CN106056149A - System and method for establishing working condition division remote damage assessment of different vehicle types based on artificial intelligence unsupervised learning principal component analysis method - Google Patents
System and method for establishing working condition division remote damage assessment of different vehicle types based on artificial intelligence unsupervised learning principal component analysis method Download PDFInfo
- Publication number
- CN106056149A CN106056149A CN201610365533.9A CN201610365533A CN106056149A CN 106056149 A CN106056149 A CN 106056149A CN 201610365533 A CN201610365533 A CN 201610365533A CN 106056149 A CN106056149 A CN 106056149A
- Authority
- CN
- China
- Prior art keywords
- data
- collision
- model
- operating mode
- principal component
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Abstract
The invention relates to a system and method for establishing working condition division remote damage assessment of different vehicle types based on an artificial intelligence unsupervised learning principal component analysis method and belongs to the vehicle damage assessment field. The objective of the invention is to solve problems in working condition detection after a vehicle collision. According to the technical schemes of the invention, a working condition detection subsystem is adopted to judge all working condition information in the vehicle collision; and the working condition detection subsystem learns working condition training data so as to generate a working condition model, wherein the working condition model is built by adopting the unsupervised learning principal component analysis method. With the system and method provided by the technical schemes of the invention adopted, working condition detection in the vehicle collision can be realized; and a machine learning method is used in the remote damage assessment technical field, so that the accuracy of judgment in a damage assessment process can be improved.
Description
Technical field
The invention belongs to car damage identification field, relate to one and build based on artificial intelligence's unsupervised learning principal component analytical method
The vertical long-range loss assessment system of different automobile types divided working status and method.
Background technology
During low-speed motion (including low speed links traveling, vehicle parking etc.), take place frequently collision accident for vehicle and lead
The Claims Resolution dispute problem caused, long-range setting loss technology by multi-signal in collection vehicle driving process (as speed, acceleration,
Angular velocity, sound etc.) and analyzed with signal processing and machine learning techniques, to judge whether collision occurs and collision rift
The damage situation of vehicle.
After vehicle crashes, headend equipment is capable of detecting when the generation of collision and intercepts the signal of collision process,
Being sent to high in the clouds by wireless network, remote server extracts the eigenvalue of design in advance from the signal received, and uses machine
Learning algorithm is analyzed, and first judges the accuracy of crash data, then judges collision object and operating mode situation, to determine Collision Number
Create the damage of which kind of grade according to what part of set pair, then go out with reference to amount for which loss settled concurrent according to part injury rating calculation
Deliver to insurance company.The detection for vehicle, operating mode, target, part and region can be related to during this.
Summary of the invention
After solving vehicle collision, for the problem of operating mode detection, the present invention proposes based on artificial intelligence without supervision
Study principal component analytical method sets up the long-range loss assessment system of different automobile types divided working status and method, to realize the operating mode during setting loss
Detection.
In order to solve above-mentioned technical problem, the technical scheme that the present invention provides is characterized by: including:
Truck type choice subsystem, selects the model data corresponding to vehicle as total data set；
Data classification subsystem, reads CAE emulation data and real vehicle data, and classifies data accordingly；
Collision detection subsystem, it is judged that the most whether vehicle collides；Described collision detection subsystem pair
Collision training data carries out learning thus generates collision model, and described collision model is set up and used unsupervised learning principal component analysis
Method；
Operating mode detection subsystem, it is judged that all work informations that collision occurs；Operating mode is instructed by described operating mode detection subsystem
White silk data carry out learning thus generate condition model, and described condition model is set up and used unsupervised learning principal component analytical method.
Beneficial effect: technique scheme, it is possible to achieve the operating mode for vehicle collision detects, at this of long-range setting loss
Technical field employs the method for machine learning, for machine learning method, during setting loss, it determines accuracy rate on
To promote；The present invention is by selecting vehicle to import the data corresponding to this vehicle, and data classification is then for model training
With the purpose of test and the step that adds；The detection of operating mode is the purpose that the program realizes, and is to obtain through sequence of operations
The result arrived.
