Body-in-white structure design method, computer equipment and storage medium

Abstract

The invention provides a body-in-white structure design method, computer equipment and a storage medium. The method comprises the steps of determining a target performance index corresponding to a to-be-designed body-in-white structure; based on the target performance index, determining a target part index corresponding to the to-be-designed body-in-white structure by using a preset system-level decision tree model; based on the target part index, a preset part level decision tree model is used for determining the value range of design parameters of parts corresponding to the body-in-white structure, and the value range of the design parameters of the parts is used for designing the body-in-white structure, so that the body-in-white structure reaches the target performance index. According to the invention, the design efficiency of the body-in-white structure can be improved.

Description

Technical Field

[0001] The present application belongs to the field of automobile design, and in particular, relates to a body-in-white structure design method, computer equipment and storage medium. Background Art

[0002] In the field of automobile design, the design of the body-in-white structure is an extremely complex system engineering, and the design result of the body-in-white structure is crucial to the performance of the whole vehicle. In related technologies, the design of the body-in-white structure mainly relies on simulation and experiment to achieve optimized design parameters.

[0003] However, since complex structures such as the body-in-white have a large number of design parameters and there is interaction between different design parameters, the above method requires a large number of simulations and experiments to obtain better design results, resulting in low design efficiency of the body-in-white structure. Summary of the invention

[0004] The embodiments of the present application provide a body-in-white structure design method, a computer device, and a storage medium to solve the problem of low design efficiency of the body-in-white structure.

[0005] In a first aspect, an embodiment of the present application provides a body-in-white structure design method, the method comprising: determining a target performance indicator corresponding to the body-in-white structure to be designed; based on the target performance indicator, using a preset system-level decision tree model, determining a target component indicator corresponding to the body-in-white structure to be designed; based on the target component indicator, using a preset component-level decision tree model, determining a value range of a design parameter of a component corresponding to the body-in-white structure, the value range of the design parameter being used to design the body-in-white structure so that the body-in-white structure achieves the target performance indicator.

[0006] In some embodiments, determining the target performance index corresponding to the body-in-white structure to be designed includes: obtaining a whole vehicle performance index corresponding to the vehicle to be designed; parsing the whole vehicle performance index to obtain the target performance index of the body-in-white structure corresponding to the vehicle.

[0007] In some embodiments, based on the target performance indicator, a preset system-level decision tree model is used to determine the target component indicator corresponding to the body-in-white structure to be designed, including: traversing the system-level decision tree model to obtain a performance indicator that matches the target performance indicator; and determining that the component performance indicator corresponding to the matched performance indicator is the target component indicator corresponding to the body-in-white structure to be designed.

[0008] In some embodiments, based on the target component index, a preset component-level decision tree model is used to determine the value range of the design parameters of the component corresponding to the body-in-white structure, including: traversing the component-level decision tree model to obtain a component performance index that matches the target component index; and determining that the value range of the design parameters of the component corresponding to the matching component performance index is the value range of the design parameters of the component corresponding to the body-in-white structure.

[0009] In some embodiments, the method further includes: constructing an indicator-parameter framework corresponding to the body-in-white structure to be designed based on the design data corresponding to the designed body-in-white structure; determining basic data using the indicator-parameter framework; and constructing the system-level decision tree model and the component-level decision tree model based on the indicator-parameter framework and the basic data.

[0010] In some embodiments, the construction of an indicator-parameter framework corresponding to the body-in-white structure to be designed based on the design data corresponding to the designed body-in-white structure includes: parsing the design data to obtain performance indicators, component performance indicators and design parameters of the body-in-white structure to be designed; using the performance indicators as the first node of the indicator-parameter framework; using the component performance indicators as the second node of the indicator-parameter framework; using the design parameters as the third node of the indicator-parameter framework; and constructing the indicator-parameter framework based on the first node, the second node and the third node.

[0011] In some embodiments, the use of the indicator-parameter framework to determine basic data includes: determining vehicle type information corresponding to the body-in-white structure to be designed; determining initial finite element models of multiple designed body-in-white structures that match the vehicle type information; obtaining a finite element model of the body-in-white structure to be designed based on multiple initial finite element models; generating simulation data of the body-in-white structure to be designed using the indicator-parameter framework and the finite element model; obtaining test data of the body-in-white structure to be designed, and combining the simulation data with the test data to obtain the basic data.

[0012] In some embodiments, constructing the system-level decision tree model and the component-level decision tree model based on the indicator-parameter framework and the basic data includes: using the indicator-parameter framework to parse the basic data to obtain first data and second data, the first data including performance indicators and component performance indicators of multiple designed body-in-white structures, and the second data including component performance indicators and design parameters of multiple designed body-in-white structures; based on the first data, using a first decision tree algorithm to obtain the system-level decision tree model; based on the second data, using a second decision tree algorithm to obtain the component-level decision tree model.

[0013] In a second aspect, an embodiment of the present application provides a body-in-white structure design device, the device comprising: an indicator determination module, used to determine a target performance indicator corresponding to the body-in-white structure to be designed; the indicator determination module is also used to determine a target component indicator corresponding to the body-in-white structure to be designed based on the target performance indicator and using a preset system-level decision tree model; a parameter determination module, used to determine a value range of a design parameter of a component corresponding to the body-in-white structure based on the target component indicator and using a preset component-level decision tree model, the value range of the design parameter being used to design the body-in-white structure so that the body-in-white structure achieves the target performance indicator.

[0014] In a third aspect, an embodiment of the present application provides a computer device, which includes a processor and a memory, and the processor is used to implement any one of the above-mentioned body-in-white structure design methods when executing a computer program stored in the memory.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to implement the above-mentioned body-in-white structure design method.

[0016] The above-mentioned body-in-white structure design method provided in the embodiment of the present application determines the target performance index corresponding to the body-in-white structure to be designed; based on the target performance index, a preset system-level decision tree model is used to determine the target component index corresponding to the body-in-white structure to be designed; based on the target component index, a preset component-level decision tree model is used to determine the value range of the design parameters of the components corresponding to the body-in-white structure, and the value range of the design parameters is used to design the body-in-white structure so that the body-in-white structure reaches the target performance index. The above-mentioned method describes the relationship between the design parameters of the components corresponding to the body-in-white structure and the performance index of the body-in-white through the system-level decision tree model and the component-level decision tree model, and uses the system-level decision tree model and the component-level decision tree model to guide the design of the body-in-white structure, which can reduce the number of simulations and experiments of the body-in-white structure in the optimization design process, reduce development costs, and improve the design efficiency of the body-in-white structure. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a diagram of a computer device for executing the body-in-white structure design method provided in an embodiment of the present application.

[0018] Figure 2 It is a flow chart of a body-in-white structure design method provided in an embodiment of the present application.

[0019] Figure 3 It is a schematic diagram of a component-level decision tree model provided in an embodiment of the present application.

[0020] Figure 4 It is a flow chart of a method for constructing a decision tree model provided in an embodiment of the present application.

[0021] Figure 5 It is a structural diagram of an indicator-parameter framework provided in an embodiment of the present application.

[0022] Figure 6 It is a structural schematic diagram of a finite element model provided in an embodiment of the present application.

[0023] Figure 7 It is a flow chart of simulation data determination and expansion provided in an embodiment of the present application.

