Automobile two-force rod buckling force prediction method, device, equipment and medium

Abstract

The invention discloses a method for predicting the buckling force of a two-force rod of an automobile. The method comprises the following steps: acquiring 3D grid data of the two-force rod of the automobile; inputting the 3D grid data of the automobile two-force rod into the trained buckling force prediction model, and predicting the buckling force of the two-force rod through the buckling force prediction model to obtain a prediction result of the buckling force of the two-force rod; wherein the buckling force prediction model is obtained by training the initial neural network model by taking 3D grid data of the two-force rod of the automobile as input and taking the buckling force of the two-force rod as output; according to the method, the grid data is directly used for the 3D model file of the automobile two-force rod, the two-force rod buckling force regression prediction task is achieved through the 3D neural network for conducting feature extraction on the grid data, the tedious finite element analysis work in the two-force rod simulation process can be replaced, the time performance is greatly improved, and the method is suitable for popularization and application. And moreover, the working efficiency of simulation engineers is improved, the AI-enabled simulation service is really realized, and the method has a wide application prospect.

Description

Technical Field

[0001] The present application relates to the technical field of buckling force analysis, and in particular to a method, device, equipment and medium for predicting the buckling force of a two-force rod of an automobile. Background Art

[0002] Buckling is a failure mode in engineering calculations that may occur when a structure is subjected to compressive stress. Buckling is characterized by sudden lateral deformation of structural members and leads to structural instability. For compressive members, buckling is the most common cause of instability. The general steps of nonlinear buckling analysis are: 1) establish a finite element model, 2) eigenvalue buckling analysis, 3) apply the first-order modal initial defect, 4) solve the stability limit load, and 5) post-processing. Traditional CAE simulation calculates the buckling force of an object. Most of them use HYPERWORKS software combined with finite element analysis of ABAQUS solver to achieve the above steps and draw conclusions. However, each simulation calculation process of the buckling force of a two-force rod of an automobile is time-consuming and laborious. For each change in shape or material, it means that the simulation engineer has to recalculate all the steps described above, which is time-consuming and inefficient. A fast buckling prediction method based on the 3D model data of the object is particularly necessary. Summary of the invention

[0003] In view of the above-mentioned shortcomings of the prior art, the present application provides a method, device, equipment and medium for predicting the buckling force of an automobile two-force rod, which are used to solve the defect of the complex buckling analysis process in the prior art.

[0004] To achieve the above objectives and other objectives, the present application provides a method for predicting the buckling force of a two-force rod of an automobile, the prediction method comprising:

[0005] Get the 3D mesh data of the two-force rod of the car;

[0006] The 3D mesh data of the automobile two-force bar is input into a trained buckling force prediction model, and the buckling force of the two-force bar is predicted by the buckling force prediction model to obtain a prediction result of the buckling force of the two-force bar; wherein the buckling force prediction model is obtained by training an initial neural network model with the 3D mesh data of the automobile two-force bar as input and the buckling force of the two-force bar as output.

[0007] In one embodiment of the present application, the buckling force prediction model includes a feature extraction layer, a global pooling layer, and a fully connected layer; the step of predicting the buckling force of the two-force bar by using the buckling force prediction model to obtain the prediction result of the buckling force of the two-force bar includes:

[0008] Performing feature extraction on the 3D mesh data through the feature extraction layer to obtain a first mesh feature;

[0009] Using the global pooling layer to perform global average pooling processing on the first grid feature to obtain a second grid feature;

[0010] All the second mesh features are connected through the fully connected layer, and regression prediction of the buckling force of the two-force bar is performed based on the connected second mesh features.

[0011] In one embodiment of the present application, the feature extraction layer includes a plurality of feature extraction modules connected in sequence, and the output of the previous feature extraction module serves as the input of the next feature extraction module; the feature extraction module includes:

[0012] Convolution unit, used to extract features from input data;

[0013] The pooling unit is used to perform pooling operations on the features output by the convolution unit to obtain new features.

[0014] In one embodiment of the present application, the 3D mesh data includes:

[0015] The ratio of the angle between a first plane consisting of three sides and a second plane consisting of three sides, the inner angle corresponding to the target line in the first plane, the inner angle corresponding to the target line in the second plane, the length of a first perpendicular line having a first perpendicular relationship with the target line in the first plane and the length of a second perpendicular line having a second perpendicular relationship with the target line in the second plane, wherein the target line is a common side of the first plane and the second plane.