Accompanying drawing explanation
Fig. 1 is the structural schematic block diagram of system of the present invention；
Fig. 2 is data variation directional diagram；
Fig. 3 is sample data coordinate transforming schematic diagram；
Fig. 4 is the point diagram of reconstruct data.
Detailed description of the invention
In order to the present invention being made apparent explanation, below the technical term that the present invention relates to is made definitions:
Operating mode: all collision informations such as collision angle, direction, target, region；
Vehicle: automobile model；
Target: collision target；
Region: position of collision；
Part: auto parts；
Operating mode detects: detect all collision informations such as this car collision angle, direction, target, region；
Vehicle detects: the automobile model that detection collides with this car；
Target detection: detect this car collision target；
Region detection: detect this car position of collision；
Piece test: detect this car auto parts.
Embodiment 1:
A kind of set up different automobile types divided working status long-range setting loss system based on artificial intelligence's unsupervised learning principal component analytical method
System, including:
Truck type choice subsystem, selects the model data corresponding to vehicle as total data set；
Data classification subsystem, reads CAE emulation data and real vehicle data, and classifies data accordingly；
Collision detection subsystem, it is judged that the most whether vehicle collides；Described collision detection subsystem pair
Collision training data carries out learning thus generates collision model, and described collision model is set up and used unsupervised learning principal component analysis
Method；
Operating mode detection subsystem, it is judged that all work informations that collision occurs；Operating mode is instructed by described operating mode detection subsystem
White silk data carry out learning thus generate condition model, and described condition model is set up and used unsupervised learning principal component analytical method.
Described collision detection subsystem includes, collision training module, crash tests module, collision authentication module, described in touch
Hitting training module and generate collision model for learning collision training data, crash tests module is for surveying collision
Examination data bring the result detecting collision model in collision model into, and collision authentication module uses true sport car data verification collision mould
The reliability of type and accuracy rate；
Described operating mode detection subsystem includes, operating mode training module, working condition measurement module, operating mode authentication module, described work
Condition training module generates condition model for learning operating mode training data, and described working condition measurement module is for by work
Condition test data bring the result detecting condition model in model into, and operating mode authentication module uses true sport car data verification operating mode mould
The reliability of type and accuracy rate.
Described unsupervised learning principal component analytical method:
Step 1. represents data set；
Step 2. spin data, each sample concentrating data rotates respectively；
Step 3. Data Dimensionality Reduction number；
Step 4. reduces approximate data；
Step 5. selects main constituent number.
Said method particularly as follows:
Step 1. represents data set:
Data set table is shown as { x(1), x(2)..., x(m), calculate matrix ∑,
The average assuming x is zero, and ∑ is the covariance matrix of x, and m is Grad；
Step 2. spin data, each sample concentrating data rotates respectively, obtains spin dataU is data variation direction, computing UTX represents and rotates to base u1,u2,...,unOn training data, n is
The number of base；
Step 3. Data Dimensionality Reduction number: dataDrop to k dimension table showChoose xrotFront k composition,
Before the most corresponding, k is according to the principal direction of change,It is xrotLast n-k element set to 0 the approximate representation of gained；
Step 4. reduces approximate data: make x=Uxrot；
Step 5. selects main constituent number: set λ1, λ2..., λnRepresent the eigenvalue of ∑ so that λjFor corresponding to feature to
Amount ujEigenvalue.If that we retain front k composition, then the variance percentage retained can be calculated as:
Embodiment 2:
One sets up the long-range setting loss side of different automobile types divided working status based on artificial intelligence's unsupervised learning principal component analytical method
Method, comprises the following steps:
Step one. select the model data corresponding to vehicle as total data set；
Step 2. read CAE emulation data and real vehicle data, and accordingly data are classified；
Step 3. judge the most whether vehicle collides；Described collision detection subsystem is to collision training
Data carry out learning thus generate collision model, and described collision model is set up and used unsupervised learning principal component analytical method；
Step 4. judge all work informations that collision occurs；Operating mode training data is entered by described operating mode detection subsystem
Row learns thus generates condition model, and described condition model is set up and used unsupervised learning principal component analytical method.
Comprise the concrete steps that:
Step 3 includes:
S3.1. use collision detection subsystem that CAE collision simulation data are processed, then classify to produce collision to it
Training data and crash tests data；
S3.2. in collision training module, collision training data learnt and produces collision model, carrying out simulated crash
The effect of training data；
S3.3. crash tests data are used to carry out the result of test collisions model in crash tests module；
S3.4. use true sport car data as collision checking data and to bring collision authentication module into, verify collision mould
The accuracy of type；
Step 4 includes:
S4.1. CAE operating mode emulation data are processed by applying working condition detection subsystem, then it is carried out classification generation operating mode instruction
Practice data and working condition measurement data；
S4.2. in operating mode training module, operating mode training data learnt and produce condition model, carrying out simulated condition
The effect of training data；
S4.3. in working condition measurement module, applying working condition test data carry out the result of measurement condition model；
S4.4. use true sport car data as operating mode checking data and to bring operating mode authentication module into, verify operating mode mould
The accuracy of type.
Described unsupervised learning principal component analytical method:
Step 1. represents data set；
Step 2. spin data, each sample concentrating data rotates respectively；
Step 3. Data Dimensionality Reduction number；
Step 4. reduces approximate data；
Step 5. selects main constituent number.
Said method particularly as follows:
Step 1. represents data set:
Data set table is shown as { x(1), x(2)..., x(m), calculate matrix ∑,
The average assuming x is zero, and ∑ is the covariance matrix of x, and m is Grad；
Step 2. spin data, each sample concentrating data rotates respectively, obtains spin dataU is data variation direction, computing UTX represents and rotates to base u1,u2,...,unOn training data, n is
The number of base；
Step 3. Data Dimensionality Reduction number: dataDrop to k dimension table showChoose xrotFront k composition,
Before the most corresponding, k is according to the principal direction of change,It is xrotLast n-k element set to 0 the approximate representation of gained；
Step 4. reduces approximate data: make x=Uxrot；
Step 5. selects main constituent number: set λ1, λ2..., λnRepresent the eigenvalue of ∑ so that λjFor corresponding to feature to
Amount ujEigenvalue.If that we retain front k composition, then the variance percentage retained can be calculated as:
Embodiment 3:
Make to supplement further to the technical scheme of embodiment 1 and 2, use the principal component analytical method of unsupervised learning, should
Method is the principal component analysis (PCA) method more like a pretreatment, and data originally can be reduced dimension by it, and make
Reduce the variance between the data of dimension maximum.
We can consider to reduce variance (such as the when of training pattern, we can take into account variance-deviation sometimes
Equilibrium), the increase variance that we can try one's best sometimes.For practice, calculate by increasing the PCA of projection variance as far as possible
Method, can improve our algorithm quality really.
Step 1. is during we use data, and data set table is shown as { x(1), x(2)..., x(m), dimension n=2 is i.e.Assume us to want data and drop to 1 dimension from 2 dimensions.Make each feature x1And x2Have identical average (zero) and
Variance.For convenience of showing, according to x1The size of value, each point has been coated one of three kinds of colors by respectively, but this color is also
It is not used in algorithm and is only used for diagram.
PCA algorithm will find a lower dimensional space to project our data.u1It is the principal direction of data variation, and u2It is
Secondary direction.
It is to say, data are at u1Change ratio on direction is at u2On direction greatly.For finding out direction u more formally1With
u2, first we calculate matrix ∑, as follows:
The average assuming x is zero, then ∑ is exactly the covariance matrix of x.(symbol ∑ is read " Sigma ", is covariance square
The standard symbol of battle array.While it seem that with summation symbolRelatively picture, but they are two different concepts in fact.)
Step 2. spin data.So far, we can be x (u1, u2) base is expressed as:
The each sample i concentrating data rotates respectively:Then after conversion
Data xrotDisplay, in coordinate diagram, can obtain:
Here it is training dataset is rotated to u1, u2Result after base.It is said that in general, computing UTX represents and rotates to base
u1,u2,...,unOn training data.Matrix U has orthogonality, i.e. meets UTU=UUT=I, if so want by postrotational to
Amount xrotIt is reduced to initial data x, by its premultiplication matrix U: x=Uxrot, check: Uxrot=UUTX=x.
Step 3. Data Dimensionality Reduction number.The principal direction of data is exactly the first dimension x of spin dataRot, 1.Therefore, if wanting this number
One-dimensional according to dropping to, can make:
More generally, if wanting dataDrop to k dimension table show(making k ＜ n), only need to choose xrot's
Front k composition, the most corresponding front k is according to the principal direction of change.
Step 4. reduces approximate data.Now, we have obtained initial dataLow-dimensional " compress " token stateIt is converted back, only needs x=Uxrot?.Further, weRegard as xrotLast n-k element quilt
Set to 0 the approximate representation of gained, if therefore givenCan be right by obtaining its end interpolation n-k 0Approximation, finally, premultiplication U just can approximate and restore former data x.Specifically, it is calculated as follows:
Above equation is based on the previous definition to U.When realizing, we the most first giveFill out 0 the most left
Take advantage of U, take advantage of 0 computing as it means that substantial amounts of.We can useCarry out the row of the front k with U to be multiplied, i.e. the rightest in above formula
, reach same purpose.The data set being applied in this example by this algorithm, can obtain as follows about reconstruct dataPoint
Figure.
As seen from the figure, what we obtained is the one-dimensional approximate reconstruction to raw data set.
When training autocoder or other nothing supervision feature learning algorithm, Riming time of algorithm will depend on input number
According to dimension.If usingReplace x as input data, then algorithm just can use low-dimensional data to be trained, and runs speed
Degree will dramatically speed up.For a lot of data sets, low-dimensional token stateIt is the splendid approximation of former data set, therefore at these
It is very suitable for closing and using PCA, the error of approximation that it introduces the least, but can significantly increase the speed of service of your algorithm.
Step 5. selects main constituent number.How we select k, i.e. retain how many PCA main constituents？Above this
In simple two dimension experiment, retain first composition and look like naturally selection.For high dimensional data, make this and determine
The most not as simple: if k is excessive, data compression rate is the highest, when limiting case k=n, using initial data in fact
(simply rotate and projected different bases)；If on the contrary, k is too small, the error of approximation Mrs of those data.
When determining k value, we would generally consider the retainable variance percentage of different value of K.Specifically, if k=n,
What so we obtained is the perfect approximation to data, namely remains all changes of the variance of 100%, i.e. initial data
All it is retained when；On the contrary, if k=0, that use in fact null vector to approach input data, the side of the most only 0%
Difference is retained when.
It is said that in general, set λ1, λ2..., λnRepresent the eigenvalue (arranging by descending order) of ∑ so that λjFor correspondence
In characteristic vector ujEigenvalue.If that we retain front k composition, then the variance percentage retained can be calculated as:
PCA additive method: maximize variance method.
Assume that we are still by the spot projection in a space to vector.First, the center in former space is given
Point:
Assuming that u1 is projection vector, the variance after projection is:
With method of Lagrange multipliers:
u1 TSu1+λ1(1-u1 Tu1)
By above formula derivation, it is allowed to be 0, obtains:
Su1=λ1u1
This is the Eigenvalue expressions of a standard, λ characteristic of correspondence value, u characteristic of correspondence vector.The left side of above formula
The condition obtaining maximum is exactly that λ 1 is maximum, the when of namely obtaining the eigenvalue of maximum.Assume that we are intended to a D dimension
Data space project in the data space of M dimension (M < D), we take the projection matrix of front M characteristic vector composition and are exactly
Enable to the matrix that variance is maximum.
Embodiment 4:Have and the identical technical scheme of embodiment 1 or 2 or 3, more specifically:
Conceptual data collection in such scheme: be entirely CAE emulation data and sport car data；Be divided into three parts as follows
1. training dataset: be used to training pattern or determine model parameter (CAE emulation data and sport car data).
2. checking data set: be used to do Model Selection (modelselection), i.e. does the final optimization pass and really of model
Fixed (CAE emulation data and sport car data).
3. test data set: the Generalization Ability of the model then trained for purely test.(CAE emulates data
With sport car data).
In the present embodiment also to the filtering related to during setting loss, weighting choose, feature extraction, normalization, eigentransformation
Have been described.
1. wave filter technology: the filtering method realized includes FIR filtering, FIR Chebyshev approximation, Chebyshev's filter
Ripple, butterworth filter etc., the Filtering.m file in mastery routine realizes.Each wave filter is common wave filter,
Matlab has corresponding function to realize, and specific algorithm refers to signal processing professional book.Provide the interior of FIR filter herein
Hold and the introduction of flow process.
Limited impulse response digital filter (FIR, FiniteImpulseResponse) is the system of a kind of full zero point,
The design of FIR filter is ensureing that amplitude characteristic meets the colleague that technology requires, it is easy to accomplish strict linear phase characteristic,
So being the outstanding advantages of FIR filter according to there being stable and linear phase characteristic.Chebyshev approximation is that the ripples such as one are forced
Nearly method, it is possible to make error frequency band be uniformly distributed, to same technical specification, this filter order sending out needs shoulder to shoulder is low, right
In the wave filter of same exponent number, this approximatioss maximum error is minimum, and the key step of its design is as follows:
Step 1: the setting of filter parameter
The parameter of wave filter includes: cut-off frequecy of passband, stopband cut-off frequency, passband maximum attenuation and stopband minimum decline
Subtract；
Step 2: be arranged on passband and the amplitude-frequency response of stopband coideal
Step 3: be scheduled on the weighting on cut-off frequecy of passband and stopband cut-off frequency point
Step 4: utilize Equation for Calculating Chebyshev approximation filter coefficient
Step 5: preserve coefficient
Step 6: extraction coefficient carries out data filtering
Wherein: the guarantee signal that is disposed to of filter parameter does not haves distortion now during processing
As, the cut-off frequency of filtered signal and sample frequency need to meet Nyquist's theorem, the most after the filtering signal
Highest frequency not can exceed that the 1/2 of original signal sample frequency, otherwise arises that Lou frequency phenomenon.According to the signal in current project
The sample frequency of collection plate is mainly 50Hz and 1KHz, according to formula F as a example by 50HzCut-off< 50/2, therefore select filter cutoff
Frequency is below 25.
See Fig. 2, for choosing of the weighting on cut-off frequecy of passband and stopband cut-off frequency point.
2. Feature Extraction Technology (seeing Fig. 3): feature extraction is carried out on collision alarm.Judge the spy that collision uses
Levy include in window in the maximum of acceleration absolute value, window between acceleration maxima and minima difference, in window
The absolute value of each point slope in the average energy of acceleration (in window the quadratic sum of acceleration a little divided by counting), window
Meansigma methods.
Judge the feature that part category is used include the average energy between speed, acceleration peak to peak,
Amplitude between maximum and minima/width between the two, acceleration maximum, acceleration minima, maximum place
The width of half-wave, minima place half-wave width, maximum and minima between difference, between peak to peak
Span, the meansigma methods of absolute value of each point slope, signal carry out each of the signal after Fourier transform in 0～38 frequency ranges
The amplitude of frequency component.
3. normalization technology: that causes classification task to eliminate the dimension between feature or order of magnitude difference is unfavorable
Impact, needs to be normalized characteristic so that have comparability between each eigenvalue, it is to avoid the spy that numerical value is bigger
Levy and flood the feature that numerical value is less.Original characteristic is after normalized, and each feature is in identical codomain scope.
Owing to the performance of Z-Score is more preferable, use Z-Score as method for normalizing.
4. feature transform technique: in the case of feature is more, for the dependency eliminated between feature and reduce redundancy
Feature, needs to convert feature, carrys out reflected sample information with the fewest new feature.In the less situation of experiment sample
Under (practical situation of this project) reduce too much intrinsic dimensionality, moreover it is possible to avoid sending out of over-fitting or poor fitting to a certain extent
Raw.According to actual needs, the eigentransformation the most realized is PCA.Being found through experiments, PCA divides for improving this project
Class performance there is no help, has declined, and this is that the feature owing to being used at present is less, does not has redundancy feature, therefore
PCA wouldn't be used, but be as being stepped up of subsequent characteristics, however not excluded that use the probability of PCA later.
In accompanying drawing 1, record: the Truck type choice subsystem that Truck type choice is in the present invention；Data categorization module is
Data classification subsystem in the present invention；The collision detection subsystem that collision judgment module is in the present invention；Operating mode detection mould
Block is the operating mode detection subsystem of the present invention；Vehicle detection module is the vehicle detection subsystem of the present invention；Piece test
Module i.e. piece test subsystem；Module of target detection is the target detection subsystem of the present invention, and region detection module is
The region detection subsystem of the present invention.
The above, only the invention preferably detailed description of the invention, but the protection domain of the invention is not
Being confined to this, any those familiar with the art is in the technical scope that the invention discloses, according to the present invention
The technical scheme created and inventive concept thereof in addition equivalent or change, all should contain the invention protection domain it
In.
Claims (8)
1. set up the long-range loss assessment system of different automobile types divided working status based on artificial intelligence's unsupervised learning principal component analytical method,
It is characterized in that, including:
Truck type choice subsystem, selects the model data corresponding to vehicle as total data set；
Data classification subsystem, reads CAE emulation data and real vehicle data, and classifies data accordingly；
Collision detection subsystem, it is judged that the most whether vehicle collides；Described collision detection subsystem is to collision
Training data carries out learning thus generates collision model, and described collision model is set up and used unsupervised learning principal component analysis side
Method；
Operating mode detection subsystem, it is judged that all work informations that collision occurs；Operating mode is trained number by described operating mode detection subsystem
Generating condition model according to carrying out learning, described condition model is set up and is used unsupervised learning principal component analytical method.
2. set up different automobile types divided working status based on artificial intelligence's unsupervised learning principal component analytical method as claimed in claim 1
Remotely loss assessment system, it is characterised in that
Described collision detection subsystem includes, collision training module, crash tests module, collision authentication module, described collision is instructed
Practicing module and generate collision model for learning collision training data, crash tests module is for by crash tests number
According to bringing the result detecting collision model in collision model into, collision authentication module uses true sport car data verification collision model
Reliability and accuracy rate；
Described operating mode detection subsystem includes, operating mode training module, working condition measurement module, operating mode authentication module, and described operating mode is instructed
Practicing module and generate condition model for learning operating mode training data, described working condition measurement module is for surveying operating mode
Examination data bring the result detecting condition model in model into, and operating mode authentication module uses true sport car data verification condition model
Reliability and accuracy rate.
3. such as claim 1 or 2, to set up different automobile types divided working status based on artificial intelligence's unsupervised learning principal component analytical method remote
Journey loss assessment system, it is characterised in that described unsupervised learning principal component analytical method includes:
Step 1. represents data set；
Step 2. spin data, each sample concentrating data rotates respectively；
Step 3. Data Dimensionality Reduction number；
Step 4. reduces approximate data；
Step 5. selects main constituent number.
4. set up different automobile types divided working status such as claim 3 based on artificial intelligence's unsupervised learning principal component analytical method remotely to determine
Damage system, it is characterised in that unsupervised learning principal component analytical method particularly as follows:
Step 1. represents data set:
Data set table is shown as { x(1), x(2)..., x(m), calculate matrix Σ,
The average assuming x is zero, and Σ is the covariance matrix of x, and m is Grad；
Step 2. spin data, each sample concentrating data rotates respectively, obtains spin dataU
For data variation direction, computing UTX represents and rotates to base u1, u2 ..., unOn training data, n is the number of base；
Step 3. Data Dimensionality Reduction number: dataDrop to k dimension table showChoose xrotFront k composition, respectively
Before corresponding, k is according to the principal direction of change,It is xrotLast n-k element set to 0 the approximate representation of gained；
Step 4. reduces approximate data: make x=Uxrot；
Step 5. selects main constituent number: set λ1, λ2..., λnRepresent the eigenvalue of Σ so that λjFor corresponding to characteristic vector uj
Eigenvalue.If that we retain front k composition, then the variance percentage retained can be calculated as:
5. set up the long-range damage identification method of different automobile types divided working status based on artificial intelligence's unsupervised learning principal component analytical method,
It is characterized in that, comprise the following steps:
Step one. select the model data corresponding to vehicle as total data set；
Step 2. read CAE emulation data and real vehicle data, and accordingly data are classified；
Step 3. judge the most whether vehicle collides；Collision training data is learnt thus generates and touch
Hitting model, described collision model is set up and is used unsupervised learning principal component analytical method；
Step 4. judge all work informations that collision occurs；Operating mode training data is learnt thus generates condition model,
Described condition model is set up and is used unsupervised learning principal component analytical method.
6. set up different automobile types divided working status based on artificial intelligence's unsupervised learning principal component analytical method as claimed in claim 5
Remotely damage identification method, it is characterised in that comprise the concrete steps that:
Step 3 includes:
S3.1. use collision detection subsystem that CAE collision simulation data are processed, then classify to produce collision training to it
Data and crash tests data；
S3.2. in collision training module, collision training data learnt and produces collision model, carrying out simulated crash training
The effect of data；
S3.3. crash tests data are used to carry out the result of test collisions model in crash tests module；
S3.4. use true sport car data as collision checking data and to bring collision authentication module into, verify collision model
Accuracy；
Step 4 includes:
S4.1. CAE operating mode emulation data are processed by applying working condition detection subsystem, then it is carried out classification generation operating mode training number
According to working condition measurement data；
S4.2. in operating mode training module, operating mode training data learnt and produce condition model, carrying out simulated condition training
The effect of data；
S4.3. in working condition measurement module, applying working condition test data carry out the result of measurement condition model；
S4.4. use true sport car data as operating mode checking data and to bring operating mode authentication module into, verify condition model
Accuracy.
7. setting up different automobile types based on artificial intelligence's unsupervised learning principal component analytical method and divide as described in claim 5 or 6
The long-range damage identification method of operating mode, it is characterised in that described unsupervised learning principal component analytical method:
Step 1. represents data set；
Step 2. spin data, each sample concentrating data rotates respectively；
Step 3. Data Dimensionality Reduction number；
Step 4. reduces approximate data；
Step 5. selects main constituent number.
8. set up different automobile types divided working status based on artificial intelligence's unsupervised learning principal component analytical method as claimed in claim 7
Remotely damage identification method, it is characterised in that unsupervised learning principal component analytical method particularly as follows:
Step 1. represents data set:
Data set table is shown as { x(1), x(2)..., x(m), calculate matrix Σ,
The average assuming x is zero, and Σ is the covariance matrix of x, and m is Grad；
Step 2. spin data, each sample concentrating data rotates respectively, obtains spin dataU
For data variation direction, computing UTX represents and rotates to base u1,u2,...,unOn training data, n is the number of base；
Step 3. Data Dimensionality Reduction number: dataDrop to k dimension table showChoose xrotFront k composition, respectively
Before corresponding, k is according to the principal direction of change,It is xrotLast n-k element set to 0 the approximate representation of gained；
Step 4. reduces approximate data: make x=Uxrot；
Step 5. selects main constituent number: set λ1, λ2..., λnRepresent the eigenvalue of Σ so that λjFor corresponding to characteristic vector uj
Eigenvalue.If that we retain front k composition, then the variance percentage retained can be calculated as:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610365533.9A CN106056149A (en) | 2016-05-27 | 2016-05-27 | System and method for establishing working condition division remote damage assessment of different vehicle types based on artificial intelligence unsupervised learning principal component analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610365533.9A CN106056149A (en) | 2016-05-27 | 2016-05-27 | System and method for establishing working condition division remote damage assessment of different vehicle types based on artificial intelligence unsupervised learning principal component analysis method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106056149A true CN106056149A (en) | 2016-10-26 |
Family
ID=57175143
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610365533.9A Pending CN106056149A (en) | 2016-05-27 | 2016-05-27 | System and method for establishing working condition division remote damage assessment of different vehicle types based on artificial intelligence unsupervised learning principal component analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106056149A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112464154A (en) * | 2020-11-27 | 2021-03-09 | 中国船舶重工集团公司第七0四研究所 | Method for automatically screening effective features based on unsupervised learning |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103303237A (en) * | 2013-06-21 | 2013-09-18 | 湖南大学 | Air bag detonation control method based on genetic neural network |
| CN104932359A (en) * | 2015-05-29 | 2015-09-23 | 大连楼兰科技股份有限公司 | Vehicle remote unattended loss assessment system based on CAE technology and loss assessment method thereof |
-
2016
- 2016-05-27 CN CN201610365533.9A patent/CN106056149A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103303237A (en) * | 2013-06-21 | 2013-09-18 | 湖南大学 | Air bag detonation control method based on genetic neural network |
| CN104932359A (en) * | 2015-05-29 | 2015-09-23 | 大连楼兰科技股份有限公司 | Vehicle remote unattended loss assessment system based on CAE technology and loss assessment method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 奋斗路上的产品狗: "主成分析法（PCA）", 《奋斗路上的产品狗，URL：HTTPS://BLOG.CSDN.NET/QQ_25491201/ARTICLE/DETAILS/51254107》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112464154A (en) * | 2020-11-27 | 2021-03-09 | 中国船舶重工集团公司第七0四研究所 | Method for automatically screening effective features based on unsupervised learning |
| CN112464154B (en) * | 2020-11-27 | 2024-03-01 | 中国船舶重工集团公司第七0四研究所 | Method for automatically screening effective features based on unsupervised learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110132598B (en) | Fault noise diagnosis algorithm for rolling bearing of rotating equipment | |
| CN105046279B (en) | A kind of analog circuit fault method for classifying modes | |
| CN103645249B (en) | Online fault detection method for reduced set-based downsampling unbalance SVM (Support Vector Machine) transformer | |
| CN110197209B (en) | Radiation source identification method based on multi-feature fusion | |
| CN104282028A (en) | Blurred image detection method fusing frequency spectrum information and cepstrum information | |
| CN106056147A (en) | System and method for establishing target division remote damage assessment of different vehicle types based artificial intelligence radial basis function neural network method | |
| CN102122355A (en) | SAR (synthetic aperture radar) target identification method based on nuclear sparse representation | |
| CN106056150A (en) | System and method for establishing part division remote damage assessment of different vehicle types based on artificial intelligence random forest method | |
| CN106548136A (en) | A kind of wireless channel scene classification method | |
| CN106055776A (en) | Regional and remote damage-assessment system and method established based on artificial-intelligence supervised learning linear regression method for different types of vehicles | |
| CN106055779A (en) | Remote damage-assessment system and method established based on artificial intelligence semi-supervised learning logistic-regression method for different types of vehicles | |
| CN106056149A (en) | System and method for establishing working condition division remote damage assessment of different vehicle types based on artificial intelligence unsupervised learning principal component analysis method | |
| CN106067035A (en) | The long-range loss assessment system of different automobile types partial objectives for and method is set up based on artificial intelligence's supervised learning traditional decision-tree | |
| CN106096624A (en) | The long-range loss assessment system of different automobile types divided working status and method is set up based on artificial intelligence | |
| CN112991394A (en) | KCF target tracking method based on cubic spline interpolation and Markov chain | |
| CN106056453A (en) | System and method for establishing working condition division remote damage assessment of different vehicle types based on artificial intelligence semi-supervised learning clustering hypothesis method | |
| CN106056152A (en) | System and method for establishing target division remote damage assessment of different vehicle types based on artificial intelligence semi-supervised learning BIRCH method | |
| CN106067038A (en) | Point long-range loss assessment system of vehicle and a method is set up based on artificial intelligence's supervised learning Nae Bayesianmethod | |
| CN106067036A (en) | Set up different automobile types based on artificial intelligence's unsupervised learning K means method and divide the long-range loss assessment system of part and method | |
| CN106055891A (en) | Remote damage-assessment system and method established based on artificial intelligence Softmax regression method for different types of vehicles | |
| CN106056140A (en) | System and method for establishing working condition division remote damage assessment of different vehicle types based on artificial intelligence supervised learning linear regression method | |
| CN115236272A (en) | Gas sensor fault diagnosis method and device under multi-working condition and storage medium | |
| CN104615613A (en) | Global feature descriptor polymerization method | |
| CN104077608A (en) | Behavior recognition method based on sparsely coded slow characteristic functions | |
| CN114692831A (en) | Method of calculating variable convolution kernel for variable resolution, storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161026 |
|
| RJ01 | Rejection of invention patent application after publication |