Method and device for determining front crossbeam structure of front seat

By obtaining structural constraint information and using the key area identification network for optimization, the problem of multi-objective conflict in the front beam structural design of the front seat is solved, and the multi-objective balance and optimization accuracy of the structural design are achieved.

CN118918304BActive Publication Date: 2024-12-03ZHEJIANG LONGJI AUTO PARTS
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411389682.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-10-08
Publication Date
2024-12-03
Estimated Expiration
2044-10-08

AI Technical Summary

Technical Problem

The front beam structure design of the front seat front beam needs to meet multiple constraints at the same time, resulting in the problem of multi-objective conflict, such as increasing strength to meet the needs of collision safety, but at the same time, weight reduction is required to improve fuel efficiency or range.

Method used

By obtaining structural constraint information, the initial structural image is generated, and inputting it into the key area identification network for key area identification, the optimization algorithm is used to determine the key areas that meet the various constraints, and optimize it to obtain the final structural image to achieve the multi-objective balance.

Benefits of technology

It effectively solves the conflict problem between different requirements in traditional design, reduces the workload of manual identification, improves the accuracy of the optimization process, and achieves multi-objective balance of structural design.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118918304B_ABST
    Figure CN118918304B_ABST
Patent Text Reader

Abstract

The present application is applicable to the field of automobile structural technology, and in particular to a method and device for determining the structure of a front crossbeam of a front row seat, the method comprising: obtaining structural constraint information of the front crossbeam of the front row seat; obtaining an initial structural image of the front crossbeam of the front row seat according to the structural constraint information; inputting the initial structural image and each constraint information in the structural constraint information into a key area recognition network for key area recognition to obtain key area information; based on the key area information and the structural constraint information, determining the key areas in all key areas of the key area information that meet the structural constraint information using an optimization algorithm to obtain target area information; optimizing the initial structural image according to the target area information to obtain the final structural image of the front crossbeam of the front row seat. The method effectively solves the conflict problem between different requirements in the traditional design process.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the technical field of automobile structures, and in particular relates to a method and device for determining a front crossbeam structure of a front row seat. Background Art

[0002] The front seat cross member refers to a part of the structure in front of the front seats inside the car. The front seat cross member is used to support the dashboard and other front components and provide additional structural strength for the vehicle. It is located in front of the driver and co-pilot seats, near the foot area. This structure plays a very important role in car design. It not only provides physical support, but also helps absorb and disperse impact force in the event of a collision to protect the safety of passengers.

[0003] In the prior art, the structural design of the front cross beam of the front seat needs to meet multiple constraints at the same time, and there may be conflicts between each constraint. For example, in order to meet the collision safety requirements, the front cross beam of the front seat needs to have sufficient strength, which may increase the weight of the cross beam, but at the same time, it is also necessary to reduce the weight of the cross beam as much as possible to improve the vehicle's fuel efficiency or the cruising range of electric vehicles.

[0004] In summary, there are multiple objective conflicts in the process of determining the front crossbeam structure of the front seats. Summary of the invention

[0005] The embodiments of the present application provide a method and device for determining a front cross beam structure of a front row seat, which can solve the problem of multiple target conflicts in the process of determining the front cross beam structure of a front row seat in the related art.

[0006] In a first aspect, an embodiment of the present application provides a method for determining a front crossbeam structure of a front seat, comprising:

[0007] Acquiring structural constraint information of the front cross beam of the front seat; wherein the structural constraint information includes vehicle structural constraint information, collision safety constraint information, structural function constraint information, and vehicle performance constraint information;

[0008] Acquiring an initial structural image of the front cross beam of the front seat according to the structural constraint information;

[0009] Inputting the initial structural image and each constraint information in the structural constraint information into a key area recognition network to perform key area recognition to obtain key area information; wherein the key area recognition network is used to identify the key area that can be optimized in the initial structural image according to each constraint information;

[0010] Based on the key area information and the structural constraint information, an optimization algorithm is used to determine the key area that satisfies the structural constraint information among all the key areas in the key area information to obtain the target area information;

[0011] The initial structural image is optimized according to the target area information to obtain a final structural image of the front cross beam of the front seat.

[0012] The above technical solutions in the embodiments of the present application have at least the following technical effects:

[0013] The method for determining the structure of the front beam of the front row seat provided in the present application first obtains the structural constraint information of the front beam of the front row seat, then obtains the initial structural image of the front beam of the front row seat according to the structural constraint information, and then inputs the initial structural image and each constraint information in the structural constraint information into the key area recognition network for key area recognition to obtain the key area information, and based on the key area information and the structural constraint information, uses the optimization algorithm to determine the key areas that meet the structural constraint information in all key areas of the key area information to obtain the target area information, and finally optimizes the initial structural image according to the target area information to obtain the final structural image of the front beam of the front row seat. This method can strike a balance between multiple objectives, avoid the performance or safety defects caused by single-objective optimization, and effectively solve the conflict between different requirements in the traditional design process. This method introduces a key area recognition network to automatically identify the optimizable areas in the structure, greatly reducing the workload of manual identification in traditional design, while improving the accuracy of the optimization process.

[0014] In a second aspect, an embodiment of the present application provides a device for determining a front crossbeam structure of a front row seat, comprising:

[0015] An acquisition unit, used to acquire structural constraint information of a front cross beam of a front row of seats; wherein the structural constraint information includes vehicle structural constraint information, collision safety constraint information, structural function constraint information, and vehicle performance constraint information;

[0016] An initial image generating unit, used for acquiring an initial structural image of the front cross beam of the front seat according to the structural constraint information;

[0017] An identification unit, used for inputting the initial structural image and each constraint information in the structural constraint information into a key area identification network for key area identification to obtain key area information; wherein the key area identification network is used for identifying the key area that can be optimized in the initial structural image according to each constraint information;

[0018] A target region obtaining unit, based on the key region information and the structural constraint information, uses an optimization algorithm to determine the key regions satisfying the structural constraint information among all the key regions in the key region information, and obtains the target region information;

[0019] The optimization unit is used to optimize the initial structural image according to the target area information to obtain a final structural image of the front cross beam of the front seat.

[0020] In a third aspect, an embodiment of the present application provides a device for determining a front cross beam structure of a front row seat, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any one of the embodiments of the first aspect when executing the computer program.

[0021] It can be understood that the beneficial effects of the second to third aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0023] Figure 1 It is a flow chart of a method for determining a front cross beam structure of a front row seat provided in an embodiment of the present application;

[0024] Figure 2 is a schematic diagram of the structure of a key area identification network for implementing step S300 in the method for determining the front cross beam structure of a front row seat provided in an embodiment of the present application;

[0025] Figure 3 4 is a schematic diagram of the implementation flow of step S400 in the method for determining the front cross beam structure of a front row seat provided in an embodiment of the present application;

[0026] Figure 4 is a schematic structural diagram of a device for determining a front crossbeam structure of a front row seat provided in an embodiment of the present application;

[0027] Figure 5 It is a structural schematic diagram of a device for determining the front cross beam structure of a front seat provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0029] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0030] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0031] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0032] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0033] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0034] In the related art, the structural design of the front cross beam of the front seat needs to meet multiple constraints at the same time, and there may be conflicts between each constraint. For example, in order to meet the collision safety requirements, the front cross beam of the front seat needs to have sufficient strength, which may increase the weight of the cross beam, but at the same time, it is also necessary to reduce the weight of the cross beam as much as possible to improve the vehicle's fuel efficiency or the cruising range of electric vehicles.

[0035] To solve the above problems, the embodiment of the present application provides a method and device for determining the structure of the front beam of the front row of seats. In the method, the structural constraint information of the front beam of the front row of seats is first obtained, and then the initial structural image of the front beam of the front row of seats is obtained according to the structural constraint information, and then the initial structural image and each constraint information in the structural constraint information are respectively input into the key area recognition network for key area recognition to obtain the key area information, and based on the key area information and the structural constraint information, the optimization algorithm is used to determine the key areas that meet the structural constraint information in all key areas of the key area information to obtain the target area information, and finally the initial structural image is optimized according to the target area information to obtain the final structural image of the front beam of the front row of seats. The method can strike a balance between multiple objectives, avoid the performance or safety defects caused by single-objective optimization, and effectively solve the conflict between different performance requirements in the traditional design process. The method introduces a key area recognition network to automatically identify the optimizable areas in the structure, greatly reducing the workload of manual identification in traditional design, while improving the accuracy of the optimization process.

[0036] The method for determining the front cross beam structure of a front row seat provided in the embodiment of the present application can be applied to a device for determining the front cross beam structure of a front row seat. In this case, the device for determining the front cross beam structure of a front row seat is the executor of the method for determining the front cross beam structure of a front row seat provided in the embodiment of the present application. The embodiment of the present application does not impose any restriction on the specific type of the device for determining the front cross beam structure of a front row seat.

