0119] Taking into account the performance requirements of the frame structure under multiple working conditions, a multi-working condition weighted strain energy objective function is established based on the compromise planning method as the objective function of multi-working condition topology optimization. Its mathematical expression is as follows:

[0120]

[0121] Where, is the weighted strain energy of the structure under M working conditions; For thek The comprehensive weight of each working condition; It is i The density of the units, It is i The volume of a unit; For the k The total structural flexibility of each working condition; For the k The maximum value of the total flexibility of the structure obtained by single-objective topology optimization under each working condition is: For the k The minimum value of the total flexibility of the structure obtained by the single-objective topology optimization of each working condition, p is the penalty factor, is the optimized structural volume, is the volume of the topological region, f is the volume fraction, which refers to the upper limit of the percentage of the optimized structure volume to the volume of the topological region, where f is a custom constant, f The value range is 0.2-0.4, preferably 0.3.

[0122] S5. Build and optimize the first-level topology optimization model to obtain the overall topology structure of the frame, including the number, size, and distribution of the frame crossbeams.

[0123] The specific process is as follows:

[0124] S5.1. Construct an initial geometric model for topology optimization based on the original geometric features and external dimensions of the sampled vehicle frame.

[0125] S5.2. Use the finite element method to discretize the initial geometric model and generate a finite element model.

[0126] S5.3. Based on the structural characteristics of the sample frame, the finite element model is divided into topological areas and non-topological areas. Figure 3 As shown, the non-topological area includes the front, rear, and left and right edge areas of the finite element model, specifically including the front crossbeam 1, the tail crossbeam 7, and the longitudinal beams. The rest of the finite element model is set as the topological area.

[0127] S5.4. Based on the load characteristics of the four typical vehicle operating conditions, set corresponding boundary conditions and construct the overall frame topology model. The boundary conditions include lug constraints, mass point constraints, acceleration constraints, and dynamic load constraints.

[0128] Among them, applying eye constraints: setting bending conditions, braking conditions, and steering conditions to constrain the translational degrees of freedom of all eyes of the front suspension in the Y and Z directions, the translational degrees of freedom of the front eyes on both sides of the rear suspension in the X, Y, and Z directions, and the translational degrees of freedom of the rear eyes on both sides of the rear suspension in the Y and Z directions.

[0129] Set the torsional working condition to constrain the translational degrees of freedom of the left hanging eye of the front suspension in the Y and Z directions, the translational degrees of freedom of the front hanging eyes on both sides of the rear suspension in the X, Y, and Z directions, and the translational degrees of freedom of the rear hanging eyes on both sides of the rear suspension in the Y and Z directions, and release all degrees of freedom of the right hanging eye of the front suspension.

[0130] Applying mass point constraints: The main loads acting on the electric commercial vehicle frame are set as mass points at their respective center of mass. The RBE3 element is used to simulate the connection between the mass points and the frame to load the frame. Specifically, the main loads acting on the electric commercial vehicle frame include: the cab assembly, saddle load (cargo box and cargo), power battery pack, battery assembly, controller assembly, and brake air compressor assembly.

[0131] Apply dynamic acceleration constraints: gravity acceleration is applied to bending, twisting, steering, and braking conditions g , additional 0.4 is applied during braking g Braking inertia force, additional 0.6 applied in cornering conditions g Lateral inertia force.

[0132] Apply dynamic load constraints: set a dynamic load factor of 2 for bending conditions, and a dynamic load factor of 1.2 for torsion conditions, braking conditions, and cornering conditions.

[0133] S5.5. Single-condition optimization of the overall frame topology model: Use the SIMP variable density method to perform single-condition topology optimization design of the overall frame structure, set flexibility minimization as the optimization goal, and obtain the maximum flexibility value of the optimization process under different typical working conditions. and minimum flexibility value .

[0134] In this embodiment, the maximum compliance value of each single working condition is obtained based on the target vehicle and minimum flexibility value As shown in Table 5:

[0135] Table 5 Compliance values ​​for each working condition

[0136]

[0137] Among them, the cell density of the topological area is used as the design variable, and the penalty factor is set p Set it to 3, set the vertical draft direction and symmetry constraint, and the minimum member size to 60mm, that is, the minimum beam width is 60mm.

[0138] S5.6. Perform multi-condition optimization on the overall topology model of the vehicle frame: Based on the objective function of the multi-condition topology optimization, use the SIMP variable density method to perform multi-condition topology optimization design on the overall topology model of the vehicle frame, thereby obtaining the overall topology structure of the vehicle frame, including the number, size, and distribution of the frame crossbeams.

