Stabilizer bar stiffness strength prediction method, device and vehicle for passenger vehicle

Abstract

The application relates to a passenger vehicle stabilizer bar rigidity strength prediction method and device and a vehicle, wherein the method comprises the following steps: determining stabilizer bar design parameters and a design space of a stabilizer bar of a passenger vehicle; based on the stabilizer bar design parameters and the design space, constructing a hollow stabilizer bar sample library and a solid stabilizer bar sample library; and respectively establishing a hollow stabilizer bar prediction model and a solid stabilizer bar prediction model by using the hollow stabilizer bar sample library and the solid stabilizer bar sample library, so as to output a rigidity strength prediction value of the stabilizer bar. Therefore, the problems that, in the related art, the applicability of an empirical formula is limited, the rigidity estimation accuracy is difficult to guarantee, and frequent iteration of simulation analysis calculation leads to high labor cost and low analysis efficiency are solved.

Description

Technical Field

[0001] This application relates to the field of vehicle component testing technology, and in particular to a method, device and vehicle for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle. Background Technology

[0002] Currently, stabilizer bars are widely used in modern vehicles and are an extremely important passive safety device. During the design phase, the stiffness and strength of stabilizer bars in passenger vehicles need to be verified, and their hard points and structure are frequently adjusted during the early chassis tuning phase to further improve vehicle performance. With the development of the automotive industry, intensified market competition, and compressed design cycles, the faster and lower-cost prediction of the stiffness and strength of passenger vehicle stabilizer bars has become an urgent need.

[0003] In related technologies, the methods for verifying the stiffness and strength of passenger car stabilizer bars mainly include the following: First, experimental verification, which requires the fabrication of actual prototypes for verification; second, estimation of stiffness using empirical formulas, which are derived from stabilizer bars with similar characteristics or structurally regular stabilizer bars from which formulas can be directly derived; and third, simulation analysis and calculation, which uses finite element simulation to obtain stiffness and strength data through steps such as drawing three-dimensional data geometry, mesh generation, attribute assignment, boundary setting, and solution calculation.

[0004] However, in related technologies, the reliance on physical prototypes for experimental verification leads to problems such as long prototype manufacturing cycles and high costs, which conflict with the need for frequent adjustments in the early stages of design. Due to factors such as design space and stiffness requirements, the stabilizer has complex geometric characteristics, which limits the applicability of empirical formulas, makes it difficult to guarantee the accuracy of stiffness estimation, and makes it impossible to evaluate strength performance. Furthermore, the frequent iterations of finite element simulation analysis result in high labor costs and low analysis efficiency, which urgently need to be improved. Summary of the Invention

[0005] This application provides a method, device, and vehicle for predicting the stiffness and strength of a stabilizer bar in passenger vehicles, in order to solve the problems in related technologies where the applicability of empirical formulas is limited, making it difficult to guarantee the accuracy of stiffness estimation, and where frequent iterations of simulation analysis calculations lead to high labor costs and low analysis efficiency.

[0006] The first aspect of this application provides a method for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle, comprising the following steps: determining the stabilizer bar design parameters and design space of the stabilizer bar in the passenger vehicle; constructing a hollow stabilizer bar sample library and a solid stabilizer bar sample library based on the stabilizer bar design parameters and the design space; and establishing a hollow stabilizer bar prediction model and a solid stabilizer bar prediction model respectively using the hollow stabilizer bar sample library and the solid stabilizer bar sample library to output the predicted stiffness and strength value of the stabilizer bar.

[0007] Through the above-mentioned technical means, the embodiments of this application can construct a hollow stabilizer sample library and a solid stabilizer sample library based on the stabilizer design parameters and design space, and establish hollow stabilizer prediction models and solid stabilizer prediction models respectively to output the stiffness and strength prediction values ​​of the stabilizer. This enables rapid prediction of stiffness and strength results directly from the geometric model, simplifies the analysis process steps, reduces complexity, and thus greatly improves efficiency and reduces labor costs while ensuring accuracy.

[0008] Optionally, in one embodiment of this application, determining the stabilizer bar design parameters and design space of a passenger vehicle includes: obtaining the ridge line parameters and cross-sectional parameters of the stabilizer bar; determining the stabilizer bar design parameters based on the ridge line parameters and the cross-sectional parameters; and determining the design space based on the stabilizer bar design parameters, historical data, ridge line trends, and a preset design range.

[0009] Through the above-mentioned technical means, the embodiments of this application can obtain ridge line parameters and cross-sectional parameters to determine the design parameters of the stabilizer bar, and combine historical data, ridge line trends and preset design ranges to determine the design space, thereby providing accurate input boundaries for the subsequent construction of the sample library, avoiding parameter values ​​from deviating from actual engineering constraints, and improving the applicability of the prediction model in the real design space.

[0010] Optionally, in one embodiment of this application, constructing the hollow stabilizer bar sample library and the solid stabilizer bar sample library includes: determining multiple ridge parameter points according to the type of stabilizer bar; sampling based on the multiple ridge parameter points to generate inner and outer radius parameter files, stiffness calculation files and / or stiffness calculation results, so as to establish the hollow stabilizer bar sample library and / or the solid stabilizer bar sample library using the inner and outer radius parameter files, the stiffness calculation files and / or the stiffness calculation results.

[0011] Through the above-mentioned technical means, the embodiments of this application can generate a sample library containing geometric parameters and corresponding stiffness calculation results by sampling ridge parameter points, thereby systematically constructing a sample library covering a variety of design variants and providing sufficient and diverse data support for subsequent model training.

[0012] Optionally, in one embodiment of this application, the step of sampling based on the plurality of ridge parameter points to generate an inner and outer radius parameter file, a stiffness calculation file, and / or a stiffness calculation result includes: sampling based on the plurality of ridge parameter points to generate a sampling file; generating the inner and outer radius parameter file based on the sampling file; applying a preset load value to the loading point of the sampling file to obtain the displacement of the loading point; and generating a stiffness calculation file and / or a stiffness calculation result based on the displacement of the loading point and a preset conversion factor.