[0037] For example, the device for determining the front cross beam structure of the front seat can be a tablet computer, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), a desktop computer, a smart large screen, a computing device or other processing device connected to a wireless modem, a computer, a laptop computer, customer premise equipment (CPE) and / or other devices used to communicate on a wireless system and a next-generation communication system, such as a mobile terminal in a 5G network or a mobile terminal in a future evolved public land mobile network (PLMN), etc.

[0038] In order to better understand the method for determining the front cross beam structure of the front row seat provided in the embodiment of the present application, the specific implementation process of the method for determining the front cross beam structure of the front row seat provided in the embodiment of the present application is exemplarily introduced below.

[0039] Figure 1A schematic flow chart of a method for determining a front cross beam structure of a front row seat provided in an embodiment of the present application is shown. The method for determining a front cross beam structure of a front row seat includes:

[0040] S100, obtaining structural constraint information of a front cross beam of a front row seat, wherein the structural constraint information includes vehicle structural constraint information, collision safety constraint information, structural function constraint information, and vehicle performance constraint information.

[0041] It can be understood that the vehicle structure constraint information describes the design limitations of the front crossbeam of the front seats in the overall structure of the vehicle. For example, the installation location, size requirements, material requirements, and the connection method with the vehicle chassis, seats, and other parts of the vehicle body. The collision safety constraint information is related to the safety performance of the vehicle during a collision, which is conducive to the crossbeam being able to provide safety protection when a collision occurs. It may include the deformation behavior requirements of the crossbeam during a collision, the ability to absorb collision energy, and the safety of protecting passengers. The structural function constraint information reflects the functional requirements of the crossbeam, such as supporting seats, fixing seat belts, and bearing loads from seats and passengers. The vehicle performance constraint information reflects the requirements related to the performance of the entire vehicle, which may include the weight requirements of the crossbeam, the balance of stiffness and strength, and the impact on the vehicle's handling, stability, and comfort.

[0042] For example, engineers can enter specific structural constraints into the system one by one through input interfaces (such as keyboards, mice, or other input devices) according to project requirements and design specifications. The input can be in the form of text data, tabular data, or parametric input of a graphical interface. The input information can be in a standardized format or a file format supported by a specific software tool, such as XML, JSON, or directly input into the user interface of the simulation analysis tool. The system will store the input structural constraint information in a database or memory for subsequent structural design, simulation, and optimization. The input constraint information is logically verified to ensure that all constraints are reasonable and complete, and the structural constraint information is parsed for use in model design and optimization algorithms.

[0043] This step lays the foundation for the subsequent design of the front crossbeam structure of the front seats.

[0044] S200: Acquire an initial structural image of a front cross beam of a front row seat according to the structural constraint information.

[0045] For example, the structural constraint information can be parsed, and key parameters such as the size, material, shape, position of the crossbeam, and related load conditions and performance indicators can be extracted from the parsed structural constraint information. Based on the extracted parameters, the initial structural model of the front crossbeam of the front seat can be automatically generated using the modeling function in the CAD (computer-aided design) software or CAE (computer-aided engineering) tool. The parametric modeling can dynamically adjust the size, shape and structure of the crossbeam according to the input constraint information. According to the material requirements in the vehicle structural constraint information, the generated initial structural model can be given corresponding material properties, such as strength, density, elastic modulus, etc. According to the collision safety constraint information, some energy absorption structures or deformation zones can be integrated into the model. The design of the energy absorption structure or deformation zone can be based on the energy absorption requirements during the collision, so that the front crossbeam of the front seat can effectively protect the passengers during the collision. At the same time, the corresponding connection points and installation locations in the initial structural model can also be designed according to the functional constraint information, such as seat support points, seat belt fixing interfaces, etc.

[0046] When the 3D model of the beam and its related mechanical properties, functions and crash safety features are all integrated, the initial structural image is output, which can be a 3D rendered model or CAD drawing, showing the geometry and key design features of the beam. The structural image is visualized by a graphics rendering engine to show the structural form of the beam for engineers to view and evaluate. The structural image may include 3D geometry (wireframe or solid rendering), material distribution and performance description, etc.

[0047] In a possible implementation, S200, acquiring an initial structural image of a front cross beam of a front seat according to structural constraint information, includes:

[0048] S210: Generate a first structure image according to the structure constraint information.

[0049] For example, geometric modeling is performed based on the vehicle structural constraint information to generate a preliminary 3D structural image of the front crossbeam of the front seat. Suitable materials such as aluminum alloy, steel or composite materials can be selected according to the requirements of lightweight vehicles. Combined with the installation space requirements of the whole vehicle structure for the crossbeam, a reasonable size and shape are designed to facilitate the correct connection of the crossbeam with other parts of the vehicle. The installation points and support points of the crossbeam are designed to facilitate the effective connection of the crossbeam with components such as seats and body structures.

[0050] When designing the geometry, you can also consider crash safety requirements, which can be the force and deformation behavior of the beam in the event of a collision. Design specific areas of the beam so that the beam can effectively absorb energy in the event of a collision and avoid excessive stress concentration.

[0051] It can be designed to support critical areas of the seat, where it can maintain sufficient strength to withstand the weight and shock loads of the passengers. It can also be designed to provide seat belt mounting points that are strong and rigid enough to protect the passengers in a collision.

[0052] Combined with vehicle performance constraint information, the design of the crossbeam is further optimized to achieve a balance between performance and structure. The weight of the crossbeam can be reduced, the distribution of the crossbeam material can be optimized, and unnecessary materials can be removed, while ensuring that the structural strength is not affected. Adding structural support to the key stress-bearing area of ​​the crossbeam is beneficial to the rigidity and anti-twisting ability of the crossbeam during vehicle driving.

[0053] Based on the above steps, the first structural image of the front cross beam of the front seat is generated. The first structural image can be a 3D geometric model, showing the complete shape of the cross beam. The first structural image is the preliminary design of the front cross beam of the front seat, which provides a basis for subsequent simulation, analysis and optimization.

[0054] S220: Fusing the first structural image with the vehicle structural image in the vehicle structural constraint information to generate a second image, wherein the second image is used to reflect the position and relative relationship of the front cross beam of the front seat in the overall vehicle structure.

[0055] It is understood that the vehicle structure image can be a 3D geometric model of the entire vehicle, including other important structures of the vehicle, such as chassis, body, seat installation positions, etc. The vehicle structure image reflects the overall structure of the vehicle and provides a relative reference for the front cross beam of the front seat.

[0056] Exemplarily, if the first structural image and the vehicle structural image are modeled in different coordinate systems, an alignment operation can be performed. Generally, the reference point of the vehicle model is the center of the vehicle body or the fixed point of the chassis, and the crossbeam model can be aligned with reference to its installation position. The front crossbeam of the front seat can be located below or in front of the front seat of the vehicle, connected to the seat bracket, chassis and other supporting structures. The connection points of the crossbeam and other vehicle structures (such as chassis, seat bracket, etc.) can be defined, and the connection points can be fixed through preset interfaces or mounting brackets. The first structural image (the model of the front crossbeam of the front seat) is imported into the vehicle structural image, and the crossbeam is aligned with the relevant position in the vehicle structure by matching the connection points of the crossbeam. For example, the mounting hole or fixing point of the crossbeam is aligned with the corresponding position of the seat bracket. Adjusting the posture of the crossbeam (rotation, movement, etc.) is conducive to the precise docking of the crossbeam with the vehicle chassis and seat mounting parts. Once the geometry and installation position of the crossbeam match the overall structure of the vehicle, the second image can begin to form. The second image not only shows the shape of the front crossbar of the front seats, but also reflects the position of the crossbar in the entire vehicle and its relationship to other components.

[0057] During the embedding and fusion process, computer-aided design (CAD) tools can be used to check whether the front crossbar of the front seat has geometric conflicts with other parts of the vehicle. For example, ensure that the crossbar does not overlap or interfere with other components of the vehicle, such as the chassis and seats. After the front crossbar of the front seat is accurately integrated with the overall structure of the vehicle, a new 3D image (second image) is generated.

[0058] S230: Generate a third image based on the first structural image and the collision safety constraint information, wherein the third image is used to reflect the stress distribution and energy absorption of the front cross beam of the front seat in the collision simulation.

[0059] For example, the first structural image can be imported into simulation software (such as ANSYS, LS-DYNA, etc.) to establish a physical model, which can accurately reflect the geometry and material properties of the beam. According to the collision safety constraint information, material properties are assigned to the beam. For example, the elastic modulus, yield strength, density, etc. of aluminum alloy, steel or composite material. The boundary conditions of the collision scene are set, such as the fixing point of the beam (the part connected to the vehicle chassis and seat bracket), the speed, direction and force distribution of the collision.