[0139] Specifically, the maximum compliance value of each typical working condition obtained by S5.5 is , minimum flexibility value And the comprehensive weight obtained through S3 , calculate the objective function of the multi-condition topology optimization constructed in S4, set the flexibility minimization as the optimization goal, and perform the multi-condition topology optimization design of the overall topology model of the frame through the SIMP variable density method.

[0140] Among them, the volume fraction of 0.3 is set as the constraint condition, the cell density of the topological area is used as the design variable, the penalty factor is set to 3, the vertical draft direction and symmetry constraint are set, and the minimum member size is 60 mm.

[0141] S5.7. Obtain the overall topological structure of the vehicle frame including the number, size, and distribution of the frame crossbars.

[0142] This embodiment is based on the multi-condition flexibility iteration curve after 11 iterations of the target vehicle parameters. Figure 4 As shown, the overall topological structure of the frame is as follows Figure 5 As shown, the topological area includes the power battery system front beam 2, the battery box first bracket crossbeam 3, the battery box second bracket crossbeam 4, the power battery system rear beam 5, and the saddle beam 6.

[0143] S6. Construct and optimize the second-level topology optimization model to obtain the longitudinal / transverse beam cross-section topology structure of the frame.

[0144] The specific process is as follows:

[0145] S6.1. Load extraction of the second-level topology optimization model: Establish a rigid-flexible coupled virtual prototype model of the entire vehicle with the sample frame as the flexible body, and simulate and analyze the typical operating conditions of the vehicle to obtain the equivalent static load at the connection between the frame longitudinal beam and the cross beam under different typical operating conditions.

[0146] Specifically, a rigid-flexible coupling virtual prototype model of the entire vehicle was established using the multi-body dynamics software ADAMS / CAR, in which the connection surfaces between the longitudinal / transverse beams of the frame were constrained together by rigid units to facilitate load extraction.

[0147] The process of payload extraction is as follows:

[0148] By compiling a Class B virtual random road and writing drive control programs for uniform speed, steering and braking conditions, a virtual test field for vehicle bending, steering and braking conditions is built.

[0149] By compiling a single-raised obstacle road, the right wheels of the target vehicle are made to pass over the triangular obstacles in turn, building a virtual test field for torsional working conditions.

[0150] Through virtual simulation tests, the load-time history of the frame longitudinal / transverse beam connection was obtained, and the equivalent static load was obtained by the equivalent static method.

[0151] S6.2. Construction of the second-level topology optimization model: Comprehensively consider the envelope space of the sampled frame longitudinal / transverse beams and the dimensions of each beam in the first-level frame overall topology optimization structure, determine the topology optimization space of each longitudinal / transverse beam of the frame, construct the initial geometric model of the second-level topology optimization, and use hexahedral mesh units to discretize the initial geometric model into a finite element model.

[0152] The nodes with possible connection structures in the frame longitudinal / transverse beam topology optimization space are captured using RBE3 elements, and the multi-condition equivalent static loads extracted from the virtual prototype model are applied step by step to each longitudinal / transverse beam connection for loading;

[0153] The inertia release method is used to perform virtual constraint processing on the second-level topology optimization model, and then the topology optimization models of each longitudinal / transverse beam section of the frame are established.

[0154] S6.3. Perform single-condition topology optimization on the topology optimization models of each longitudinal / transverse beam section of the vehicle frame: Use the SIMP variable density method to perform single-condition topology optimization design on each longitudinal / transverse beam section structure to obtain the maximum flexibility value of the optimization process under different working conditions for each longitudinal / transverse beam structure. and minimum flexibility value .

[0155] Minimizing flexibility is set as the optimization goal, the volume fraction of 0.3 is set as the constraint condition, the cell density of the topological area is used as the design variable, the penalty factor is set to 3, the extrusion constraint is set, and the minimum member size is 60mm; the maximum flexibility value of each longitudinal / transverse beam structure of the frame under different working conditions is obtained. and minimum flexibility value .

[0156] S6.4. Perform multi-condition topology optimization on the topology optimization models of each longitudinal / transverse beam section of the frame: Based on the objective function of the multi-condition topology optimization, use the SIMP variable density method to perform multi-condition topology optimization design on the topology models of each longitudinal / transverse beam section to obtain the topological structure of each longitudinal / transverse beam section.

[0157] Specifically, the maximum flexibility of each working condition calculated above is , minimum flexibility The comprehensive weight W is brought into the multi-condition topology optimization objective function, and the SIMP variable density method is used to perform multi-condition topology optimization design of each longitudinal / transverse beam section structure of the frame. Minimization of flexibility is set as the optimization goal, the volume fraction 0.3 is set as the constraint condition, the unit density of the topological area is used as the design variable, the penalty factor 3 is set, the extrusion constraint is set, and the minimum member size is 60 mm. Finally, the second-level frame longitudinal / transverse beam section structure is obtained.

