CN114547963A - Tire modeling method and medium based on data driving - Google Patents
Tire modeling method and medium based on data driving Download PDFInfo
- Publication number
- CN114547963A CN114547963A CN202111424322.5A CN202111424322A CN114547963A CN 114547963 A CN114547963 A CN 114547963A CN 202111424322 A CN202111424322 A CN 202111424322A CN 114547963 A CN114547963 A CN 114547963A
- Authority
- CN
- China
- Prior art keywords
- data
- tire
- model
- driven
- modeling method
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000002474 experimental method Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000012360 testing method Methods 0.000 claims abstract description 8
- 238000012795 verification Methods 0.000 claims abstract description 4
- 230000008569 process Effects 0.000 claims description 13
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 238000005096 rolling process Methods 0.000 claims description 4
- 238000005299 abrasion Methods 0.000 claims description 3
- 230000007613 environmental effect Effects 0.000 claims description 2
- 230000009466 transformation Effects 0.000 claims 1
- 230000006870 function Effects 0.000 abstract description 11
- 238000013135 deep learning Methods 0.000 abstract description 2
- 238000013461 design Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000013213 extrapolation Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000005312 nonlinear dynamic Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010561 standard procedure 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]
- G06F30/20—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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- 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
- 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
- G06N3/045—Combinations of networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Automation & Control Theory (AREA)
- Aviation & Aerospace Engineering (AREA)
- Computing Systems (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Tires In General (AREA)
Abstract
The invention discloses a tire modeling method based on data driving, which specifically comprises the following steps: determining input variables and output variables of tire modeling; carrying out a tire experiment, and acquiring input data and output data of the tire on different roads and under different working conditions; processing the acquired input data and the acquired output data, constructing a data set for tire modeling, and dividing a database into a training sample and a verification sample; selecting a proper data driving model, determining a driving model structure and a driving model method, initializing, training the tire model by using a training sample, and testing and verifying the model by using a verifying sample. The method of the invention is based on experimental data, utilizes deep learning to train Koopman operator and search high-dimensional function space at the same time, updates dictionary set and automatically carries out tire modeling. The tire modeling method can be used for processing dynamic problems, improves tire modeling precision, has interpretability, and has practical and practical value and wide application prospect.
Description
Technical Field
The present invention relates to tire modeling, and more particularly, to a data-driven tire modeling method and medium.
Background
The force of the tire against the road surface, as the only part that contacts the road surface, is of great importance to the control and driving safety of the vehicle system. The tire has nonlinear mechanical characteristics, a complex mapping relation exists, and the establishment of an accurate tire mechanical model is always the focus of attention in the field of vehicle system dynamics.
The existing tire modeling method can be mainly divided into three categories, namely a physical model, an empirical model, a semi-empirical model and the like, is built by fitting experimental data based on a mechanical mechanism, and is mostly a steady-state model; the model driving method based on the mechanism has the disadvantages of complex form and difficult parameter multi-fitting, and causes low precision and efficiency of model construction and poor extrapolation property; most of the models are steady-state models, which are not enough to deal with dynamic problems, and therefore, the design of the control system of the vehicle is hindered.
The data-driven modeling does not need the complex relation in the tire, and the model between the input and the output can be established through the training of the acquired experimental data, so that the experimental test result is infinitely approximated. The deep learning is taken as the heat tide of the current research, has the capability of approximating various nonlinear complex functions, and a single neural network has the defect of inexplicability in modeling; at present, koopman operators are particularly powerful in inferring the characteristics of dynamic systems that are partially or completely unknown or too complex to analyze using standard methods; similarly, a dictionary base is established, a nonlinear dynamics sparse representation method is adopted, and the model is simple in structure and easy to explain. Therefore, it is important to find a method for data-driven modeling with simple calculation process and interpretability.
Disclosure of Invention
The purpose of the invention is as follows: aiming at the defects of the prior art, the tire modeling method and medium based on data driving are provided, the accuracy of the model is improved, the calculation complexity is reduced, the interpretable data driving modeling is realized, and convenience is provided for the design of a vehicle control system.
The technical scheme is as follows: a data-driven tire modeling method, comprising:
step 1: determining input variables and output variables of tire modeling;
step 2: carrying out a tire experiment, and acquiring input data and output data of the tire on different roads and under different working conditions;
and step 3: processing the acquired input data and the acquired output data, constructing a data set for tire modeling, and dividing a database into a training sample and a verification sample;
and 4, step 4: selecting a proper data driving model, determining a driving model structure and a driving model method, initializing, training the tire model by using a training sample, and testing and verifying the model by using a verifying sample.