[0024] Figure 8 A schematic diagram illustrating a process of training a decision tree model provided in an embodiment of the present application.

[0025] Fig. 9 It is a structural schematic diagram of a body-in-white structure design device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] It should be noted that the terms "first" and "second" in the specification, claims and drawings of the present application are used to distinguish similar objects rather than to describe a specific order or sequence.

[0027] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way. The following embodiments and features in the embodiments may be combined with each other without conflict.

[0028] The body-in-white structure is a rigid spatial structure, usually including main load-bearing components such as longitudinal beams, cross beams and pillars, as well as sheet metal parts connected to them. The body-in-white structure is the installation basis for body parts (such as wheels, seats, steering wheels, etc.). The design result of the body-in-white structure is crucial to the performance of the whole vehicle, including vehicle dynamics, lightweight, noise (Noise), vibration (Vibration) and harshness (Harshness), referred to as NVH, safety, durability, etc., which will be affected by the body-in-white structure.

[0029] In the related art, the design of the body-in-white structure mainly relies on simulation and experiment to optimize the design parameters. For example, the design parameters are studied one by one, and a single design parameter is changed under the premise of limiting other design parameters to analyze the impact of the design parameter on the performance of the body-in-white structure. However, since there are a large number of design parameters in complex structures such as the body-in-white structure, and there are interactive effects between different design parameters, the above method requires a large number of simulations and experiments to obtain a better design result, and the design efficiency of the body-in-white structure is low. For another example, a large number of simulations and experiments can be carried out to determine the better body-in-white structure under the premise of setting the optimization goal. However, the above method requires a large number of simulations and experiments, and the body-in-white structure obtained by this method is unique. The body-in-white structure cannot be used to guide the secondary improvement of the structure, resulting in a large amount of time and cost to carry out similar body-in-white structure optimization design research again during the development of each vehicle, and the design efficiency of the body-in-white structure is low.

[0030] In view of this, the present application provides a body-in-white structure design method, computer equipment and storage medium, which can reduce the number of simulations and tests during the optimization design process of the body-in-white structure, reduce development costs, and improve the design efficiency of the body-in-white structure.

[0031] Some embodiments will be described below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments may be combined with each other.

[0032] Combination Figure 1 A diagram of a computer device for executing the body-in-white structure design method provided in an embodiment of the present application is provided. The body-in-white structure design method can be applied to a computer device 10, which can include a computer, a mobile phone, a laptop, a tablet, a server, and a vehicle-mounted device, etc. The server can be a cloud server or a server cluster. The present application does not limit the type of the computer device 10. Figure 1 As shown, the computer device 10 includes a communication module 101, a memory 102, a processor 103, an input / output (I / O) interface 104, and a bus 105. The processor 103 is coupled to the communication module 101, the memory 102, and the input / output interface 104 through the bus 105.

[0033] In some embodiments, the communication module 101 may include a wired communication module and / or a wireless communication module.

[0034] In some embodiments, the memory 102 is used to store one or more computer programs. The one or more computer programs are configured to be executed by the processor 103. The one or more computer programs include multiple instructions, and when the multiple instructions are executed by the processor 103, the body-in-white structure design method executed on the computer device 10 can be implemented.

[0035] In some embodiments, the processor 103 provides computing and control capabilities. For example, the processor 103 is used to execute a computer program stored in the memory 102 to implement the above-mentioned body-in-white structure design method.

[0036] In some embodiments, the input / output interface 104 is used to provide a channel for user input or output. For example, the input / output interface 104 can be used to connect various input and output devices, such as a mouse, keyboard, touch device, display screen, etc., so that the user can enter information or visualize information.

[0037] In some embodiments, the bus 105 is at least used to provide a channel for mutual communication among the communication module 101 , the memory 102 , the processor 103 , and the input / output interface 104 in the computer device 10 .

[0038] In some embodiments, in a body-in-white structure design scenario, a user (for example, a designer) can input vehicle performance indicators corresponding to the vehicle to be designed into the computer device 10 through the input / output interface 104. The processor 103 obtains target performance indicators corresponding to the body-in-white structure to be designed by analyzing the vehicle performance indicators, and determines the value range of design parameters of the components corresponding to the body-in-white structure to be designed based on the target performance indicators, the system-level decision tree model and the component-level decision tree model. The value range of the design parameters is used to design the body-in-white structure so that the body-in-white structure achieves the target performance indicators.

[0039] In the computer device 10 provided in the embodiment of the present application, the relationship between the design parameters of the components corresponding to the body-in-white structure and the performance indicators of the body-in-white are described by a system-level decision tree model and a component-level decision tree model, and the system-level decision tree model and the component-level decision tree model are used to guide the design of the body-in-white structure, which can reduce the number of simulations and tests of the body-in-white structure during the optimization design process, reduce development costs, and improve the design efficiency of the body-in-white structure.

[0040] It is to be understood that the structure illustrated in the embodiment of the present application does not constitute a specific limitation on the computer device 10. In other embodiments of the present application, the computer device 10 may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0041] Figure 2 is a flow chart of a body-in-white structure design method provided by an embodiment of the present application, the body-in-white structure design method is applied to a computer device (for example, Figure 1 Computer device 10 in FIG. Figure 2 As shown, the body-in-white structure design method includes the following steps. According to different requirements, the order of the steps in the flowchart can be changed, and some can be omitted.

[0042] S11, determining target performance indicators corresponding to the body-in-white structure to be designed.

[0043] In at least one embodiment of the present application, the target performance index may represent a preset performance index of the body-in-white structure. The target performance index may be input by a user, or may be obtained by parsing a preset vehicle performance index.

[0044] The embodiment of the present application is explained by parsing the preset vehicle performance index to obtain the target performance index. In some embodiments, the determination of the target performance index corresponding to the body-in-white structure to be designed may include: obtaining the vehicle performance index corresponding to the vehicle to be designed; parsing the vehicle performance index to obtain the target performance index of the body-in-white structure corresponding to the vehicle. The embodiment of the present application obtains the target performance index of the body-in-white structure by parsing the vehicle performance index of the vehicle to be designed, which can accurately match the vehicle design requirements and improve the accuracy of the body-in-white structure design.

[0045] Among them, the vehicle performance index can be entered into the computer device by the user. The vehicle performance index can be set according to actual needs. For example, the vehicle performance index can include dynamic index, lightweight index, NVH index, safety index and durability index, etc. There is a correlation between the vehicle performance index and the performance index of the body-in-white structure. By parsing the vehicle performance index, the performance index of the body-in-white structure related to the vehicle performance index can be obtained. Exemplarily, taking the vehicle performance index as the NVH index as an example, the performance index of the body-in-white structure can include the first-order bending / torsion mode, static bending / torsion stiffness and the dynamic stiffness characteristics of the suspension system. Among them, the first-order bending / torsion can represent the maximum deformation that the body-in-white structure can withstand when subjected to bending and torsional loads. Static bending / torsion stiffness can represent the ability of the body-in-white structure to resist bending and torsional deformation under static loads. The dynamic stiffness characteristics of the suspension system can represent the natural frequency and stiffness characteristics of the front and rear suspension systems during torsional vibration.

