CN116415494A - Road noise optimization method, system and storage medium - Google Patents
Road noise optimization method, system and storage medium Download PDFInfo
- Publication number
- CN116415494A CN116415494A CN202310330213.XA CN202310330213A CN116415494A CN 116415494 A CN116415494 A CN 116415494A CN 202310330213 A CN202310330213 A CN 202310330213A CN 116415494 A CN116415494 A CN 116415494A
- Authority
- CN
- China
- Prior art keywords
- road noise
- prediction model
- decomposition
- optimization
- level
- 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
- 238000005457 optimization Methods 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 81
- 238000013461 design Methods 0.000 claims abstract description 39
- 230000005540 biological transmission Effects 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims description 44
- 238000004422 calculation algorithm Methods 0.000 claims description 37
- 239000000725 suspension Substances 0.000 claims description 35
- 238000012360 testing method Methods 0.000 claims description 34
- 238000013527 convolutional neural network Methods 0.000 claims description 17
- 230000004044 response Effects 0.000 claims description 17
- 238000010206 sensitivity analysis Methods 0.000 claims description 17
- 230000005284 excitation Effects 0.000 claims description 16
- 230000035945 sensitivity Effects 0.000 claims description 11
- 230000002068 genetic effect Effects 0.000 claims description 10
- 238000013523 data management Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 4
- 238000011161 development Methods 0.000 abstract description 9
- 238000012937 correction Methods 0.000 abstract description 5
- 230000006872 improvement Effects 0.000 abstract description 4
- 230000001133 acceleration Effects 0.000 description 40
- 230000006870 function Effects 0.000 description 18
- 238000004088 simulation Methods 0.000 description 17
- 238000004458 analytical method Methods 0.000 description 15
- 238000004364 calculation method Methods 0.000 description 12
- 238000005259 measurement Methods 0.000 description 10
- 239000006096 absorbing agent Substances 0.000 description 9
- 230000035939 shock Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 7
- 238000013016 damping Methods 0.000 description 6
- 238000013507 mapping Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 238000012800 visualization Methods 0.000 description 3
- 208000009205 Tinnitus Diseases 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 239000010426 asphalt Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 210000005069 ears Anatomy 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000013433 optimization analysis Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 230000000007 visual effect Effects 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
- 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
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- 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
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/10—Noise analysis or noise optimisation
-
- 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)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Computer Hardware Design (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Computational Mathematics (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Mathematical Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention relates to the field of NVH performance of automobiles, in particular to a road noise optimization method, a system and a storage medium, which comprise the following steps: s1, decomposing the road noise of the whole vehicle to chassis parts layer by layer along a vibration transmission path to obtain a plurality of decomposition parameters, constructing a road noise performance level decomposition framework of the whole vehicle, and establishing a road noise prediction model which is input as the decomposition parameters and output as noise in the vehicle based on the road noise multi-layer decomposition framework; s2, determining a value range of the decomposition parameters by taking the decomposition parameters as design variables, setting an optimization target and constraints, carrying out optimizing solution based on a road noise prediction model, and calculating to obtain an optimal solution of the decomposition parameters. The method can efficiently and accurately realize the prediction and optimization of the structure sound in the vehicle, combines the product performance positioning, provides a menu type combination scheme for the design of NVH performance parameters of the chassis of the product, guides the forward development and problem correction of the product, and realizes the improvement of the road noise performance.
Description
Technical Field
The invention relates to the field of NVH performance of automobiles, in particular to a road noise optimization method, a road noise optimization system and a storage medium.
Background
Road noise is one of the main factors influencing NVH performance of automobiles, and with the rapid development of new energy automobiles, the noise of the traditional engine is obviously reduced, and the masking effect of power transmission noise is weakened, so that the road noise is more prominent. Road noise often causes complaints such as "rattle", "rumble", "tire cavity" and the like, which are concentrated in the frequency range of 20 to 300Hz, and are mainly represented by structural path transfer. Road noise is used as an important component of noise in a vehicle, so that the improvement of the road noise can effectively improve the market competitiveness of a product, and the research on the prediction and optimization control method of the road noise of the whole vehicle has important engineering significance.
The method widely applied at present is to establish a whole vehicle model by a CAE (Computer Aided Engineering) simulation method to develop road noise prediction. The CAE simulation prediction method for the vibration noise of the whole vehicle pavement disclosed by CN201610415861.5 establishes a whole vehicle finite element simulation model comprising a chassis system and a vehicle body system, loads the wheel center load to the wheel center of the finite element simulation model, can more comprehensively analyze the vibration and the noise caused by the excitation of the whole vehicle loaded wheel center load pavement so as to facilitate the later rectification and the improvement of the NVH performance of the vehicle, and comprises the following steps: a: testing the steering knuckle acceleration of the sample car to obtain an acceleration signal matrix Ga of the steering knuckle; b: obtaining a transfer function Hs from the wheel center to the knuckle by using a finite element method or an experimental test method; c: calculating the wheel center load Gf based on a wheel center load theoretical formula gf=hs+gahs+h according to a transfer function Hs from the wheel center to the knuckle and an acceleration signal matrix Ga of the knuckle, wherein the symbol+ represents a pseudo-inverse and H represents a conjugate transpose; d: and (3) establishing a vehicle NVH finite element simulation model, loading the wheel center load Gf to the wheel center of the finite element simulation model, and calculating the vibration and noise curve in the vehicle. The vibration and noise in the vehicle are calculated and analyzed. The suspension system multidisciplinary optimization design method based on brake shake and road noise performance disclosed in CN201911017187.5 establishes a whole vehicle finite element simulation model comprising a modal tire system, a chassis system, a power system and an interior vehicle body system, converts acquired road spectrum PSD (power spectral density) data into tire displacement excitation and applies the tire displacement excitation to the whole vehicle model to develop road noise performance simulation analysis, and specifically comprises the following steps: step 1, determining relevant parameters of multidisciplinary optimization design of a suspension system; step 2, a brake shake simulation analysis model and a road noise simulation model are established and simulation analysis is carried out; step 3, determining a plurality of design variables of the suspension system to be optimally designed, and carrying out parameterized modeling on each design variable; step 4, DOE sampling calculation of each design variable is carried out; step 5, extracting DOE sample points and calculation results, constructing a response surface approximation model I meeting the precision requirement based on brake shake performance, and constructing a response surface approximation model II meeting the precision requirement based on road noise performance; and 6, performing multidisciplinary optimization design on each design variable based on the two approximate models in the step 5, and obtaining an optimization scheme.
Because the road noise relates to more parts, the road noise level is influenced by a plurality of parameters such as road surface excitation, tire rigidity, bushing rigidity, shock absorber damping characteristics, vehicle body dynamic rigidity and the like, the nonlinear characteristics are stronger, and the formation mechanism is complex. If a complete car road noise model is established through traditional CAE simulation, nonlinear characteristic expression of parts such as tires, bushings, shock absorbers and the like is involved, the reality problem that many parameters are difficult to obtain exists, model accuracy is difficult to guarantee, and even if a complete car CAE refined model is established, the problems of low efficiency, high cost and the like exist.
Along with the development of big data technology and the continuous accumulation of simulation and test data in the development process of automobile factories, road noise analysis by adopting a data mining method is possible. The method is characterized in that the method is an effective method, based on a mathematical statistics method, the corresponding relation between an input variable and a response is obtained by fitting by using historical sample data, the method passes through a complex finite element model, an approximate model is built, and on the premise that the accuracy of the approximate model is effectively ensured, the analysis result has enough referential property. The optimization process based on the approximate model has the advantages that the calculation and analysis of a complex finite element model are avoided, the time required by the real iterative calculation process is greatly reduced, the calculation amount is small, the calculation period is short, the optimization efficiency is greatly improved, and meanwhile, the requirements on computer hardware are also relieved.
Disclosure of Invention
The invention aims to provide a road noise optimization method, a road noise optimization system and a storage medium, which can efficiently and accurately realize prediction and optimization of in-vehicle structural sound, and provide a menu type combination scheme for design of NVH performance parameters of a product chassis by combining product performance positioning, guide forward development and problem correction of the product, and realize improvement of road noise performance.
In order to achieve the above object, the present invention adopts the following technical scheme.
In a first aspect, the present invention provides a road noise optimization method, which includes the following steps:
s1, decomposing the road noise of the whole vehicle to chassis parts layer by layer along a vibration transmission path to obtain a plurality of decomposition parameters, constructing a road noise performance level decomposition framework of the whole vehicle, and establishing a road noise prediction model which is input as the decomposition parameters and output as noise in the vehicle based on the road noise multi-layer decomposition framework;
s2, determining a value range of the decomposition parameters by taking the decomposition parameters as design variables, setting an optimization target and constraints, carrying out optimizing solution based on a road noise prediction model, and calculating to obtain an optimal solution of the decomposition parameters.
Further, the overall vehicle road noise performance level decomposition architecture in S1 includes a first level, a second level, a third level and a fourth level, where the first level is in-vehicle noise, the second level is a passive side vibration response of a suspension and a vehicle body attachment point, the third level is an active side vibration response of the suspension and the vehicle body attachment point, and the fourth level is vibration excitation from a knuckle and a dynamic parameter of a spherical hinge bushing.
Further, the construction of the road noise prediction model in S1 specifically includes the following steps:
s11, acquiring training sample data based on the constructed whole vehicle road noise performance hierarchical decomposition architecture;
and S12, training the whole vehicle road noise prediction model by using training sample data to obtain a trained road noise prediction model meeting the precision requirement.
Further, S12 is specifically: the training sample data acquired in the step S11 are divided into a training set and a testing set, the training set is input into a road noise prediction model to train the model, and a trained whole road noise prediction model is obtained; and then inputting training sample data of the test set into the trained road noise prediction model for verification and evaluation, if the prediction precision reaches the design requirement, the test model is proved to pass, otherwise, the road noise prediction model is reconstructed.
Further, sensitivity analysis is performed based on the road noise prediction model by adopting a sensitivity analysis method, cost and performance influence are comprehensively considered, at least one of a plurality of decomposition parameters is determined as a design variable, and the value range of the design variable is determined.
Further, the road noise prediction model in the S1 is a mixed model of a support vector regression algorithm and a convolutional neural network algorithm.
And further, in S2, optimizing based on the road noise prediction model by adopting a genetic algorithm, and calculating to obtain an optimal value of the decomposition parameter.
Further, after the optimal solution of the decomposition parameters is obtained in the step S2, the optimal solution can be expanded into an optimal section through the interval search of the design variable, whether the optimization target is achieved is judged, if not, the constraint range of the optimization target is adjusted, or the design variable and the value range are readjusted until the optimization target is achieved.
Further, before S2, the road noise prediction model, the sensitivity analysis method, the genetic algorithm and the interval optimization algorithm are integrated and packaged, and a road noise optimization platform and software which are convenient for engineering operation are developed and obtained.
In a second aspect, the present invention provides a road noise optimization system capable of performing the steps of the road noise optimization method according to the present invention, including: the road noise prediction model module is used for establishing a road noise prediction model which is input into decomposition parameters and output into noise in the vehicle based on the road noise multi-layer decomposition architecture; and the optimization solving module is used for carrying out optimization solving based on the road noise prediction model and calculating to obtain an optimal solution of the decomposition parameters.
The system further comprises a data management module and a sensitivity module, wherein the data management module is used for collecting road noise data and generating training sample data based on the collected road noise data, and the sensitivity module is used for performing sensitivity analysis based on a road noise prediction model.
In a third aspect, the present invention provides a storage medium having stored therein a computer readable program which when invoked is capable of performing the steps of the road noise optimization method according to the present invention.
The invention has the beneficial effects of.
1. According to the invention, the in-vehicle vibration noise performance, namely the whole vehicle road noise is decomposed to chassis parts layer by layer along a vibration transmission path to obtain a plurality of decomposition parameters, and a whole vehicle road noise performance hierarchical decomposition framework is established on the basis, so that the vehicle physical model is replaced for subsequent analysis. On the basis, a road noise prediction model based on a support vector regression algorithm and a convolutional neural network algorithm is established, accurate prediction from suspension chassis parameters to the noise of the right ear of a driver is realized, and model accuracy is high. The road noise prediction model can also be independently learned in the process of continuously expanding data, so that the accuracy and generalization capability of the road noise prediction model are further improved. And then, global optimization based on the road noise level decomposition architecture from top to bottom is realized through a genetic algorithm and an interval optimization algorithm, so that an optimized interval of each component parameter of the bottom level is obtained, and the forward development design of engineers is guided.
2. The invention uses the road noise prediction model to carry out chassis parameter sensitivity analysis, identifies the high-sensitivity chassis parameters and the main transmission path of noise vibration in the vehicle, and has important guiding function for road noise problem optimization in engineering practice. In addition, aiming at the problem of late road noise correction, the sensitivity analysis result can be combined to optimize one or more paths with higher sensitivity, so that the problem correction efficiency is improved, and the test cost is reduced.
3. The road noise prediction model, the sensitivity analysis method, the genetic algorithm and the interval optimization algorithm are integrated and packaged, the road noise optimization platform and software which are convenient for engineering operation are developed, the product design and development efficiency is improved, the utilization rate of enterprise test data is effectively improved, and reliable guidance is provided for forward design and subsequent rectification of the product.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the description of the embodiments or the prior art will be briefly introduced below, it being obvious that the drawings in the description below are only some examples of the present invention.
FIG. 1 is a flow chart of a road noise optimization method according to the invention;
FIG. 2 is a schematic diagram of a road noise structure sound transmission path of a certain vehicle type;
FIG. 3 is a schematic representation of a Macpherson front suspension hierarchy decomposition in a road noise multilayer decomposition architecture;
FIG. 4 is a graph of chassis parameters versus road noise sensitivity analysis results;
FIG. 5 is a schematic diagram of a road noise optimization system according to the present invention;
FIG. 6 is a schematic diagram of a software interface of the road noise parameterized development platform.
Detailed Description
Further advantages and effects of the present invention will become readily apparent to those skilled in the art from the disclosure herein, by referring to the accompanying drawings and the preferred embodiments. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be understood that the preferred embodiments are presented by way of illustration only and not by way of limitation.
It should be noted that the illustrations provided in the following embodiments merely illustrate the basic concept of the present invention by way of illustration, and only the components related to the present invention are shown in the drawings and are not drawn according to the number, shape and size of the components in actual implementation, and the form, number and proportion of the components in actual implementation may be arbitrarily changed, and the layout of the components may be more complicated.
In a first embodiment, a road noise optimization method includes the following steps:
s1, decomposing the road noise of the whole vehicle layer by layer to chassis parts along a vibration transmission path to obtain a plurality of decomposition parameters, and constructing a road noise performance level decomposition framework of the whole vehicle. And establishing a road noise prediction model with the input as a decomposition parameter and the output as in-vehicle noise based on the road noise multi-layer decomposition architecture.
The road noise structure has a plurality of influencing factors, and is mainly related to parameters such as dynamic stiffness of the bushing, damping characteristics of the shock absorber and the like. In actual driving, the automobile system is a multiple-input multiple-output system, and the road surface excitation transmission process is simplified into: road surface excitation transmitted from the tire passes through the suspension system, vibration is transmitted to the vehicle body, and the vehicle body vibration and the in-vehicle acoustic cavity jointly act to generate structural radiation sound, and finally the structural radiation sound is transmitted to human ears. In the case of the front macpherson suspension and the rear multi-link suspension, the structural transmission path is shown in fig. 2.
The root of road noise is the (system, subsystem, component) parameter matching under each structure. Because road noise relates to a complex system of the whole vehicle, prediction analysis is difficult to carry out through an input-output double-level model, the invention provides a multi-level analysis method for NVH performance of the vehicle, and the road noise is subjected to level decomposition based on the NVH performance of a suspension. The road noise problem of the whole vehicle is decomposed into parameters of all parts of the chassis layer by layer along a vibration transmission path, the parameters with higher sensitivity are selected from a plurality of influence factors of road noise to serve as input features of a road noise prediction model, a road noise level decomposition system is constructed, and a road noise performance multi-level decomposition framework is built on the basis. The whole vehicle road noise performance level decomposition architecture comprises a first level, a second level, a third level and a fourth level, wherein the first level is in-vehicle noise, the second level is the passive side vibration response of a suspension and a vehicle body attachment point, the third level is the active side vibration response of the suspension and the vehicle body attachment point, and the fourth level is vibration excitation from a steering knuckle and dynamic parameters of a spherical hinge bush. The decomposition principle is as follows: the design objective (noise or vibration) of the upper level is decomposed down into excitation (vibration) of the adjacent lower level and the connecting element dynamics parameters (bushing stiffness, spring stiffness, shock absorber damping force).
Taking the macpherson front suspension and the multi-link rear suspension vehicle type as an example, the road noise performance of the vehicle type is distributed symmetrically in a left-right manner by a multi-level decomposition structure, see fig. 3, and the illustrated macpherson front suspension has a level decomposition structure, wherein the first level is noise in the vehicle and comprises right ear noise of a driver. The second level is the passive side vibration response of the attachment points of the suspension and the vehicle body, and comprises passive side vibration acceleration of the front mounting point of the front auxiliary frame, passive side vibration acceleration of the rear mounting point of the front auxiliary frame and passive side vibration acceleration of the upper mounting point of the front shock absorber. The third level is the active side vibration response of the suspension and the vehicle body attachment point, and comprises front swing arm front mounting point active side vibration acceleration and front swing arm bushing dynamic stiffness, wherein the front swing arm front mounting point active side vibration acceleration and the front swing arm bushing dynamic stiffness are decomposed downwards by front auxiliary frame front mounting point passive side vibration acceleration, front swing arm rear mounting point active side vibration acceleration and front swing arm rear bushing dynamic stiffness, front steering knuckle vibration acceleration and front shock absorber speed-damping, front shock absorber on-front support post mounting bushing dynamic stiffness and front spring stiffness. The fourth level is vibration excitation from the knuckle and dynamic parameters of the spherical hinge bush, and comprises front knuckle vibration acceleration, front swing arm and knuckle connecting spherical hinge bush dynamic stiffness, front knuckle vibration acceleration and knuckle connecting spherical hinge bush dynamic stiffness, and front knuckle vibration acceleration.
The hierarchical decomposition of the multi-link rear suspension specifically includes: the first level is in-vehicle noise, including driver right ear noise. The second level is the passive side vibration response of the suspension and the vehicle body attachment point, and comprises passive side vibration acceleration of a front bushing of a rear longitudinal arm, passive side vibration acceleration of a front mounting point of a rear auxiliary frame, passive side vibration acceleration of a rear mounting point of the rear auxiliary frame, passive side vibration acceleration of the inner side of a rear lower swing arm, passive side vibration acceleration of a rear spring and passive side vibration acceleration of a rear shock absorber.
The third level is the active side vibration response of the suspension and the vehicle body attachment point, and comprises a rear knuckle vibration acceleration and a rear trailing arm bushing dynamic stiffness, wherein the rear knuckle vibration acceleration and the rear trailing arm bushing dynamic stiffness are decomposed downwards, the rear transverse pull rod active side vibration acceleration and the rear transverse pull rod inner side bushing dynamic stiffness are decomposed downwards, the rear transverse control arm active side acceleration and the rear transverse control arm inner side bushing dynamic stiffness are decomposed downwards, the rear lower swing arm inner side active side vibration acceleration and the rear lower swing arm inner side bushing dynamic stiffness are decomposed downwards, the rear lower swing arm inner side active side vibration acceleration and the rear spring stiffness are decomposed downwards, the rear spring upper rubber gasket stiffness and the rear spring lower rubber gasket stiffness are decomposed downwards, the rear knuckle vibration acceleration and the rear damper dynamic stiffness are decomposed downwards, and the rear damper speed-damping speed is realized.
The fourth level is vibration excitation from the knuckle and dynamic parameters of the spherical hinge bushing, and comprises rear knuckle vibration acceleration with downward decomposed rear transverse pull rod driving side vibration acceleration, rear knuckle vibration acceleration with downward decomposed rear transverse pull rod and knuckle connecting bushing dynamic stiffness, rear transverse control arm and knuckle connecting bushing dynamic stiffness, rear knuckle vibration acceleration with downward decomposed rear lower swing arm inner side driving side vibration acceleration and rear lower swing arm and knuckle connecting bushing dynamic stiffness.
In this embodiment, the construction of the road noise prediction model specifically includes the following steps.
S11, acquiring training sample data based on the constructed whole vehicle road noise performance hierarchical decomposition architecture. The training sample data can be obtained through road tests, simulation analysis or combination of the road tests and the simulation analysis, and the like, and can be specifically selected according to actual conditions of projects. In the embodiment, noise and vibration data are acquired for the test vehicle in a road test mode.
Road surface excitation is transmitted to steering knuckles through four tires, and each steering knuckle is provided with a three-way acceleration sensor, and the total number of the steering knuckles is four. The suspension and the vehicle body end and the suspension end of the vehicle body attachment point are respectively provided with a three-way acceleration sensor. In order to measure the transmission parameters of vibration of each transmission path as much as possible, the sensor direction should be perpendicular to the whole vehicle coordinate system. The sound pressure sensor is arranged in the vehicle at the right ear of the driver.
After the sound pressure sensor and the acceleration sensor are arranged, setting data acquisition parameters: the vibration bandwidth is 0-2000 Hz, the noise bandwidth is 0-2000 Hz, and the frequency resolution is 1Hz. The test working condition is that the rough asphalt road runs at a constant speed, and the running speed is 60km/h. In order to ensure the reliability of the measurement result, the measurement is repeated at least three times, the sound pressure level difference of each measurement is not more than 0.5dB (A), if the difference of the measurement values exceeds 0.5dB (A), multiple tests are required, and the sound pressure level difference of three continuous measurement data is required to be controlled within a specified range. And finally, testing to obtain the sound pressure signals of the sound pressure sensor and the vibration signals of all the three-way acceleration sensors.
And S12, training the whole vehicle road noise prediction model by using training sample data to obtain a trained road noise prediction model meeting the precision requirement. In this embodiment, the road noise prediction model is a hybrid model combining a support vector regression algorithm and a convolutional neural network algorithm. The SVR algorithm is a machine learning method based on a statistical learning theory and a structural risk minimization principle, and the method maps multidimensional input to a feature space with higher dimensionality through a nonlinear kernel function and then carries out regression operation so as to obtain a nonlinear mapping relation between input features and output indexes. The support vector regression is based on the minimum error bound of all data, so that the support vector regression only needs a part of subsets in the training data, the calculation efficiency is high, and the accuracy is ensured.
The expression of the road noise prediction model is as follows:
wherein f (x) is structural acoustic road noise output by the road noise prediction model, and m is road noise predictionThe number of sample points, alpha, in the training sample dataset used by the model i And
is Lagrangian multiplier, k (x i ,x j ) As a kernel function, b is a bias value.
Since radial basis functions (Radial Basis Function, RBF) have high fitting accuracy among many problems, the present embodiment selects the RBF function as a kernel function, whose expression is: k (x) i ,x j )＝exp(-γx i -x j 2 ),γ＞0。
Wherein x is i Ith sample point data, x, used for training a road noise prediction model j And the j-th sample point data used for training the road noise prediction model is gamma which is the RBF kernel width.
Convolutional neural networks are used as a representative algorithm for deep learning, and have excellent performances in various fields such as image classification, target detection and the like. The convolutional neural network directly drives features by the data, and establishes a mapping relation between input features and output indexes through multidimensional nonlinear feature extraction, so that the convolutional neural network has extremely strong data characterization and mapping capability, and compared with the traditional shallow artificial neural network, the convolutional neural network model has higher prediction precision and robustness. The typical structure mainly comprises five layers, namely: input layer, convolution layer, pooling layer, full tie layer and output layer, in this embodiment, the output layer is SVR regression output layer. According to the invention, basic road noise data of the same suspension type vehicle type are obtained in a test mode, and then a road noise prediction model is built based on a road noise level decomposition architecture, an SVR algorithm and a convolutional neural network algorithm.
And (3) dividing the training sample data acquired in the step (S11) into a training set and a testing set, inputting the training set into a road noise prediction model, and performing parameter optimization training on the model to obtain a trained whole road noise prediction model. And then inputting training sample data of the test set into the trained road noise prediction model for verification and evaluation, if the prediction precision reaches the design requirement, the test model is proved to pass, otherwise, the road noise prediction model is reconstructed.
S2, determining a value range of the decomposition parameters by taking the decomposition parameters as design variables, setting an optimization target and constraints, carrying out optimizing solution based on a road noise prediction model by adopting a genetic algorithm, and calculating to obtain an optimal solution of the decomposition parameters. In engineering practice, the sound pressure level of noise in a vehicle may be defined as an optimization target, and the vibration level of components such as a vehicle seat and a steering wheel may be defined as an optimization target.
In this embodiment, in order to reduce the calculation amount, a sensitivity analysis method is adopted to perform sensitivity analysis based on a road noise prediction model, the cost and performance effects are comprehensively considered, at least one of a plurality of decomposition parameters is determined as a design variable, the value range of the design variable is determined, and optimization solution is not required for all the decomposition parameters. The front suspension frame is of a Macpherson structure, the rear suspension frame is of a road noise level decomposition structure built by a multi-link structure, a road noise prediction model is built by combining a support vector regression algorithm and a convolutional neural network algorithm, the influence of the model on high-level response is calculated by carrying out disturbance to low-level parameters in a certain range on the premise of guaranteeing the prediction accuracy of the model, and the numerical calculation result is visualized. The analysis result of the chassis parameters on the road noise sensitivity is shown in fig. 4, the color of the line in the graph indicates the sensitivity of each level, the darker the color indicates the higher the sensitivity, and thus the visual result graph is beneficial to engineers to quickly find out paths with relatively higher sensitivity, and the follow-up specific analysis on a certain path is facilitated.
The genetic algorithm GA is a probability optimization method combining nature genetics and computer science, is based on the principle of biological evolutionary theory, conforms to the thought of 'object bid, survival of the fittest and the culling' of the biological evolutionary theory, has the advantages of strong adaptability, strong stability, high convergence speed and the like, and utilizes the genetic algorithm to solve the optimal solution of the global variable.
In a second embodiment, a road noise optimization method includes the following steps:
s1, decomposing the road noise of the whole vehicle layer by layer to chassis parts along a vibration transmission path to obtain a plurality of decomposition parameters, and constructing a road noise performance level decomposition framework of the whole vehicle. And establishing a road noise prediction model with the input as a decomposition parameter and the output as in-vehicle noise based on the road noise multi-layer decomposition architecture. The whole vehicle road noise performance level decomposition architecture comprises a first level, a second level, a third level and a fourth level, wherein the first level is in-vehicle noise, the second level is the passive side vibration response of a suspension and a vehicle body attachment point, the third level is the active side vibration response of the suspension and the vehicle body attachment point, and the fourth level is vibration excitation from a steering knuckle and dynamic parameters of a spherical hinge bush. The decomposition principle is as follows: the design objective (noise or vibration) of the upper level is decomposed down into excitation (vibration) of the adjacent lower level and the connecting element dynamics parameters (bushing stiffness, spring stiffness, shock absorber damping force).
The construction of the road noise prediction model specifically comprises the following steps.
S11, acquiring training sample data based on the constructed whole vehicle road noise performance hierarchical decomposition architecture. The training sample data can be obtained through road tests, simulation analysis or combination of the road tests and the simulation analysis, and the like, and can be specifically selected according to actual conditions of projects. In the embodiment, noise and vibration data are acquired for the test vehicle in a road test mode.
Road surface excitation is transmitted to steering knuckles through four tires, and each steering knuckle is provided with a three-way acceleration sensor, and the total number of the steering knuckles is four. The suspension and the vehicle body end and the suspension end of the vehicle body attachment point are respectively provided with a three-way acceleration sensor. In order to measure the transmission parameters of vibration of each transmission path as much as possible, the sensor direction should be perpendicular to the whole vehicle coordinate system. The sound pressure sensor is arranged in the vehicle at the right ear of the driver.
After the sound pressure sensor and the acceleration sensor are arranged, setting data acquisition parameters: the vibration bandwidth is 0-2000 Hz, the noise bandwidth is 0-2000 Hz, and the frequency resolution is 1Hz. The test working condition is that the rough asphalt road runs at a constant speed, and the running speed is 60km/h. In order to ensure the reliability of the measurement result, the measurement is repeated at least three times, the sound pressure level difference of each measurement is not more than 0.5dB (A), if the difference of the measurement values exceeds 0.5dB (A), multiple tests are required, and the sound pressure level difference of three continuous measurement data is required to be controlled within a specified range. And finally, testing to obtain the sound pressure signals of the sound pressure sensor and the vibration signals of all the three-way acceleration sensors.
And S12, training the whole vehicle road noise prediction model by using training sample data to obtain a trained road noise prediction model meeting the precision requirement. In this embodiment, the road noise prediction model is a hybrid model combining a support vector regression algorithm and a convolutional neural network algorithm. The SVR algorithm is a machine learning method based on a statistical learning theory and a structural risk minimization principle, and the method maps multidimensional input to a feature space with higher dimensionality through a nonlinear kernel function and then carries out regression operation so as to obtain a nonlinear mapping relation between input features and output indexes. The support vector regression is based on the minimum error bound of all data, so that the support vector regression only needs a part of subsets in the training data, the calculation efficiency is high, and the accuracy is ensured.
The expression of the road noise prediction model is as follows:
wherein f (x) is structural acoustic road noise output by the road noise prediction model, m is the number of sample points in the training sample data set used by the road noise prediction model, and alpha i And
is Lagrangian multiplier, k (x i ,x j ) As a kernel function, b is a bias value.
Since radial basis functions (Radial Basis Function, RBF) have high fitting accuracy among many problems, the present embodiment selects the RBF function as a kernel function, whose expression is: k (x) i ,x j )＝exp(-γx i -x j 2 ),γ＞0。
Wherein x is i Ith sample point data, x, used for training a road noise prediction model j And the j-th sample point data used for training the road noise prediction model is gamma which is the RBF kernel width.
Convolutional neural networks are used as a representative algorithm for deep learning, and have excellent performances in various fields such as image classification, target detection and the like. The convolutional neural network directly drives features by the data, and establishes a mapping relation between input features and output indexes through multidimensional nonlinear feature extraction, so that the convolutional neural network has extremely strong data characterization and mapping capability, and compared with the traditional shallow artificial neural network, the convolutional neural network model has higher prediction precision and robustness. The typical structure mainly comprises five layers, namely: input layer, convolution layer, pooling layer, full tie layer and output layer, in this embodiment, the output layer is SVR regression output layer. According to the invention, basic road noise data of the same suspension type vehicle type are obtained in a test mode, and then a road noise prediction model is built based on a road noise level decomposition architecture, an SVR algorithm and a convolutional neural network algorithm.
And (3) dividing the training sample data acquired in the step (S11) into a training set and a testing set, inputting the training set into a road noise prediction model, and performing parameter optimization training on the model to obtain a trained whole road noise prediction model. And then inputting training sample data of the test set into the trained road noise prediction model for verification and evaluation, if the prediction precision reaches the design requirement, the test model is proved to pass, otherwise, the road noise prediction model is reconstructed.
S2, determining a value range of the decomposition parameters by taking the decomposition parameters as design variables, setting an optimization target and constraints, carrying out optimizing solution based on a road noise prediction model by adopting a genetic algorithm, and calculating to obtain an optimal solution of the decomposition parameters. In engineering practice, the sound pressure level of noise in a vehicle may be defined as an optimization target, and the vibration level of components such as a vehicle seat and a steering wheel may be defined as an optimization target.
In this embodiment, in order to reduce the calculation amount, a sensitivity analysis method is adopted to perform sensitivity analysis based on a road noise prediction model, the cost and performance effects are comprehensively considered, at least one of a plurality of decomposition parameters is determined as a design variable, the value range of the design variable is determined, and optimization solution is not required for all the decomposition parameters. The front suspension frame is of a Macpherson structure, the rear suspension frame is of a road noise level decomposition structure built by a multi-link structure, a road noise prediction model is built by combining a support vector regression algorithm and a convolutional neural network algorithm, the influence of the model on high-level response is calculated by carrying out disturbance to low-level parameters in a certain range on the premise of guaranteeing the prediction accuracy of the model, and the numerical calculation result is visualized.
In this embodiment, after obtaining the optimal solution of the decomposition parameters, the interval optimization algorithm is adopted to search the designed variable interval to expand the optimal interval, and whether the optimization target is achieved is determined, if not, the constraint range of the optimization target is adjusted, or the designed variable and the value range are readjusted until the optimization target is achieved. The design variable interval searching strategy comprises a large-range variable interval expansion searching and a variable interval refinement expansion searching. For large-scale variable interval expansion search, the main purpose is to quickly find out the approximate edge positions of all the optimal design variable intervals, so that the subsequent refined search is located at a better initial position. And (3) carrying out small-range refined search on each design variable on the rough edge of the previously found optimal interval for the refined expansion search of the variable interval, so that the accurate edge of the optimal interval of the design variable can be obtained, and the optimal design is completed. The optimizing strategy and the section expanding technology can be adjusted or improved according to the needs in actual work.
In a third embodiment, referring to fig. 5, a road noise optimization system is shown, which is capable of executing the steps of the road noise optimization method according to the first embodiment of the present invention, and includes: the road noise prediction model module is used for establishing a road noise prediction model which is input into decomposition parameters and output into noise in the vehicle based on the road noise multi-layer decomposition architecture; the optimization solving module is used for carrying out optimization solving based on the road noise prediction model and calculating to obtain an optimal solution of the decomposition parameters; the data management module is used for collecting road noise data, generating training sample data based on the collected road noise data, and the sensitivity module is used for carrying out sensitivity analysis based on the road noise prediction model.
The data management module can realize data updating, data screening and model training, the prediction module, namely the road noise prediction model module, can realize data importing, data prediction and result visualization, the sensitivity module can realize sensitivity analysis and result visualization, and the optimization module, namely the optimization solving module, can realize optimization analysis and result visualization.
And integrating and packaging the road noise prediction model, the sensitivity analysis method, the genetic algorithm and the interval optimization algorithm, and developing based on Python to obtain a road noise optimization platform and software which are convenient for engineering operation. Referring to FIG. 6, the software platform can provide a menu type combination scheme for NVH performance parameter design of chassis of each vehicle type, guide forward development of products in the early stage of the project, improve road noise performance, solve the problem of quick locking in the later stage of the project, reduce test times, improve correction efficiency of road noise problems, and reduce test cost.
In a fourth embodiment, a storage medium has a computer readable program stored therein, where the computer readable program is capable of executing the steps of the road noise optimization method according to the first embodiment of the present invention when the computer readable program is called.
It should be noted that the illustrated storage medium may be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may be, for example, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-Only Memory (ROM), an erasable programmable read-Only Memory (Erasable Programmable Read Only Memory, EPROM), flash Memory, an optical fiber, a portable compact disc read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this embodiment, the computer-readable signal medium may comprise a data signal propagated in baseband or as part of a carrier wave, with a computer-readable computer program embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. A computer program embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. Where each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The above embodiments are merely preferred embodiments for fully explaining the present invention, and the scope of the present invention is not limited thereto. Equivalent substitutions and modifications will occur to those skilled in the art based on the present invention, and are intended to be within the scope of the present invention.
Claims (12)
1. The road noise optimization method is characterized by comprising the following steps of:
s1, decomposing the road noise of the whole vehicle to chassis parts layer by layer along a vibration transmission path to obtain a plurality of decomposition parameters, constructing a road noise performance level decomposition framework of the whole vehicle, and establishing a road noise prediction model which is input as the decomposition parameters and output as noise in the vehicle based on the road noise multi-layer decomposition framework;
s2, determining a value range of the decomposition parameters by taking the decomposition parameters as design variables, setting an optimization target and constraints, carrying out optimizing solution based on a road noise prediction model, and calculating to obtain an optimal solution of the decomposition parameters.
2. The road noise optimization method according to claim 1, wherein: the system comprises a vehicle road noise performance level decomposition framework in S1, wherein the vehicle road noise performance level decomposition framework comprises a first level, a second level, a third level and a fourth level, the first level is in-vehicle noise, the second level is in vibration response of a suspension and a vehicle body attachment point passive side, the third level is in vibration response of a suspension and a vehicle body attachment point active side, and the fourth level is vibration excitation from a steering knuckle and dynamic parameters of a spherical hinge bush.
3. The road noise optimization method according to claim 1 or 2, characterized in that: the construction of the road noise prediction model in S1 specifically comprises the following steps:
s11, acquiring training sample data based on the constructed whole vehicle road noise performance hierarchical decomposition architecture;
and S12, training the whole vehicle road noise prediction model by using training sample data to obtain a trained road noise prediction model meeting the precision requirement.
4. The road noise optimization method according to claim 3, wherein S12 specifically is: the training sample data acquired in the step S11 are divided into a training set and a testing set, the training set is input into a road noise prediction model to train the model, and a trained whole road noise prediction model is obtained; and then inputting training sample data of the test set into the trained road noise prediction model for verification and evaluation, if the prediction precision reaches the design requirement, the test model is proved to pass, otherwise, the road noise prediction model is reconstructed.
5. The road noise optimization method according to claim 1 or 2, characterized in that sensitivity analysis is performed based on a road noise prediction model, cost and performance effects are comprehensively considered, at least one of a plurality of decomposition parameters is determined as a design variable, and a value range of the design variable is determined.
6. The road noise optimization method according to claim 1 or 2, wherein the road noise prediction model in S1 is a hybrid model of a support vector regression algorithm and a convolutional neural network algorithm.
7. The road noise optimization method according to claim 1 or 2, characterized in that: and S2, optimizing based on the road noise prediction model by adopting a genetic algorithm, and calculating to obtain an optimal value of the decomposition parameter.
8. The road noise optimization method according to claim 1 or 2, characterized in that: and S2, after the optimal solution of the decomposition parameters is obtained, the optimal solution can be expanded into an optimal interval through searching the designed variable interval, whether the optimization target is achieved is judged, and if not, the constraint range of the optimization target is adjusted, or the designed variable and the value range are readjusted until the optimization target is achieved.
9. A road noise optimization system, characterized in that it is capable of performing the steps of the road noise optimization method according to any one of claims 1 to 8, comprising:
the road noise prediction model module is used for establishing a road noise prediction model which is input into decomposition parameters and output into noise in the vehicle based on the road noise multi-layer decomposition architecture;
and the optimization solving module is used for carrying out optimization solving based on the road noise prediction model and calculating to obtain an optimal solution of the decomposition parameters.
10. The road noise optimization system of claim 9, further comprising a data management module for collecting road noise data and generating training sample data based on the collected road noise data.
11. The road noise optimization system of claim 9, further comprising a sensitivity module for performing sensitivity analysis based on a road noise prediction model.
12. A storage medium, characterized by: a computer readable program stored therein, which when invoked is capable of performing the steps of the road noise optimization method according to any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310330213.XA CN116415494A (en) | 2023-03-30 | 2023-03-30 | Road noise optimization method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310330213.XA CN116415494A (en) | 2023-03-30 | 2023-03-30 | Road noise optimization method, system and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116415494A true CN116415494A (en) | 2023-07-11 |
Family
ID=87049109
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310330213.XA Pending CN116415494A (en) | 2023-03-30 | 2023-03-30 | Road noise optimization method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116415494A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113761798A (en) * | 2021-08-30 | 2021-12-07 | 重庆长安汽车股份有限公司 | Method for building and predicting model for sound insulation performance of vehicle acoustic package |
-
2023
- 2023-03-30 CN CN202310330213.XA patent/CN116415494A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113761798A (en) * | 2021-08-30 | 2021-12-07 | 重庆长安汽车股份有限公司 | Method for building and predicting model for sound insulation performance of vehicle acoustic package |
| CN113761798B (en) * | 2021-08-30 | 2024-05-03 | 重庆长安汽车股份有限公司 | Construction method and prediction method of vehicle acoustic sound-wrapping and insulation performance prediction model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rauh | Virtual development of ride and handling characteristics for advanced passenger cars | |
| CN112287551B (en) | Driving performance system level index decomposition method based on whole vehicle conceptual model | |
| CN110910531B (en) | Rapid pavement friction coefficient detection method based on vehicle-mounted OBD information | |
| Pan et al. | Machine learning approaches to estimate suspension parameters for performance degradation assessment using accurate dynamic simulations | |
| CN116415494A (en) | Road noise optimization method, system and storage medium | |
| CN108228994B (en) | Stress calculation method for vehicle and working device under random excitation of off-road | |
| CN113886974B (en) | Method for predicting sound path noise of in-vehicle structure | |
| Llopis-Albert et al. | Multiobjective optimization framework for designing a vehicle suspension system. A comparison of optimization algorithms | |
| CN115659492A (en) | Complex equipment nonlinear vibration model parameter optimization method | |
| Wang et al. | A road level identification method for all-terrain crane based on Support Vector Machine | |
| Yarmohammadisatri et al. | A family base optimization of a developed nonlinear vehicle suspension model using gray family design algorithm | |
| CN115130218A (en) | Self-learning prediction method for road noise of automotive suspension structure | |
| CN115270284A (en) | Shock absorber dynamic characteristic identification model establishing method and simulation method | |
| Xu et al. | Identification of spring-force factors of suspension systems using progressive neural network on a validated computer model | |
| CN108733944B (en) | Method and device for evaluating steering wheel buffeting performance | |
| Abubakar et al. | Robust polytopic LPV based Adaptive Cruise Control design for Autonomous Vehicle System | |
| CN114880773A (en) | Road noise sensitivity analysis method based on whole vehicle road noise multilevel decomposition framework | |
| Kanchwala | Studies in simplified dynamic modeling and characterization of vehicle suspensions | |
| Zhang et al. | Serial combinational optimization method for double wishbone suspension’s pseudo damage improvement | |
| Jia et al. | Vehicle interior road noise predicted using support vector regression | |
| Hugo | Virtual model of a vehicle adaptive damper system | |
| CN116484494A (en) | Method and device for predicting sound path noise of in-vehicle structure, vehicle and readable storage medium | |
| Chen et al. | Ride Comfort Optimization Method for Commercial Vehicle Based on Nonlinear Damping and PSO | |
| CN115358112A (en) | Road noise prediction method and system and computer equipment | |
| DiIenna | An Optimization Approach to the Design of a Formula-Style Racecar Through Artificial Neural Network Surrogate Models |
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 |