[0060] The collision speed and angle of the beam can be set when the vehicle is hit head-on or sideways according to the collision safety standards. For example, the collision scenario includes a head-on collision or side collision at 40 km / h. The shape and material properties of the collision objects, such as rigid obstacles or flexible vehicle models, can be defined to simulate actual collision scenarios.

[0061] The finite element method can be used for collision simulation to simulate the stress, deformation process and energy absorption mechanism of the front crossbeam of the front seat when a collision occurs, and output the stress distribution of each node of the crossbeam during the collision process. According to the stress distribution of each node of the crossbeam during the collision, a stress distribution diagram of the crossbeam in the collision simulation and an image of the energy absorbed by the crossbeam during the collision are generated. The stress distribution image can show the stress level of the crossbeam in different areas through color gradients. For example, red may represent high stress areas, while blue represents low stress areas. High stress areas are places where stress is concentrated during a collision, which can be connection points or parts with the greatest force, and may require enhanced design; low stress areas indicate that the material is used less efficiently and may be potential optimization areas. The energy absorption image can show how much energy the crossbeam absorbs at different time nodes and how the energy is distributed on the structure through time series graphs or color markings.

[0062] The stress distribution diagram and the energy absorption diagram are integrated to obtain a third image, which provides a visual collision simulation result and helps to evaluate whether the design of the beam meets the collision safety requirements and whether further optimization is needed.

[0063] S240, identifying the functional area in the first structural image according to the structural functional constraint information to obtain a fourth image, wherein the fourth image is used to reflect the functional area of ​​the front cross beam of the front seat.

[0064] It can be understood that the functional areas of the front cross beam of the front seats may include: seat support points, which are the installation points of the front seats. The cross beam is required to have sufficient rigidity and strength at the installation points to bear the weight of the seats and passengers; seat belt fixing points, which are the connection points of the seat belts. This area is required to be able to withstand a large tensile force during a collision, which is conducive to the normal functioning of the seat belts; chassis connection points, which are the connection areas between the cross beam and the vehicle chassis, which are conducive to the cross beam being firmly fixed in the vehicle frame; energy absorption areas. In some cases, the front cross beam of the front seats may also have a specific energy absorption function to protect the occupants in the vehicle during a vehicle collision.

[0065] For example, the first structural image of the front cross beam of the front seat can be imported into the design software or simulation platform. According to the structural function constraint information, different functional areas in the front cross beam of the front seat are marked, and the functional areas can clearly reflect the connection points between the cross beam and the seat, chassis, support structure, etc. The parts of the cross beam related to the seat installation can be identified and marked; the mounting holes or brackets related to the seat belt can be marked on the first structural drawing; the parts where the cross beam is connected to the vehicle chassis can be identified; if the design of the cross beam includes an energy absorption area, the energy absorption area can be marked.

[0066] Based on the first structural image, an image with functional area markings (fourth image) can be generated. Each functional area can be marked with different colors, shapes or symbols to facilitate the identification of different areas at a glance. The seat support points can be marked with a specific color (such as blue) to show their location and range. The seat belt anchorage points can be marked with different colors (such as green) to show the specific structural details of the area. The chassis connection points can mark the connection between the crossbeam and the chassis, which may be marked in red or other colors. The energy absorption area can be marked with orange or yellow, and its specific energy absorption function can be indicated with annotations or labels.

[0067] S250: Generate a fifth image based on the first structural image and the vehicle performance constraint information, wherein the fifth image is used to reflect the influence of the front cross beam of the front seat on the overall performance of the vehicle.

[0068] For example, the content in the vehicle performance constraint information can be analyzed. It can be: weight requirements, the weight of the front crossbeam of the front seats has a direct impact on the fuel economy, acceleration performance and cruising range of the vehicle, especially in electric vehicles and energy-saving vehicles, weight reduction is a key goal; stiffness and strength requirements, the stiffness and strength of the front crossbeam of the front seats can affect the overall body rigidity of the vehicle, affecting the handling, comfort and body anti-twisting ability; vibration and noise performance, the crossbeam can suppress vibration and reduce noise during vehicle driving, which is beneficial to riding comfort and quietness in the car; power transmission and structural safety, in the event of a collision or vehicle bumps, the front crossbeam of the front seats can effectively transmit impact force, which is beneficial to the safety of the vehicle.

[0069] The first structural image can be subjected to weight analysis. The total weight of the crossbeam can be calculated based on the material density and the geometry of the crossbeam, which can affect the vehicle's curb weight, thereby affecting the vehicle's fuel consumption, acceleration performance and cruising range. The weight distribution of the crossbeam in the vehicle can be analyzed, especially the weight concentration in the front of the vehicle body or the seat support area, which has an important impact on the center of gravity position and balance of the vehicle. An image showing the weight distribution of the crossbeam in the vehicle is generated to reflect the impact of the crossbeam on the overall center of gravity and front-to-rear balance of the vehicle.

[0070] The first structural image can be analyzed for stiffness and strength. Finite element analysis (FEA) can be used to simulate the stress and deformation of the beam under different loads (such as seat load, driving bumps, and collision impacts) to evaluate the contribution of the beam's stiffness and strength to the overall rigidity of the vehicle. The effect of the beam on the torsional rigidity of the vehicle can be analyzed. The torsional rigidity of the vehicle body is closely related to the handling performance of the vehicle, especially when driving at high speeds or turning. It is possible to evaluate how the beam shares and transmits loads from the vehicle chassis, seats, and external impacts under different conditions, which helps to optimize the structural design of the beam and avoid stress concentration at key stress points. Generate a stiffness analysis diagram, which uses color gradients or vector diagrams to show the stress and deformation of the beam under different stress states, reflecting the contribution of the beam to the rigidity of the vehicle.

[0071] The first structural image can be used to analyze the vibration performance. Modal Analysis can be used to simulate the vibration characteristics of the beam and evaluate the resonant modes of the beam at different frequencies. It can be used to evaluate how the beam transmits vibration and noise from the vehicle when it is running, which can help find out whether the design of the beam causes additional noise in the cockpit. Generate a vibration mode image showing the beam, reflecting the resonance phenomenon of the beam at different frequencies, marking potential noise and vibration sources.

[0072] The first structural image can be used for power transmission and safety analysis. It can simulate how the beam transfers energy to the body structure during a collision, which is beneficial for the effective protection of the occupants in the vehicle during a collision. Combined with the overall structure of the vehicle, the impact force and deformation of the beam during a collision can be analyzed to ensure that the beam will not fail. The failure mode of the beam under extreme conditions (such as collisions or severe bumps) can be analyzed, which is beneficial for the beam to have sufficient redundancy within the design range to avoid premature failure. A collision energy absorption diagram is generated to show the path of the beam to transfer energy and absorb impact force during a collision, showing its role in collision safety.

[0073] The weight distribution image, the stiffness analysis diagram, the vibration mode image, and the collision energy absorption diagram may be recorded in a document, and the fifth image may include the weight distribution image, the stiffness analysis diagram, the vibration mode image, and the collision energy absorption diagram.

[0074] S260: Integrate the first structural image, the second image, the third image, the fourth image and the fifth image to obtain an initial structural image of the front cross beam of the front seat.

[0075] For example, image editing tools (such as Adobe Photoshop, GIMP) or professional image management software can be used to integrate all images. All images can be inserted into a single document (such as a PDF file or a Word document), and each image can be equipped with corresponding captions and titles, so that the images can be arranged in a certain logical order for easy viewing and understanding.

[0076] Through these steps, a comprehensive visual reference can be provided for design and analysis, and a basis for subsequent optimization and verification work.

[0077] S300, inputting the initial structure image and each constraint information in the structure constraint information into a key region identification network to perform key region identification to obtain key region information. The key region identification network is used to identify the key region that can be optimized in the initial structure image according to each constraint information.

[0078] It can be understood that the critical area recognition network can automatically analyze the potential optimization areas in the structural image based on deep learning or other machine learning methods, combined with image processing and physical simulation. The critical area recognition network can accept various types of information input, which can include: the initial structural image, including the geometry and shape of the beam; structural constraint information, specific constraints, such as vehicle structure, collision safety, etc. The role of the critical area recognition network is to mark the optimizable areas in the structural image according to different constraint information. The critical area recognition network can effectively process three-dimensional image information, and can analyze each area in the structural image for each constraint information according to the different constraint information input.