[0046] In some embodiments, the user can analyze the performance index of the whole vehicle based on relevant experience to obtain the target performance index of the body-in-white structure corresponding to the vehicle. In other embodiments, the index analysis model can also be pre-trained, and the index analysis model can be used to analyze the performance index of the whole vehicle to obtain the target performance index of the body-in-white structure corresponding to the vehicle. The input data of the index analysis model is the performance index of the whole vehicle, and the output data is the performance index of the body-in-white structure. The index analysis model can be a convolutional neural network (CNNs) model, a recurrent neural network (RNNs) model, and a generative adversarial network (GANs) model, etc., which are not limited here. The training method of the index analysis model can include a supervised training method and an unsupervised training method. The training process of the model can refer to the relevant technology and will not be repeated here.

[0047] S12, based on the target performance index, using a preset system-level decision tree model, determining target component indexes corresponding to the body-in-white structure to be designed.

[0048] In at least one embodiment of the present application, a system-level decision tree model is used to predict related component performance indicators based on the performance indicators of the body-in-white structure. The system-level decision tree model uses the performance indicators of the body-in-white structure as leaf node content and the component performance indicators as decision node content. Among them, the component performance indicators can represent the performance indicators of the components that have a major impact on the performance indicators of the body-in-white structure. Taking the performance indicators of the body-in-white structure as the first-order bending / torsion mode as an example, the component performance indicators can include the beam natural frequency, beam bending / torsion stiffness, test point stiffness, joint stiffness, joint quality, constraint point stiffness and the number of joints.

[0049] In some embodiments, based on the target performance index, the target component index corresponding to the body-in-white structure to be designed is determined using a preset system-level decision tree model, including: traversing the system-level decision tree model to obtain a performance index that matches the target performance index; determining that the component performance index corresponding to the matched performance index is the target component index corresponding to the body-in-white structure to be designed. The embodiment of the present application describes the relationship between the performance index of the body-in-white structure and the component performance index through a system-level decision tree model, and uses the system-level decision tree model to determine the target component index corresponding to the target performance index of the body-in-white structure, which can reduce the number of simulations and experiments of the body-in-white structure during the optimization design process, reduce development costs, and improve the design efficiency of the body-in-white structure.

[0050] Among them, the performance indicators of the body-in-white structure can be set to excellent, general and poor according to actual needs, and the performance indicators of components can be set to an indicator range according to actual needs, and the indicator range can include a numerical range. Traverse the system-level decision tree model, select the performance indicators that are the same as the target performance indicators as the matching performance indicators, and use the component performance indicators on the same branch as the matching performance indicators as the target component indicators corresponding to the body-in-white structure to be designed. In some embodiments, the number of matching performance indicators may be 1 or more. When the number of matching performance indicators is multiple, a preset number of performance indicators can be selected from multiple matching performance indicators using preset screening rules. Among them, the screening rules can be set according to actual needs. For example, screening can be performed based on the discreteness of the branches. The preset number can be 1 or more, and there is no restriction here.

[0051] Following the above embodiment, when the performance index of the body-in-white structure is a first-order bending / torsion mode, the component performance index may include beam natural frequency, beam bending / torsion stiffness, test point stiffness, joint stiffness, joint mass, constraint point stiffness, and the number of joints. Exemplarily, if the target performance index is used to indicate that the performance of the first-order bending / torsion mode is excellent, by traversing the system-level decision tree model, the target component index is used to indicate that the performance index of the beam natural frequency is in the first numerical range, the performance index of the beam bending / torsion stiffness is in the second numerical range, and the performance index of the test point stiffness is in the third numerical range. Among them, the first numerical range, the second numerical range, and the third numerical range can be set according to actual needs and are not limited here.

[0052] S13, based on the target component index, using a preset component-level decision tree model, determining the value range of the design parameters of the component corresponding to the body-in-white structure, the value range of the design parameters of the component is used to design the body-in-white structure so that the body-in-white structure achieves the target performance index.

[0053] In at least one embodiment of the present application, a component-level decision tree model is used to predict the value range of the design parameters of related components based on the component performance index. The component-level decision tree model uses the component performance index as the leaf node content and the value range of the component design parameter as the decision node content. Among them, the component performance index can represent the performance index of the component that has a major impact on the performance index of the body-in-white structure, and the value range of the component design parameter can represent the value range corresponding to the design parameter that has a major impact on the performance index of the body-in-white structure and has adjustability. Taking the component performance index as the beam natural frequency as an example, the value range of the component design parameter can include the threshold beam thickness range, the cross beam thickness range, the longitudinal beam thickness range, the threshold beam material, the cross beam material and the longitudinal beam material. The embodiment of the present application diversifies the design results of the body-in-white structure by providing the value range of the design parameter, and has a stronger design fault tolerance and inclusiveness.

[0054] In some embodiments, based on the target component index, the preset component-level decision tree model is used to determine the value range of the design parameters of the component corresponding to the body-in-white structure, including: traversing the component-level decision tree model to obtain the component performance index that matches the target component index; determining that the value range of the design parameters of the component corresponding to the matching component performance index is the value range of the design parameters of the component corresponding to the body-in-white structure. The embodiment of the present application describes the relationship between the value range of the component performance index and the design parameter of the component through the component-level decision tree model, and uses the component-level decision tree model to determine the value range of the design parameters of the component corresponding to the target component index, which can reduce the number of simulations and experiments of the body-in-white structure in the optimization design process, reduce development costs, and improve the design efficiency of the body-in-white structure.

[0055] Among them, the component performance index can be represented by a numerical range, and the value range of the design parameter can be a numerical range, a category range (for example, material categories such as steel, aluminum alloy, carbon fiber, etc.), and a logical range (for example, "yes" or "no"), which is not limited here. Traverse the component-level decision tree model, select the component performance index that is the same as the target component index as the matching component performance index, and use the value range of the design parameters on the same branch as the matching component performance index as the value range of the design parameters of the corresponding component of the body-in-white structure. In some embodiments, the number of matching component performance indicators may be 1 or more. When the number of matching component performance indicators is multiple, a preset screening rule can be used to select a preset number of component performance indicators from multiple matching component performance indicators. Among them, the screening rules can be set according to actual needs. For example, screening can be performed based on the discreteness of the branch. The preset number can be 1 or more, which is not limited here.

[0056] For example, the target performance index of the body-in-white structure is collision performance. The performance indexes of the components corresponding to the collision performance include the specific energy absorption (SEA) of the beam and the crush force efficiency (CFE) of the side structure. Among them, the specific energy absorption is used to indicate the energy that a material or structure per unit mass can absorb during a collision. In the body-in-white structure, the SEA values ​​of the main energy-absorbing components such as the transverse and longitudinal beams are usually evaluated. These components are the first to contact and absorb a large amount of energy in a frontal collision to reduce the impact of the collision on the passenger compartment. The crush force efficiency is used to indicate the ratio of the average crush force actually generated by the structure during a collision to the theoretical crush force. In the body-in-white structure, CFE is mainly used to evaluate side structural components such as the B-pillar and the door anti-collision beam. These components play a key supporting and energy-absorbing role in side collisions, and the stability and consistency of their crush force are crucial to protecting the occupants in the car. Based on this, the design parameters corresponding to the specific energy absorption of the beam may include the transition curvature radius of the joint between the rear cross beam of the roof and the upper side beam, the size of the rear cross beam of the roof in the width direction of the frame, and the Y-direction width of the sill beam, where the Y-direction may represent the direction perpendicular to the length direction of the vehicle. The design parameters corresponding to the crushing force efficiency of the side structure may include the thickness of the B-pillar and the thickness of the sill beam.