[0158] In this embodiment, the topology optimization results of the second-level frame longitudinal / transverse beam cross-section structure obtained based on the above method are as follows: Figure 6 shown.

[0159] S7. Based on the overall topological structure of the frame and the topological structures of each longitudinal / transverse beam section of the frame, combined with process constraints and optimization design, the aluminum alloy frame structure engineering design is obtained.

[0160] Among them, the final design scheme obtained based on the target vehicle in this embodiment is as follows Figure 7 shown.

[0161] like Figure 5-Figure 7 As shown, Figure 5 This is the axonometric drawing of the first-level topology optimization result; Figure 6 This is the result of the second-level topology optimization. Specifically, the positions of the beams are marked on the top view of the first-level topology optimization result to facilitate the visualization of their specific positions. Figure 7 The frame engineering design structure is obtained based on the first-level topology optimization results and the second-level topology optimization results.

[0162] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein with equivalents. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data, characterized by: The steps include: S1. Collect and analyze big data from the Internet of Vehicles, and determine objective weights based on the proportion of typical vehicle operating conditions. The typical vehicle operating conditions include bending conditions, twisting conditions, steering conditions and braking conditions; S2. Determine subjective weights based on the analytic hierarchy process ; S3. Objective weight , subjective weight Perform weight fusion to obtain comprehensive weight ; S4. Construct multi-condition topology optimization objective function; S5. Build and optimize a first-level topology optimization model to obtain an overall topology structure of the vehicle frame, wherein the overall topology structure of the vehicle frame includes the number, size, and distribution of crossbeams of the frame; S6. Build and optimize the second-level topology optimization model to obtain the topological structure of the longitudinal / transverse beam section of the frame; S7, based on the overall frame topology obtained in S5 and the longitudinal / transverse beam cross-section topology obtained in S6, combined with process constraints, an optimization design is performed to obtain the final aluminum alloy frame structure; in: The specific process of step S5 is as follows: S5.

1. Construct an initial geometric model for topology optimization based on the original geometric features and external dimensions of the sampled vehicle frame. S5.

2. Discretize the initial geometric model using the finite element method to generate a finite element model; S5.

3. Based on the structural characteristics of the sample frame, divide the finite element model into topological areas and non-topological areas; S5.

4. Based on the load characteristics of various typical vehicle operating conditions, set corresponding boundary conditions and construct a topological model of the overall vehicle frame; S5.

5. Perform single-condition optimization on the overall topological model of the frame: Based on the SIMP variable density method, set flexibility minimization as the optimization target and obtain the maximum flexibility value of the optimization process under different typical working conditions. and minimum flexibility value ; S5.6, multi-condition optimization of the overall topology model of the frame: the maximum flexibility value of each typical working condition obtained by S5.5 , minimum flexibility value And the comprehensive weight obtained through S3 , calculate the objective function of the multi-condition topology optimization constructed in S4, set the flexibility minimization as the optimization goal, and perform multi-condition topology optimization design on the overall topology model of the frame using the SIMP variable density method; S5.

7. Obtain the overall topological structure of the vehicle frame, including the number, size, and distribution of the frame crossbars; The specific process of step S6 is as follows: S6.

1. Load extraction of the second-level topology optimization model: Build a rigid-flexible coupled virtual prototype model of the entire vehicle with the sample frame as the flexible body. Perform simulation analysis of typical vehicle operating conditions to obtain the equivalent static load at the connection between the frame longitudinal beam and cross beam under different typical operating conditions. S6.

2. Second-level topology optimization model construction: Comprehensively consider the envelope space of the sampled frame longitudinal / transverse beams and the dimensions of each beam in the first-level frame overall topology optimization structure, determine the topology optimization space of each frame longitudinal / transverse beam, and construct the initial geometric model for the second-level topology optimization; Discretize the initial geometric model into a finite element model using hexahedral mesh elements; The nodes with connection structures in the frame longitudinal / transverse beam topology optimization space are captured using RBE3 elements, and the multi-condition equivalent static loads extracted from the virtual prototype model are applied step by step to each longitudinal / transverse beam connection for loading; The inertia release method is used to perform virtual constraint processing on the second-level topology optimization model, and then the topology optimization model of each longitudinal / transverse beam section of the frame is established; S6.

3. Perform single-condition topology optimization on the topology optimization models of each longitudinal / transverse beam section of the frame: Based on the SIMP variable density method, set flexibility minimization as the optimization goal and obtain the maximum flexibility value of the optimization process under different typical working conditions. and minimum flexibility value ; S6.