And 5: and establishing a tire model based on data driving.
Further, in step 1, the input variables of the tire model include a slip rate, a yaw angle, a radial deformation amount, a roll angle, a wheel speed, and a yaw angle, and the output variables include a longitudinal force, a lateral force, a normal force, a roll moment, a rolling resistance moment, and a aligning moment.
Further, in the step 2, the same model of tire with a known radius is adopted for the tire experiment, a sensor is installed, and a complete information acquisition system is configured; in the data acquisition process, enough sampling frequency and time are ensured, and accurate measurement of input and output data is ensured.
Specifically, the tire experiment included:
step 2.1: selecting an input variable;
step 2.2: the tires with the same abrasion loss move relative to the road surface and keep relatively stable under the condition of consistent environmental factors;
step 2.3: changing an input variable, and recording corresponding output data in a changing process;
step 2.4: and (4) replacing the variables in the step (1), and repeating the step (2.1) to the step (2.3) until all the variables are traversed and the data acquisition is finished.
Further, in step 2, different road surfaces comprise a dry road surface and a wet road surface, and different working conditions comprise high-speed steering, braking and accelerating.
Further, in step 3, the collected input and output data are processed into data normalization, and the magnitude order is unified.
Further, the normalization process is to perform linear change on the original data and map the measured data to [0,1]The treatment process is as follows:in the formula, x0For pre-processing data, xmaxFor pre-processing data maximum, xminFor pre-processing data minimum, x1The output data is similarly processed for the processed variables, denoted as y1。
Further, in step 4, a deep neural network DNN is adopted to automatically and effectively generate a dictionary Ng(xk) Simultaneously traversing the function space, mapping the dictionary to a high-dimensional space in which a relational expression KN existsg(xk)=Ng(xk+1) And calculating an error loss function to obtain accurate estimation of the koopman operator K, automatically updating a dictionary set at the same time, approximating experimental data, and establishing a data-driven tire model.
Further, the data driving tire model is an unsteady state model and is used for processing the dynamic working conditions of the tire.
A computer readable storage medium comprising one or more programs for execution by one or more processors, the one or more programs including instructions for performing any of the methods described above.
Has the advantages that: compared with the prior art, the invention has the following remarkable advantages: the invention takes the tire experimental data as the basis, adopts a data-driven modeling method, and solves the problems of complex mechanism model, difficult fitting of statistical data, low model precision, poor extrapolation property and the like; by utilizing the deep neural network DNN, a dictionary set is automatically and effectively generated, and the subjectivity of a traditional manual dictionary is avoided; the tire experimental process relates to input and output data under unsteady state motion under different road surfaces, can train out a plurality of unsteady state models, is convenient for handle dynamic problem, provides convenience for vehicle control system design.
Drawings
FIG. 1 is a schematic flow chart of data driven tire modeling in accordance with the present invention.
FIG. 2 is a diagram illustrating an example of a process of acquiring experimental data of a tire according to the present invention.
FIG. 3 is a data-driven model structure and flow chart of the present invention.
FIG. 4 is a diagram of a DNN structure in the data-driven model of the present invention.
Detailed Description
The technical scheme of the invention is further explained by combining the attached drawings.
As shown in fig. 1, the present embodiment provides a data-driven-based tire modeling method, including the steps of:
step 1: determining input variables and output variables of tire modeling according to the actual tire modeling problem;
step 2: carrying out a tire experiment, and collecting input and output data of the tire under different working conditions on different road surfaces;
and step 3: processing the acquired input data and the acquired output data, constructing a data set for tire modeling, and dividing a database into a training sample and a verification sample;
and 4, step 4: selecting a proper data driving model, determining a driving model structure and a driving model method, initializing, training the tire model by using a training sample, and testing and verifying the model by using a verifying sample.
And 5: and establishing a tire model based on data driving.
As shown in FIG. 2, the main features of the tire are various working conditions based on the prior dynamicsInfluencing factors, the invention selects the input parameters of the model for tire modeling as the vertical load FzThe slip rate lambda, the slip angle alpha, the radial deformation r, the roll angle gamma, the wheel speed v and the yaw angle beta, and the output parameters are longitudinal force FxLateral force FyNormal force FzRoll moment MxRolling resistance moment MySum and return moment Mz. The variables have mutual influence, and a single variable is not only used for representing a certain working condition.