[0079] Exemplarily, for each constraint information, the critical area recognition network processes it separately and gradually identifies different critical areas in the structural image. The critical area recognition network can identify the critical parts that need to be retained in the overall structure of the vehicle, such as areas with greater force and areas connected to other body parts. The critical area recognition network can analyze the force and deformation behavior of the beam during a collision, and mark the areas that need to be particularly strengthened during a collision, such as collision energy absorption areas and weak areas that may deform too much during a collision. The critical area recognition network can analyze the functional requirements of the beam and mark the parts that can be optimized or improved, such as seat support points, mounting holes, and other areas that must be retained or strengthened. The critical area recognition network can identify areas that affect the overall performance of the vehicle (such as weight, rigidity), which can be parts that need to be optimized in terms of reducing weight or improving stiffness, such as non-critical force areas, which can reduce weight through material substitution or shape optimization.

[0080] After processing each constraint information, the key region identification network outputs a set of key region information, which may include the regions in the initial structure image that can be optimized or the regions that remain unchanged, and marks the corresponding regions according to each constraint type. The key region identification network can mark which regions in the structure image belong to the key regions, such as in the form of color coding to distinguish the parts that need to be strengthened, optimized or maintained.

[0081] Through this step, the key area information identified by the key area identification network will be used as input for the subsequent optimization stage to help the optimization algorithm better adjust to these areas.

[0082] In a possible implementation, S300, inputting the initial structure image and each constraint information in the structure constraint information into a key region recognition network to perform key region recognition to obtain key region information includes:

[0083] S310, according to each constraint information in the structural constraint information, using a key area recognition network to identify a key area that can be optimized in the image corresponding to each constraint information, and obtain key area information corresponding to each constraint information, wherein the key area information corresponding to each constraint information includes position information of the key area corresponding to each constraint information.

[0084] Exemplarily, a key area recognition network of machine learning or deep learning can be used to identify the key areas that can be optimized in the image corresponding to each constraint for each constraint information. The key area recognition network can identify which parts of the material or geometry in the second image can be optimized based on the vehicle structure constraint information; the key area recognition network can identify the energy absorption area that may be optimized and the weak points where the collision stress is concentrated in the third image based on the collision safety constraint information; the key area recognition network can identify the areas in the fourth image that do not involve functional loss and can be optimized based on the structural function constraint information. The key area recognition network can identify the areas in the fifth image that can be lightweighted or have improved stiffness based on the vehicle performance constraint information. After the recognition is completed, the key area recognition network can output the key areas under each constraint condition, which may include location information, such as the coordinates or marking positions of the key areas in the 3D model.

[0085] Optionally, S310, according to each constraint information in the structural constraint information, using a key area recognition network to identify a key area that can be optimized in the image corresponding to each constraint information, and obtaining key area information corresponding to each constraint information, including:

[0086] S311, using a key area recognition network to extract features from an image corresponding to each constraint information, to obtain a feature image corresponding to each constraint information.

[0087] Exemplarily, each image (the second image to the fifth image) is processed by a key area recognition network to extract the key features of each image. Convolutional neural networks (CNNs) or other deep learning methods can be used to extract features in each image to capture spatial information, stress distribution, functional layout, and performance impact in different images. Features in the second image may include geometric shapes, connection points, installation locations, etc.; features in the third image may include stress concentration areas, energy absorption areas, deformation modes, etc.; features in the fourth image may be identification of functional areas, such as seat support points, seat belt fixing points, etc.; features in the fifth image may involve the impact of beams on weight, stiffness, vibration and other performance. After feature extraction, each image generates a corresponding feature image, which retains the key information in the original image to help with subsequent key area identification.

[0088] Exemplarily, in S311, using a key area recognition network to extract features from an image corresponding to each constraint information to obtain a feature image corresponding to each constraint information, including:

[0089] S3111, using the convolutional neural network of the key area recognition network to extract features of the image corresponding to each constraint information, to obtain a first feature image corresponding to each constraint information.

[0090] It can be understood that the structure of a convolutional neural network (CNN) includes: an input layer that accepts input images and converts them into tensor form; a convolution layer that uses multiple convolution kernels (filters) to perform convolution operations on the input image to extract local features, such as edges, corners, etc.; an activation function that can use ReLU (Rectified Linear Unit) as an activation function to increase the nonlinear expression ability of the network; a pooling layer that uses max pooling or average pooling to reduce the dimension of the image while retaining important features; a fully connected layer that integrates the extracted local features and ultimately outputs recognition results or feature extraction. The network contains multiple layers of convolution and pooling operations, and each layer extracts features at different levels. The early layers extract low-level features (such as edges and textures), and the subsequent layers extract more advanced features (such as local structures, shapes, and complex patterns).

[0091] Exemplarily, the image corresponding to each constraint information (the second image to the fifth image) can be used as input and input into the input layer of the convolutional neural network. The convolution layer uses multiple convolution kernels of different sizes to scan each local area of ​​the image and extract low-level features (such as edges, textures, shapes, etc.). The convolution operation generates feature maps, which reflect the distribution of different features in the input image. After each convolution, an activation function (such as ReLU) is applied to enhance the nonlinear expression ability of the network and help the network recognize more complex image patterns. After each convolution, the pooling layer downsamples the feature map to retain the most important features in the local area and reduce the spatial size of the feature image, reducing the amount of calculation while retaining key information, and obtaining a reduced-dimensional feature map. The reduced-dimensional feature map retains the key information of the input image, but is smaller in size, which is helpful for subsequent processing. After multiple layers of convolution and pooling, the fully connected layer outputs the first feature image corresponding to each constraint information, and each feature image will be used for subsequent multi-scale feature extraction and optimization.

[0092] S3112, using the multi-scale feature extraction module of the key area recognition network to perform multi-scale feature extraction on the first feature image corresponding to each constraint information, to obtain a multi-scale feature image corresponding to each constraint information.

[0093] Exemplarily, in order to enhance feature representation, a multi-scale feature extraction module may be used to perform multi-scale feature processing on the first feature image corresponding to each constraint information. The purpose of multi-scale feature extraction is to capture the global and local information of the image at different scales, so as to better understand the different hierarchical structures of the image.

[0094] The first feature image can be processed using convolution kernels of different sizes (such as 3x3, 5x5, 7x7, etc.) to capture details, global structures, and local patterns in the image. By combining large-scale and small-scale features, more diversified information can be extracted, including macro and micro patterns of the image. Features can include large geometric shapes, large-scale changes in stress distribution, small-scale local textures, etc. The multi-scale feature image processed by the multi-scale feature extraction module is output. The multi-scale feature image corresponding to each constraint information includes feature representations from different scales, which can reflect the complex structure in the image.

[0095] S3113, using the feature fusion module of the key area recognition network to perform image fusion on the multi-scale feature images corresponding to each constraint information, generating a fused feature image corresponding to each constraint information, and determining the fused feature image corresponding to each constraint information as the feature image corresponding to each constraint information.

[0096] For example, in order to integrate the scale features extracted from the multi-scale feature extraction module, a feature fusion module may be used to perform image fusion. The function of the feature fusion module is to combine different levels of information of the multi-scale feature image to generate a unified feature image.

[0097] Multi-scale features can be fused through weighting mechanisms or other fusion strategies (such as splicing, pixel-by-pixel addition, etc.). The fusion process can fully utilize the advantages of each scale to generate a feature image with comprehensive information. If feature images of different scales are stored in multiple channels, the feature fusion module can integrate multi-scale features into a unified representation through channel fusion operations. In the fusion process, global features and local details are organically combined, so that the fused image has both macroscopic structure and subtle features. The fused feature image corresponding to each constraint information is output. The fused feature image contains information at multiple scales and can more comprehensively represent the features of the input image.

[0098] Through the above steps, information at different levels can be captured and fused into the final feature image, which provides a basis for further optimization and key area identification.

[0099] S312, according to each constraint information in the structural constraint information, using a key area recognition network to perform key area recognition on the feature image corresponding to each constraint information, to obtain key area information corresponding to each constraint information.

[0100] For example, a key area recognition network can be used to identify areas in the crossbeam and the overall structure of the vehicle in the feature image of the vehicle structural constraint information that may need to be optimized, such as structural connection points, stress concentration areas, etc. A key area recognition network can be used to identify weak points of stress concentration, energy absorption areas, or areas that need to be strengthened during a collision in the feature image of the collision safety constraint information. A key area recognition network can be used to identify functional areas in the feature image of the structural function constraint information, such as important functional areas such as seat support points and seat belt fixing points, to determine whether there is room for optimization. A key area recognition network can be used to identify key areas in the feature image of the vehicle performance constraint information that may affect the overall performance of the vehicle (such as weight, stiffness, vibration, etc.).

[0101] Exemplarily, S312, according to each constraint information in the structural constraint information, using a key area recognition network to perform key area recognition on a feature image corresponding to each constraint information, to obtain key area information corresponding to each constraint information, including:

[0102] S3121, according to each constraint information in the structural constraint information and the feature image corresponding to each constraint information, a candidate region generation module of the key region recognition network is used to generate a candidate region corresponding to each constraint information, and obtain candidate region information corresponding to each constraint information. The candidate region information corresponding to each constraint information includes the candidate region corresponding to each constraint information and the feature of the candidate region corresponding to each constraint information.

[0103] It can be understood that the function of the candidate region generation module is to generate a series of candidate regions according to the feature image corresponding to each constraint information. The candidate regions may include parts that can be optimized, or may be more important structural regions.

[0104] Exemplarily, the structural constraint information is used to guide the key area recognition network to generate candidate areas corresponding to each constraint information. The key area recognition network uses a sliding window on the feature image to gradually scan the image and identify potential candidate areas, each of which may include local features such as geometry, material distribution, stress concentration, etc. The generated candidate area is composed of multiple feature windows, reflecting the part of the image that may need further processing. The candidate area information may include the location information of the candidate area corresponding to each constraint information (specific coordinates in the feature image) and the feature information of the candidate area corresponding to each constraint information (such as local stress, material distribution, etc.).

[0105] S3122, based on the classification module of the key area recognition network, the features of the candidate area corresponding to each constraint information are classified, and whether the candidate area corresponding to each constraint information is an optimizable key area is identified to obtain an identification result, wherein the identification result includes the optimizable key area corresponding to each constraint information.

[0106] Exemplarily, the feature information of each candidate region is input as input to the classification module of the key region recognition network, and the classification module is responsible for determining whether the candidate region belongs to the key region that can be optimized.

[0107] The classification module can classify the features of the candidate area through a deep learning model (such as a fully connected neural network). The geometric shape, local stress concentration, functional interface and other information in the feature information are used to determine whether the area is a key area that can be optimized. Each candidate area is classified as a "key area that can be optimized" or a "non-key area" according to the characteristics. The classification results of each constraint information output can include key areas that can be optimized, which are determined to be areas that can be further optimized after classification; non-key areas and areas that do not need to be optimized will be excluded.

[0108] S3123, based on the regression module of the key area recognition network, predict the position of the key area that can be optimized corresponding to each constraint information, and obtain the position information of the key area corresponding to each constraint information.

[0109] Exemplarily, for the optimizable key areas that have been identified by the classification module, they are input into the regression module of the key area identification network for accurate position prediction.

[0110] The regression module can accurately predict the specific position of each key area that can be optimized in the entire structural image based on the feature image corresponding to each constraint information through the regression algorithm. The regression module outputs the specific coordinates and size of the key area that can be optimized in the image, such as the coordinates of the upper left corner and the lower right corner, the width and height of the area, etc. Each key area that can be optimized will obtain the corresponding position information, indicating the specific position of the key area on the front beam of the front seat.

[0111] S3124, integrating the optimizable key area corresponding to each constraint information and the position information of the key area corresponding to each constraint information, to obtain the key area information corresponding to each constraint information.

[0112] For example, the optimizable key area corresponding to each constraint information (such as the identifier of the optimizable area under each constraint information) and the location information of the optimizable key area corresponding to each constraint information (such as the specific coordinates and spatial range of the key area) can be integrated to generate complete key area information. The final key area information includes the key area corresponding to each constraint information and its precise location information. The final key area information will be used for subsequent design optimization and structural improvement.

[0113] S320 , marking the key area corresponding to each constraint information in the first structure image according to the key area information corresponding to each constraint information, to obtain a marked image.

[0114] Exemplarily, the key area information corresponding to each constraint information can be applied to the first structural image to mark the key areas in the first structural image. Different colors or symbols can be used to identify the key areas corresponding to different constraint information to clearly display different key areas. For example, red can represent key areas related to collision safety; blue can represent key areas related to vehicle performance; green can represent key areas related to structural functions; and yellow can represent optimized areas related to structural materials. A marked structural image (marked image) is generated, and the marked image clearly marks the key areas under different constraint conditions on the structural model of the front crossbeam of the front seat.

[0115] S330, integrating the marked image and the key area information corresponding to each constraint information to obtain the key area information.

[0116] Exemplarily, the specific coordinates or position of each region may be marked on the marked image, and an image containing all optimizable regions under all constraints may be obtained.

[0117] The above steps not only help identify potential optimization points in beam design, but also provide direction and basis for subsequent optimization and adjustment.

[0118] S400, based on the key region information and the structural constraint information, using an optimization algorithm to determine the key regions in the key region information that meet the structural constraint information, and obtain the target region information.

[0119] It can be understood that optimization algorithms may include: Topology optimization, which is used to find the best solution for material distribution in key areas. Optimize material distribution and remove unnecessary parts while maintaining structural strength and functional requirements to achieve weight reduction or performance improvement. Multi-objective optimization based on genetic algorithms, by simulating the natural selection process, iteratively generates multiple optimization schemes, and selects the best design that meets multiple goals (such as weight, rigidity, and collision safety) at the same time. Gradient descent method can be used for single-objective or multi-objective optimization to find the optimal solution by continuously adjusting design parameters. Finite element analysis (FEA) combined with optimization, simulates the force conditions of each key area through finite element analysis, and gradually adjusts the design in combination with the optimization algorithm to find the best material distribution or shape.

[0120] For example, an appropriate optimization algorithm can be selected to optimize the key areas, and each constraint in the structural constraint information can be set as the boundary of the optimization algorithm, which is conducive to the optimization algorithm not violating the constraint during the optimization process. The optimization algorithm can gradually adjust and calculate each key area through repeated iterations, and each iteration will be refined and adjusted according to the results of the previous one until the goal is met or the optimal solution is reached. The optimization algorithm generates multiple optimization schemes and screens out designs that meet multiple goals at the same time (such as reducing weight and increasing stiffness). After the optimization is completed, the optimization algorithm outputs the final target area information to determine which areas can be adjusted.

[0121] In a possible implementation, S400, based on the key region information and the structural constraint information, an optimization algorithm is used to determine the key regions satisfying the structural constraint information in all the key regions in the key region information, and obtain the target region information, including:

[0122] S410, constructing an optimization function according to the structural constraint information.

[0123] Exemplarily, weighted multi-objective optimization and Pareto optimal multi-objective optimization can be used. Pareto optimal multi-objective optimization is directly processed by the multi-objective optimization algorithm, and each objective function is optimized independently. Weighted multi-objective optimization is to generate a single optimization function by weighted combination of multiple objective functions according to different priorities.

[0124] Using Pareto optimal multi-objective optimization, the optimization objectives can be determined based on the structural constraint information. The optimization objectives may include: reducing weight, minimizing the weight of the front crossbeam of the front seats; increasing stiffness and strength, which is conducive to the stiffness and strength of key areas meeting the design requirements; improving energy absorption capacity, for collision safety, which is conducive to the key areas being able to absorb enough collision energy to protect the passengers in the car; functional requirements, ensuring the strength and durability of functional areas, such as seat support points and seat belt fixing points. One optimization objective corresponds to an objective function, for example: , , , .

[0125] Using weighted multi-objective optimization, each objective function has a corresponding weight , combining multiple objectives by weighting to form a single optimization objective: .

[0126] Optionally, S410, constructing an optimization function according to the structural constraint information includes:

[0127] S411, obtaining a weight value corresponding to each constraint information in the structural constraint information.

[0128] For example, the weight value corresponding to each constraint information is manually defined and input into the system through the human-computer interaction window. For example, the weight value of the vehicle structure constraint information is , the weight value of collision safety constraint information is , the weight value of the structural function constraint information is , the weight value of vehicle performance constraint information is The weight value reflects the relative importance of the constraint condition. For example, collision safety may be more important. Larger, weight loss may be a secondary goal, then Smaller.

[0129] S412: construct a first objective function according to the vehicle structure constraint information.

[0130] For example, the vehicle structure constraint information involves the weight, size, material, etc. of the crossbeam, so the optimization goal corresponding to the vehicle structure constraint information may be to reduce the weight of the crossbeam to improve fuel economy or electric vehicle endurance, that is, The first objective function can calculate the total weight based on the material density and the beam volume. The first objective function is: ,in, is the density of the material, is the volume of the beam.

[0131] S413: construct a second objective function according to the collision safety constraint information.

[0132] For example, the collision safety constraint information involves the energy absorption, stress distribution and deformation control of the cross beam during a collision. Therefore, the optimization target corresponding to the collision safety constraint information may be that the cross beam can absorb the maximum energy during a collision to protect the occupants in the vehicle, that is, The second objective function can be expressed as the energy absorbed by the beam during the collision, based on the integral calculation of stress and strain: ,in, is the stress in each part of the beam, It's strain.