[0016] In one embodiment of the present application, when the feature extraction layer extracts features from the 3D mesh data, the feature extraction layer extracts features from the 3D mesh data through edge convolution, including:

[0017] Get the features of the common edge and the features of the neighborhood of the common edge;

[0018] A weighted sum is performed on the features of the common edge and the features of the neighborhood of the common edge to extract features from the input data.

[0019] In one embodiment of the present application, before performing weighted summation on the features of the common edge and the features of the neighborhood of the common edge, the method further includes:

[0020] The features of the neighborhood of the common edge are changed by a transformation relationship; the transformation relationship is expressed as:

[0021] (e 1 ,e 2 ,e 3 ,e 4 )=(|ac|,a+c,|bd|,b+d)

[0022] Among them, (e1 ,e 2 ,e 3 ,e 4 ) represents the characteristics of the neighborhood of the common edge after the transformation, a, b, c, d represent the four neighboring edges of the common edge, edge a and edge b are located in the same plane, edge c and edge d are located in the same plane, edge c and edge b have a common point, and edge a and edge d have a common point.

[0023] In one embodiment of the present application, the pooling unit performs a pooling operation on the features output by the convolution unit through a pooling relationship, and the pooling relationship is expressed as:

[0024]

[0025] Among them, a, b, c, and d represent the four adjacent edges of the common edge e, edge a and edge b are located in the same plane, edge c and edge d are located in the same plane, edge c and edge b have a common point, and edge a and edge d have a common point; f(e) represents the length of the common edge, f(a) represents the length of edge a, f(b) represents the length of edge b, f(c) represents the length of edge c, and f(d) represents the length of edge d.

[0026] In order to achieve the above-mentioned and other purposes, the present application provides a device for predicting the buckling force of a two-force rod of an automobile, the prediction device comprising:

[0027] A data acquisition module is used to acquire 3D mesh data of the two-force rod of the automobile;

[0028] A prediction module is used to input the 3D mesh data of the automobile two-force bar into a trained buckling force prediction model, predict the buckling force of the two-force bar through the buckling force prediction model, and obtain a prediction result of the buckling force of the two-force bar; wherein the buckling force prediction model is obtained by training an initial neural network model with the 3D mesh data of the automobile two-force bar as input and the buckling force of the two-force bar as output.

[0029] In order to achieve the above-mentioned purpose and other related purposes, the present application provides a vehicle two-force rod buckling force prediction device, comprising:

[0030] one or more processors; and

[0031] The memory is used to store one or more programs. When the one or more programs are executed by the one or more processors, the memory implements the method described.

[0032] To achieve the above objectives and other related objectives, the present application provides one or more machine-readable media having instructions stored thereon, which, when executed by one or more processors, enable the processors to execute the described methods.

[0033] Beneficial effects of this application:

[0034] The present application provides a method for predicting the buckling force of a two-force bar of an automobile, comprising: obtaining 3D mesh data of a two-force bar of an automobile; inputting the 3D mesh data of the two-force bar of the automobile into a trained buckling force prediction model, predicting the buckling force of the two-force bar by means of the buckling force prediction model, and obtaining a prediction result of the buckling force of the two-force bar; wherein the buckling force prediction model is obtained by training an initial neural network model with the 3D mesh data of the two-force bar of the automobile as input and the buckling force of the two-force bar as output; the present application directly uses mesh data for the 3D model file of the two-force bar of the automobile, and realizes the buckling force regression prediction task of the two-force bar by means of a 3D neural network that extracts features from the mesh data, which can not only replace the tedious finite element analysis work in the simulation process of the two-force bar, greatly improve the time performance, but also improve the work efficiency of simulation engineers, and truly realize the use of AI to empower simulation business, and has broad application prospects.