[0013] Through the above-mentioned technical means, the embodiments of this application can derive stiffness indices through displacement prediction, thereby reducing the sample size requirement and data dimensionality, and then generating stiffness calculation results of samples in batches, which greatly improves the efficiency of sample library construction and reduces the manpower cost of simulation analysis.

[0014] Optionally, in one embodiment of this application, the step of establishing a hollow stabilizer prediction model and a solid stabilizer prediction model using the hollow stabilizer sample library and the solid stabilizer sample library respectively includes: establishing the hollow stabilizer prediction model based on a graph convolutional network and the solid stabilizer prediction model based on a graph convolutional network based on the hollow stabilizer sample library and the solid stabilizer sample library; generating a hollow stabilizer training set, a hollow stabilizer validation set, a hollow stabilizer test set, a solid stabilizer training set, a solid stabilizer validation set, and a solid stabilizer test set based on the hollow stabilizer sample library and the solid stabilizer sample library; determining the first training parameters according to the hollow stabilizer sample library, and according to the... The following steps are taken: First, a second training parameter is determined using a solid stabilizer bar sample library. Then, based on the hollow stabilizer bar training set, the hollow stabilizer bar validation set, the hollow stabilizer bar test set, and the first training parameter, the hollow stabilizer bar prediction model is trained and its parameters are tuned until the accuracy of the hollow stabilizer bar test set meets a first preset test condition. Finally, the trained hollow stabilizer bar prediction model is saved. Next, based on the solid stabilizer bar training set, the solid stabilizer bar validation set, the solid stabilizer bar test set, and the second training parameter, the solid stabilizer bar prediction model is trained and its parameters are tuned until the accuracy of the solid stabilizer bar test set meets a second preset test condition.

[0015] Through the above-mentioned technical means, the embodiments of this application can use graph convolutional networks to establish hollow stabilizer prediction models and solid stabilizer prediction models respectively, and generate independent training sets, validation sets and test sets based on the sample library. Combined with training parameters, the models are trained and optimized to ensure that the prediction accuracy meets the standards, improve the reliability and accuracy of the prediction results, thereby improving the pertinence and accuracy of the prediction model, and realizing the rapid and accurate prediction of stabilizer stiffness and strength.

[0016] A second aspect of this application provides a device for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle, comprising: a determination module for determining the stabilizer bar design parameters and design space of the stabilizer bar in the passenger vehicle; a construction module for constructing a hollow stabilizer bar sample library and a solid stabilizer bar sample library based on the stabilizer bar design parameters and the design space; and a prediction module for establishing a hollow stabilizer bar prediction model and a solid stabilizer bar prediction model using the hollow stabilizer bar sample library and the solid stabilizer bar sample library, respectively, to output the predicted stiffness and strength value of the stabilizer bar.

[0017] Through the above-mentioned technical means, the embodiments of this application can construct a hollow stabilizer sample library and a solid stabilizer sample library based on the stabilizer design parameters and design space, and establish hollow stabilizer prediction models and solid stabilizer prediction models respectively to output the stiffness and strength prediction values ​​of the stabilizer. This enables rapid prediction of stiffness and strength results directly from the geometric model, simplifies the analysis process steps, reduces complexity, and thus greatly improves efficiency and reduces labor costs while ensuring accuracy.

[0018] Optionally, in one embodiment of this application, the determining module includes: an acquisition unit, configured to acquire the ridge line parameters and cross-sectional parameters of the stabilizer bar; and a first determining unit, configured to determine the stabilizer bar design parameters based on the ridge line parameters and the cross-sectional parameters, and to determine the design space based on the stabilizer bar design parameters, historical data, ridge line trends, and a preset design range.

[0019] Through the above-mentioned technical means, the embodiments of this application can obtain ridge line parameters and cross-sectional parameters to determine the design parameters of the stabilizer bar, and combine historical data, ridge line trends and preset design ranges to determine the design space, thereby providing accurate input boundaries for the subsequent construction of the sample library, avoiding parameter values ​​from deviating from actual engineering constraints, and improving the applicability of the prediction model in the real design space.

[0020] Optionally, in one embodiment of this application, the construction module includes: a second determining unit, configured to determine multiple ridge parameter points according to the type of the stabilizer bar; and a first establishing unit, configured to sample based on the multiple ridge parameter points to generate inner and outer radius parameter files, stiffness calculation files and / or stiffness calculation results, so as to establish the hollow stabilizer bar sample library and / or the solid stabilizer bar sample library using the inner and outer radius parameter files, the stiffness calculation files and / or the stiffness calculation results.

[0021] Through the above-mentioned technical means, the embodiments of this application can generate a sample library containing geometric parameters and corresponding stiffness calculation results by sampling ridge parameter points, thereby systematically constructing a sample library covering a variety of design variants and providing sufficient and diverse data support for subsequent model training.

[0022] Optionally, in one embodiment of this application, the first establishing unit includes: a first generation subunit, used to sample based on the plurality of ridge parameter points to generate a sampling file; a second generation subunit, used to generate the inner and outer radius parameter file based on the sampling file; an application subunit, used to apply a preset load value to the loading point of the sampling file to obtain the displacement of the loading point; and a third generation subunit, used to generate a stiffness calculation file and / or stiffness calculation results based on the displacement of the loading point and a preset conversion coefficient.

[0023] Through the above-mentioned technical means, the embodiments of this application can derive stiffness indices through displacement prediction, thereby reducing the sample size requirement and data dimensionality, and then generating stiffness calculation results of samples in batches, which greatly improves the efficiency of sample library construction and reduces the manpower cost of simulation analysis.