[0133] S414: Construct a third objective function according to the structural function constraint information.

[0134] For example, the structural function constraint information involves the strength and load-bearing capacity of functional areas such as seat support points and seat belt anchor points. Therefore, the optimization goal corresponding to the structural function constraint information can be that the strength of the functional area (such as the seat support point) is sufficient to prevent failure in daily use, that is, The third objective function can control the stress and displacement limits of key functional areas, which is beneficial for the beam not to exceed the design limit. The third objective function is: ,in, It is the location of specific functional areas, with the goal of maximizing the carrying capacity of these areas.

[0135] S415: Construct a fourth objective function according to the vehicle performance constraint information.

[0136] For example, the vehicle performance constraint information may include the control of the overall body stiffness, vibration, comfort and noise. Therefore, the optimization target corresponding to the vehicle performance constraint information may be to optimize the stiffness of the crossbeam to reduce body deformation while ensuring weight reduction, that is, The fourth function can be optimized based on the deformation of the beam after the load is applied, expressed as the minimization of displacement, the fourth objective function: ,in, is the displacement of the beam under load, is the applied force.

[0137] S416, integrating the first objective function, the second objective function, the third objective function and the fourth objective function according to the weight value corresponding to each constraint information in the structural constraint information to obtain an optimization function.

[0138] Exemplarily, the final weighted optimization function may be obtained by using the weight value corresponding to each constraint information in the structural constraint information and the first objective function, the second objective function, the third objective function, and the fourth objective function: .

[0139] S420, solving the optimization function using an optimization algorithm according to the key region information to obtain a set of equilibrium solutions, wherein the equilibrium solution is used to characterize the key regions that satisfy the structural constraint information in all the key regions in the key region information.

[0140] For example, in Pareto optimal multi-objective optimization, there is no single global optimal solution, but rather a set of Pareto optimal solutions. Each solution on the Pareto frontier achieves the best results in some objectives, but may compromise on other objectives. That is, when optimizing one objective, it is impossible to further improve the objective without sacrificing other objectives. If the solution Relative to the solution Superior to a certain goal , but at least equal to Equally excellent, A multi-objective optimization algorithm (such as NSGA-II or SPEA2) is used to find a Pareto optimal solution set. The optimization algorithm searches for all objectives simultaneously and generates a set of solutions, each of which has advantages in some objectives but may compromise on other objectives.

[0141] In weight-based multi-objective optimization, optimization algorithms (such as genetic algorithm (GA) and particle swarm algorithm (PSO)) are used to iteratively solve the problem. In each iteration, the solution set is continuously adjusted to gradually approach the optimal solution and generate a set of balanced solutions.

[0142] Optionally, S420, solving the optimization function using an optimization algorithm according to the key area information to obtain a balanced solution set, including:

[0143] S421, determining the variable range and constraint conditions of each key region according to the key region information, wherein the variable range is used to characterize the variable range of the target variable in the target function.

[0144] It can be understood that the key area information may include geometric information of the key area, geometric characteristics of the key area, such as size (thickness, length, width), shape and position; material properties of the key area, such as density, elastic modulus, yield strength, etc. The target variables of the key area can be geometric variables and material variables.

[0145] For example, the minimum and maximum possible geometric dimensions can be calculated by geometric analysis. A size range is assigned to each critical area based on design specifications, empirical rules, or historical data. For example, the thickness of a critical area may be set to You can extract material strength, stiffness, density and other data from the material database, and assign material variable ranges to each key area based on the data in the material database. For example, if you need to optimize the balance between strength and weight, select the appropriate material from [Aluminum, Steel, Composite] and set the corresponding properties (such as density, strength, etc.) for each material.

[0146] The maximum allowable stress (strength constraint) for each critical area can be generated based on the yield strength of the material. For example, if aluminum alloy is used in the critical area, the stress upper limit is set to For critical load-bearing areas, such as energy absorption areas during collision, the maximum deformation is calculated based on the requirements during collision or the simulation analysis results. For example, the displacement constraint of a certain area is .

[0147] S422, generating an optimization model according to the optimization function, the variable range of each key area and the constraint conditions.

[0148] Exemplarily, the final constrained optimization model is generated by combining the variable range and constraint conditions of each optimization function and each key area. Optimization model: ,in, are inequality constraints (such as maximum stress, maximum deformation), is an equality constraint (such as a geometric or material requirement).

[0149] S423, using an optimization algorithm to iteratively calculate the solution of the optimization model to obtain a set of equilibrium solutions.

[0150] For example, an optimization algorithm can be selected according to the complexity of the problem and the nature of the optimization target, such as genetic algorithm (GA), particle swarm optimization (PSO), simulated annealing algorithm (SA), etc. In the genetic algorithm, an initial population is generated, each individual represents a solution, and the initial population can be randomly generated according to the variable range. In the particle swarm optimization algorithm or simulated annealing algorithm, the solution set is initialized to generate multiple initial solutions. For each individual (solution), the optimization model is calculated. , represents the performance of the solution in multi-objective optimization. When calculating, the effectiveness of the solution can be evaluated according to the optimization function and constraints. Evaluate the quality of each solution and calculate the performance of each objective (such as weight, collision energy absorption, functional area strength and stiffness) on the solution. According to the selected optimization algorithm, select a suitable update strategy to gradually improve the solution. For example, the genetic algorithm (GA) selects individuals with better performance according to the fitness function, performs a crossover operation between two parent individuals, generates a new solution, performs a small probability mutation operation on the individual to explore the new search space, and updates the population with the newly generated solution; the speed of each particle in the particle swarm optimization (PSO) is updated according to its personal best position and the global best position, and the position of the particle (that is, the value of the solution) is updated according to the updated speed; the simulated annealing algorithm (SA) reduces the "temperature" through simulated annealing, gradually shrinks the search range of the solution, avoids the local optimal solution, and generates a new solution based on the neighborhood of the current solution. If the new solution is better, it replaces the current solution, otherwise it is replaced according to the temperature acceptance probability. After each iteration, check whether the optimization process has reached the convergence condition, for example, the preset maximum number of iterations has been reached; the change in the objective function value is lower than a certain threshold (such as 0.001), indicating that the solution is stable. Finally, the detailed data of the solution set is output, and each solution contains its value on each objective function, as well as the optimal value of each variable (such as thickness, material selection, stress distribution, etc.).

[0151] S430, determining the best balance solution in the balance solution set, and determining the key area in the best balance solution as the target area, and obtaining target area information.

[0152] For example, the selection of the best solution from the Pareto frontier solution set can be based on the designer's preference or actual needs. For example, if the design goal is to reduce weight, a solution with a lighter weight is selected from the Pareto solution set. If multiple goals have similar priorities, the Pareto solution set can be further analyzed through manual or automated decision-making tools.

[0153] In weighted multi-objective optimization, different weights can be assigned to each objective in a weighted manner, and the weighted total score can be calculated to select the best solution. The corresponding solution is selected based on the highest priority objective (such as collision safety or weight reduction). If the weights of the various objectives are similar, the solution with the best overall performance is selected.

[0154] Each solution includes information of multiple key areas, and the key area in the optimal solution is the target area.

[0155] Optionally, in step S430, determining the best equilibrium solution in the equilibrium solution set includes:

[0156] S431, according to the weight value corresponding to each constraint information in the structural constraint information, calculate the score value of each equilibrium solution in the equilibrium solution set to obtain a score value set, wherein the score value set includes each equilibrium solution and the score value of each equilibrium solution.

[0157] For example, each equilibrium solution can reflect the overall performance of the equilibrium solution on different objectives by calculating a weighted score value. The score calculation formula is: ,in, It is a balanced solution For each equilibrium solution , for each objective function value Perform weighted processing and use the score value formula to calculate the weighted score of each horizontal solution In order to ensure that each objective function value is within the same range, the objective function value can be standardized and then weighted summed based on the standardized value. For example, the calculation formula can be used: , each objective function value is scaled to the range of [0,1] according to the maximum and minimum values.

[0158] The representation of a set of fractional values ​​can be , where each equilibrium solution And its corresponding score An entry is formed. The calculated score value set includes each equilibrium solution and the score value corresponding to the equilibrium solution, which serves as the basis for further screening the best solution.

[0159] S432, screening out the equilibrium solution with the highest score value from the set of score values, and determining the equilibrium solution with the highest score value as the best equilibrium solution in the set of equilibrium solutions.