[0035] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0037] Figure 1 This is a schematic diagram of the structure of 3D mesh data according to an embodiment of the present application;

[0038] Figure 2 A schematic diagram of the structure of a buckling force prediction model according to an embodiment of the present application;

[0039] Figure 3 This is a flow chart of a method for predicting buckling force of a two-force rod of an automobile according to an embodiment of the present application;

[0040] Figure 4 A schematic diagram of the composition of 3D mesh data according to an embodiment of the present application;

[0041] Figure 5 A schematic diagram of 3D mesh data according to an embodiment of the present application;

[0042] Figure 6 A schematic diagram of a pooling process according to an embodiment of the present application;

[0043] Figure 7 This is a principle block diagram of a vehicle two-force rod buckling force prediction device according to one embodiment of the present application;

[0044] Figure 8 A schematic diagram of the structure of a computer system suitable for implementing the memory of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0045] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0046] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0047] Although the terms "first", "second", "A", and "B", etc. may be used herein to describe various elements, these elements should not be limited by these terms and are only used to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element without departing from the scope of the following technology. The term "and / or" includes a combination of multiple related items or any of the multiple related items.

[0048] As used herein, unless the context indicates otherwise, the singular form is intended to include the plural form, and it will be understood that the term "comprising" means the presence of stated features, quantities, steps, operations, elements, or combinations thereof, but does not exclude the presence or addition of one or more other features, quantities, steps, operations, elements, components, or combinations thereof.

[0049] Before the detailed description, it is intended to clarify that the division of components in this specification is divided only by the main function of each component. That is, two or more components to be described below can be combined into one component, or can be divided into two or more components according to more detailed functions. In addition to the main function of the component, each component to be described below can also perform some or all of the functions of other components, and some of the main functions of each component can be exclusively performed by other components.

[0050] Buckling is a failure mode in engineering calculations and may occur when a structure is subjected to compressive stress. Buckling is characterized by sudden lateral deformation of structural members and resulting in structural instability. For compressive members, buckling is the most common cause of instability. When the current simulation algorithm performs nonlinear buckling analysis, it uses a series of simulation steps such as establishing a local coordinate system through specified rules, applying loads along specified directions, applying constraints of different degrees of freedom at both ends, and calculating buckling analysis. This analysis method is time-consuming and labor-intensive, and prone to errors. Based on this, the present application provides a method for predicting the buckling force of an automobile two-force bar based on 3D mesh data. The method first extracts mesh data from a 3D model file of an automobile two-force bar, and then extracts features by using a 3D neural network model MeshCNN backbone network based on mesh data. Finally, the classification task of the model head network is partially changed to a regression task, and the buckling value of the Mesh feature is regressively predicted.

[0051] In order to implement the buckling force prediction method of the automobile two-force rod disclosed in the present application, it is necessary to train the corresponding buckling force prediction model in advance. The training process of the model includes:

[0052] S1. Obtain a historical data sample set and divide it into a training sample set and a test sample set;

[0053] The data in the historical data sample set are obj files saved by simulation engineers when performing CAE simulation. In the file data, an object is composed of many meshes. For example, in common mesh data, three points form a face to form the entire object. The mesh data structure is as follows: Figure 1 shown.

[0054] S2. training a preset buckling force prediction model using the training sample set;

[0055] Specifically, during training, 60% of the training samples in the training sample set can be used to train the deep learning model, 20% of the training samples can be used to verify the deep learning model, and 20% of the training samples can be used to test the deep learning model. Specifically, the MATLAB deep learning toolbox can be used to train, verify, and test the initial neural network model.

[0056] S3. Using the test sample set to test the trained buckling force prediction model, when the test result meets the preset conditions, the neural network model is determined to be the trained buckling force prediction model.

[0057] This application uses MeshCNN (Mesh Convolutional Neural Network) as a buckling force prediction model. This model uses the three-dimensional model as a graph or manifold to jointly study mesh information such as vertex connections or edge connections. The network mainly directly uses convolution, pooling and other basic components of convolutional neural networks for three-dimensional mesh data, and finally realizes the prediction of buckling force. The specific structure of the model is as follows Figure 2 The parameters of each layer in the model are shown in Table 1.

[0058] In the embodiment of the present application, the composition and parameters of the buckling force prediction model are shown in Table 1.

[0059] Table 1

[0060] <![CDATA[MeshConv f in ×32]]> MeshPool → 600 MeshConv 32×64 MeshPool → 450 MeshConv 64×128 MeshPool → 300 MeshConv 128×256 MeshPool → 279 GlobaAvgPool FC 256×100 FC 100×30 Regression

[0061] After obtaining the trained buckling force prediction model, refer to Figure 3 , Figure 3 This is a flow chart of a method for predicting the buckling force of a two-force rod of an automobile according to an embodiment of the present application. Figure 3 As shown, the method for predicting the buckling force of a two-force rod of an automobile at least includes steps S310 to S320:

[0062] Step S310, obtaining 3D mesh data of the two-force rod of the vehicle;

[0063] 3D mesh data is an obj file saved by simulation engineers when performing CAE simulation. In the file data, an object is composed of many meshes. For example, in common mesh data, three points form a face to form the entire object.

[0064] Step S320, input the 3D mesh data of the automobile two-force bar into the trained buckling force prediction model, predict the buckling force of the two-force bar through the buckling force prediction model, and obtain the prediction result of the buckling force of the two-force bar; wherein the buckling force prediction model is obtained by training the initial neural network model with the 3D mesh data of the automobile two-force bar as input and the buckling force of the two-force bar as output.

[0065] In one embodiment of the present application, the buckling force prediction model includes a feature extraction layer, a global pooling layer, and a fully connected layer; the buckling force of the two-force bar is predicted by the buckling force prediction model, and the step of obtaining the prediction result of the buckling force of the two-force bar includes:

[0066] Step S410, extracting features from the 3D mesh data through a feature extraction layer of the buckling force prediction model to obtain a first mesh feature;

[0067] In the embodiment of the present application, the feature extraction layer includes a plurality of feature extraction modules connected in sequence, and the output of the previous feature extraction module serves as the input of the next feature extraction module;

[0068] The buckling force prediction model performs multiple mesh convolution and mesh pooling operations on the input 3D mesh data through a feature extraction layer including multiple feature extraction modules connected in sequence, and the latter feature extraction module of two adjacent feature extraction modules in the multiple feature extraction modules extracts features again on the features output by the previous feature extraction module.

[0069] Specifically, in the embodiment of the present application, the feature extraction layer includes three feature extraction modules, each of which includes:

[0070] Convolution unit, used to extract features from input data; weighted sum is calculated by sliding the convolution kernel over the input data;

[0071] The pooling unit is used to perform pooling operations on the features output by the convolution unit to obtain new features. Pooling is used for feature dimensionality reduction, which reduces the size of the data space by aggregating and counting the elements in the pooling window.

[0072] In the embodiments of this application, please refer to Figure 4 , Figure 4 Schematic diagram of grid data of an embodiment of the present application. Figure 4 As shown, the 3D mesh data includes:

[0073] The angle between the first plane formed by the three sides and the second plane formed by the three sides, the inner angle corresponding to the target line in the first plane, the inner angle corresponding to the target line in the second plane, the length ratio of the first perpendicular line having a first perpendicular relationship with the target line in the first plane and the second perpendicular line having a second perpendicular relationship with the target line in the second plane (wherein the first perpendicular line and the second perpendicular line are Figure 4 ), wherein the target line is the common edge of the first plane and the second plane.

[0074] In an embodiment of the present application, when the feature extraction layer extracts features from 3D mesh data, it performs feature extraction on the 3D mesh data through edge convolution, including: obtaining features of common edges and features of the neighborhood of common edges; performing weighted summation of features of common edges and features of the neighborhood of common edges to extract features from the input data.

[0075] The side convolution is:

[0076]

[0077] Among them, e represents the feature of the common edge, k0 represents the weight of the feature of the common edge, and e jrepresents the characteristics of the jth adjacent edge, k j Represents the weight of the feature of the jth adjacent edge.

[0078] When performing edge convolution on 3D mesh data, consider a neighborhood of the edge, such as Figure 5 As shown in the figure, for the common edge e, considering the features of its adjacent edges in the counterclockwise direction will include two cases: (a, b, c, d) and (c, d, a, b). This is actually the same for the same edge, but directly performing edge convolution in different orders will produce different point value results.

[0079] In order to eliminate the influence of different orders, before weighted summing the features of the common edge and the features of the neighborhood of the common edge, the method further includes:

[0080] The features of the neighborhood of the common edge are changed through the transformation relationship; the transformation relationship is expressed as:

[0081] (e 1 ,e 2 ,e 3 ,e 4 )=(|ac|,a+c,|bd|,b+d)

[0082] Among them, (e 1 ,e 2 ,e 3 ,e 4 ) represents the characteristics of the neighborhood of the common edge after the transformation, a, b, c, d represent the four neighboring edges of the common edge, edge a and edge b are located in the same plane, edge c and edge d are located in the same plane, edge c and edge b have a common point, and edge a and edge d have a common point.

[0083] Step S420, performing global average pooling processing on the first mesh feature using the global pooling layer of the buckling force prediction model to obtain a second mesh feature;

[0084] In an embodiment of the present application, the global average pooling operation is to calculate the average value of each channel of the feature to obtain a one-dimensional vector with the same number of channels. Each element of this vector represents the average activation value on the corresponding channel, which can be regarded as a global statistical feature of the channel. The advantage of global average pooling is that it can extract the overall information of the feature and average the information in the channel, which helps to reduce the sensitivity of the model to specific positions. After the first grid feature is pooled by the global pooling layer, the first grid feature is pooled to a fixed dimension to obtain a first grid feature of a fixed dimension, that is, a second grid feature. The pooling process can be seen in FIG. Figure 6 shown.

[0085] In the embodiment of the present application, the pooling unit performs a pooling operation on the features output by the convolution unit through a pooling relationship, and the pooling relationship is expressed as:

[0086]

[0087] Among them, a, b, c, and d represent the four adjacent edges of the common edge e, edge a and edge b are located in the same plane, edge c and edge d are located in the same plane, edge c and edge b have a common point, and edge a and edge d have a common point; f(e) represents the length of the common edge, f(a) represents the length of edge a, f(b) represents the length of edge b, f(c) represents the length of edge c, and f(d) represents the length of edge d.

[0088] Step S430, all the second mesh features are connected through a fully connected layer, and regression prediction of the two-force bar buckling force is performed based on the connected second mesh features.

[0089] The fully connected layer includes a first connection layer and a second connection layer;

[0090] The first connection layer connects all the second grid features, and accesses the connected second grid features to the second connection layer, and the buckling force of the two-force bar is regressed and predicted through the second connection layer.

[0091] In summary, the present application directly uses mesh data for the 3D model file of the automobile two-link, and realizes the two-link buckling force regression prediction task through a 3D neural network that performs feature extraction on the mesh; this can not only replace the tedious finite element analysis work in the simulation process, but most importantly, it has a significant improvement in time performance. On the one hand, it improves the work efficiency of simulation engineers, and truly realizes the use of AI to empower simulation business, which has broad application prospects; on the other hand, the present application is not limited to the buckling force prediction of automobile two-links, but can also be used for other industrial business numerical regression tasks to meet more different scenarios and needs, and is also flexible and has application prospects.

[0092] 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.

[0093] Figure 7 FIG. 1 is a block diagram of a vehicle two-force rod buckling force prediction device according to an embodiment of the present application. Figure 6 As shown, a device for predicting the buckling force of a two-force rod of an automobile comprises:

[0094] The data acquisition module 710 is used to acquire 3D mesh data of the two-force rod of the vehicle;

[0095] The prediction module 720 is used to input the 3D mesh data of the automobile two-force bar into the trained buckling force prediction model, predict the buckling force of the two-force bar through the buckling force prediction model, and obtain the prediction result of the buckling force of the two-force bar; wherein the buckling force prediction model is obtained by training the initial neural network model with the 3D mesh data of the automobile two-force bar as input and the buckling force of the two-force bar as output.

[0096] In an embodiment of the present application, the buckling force prediction model includes:

[0097] The feature extraction layer is used to extract features from the 3D mesh data to obtain a first mesh feature;

[0098] A global pooling layer, used to perform global average pooling processing on the first grid features to obtain second grid features;

[0099] The fully connected layer is used to connect all the second mesh features and perform regression prediction of the buckling force of the two-force bar based on the connected second mesh features.

[0100] In the embodiment of the present application, the feature extraction layer includes a plurality of feature extraction modules connected in sequence, and the output of the previous feature extraction module is used as the input of the next feature extraction module; the feature extraction module includes:

[0101] Convolution unit, used to extract features from input data;

[0102] The pooling unit is used to perform pooling operations on the features output by the convolution unit to obtain new features.

[0103] In the embodiment of the present application, the 3D mesh data includes:

[0104] The ratio of the angle between a first plane consisting of three sides and a second plane consisting of three sides, the inner angle corresponding to the target line in the first plane, the inner angle corresponding to the target line in the second plane, the length of a first perpendicular line having a first perpendicular relationship with the target line in the first plane and the length of a second perpendicular line having a second perpendicular relationship with the target line in the second plane, wherein the target line is a common side of the first plane and the second plane.