[0057] Taking the collision performance as an example, the target performance indicator of the body-in-white structure is combined with Figure 3 A schematic diagram illustrating a component-level decision tree model provided in an embodiment of the present application. Figure 3 As shown in Figure 2, the component performance indicators of the component-level decision tree model include the specific energy absorption of the beam and the crushing force efficiency of the side structure. Figure 3 The above two component performance indicators are integrated, and the working condition where the specific energy absorption of the beam is greater than or equal to 16KJ / kg, and the crushing force efficiency of the side structure is greater than or equal to 45% is marked as excellent collision performance of the body-in-white structure, marked with e. The working condition where the specific energy absorption of the beam is greater than or equal to 13.64KJ / kg and less than 16KJ / kg, and the crushing force efficiency of the side structure is greater than or equal to 35% and less than 45% is marked as average collision performance of the body-in-white structure, marked with g. The working condition where the specific energy absorption of the beam is less than 13.64KJ / kg, and the crushing force efficiency of the side structure is less than 35% is marked as poor collision performance of the body-in-white structure, marked with b. The design parameters of the components include the thickness of the B-pillar, the thickness of the sill beam, the transition curvature radius of the joint between the rear cross beam of the roof and the upper side beam, the size of the rear cross beam of the roof in the width direction of the frame, and the Y-direction width of the sill beam, which are marked with d, n, m, h, and t respectively. Figure 3As shown, there are 5 branches with excellent collision performance of the body-in-white structure, 7 branches with average collision performance of the body-in-white structure, and 5 branches with poor collision performance of the body-in-white structure. Exemplarily, when the target performance indicator of the body-in-white structure is excellent collision performance, by traversing the system-level decision tree model (not shown in the figure), it can be known that the target component indicators corresponding to the collision performance include the specific energy absorption of the beam and the crushing force efficiency of the side structure, and the specific energy absorption of the beam is greater than or equal to 16KJ / kg, and the crushing force efficiency of the side structure is greater than or equal to 45%. Afterwards, traverse as shown in the figure. Figure 3 The component-level decision tree model shown selects five branches whose specific energy absorption of the beam is greater than or equal to 16 kJ / kg and whose crushing force efficiency of the side structure is greater than or equal to 45%. By screening the five branches, the value range of the design parameters of the component corresponding to one branch is selected.

[0058] In some embodiments, after the component-level decision tree model training is completed, the value ranges of multiple design parameters in the component-level decision tree model can be simulated and verified to obtain simulation verification results. The simulation verification results are used to indicate whether the performance indicators of the body-in-white structure corresponding to the value ranges of the design parameters meet the target performance indicators. The simulation verification results include the number of performance indicators of the body-in-white structure corresponding to the value ranges of the design parameters that meet the target performance indicators and the number of performance indicators of the body-in-white structure corresponding to the value ranges of the design parameters that do not meet the target performance indicators. Please continue to refer to Figure 3 , Figure 3 b(30.0) is used to indicate that the number of performance indicators of the body-in-white structure corresponding to the value range of the design parameter that meet the target performance indicators (for example, poor collision performance) is 30, and g(11.0 / 33.0) is used to indicate that the number of performance indicators of the body-in-white structure corresponding to the value range of the design parameter that meet the target performance indicators (for example, average collision performance) is 11, and the number of performance indicators of the body-in-white structure corresponding to the value range of the design parameter that do not meet the target performance indicators is 3.

[0059] In some embodiments, after determining the value range of the design parameters of the components corresponding to the body-in-white structure, the method further includes: verifying the value range of the design parameters to obtain a verification result. Wherein, verifying the value range of the design parameters may include performing simulation verification according to the value range of the design parameters to obtain a simulation verification result; and performing tests according to the value range of the design parameters to obtain a test result. When performing tests according to the value range of the design parameters, corresponding components may be produced according to the value range of the design parameters, the body-in-white structure may be assembled using the produced components, and the performance indicators of the body-in-white structure may be confirmed. If the performance indicators of the body-in-white structure meet the preset requirements, the body-in-white structure may be produced using the value range of the design parameters; if the performance indicators of the body-in-white structure do not meet the preset requirements, the value range of the design parameters may be readjusted. Wherein, the preset requirements may be determined based on the vehicle performance indicators output by the user.

[0060] The above-mentioned body-in-white structure design method provided in the embodiment of the present application describes the relationship between the design parameters of the corresponding components of the body-in-white structure and the performance indicators of the body-in-white through a system-level decision tree model and a component-level decision tree model, and uses the system-level decision tree model and the component-level decision tree model to guide the design of the body-in-white structure, which can reduce the number of simulations and experiments of the body-in-white structure in the optimization design process, reduce development costs, and improve the design efficiency of the body-in-white structure; in addition, the system-level decision tree model and the component-level decision tree model can be quickly rebuilt according to the different design requirements of the user without the need for huge calculations, and a decision tree model for different design requirements can be obtained, which further reduces development costs and improves the design efficiency of the body-in-white structure.

[0061] In at least one embodiment of the present application, the system-level decision tree model and the component-level decision tree model can predict the value range of the design parameters of the components corresponding to the body-in-white structure based on the known vehicle performance indicators, and derive the vehicle performance indicators corresponding to the value range of the design parameters based on the known value range of the design parameters, thereby realizing the bidirectional functions of prediction and decision-making. Figure 4 is a flow chart of a method for constructing a decision tree model provided in an embodiment of the present application, and the method for constructing a decision tree model is applied to a computer device. Figure 4 As shown, the following steps are included: S21, constructing an index-parameter framework corresponding to the body-in-white structure to be designed based on the design data corresponding to the designed body-in-white structure.

[0062] In at least one embodiment of the present application, the design data can be obtained based on the design case of the designed body-in-white structure, and the design data can include the performance index of the whole vehicle, the performance index of the body-in-white structure, the performance index of the components that have a major impact on the performance index of each body-in-white structure, and the adjustable design parameters and the value range corresponding to the design parameters. The index-parameter framework is used to guide the generation of basic data for training decision tree models (for example, system-level decision tree models and component-level decision tree models). The index-parameter framework includes the performance index of the body-in-white structure, the performance index of the components and the design parameters. According to the index-parameter framework, the corresponding relationship between the performance index of the body-in-white structure and the performance index of the components, as well as the corresponding relationship between the performance index of the components and the design parameters can be determined.

[0063] In some embodiments, the design data corresponding to the designed body-in-white structure is used to construct an indicator-parameter framework corresponding to the body-in-white structure to be designed, including: parsing the design data to obtain the performance indicators, component performance indicators and design parameters of the body-in-white structure to be designed; using the performance indicators as the first node of the indicator-parameter framework; using the component performance indicators as the second node of the indicator-parameter framework; using the design parameters as the third node of the indicator-parameter framework; constructing the indicator-parameter framework based on the first node, the second node and the third node. Among them, the second node is used to indicate the next level node of the first node, and the third node is used to indicate the next level node of the second node. The embodiment of the present application uses the performance indicators, component performance indicators and design parameters of the body-in-white structure to construct an indicator-parameter framework, and uses the indicator-parameter framework to guide the generation of basic data for training a decision tree model, improve the accuracy of basic data determination, and then improve the accuracy of decision tree model training, and enhance the design effect of the body-in-white structure.