[0160] For example, the score value set is traversed to filter out the score value The highest equilibrium solution ,Right now, ,in, is the solution with the highest weighted score, representing the solution with the best overall performance on all objectives. as the final optimal equilibrium solution.

[0161] The above steps are helpful to obtain the best optimizable key areas, laying the foundation for subsequent beam structure optimization.

[0162] S500: Optimize the initial structural image according to the target area information to obtain a final structural image of the front cross beam of the front seat.

[0163] For example, the initial structural image can be geometrically adjusted. According to the target area information, the distribution of materials can be optimized. For example, unnecessary materials can be removed, or the material thickness can be increased in key stress areas. The shape of the beam can be adjusted, such as reducing weight or increasing strength. Possible operations include changing the cross-sectional shape of the beam (for example, from a solid section to an I-shaped section or a hollow section) to reduce weight but maintain rigidity. Structural support or strengthening of the material strength can be added in areas that need to withstand high loads or collision forces (such as collision energy absorption areas).

[0164] During the optimization process, the materials of certain areas may be replaced according to the target area information, such as replacing some steel with aluminum alloy or composite materials to achieve lightweight or collision energy absorption. The new material is assigned so that the updated model has the correct physical properties, including density, elastic modulus, and yield strength, which is conducive to correct performance in subsequent simulation and performance analysis.

[0165] The initial structural image can be optimized for structural function. When optimizing the structural function, it is necessary to check that the functionality of the front crossbeam of the front seat will not be affected by the structural changes. For example, the functional areas such as the seat support points and the seat belt fixing points can maintain the original geometry and strength requirements to prevent functional failure due to optimization. In some special cases, some functional areas may need additional reinforcement or redesign after optimization to ensure functional requirements.

[0166] After the optimization is completed, the final structural image of the front crossbeam of the front seat is generated. The final structural image can include: 3D model file. The optimized crossbeam structure can be output in standard 3D file format (such as STEP, IGES, STL, etc.) for further use and manufacturing; 2D engineering drawings. Generate corresponding engineering drawings with detailed annotation of dimensions, tolerances, materials and functional areas to facilitate manufacturing and assembly.

[0167] In a possible implementation, S500 optimizes the initial structural image according to the target area information to obtain a final structural image of the front cross beam of the front seat, including:

[0168] S510, optimizing the material in the first structural image according to the target region information to obtain material optimization information.

[0169] Exemplarily, appropriate materials can be selected or the properties of existing materials can be adjusted according to the optimization requirements of each target area (such as reducing weight and improving strength). The most suitable material can be selected from the material library. For example, aluminum alloy, steel or composite materials can be selected from the material library according to the optimization requirements of the target area. According to the design requirements, the elastic modulus, density, yield strength and other properties of the material are adjusted. If the existing material in the target area does not meet the optimization requirements, it can be replaced with a new material. The material optimization information can include the target area, which area the material optimization is applicable to; the material type, the newly selected material and its properties (such as elastic modulus, density, strength, etc.).

[0170] S520: Optimize the connection points of the first structure image according to the target region information to obtain connection point optimization information.

[0171] Exemplarily, connection point optimization is conducive to the various connection points of the front cross beam of the front seat (such as the connection points with the seat and the body) to bear the design load. The load capacity and stress concentration area of ​​each connection point can be determined based on the simulation analysis results. For connection points that can be enhanced, the geometry is optimized (such as adding support and adjusting thickness) to enhance the load-bearing capacity of the target area. The material selection of the connection point can also be optimized according to the load requirements to ensure the strength and durability of the connection point. The connection point optimization information can include the target area, the optimized connection point position; optimization measures, specific optimization operations, such as material adjustment, geometry optimization, adding support, etc.

[0172] S530: Optimize the structure of the first structural image according to the target region information to obtain structural optimization information.

[0173] Exemplarily, structural optimization is to optimize the overall structure of the beam to improve stiffness, reduce weight, or enhance collision safety. The geometric structure of the beam can be adjusted according to the simulation analysis results and the optimization algorithm results. For example, through topology optimization, unnecessary materials are removed and key parts of structural strength are retained. The structural stress distribution in the simulation results can be analyzed to check the rationality of the load path. If the stress concentration in certain areas is too high, the structural design of the area can be optimized. According to the results of the optimization algorithm, the structural shape can be optimized to make the beam more in line with the balance requirements of strength and weight. Structural optimization information can include optimization areas, structural areas that need to be optimized; optimization measures, such as geometry adjustment, load path optimization, material removal or increase.

[0174] S540: Based on the material optimization information, the connection point optimization information and the structure optimization information, the first structure image is adjusted to generate a final structure image of the front cross beam of the front seat.

[0175] For example, according to the material optimization information, the old material in the target area is replaced with the optimized new material. In the geometric structure of the target area, the material distribution is adjusted according to the material optimization information to provide better mechanical properties in the local area. For example, if a certain area needs to reduce weight, the system replaces the original steel with a lightweight aluminum alloy or composite material while maintaining the strength requirements of the area.

[0176] Based on the connection point optimization information, the geometry of the connection point can be adjusted, such as thickening or adding support structures to enhance the strength and stiffness of the connection point. Adjusting the location or number of connection points will help make the load path more reasonable and the force distribution more uniform in the new design. For example, at the connection point between the seat and the crossbeam, the system may add a reinforcement plate or adjust the support structure to enhance the strength of this location.

[0177] The thickness can be adjusted for specific areas based on the structural optimization information. For example, the system can increase the thickness in areas with greater stress to reduce deformation, while reducing materials and weight in non-critical areas. Based on the collision simulation results, special energy absorption areas can be designed to absorb more energy and improve safety through a reasonable combination of geometric deformation and materials. For example, the system generates energy absorption grooves or other energy absorption structures on the beam to optimize the collision performance of the beam.

[0178] Through the above steps, the design efficiency of the beam structure is improved, which is conducive to the final structure meeting the safety, functionality and performance requirements.

[0179] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0180] Corresponding to the method for determining the front cross beam structure of the front row seat described in the above embodiment, the embodiment of the present application also provides a device for determining the front cross beam structure of the front row seat, and each unit of the device can implement each step of the method for determining the front cross beam structure of the front row seat. Figure 4 A structural block diagram of a device for determining a front cross beam structure of a front row seat provided in an embodiment of the present application is shown. For ease of explanation, only portions related to the embodiment of the present application are shown.

[0181] Reference Figure 4 , the device comprises:

[0182] The acquisition unit is used to acquire the structural constraint information of the front cross beam of the front seat, wherein the structural constraint information includes vehicle structural constraint information, collision safety constraint information, structural function constraint information, and vehicle performance constraint information.

[0183] The initial image generating unit is used to obtain the initial structural image of the front cross beam of the front seat according to the structural constraint information.

[0184] The identification unit is used to input the initial structure image and each constraint information in the structure constraint information into the key area identification network for key area identification to obtain key area information. The key area identification network is used to identify the key area that can be optimized in the initial structure image according to each constraint information.

[0185] The target region obtaining unit is used to determine the key regions satisfying the structural constraint information in all the key regions in the key region information by using an optimization algorithm based on the key region information and the structural constraint information, and obtain the target region information.

[0186] The optimization unit is used to optimize the initial structural image according to the target area information to obtain the final structural image of the front cross beam of the front seat.

[0187] It should be noted that the information interaction, execution process and other contents between the above-mentioned units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0188] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0189] The embodiment of the present application also provides a device for determining the structure of the front crossbeam of a front row seat, Figure 5 This is a schematic diagram of the structure of a device for determining the front cross beam structure of a front row seat provided in an embodiment of the present application. Figure 5 As shown, the front cross beam structure determination device 6 of the front seat of this embodiment includes: at least one processor 60 ( Figure 5 Only one is shown), at least one memory 61 ( Figure 5Only one is shown in the figure) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, the front row seat front cross beam structure determination device 6 implements the steps in any of the above-mentioned front row seat front cross beam structure determination method embodiments, or the front row seat front cross beam structure determination device 6 implements the functions of the units in the above-mentioned device embodiments.

[0190] Exemplarily, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 62 in the front cross beam structure determination device 6 for the front row of seats.

[0191] The front crossbeam structure determination device 6 of the front row seat can be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The front crossbeam structure determination device of the front row seat can include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art can understand that Figure 5 It is only an example of the device 6 for determining the structure of the front cross beam of the front row of seats and does not constitute a limitation on the device 6 for determining the structure of the front cross beam of the front row of seats. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components, for example, it may also include input and output devices, network access devices, buses, etc.