[0105] In an embodiment of the present application, when the feature extraction layer extracts features from 3D mesh data, it performs feature extraction on the 3D mesh data through edge convolution; the feature extraction layer obtains features of common edges and features of the neighborhood of common edges; and performs weighted summation of features of common edges and features of the neighborhood of common edges to extract features from the input data.

[0106] In the embodiment of the present application, the feature extraction layer performs feature changes on the features of the neighborhood of the common edge through a transformation relationship; the transformation relationship is expressed as:

[0107] (e 1 ,e 2 ,e 3 ,e 4 )=(|ac|,a+c,|bd|,b+d)

[0108] Among them, (e 1 ,e 2 ,e 3 ,e 4 ) represents the characteristics of the neighborhood of the common edge after the transformation, a, b, c, d represent the four neighboring edges of the common edge, edge a and edge b are located in the same plane, edge c and edge d are located in the same plane, edge c and edge b have a common point, and edge a and edge d have a common point.

[0109] In the embodiment of the present application, the pooling unit performs a pooling operation on the features output by the convolution unit through a pooling relationship, and the pooling relationship is expressed as:

[0110]

[0111] Among them, a, b, c, and d represent the four adjacent edges of the common edge e, edge a and edge b are located in the same plane, edge c and edge d are located in the same plane, edge c and edge b have a common point, and edge a and edge d have a common point; f(e) represents the length of the common edge, f(a) represents the length of edge a, f(b) represents the length of edge b, f(c) represents the length of edge c, and f(d) represents the length of edge d.

[0112] It should be noted that the automobile two-force bar buckling force prediction device provided in the above embodiment and the automobile two-force bar buckling force prediction method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment, and will not be repeated here. In actual application, the automobile two-force bar buckling force prediction device provided in the above embodiment can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0113] An embodiment of the present application also provides a device, including: one or more processors; and a memory for storing one or more programs, when the one or more programs are executed by one or more processors, the memory implements the automobile two-force rod buckling force prediction method in the above embodiment.

[0114] The embodiments of the present application further provide one or more machine-readable media having instructions stored thereon, which, when executed by one or more processors, enable the processors to execute the automotive two-force rod buckling force prediction method in the above embodiments.

[0115] Figure 8The schematic diagram of the structure of a computer system suitable for implementing the memory of the embodiment of the present application is shown. It should be noted that: Figure 8 The computer system of the memory shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0116] like Figure 8 As shown, the computer system includes a central processing unit (CPU), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) or the program loaded from the storage part to the random access memory (RAM), such as the method in the above embodiment. In the RAM, various programs and data required for system operation are also stored. The CPU, ROM and RAM are connected to each other through a bus. The input / output (I / O) interface is also connected to the bus.

[0117] The following components are connected to the I / O interface: an input part including a keyboard, a mouse, etc.; an output part including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage part including a hard disk, etc.; and a communication part including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication part performs communication processing via a network such as the Internet. A drive is also connected to the I / O interface as needed. Removable media, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., are installed on the drive as needed so that a computer program read therefrom is installed into the storage part as needed.

[0118] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a computer program for executing the aforementioned automotive two-force rod buckling force prediction method. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), various functions defined in the system of the present application are executed.

[0119] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0120] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0121] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0122] Another aspect of the present application provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor of a computer, the computer executes the above-mentioned automobile two-force rod buckling force prediction method. The computer-readable storage medium may be included in the memory described in the above embodiment, or may exist independently without being assembled into the memory.

[0123] Another aspect of the present application also provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the automotive two-force rod buckling force prediction method provided in each of the above embodiments.

[0124] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A method for predicting the buckling force of a two-force rod of an automobile, characterized in that: The prediction method comprises: Get the 3D mesh data of the two-force rod of the car; The 3D mesh data of the automobile two-force bar is input into a trained buckling force prediction model, and the buckling force of the two-force bar is predicted by the buckling force prediction model to obtain a prediction result of the buckling force of the two-force bar; wherein the buckling force prediction model is obtained by training an initial neural network model with the 3D mesh data of the automobile two-force bar as input and the buckling force of the two-force bar as output.