[0064] Combination Figure 5 A schematic diagram illustrating the structure of the indicator-parameter framework provided in the embodiment of the present application. Figure 5As shown, taking the vehicle performance index as the NVH index as an example, the performance index of the body-in-white structure can include the first-order bending / torsion mode, static bending / torsion stiffness and suspension system dynamic stiffness characteristics, and the first-order bending / torsion mode, static bending / torsion stiffness and suspension system dynamic stiffness characteristics are respectively used as the first node. The above three performance indexes of the body-in-white structure are all related to the beam natural frequency, beam bending / torsion stiffness, test point stiffness, joint stiffness, joint quality, constraint point stiffness and the number of joints, and the beam natural frequency, beam bending / torsion stiffness, test point stiffness, joint stiffness, joint quality, constraint point stiffness and the number of joints are used as the second node. The beam natural frequency, beam bending / torsion stiffness, test point stiffness, joint stiffness and joint quality are all related to the beam design parameters and joint design parameters, and the beam design parameters and joint design parameters are used as the third node of the beam natural frequency, beam bending / torsion stiffness, test point stiffness, joint stiffness and joint quality. The constraint point design parameters are related to the constraint point stiffness, and the constraint point design parameters are used as the third node of the constraint point stiffness. The judgment parameter is related to the number of joints, and the judgment parameter is used as the third node of the number of joints. Among them, the beam design parameters may include beam material parameters (for example, aluminum alloy, high-strength steel, composite materials, etc.), beam geometric parameters (for example, the cross-sectional shape, cross-sectional size, length and curvature radius of the beam, etc.). The joint design parameters may include the material parameters of the joint, the geometric parameters of the joint, etc. The constraint point design parameters may include the coordinate position of the constraint point, the number of constraint points, etc. The judgment parameters may include the joint spacing, the uniformity of the joint strength distribution, the joint formation efficiency, etc.

[0065] S22, using the indicator-parameter framework to determine basic data.

[0066] In at least one embodiment of the present application, basic data is used to train a system-level decision tree model and a component-level decision tree model. The basic data may include test data and simulation data, wherein the test data is obtained by collecting data on the development of the body-in-white structure of multiple models of related models, and the test data includes performance indicators of the body-in-white structure, performance indicators of components, and value ranges of design parameters, etc. The simulation data is obtained by performing simulation processing using a system-level simulation model and a component-level simulation model, and the simulation data includes performance indicators of the body-in-white structure, performance indicators of components, and value ranges of design parameters, etc.

[0067] In some embodiments, the use of the indicator-parameter framework to determine basic data includes: determining vehicle type information corresponding to the body-in-white structure to be designed; determining initial finite element models of multiple designed body-in-white structures that match the vehicle type information; obtaining a finite element model of the body-in-white structure to be designed based on multiple initial finite element models; generating simulation data of the body-in-white structure to be designed using the indicator-parameter framework and the finite element model; obtaining test data of the body-in-white structure to be designed, and combining the simulation data with the test data to obtain the basic data.

[0068] Among them, the same vehicle model information may include multiple car styles. For example, the same vehicle model information may include ordinary styles, sports styles, luxury styles, etc. The body-in-white structures of different car styles may be different, which is not limited here. In some embodiments, a first correspondence between the vehicle model information and the car style and a second correspondence between the car style and the finite element model are pre-set. By querying the first correspondence, other car styles with the same vehicle model information as the body-in-white structure to be designed can be obtained. By querying the second correspondence, a finite element model corresponding to the car style can be obtained (for ease of description, this application is referred to as the "initial finite element model"). Among them, the finite element model is used to indicate that a continuous physical structure (for example, a designed body-in-white structure) is discretized into a finite number of units, and specific physical and mechanical properties are given to these units, so that the behavior of complex structures can be simulated and analyzed on a computer device. The method for establishing a finite element model can refer to the relevant technology and will not be repeated here. Each car style has a corresponding initial finite element model.

[0069] In some embodiments, the finite element model of the body-in-white structure to be designed includes initial finite element models of multiple car models under the same vehicle model information. The multiple initial finite element models correspond to the same design parameters affecting the performance of the body-in-white structure. In this way, when determining the parameter range of the same design parameter, the design results of the multiple initial finite element models can be obtained, thereby realizing the mining of multiple design results at the same time.

[0070] Combination Figure 6 The schematic diagram of the structure of the finite element model provided in the embodiment of the present application is shown in FIG. Figure 6 As shown, there are four car models with the same vehicle model information as the body-in-white structure to be designed, which are recorded as the first car, the second car, the third car and the fourth car respectively, and the body-in-white structure of each car model is different. This application uses the initial finite element models of multiple designed car models with the same vehicle model information to construct a finite element model of the body-in-white structure to be designed, which can achieve the effect of mining a large number of design results of multiple car models at the same time, and improve the design effect of the body-in-white structure.

[0071] In some embodiments, using the indicator-parameter framework and the finite element model to generate simulation data of the body-in-white structure to be designed can include: determining the simulation conditions of the finite element model based on the indicator-parameter framework; building a simulation platform based on the simulation conditions and the finite element model, and generating simulation data using the simulation platform.

[0072] Among them, the simulation conditions are used to represent the decomposition strategy of decomposing the performance indicators of the body-in-white structure into component performance indicators, and decomposing the component performance indicators into the value range of the design parameters. Using the indicator-parameter framework, the relationship between the performance indicators of the body-in-white structure and the performance indicators of the components, as well as the relationship between the performance indicators of the components and the design parameters can be determined, and the simulation conditions can be determined based on the above relationship. For example, the simulation conditions may include decomposing the first-order bending / torsion mode into the natural frequency of the beam, the bending / torsion stiffness of the beam, the test point stiffness, the joint stiffness, the joint mass, the constraint point stiffness and the number of joints. The embodiment of the present application determines the simulation conditions of the finite element model by using the indicator-parameter framework, so that the simulation at the component level can truly reflect the performance of the components.