[0192] The processor 60 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0193] In some embodiments, the memory 61 may be an internal storage unit of the front-row seat front beam structure determination device 6, such as a hard disk or memory of the front-row seat front beam structure determination device 6. In other embodiments, the memory 61 may also be an external storage device of the front-row seat front beam structure determination device 6, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the front-row seat front beam structure determination device 6. Further, the memory 61 may also include both the internal storage unit and the external storage device of the front-row seat front beam structure determination device 6. The memory 61 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program, etc. The memory 61 may also be used to temporarily store data that has been output or is to be output.

[0194] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0195] An embodiment of the present application provides a computer program product. When the computer program product runs on a device for determining the structure of a front cross beam of a front row of seats, the device for determining the structure of a front cross beam of a front row of seats implements the steps in any of the above-mentioned method embodiments.

[0196] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the front cross beam structure determination device of the front seat, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0197] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0198] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0199] In the embodiments provided in the present application, it should be understood that the disclosed apparatus, device and method for determining the front crossbeam structure of the front row of seats can be implemented in other ways. For example, the above-described apparatus and device embodiments for determining the front crossbeam structure of the front row of seats are merely schematic. For example, the division of the modules or units is merely a logical function division. There may be other division methods in actual implementation, such as multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices or units, which may be electrical, mechanical or other forms.

[0200] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0201] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for determining the front crossbeam structure of a front row seat, characterized in that: include: Acquiring structural constraint information of the front cross beam of the front seat; wherein the structural constraint information includes vehicle structural constraint information, collision safety constraint information, structural function constraint information, and vehicle performance constraint information; Acquiring an initial structural image of the front cross beam of the front seat according to the structural constraint information; Inputting the initial structural image and each constraint information in the structural constraint information into a key area recognition network to perform key area recognition to obtain key area information; wherein the key area recognition network is used to identify the key area that can be optimized in the initial structural image according to each constraint information; Based on the key area information and the structural constraint information, an optimization algorithm is used to determine the key area that satisfies the structural constraint information among all the key areas in the key area information to obtain the target area information; Optimizing the initial structural image according to the target area information to obtain a final structural image of the front cross beam of the front seat; The step of acquiring the initial structural image of the front cross beam of the front seat according to the structural constraint information comprises: generating a first structural image according to the structural constraint information; The first structural image is merged with the vehicle structural image in the vehicle structural constraint information to generate a second image; wherein the second image is used to reflect the position and relative relationship of the front cross beam of the front seat in the overall structure of the vehicle; generating a third image according to the first structural image and the collision safety constraint information; wherein the third image is used to reflect the stress distribution and energy absorption of the front cross beam of the front seat in the collision simulation; According to the structural function constraint information, the functional area in the first structural image is identified to obtain a fourth image; wherein the fourth image is used to reflect the functional area of ​​the front cross beam of the front seat; Generate a fifth image according to the first structural image and the vehicle performance constraint information; wherein the fifth image is used to reflect the influence of the front cross beam of the front seat on the overall performance of the vehicle; Integrate the first structural image, the second image, the third image, the fourth image and the fifth image to obtain an initial structural image of the front cross beam of the front seat; The step of determining the key areas satisfying the structural constraint information in all the key areas in the key area information by using an optimization algorithm based on the key area information and the structural constraint information to obtain the target area information includes: constructing an optimization function according to the structural constraint information; According to the key region information, the optimization function is solved by using the optimization algorithm to obtain a set of balanced solutions; wherein the balanced solutions are used to characterize the key regions in all the key regions in the key region information that satisfy the structural constraint information; Determine the best equilibrium solution in the equilibrium solution set, and determine the key area in the best equilibrium solution as the target area, and obtain the target area information; The step of constructing an optimization function according to the structural constraint information comprises: Obtain a weight value corresponding to each constraint information in the structural constraint information; wherein the weight value corresponding to each constraint information in the structural constraint information includes the weight value of the vehicle structural constraint information , the weight value of the collision safety constraint information , the weight value of the structural function constraint information , the weight value of the vehicle performance constraint information ; According to the vehicle structure constraint information, a first objective function is constructed; wherein the first objective function is , is the density of the material, is the volume of the beam; According to the collision safety constraint information, a second objective function is constructed; wherein the second objective function is , is the stress in each part of the beam, It is strain; According to the structural function constraint information, a third objective function is constructed; wherein the third objective function is , It is the location of a specific functional area; According to the vehicle performance constraint information, a fourth objective function is constructed; wherein the fourth objective function is , is the displacement of the beam under load, is the applied force; According to the weight value corresponding to each constraint information in the structural constraint information, the first objective function, the second objective function, the third objective function and the fourth objective function are integrated to obtain the optimization function; wherein the optimization function is .

2. The method for determining the front cross beam structure of a front row seat as claimed in claim 1, characterized in that: The step of inputting the initial structure image and each constraint information in the structure constraint information into a key area recognition network to perform key area recognition to obtain key area information comprises: According to each constraint information in the structural constraint information, the key area recognition network is used to identify the key area that can be optimized in the image corresponding to each constraint information, so as to obtain the key area information corresponding to each constraint information; wherein the key area information corresponding to each constraint information includes the position information of the key area corresponding to each constraint information; Marking the key area corresponding to each constraint information in the first structure image according to the key area information corresponding to each constraint information to obtain a marked image; The key area information corresponding to the marked image and each constraint information is integrated to obtain the key area information.

3. The method for determining the front cross beam structure of a front row seat as claimed in claim 2, characterized in that: The step of identifying, according to each constraint information in the structural constraint information, a key region that can be optimized in the image corresponding to each constraint information by using the key region recognition network to obtain key region information corresponding to each constraint information includes: Using the key area recognition network to extract features from the image corresponding to each constraint information, to obtain a feature image corresponding to each constraint information; According to each constraint information in the structural constraint information, the key area recognition network is used to perform key area recognition on the feature image corresponding to each constraint information to obtain the key area information corresponding to each constraint information.

4. The method for determining the front cross beam structure of a front row seat as claimed in claim 3, characterized in that: The step of extracting features from the image corresponding to each constraint information by using the key area recognition network to obtain a feature image corresponding to each constraint information includes: Using the convolutional neural network of the key area recognition network to extract features from the image corresponding to each constraint information, to obtain a first feature image corresponding to each constraint information; Using the multi-scale feature extraction module of the key area recognition network to perform multi-scale feature extraction on the first feature image corresponding to each constraint information, to obtain a multi-scale feature image corresponding to each constraint information; The feature fusion module of the key area recognition network is used to perform image fusion on the multi-scale feature images corresponding to each constraint information to generate a fused feature image corresponding to each constraint information, and the fused feature image corresponding to each constraint information is determined as the feature image corresponding to each constraint information.

5. The method for determining the front cross beam structure of a front row seat as claimed in claim 3, characterized in that: According to each constraint information in the structural constraint information, using the key area recognition network to perform key area recognition on the feature image corresponding to each constraint information to obtain key area information corresponding to each constraint information, including: According to each constraint information in the structural constraint information and the feature image corresponding to each constraint information, a candidate region corresponding to each constraint information is generated using a candidate region generation module of the key region recognition network to obtain candidate region information corresponding to each constraint information; wherein the candidate region information corresponding to each constraint information includes the candidate region corresponding to each constraint information and the feature of the candidate region corresponding to each constraint information; Based on the classification module of the key area recognition network, the features of the candidate area corresponding to each constraint information are classified, and whether the candidate area corresponding to each constraint information is an optimizable key area is identified to obtain an identification result; wherein the identification result includes the optimizable key area corresponding to each constraint information; Based on the regression module of the key area recognition network, predict the position of the key area that can be optimized corresponding to each constraint information, and obtain the position information of the key area corresponding to each constraint information; The optimizable key area corresponding to each constraint information and the position information of the key area corresponding to each constraint information are integrated to obtain the key area information corresponding to each constraint information.

6. The method for determining the front cross beam structure of a front row seat as claimed in claim 1, characterized in that: The step of solving the optimization function using the optimization algorithm according to the key area information to obtain a balanced solution set includes: Determine the variable range and constraint conditions of each key area according to the key area information; wherein the variable range is used to characterize the variable range of the target variable in the objective function; Generate an optimization model according to the optimization function, the variable range and the constraint conditions of each key area; Iteratively calculating the solution of the optimization model using the optimization algorithm to obtain the equilibrium solution set; The determining of the best equilibrium solution in the equilibrium solution set comprises: According to the weight value corresponding to each constraint information in the structural constraint information, the score value of each equilibrium solution in the equilibrium solution set is calculated to obtain a score value set; wherein the score value set includes each equilibrium solution and the score value of each equilibrium solution; A balance solution with a highest score value is screened out from the score value set, and the balance solution with the highest score value is determined as the best balance solution in the balance solution set.

7. A device for determining the structure of a front cross beam of a front row seat, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.