As shown in fig. 3, the tire experiment data acquisition in step 2 includes the following steps:
step 2.1: selecting one of the input variables;
step 2.2: the tires with the same abrasion amount move relative to the road surface under the condition that other factors are consistent;
step 2.3: the controller changes the variable value and records the corresponding output data;
step 2.4, changing the working condition and recording corresponding output data;
step 2.5: and (3) selecting different variables from the step 2.1, and repeating the steps until all the variables are traversed and the data acquisition is finished. In the embodiment, a high-rotation-speed tire sharp-turning working condition is selected, the initial speed of the tire is 60km/h, the controller changes the tire slip angle alpha, the condition that other input variables are changed simultaneously exists at the moment, the sensor group measures the input and output variables, the tire slip angle alpha is changed every time, the sampling time is recorded as 10s, each tire is specified to rotate for one circle, at least 32 points are collected, and the collected input data is recorded as x0,N is the number of collected samples, and the output data is recorded as y0,The data is recorded by the data acquisition unit and is input into the PC in real time for recording and storage, and the data is used as the source of the following data set.
In the step 2, the same model of tire with known radius is adopted for the experiment, and a sensor for acquiring related parameters is installed, so that the input data and the output data can be measured.
Processing the collected input and output data into data normalization, wherein the normalization processing is to perform linear change on the original data and map the measured data to [0,1 ]]The treatment process is as follows:in the formula x0For pre-processing data, xmax,xminMaximum and minimum values of data before processing, x1The output data is similarly processed for the processed variables, denoted y1。
For a tire model, a data set is givenxiTo input, yiFor output, N is the number of samples, taking m training samples { (x)1,y1),(x2,y2),...,(xm,ym) Find the following function: y isi=f(xi) The function f is the model to be built; such that the test sample { (x)m+1,ym+1),(xm+2,ym+2),...,(xN,yN) As much as ym+i=f(xm+i) N-m, 70% of the model training samples are selected as the model training samples, and the rest 30% are used as the test samples.
As shown in fig. 3, the data-driven model is built as follows: discretizing a dataset of training samples to a representation xk=x(kΔt),ykY (k Δ t), where k is 1, 2. Training sample input vertical load FzThe slip rate lambda, the slip angle alpha, the radial deformation r, the roll angle gamma, the wheel speed v and the yaw angle beta are used as control input, and the control input is uk=xkTire output variable longitudinal force FxLateral force FyNormal force FzRoll moment MxRolling resistance moment MySum and return moment MzOne or more of which are used as state variables, and outputs in the process of establishing the modelThe data set is output as a state variable, and in the state space, x is enabledk=yk(ii) a Using dictionary sets as the output of a deep neural network, i.e. DNN=Ng(xk),Ng() As a combinatorial function of the neurons, ukAnd xkLinear combination in high dimensional space, using DNN model to combine xk、xk+1、ukRespectively mapped to a high dimensional space g (x)k)、g(xk+1),g(uk) The function of g () is a space function, g (x)k+1) And g (u)k)、g(xk) A linear relation exists among the three, and the structure of the deep neural network is shown in figure 4;
the dictionary set is expressed as Representing a mapping to a state variable g (x) in a high dimensional spacek)=(Fxk,Fyk,Fzk,Mxk,Myk,Mzk) And control input g (u)k)=(Fzk,λk,αk,rk,γk,vk,βk) A set of all possible combinations as mapping bases;
presence of KN in high dimensional spaceg(xk)=Ng(xk+1) Calculating an error loss function:
in the formula of1、λ2Is the sparse parameter of the dictionary set, | | | | non-woven phosphor2、|| ||1Respectively representing 2-norm and 1-norm, theta is a parameter of the deep neural network, K is a required koopman operator, the koopman operator is updated according to a loss function, and meanwhile, the dictionary is automatically perfectedComplete modelingThe koopman problem will now translate to:
in the formula, K is the required koopman operator, and in the scheme, K is a matrix; and finally, initializing a neural network, obtaining a random parameterized dictionary, completing the process, verifying the reliability of the model by using the test sample, and establishing a data-driven tire model.
Embodiments of the present invention may be provided as methods or computer program products and, thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The embodiments of the present invention are not described in detail, but are known in the art, and can be implemented by referring to the known techniques.
Claims (10)
1. A data-driven tire modeling method, comprising:
step 1: determining input variables and output variables of tire modeling;
step 2: carrying out a tire experiment, and acquiring input data and output data of the tire on different roads and under different working conditions;
and step 3: processing the acquired input data and the acquired output data, constructing a data set for tire modeling, and dividing a database into a training sample and a verification sample;
and 4, step 4: selecting a proper data driving model, determining a driving model structure and a driving model method, initializing, training the tire model by using a training sample, and testing and verifying the model by using a verifying sample.
And 5: and establishing a tire model based on data driving.