[0073] In some embodiments, a simulation platform can be built based on simulation conditions and finite element models. The simulation platform can include a system-level simulation platform and a component-level simulation platform. Figure 7 , Figure 7It is a flow chart of simulation data determination and expansion provided by the embodiment of the present application. In the simulation data determination stage, the simulation platform sequentially performs pre-processing, simulation calculation, post-processing and output steps. Among them, the pre-processing includes generating multiple design schemes according to the design of experiment (DOE) method according to the preset value range of the design parameters. Exemplarily, the value ranges of the design parameters of multiple parts in the body-in-white structure are pre-set, for example, the thickness range of the sill beam, the thickness range of the cross beam, the thickness range of the longitudinal beam, the material of the sill beam, the material of the cross beam and the material of the longitudinal beam. By combining the value ranges of the design parameters of the above-mentioned multiple parts using the DOE method, multiple design schemes can be obtained. Among them, the DOE method can include full factorial experimental design, partial factorial experimental design, orthogonal experimental design, etc. The embodiment of the present application takes the DOE method as an example of full factorial experimental design. By using the full factorial experimental design method, the value range of each design parameter and the influence of its interaction on the design result can be fully understood. The simulation calculation includes determining the performance index and component index of the body-in-white structure using a finite element model according to multiple design schemes. Post-processing may include selecting relevant data whose performance indicators of the body-in-white structure meet specified requirements, and the relevant data include data such as the value range of design parameters, component performance indicators, and performance indicators of the body-in-white structure. Afterwards, the value range of design parameters, component performance indicators, and performance indicators of the body-in-white structure are output as simulation data. Exemplarily, the component-level simulation platform is used to process the preset value range of design parameters to obtain the component performance indicators; then the system-level simulation platform is used to process the component performance indicators to obtain the performance indicators of the body-in-white structure. In this way, using the simulation platform, the value range of design parameters, component performance indicators, and performance indicators of the body-in-white structure can be obtained, and the above data is used as simulation data for subsequent training and verification of system-level decision tree models and component-level decision tree models.

[0074] In some embodiments, after outputting the simulation data, the method further includes: detecting whether the simulation data meets the data volume requirement; if the simulation data meets the data volume requirement, entering the simulation data expansion phase; if the simulation data does not meet the data requirement, continuing to generate simulation data. Wherein, "reach" may include greater than or equal to. The embodiment of the present application performs data volume detection on the simulation data, and continues to generate simulation data when the simulation data does not meet the data requirement. By generating a large amount of simulation data, the robustness and stability of the decision tree model can be enhanced.

[0075] In at least one embodiment of the present application, in order to increase the diversity of data and avoid the problem of model overfitting, data expansion can be performed, for example, the test data is expanded, or the simulation data is expanded, or the basic data is expanded. The embodiment of the present application is explained by taking the expansion of simulation data as an example, and the method also includes: using a preset data expansion model to expand the simulation data to obtain the expanded simulation data; updating the basic data according to the expanded simulation data, and using the updated basic data to train the system-level decision tree model and the component-level decision tree model. Among them, the expanded simulation data and the test data are used as the updated basic data. The simulation data is used as the source of the expanded data, and is split into an expanded training set and an expanded verification set. The expanded training set is used to train the data expansion model, and the expanded verification set is used to evaluate the performance of the trained data expansion model. For example, 70% to 80% of the data in the simulation data is used as an expanded training set, and the remaining data is used as an expanded verification set. Among them, the data expansion model can include a generative adversarial network model, a variational autoencoder model, a support vector machine regression model, a Gaussian mixture model, etc., which are not limited here.

[0076] Please continue reading Figure 7 In the expansion stage of simulation data, the simulation data is divided into an expanded training set and an expanded verification set; based on the expanded training set, multiple preset models are used to perform model training to obtain multiple trained data expansion models; a target data expansion model is selected from multiple trained data expansion models based on model performance; the expanded verification set is used to verify whether the target data expansion model passes the verification, and if the verification passes, the verified target data expansion model is used to perform data expansion to obtain expanded simulation data; if the verification fails, the steps of performing model training based on the expanded training set using multiple preset models to obtain multiple trained data expansion models are repeated until the data expansion model passes the verification.

[0077] In some embodiments, model performance may include but is not limited to data expansion rate, model complexity, training time, memory usage and other performance, wherein the data expansion rate is used to evaluate the rate at which the data expansion model generates new data. The greater the data expansion rate, the better the model performance; the smaller the data expansion rate, the worse the model performance. Model complexity is used to evaluate the complexity of the model structure. Training time is used to indicate the training iteration time of the model. Memory usage is used to indicate the memory usage of the model during training and prediction. The embodiment of the present application selects a model with better model performance from multiple trained data expansion models as the target data expansion model based on model performance, which can improve the effect of data expansion. In some embodiments, using an expanded validation set to verify whether the target data expansion model passes may include: using the target data expansion model to generate new data, and using a preset error function to determine the error value between the new data and the expanded validation set; if the error value is less than or equal to the error threshold, then it is determined that the verification is passed; if the error value is greater than the error threshold, then it is determined that the verification fails. Among them, the error function may include a mean square error function, a cross entropy loss function, etc., and the error threshold can be set according to actual needs and is not limited here. The embodiment of the present application can evaluate the quality of the generated data and the generalization ability of the model on the new data by comparing the new data generated by the data expansion model with the expanded validation set, thereby improving the accuracy of the data expansion model training.

[0078] In the body-in-white structure design method provided in the embodiment of the present application, the basic data is expanded using a preset data expansion model to obtain the expanded basic data, and the expanded basic data is used to train a system-level decision tree model and a component-level decision tree model, so as to increase the diversity of the data, improve the accuracy of the system-level decision tree model and the component-level decision tree model, and then improve the effect of the body-in-white structure design.

[0079] S23, constructing the system-level decision tree model and the component-level decision tree model according to the indicator-parameter framework and the basic data.

[0080] In at least one embodiment of the present application, during training, the system-level decision tree model is first trained, and then the component-level decision tree model to be trained is determined based on the determined system-level decision tree model. Based on this, the basic data is first split according to the indicator-parameter framework to obtain the first data suitable for the system-level decision tree model training; then the basic data is split according to the indicator-parameter framework to obtain the second data suitable for the component-level decision tree model training.

[0081] In some embodiments, the system-level decision tree model and the component-level decision tree model are constructed based on the indicator-parameter framework and the basic data, including: using the indicator-parameter framework to parse the basic data to obtain first data and second data; based on the first data, using the first decision tree algorithm to obtain a system-level decision tree model; based on the second data, using the second decision tree algorithm to obtain a component-level decision tree model. Among them, the first data includes performance indicators and component performance indicators of multiple designed body-in-white structures, and the second data includes component performance indicators and design parameters of multiple designed body-in-white structures. The first decision tree algorithm and the second decision tree algorithm may be the same or different, and there is no limitation here. The first decision tree algorithm and the second decision tree algorithm may include an iterative dichotomiser 3 (ID3) algorithm, a C4.5 algorithm, a classification and regression tree (CART) algorithm, etc. Among them, the first decision tree algorithm and the second decision tree algorithm can be determined according to the data type and / or data scale. For example, if the data set contains a large number of numerical features, the CART algorithm can be selected; if the data set contains more categorical features, the ID3 algorithm or the C4.5 algorithm can be selected; if the data volume is small and the feature dimension is low, the ID3 algorithm or the C4.5 algorithm can be selected; if the data volume is large, the CART algorithm can be selected. The embodiment of the present application uses an indicator-parameter framework to decompose the basic data to obtain the first data used for system-level decision tree model training and the second data for component-level decision tree model training. The adapted data features are selected according to the model requirements, which can reduce noise interference and improve the training accuracy of the system-level decision tree model and the component-level decision tree model.

