Automobile air wind resistance friction simulation system based on CFD
Through the CFD simulation system combined with the deep learning optimization module, the high cost and time consumption problems of traditional wind tunnel experiments are solved, efficient and flexible automobile aerodynamic optimization is achieved, and the wind resistance friction simulation efficiency is improved.
Patent Information
- Application Number
- CN202510150545.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-11
- Publication Date
- 2025-07-11
AI Technical Summary
Traditional wind tunnel experiments are costly, time-consuming and limited in automotive aerodynamic testing, making it difficult to fully simulate complex airflow in real driving environments.
The CFD-based automotive air resistance friction simulation system is adopted, including data acquisition, processing, simulation, real-time monitoring visualization and deep learning optimization modules, and the resistance coefficient and friction index are optimized through unstructured grid processing and deep learning models.
It improves the wind resistance friction simulation efficiency, reduces costs, shortens simulation time, enhances design flexibility, and can meet different design needs in a short time.
Smart Images
Figure CN120296860A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of CFD simulation, and specifically to an automotive air drag and friction simulation system based on CFD. Background Art
[0002] The aerodynamic performance of an automobile has an important impact on the vehicle's fuel economy, stability, and handling. Especially with the increasing environmental protection and energy crisis, reducing the aerodynamic drag coefficient of automobiles has become one of the key technical means for automobile manufacturers to improve fuel efficiency and reduce carbon emissions. During the automobile design process, optimizing the automobile's exterior design through wind tunnel experiments and Computational Fluid Dynamics (CFD) simulations to reduce air resistance is a commonly used method.
[0003] Traditional wind tunnel experiments are a direct and effective method for aerodynamic testing. Wind tunnel experiments simulate the airflow when the vehicle is moving in a controlled laboratory environment and measure the surface pressure distribution and air resistance. However, wind tunnel experiments have the following disadvantages: 1. High cost: Building and maintaining wind tunnel experimental facilities require a large amount of capital investment.
[0004] 2. Time-consuming: The experimental preparation and data collection processes are time-consuming.
[0005] 3. Limited test conditions: The test conditions of wind tunnel experiments are limited, and it is difficult to comprehensively simulate the complex airflow in the real driving environment.
[0006] Compared with wind tunnel experiments, CFD simulation is a more economical and efficient method for aerodynamic analysis. CFD simulates the process of air flowing over the vehicle's surface by numerically solving the fluid mechanics equations. CFD simulation has the following advantages in automotive aerodynamic optimization: 1. Low cost: CFD simulation only requires computing resources and corresponding software tools, and the cost is much lower than that of wind tunnel experiments.
[0007] 2. High efficiency: CFD simulation can complete the calculation and analysis of complex flow fields in a relatively short time.
[0008] 3. Strong flexibility: CFD simulation can easily change the model and boundary conditions to adapt to different design requirements.
[0009] Therefore, an automotive air drag and friction simulation system based on CFD is provided. Summary of the Invention
[0010] In order to solve the above technical problems, the purpose of the present invention is to provide an automotive air drag and friction simulation system based on CFD.
[0011] To achieve the above object, the present invention provides the following technical solution: A CFD-based automotive air drag and friction simulation system, including a monitoring center, which is communicatively connected with a data acquisition module, a data processing module, a CFD simulation module, a real-time monitoring and visualization module, and a deep learning optimization module; The data acquisition module is used to collect vehicle shape geometric data, vehicle operation state data, and environmental condition data of the vehicle to be tested; The data processing module is used to perform model data processing on the vehicle shape geometric data to obtain vehicle geometric model data, and construct a vehicle geometric model according to the vehicle geometric model data; The CFD simulation module is used to perform unstructured grid processing on the vehicle geometric model to obtain unstructured grid data; obtain the initial simulation conditions of the vehicle geometric model according to the vehicle operation state data and environmental condition data, and obtain the drag coefficient and friction force index according to the unstructured grid data and the initial simulation conditions; The real-time monitoring and visualization module is used to monitor the drag coefficient and friction force index during the CFD simulation in real time, perform visualization processing on the real-time monitored drag coefficient and friction force index, and then judge whether the drag coefficient and friction force index of the vehicle to be tested meet the requirements; The deep learning optimization module is used to construct a deep learning model for training the CFD simulation results according to the drag coefficient and friction force index, obtain the drag and friction optimization coefficient according to the deep learning model for training the CFD simulation results; optimize the vehicle to be tested according to the drag and friction optimization coefficient.