4. Perform multi-condition topology optimization on the topology optimization models of each longitudinal / transverse beam section of the frame: the maximum flexibility value of each typical working condition obtained by S6.3 , minimum flexibility value And the comprehensive weight obtained through S3 , calculate the objective function of the multi-condition topology optimization constructed in S4, set flexibility minimization as the optimization goal, and perform multi-condition topology optimization design of the topology optimization model of each longitudinal / transverse beam section of the frame using the SIMP variable density method; S6.5 obtains the longitudinal / transverse beam cross-section topology of the frame.

2. The multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data according to claim 1 is characterized in that: The specific process of step S1 is as follows: S1.

1. Collecting IoV big data to construct an analysis sample, wherein the IoV big data includes at least: time information, latitude and longitude information, road condition information, acceleration information, braking information, and steering information; S1.

2. Preprocess the collected Internet of Vehicles big data; S1.

3. Statistical analysis is performed on the pre-processed Internet of Vehicles big data to obtain the proportion of typical vehicle operating conditions and form objective weights. .

3. The multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data according to claim 2 is characterized in that: The specific process of preprocessing the collected Internet of Vehicles big data is as follows: S1.2.

1. Convert the format of the raw data of the analysis sample; S1.2.

2. Establish a spatiotemporal correlation model based on time information and latitude and longitude information to obtain the corresponding relationship between time and vehicle position changes; S1.2.

3. Identify anomalies based on spatiotemporal correlation models and perform data cleaning on anomalies; S1.2.

4. Smooth and denoise the cleaned data.

4. The multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data according to claim 2 is characterized in that: Perform statistical analysis on pre-processed Internet of Vehicles big data to obtain the proportion of typical vehicle operating conditions and form objective weights The specific process is as follows: S1.3.

1. Operating condition feature extraction: Based on the parameters of the analysis sample, a comprehensive vector of the vehicle operating condition is constructed through principal component analysis; S1.3.

2. Operating Condition Cluster Analysis: Classify the comprehensive vector of vehicle operating conditions through the unsupervised self-learning K-means cluster analysis algorithm, construct the vehicle's typical user usage scenario and driving condition state space, determine the proportion of vehicles in different operating conditions, and set the proportion of typical vehicle conditions as the objective weight. .

5. The multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data according to claim 1 is characterized in that: The specific process of step S2 is as follows: S2.

1. Construct a judgment matrix based on the importance scale of typical vehicle operating conditions; S2.

2. Calculate the eigenvectors, weights, and eigenroots of the judgment matrix using the analytic hierarchy process. S2.

3. Calculate the random consistency ratio CR and determine whether the constructed judgment matrix meets the consistency requirements; The calculation formula of random consistency ratio CR is as follows: Where, is the maximum characteristic root, CI is the consistency index, n is the order of the judgment matrix, RI is the random consistency index; S2.

4. When the judgment matrix meets the consistency requirements, the weight value of each typical working condition is obtained to determine the subjective weight ,When the judgment matrix does not meet the consistency requirements, the judgment matrix is ​​reconstructed.

6. The multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data according to claim 1 is characterized in that: Comprehensive weight The calculation formula is as follows: Where, For the j The comprehensive weight of each working condition, For the j The objective weight of each working condition, No. j The subjective weight of each working condition.

7. The multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data according to claim 1 is characterized in that: The calculation formula of the multi-condition topology optimization objective function is as follows: Where, is the weighted strain energy of the structure under M working conditions; For the k The comprehensive weight of each working condition; It is i The density of the units, It is i The volume of a unit; For the k The total structural flexibility of each working condition; For the k The maximum value of the total flexibility of the structure obtained by single-objective topology optimization under each working condition is: For the k The minimum value of the total flexibility of the structure obtained by the single-objective topology optimization of each working condition, p is the penalty factor, is the optimized structural volume, is the volume of the topological region, f is the volume fraction, which refers to the upper limit of the percentage of the optimized structure volume to the volume of the topological region.

8. The multi-level collaborative topology optimization method for aluminum alloy frames based on Internet of Vehicles big data according to claim 1 is characterized in that: The extraction process of the equivalent static load in step S6.1 is as follows: By compiling a Class B virtual random road and writing drive control programs for constant speed, steering, and braking conditions, a virtual test field for vehicle bending, steering, and braking conditions was built; By programming a single-raised obstacle course, the right wheels of the target vehicle are forced to pass over the triangular obstacles in sequence, thus building a virtual test field for torsional conditions. Through virtual simulation tests, the load-time history of the frame longitudinal / transverse beam connection was obtained, and the equivalent static load was obtained by the equivalent static method.

Images