[0082] For example, taking the vehicle performance index as the NVH index as an example, according to the index-parameter framework, the performance index of the body-in-white structure can include the first-order bending / torsion mode, static bending / torsion stiffness, and the dynamic stiffness characteristics of the suspension system. When the performance index of the body-in-white structure is the first-order bending / torsion mode, static bending / torsion stiffness, and the dynamic stiffness characteristics of the suspension system, the component performance index can include the beam natural frequency, beam bending / torsion stiffness, test point stiffness, joint stiffness, joint mass, constraint point stiffness, and the number of joints. Based on this, the basic data is analyzed, and the first-order bending / torsion mode and its corresponding beam natural frequency, beam bending / torsion stiffness, test point stiffness, joint stiffness, joint mass, constraint point stiffness and number of joints; the static bending / torsion stiffness and its corresponding beam natural frequency, beam bending / torsion stiffness, test point stiffness, joint stiffness, joint mass, constraint point stiffness and number of joints; and the dynamic stiffness characteristics of the suspension system and its corresponding beam natural frequency, beam bending / torsion stiffness, test point stiffness, joint stiffness, joint mass, constraint point stiffness and number of joints are taken as the first data.

[0083] For example, taking the beam natural frequency as an example of the performance index of the component, according to the index-parameter framework, the value range of the design parameter of the component may include the threshold beam thickness range, the cross beam thickness range, the longitudinal beam thickness range, the threshold beam material, the cross beam material and the longitudinal beam material, etc. Based on this, the beam natural frequency and its corresponding threshold beam thickness range, cross beam thickness range, longitudinal beam thickness range, threshold beam material, cross beam material and longitudinal beam material are used as the second data.

[0084] Combination Figure 8 A schematic diagram illustrating the process of training a decision tree model provided in an embodiment of the present application. Figure 8 As shown, the training process of the decision tree model includes the training process of the system-level decision tree model and the training process of the component-level decision tree model, and the training process of the system-level decision tree model is first and the training process of the component-level decision tree model is later.

[0085] In the training process of the system-level decision tree model, the performance index of the body-in-white structure is set as the leaf node content of the system-level decision tree model as the model requirement of the system-level decision tree model; according to the determined model requirements, the first data is divided into a first training set and a first validation set; according to the first training set, the first decision tree algorithm is used for training; the trained system-level decision tree model is pruned to obtain an initial system-level decision tree model; the initial system-level decision tree model is verified using the first validation set to verify whether the initial system-level decision tree model meets the accuracy requirements; if the verification result is that the initial system-level decision tree model meets the accuracy requirements, the initial system-level decision tree model is used as the target system-level decision tree model, and the training process of the component-level decision tree model is executed; if the verification result is that the initial system-level decision tree model does not meet the accuracy requirements, the first decision tree algorithm is used to train again according to the first training set until the verification passes and the target system-level decision tree model is obtained. Among them, the accuracy requirements can be set according to actual needs and are not limited here. Pruning is used to remove some branches in the model to avoid overfitting of the model. For pruning the decision tree model, please refer to the relevant technology, which will not be described in detail here.

[0086] In the training process of the component-level decision tree model, the component performance index is set as the leaf node content of the component-level decision tree model as the model requirement of the component-level decision tree model; according to the determined model requirements, the second data is divided into a second training set and a second validation set; according to the second training set, the second decision tree algorithm is used for training; the trained component-level decision tree model is pruned to obtain an initial component-level decision tree model; the initial component-level decision tree model is verified using the second validation set to verify whether the initial component-level decision tree model meets the accuracy requirements; if the verification result shows that the initial component-level decision tree model meets the accuracy requirements, the initial component-level decision tree model is used as the target component-level decision tree model; if the verification result shows that the initial component-level decision tree model does not meet the accuracy requirements, the second decision tree algorithm is used for training again according to the second training set until the verification passes to obtain the target component-level decision tree model. Among them, the accuracy requirements can be set according to actual needs and are not limited here.

[0087] In the body-in-white structure design method provided in the embodiment of the present application, the index-parameter framework corresponding to the body-in-white structure to be designed is used to guide the generation of basic data, which can improve the accuracy of the basic data; and the present application uses the index-parameter framework and the basic data to construct the system-level decision tree model and the component-level decision tree model, so that the system-level decision tree model can describe the relationship between the performance index of the body-in-white structure and the performance index of the component, and the component-level decision tree model can describe the correspondence between the performance index of the component and the value range of the design parameter. The subsequent use of the system-level decision tree model and the component-level decision tree model to guide the design of the body-in-white structure can reduce the number of simulations and experiments of the body-in-white structure in the optimization design process, reduce development costs, and improve the design efficiency of the body-in-white structure.

[0088] See also Fig. 9 , Fig. 9 1 is a schematic diagram of a structure of a body-in-white structure design device provided in an embodiment of the present application. In some embodiments, the body-in-white structure design device 20 may include a plurality of functional modules composed of computer program segments. The computer program of each program segment in the body-in-white structure design device 20 may be stored in the memory of the computer device 10 and executed by at least one processor to execute (see Figure 2 Description) Function of body-in-white structural design.

[0089] In this embodiment, the body-in-white structure design device 20 can be divided into multiple functional modules according to the functions it performs. The functional modules may include: an index determination module 201 and a parameter determination module 202. The module referred to in this application refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, which are stored in a memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0090] The index determination module 201 may be used to determine a target performance index corresponding to the body-in-white structure to be designed.

[0091] The index determination module 201 is also used to determine the target component index corresponding to the body-in-white structure to be designed based on the target performance index and using a preset system-level decision tree model.

[0092] The parameter determination module 202 is used to determine the value range of the design parameters of the components corresponding to the body-in-white structure based on the target component indicators and using a preset component-level decision tree model. The value range of the design parameters is used to design the body-in-white structure so that the body-in-white structure achieves the target performance indicators.

[0093] In some embodiments, the index determination module 201 can also be used to obtain the whole vehicle performance index corresponding to the vehicle to be designed; analyze the whole vehicle performance index to obtain the target performance index of the body-in-white structure corresponding to the vehicle.

[0094] In some embodiments, the indicator determination module 201 can also be used to traverse the system-level decision tree model to obtain performance indicators that match the target performance indicators; and determine that the component performance indicators corresponding to the matching performance indicators are the target component indicators corresponding to the body-in-white structure to be designed.

[0095] In some embodiments, the parameter determination module 202 can also be used to traverse the component-level decision tree model to obtain component performance indicators that match the target component indicators; determine that the value range of the design parameters of the component corresponding to the matching component performance indicators is the value range of the design parameters of the component corresponding to the body-in-white structure.

[0096] In some embodiments, the parameter determination module 202 can also be used to construct an indicator-parameter framework corresponding to the body-in-white structure to be designed based on the design data corresponding to the designed body-in-white structure; use the indicator-parameter framework to determine basic data; and construct the system-level decision tree model and the component-level decision tree model based on the indicator-parameter framework and the basic data.

[0097] In some embodiments, the parameter determination module 202 can also be used to parse the design data to obtain the performance indicators, component performance indicators and design parameters of the body-in-white structure to be designed; use the performance indicators as the first node of the indicator-parameter framework; use the component performance indicators as the second node of the indicator-parameter framework; use the design parameters as the third node of the indicator-parameter framework; and construct the indicator-parameter framework based on the first node, the second node and the third node.