[0024] Optionally, in one embodiment of this application, the prediction module includes: a second establishment unit, configured to establish a hollow stabilizer prediction model based on a graph convolutional network and a solid stabilizer prediction model based on a graph convolutional network, based on the hollow stabilizer sample library and the solid stabilizer sample library; a generation unit, configured to generate a hollow stabilizer training set, a hollow stabilizer validation set, a hollow stabilizer test set, a solid stabilizer training set, a solid stabilizer validation set, and a solid stabilizer test set, based on the hollow stabilizer sample library and the solid stabilizer sample library; and a third determination unit, configured to determine a first training parameter based on the hollow stabilizer sample library and a second training parameter based on the solid stabilizer sample library. The first training unit is used to train and fine-tune the hollow stabilizer prediction model based on the hollow stabilizer training set, the hollow stabilizer validation set, the hollow stabilizer test set, and the first training parameters, until the accuracy of the hollow stabilizer test set meets the first preset test condition, and then saves the trained hollow stabilizer prediction model; the second training unit is used to train and fine-tune the solid stabilizer prediction model based on the solid stabilizer training set, the solid stabilizer validation set, the solid stabilizer test set, and the second training parameters, until the accuracy of the solid stabilizer test set meets the second preset test condition, and then saves the trained solid stabilizer prediction model.

[0025] Through the above-mentioned technical means, the embodiments of this application can use graph convolutional networks to establish hollow stabilizer prediction models and solid stabilizer prediction models respectively, and generate independent training sets, validation sets and test sets based on the sample library. Combined with training parameters, the models are trained and optimized to ensure that the prediction accuracy meets the standards, improve the reliability and accuracy of the prediction results, thereby improving the pertinence and accuracy of the prediction model, and realizing the rapid and accurate prediction of stabilizer stiffness and strength.

[0026] A third aspect of this application provides a vehicle, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for predicting the stiffness and strength of a stabilizer bar for a passenger vehicle as described in the above embodiments.

[0027] A fourth aspect of this application provides a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle.

[0028] A fifth aspect of this application provides a computer program product that stores a computer program that, when executed by a processor, implements the above-described method for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle.

[0029] This application's embodiments can construct hollow and solid stabilizer bar sample libraries based on stabilizer bar design parameters and design space, and establish hollow and solid stabilizer bar prediction models respectively to output predicted values ​​of stabilizer bar stiffness and strength. This enables rapid prediction of stiffness and strength results directly from the geometric model, simplifying the analysis process, reducing complexity, and thus greatly improving efficiency and reducing labor costs while ensuring accuracy. Therefore, it solves the problems in related technologies where the limited applicability of empirical formulas makes it difficult to guarantee the accuracy of stiffness estimation, and frequent iterations in simulation analysis lead to high labor costs and low analysis efficiency.

[0030] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0031] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a method for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle according to an embodiment of this application; Figure 2 This is a schematic diagram of a ridgeline and cross-section according to an embodiment of this application; Figure 3 This is a schematic diagram illustrating the selection of example stabilizer bar parameters according to one embodiment of this application; Figure 4 This is a reverse-view schematic diagram according to an embodiment of this application; Figure 5 This is a schematic diagram illustrating the differences between hollow and solid stabilizer bars according to an embodiment of this application; Figure 6 This is a schematic diagram of an automatic generation process for stabilizer bar samples according to an embodiment of this application; Figure 7 This is a schematic diagram of the training process loss curve according to an embodiment of this application; Figure 8 This is a schematic diagram illustrating the stress and displacement (stiffness) prediction of a hollow stabilizer bar A for a passenger vehicle according to an embodiment of this application. Figure 9This is a schematic diagram illustrating the stress and displacement (stiffness) prediction of a solid stabilizer bar C for a passenger vehicle according to an embodiment of this application. Figure 10 This is a flowchart of a method for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle according to an embodiment of this application; Figure 11 This is a schematic diagram of the structure of a stabilizer bar stiffness and strength prediction device for a passenger vehicle according to an embodiment of this application; Figure 12 This is a structural schematic diagram of a vehicle provided according to an embodiment of this application.

[0032] Figure label: 10-Stiffness and strength prediction device for stabilizer bar of passenger vehicle; 100-Determination module, 200-Construction module, 300-Prediction module; 1201-Memory, 1202-Processor, 1203-Communication interface. Detailed Implementation

[0033] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0034] The following description, with reference to the accompanying drawings, outlines a method, apparatus, and vehicle for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle, according to embodiments of this application. Addressing the issues raised in the background section regarding the limited applicability of empirical formulas, which makes it difficult to guarantee the accuracy of stiffness estimation, and the high labor costs and low analysis efficiency resulting from frequent iterations in simulation analysis, this application provides a method for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle. This method constructs a hollow stabilizer bar sample library and a solid stabilizer bar sample library based on the stabilizer bar design parameters and design space, and establishes prediction models for hollow and solid stabilizer bars respectively to output predicted values ​​of the stabilizer bar's stiffness and strength. This achieves rapid prediction directly from the geometric model to the stiffness and strength results, simplifying the analysis process, reducing complexity, and significantly improving efficiency and reducing labor costs while ensuring accuracy. Therefore, this solves the problems in the related technologies where the limited applicability of empirical formulas makes it difficult to guarantee the accuracy of stiffness estimation, and the high labor costs and low analysis efficiency resulting from frequent iterations in simulation analysis.

[0035] Specifically, Figure 1 This is a flowchart illustrating a method for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle, as provided in an embodiment of this application.

[0036] like Figure 1As shown, the method for predicting the stiffness and strength of the stabilizer bar in this passenger vehicle includes the following steps: In step S101, the stabilizer bar design parameters and design space of the passenger vehicle are determined.

[0037] It is understood that the stabilizer bar design parameters in the embodiments of this application may include, but are not limited to, ridge line parameters and cross-sectional parameters. The design space can be understood as the reasonable range of values ​​for the stabilizer bar design parameters, which defines the adjustable boundary of the stabilizer bar design.