[0012] Further, the process of the data acquisition module collecting the vehicle shape geometric data, vehicle operation state data, and environmental condition data of the vehicle to be tested includes: The data acquisition module is provided with an image scanning unit, a vehicle sensor unit, and an environmental data acquisition unit; The image scanning unit is used to scan the vehicle to be tested to obtain the vehicle shape geometric data of the vehicle to be tested; the vehicle sensor unit is used to collect the vehicle operation state data of the vehicle to be tested; the environmental data acquisition unit is used to collect the environmental condition data when the vehicle to be tested is driving The vehicle operation state data includes the driving speed; the environmental condition data includes the fluid temperature, fluid pressure, and fluid humidity.
[0013] Further, the process of the data processing module performing model data processing on the vehicle shape geometric data includes: Perform preprocessing operations on the obtained vehicle shape geometric data; The preprocessed vehicle exterior geometry data is discretized into a number of discrete points through equal-width discretization technology. Arbitrarily select one of the discrete points as the reference point, and assign different coordinates to the remaining discrete points according to the position information of the remaining discrete points relative to the reference point. Denote the discretized points after coordinate transformation as vehicle geometry model data. Preset a three-dimensional coordinate system. Initialize the coordinates of the vehicle geometry model data and import it into the three-dimensional coordinate system. Obtain the vehicle geometry model according to the non-uniform rational B-spline fitting technology.
[0014] Furthermore, the process by which the CFD simulation module performs unstructured grid processing on the vehicle geometry model includes: Divide different regions on the surface of the vehicle geometry model. According to the contact relationship between the air fluid and the vehicle geometry model in the CFD simulation process for different regions, select the corresponding element shape and perform unstructured grid division on the regions where the fluid flows to different surfaces of the vehicle geometry model, thereby obtaining unstructured grid data.
[0015] Furthermore, the process by which the CFD simulation module obtains the initial simulation conditions of the vehicle geometry model based on the vehicle operation state data and environmental condition data includes: Set the boundaries, inlets, and outlets of the vehicle geometry model. Perform data simulation on the vehicle operation state data of the vehicle to be tested and the environmental condition data during the driving of the vehicle to be tested to obtain simulated vehicle operation state data and simulated environmental condition data. The simulated vehicle operation state data includes the simulated driving speed; the simulated environmental condition data includes the simulated fluid temperature, simulated fluid pressure, and simulated fluid humidity. Mark the boundaries, inlets, outlets, simulated vehicle operation state data, and simulated environmental condition data as the initial simulation conditions of the vehicle geometry model.
[0016] Furthermore, the process by which the CFD simulation module obtains the drag coefficient and friction force index based on the unstructured grid data and the initial simulation conditions includes: Label the simulated vehicle operation state data and simulated environmental condition data in the initial simulation conditions. Use the finite difference method technology on the unstructured grid data to obtain the unstructured coefficient on the surface of the vehicle geometry model. Obtain the drag coefficient and friction force index based on the labeled simulated vehicle operation state data, simulated environmental condition data, and the unstructured coefficient.
[0017] Furthermore, the process by which the real-time monitoring and visualization module performs visualization processing on the drag coefficient and friction force index includes: Set the CFD simulation period; Obtain the wind resistance coefficient and friction force index for different CFD simulation periods; Plot a two-dimensional line graph with the CFD simulation period as the abscissa and the wind resistance coefficient or friction force index as the ordinate, which are respectively recorded as the wind resistance coefficient line graph and the friction force index line graph.
[0018] Furthermore, the process by which the real-time monitoring and visualization module determines whether the wind resistance coefficient and friction force index of the vehicle to be tested meet the requirements includes: Preset the standard wind resistance coefficient and standard friction force index; And import the standard wind resistance coefficient and standard friction force index into the wind resistance coefficient line graph and the friction force index line graph respectively; If the wind resistance coefficient within the CFD simulation period is greater than the standard wind resistance coefficient and the friction force index is greater than the standard friction force index, it indicates that the wind resistance coefficient and friction force index of the vehicle to be tested do not meet the requirements, and feedback to the deep learning optimization module; If the wind resistance coefficient within the CFD simulation period is less than or equal to the standard wind resistance coefficient and the friction force index is less than or equal to the standard friction force index, it indicates that the wind resistance coefficient and friction force index of the vehicle to be tested meet the requirements.
[0019] Furthermore, the process by which the deep learning optimization module constructs a deep learning model for training the CFD simulation results based on the wind resistance coefficient and friction force index and obtains the wind resistance and friction optimization coefficient includes: Construct a standard deep learning model according to the convolutional neural network technology; Obtain several groups of wind resistance coefficients and friction force indexes that meet the requirements; Divide the obtained wind resistance coefficient and friction force index into a training set, a validation set, and a test set, and input the training set, validation set, and test set into the standard deep learning model to train the standard deep learning model, and obtain the trained standard deep learning model, and record the trained standard deep learning model as the deep learning model for training the CFD simulation results; Obtain the wind resistance and friction optimization coefficient according to the deep learning model for training the CFD simulation results; optimize the wind resistance coefficient and friction force index of the vehicle to be tested that are determined to not meet the requirements according to the wind resistance and friction optimization coefficient.
[0020] Compared with the prior art, the beneficial effects of the present invention are as follows: by collecting the vehicle shape geometric data, vehicle operation state data, and environmental condition data of the vehicle to be experimented, a vehicle geometric model and initial simulation conditions are constructed; the vehicle geometric model is subjected to unstructured grid processing to obtain unstructured grid data; and based on the unstructured grid data and the initial simulation conditions, the wind resistance coefficient and the friction force index are obtained, and then it is judged whether the wind resistance coefficient and the friction force index of the vehicle to be experimented meet the requirements; and based on the wind resistance coefficient and the friction force index, a CFD simulation result training deep learning model is constructed to obtain a wind resistance and friction optimization coefficient, thereby improving the simulation efficiency of wind resistance and friction. Brief Description of the Drawings
[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.
[0022] Figure 1 It is a schematic diagram of the present invention. Detailed Embodiments
[0023] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other implementation manners obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present invention.
[0024] As Figure 1 shown, a CFD-based vehicle air resistance and friction simulation system includes a monitoring center, and the monitoring center is communicatively connected to a data acquisition module, a data processing module, a CFD simulation module, a real-time monitoring and visualization module, and a deep learning optimization module; The data acquisition module is used to collect the vehicle shape geometric data, vehicle operation state data, and environmental condition data of the vehicle to be experimented; The data processing module is used to perform model data processing on the vehicle shape geometric data to obtain vehicle geometric model data, and construct a vehicle geometric model according to the vehicle geometric model data; The CFD simulation module is used to perform unstructured grid processing on the vehicle geometric model to obtain unstructured grid data; obtain the initial simulation conditions of the vehicle geometric model according to the vehicle operation state data and the environmental condition data, and obtain the wind resistance coefficient and the friction force index according to the unstructured grid data and the initial simulation conditions; The real-time monitoring and visualization module is used to monitor the wind resistance coefficient and friction force index during the CFD simulation in real time, and perform visualization processing on the wind resistance coefficient and friction force index monitored in real time, so as to judge whether the wind resistance coefficient and friction force index of the vehicle to be tested meet the requirements; The deep learning optimization module is used to construct a deep learning model for training the CFD simulation results according to the wind resistance coefficient and friction force index, and obtain the wind resistance and friction optimization coefficient according to the deep learning model trained by the CFD simulation results; optimize the vehicle to be tested according to the wind resistance and friction optimization coefficient.
[0025] It should be further noted that in the specific implementation process, the specific process of the data acquisition module collecting the vehicle shape geometric data, vehicle operation state data and environmental condition data of the vehicle to be tested includes: The data acquisition module is provided with an image scanning unit, a vehicle sensor unit and an environmental data acquisition unit; The image scanning unit is used to scan the vehicle to be tested to obtain the vehicle shape geometric data of the vehicle to be tested; the vehicle sensor unit is used to collect the vehicle operation state data of the vehicle to be tested; the environmental data acquisition unit is used to collect the environmental condition data when the vehicle to be tested is driving; It should be further noted that the vehicle operation state data includes the driving speed; the environmental condition data includes the fluid temperature, fluid pressure and fluid humidity.
[0026] The data acquisition module uploads the collected vehicle shape geometric data to the data processing module; uploads the collected vehicle operation state data and environmental condition data to the CFD simulation module.
[0027] It should be further noted that in the specific implementation process, the specific process of the data processing module performing model data processing on the vehicle shape geometric data includes: Perform preprocessing operations on the obtained vehicle shape geometric data, and the preprocessing operations include denoising and filling missing values, so as to remove noise and error points in the vehicle shape geometric data and ensure data continuity; Generate a number of discrete points from the preprocessed vehicle shape geometric data through equal-width discretization technology, select any one of the discrete points as the reference point, and assign different coordinates to the remaining discrete points according to the position information of the remaining discrete points and the reference point; record the discretized points after coordinate transformation as vehicle geometric model data; Preset a three-dimensional coordinate system; Initialize the coordinates of the vehicle geometric model data and import it into the three-dimensional coordinate system. According to the non-uniform rational B-spline fitting technology, fit the vehicle geometric model data in the three-dimensional coordinate system to form a complete geometric model, and denote this model as the vehicle geometric model.
[0028] The data processing module uploads the constructed vehicle geometric model to the CFD simulation module.
[0029] It should be further noted that in the specific implementation process, the specific process of the CFD simulation module performing unstructured grid processing on the vehicle geometric model includes: Divide different regions on the surface of the vehicle geometric model. According to the contact relationship between the air fluid and the vehicle geometric model during the CFD simulation in different regions, select the corresponding element shape, and perform unstructured grid division on different regions flowing to the surface of the vehicle geometric model, thereby obtaining unstructured grid data; It should be further noted that the element shapes include tetrahedral element shapes and hexahedral element shapes; the contact relationships include direct contact and indirect contact; the accuracy of the hexahedral element shape is higher than that of the tetrahedral element shape, and it is used to divide the region where the contact relationship between the air fluid and the vehicle geometric model is direct contact; the tetrahedral element shape is used to divide the region where the contact relationship between the air fluid and the vehicle geometric model is indirect contact.
[0030] It should be further noted that in the specific implementation process, the specific process of the CFD simulation module obtaining the initial simulation conditions of the vehicle geometric model according to the vehicle operation state data and the environmental condition data includes: Set the boundaries, inlets, and outlets of the vehicle geometric model; It should be further noted that the boundary is the edge where the vehicle geometric model is in contact with the air fluid; the inlet is the position where the air fluid flows towards the vehicle geometric model; the outlet is the position where the air fluid flows out of the vehicle geometric model; Perform data simulation on the vehicle operation state data of the vehicle to be tested and the environmental condition data during the driving of the vehicle to be tested. The specific process includes: According to the actually collected vehicle operation state data, set the same vehicle operation state data for the vehicle geometric model, and denote it as the simulated vehicle operation state data; According to the actually collected environmental condition data, set the same environmental condition data for the air fluid during the CFD simulation, and denote it as the simulated environmental condition data; It should be further noted that the simulated vehicle operation state data includes the simulated driving speed; the simulated environmental condition data includes the simulated fluid temperature, the simulated fluid pressure, and the simulated fluid humidity; Mark the boundary, inlet, outlet, simulation vehicle operation state data, and simulation environment condition data as the initial simulation conditions of the vehicle geometric model.
[0031] It should be further noted that in the specific implementation process, the specific process by which the CFD simulation module obtains the drag coefficient and friction force index according to the unstructured grid data and the initial simulation conditions includes: Number the simulation vehicle operation state data and simulation environment condition data in the initial simulation conditions. The specific process includes: Mark the simulation driving speed as V; mark the simulation fluid temperature, simulation fluid pressure, and simulation fluid humidity as T, P, and S respectively; Obtain the unstructured coefficient γ on the surface of the vehicle geometric model from the unstructured grid data through the finite difference method technology; It should be further noted that in the specific implementation process, the specific process of obtaining the drag coefficient includes: According to the unstructured coefficient γ, simulation driving speed V, simulation fluid temperature T, simulation fluid pressure P, and simulation fluid humidity S, the calculation formula for obtaining the drag coefficient is: ; where δ is the drag coefficient.
[0032] It should be further noted that in the specific implementation process, the specific process of obtaining the friction force index includes: According to the unstructured coefficient γ, simulation driving speed V, simulation fluid temperature T, simulation fluid pressure P, simulation fluid humidity S, and drag coefficient δ, the calculation formula for obtaining the friction force index is:
[0033] where ∂ is the friction force index; R is the air fluid constant; A is the contact area; τ is the friction coefficient adjustment factor.
[0034] The CFD simulation module uploads the drag coefficient and the friction force index to the real-time monitoring visualization module and the deep learning optimization module.
[0035] It should be further noted that in the specific implementation process, the specific process by which the real-time monitoring visualization module performs visualization processing on the drag coefficient and the friction force index includes: Set the CFD simulation period; Obtain the drag coefficient and the friction force index for different CFD simulation periods; Draw a two-dimensional line chart with the CFD simulation period as the abscissa and the drag coefficient or the friction force index as the ordinate, which are respectively recorded as the drag coefficient line chart and the friction force index line chart.
[0036] It should be further noted that in the specific implementation process, the specific process by which the real-time monitoring and visualization module determines whether the wind resistance coefficient and the friction force index of the vehicle to be tested meet the requirements includes: Preset the standard wind resistance coefficient and the standard friction force index; And import the standard wind resistance coefficient and the standard friction force index into the wind resistance coefficient line chart and the friction force index line chart respectively; If the wind resistance coefficient during the CFD simulation period is greater than the standard wind resistance coefficient and the friction force index is greater than the standard friction force index, it indicates that the wind resistance coefficient and the friction force index of the vehicle to be tested do not meet the requirements and need to be further optimized, and feedback to the deep learning optimization module; If the wind resistance coefficient during the CFD simulation period is less than or equal to the standard wind resistance coefficient and the friction force index is less than or equal to the standard friction force index, it indicates that the wind resistance coefficient and the friction force index of the vehicle to be tested meet the requirements.
[0037] It should be further noted that in the specific implementation process, the specific process by which the deep learning optimization module constructs a deep learning model for training the CFD simulation results based on the wind resistance coefficient and the friction force index and obtains the wind resistance and friction optimization coefficient includes: Construct a standard deep learning model according to the convolutional neural network technology; Obtain several groups of wind resistance coefficients and friction force indexes that meet the requirements; Divide the obtained wind resistance coefficients and friction force indexes into a training set, a validation set, and a test set, and input the training set, the validation set, and the test set into the standard deep learning model to train the standard deep learning model, and obtain the standard deep learning model after completion of training, and denote the standard deep learning model after completion of training as the deep learning model for training the CFD simulation results; According to the deep learning model for training the CFD simulation results, the calculation formula for obtaining the wind resistance and friction optimization coefficient is: ; Where, ω is the wind resistance and friction optimization coefficient; 𝑁 is the total number of wind resistance coefficients and friction force indexes; δt is the wind resistance coefficient of the t-th group; ∂t is the friction force index of the t-th group.
[0038] Optimize the wind resistance coefficient δ and the friction force index ∂ of the vehicle to be tested that are determined to not meet the requirements according to the wind resistance and friction optimization coefficient ω.
[0039] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.
Claims
1. The CFD-based automotive air drag and friction simulation system, including a monitoring center, is characterized in that The monitoring center is communicatively connected to a data acquisition module, a data processing module, a CFD simulation module, a real-time monitoring visualization module, and a deep learning optimization module; The data acquisition module is used to acquire the vehicle external geometry data, vehicle operating state data, and environmental condition data of the vehicle to be tested; The data processing module is used to perform model data processing on the vehicle external geometry data to obtain vehicle geometry model data, and construct a vehicle geometry model according to the vehicle geometry model data; The CFD simulation module is used to perform unstructured grid processing on the vehicle geometry model to obtain unstructured grid data; According to the vehicle operating state data and environmental condition data, obtain the initial simulation conditions of the vehicle geometry model, and according to the unstructured grid data and the initial simulation conditions, obtain the drag coefficient and the friction index; The real-time monitoring visualization module is used to monitor the drag coefficient and the friction index during the CFD simulation in real time, and perform visualization processing on the drag coefficient and the friction index monitored in real time, so as to judge whether the drag coefficient and the friction index of the vehicle to be tested meet the requirements; The deep learning optimization module is used to construct a deep learning model for training the CFD simulation results according to the drag coefficient and the friction index, and obtain the drag and friction optimization coefficient according to the deep learning model for training the CFD simulation results; Optimize the vehicle to be tested according to the drag and friction optimization coefficient.
2. The CFD-based automotive air drag friction simulation system according to claim 1, characterized in that, The process of the data acquisition module acquiring the vehicle external geometry data, vehicle operating state data, and environmental condition data of the vehicle to be tested includes: The data acquisition module is provided with an image scanning unit, a vehicle sensor unit, and an environmental data acquisition unit; The image scanning unit is used to scan the vehicle to be tested to obtain the vehicle external geometry data of the vehicle to be tested; the vehicle sensor unit is used to acquire the vehicle operating state data of the vehicle to be tested; the environmental data acquisition unit is used to acquire the environmental condition data when the vehicle to be tested is driving The vehicle operating state data includes the driving speed; the environmental condition data includes the fluid temperature, fluid pressure, and fluid humidity.
3. The CFD-based automotive air drag and friction simulation system according to claim 2, characterized in that, The process of the data processing module performing model data processing on the vehicle external geometry data includes: Perform preprocessing operations on the obtained vehicle external geometry data; Generate a number of discrete points from the preprocessed vehicle external geometry data through equal-width discretization technology, select any one of the discrete points as the reference point, and assign different coordinates to the remaining discrete points according to the position information of the remaining discrete points and the reference point; record the discretized points after coordinate transformation as vehicle geometry model data; Preset a three-dimensional coordinate system; Perform coordinate initialization on the vehicle geometry model data and import it into the three-dimensional coordinate system, and obtain a vehicle geometry model according to the non-uniform rational B-spline fitting technology.
4. The CFD-based automotive air drag friction simulation system according to claim 3, characterized in that, The process of the CFD simulation module performing unstructured grid processing on the vehicle geometry model includes: Divide different regions on the surface of the vehicle geometric model. According to the contact relationship between the air fluid and the vehicle geometric model in the CFD simulation process for different regions, select the corresponding element shape, perform unstructured grid division on different regions flowing towards the surface of the vehicle geometric model, and then obtain unstructured grid data.
5. The CFD-based automotive air drag and friction simulation system according to claim 4, wherein The process by which the CFD simulation module obtains the initial simulation conditions of the vehicle geometric model based on the vehicle operating state data and environmental condition data includes: Set the boundaries, inlets, and outlets of the vehicle geometric model; Perform data simulation on the vehicle operating state data of the vehicle to be tested and the environmental condition data during the driving of the vehicle to be tested to obtain simulated vehicle operating state data and simulated environmental condition data; The simulated vehicle operating state data includes the simulated driving speed; the simulated environmental condition data includes the simulated fluid temperature, simulated fluid pressure, and simulated fluid humidity; Mark the boundaries, inlets, outlets, simulated vehicle operating state data, and simulated environmental condition data as the initial simulation conditions of the vehicle geometric model.
6. The CFD-based automotive air drag friction simulation system according to claim 5, wherein The process by which the CFD simulation module obtains the drag coefficient and friction force index based on the unstructured grid data and the initial simulation conditions includes: Number the simulated vehicle operating state data and simulated environmental condition data in the initial simulation conditions; Obtain the unstructured coefficient on the surface of the vehicle geometric model by using the finite difference method technology for the unstructured grid data; Obtain the drag coefficient and friction force index based on the numbered simulated vehicle operating state data, simulated environmental condition data, and unstructured coefficient.
7. The CFD-based automotive air drag and friction simulation system according to claim 6, characterized in that The process by which the real-time monitoring and visualization module performs visualization processing on the drag coefficient and friction force index includes: Set the CFD simulation period; Obtain the drag coefficient and friction force index for different CFD simulation periods; Draw a two-dimensional line chart with the CFD simulation period as the abscissa and the drag coefficient or friction force index as the ordinate, which are respectively recorded as the drag coefficient line chart and the friction force index line chart.
8. The CFD-based automotive air drag and friction simulation system according to claim 7, characterized in that The process by which the real-time monitoring and visualization module determines whether the drag coefficient and friction force index of the vehicle to be tested meet the requirements includes: Preset the standard drag coefficient and standard friction force index; And import the standard drag coefficient and standard friction force index into the drag coefficient line chart and the friction force index line chart respectively; If the drag coefficient within the CFD simulation period is greater than the standard drag coefficient and the friction force index is greater than the standard friction force index, it means that the drag coefficient and friction force index of the vehicle to be tested do not meet the requirements, and feedback to the deep learning optimization module; If the drag coefficient within the CFD simulation period is less than or equal to the standard drag coefficient and the friction force index is less than or equal to the standard friction force index, it means that the drag coefficient and friction force index of the vehicle to be tested meet the requirements.
9. The CFD-based automotive air drag friction simulation system according to claim 8, characterized in that, The process by which the deep learning optimization module constructs a CFD simulation result training deep learning model based on the drag coefficient and friction force index and obtains the drag and friction optimization coefficient includes: Construct a standard deep learning model according to the convolutional neural network technology; Obtain several groups of drag coefficients and friction force indices that meet the requirements; Divide the obtained drag coefficient and friction index into a training set, a validation set, and a test set, and input the training set, validation set, and test set into a standard deep learning model to train the standard deep learning model, obtaining a trained standard deep learning model, and denote the trained standard deep learning model as the CFD simulation result trained deep learning model; According to the CFD simulation result trained deep learning model, obtain the drag and friction optimization coefficient; optimize the drag coefficient and friction index of the vehicle to be tested that are judged to not meet the requirements according to the drag and friction optimization coefficient.