[0098] In some embodiments, the parameter determination module 202 can also be used to determine the vehicle model information corresponding to the body-in-white structure to be designed; determine the initial finite element models of multiple designed body-in-white structures that match the vehicle model information; obtain the finite element model of the body-in-white structure to be designed based on multiple initial finite element models; use the indicator-parameter framework to determine the simulation conditions of the finite element model, and obtain the basic data based on the simulation conditions and the finite element model.

[0099] In some embodiments, the parameter determination module 202 can also be used to utilize the indicator-parameter framework to parse the basic data to obtain first data and second data; based on the first data, using a first decision tree algorithm, obtain a system-level decision tree model; based on the second data, using a second decision tree algorithm, obtain a component-level decision tree model.

[0100] It can be understood that the body-in-white structure design device 20 and the body-in-white structure design method of the above-mentioned embodiment belong to the same inventive concept, and the specific implementation method of each module of the body-in-white structure design device 20 corresponds to the various steps of the body-in-white structure design method in the above-mentioned embodiment, which will not be repeated in this application.

[0101] The module division described above is a logical function division, and there may be other division methods in actual implementation. In addition, the functional modules in each embodiment of the present application may be integrated in the same processing unit, or each module may exist physically separately, or two or more modules may be integrated in the same unit. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0102] then Figure 1 In the description of the computer device, the communication module 101 may include a wired communication module and / or a wireless communication module. The wired communication module may provide one or more wired communication solutions such as universal serial bus (USB), controller area network bus (CAN, Controller Area Network). The wireless communication module may provide one or more wireless communication solutions such as wireless fidelity (Wi-Fi), Bluetooth (BT), mobile communication network, frequency modulation (FM), near field communication technology (NFC), infrared technology (IR), etc.

[0103] In some embodiments, the memory 102 may include one or more random access memories (RAM) and one or more non-volatile memories (NVM). The random access memory can be directly read and written by the processor 103, and can be used to store executable programs (such as machine instructions) of other running programs, and can also be used to store user and application data. The random access memory may include static random-access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), etc.

[0104] In some embodiments, the non-volatile memory may also store executable programs and user and application data, etc., which may be pre-loaded into the random access memory for direct reading and writing by the processor 103. The non-volatile memory may include a disk storage device and a flash memory.

[0105] In other embodiments, the computer device 10 further includes an external memory interface for connecting to an external memory to expand the storage capacity of the computer device 10 .

[0106] In some embodiments, the processor 103 may include one or more processing units, for example, the processor 103 may include an application processor (AP), a modem processor, a graphics processor (GPU), an image signal processor (ISP), a controller, a video codec, a digital signal processor (DSP), and / or a neural-network processing unit (NPU), etc. Different processing units may be independent devices or integrated into one or more processors.

[0107] It is to be understood that the structure illustrated in the embodiment of the present application does not constitute a specific limitation on the computer device 10. In other embodiments of the present application, the computer device 10 may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0108] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. The computer program includes program instructions. The method implemented when the program instructions are executed can refer to the methods in the above-mentioned embodiments of the present application.

[0109] The computer-readable storage medium may be an internal memory of the computer device described in the above embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device.

[0110] In some embodiments, the computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of the computer device, etc.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0112] The modules described as separate components may or may not be physically separated, and the components shown as modules 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 modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0113] In addition, each functional module in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0114] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present application is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present application. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0115] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in this application can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical solution of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present application.

Claims

1. A method for designing a body-in-white structure, characterized in that: The method comprises: Determine the target performance indicators corresponding to the body-in-white structure to be designed; Based on the target performance index, a preset system-level decision tree model is used to determine the target component index corresponding to the body-in-white structure to be designed; Based on the target component index, a preset component-level decision tree model is used to determine the value range of the design parameters of the components corresponding to the body-in-white structure. The value range of the design parameters of the components is used to design the body-in-white structure so that the body-in-white structure achieves the target performance index.

2. The body-in-white structure design method according to claim 1, characterized in that: The step of determining the target performance index corresponding to the body-in-white structure to be designed includes: Obtain vehicle performance indicators corresponding to the vehicle to be designed; The whole vehicle performance index is analyzed to obtain the target performance index of the vehicle corresponding to the body-in-white structure.

3. The body-in-white structure design method according to claim 1, characterized in that: The method of determining target component indicators corresponding to the body-in-white structure to be designed based on the target performance indicators and using a preset system-level decision tree model includes: Traversing the system-level decision tree model to obtain a performance indicator that matches the target performance indicator; The component performance index corresponding to the matched performance index is determined to be the target component index corresponding to the body-in-white structure to be designed.

4. The body-in-white structure design method according to claim 1, characterized in that: The method of determining the value range of the design parameter of the component corresponding to the body-in-white structure by using a preset component-level decision tree model based on the target component index includes: Traversing the component-level decision tree model to obtain component performance indicators that match the target component indicators; It is determined that the value range of the design parameter of the component corresponding to the matching component performance index is the value range of the design parameter of the component corresponding to the body-in-white structure.

5. The method for designing a body-in-white structure according to claim 1, wherein: The method further comprises: Based on the design data corresponding to the designed body-in-white structure, an index-parameter framework corresponding to the body-in-white structure to be designed is constructed; Using the indicator-parameter framework, determine the basic data; The system-level decision tree model and the component-level decision tree model are constructed based on the indicator-parameter framework and the basic data.

6. The body-in-white structure design method according to claim 5, characterized in that: The step of constructing an index-parameter framework corresponding to the body-in-white structure to be designed based on the design data corresponding to the designed body-in-white structure includes: Analyzing the design data to obtain performance indicators, component performance indicators and design parameters of the body-in-white structure to be designed; Using the performance indicator as the first node of the indicator-parameter framework; Using the component performance index as the second node of the index-parameter framework; Using the design parameter as the third node of the indicator-parameter framework; The indicator-parameter framework is constructed based on the first node, the second node and the third node.

7. The body-in-white structure design method according to claim 5, characterized in that: The use of the indicator-parameter framework to determine basic data includes: Determining vehicle model information corresponding to the body-in-white structure to be designed; determining initial finite element models of a plurality of designed body-in-white structures matching the vehicle model information; Based on the multiple initial finite element models, a finite element model of the body-in-white structure to be designed is obtained; Generate simulation data of the body-in-white structure to be designed by using the index-parameter framework and the finite element model; The test data of the body-in-white structure to be designed is obtained, and the simulation data and the test data are combined to obtain the basic data.

8. The method for designing a body-in-white structure according to claim 5, wherein: The step of constructing the system-level decision tree model and the component-level decision tree model based on the indicator-parameter framework and the basic data includes: Utilizing the index-parameter framework, parsing the basic data to obtain first data and second data, wherein the first data includes a plurality of performance indexes and component performance indexes of the designed body-in-white structure, and the second data includes a plurality of component performance indexes and design parameters of the designed body-in-white structure; Based on the first data, using a first decision tree algorithm, obtaining the system-level decision tree model; Based on the second data, the component-level decision tree model is obtained using a second decision tree algorithm.

9. A computer device, characterized in that: The computer device includes a processor and a memory, and the processor is used to implement the body-in-white structure design method as claimed in any one of claims 1 to 8 when executing a computer program stored in the memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of a computer device, the method for designing a body-in-white structure according to any one of claims 1 to 8 is implemented.

Images