2. The method for predicting the buckling force of a two-force rod of an automobile according to claim 1, characterized in that: The buckling force prediction model includes a feature extraction layer, a global pooling layer, and a fully connected layer; the buckling force of the two-force bar is predicted by the buckling force prediction model, and the step of obtaining the prediction result of the buckling force of the two-force bar includes: Performing feature extraction on the 3D mesh data through the feature extraction layer to obtain a first mesh feature; Using the global pooling layer to perform global average pooling processing on the first grid feature to obtain a second grid feature; All the second mesh features are connected through the fully connected layer, and regression prediction of the buckling force of the two-force bar is performed based on the connected second mesh features.

3. The method for predicting the buckling force of a two-force rod of an automobile according to claim 2, characterized in that: The feature extraction layer includes a plurality of feature extraction modules connected in sequence, and the output of the previous feature extraction module serves as the input of the next feature extraction module; the feature extraction module includes: Convolution unit, used to extract features from input data; The pooling unit is used to perform pooling operations on the features output by the convolution unit to obtain new features.

4. The method for predicting the buckling force of a two-force rod of an automobile according to claim 3, characterized in that: The 3D mesh data includes: The ratio of the angle between a first plane consisting of three sides and a second plane consisting of three sides, the inner angle corresponding to the target line in the first plane, the inner angle corresponding to the target line in the second plane, the length of a first perpendicular line having a first perpendicular relationship with the target line in the first plane and the length of a second perpendicular line having a second perpendicular relationship with the target line in the second plane, wherein the target line is a common side of the first plane and the second plane.

5. The method for predicting the buckling force of a two-force rod of an automobile according to claim 4, characterized in that: When the feature extraction layer extracts features from the 3D mesh data, the feature extraction layer extracts features from the 3D mesh data through edge convolution, including: Get the features of the common edge and the features of the neighborhood of the common edge; A weighted sum is performed on the features of the common edge and the features of the neighborhood of the common edge to extract features from the input data.

6. The method for predicting the buckling force of a two-force rod of an automobile according to claim 5, characterized in that: Before performing weighted summation on the features of the common edge and the features of the neighborhood of the common edge, the method further comprises: The features of the neighborhood of the common edge are changed by a transformation relationship; the transformation relationship is expressed as: (e 1 ,e 2 ,e 3 ,e 4 )=(|a-c|,a+c,|b-d|,b+d) Among them, (e 1 ,e 2 ,e 3 ,e 4 ) represents the characteristics of the neighborhood of the common edge after the transformation, a, b, c, d represent the four neighboring edges of the common edge, edge a and edge b are located in the same plane, edge c and edge d are located in the same plane, edge c and edge b have a common point, and edge a and edge d have a common point.

7. The method for predicting the buckling force of a two-force rod of an automobile according to claim 5, characterized in that: The pooling unit performs a pooling operation on the features output by the convolution unit through a pooling relationship, and the pooling relationship is expressed as: Among them, a, b, c, and d represent the four adjacent edges of the common edge e, edge a and edge b are located in the same plane, edge c and edge d are located in the same plane, edge c and edge b have a common point, and edge a and edge d have a common point; f(e) represents the length of the common edge, f(a) represents the length of edge a, f(b) represents the length of edge b, f(c) represents the length of edge c, and f(d) represents the length of edge d.

8. A device for predicting the buckling force of a two-force rod of an automobile, characterized in that: The prediction device comprises: A data acquisition module is used to acquire 3D mesh data of the two-force rod of the automobile; A prediction module is used to input the 3D mesh data of the automobile two-force bar into a trained buckling force prediction model, predict the buckling force of the two-force bar through the buckling force prediction model, and obtain a prediction result of the buckling force of the two-force bar; wherein the buckling force prediction model is obtained by training an initial neural network model with the 3D mesh data of the automobile two-force bar as input and the buckling force of the two-force bar as output.

9. An automobile two-force rod buckling force prediction device, characterized in that: include: one or more processors; and A memory for storing one or more programs. When the one or more programs are executed by the one or more processors, the memory implements the automotive two-force rod buckling force prediction method as described in any one of claims 1-7.

10. A machine-readable medium, characterized in that Instructions are stored thereon, and when executed by one or more processors, the processors execute the automotive two-force rod buckling force prediction method as described in any one of claims 1-7.

Images