[0038] In actual implementation, the embodiments of this application can determine the design parameters and design space of the stabilizer bar. Historical data of the stabilizer bar shows that most of them are symmetrical results. In order to simplify modeling and improve calculation efficiency, the embodiments of this application can select symmetrical structures for analysis.

[0039] For example, in the embodiments of this application, the design parameters of the stabilizer bar can be extracted and determined by combining the requirements of passenger vehicle model, chassis configuration, etc. At the same time, by combining the design goals of the stabilizer bar, industry design specifications, vehicle driving condition requirements and historical data, a reasonable range of values ​​for the design parameters can be defined to form a design space.

[0040] The embodiments of this application can determine the stabilizer bar design parameters and design space of passenger vehicles, thereby laying the foundation for the automated generation of subsequent sample libraries, while providing a clear adjustable range for frequent adjustments in the early stages of design, reducing invalid design iterations, and laying the foundation for rapid stiffness and strength prediction.

[0041] In step S102, based on the stabilizer bar design parameters and design space, a hollow stabilizer bar sample library and a solid stabilizer bar sample library are constructed.

[0042] It is understood that the sample library in this application embodiment can be understood as a collection that stores the design parameters and stiffness data of the stabilizer bar, which can be used to provide data support for the training of subsequent prediction models.

[0043] In actual implementation, the embodiments of this application can filter out the parameter differences between hollow and solid stabilizer bars based on stabilizer bar design parameters and design space. Then, through parameter combination, sampling, mirroring and other methods, a large number of sample data that meet the design space requirements are generated. Each sample contains complete design parameters and corresponding stiffness and strength related data. Finally, they are classified and organized to form a hollow stabilizer bar sample library and a solid stabilizer bar sample library.

[0044] The embodiments of this application can classify and construct hollow and solid stabilizer bar sample libraries based on design parameters and design space, providing comprehensive training data support for subsequent targeted prediction model building. At the same time, the construction of the sample library fits the design space requirements and can accurately match the adjustment needs in the early stage of design, providing data guarantee for the rapid prediction of stabilizer bar stiffness and strength.

[0045] In step S103, hollow stabilizer bar prediction models and solid stabilizer bar prediction models are established using hollow stabilizer bar sample libraries and solid stabilizer bar sample libraries, respectively, to output the predicted values ​​of the stabilizer bar stiffness and strength.

[0046] It is understood that the stiffness and strength prediction values ​​in the embodiments of this application can be understood as the numerical results of the stiffness and strength of the stabilizer bar calculated by the prediction model.

[0047] In actual implementation, the embodiments of this application can use machine learning, deep learning and other technologies to train, verify and optimize the initial model using sample data from hollow stabilizer bar sample libraries and solid stabilizer bar sample libraries, so that the model can accurately capture the correlation between stabilizer bar design parameters and stiffness and strength; after training, the stabilizer bar design parameters to be predicted are input and the model can quickly output the corresponding stiffness and strength prediction values.

[0048] For example, embodiments of this application may employ model architectures such as graph convolutional networks and neural networks, taking design parameters from a hollow stabilizer bar sample library as input and corresponding stiffness and strength data as output. The model parameters are optimized through multiple rounds of training until the model prediction accuracy reaches a preset standard. When it is necessary to predict the stiffness and strength of a hollow stabilizer bar, the design parameters such as the diameter, wall thickness, and inner hole size of the stabilizer bar are input, and the model can output the corresponding stiffness and strength values ​​within a few seconds without the need to conduct complex experiments or simulation calculations.

[0049] The embodiments of this application can establish hollow stabilizer prediction models and solid stabilizer prediction models based on stabilizer sample libraries, and quickly output the predicted values ​​of stabilizer stiffness and strength. This effectively solves the problems of low efficiency, high cost and long cycle of related technologies. At the same time, the classification model adapts to the needs of stabilizer structures, improves prediction accuracy, and provides a fast and reliable reference for stabilizer design and adjustment.

[0050] Optionally, in one embodiment of this application, determining the stabilizer bar design parameters and design space for a passenger vehicle includes: obtaining the ridge line parameters and cross-sectional parameters of the stabilizer bar; determining the stabilizer bar design parameters based on the ridge line parameters and cross-sectional parameters; and determining the design space based on the stabilizer bar design parameters, historical data, ridge line trends, and a preset design range.

[0051] It is understood that the ridge line parameters in this embodiment may include the loading endpoint, left fixed point, zero point, right fixed point, and symmetry point of the stabilizer bar. For ease of parameter modeling, the coordinates of the stabilizer bar and subframe fixed points are set to zero, and other ridge line points are used as references. Cross-sectional parameters may include the inner and outer radii of the stabilizer bar. The preset design range can be understood as the parameter value boundaries set according to manufacturing process or platform requirements. The preset design range can be set by those skilled in the art based on actual conditions, and no specific limitations are imposed here.

[0052] In actual implementation, the embodiments of this application can select a symmetrical structure for analysis, and select ridge line parameters (to facilitate parameter modeling, the coordinates of the stabilizer bar and subframe fixing points are set to zero, and other ridge line points are based on this) and cross-sectional parameters (inner radius, outer radius), as illustrated in the figure below. Figure 2 As shown in Table 1. Table 1 is a schematic table of design parameters.

[0053] Table 1

[0054] The embodiments of this application can obtain ridge line parameters and cross-sectional parameters to determine the design parameters of the stabilizer bar, and combine historical data, ridge line trends and preset design ranges to determine the design space, thereby providing accurate input boundaries for the subsequent construction of the sample library, avoiding parameter values ​​from deviating from actual engineering constraints, and improving the applicability of the prediction model in the real design space.

[0055] Optionally, in one embodiment of this application, constructing a hollow stabilizer bar sample library and a solid stabilizer bar sample library includes: determining multiple ridge parameter points according to the type of stabilizer bar; sampling based on the multiple ridge parameter points to generate inner and outer radius parameter files, stiffness calculation files and / or stiffness calculation results, so as to establish a hollow stabilizer bar sample library and / or a solid stabilizer bar sample library using the inner and outer radius parameter files, stiffness calculation files and / or stiffness calculation results.

[0056] It is understood that the types of stabilizers in this application embodiment may include hollow stabilizers and solid stabilizers. The inner and outer radius parameter file is a file storing the outer and inner radius parameters of the hollow stabilizer, such as a JSON file. The stiffness and strength calculation file can be understood as a file containing all the model information required for finite element analysis, such as a FEM file, which defines nodal coordinates, element connection relationships, material properties (elastic modulus, Poisson's ratio), boundary constraints, and the location and magnitude of applied loads, and can be directly submitted to the solver for numerical calculation; the stiffness and strength calculation results can be understood as a file storing the physical field data obtained after the finite element solution, such as an H3D file, containing calculation results such as nodal displacements, element stresses, stiffness values, and strength values.

[0057] For example, in this embodiment, the stabilizing bar ridge line can first be mirrored to a single quadrant, then classified according to the ridge line trend, and finally the Morph method is used to expand the sample. Taking a certain type of ridge line trend as an example, 7 ridge line parameter points are selected, i.e., 21 ridge line parameters + 2 cross-sectional parameters, for a total of 23 parameters, such as... Figure 3 As shown. Simultaneously, to expand the design space and facilitate design use, the sample is reverse-mirrorized (the process is as follows). Figure 4 (As shown) A total of 1600 samples. The above is the sample library preparation for hollow stabilizer bars. Given the difference in inner and outer radius parameters between hollow and solid stabilizer bars (the inner radius of solid stabilizer bars can be uniformly set to zero; the difference is as follows...), Figure 5 As shown in the figure, with other parameters remaining unchanged, 1600 samples were established similarly, and the solid stabilizer and the hollow stabilizer were predicted separately in the future.

[0058] This application embodiment can generate a sample library containing geometric parameters and corresponding stiffness calculation results by sampling ridge parameter points, thereby systematically constructing a sample library covering a variety of design variants and providing sufficient and diverse data support for subsequent model training.

[0059] Optionally, in one embodiment of this application, sampling is performed based on multiple ridge parameter points to generate an inner and outer radius parameter file, a stiffness calculation file, and / or a stiffness calculation result, including: sampling based on multiple ridge parameter points to generate a sampling file; generating an inner and outer radius parameter file based on the sampling file; applying a preset load value to the loading point in the sampling file to obtain the displacement of the loading point; and generating a stiffness calculation file and / or a stiffness calculation result based on the displacement of the loading point and a preset conversion factor.

[0060] It is understood that the preset load value in this embodiment can be 1000N. The preset load value can be set by those skilled in the art according to the actual situation, and no specific limitation is made here. The preset conversion factor can be understood as a scaling factor that converts the displacement of the loading point into stiffness. It can be obtained through historical data and comparative tests. The preset conversion factor can be set by those skilled in the art according to the actual situation, and no specific limitation is made here.

[0061] Specifically, in this application embodiment, a conversion factor can be obtained through historical data and comparative experiments. After loading a fixed load value, the stress level under different loads can be calculated by a proportional coefficient instead of directly increasing the load value parameter.

[0062] Due to the large sample size in actual implementation, this application embodiment can establish an automated process to automatically generate JSON files (inner and outer radius parameters), FEM files (stiffness calculation files), and H3D files (stiffness calculation results) to save manpower and improve efficiency. A schematic diagram is shown below. Figure 6 As shown. Among them, as... Figure 6 The working condition shown is that a force of 1000N is applied to the Z-direction of the vehicle at the loading end of the stabilizer bar. That is, the stiffness index is the state under a fixed load, which is the force divided by the displacement of the loading point (later predicted response 1) and the stress (later predicted response 2). According to historical experience, the error of the maximum stress value under different loads calculated by the preset conversion factor (derived from historical data and benchmark tests) is acceptable in the early tuning and iteration stage.

[0063] For example, in this embodiment of the application, a force of 1000N is applied at the loading point. The preset conversion factor can be 1.0, meaning the stiffness is directly obtained by dividing 1000N by the displacement at the loading point. If it is necessary to convert the stress level of 1000N to other loads (such as 1200N), the scaling factor is 0.001N. -1 Therefore, the stress at 1200N = the stress at 1000N × 1200N * 0.001N -1 That is, the preset conversion factor can be 1.2 at this time.

[0064] The embodiments of this application can derive stiffness indices through displacement prediction, thereby reducing the sample size requirement and data dimensionality, and then generating stiffness and strength calculation results of samples in batches, which greatly improves the efficiency of sample library construction and reduces the manpower cost of simulation analysis.

[0065] Optionally, in one embodiment of this application, a hollow stabilizer prediction model and a solid stabilizer prediction model are established using a hollow stabilizer sample library and a solid stabilizer sample library, respectively, including: establishing a hollow stabilizer prediction model based on a graph convolutional network and a solid stabilizer prediction model based on a graph convolutional network based on the hollow stabilizer sample library and the solid stabilizer sample library; generating a hollow stabilizer training set, a hollow stabilizer validation set, a hollow stabilizer test set, a solid stabilizer training set, a solid stabilizer validation set, and a solid stabilizer test set based on the hollow stabilizer sample library and the solid stabilizer sample library; and determining a first training parameter according to the hollow stabilizer sample library. The second training parameters are determined based on the solid stabilizer bar sample library. The hollow stabilizer bar prediction model is trained and its parameters are tuned based on the hollow stabilizer bar training set, hollow stabilizer bar validation set, hollow stabilizer bar test set, and the first training parameters until the accuracy of the hollow stabilizer bar test set meets the first preset test condition. The trained hollow stabilizer bar prediction model is then saved. Similarly, the solid stabilizer bar prediction model is trained and its parameters are tuned based on the solid stabilizer bar training set, solid stabilizer bar validation set, solid stabilizer bar test set, and the second training parameters until the accuracy of the solid stabilizer bar test set meets the second preset test condition. The trained solid stabilizer bar prediction model is then saved.

[0066] It is understood that the graph convolutional network in this embodiment can be a neural network capable of performing convolution operations directly on a graph. The first preset test condition can be an accuracy greater than 90%, and this first preset test condition can be set by those skilled in the art according to actual conditions, without specific limitations here. The second preset test condition can also be an accuracy greater than 90%, and this second preset test condition can be set by those skilled in the art according to actual conditions, without specific limitations here.

[0067] For example, embodiments of this application can train a prediction model for the stiffness and strength response of a stabilizer bar using a graph convolutional network of the Physical-ai module in Altair software, based on a large number of finite element analysis results of the stabilizer bar stiffness and strength.

[0068] In actual implementation, the embodiments of this application can build a hollow stabilizer prediction model. First, the sample data is divided into a training set: validation set: test set = 68%: 12%: 20%. Based on the hollow stabilizer sample library, the first training parameters are determined: Epoch = 1500, Batch size = 1, Width * Depth = 30 * 1, Learning rate = 0.001, and Early stopping = yes.

[0069] The loss curve during training is as follows Figure 7 As shown, the blue dashed line represents the training loss (decreasing with each epoch, reflecting the model's performance on the training data); the red solid line represents the validation loss (used to evaluate the model's generalization ability on unseen data); and the black asterisk marks the final selected model position. Both the training loss and validation loss decreased to low values ​​(training loss 0.29, validation loss 0.72) and gradually approached each other, indicating that the model learned the patterns from the training data and could generalize well to new data (without overfitting or underfitting). The test set stiffness accuracy reached 97.8%, and the stress prediction accuracy reached 97.9%. To ensure model usability, model tests were conducted using passenger car stabilizer bars. The test results for hollow stabilizer bars A and B are shown in Table 2, with accuracy rates exceeding 90%. A schematic diagram of the prediction results for stabilizer bar A is shown below. Figure 8 As shown in Table 2, the stiffness and stress test results of hollow stabilizer bars A and B for passenger vehicles are presented.

[0070] Table 2

[0071] Furthermore, a prediction model for solid stabilizers was constructed, dividing the sample data into a training set: validation set: test set = 68%: 12%: 20%. Based on the solid stabilizer sample library, a second training parameter was determined: Epoch = 1500, Batch size = 1, Width * Depth = 30 * 1, Learning rate = 0.001, and Early stopping = yes. Based on the solid stabilizer training set, validation set, test set, and the second training parameter, the solid stabilizer prediction model was trained and its parameters were tuned until the accuracy of the solid stabilizer test set met the second preset testing condition. The trained solid stabilizer prediction model was then saved. The average accuracy of the test set also reached over 90%. A solid stabilizer C was tested, and the results are shown in Table 3. Figure 9 As shown in Table 3, the stiffness and stress test results of the solid stabilizer bar C for passenger vehicles are presented.

[0072] Table 3

[0073] Therefore, the rapid prediction method for the stiffness and strength of passenger vehicle stabilizer bars proposed in this application has been tested and proven to be accurate, while also significantly improving efficiency and saving labor costs. Furthermore, it is extremely user-friendly, requiring only the dragging and dropping of geometric or mesh models into the interface, avoiding cumbersome steps and lowering the barrier to entry (no complex simulation analysis knowledge is required), thus enhancing the forward development capabilities of passenger vehicle stabilizer bars. Of course, leveraging its speed advantage, it can be further embedded into optimization modules to improve the reverse optimization capabilities of passenger vehicle stabilizer bars.

[0074] In this embodiment, graph convolutional networks can be used to establish hollow stabilizer prediction models and solid stabilizer prediction models respectively. Independent training sets, validation sets and test sets are generated based on the sample library. The training parameters are combined to train and optimize the models separately to ensure that the prediction accuracy of the models meets the standards, improve the reliability and accuracy of the prediction results, thereby improving the pertinence and accuracy of the prediction models and realizing the rapid and accurate prediction of stabilizer stiffness and strength.

[0075] Specifically, it can be combined with Figure 10 As shown, the working principle of the stabilizer bar stiffness and strength prediction method for passenger vehicles in this application is explained in detail with a specific embodiment.

[0076] like Figure 10As shown, the embodiments of this application can determine the stabilizer bar parameters and design space. Specifically, based on the skeleton parameters, cross-sectional parameters, historical data, and skeleton trends, the design range is defined, thereby completing the setting of the stabilizer bar parameters and design space. On this basis, parameter points are determined and samples are generated using LHS (Latin hypercube sampling).

[0077] Subsequently, the automated process generates JSON files (inner and outer radius parameters) based on the sampling results, batch generates mesh models using Morph technology, and then automatically generates corresponding FEM files (stiffness calculation files) using preprocessing software. After solving the calculations, H3D result files (stiffness calculation results) are obtained, thus completing the construction of the hollow stabilizer bar sample library. Similarly, the solid stabilizer bar sample library is constructed following the same process.

[0078] Furthermore, the samples are divided into training, validation, and test sets. Training parameters are set, and model training, loss evaluation, and parameter tuning are performed. Once the accuracy of the prediction model on the test set reaches a preset standard, such as 90% accuracy, the model is saved. In the practical application stage, the user drags the geometric model or mesh model into the prediction main interface, selects displacement prediction or stress prediction, and uses preset conversion coefficients to calculate the stiffness and stress indicators of the stabilizer bar through relevant functions. The prediction results are then used to determine whether the design requirements are met.

[0079] The method for predicting the stiffness and strength of stabilizer bars in passenger vehicles proposed in this application can construct hollow stabilizer bar sample libraries and solid stabilizer bar sample libraries based on stabilizer bar design parameters and design space. Prediction models for hollow and solid stabilizer bars are then established respectively to output predicted values ​​of stabilizer bar stiffness and strength. This achieves rapid prediction of stiffness and strength results directly from the geometric model, simplifying the analysis process, reducing complexity, and significantly improving efficiency and reducing labor costs while ensuring accuracy. Therefore, it solves the problems in related technologies where the limited applicability of empirical formulas makes it difficult to guarantee the accuracy of stiffness estimation, and frequent iterations in simulation analysis lead to high labor costs and low analysis efficiency.

[0080] Next, referring to the accompanying drawings, a stabilizer bar stiffness and strength prediction device for passenger vehicles according to an embodiment of this application is described.

[0081] Figure 11 This is a schematic diagram of the structure of the stabilizer bar stiffness and strength prediction device for passenger vehicles according to an embodiment of this application.

[0082] like Figure 11 As shown, the stabilizer bar stiffness and strength prediction device 10 for passenger vehicles includes: a determination module 100, a construction module 200, and a prediction module 300.

[0083] The determination module 100 is used to determine the design parameters and design space of the stabilizer bar of the passenger vehicle.

[0084] Module 200 is used to build a hollow stabilizer bar sample library and a solid stabilizer bar sample library based on stabilizer bar design parameters and design space.

[0085] The prediction module 300 is used to establish hollow stabilizer prediction models and solid stabilizer prediction models using hollow stabilizer sample libraries and solid stabilizer sample libraries respectively, so as to output the predicted values ​​of the stiffness and strength of the stabilizer.

[0086] Optionally, in one embodiment of this application, the determining module 100 includes: an acquisition unit and a first determining unit.

[0087] The acquisition unit is used to acquire the ridge line parameters and cross-sectional parameters of the stabilizer bar.

[0088] The first determining unit is used to determine the stabilizer bar design parameters based on the ridge line parameters and cross-sectional parameters, and to determine the design space based on the stabilizer bar design parameters, historical data, ridge line trends, and preset design range.

[0089] Optionally, in one embodiment of this application, the construction module 200 includes a second determining unit and a first establishing unit.

[0090] The second determining unit is used to determine multiple ridge parameter points according to the type of stabilizer bar.

[0091] The first establishment unit is used to sample based on multiple ridge parameter points to generate inner and outer radius parameter files, stiffness calculation files and / or stiffness calculation results, so as to establish a hollow stabilizer sample library and / or a solid stabilizer sample library using the inner and outer radius parameter files, stiffness calculation files and / or stiffness calculation results.

[0092] Optionally, in one embodiment of this application, the first creation unit includes: a first generation subunit, a second generation subunit, an application subunit, and a third generation subunit.

[0093] The first generation subunit is used to sample based on multiple ridge parameter points to generate a sampling file.

[0094] The second generation sub-unit is used to generate inner and outer radius parameter files based on the sampling file.

[0095] The application sub-unit is used to apply a preset load value to the loading point of the sample file to obtain the displacement of the loading point.

[0096] The third generation sub-unit is used to generate stiffness calculation files and / or stiffness calculation results based on the displacement of the loading point and preset conversion coefficients.

[0097] Optionally, in one embodiment of this application, the prediction module 300 includes: a second establishment unit, a generation unit, a third determination unit, a first training unit, and a second training unit.

[0098] The second establishment unit is used to establish a hollow stabilizer prediction model based on a graph convolutional network and a solid stabilizer prediction model based on a hollow stabilizer sample library and a solid stabilizer sample library.

[0099] The generation unit is used to generate hollow stabilizer training sets, hollow stabilizer validation sets, hollow stabilizer test sets, solid stabilizer training sets, solid stabilizer validation sets, and solid stabilizer test sets based on hollow stabilizer sample libraries and solid stabilizer sample libraries.

[0100] The third determining unit is used to determine the first training parameters based on the hollow stabilizer bar sample library and the second training parameters based on the solid stabilizer bar sample library.

[0101] The first training unit is used to train and fine-tune the hollow stabilizer prediction model based on the hollow stabilizer training set, the hollow stabilizer validation set, the hollow stabilizer test set, and the first training parameters, until the accuracy of the hollow stabilizer test set meets the first preset test conditions, and then saves the trained hollow stabilizer prediction model.

[0102] The second training unit is used to train and fine-tune the solid stabilizer prediction model based on the solid stabilizer training set, solid stabilizer validation set, solid stabilizer test set, and second training parameters until the accuracy of the solid stabilizer test set meets the second preset test conditions, and then saves the trained solid stabilizer prediction model.

[0103] It should be noted that the explanation of the aforementioned embodiment of the method for predicting the stiffness and strength of the stabilizer bar of a passenger vehicle also applies to the stabilizer bar stiffness and strength prediction device of the passenger vehicle in this embodiment, and will not be repeated here.

[0104] The stabilizer bar stiffness and strength prediction device for passenger vehicles proposed in this application can construct hollow stabilizer bar sample libraries and solid stabilizer bar sample libraries based on stabilizer bar design parameters and design space, and establish hollow stabilizer bar prediction models and solid stabilizer bar prediction models respectively to output the predicted values ​​of stabilizer bar stiffness and strength. This achieves rapid prediction of stiffness and strength results directly from the geometric model, simplifies the analysis process, reduces complexity, and greatly improves efficiency and reduces labor costs while ensuring accuracy. Therefore, it solves the problems in related technologies where the applicability of empirical formulas is limited, making it difficult to guarantee the accuracy of stiffness estimation, and the high labor costs and low analysis efficiency caused by frequent iterations of simulation analysis calculations.

[0105] Figure 12 A schematic diagram of the structure of a vehicle provided in an embodiment of this application. The vehicle may include: The memory 1201, the processor 1202, and the computer program stored on the memory 1201 and executable on the processor 1202.

[0106] When the processor 1202 executes the program, it implements the method for predicting the stiffness and strength of the stabilizer bar of a passenger vehicle provided in the above embodiments.

[0107] Furthermore, the vehicle also includes: Communication interface 1203 is used for communication between memory 1201 and processor 1202.

[0108] The memory 1201 is used to store computer programs that can run on the processor 1202.

[0109] The memory 1201 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage.

[0110] If the memory 1201, processor 1202, and communication interface 1203 are implemented independently, then the communication interface 1203, memory 1201, and processor 1202 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 12 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0111] Optionally, in a specific implementation, if the memory 1201, processor 1202, and communication interface 1203 are integrated on a single chip, then the memory 1201, processor 1202, and communication interface 1203 can communicate with each other through an internal interface.

[0112] The processor 1202 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0113] This application also provides a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle.

[0114] This application also provides a computer program product storing a computer program that, when executed by a processor, implements the above-described method for predicting the stiffness and strength of a stabilizer bar for passenger vehicles.

[0115] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0116] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0117] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0118] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0119] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0120] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0121] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0122] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle, characterized in that, Includes the following steps: Determine the stabilizer bar design parameters and design space for passenger vehicles; Based on the stabilizer bar design parameters and the design space, a hollow stabilizer bar sample library and a solid stabilizer bar sample library are constructed. Using the hollow stabilizer bar sample library and the solid stabilizer bar sample library, respectively, a hollow stabilizer bar prediction model and a solid stabilizer bar prediction model are established to output the predicted values ​​of the stabilizer bar's stiffness and strength.

2. The method according to claim 1, characterized in that, The determination of the stabilizer bar design parameters and design space for passenger vehicles includes: Obtain the ridge line parameters and cross-sectional parameters of the stabilizer bar; The stabilizer bar design parameters are determined based on the ridge line parameters and the cross-sectional parameters, and the design space is determined based on the stabilizer bar design parameters, historical data, ridge line trends, and preset design range.

3. The method according to claim 1, characterized in that, The construction of the hollow stabilizer sample library and the solid stabilizer sample library includes: Multiple ridge parameter points are determined based on the type of stabilizer bar; Sampling is performed based on the multiple ridge parameter points to generate inner and outer radius parameter files, stiffness calculation files and / or stiffness calculation results, so as to establish the hollow stabilizer sample library and / or the solid stabilizer sample library using the inner and outer radius parameter files, the stiffness calculation files and / or the stiffness calculation results.

4. The method according to claim 3, characterized in that, The sampling based on the multiple ridge parameter points to generate inner and outer radius parameter files, stiffness calculation files, and / or stiffness calculation results includes: Sampling is performed based on the multiple ridge parameter points to generate a sampling file; The inner and outer radius parameter file is generated based on the sampling file; A preset load value is applied to the loading point of the sampled file to obtain the displacement of the loading point; Based on the displacement of the loading point and the preset conversion coefficient, generate a stiffness strength calculation file and / or stiffness strength calculation results.

5. The method according to claim 1, characterized in that, The step of establishing a hollow stabilizer prediction model and a solid stabilizer prediction model using the hollow stabilizer sample library and the solid stabilizer sample library, respectively, includes: Based on the hollow stabilizer sample library and the solid stabilizer sample library, a hollow stabilizer prediction model based on a graph convolutional network and a solid stabilizer prediction model based on a graph convolutional network are established. Based on the hollow stabilizer bar sample library and the solid stabilizer bar sample library, a hollow stabilizer bar training set, a hollow stabilizer bar verification set, a hollow stabilizer bar test set, a solid stabilizer bar training set, a solid stabilizer bar verification set, and a solid stabilizer bar test set are generated. The first training parameters are determined based on the hollow stabilizer bar sample library, and the second training parameters are determined based on the solid stabilizer bar sample library. Based on the hollow stabilizer training set, the hollow stabilizer validation set, the hollow stabilizer test set, and the first training parameters, the hollow stabilizer prediction model is trained and its parameters are tuned until the accuracy of the hollow stabilizer test set meets the first preset test conditions, and the trained hollow stabilizer prediction model is saved. Based on the solid stabilizer training set, the solid stabilizer validation set, the solid stabilizer test set, and the second training parameters, the solid stabilizer prediction model is trained and its parameters are tuned until the accuracy of the solid stabilizer test set meets the second preset test conditions, and the trained solid stabilizer prediction model is saved.

6. A device for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle, characterized in that, include: The determination module is used to determine the stabilizer bar design parameters and design space for passenger vehicles; The construction module is used to construct a hollow stabilizer bar sample library and a solid stabilizer bar sample library based on the stabilizer bar design parameters and the design space. The prediction module is used to establish a hollow stabilizer prediction model and a solid stabilizer prediction model using the hollow stabilizer sample library and the solid stabilizer sample library, respectively, so as to output the predicted value of the stiffness and strength of the stabilizer.

7. The apparatus according to claim 6, characterized in that, The determining module includes: The acquisition unit is used to acquire the ridge line parameters and cross-sectional parameters of the stabilizer bar; The first determining unit is used to determine the stabilizer bar design parameters based on the ridge line parameters and the cross-sectional parameters, and to determine the design space based on the stabilizer bar design parameters, historical data, ridge line trends, and a preset design range.

8. A vehicle, characterized in that, include: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, the processor executing the program to implement the method for predicting the stiffness and strength of a stabilizer bar for a passenger vehicle as described in any one of claims 1-5.

9. A non-volatile computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the method for predicting the stiffness and strength of the stabilizer bar of a passenger vehicle as described in any one of claims 1-5.

10. A computer program product, comprising a computer program, characterized in that, The computer program is executed to implement the method for predicting the stiffness and strength of a stabilizer bar in a passenger vehicle as described in any one of claims 1-5.

Images