2. The data-driven-based tire modeling method as claimed in claim 1, wherein in step 1, the input variables of the tire model include slip rate, slip angle, radial deformation amount, roll angle, wheel speed, and yaw angle, and the output variables include longitudinal force, lateral force, normal force, roll moment, rolling resistance moment, and aligning moment.
3. The data-driven tire modeling method as claimed in claim 1, wherein in step 2, the same type of tire with a known radius is used for tire experiment, the sensor is installed, and the complete information acquisition system is configured.
4. The data-driven-based tire modeling method as claimed in claim 3, wherein said tire experiment specifically comprises:
step 2.1: selecting an input variable;
step 2.2: the tires with the same abrasion loss move relative to the road surface and keep relatively stable under the condition of consistent environmental factors;
step 2.3: changing an input variable, and recording corresponding output data in a changing process;
step 2.4: and (4) replacing the variables in the step 1, and repeating the step 2.1 to the step 2.3 until all the variables are traversed and the data acquisition is finished.
5. The data-driven-based tire modeling method as claimed in claim 1, wherein in step 2, different road surfaces comprise a dry road surface and a wet road surface, and different working conditions comprise high-speed steering, braking and accelerating.
6. The data driving-based tire modeling method according to claim 1, wherein in step 3, the collected input and output data are normalized by a uniform order of magnitude.
7. The data-driven tire modeling method as claimed in claim 6, wherein said normalization process is a linear transformation of the raw data, mapping the measured data to [0,1 ]]The treatment process is as follows:in the formula, x0For pre-processing data, xmaxAs maximum value of data before processing, xminFor pre-processing data minimum, x1The output data is similarly processed for the processed variables, denoted as y1。
8. The data-driven-based tire modeling method as claimed in claim 1, wherein in step 4, a Deep Neural Network (DNN) is used to automatically and efficiently generate a dictionary, and a function space is traversed to map the dictionary to a high-dimensional space, so that accurate estimation of koopman operators is obtained in the high-dimensional space, experimental data is approximated, and a data-driven tire model is built.
9. The data-driven-based tire modeling method of claim 8, wherein said data-driven tire model is an unsteady state model for dealing with tire dynamic conditions.
10. A computer readable storage medium comprising one or more programs for execution by one or more processors, the one or more programs comprising instructions for performing the method of any of claims 1-9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111424322.5A CN114547963B (en) | 2021-11-26 | 2021-11-26 | Tire modeling method and medium based on data driving |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111424322.5A CN114547963B (en) | 2021-11-26 | 2021-11-26 | Tire modeling method and medium based on data driving |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114547963A true CN114547963A (en) | 2022-05-27 |
CN114547963B CN114547963B (en) | 2024-06-25 |
Family
ID=81669972
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111424322.5A Active CN114547963B (en) | 2021-11-26 | 2021-11-26 | Tire modeling method and medium based on data driving |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114547963B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115600480A (en) * | 2022-06-13 | 2023-01-13 | 哈尔滨工业大学(Cn) | Global linearization frequency dispersion hysteresis modeling method and device for piezoelectric transducer, computer equipment and storage medium |
WO2024082213A1 (en) * | 2022-10-20 | 2024-04-25 | 华为技术有限公司 | Method and apparatus for constructing vehicle dynamics model, device, and storage medium |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080059134A1 (en) * | 2006-08-22 | 2008-03-06 | The Yokohama Rubber Co., Ltd. | Tire characteristic calculation method, tire dynamic element parameter value derivation method, vehicle traveling simulation method, and tire designing method and vehicle designing method in which consideration is given to tire friction ellipse |
CN104236928A (en) * | 2014-05-09 | 2014-12-24 | 吉林大学 | Method for testing tire steady-state laterodeviation mechanical properties |
CN104517039A (en) * | 2014-12-30 | 2015-04-15 | 吉林大学 | Tire side-tipping side-inclining steady-state aligning torque characteristic radius semi-empirical modeling method |
CN106997581A (en) * | 2017-03-01 | 2017-08-01 | 杭州电子科技大学 | A kind of method that utilization deep learning rebuilds high spectrum image |
US20210012199A1 (en) * | 2019-07-04 | 2021-01-14 | Zhejiang University | Address information feature extraction method based on deep neural network model |
CN112911530A (en) * | 2020-12-09 | 2021-06-04 | 广西电网有限责任公司电力科学研究院 | Method for establishing small and micro intelligent sensor network congestion identification model |
CN113408047A (en) * | 2021-08-05 | 2021-09-17 | 江苏大学 | Vehicle dynamics prediction model based on time-lag feedback neural network, training data acquisition method and training method |
-
2021
- 2021-11-26 CN CN202111424322.5A patent/CN114547963B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080059134A1 (en) * | 2006-08-22 | 2008-03-06 | The Yokohama Rubber Co., Ltd. | Tire characteristic calculation method, tire dynamic element parameter value derivation method, vehicle traveling simulation method, and tire designing method and vehicle designing method in which consideration is given to tire friction ellipse |
CN104236928A (en) * | 2014-05-09 | 2014-12-24 | 吉林大学 | Method for testing tire steady-state laterodeviation mechanical properties |
CN104517039A (en) * | 2014-12-30 | 2015-04-15 | 吉林大学 | Tire side-tipping side-inclining steady-state aligning torque characteristic radius semi-empirical modeling method |
CN106997581A (en) * | 2017-03-01 | 2017-08-01 | 杭州电子科技大学 | A kind of method that utilization deep learning rebuilds high spectrum image |
US20210012199A1 (en) * | 2019-07-04 | 2021-01-14 | Zhejiang University | Address information feature extraction method based on deep neural network model |
CN112911530A (en) * | 2020-12-09 | 2021-06-04 | 广西电网有限责任公司电力科学研究院 | Method for establishing small and micro intelligent sensor network congestion identification model |
CN113408047A (en) * | 2021-08-05 | 2021-09-17 | 江苏大学 | Vehicle dynamics prediction model based on time-lag feedback neural network, training data acquisition method and training method |
Non-Patent Citations (1)
Title |
---|
肖婷: "智能车在复杂环境下的路径规划与跟随控制算法研究", 《硕士电子期刊》, 30 November 2019 (2019-11-30), pages 1 - 83 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115600480A (en) * | 2022-06-13 | 2023-01-13 | 哈尔滨工业大学(Cn) | Global linearization frequency dispersion hysteresis modeling method and device for piezoelectric transducer, computer equipment and storage medium |
WO2024082213A1 (en) * | 2022-10-20 | 2024-04-25 | 华为技术有限公司 | Method and apparatus for constructing vehicle dynamics model, device, and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN114547963B (en) | 2024-06-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ravikumar et al. | Gearbox fault diagnosis based on Multi-Scale deep residual learning and stacked LSTM model | |
CN109765053B (en) | Rolling bearing fault diagnosis method using convolutional neural network and kurtosis index | |
Wu et al. | Modeling and analysis of tool wear prediction based on SVD and BiLSTM | |
CN114547963B (en) | Tire modeling method and medium based on data driving | |
CN109255094B (en) | Commercial truck quality estimation method based on SVR-Adaboost improved algorithm | |
CN112964469B (en) | Online fault diagnosis method for rolling bearing under variable load of transfer learning | |
CN112801143B (en) | Steering road feel simulation method based on K-Means and Gaussian process regression | |
CN112613536B (en) | Near infrared spectrum diesel fuel brand recognition method based on SMOTE and deep learning | |
CN114169110B (en) | Motor bearing fault diagnosis method based on feature optimization and GWAA-XGboost | |
CN116448419A (en) | Zero sample bearing fault diagnosis method based on depth model high-dimensional parameter multi-target efficient optimization | |
CN111126438B (en) | Driving behavior recognition method and system | |
CN112298155B (en) | Method for predicting energy consumption of hybrid power truck based on variable time domain model | |
CN110991471B (en) | Fault diagnosis method for high-speed train traction system | |
CN113268833A (en) | Migration fault diagnosis method based on deep joint distribution alignment | |
CN112541510A (en) | Intelligent fault diagnosis method based on multi-channel time series data | |
CN115859077A (en) | Multi-feature fusion motor small sample fault diagnosis method under variable working conditions | |
CN116341099A (en) | Method for constructing LSTM neural network vehicle suspension system state observer based on attention mechanism | |
CN112985830A (en) | Abs result automatic judging algorithm | |
CN114781245B (en) | Transient automobile fuel consumption estimation method and system based on fuel consumption strong correlation parameter correction | |
CN117131767A (en) | Method for predicting thermal convexity of working roll of hot rolling four-high mill based on random forest algorithm | |
CN112632706B (en) | Road feel simulation method based on GMM and CART regression tree | |
CN112632705B (en) | Road feel simulation method based on GMM and Gaussian process regression | |
CN113743010A (en) | Rolling bearing running state evaluation method based on EEMD energy entropy | |
CN116304551A (en) | Motor bearing fault diagnosis and feature extraction method based on BCB model | |
CN113033683B (en) | Industrial system working condition monitoring method and system based on static and dynamic joint analysis |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |