CN113505440A - Automobile pneumatic performance parameter real-time prediction method based on three-dimensional deep learning - Google Patents
Automobile pneumatic performance parameter real-time prediction method based on three-dimensional deep learning Download PDFInfo
- Publication number
- CN113505440A CN113505440A CN202110859887.XA CN202110859887A CN113505440A CN 113505440 A CN113505440 A CN 113505440A CN 202110859887 A CN202110859887 A CN 202110859887A CN 113505440 A CN113505440 A CN 113505440A
- Authority
- CN
- China
- Prior art keywords
- dimensional
- data
- model
- deep learning
- performance parameters
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000013135 deep learning Methods 0.000 title claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000013528 artificial neural network Methods 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 7
- 238000012360 testing method Methods 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000004088 simulation Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 238000013145 classification model Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 2
- 238000012986 modification Methods 0.000 abstract description 2
- 230000004048 modification Effects 0.000 abstract description 2
- 230000001737 promoting effect Effects 0.000 abstract description 2
- 230000003044 adaptive effect Effects 0.000 abstract 1
- 238000013461 design Methods 0.000 description 8
- 239000012530 fluid Substances 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Abstract
The invention relates to a real-time prediction method of automobile aerodynamic performance parameters based on three-dimensional deep learning, which comprises the steps of constructing an automobile body three-dimensional model by using a T sample strip; simulating and calculating the pneumatic performance parameters of the model by using CFD software, and adding a wind speed label into the three-dimensional point cloud data to construct a training data set; and performing adaptive modification based on a PointNet network, taking the processed point cloud data as the input of a deep neural network, adjusting training parameters, and predicting the aerodynamic performance parameters of the automobile by using the trained deep neural network. The method realizes the rapid real-time prediction of the aerodynamic performance parameters of the automobile, avoids the complicated CFD calculation by directly applying the experimental data, and has great significance for promoting the application of deep learning in aerodynamics.
Description
Technical Field
The invention relates to the technical field of aerodynamics and deep learning, in particular to a method for predicting automobile aerodynamic performance parameters in real time based on three-dimensional deep learning.
Background
The design of the pneumatic layout of the automobile is an important link in the design of the automobile. The aerodynamic performance of the automobile can be improved by reducing the aerodynamic resistance coefficient of the automobile, and the aerodynamic performance has important significance for saving energy, reducing emission of a fuel vehicle and increasing the driving range of the electric vehicle.
At present, the aerodynamic performance of automobiles is mainly researched by a method of matching numerical analysis and experimental analysis based on theoretical research. The main approaches of test analysis include road tests and automobile wind tunnel tests, the test sites of the road tests are scarce, the test period is long, and the time and the capital cost are high. The method can artificially design the flow field property of the wind tunnel, control the test condition and obtain relatively reliable results, but has high test difficulty and high cost. The numerical simulation method is developed based on Computational Fluid Dynamics (CFD), and is mainly used for detecting the pneumatic performance of an automobile under an artificially set condition by using a computer, but the high-freedom large-scale engineering problem depends on a supercomputer, so that the method has high requirements on the performance of the computer and consumes a large amount of time. Therefore, for automobile designers, a method for predicting the aerodynamic performance of an automobile in real time is urgently needed, and the aerodynamic layout and appearance design of the automobile can be carried out more quickly, so that the automobile design and research and development period is reduced. The deep learning technology brings possibility for realizing rapid prediction of automobile pneumatic performance parameters.
Disclosure of Invention
In view of the above problems, an object of the present invention is to provide a method for predicting aerodynamic performance parameters of an automobile in real time based on three-dimensional deep learning, which, in the context of applying deep learning to fluid analysis, realizes fast real-time prediction of aerodynamic performance parameters of an automobile by directly applying experimental data without complicated CFD calculation, and has great significance in promoting application of deep learning to aerodynamics.
The technical scheme adopted by the invention is as follows:
the invention provides a real-time automobile pneumatic performance parameter prediction method based on three-dimensional deep learning, which specifically comprises the following steps:
s1, acquiring a three-dimensional digital model and point cloud data of the vehicle body, and processing the data; the specific process is as follows: constructing an automobile shape curved surface model by using T-Splines, then extracting surface control points and outputting three-dimensional coordinates of each point, encrypting point cloud data, and labeling the data;
s2, acquiring the pneumatic performance parameters of the vehicle; the specific process is as follows: carrying out grid division in ICEM, and carrying out pneumatic performance simulation analysis on the constructed three-dimensional model of the vehicle body by using CFD software CFX to obtain parameters such as air resistance coefficient;
s3, acquiring a neural network model capable of predicting the pneumatic performance in real time, and taking the output of the steps S1 and S2 as input data; the specific process is as follows: modifying a PointNet network, comprising:
for the type of input data, setting a three-dimensional transformation matrix which is used for carrying out normalization processing on original data and primary features as a four-dimensional matrix so as to adapt to the change of data dimensions; changing the classification model into a regression model, specifically, adding a full connection layer after the last layer of the multilayer perceptron, and corresponding to a specific numerical value;
aiming at the change of the structure, a related calculation method of visual indexes such as a Loss function, accuracy and the like is changed, and the original Loss function is modified into a mean square error index which is commonly used for evaluating a regression model;
importing the encrypted point cloud data into a modified PointNet network, and adjusting training parameters; and evaluating the effect and the property of the neural network, adopting the MAE to evaluate the point-to-point dispersion degree of the true value and the predicted value, and evaluating the error rate by using the ME.
Further, in step S2, when performing aerodynamic performance simulation analysis on the constructed three-dimensional model of the vehicle body, aerodynamic performance parameters at different wind speeds are obtained by changing wind speed conditions, and wind speed labels are added to all points of each model, so as to obtain a processed data set.
Further, in step S3, according to the characteristics of the PointNet network and the number of the data sets, all the data are divided into two training sets and two testing sets, and the two training sets and the two testing sets are stored as text files respectively.
Further, in the step S3, the main indicators of the evaluation neural network are SSE, MSE, R-Square, MAE, and the like.
Further, in step S3, the training parameters include an initial learning rate, a maximum Epoch, and a Batch-Size.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the pneumatic performance prediction method based on deep learning, provided by the invention, the three-dimensional data of the surface characteristics of the required vehicle body is obtained by means of a three-dimensional neural network, computational fluid mechanics simulation analysis software and a T-Splines plug-in, the pneumatic performance of the vehicle body model is analyzed to obtain performance parameters, the obtained data is led into the three-dimensional neural network for iterative training, the trained model can predict the pneumatic performance parameters of the vehicle in real time, the pneumatic layout appearance design of the vehicle is accelerated, and therefore the design and research and development period of the vehicle is reduced;
2. compared with the traditional road experiment and wind tunnel experiment, the aerodynamic performance prediction method based on deep learning has the advantages of short period and low cost, does not need complex parameter design and iterative calculation, can predict data quickly according to a neural network, and avoids complicated CFD calculation by directly applying experimental data.
Drawings
FIG. 1 is a structural block diagram of a method for predicting aerodynamic performance parameters of an automobile in real time based on three-dimensional deep learning, which is provided by the invention;
FIG. 2 is a schematic diagram illustrating the effect of using a spline curve T-spline to construct a curved surface of a vehicle body in the present invention;
FIG. 3 is a schematic diagram showing a comparison of an encrypted point cloud model, an unencrypted point cloud model, and an original model in accordance with the present invention.
Wherein, the reference numbers: 10-constructing a data module; 20-a simulation analysis module; 30-neural network training module.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
The invention provides a method for predicting automobile aerodynamic performance parameters in real time based on three-dimensional deep learning, which specifically comprises the following steps as shown in figure 1:
s1, acquiring a three-dimensional digital model and point cloud data of the vehicle body, and processing the data; the specific process is as follows: constructing an automobile shape curved surface model by using T-Splines, then extracting surface control points and outputting three-dimensional coordinates of each point, carrying out grid division in ICEM, carrying out encryption processing on three-dimensional coordinate data and point cloud data, and labeling the data; the constructed body contour curved surface model needs to be capable of reflecting the aerodynamic contour characteristics and proportion of the automobile; firstly, it is necessary to ensure that the model has no gaps, because the gaps will affect the subsequent simulation steps; secondly, the curved surface model should include key components which have a large influence on the aerodynamic parameters, such as wheels, tail spoilers, and the like; thirdly, the built three-dimensional model library should contain various vehicle types such as cars, off-road vehicles, sports cars, trucks and the like as much as possible so as to improve the universality, and the spline curve includes but is not limited to a T spline;
s2, acquiring the aerodynamic performance parameters of the vehicle, wherein the aerodynamic performance parameters include but are not limited to air resistance coefficients; the specific process is as follows: performing pneumatic performance simulation analysis on the constructed three-dimensional model of the vehicle body by using CFD software CFX, wherein the computational fluid dynamics simulation software comprises but is not limited to CFX software; carrying out grid division on the prepared model by using ICEM, setting different wind speed parameters, then establishing a computational domain simulation wind tunnel test in CFX, and carrying out pneumatic performance parameter simulation; when the constructed three-dimensional model of the vehicle body is subjected to aerodynamic performance simulation analysis, aerodynamic performance parameters under different wind speeds are obtained by changing wind speed conditions, and wind speed labels are added to all points of each model respectively, so that a processed data set is obtained.
S3, acquiring a neural network model capable of predicting the pneumatic performance in real time, and taking the output of the steps S1 and S2 as input data; the specific process is as follows: firstly, modifying a PointNet network, specifically comprising the following steps:
for the type of input data, setting a three-dimensional transformation matrix which is used for carrying out normalization processing on original data and primary features as a four-dimensional matrix so as to adapt to the change of data dimensions; changing the classification model into a regression model, specifically, adding a full connection layer after the last layer of the multilayer perceptron, and corresponding to a specific numerical value;
aiming at the change of the structure, a related calculation method of visual indexes such as a Loss function, accuracy and the like is changed, and the original Loss function is modified into a mean square error index which is commonly used for evaluating a regression model;
dividing all data into two training sets and two testing sets according to the characteristics of the PointNet network and the number of the processed data sets, and respectively storing the two training sets and the two testing sets as text files;
importing the encrypted point cloud data into a modified PointNet network, and adjusting training parameters, wherein the training parameters include but are not limited to initial learning rate, maximum Epoch and Batch-Size; evaluating the effect and the property of the neural network, adopting MAE to evaluate the point-to-point discrete degree of the true value and the predicted value, and using ME to evaluate the error rate; the main indicators for evaluating neural networks include, but are not limited to SSE, MSE, R-Square, and MAE.
The method combines deep learning and computational fluid dynamics, carries out aerodynamic performance analysis on the constructed three-dimensional model through CFD simulation software, processes the obtained data to be used as input information of a three-dimensional neural network, adaptively modifies the neural network according to the input data, and obtains the neural network model capable of predicting the automobile aerodynamic performance parameters in real time through training the input data.
The above-mentioned embodiments are merely illustrative of the preferred embodiments of the present invention, and do not limit the scope of the present invention, and various modifications and improvements of the technical solution of the present invention by those skilled in the art should fall within the protection scope defined by the claims of the present invention without departing from the spirit of the present invention.
Claims (5)
1. A real-time automobile pneumatic performance parameter prediction method based on three-dimensional deep learning is characterized by specifically comprising the following steps of:
s1, acquiring a three-dimensional digital model and point cloud data of the vehicle body, and processing the data; the specific process is as follows: constructing an automobile shape curved surface model by using T-Splines, then extracting surface control points and outputting three-dimensional coordinates of each point, encrypting point cloud data, and labeling the data;
s2, acquiring the pneumatic performance parameters of the vehicle; the specific process is as follows: carrying out grid division in ICEM, and carrying out pneumatic performance simulation analysis on the constructed three-dimensional model of the vehicle body by using CFD software CFX to obtain parameters such as air resistance coefficient;
s3, acquiring a neural network model capable of predicting the pneumatic performance in real time, and taking the output of the steps S1 and S2 as input data; the specific process is as follows: modifying a PointNet network, comprising:
for the type of input data, setting a three-dimensional transformation matrix which is used for carrying out normalization processing on original data and primary features as a four-dimensional matrix so as to adapt to the change of data dimensions; changing the classification model into a regression model, specifically, adding a full connection layer after the last layer of the multilayer perceptron, and corresponding to a specific numerical value;
aiming at the change of the structure, a related calculation method of visual indexes such as a Loss function, accuracy and the like is changed, and the original Loss function is modified into a mean square error index which is commonly used for evaluating a regression model;
importing the encrypted point cloud data into a modified PointNet network, and adjusting training parameters; and evaluating the effect and the property of the neural network, adopting the MAE to evaluate the point-to-point dispersion degree of the true value and the predicted value, and evaluating the error rate by using the ME.
2. The method for predicting the pneumatic performance parameters in real time based on the three-dimensional deep learning according to claim 1, wherein the method comprises the following steps: in the step S2, when the aerodynamic performance simulation analysis is performed on the constructed three-dimensional model of the vehicle body, aerodynamic performance parameters at different wind speeds are obtained by changing the wind speed conditions, and wind speed labels are added to all points of each model, so as to obtain a processed data set.
3. The method for predicting the pneumatic performance parameters in real time based on the three-dimensional deep learning according to claim 2, wherein the method comprises the following steps: in step S3, according to the characteristics of the PointNet network and the number of the data sets, all the data are divided into two training sets and two testing sets, which are stored as text files.
4. The method for predicting the pneumatic performance parameters in real time based on the three-dimensional deep learning according to claim 1, wherein the method comprises the following steps: in step S3, the main indicators for evaluating the neural network are SSE, MSE, R-Square, and MAE.
5. The method for predicting the pneumatic performance parameters in real time based on the three-dimensional deep learning according to claim 1, wherein the method comprises the following steps: in step S3, the training parameters include an initial learning rate, a maximum Epoch, and a Batch-Size.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110859887.XA CN113505440A (en) | 2021-07-28 | 2021-07-28 | Automobile pneumatic performance parameter real-time prediction method based on three-dimensional deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110859887.XA CN113505440A (en) | 2021-07-28 | 2021-07-28 | Automobile pneumatic performance parameter real-time prediction method based on three-dimensional deep learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113505440A true CN113505440A (en) | 2021-10-15 |
Family
ID=78014957
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110859887.XA Pending CN113505440A (en) | 2021-07-28 | 2021-07-28 | Automobile pneumatic performance parameter real-time prediction method based on three-dimensional deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113505440A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114117954A (en) * | 2021-11-08 | 2022-03-01 | 北京工业大学 | Dynamic real-time visualization method for three-dimensional reaction field in reactor |
CN114972165A (en) * | 2022-03-24 | 2022-08-30 | 中山大学孙逸仙纪念医院 | Method and device for measuring time-average shearing force |
WO2023155414A1 (en) * | 2022-02-17 | 2023-08-24 | 中汽研(天津)汽车工程研究院有限公司 | Intelligent prediction method for automobile aerodynamic performance |
CN116644522A (en) * | 2023-07-27 | 2023-08-25 | 中汽研(天津)汽车工程研究院有限公司 | Method, device and medium for determining automobile modeling parameters for pneumatic development |
CN117634365A (en) * | 2024-01-25 | 2024-03-01 | 中国航空工业集团公司沈阳空气动力研究所 | Airplane aerodynamic force prediction method, electronic equipment and storage medium |
CN117634365B (en) * | 2024-01-25 | 2024-04-26 | 中国航空工业集团公司沈阳空气动力研究所 | Airplane aerodynamic force prediction method, electronic equipment and storage medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110073301A (en) * | 2017-08-02 | 2019-07-30 | 强力物联网投资组合2016有限公司 | The detection method and system under data collection environment in industrial Internet of Things with large data sets |
US20200320386A1 (en) * | 2017-12-26 | 2020-10-08 | Landmark Graphics Corporation | Effective Representation of Complex Three-Dimensional Simulation Results for Real-Time Operations |
CN112034730A (en) * | 2019-05-20 | 2020-12-04 | 辉达公司 | Autonomous vehicle simulation using machine learning |
CN112163380A (en) * | 2020-09-24 | 2021-01-01 | 哈尔滨锅炉厂有限责任公司 | System and method for predicting furnace oxygen concentration based on numerical simulation neural network |
CN112417708A (en) * | 2020-12-11 | 2021-02-26 | 杭州博镨科技有限公司 | Method for predicting brake pad formula performance based on neural network |
CN112703457A (en) * | 2018-05-07 | 2021-04-23 | 强力物联网投资组合2016有限公司 | Method and system for data collection, learning and machine signal streaming for analysis and maintenance using industrial internet of things |
CN112733442A (en) * | 2020-12-31 | 2021-04-30 | 交通运输部公路科学研究所 | Road surface long-term performance prediction model based on deep learning and construction method thereof |
-
2021
- 2021-07-28 CN CN202110859887.XA patent/CN113505440A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110073301A (en) * | 2017-08-02 | 2019-07-30 | 强力物联网投资组合2016有限公司 | The detection method and system under data collection environment in industrial Internet of Things with large data sets |
US20200320386A1 (en) * | 2017-12-26 | 2020-10-08 | Landmark Graphics Corporation | Effective Representation of Complex Three-Dimensional Simulation Results for Real-Time Operations |
CN112703457A (en) * | 2018-05-07 | 2021-04-23 | 强力物联网投资组合2016有限公司 | Method and system for data collection, learning and machine signal streaming for analysis and maintenance using industrial internet of things |
CN112034730A (en) * | 2019-05-20 | 2020-12-04 | 辉达公司 | Autonomous vehicle simulation using machine learning |
CN112163380A (en) * | 2020-09-24 | 2021-01-01 | 哈尔滨锅炉厂有限责任公司 | System and method for predicting furnace oxygen concentration based on numerical simulation neural network |
CN112417708A (en) * | 2020-12-11 | 2021-02-26 | 杭州博镨科技有限公司 | Method for predicting brake pad formula performance based on neural network |
CN112733442A (en) * | 2020-12-31 | 2021-04-30 | 交通运输部公路科学研究所 | Road surface long-term performance prediction model based on deep learning and construction method thereof |
Non-Patent Citations (1)
Title |
---|
郑文韬: "基于三维深度学习的汽车气动性能实时预测", 《专业学位硕士学位论文》》, 15 February 2020 (2020-02-15), pages 1 - 57 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114117954A (en) * | 2021-11-08 | 2022-03-01 | 北京工业大学 | Dynamic real-time visualization method for three-dimensional reaction field in reactor |
CN114117954B (en) * | 2021-11-08 | 2023-04-18 | 北京工业大学 | Dynamic real-time visualization method for three-dimensional reaction field in reactor |
WO2023155414A1 (en) * | 2022-02-17 | 2023-08-24 | 中汽研(天津)汽车工程研究院有限公司 | Intelligent prediction method for automobile aerodynamic performance |
CN114972165A (en) * | 2022-03-24 | 2022-08-30 | 中山大学孙逸仙纪念医院 | Method and device for measuring time-average shearing force |
CN114972165B (en) * | 2022-03-24 | 2024-03-15 | 中山大学孙逸仙纪念医院 | Method and device for measuring time average shearing force |
CN116644522A (en) * | 2023-07-27 | 2023-08-25 | 中汽研(天津)汽车工程研究院有限公司 | Method, device and medium for determining automobile modeling parameters for pneumatic development |
CN116644522B (en) * | 2023-07-27 | 2023-09-22 | 中汽研(天津)汽车工程研究院有限公司 | Method, device and medium for determining automobile modeling parameters for pneumatic development |
CN117634365A (en) * | 2024-01-25 | 2024-03-01 | 中国航空工业集团公司沈阳空气动力研究所 | Airplane aerodynamic force prediction method, electronic equipment and storage medium |
CN117634365B (en) * | 2024-01-25 | 2024-04-26 | 中国航空工业集团公司沈阳空气动力研究所 | Airplane aerodynamic force prediction method, electronic equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113505440A (en) | Automobile pneumatic performance parameter real-time prediction method based on three-dimensional deep learning | |
US20190087713A1 (en) | Compression of sparse deep convolutional network weights | |
CN105868787A (en) | Electric car driving range evaluation method based on working condition identification and fuzzy energy consumption | |
CN111523579B (en) | Vehicle type recognition method and system based on improved deep learning | |
CN111597631B (en) | Automobile wind resistance coefficient optimization method based on self-adaptive agent model | |
CN110991631A (en) | Neural network acceleration system based on FPGA | |
CN114821217B (en) | Image recognition method and device based on quantum classical hybrid neural network | |
Singh et al. | Deep learning in the automotive industry: Recent advances and application examples | |
Hamed et al. | Fuel consumption prediction model using machine learning | |
CN113591717A (en) | Non-motor vehicle helmet wearing detection method based on improved YOLOv3 algorithm | |
Hu et al. | Driver identification through vehicular CAN bus data: An ensemble deep learning approach | |
Fu et al. | Prediction of automobile aerodynamic drag coefficient for SUV cars based on a novel XGBoost model | |
CN112287461B (en) | Automobile driving simulator braking system modeling method based on Gaussian process regression | |
CN116049989A (en) | Bogie digital twin model construction method, system, electronic equipment and medium | |
CN114282680A (en) | Vehicle exhaust emission prediction method and system based on machine learning algorithm | |
CN103207083B (en) | A kind of method and device analyzing turbine efficiency | |
Kibalama et al. | Route Generation Methodology for Energy Efficiency Evaluation of Connected and Automated Vehicles⋆ | |
Xiong et al. | Study on NVH robustness evaluation method of high-mileage automobile based on systematic sampling | |
CN116894225B (en) | Driving behavior abnormality analysis method, device, equipment and medium thereof | |
Wang et al. | Identification of High Emission Mobile Sources Based on Self-supervised Representation Network | |
CN116628395B (en) | Urban terrain feature-based road vehicle carbon emission measuring and calculating method and system | |
Al-Saadi et al. | Improvement of the aerodynamic behavior of a sport utility vehicle numerically by using some modifications and aerodynamic devices | |
CN114861516B (en) | Vehicle dynamic performance determining method, device and storage medium | |
Miller et al. | A Deterministic Multivariate Clustering Method for Drive Cycle Generation from In-Use Vehicle Data | |
Geng | Computer Modeling: The Future of the Car Industry |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |