CN106650157A - Method, device and system for vehicle part fault probability estimation - Google Patents
Method, device and system for vehicle part fault probability estimation Download PDFInfo
- Publication number
- CN106650157A CN106650157A CN201611264872.4A CN201611264872A CN106650157A CN 106650157 A CN106650157 A CN 106650157A CN 201611264872 A CN201611264872 A CN 201611264872A CN 106650157 A CN106650157 A CN 106650157A
- Authority
- CN
- China
- Prior art keywords
- parts
- driving behavior
- wear
- parts wear
- fault rate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Aviation & Aerospace Engineering (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Automation & Control Theory (AREA)
- Vehicle Cleaning, Maintenance, Repair, Refitting, And Outriggers (AREA)
Abstract
Disclosed are a method, a device and a system for vehicle part fault probability estimation. The method includes acquiring driving behavior data of vehicle within a first duration preset, processing the driving behavior data within the first duration, inputting the processed driving behavior data to a part wear model has been trained, calculating the corresponding fault probability of parts at the end of the first duration according to the part wear model, and then outputting the fault probability. Input parameter of the part wear model includes driving behavior; the middle parameter refers to wear degree, the output parameter refers to the fault probability of the parts; the wear degree and the driving behavior are of first mapping relation, and the fault probability of the parts and the wear degree of the parts are of a second mapping relation. According to the scheme, by the method, the device and the system, the fault probability of vehicle parts can be acquired in time so as to improve driving safety.
Description
Technical field
The present invention relates to vehicle component technical field, more particularly to vehicle component fault rate evaluation method,
Apparatus and system.
Background technology
With the improvement of people's living standards, increasing family possesses automobile.Automobile in the life of people, as
Important walking-replacing tool, occupies more and more important role.As time goes on and continue on, its each parts can go out
Now different degrees of abrasion.
Under normal circumstances, user only can know each parts when 4s shops are overhauled to vehicle or maintained,
Abrasion condition, and then calculate vehicle parts fault rate.For infrequently vehicle being overhauled or protected
Foster user, when only vehicle breaks down, just knows that the parts in vehicle are damaged.
In prior art, user is difficult to know the fault rate of parts in vehicle in time, when using in vehicle
When breaking down in journey, traffic safety may be affected.
The content of the invention
Present invention solves the technical problem that being the fault rate for how knowing vehicle component in time, driving peace is improved
Entirely.
To solve above-mentioned technical problem, the embodiment of the present invention provides a kind of vehicle component fault rate estimation side
Method, including:Obtain driving behavior data of the vehicle in default first duration;To the driving behavior data in first duration
Processed;By the driving behavior data input after process to the parts wear model trained, ground using the parts
Model is damaged, finish time corresponding fault rate of the parts in first duration is calculated, and is exported, wherein,
The |input paramete of the parts wear model includes driving behavior, and intermediate parameters are the degree of wear, and output parameter is parts
Fault rate, the degree of wear and driving behavior have the first mapping relations, and the failure of the parts occurs general
Rate has the second mapping relations with the degree of wear of the parts, obtains respectively being driven in the parts wear model by training
Sail behavior corresponding weight and the degree of wear corresponding power in second mapping relations in first mapping relations
Weight.
Alternatively, the driving behavior includes following at least one:It is anxious accelerate, anxious deceleration, idling, hypervelocity, zig zag and
Bring to a halt.
Alternatively, calculate the parts first duration finish time corresponding fault rate it
Afterwards, also include:According to the fault rate of calculated parts, with reference to the driving behavior number in first duration
According to using the parts wear model trained, impact of each driving behavior of analysis to the fault rate of parts;
For impact of each driving behavior to the component failure probability of happening, driving behavior Improving advice is generated, and exported.
Alternatively, the driving behavior data in first duration are processed, including:
Alternatively, driving range of the vehicle in first duration is obtained;Count and respectively driven in first duration
The generation total degree of behavior is sailed, and calculates the number of times of each driving behavior generation in default mileage as the driving behavior after process
Data.
Alternatively, methods described also includes:When the fault rate of calculated parts is sent out more than preset failure
During raw probability, failure prompting message is sent.
Alternatively, the |input paramete of the parts wear model also includes parts wear relation factor, the abrasion
Degree has the 3rd mapping relations with the driving behavior and parts wear relation factor, and by training described zero is respectively obtained
The each driving behavior and parts wear relation factor corresponding weight in the 3rd mapping relations in component wear model.
Alternatively, methods described also includes:Obtain the corresponding parts wear association of the vehicle in first duration
Factor data, and be input into the parts wear model trained.
Alternatively, the parts wear relation factor includes following at least one:Weather conditions, Road Factor, region
Factor, nature factor, record of examination and load-carrying.
Alternatively, the parts wear model is trained in the following way:Training sample is obtained, in the training sample
Including the degree value of driving behavior data, parts wear relation factor data and corresponding parts wear;Parts are ground
The degree value of parts wear is less than institute by the degree value of damage more than or equal to the training sample mark positive sample of predetermined threshold value
The training sample for stating threshold value is labeled as negative sample;Using logistic regression algorithm, ground according to the driving behavior data and parts
Relation factor data are damaged, logistic regression training is carried out to the positive sample and the negative sample, obtain the parts wear mould
Type.
Alternatively, the parts wear model is trained in the following way:Training sample is obtained, in the training sample
Including driving behavior data and the degree value of parts and corresponding parts wear;By the degree value of parts wear be more than or
Person is equal to the training sample mark positive sample of predetermined threshold value, and the degree value of parts wear is less than into the training sample of the threshold value
It is labeled as negative sample;Using logistic regression algorithm, according to the driving behavior data, the positive sample and the negative sample are entered
Row logistic regression is trained, and obtains the parts wear model.
Alternatively, when to the parts wear model training, also include:Select at random from the training sample pre-
If the training sample of number is used as test sample;The parts wear model is carried out accurately using test sample checking
Degree checking.
Alternatively, it is described that accuracy validation is carried out to the parts wear model using test sample checking, wrap
Include:The test sample is randomly divided into into N groups test subsample, N is natural number, and N >=3;Using wherein any N-1 groups test
Subsample carries out accuracy validation to one group of test subsample outside the N-1 groups, until N groups test subsample quilt
Checking.
Alternatively, methods described also includes:When default model modification trigger event is detected, the vehicle is obtained
Driving behavior data are updated training using the more new samples as more new samples to the parts wear model, and
To update and train the parts wear model for obtaining as the parts wear model trained.
The embodiment of the present invention also provides a kind of vehicle component fault rate estimation device, including:Acquiring unit, place
Reason unit, input block, parts wear model, computing unit and the first output unit, wherein:The acquiring unit, is suitable to
Obtain driving behavior data of the vehicle in default first duration;The processing unit, is suitable to driving in first duration
Sail behavioral data to be processed;The input block, is suitable to the driving behavior data input after the processing unit processes extremely
The parts wear model trained;The computing unit, is suitable for use with the parts wear model, calculates described zero
Finish time corresponding fault rate of the part in first duration;The |input paramete bag of the parts wear model
Include driving behavior, intermediate parameters are the degree of wear, output parameter for parts fault rate, the degree of wear with drive
Behavior is sailed with the first mapping relations, the fault rate of the parts has second with the degree of wear of the parts
Mapping relations, obtain each driving behavior in the parts wear model corresponding in first mapping relations by training
The weight and the degree of wear corresponding weight in second mapping relations;First output unit, is suitable to output and calculates
The fault rate of the parts for obtaining.
Alternatively, the driving behavior includes following at least one:It is anxious accelerate, anxious deceleration, idling, hypervelocity, zig zag and
Bring to a halt.
Alternatively, described device also includes:Analytic unit and the second output unit, wherein:The analytic unit, is suitable to
The parts are calculated after the finish time corresponding fault rate of first duration, according to calculated zero
The fault rate of part, with reference to the driving behavior data in first duration, using the parts mill trained
Model is damaged, impact of each driving behavior to the fault rate of parts is analyzed, for each driving behavior to the parts
Fault rate impact, generate driving behavior Improving advice;Second output unit, is suitable to export the analysis list
The driving behavior Improving advice that unit generates.
Alternatively, the processing unit, is suitable to obtain driving range of the vehicle in first duration, counts institute
The generation total degree of each driving behavior in the first duration is stated, and calculates the number of times conduct that each driving behavior occurs in default mileage
Driving behavior data after process.
Alternatively, described device also includes:Reminding unit, is suitable to when calculated component failure probability of happening is more than
During preset failure probability of happening, failure prompting message is sent.
Alternatively, the |input paramete of the parts wear model also includes parts wear relation factor, the abrasion
Degree has the 3rd mapping relations with the driving behavior and parts wear relation factor, and by training described zero is respectively obtained
The each driving behavior and parts wear relation factor corresponding weight in the 3rd mapping relations in component wear model.
Alternatively, the acquiring unit, is further adapted for obtaining the corresponding parts wear of the vehicle in first duration
Relation factor data, and be input into the parts wear model trained.
Alternatively, the parts wear relation factor includes following at least one:Weather conditions, Road Factor, region
Factor, nature factor, record of examination and load-carrying.
Alternatively, described device also includes the first training unit, is suitable for use with following manner and trains the parts wear
Model:Training sample is obtained, the training sample includes driving behavior data, parts wear relation factor data and correspondence
Parts wear degree value;The degree value of parts wear is marked just more than or equal to the training sample of predetermined threshold value
Sample, negative sample is labeled as by the degree value of parts wear less than the training sample of the threshold value;Using logistic regression algorithm,
According to the driving behavior data and parts wear relation factor data, logic is carried out to the positive sample and the negative sample
Regression training, obtains the parts wear model.
Alternatively, described device also includes the second training unit, is suitable for use with following manner and trains the parts wear
Model:Training sample is obtained, the training sample includes the degree value of driving behavior data and corresponding parts wear;Will
The degree value of parts wear marks positive sample more than or equal to the training sample of predetermined threshold value, by the degree of parts wear
Value is labeled as negative sample less than the training sample of the threshold value;It is right according to the driving behavior data using logistic regression algorithm
The positive sample and the negative sample carry out logistic regression training, obtain the parts wear model.
Alternatively, described device also includes test cell, is suitable to select preset number at random from the training sample
Training sample carries out accuracy validation using test sample checking as test sample to the parts wear model.
Alternatively, the test cell, is suitable to for the test sample to be randomly divided into N groups test subsample, and N is nature
Number, and N >=3, are carried out accurately using wherein any N-1 groups test subsample to one group of test subsample outside the N-1 groups
Degree checking, until N groups test subsample is verified.
Alternatively, described device also includes updating block, is suitable to, when default model modification trigger event is detected, obtain
The driving behavior data of the vehicle are taken as more new samples, the parts wear model is carried out using the more new samples
Training, and the parts wear model that renewal training is obtained are updated as the parts wear model trained.
The embodiment of the present invention also provides a kind of vehicle component fault rate estimating system, including:Data acquisition is filled
Put and the vehicle component fault rate of any of the above-described estimates device, wherein:The data acquisition unit is suitable to obtain
Driving behavior data of the vehicle in default first duration.
Alternatively, the data acquisition unit, is suitable to the driving states sensor by installing on the vehicle and gathers institute
Driving behavior data of the vehicle in default first duration are stated, or the vehicle is obtained from vehicle storage device default first
Driving behavior data in duration.
Compared with prior art, the technical scheme of the embodiment of the present invention has the advantages that:
According to the driving behavior data in the vehicle in default first duration for collecting, using zero for having trained
Part wear model, according to the first mapping relations in the parts wear model between each driving behavior and the degree of wear, obtains
To the degree of wear of the parts in default first duration, according to the abrasion of the fault rate of parts and the parts
The second mapping relations between degree, obtain the fault rate of parts.Different driving behaviors affects vehicle component
Rate of depreciation, and the degree of wear of vehicle component affect fault rate, due to the driving behavior data ratio in vehicle
It is easier to obtain, is user to vehicle such that it is able to know the fault rate of vehicle component in time according to driving behavior
Maintenance provides more objective reference frame, improves vehicle safety and reliability.
Further, according to the driving behavior number in the fault rate and the first duration of calculated parts
According to, impact of the driving behavior to the fault rate of parts is analyzed, and driving behavior Improving advice is generated, contribute to culture
The good driving behavior of user, and reduce the impact that bad steering behavior brings to the fault rate of parts.
Further, when the fault rate of calculated parts is more than preset failure probability of happening, send
Failure prompting message so that user can in time know the fault rate of parts, is easy in time keep in repair parts
Or change, improve traffic safety.
Further, when the fault rate of parts is estimated, it is considered to parts wear relation factor, can improve
The accuracy of the fault rate estimation of parts.
Further, when default model modification trigger event is detected, the driving behavior data of the vehicle are obtained
As more new samples, be updated training to the parts wear model, and using the parts wear model after renewal as
The parts wear model trained, by being updated to the parts wear model, can cause the parts to grind
Damage model is higher with the matching degree of the vehicle, such that it is able to the accuracy of further parts estimation.
Description of the drawings
Fig. 1 is a kind of flow chart of vehicle component fault rate evaluation method in the embodiment of the present invention;
Fig. 2 is the structural representation that a kind of vehicle component fault rate estimates device in the embodiment of the present invention;
Fig. 3 is the structural representation that another kind of vehicle component fault rate estimates device in the embodiment of the present invention.
Specific embodiment
In prior art, user to vehicle generally when being overhauled or being maintained, or vehicle is when breaking down, and just knows
Certain parts in dawn vehicle have been damaged.And in other cases, user then cannot be known in time a certain zero in vehicle
The fault rate of part.
To solve the above problems, in embodiments of the present invention, according in the vehicle in default first duration for collecting
Driving behavior data, using the parts wear model trained, according to each in the parts wear model row are driven
It is the first mapping relations between the degree of wear, the degree of wear of the parts in default first duration is obtained, according to zero
The second mapping relations between the degree of wear of the fault rate of part and the parts, the failure for obtaining parts occurs
Probability.Different driving behaviors affects the rate of depreciation of vehicle component, and the degree of wear of vehicle component affects failure to send out
Raw probability, because the driving behavior data in vehicle are easier to obtain, such that it is able to know vehicle in time according to driving behavior
The fault rate of parts, for user more objective reference frame is provided car inspection and repair, improves vehicle safety and can
By property.
It is understandable to enable above-mentioned purpose of the invention, feature and beneficial effect to become apparent from, below in conjunction with the accompanying drawings to this
The specific embodiment of invention is described in detail.
With reference to Fig. 1, a kind of flow process of vehicle component fault rate evaluation method in the embodiment of the present invention is given
Figure.It is described in detail with reference to concrete steps.
Step 11, obtains driving behavior data of the vehicle in default first duration.
In being embodied as, can be obtained by the vehicle diagnosing system (On-Board Diagnostic, OBD) of vehicle
Driving behavior data of the vehicle in first duration, it is also possible to carried out with the corresponding controllers in vehicle by vehicle-mounted TBOX
Communication, obtains driving behavior data of the vehicle in first duration, can also pass through the various biographies installed on vehicle
Sensor obtains in real time driving behavior data, and stores.It is understood that other intelligence that can also pass through to be installed on vehicle are hard
Part is obtaining driving behavior data of the vehicle in first duration.
In being embodied as, the driving behavior can include:Urgency acceleration, anxious deceleration, idling, hypervelocity, zig zag and urgency
One or more in brake.
In being embodied as, fault rate of the various driving behaviors to vehicle component can be counted according to big data
Impact, and using zero employed in fault rate estimation process of the higher driving behavior of the degree of association as parts
The |input paramete of part wear model, and it is not limited to driving behavior enumerated above.
In being embodied as, because the abrasion of parts is a long-term process, first duration can be nature
Week, calendar month, natural season, it is also possible to be arranged as required to concrete duration.For example, 45 days, 52 days etc., in the embodiment of the present invention
In the concrete value of first duration is not limited.
Driving behavior data in first duration are processed by step 12.
In being embodied as, the driving behavior data in first duration can be processed, count each and drive row
It is the number of times occurred in the first duration.
Step 13, by the driving behavior data input after process to the parts wear model trained, using described zero
Component wear model, calculates the fault rate of the parts, and exports.
In being embodied as, the |input paramete of the parts wear model includes driving behavior, and intermediate parameters are abrasion
Degree, output parameter is the fault rate of parts, and the degree of wear has the first mapping relations, institute with driving behavior
The degree of wear of the fault rate and the parts of stating parts has the second mapping relations, obtains described by training
In parts wear model each driving behavior in first mapping relations corresponding weight and the degree of wear described
Corresponding weight in two mapping relations.
The number of times that calculated each driving behavior is occurred in default mileage is input into as |input paramete
Training the parts wear model in, according to each driving behavior in first mapping relations corresponding weight and mill
Damage degree corresponding weight in second mapping relations, calculates the fault rate of the parts.When being calculated
After the fault rate of the parts of the vehicle, result of calculation is exported.
As shown in the above, the driving behavior data in the vehicle in default first duration for collecting, adopt
The parts wear model trained, according in the parts wear model between each driving behavior and the degree of wear
First mapping relations, obtain the degree of wear of the parts in default first duration, according to the fault rate of parts with
The second mapping relations between the degree of wear of the parts, obtain the fault rate of parts.Different driving rows
To affect the rate of depreciation of vehicle component, and the degree of wear of vehicle component affects fault rate, due in vehicle
Driving behavior data be easier to obtain, such that it is able to know that it is general that the failure of vehicle component occurs in time according to driving behavior
Rate, for user more objective reference frame is provided car inspection and repair, improves vehicle safety and reliability.
In being embodied as, impact of the bad steering behavior to the fault rate of the parts of vehicle is avoided as far as possible,
Traffic safety is improved, and improves the bad steering behavior of user.In an embodiment of the present invention, the parts are being calculated
After the finish time corresponding fault rate of first duration, can be with according to the generation of the failure of the parts
Probability, with reference to the driving behavior data in first duration, using the parts wear model trained, analyzes user
Impact of each driving behavior to the fault rate of parts, the failure of the parts is occurred for each driving behavior
The impact of probability, generates driving behavior Improving advice, and exports.
For example, show in the driving behavior data in first duration, the number of times brought to a halt is 15 times, meeting of bringing to a halt
The abrasion of the brake block of vehicle is brought, as the degree of wear of brake block increases, the fault rate of brake block is also carried therewith
It is high.After the fault rate for being calculated brake block, can provide driving behavior improvement and build according to the number of times brought to a halt
View:Set up and reduce the number of times brought to a halt.
And for example, the driving behavior data display in first duration, the number of times for suddenly accelerating is 10 times, and what is suddenly slowed down is secondary
Number is 8 times, and the number of times of zig zag is 13 times, because anxious acceleration, anxious deceleration and zig zag can affect the mill of the tire of vehicle
Damage, so as to increase with the degree of wear of tire, the fault rate of tire is also improved therewith.According in first duration
The calculated tire of driving behavior data fault rate, and occur anxious accelerate, anxious slow down and take a sudden turn
Number of times, provide driving behavior Improving advice:Smooth ride, reduces the anxious number of times for accelerating, suddenly slowing down and take a sudden turn as far as possible.
In being embodied as, after driving behavior Improving advice is obtained, the car entertainment device output of association can be passed through
The driving behavior Improving advice, it is also possible to by the output of the mobile terminals such as the mobile phone driving behavior Improving advice, may be used also
With the output of other-end equipment.It is understood that audio devices can also be adopted with audible output.
In being embodied as, for the counting accuracy of the fault rate of the parts of vehicle.It is real in the present invention one
In applying example, driving range of the vehicle in first duration is obtained, count each driving behavior in first duration
Generation total degree, and the number of times of each driving behavior generation in default mileage is calculated as the driving behavior data after process.
In being embodied as, when the fault rate of calculated parts is more than preset failure probability of happening,
Failure prompting message is sent, so that user can in time know the fault rate of parts, is easy in time to parts
Keeped in repair or changed, reduce, when user uses vehicle, because of the inconvenience that component failure brings to user, and improving driving
Safety.
In being embodied as, due to vehicle environment in the process of moving it is complex, the failure of vehicle component
Probability of happening is not only affected by driving behavior, while also suffering from the impact of other various factors.To cause to vehicle
The degree of wear of parts carries out more analyzing comprehensively and accurately, accurate with the fault rate that raising obtains parts
Degree.In an embodiment of the present invention, the |input paramete of the parts wear model can also include parts wear association because
Element, the degree of wear has the 3rd mapping relations with the driving behavior and parts wear relation factor, by training point
Each driving behavior and parts wear relation factor are not obtained in the parts wear model in the 3rd mapping relations
Corresponding weight.
In being embodied as, the corresponding parts wear relation factor number of the vehicle in first duration can be obtained
According to, and be input into the parts wear model trained.
In being embodied as, can be using the driving behavior data and parts wear relation factor data, by instruction
Driving behavior data described in the parts wear model and parts wear relation factor are got in the 3rd mapping
Corresponding weight in relation, calculates the fault rate of the parts.
In being embodied as, the parts wear relation factor can include weather conditions, Road Factor, region because
One or more in element, nature factor, record of examination and load-carrying.
In being embodied as, can adopt Controlling UEP method, analyze choose parts abrasion relation factor with
Relation between parts wear degree, and according to analysis result, using degree of correlation large effect factor as component failure
The corresponding abrasion relation factor of probability of happening.
For example, the smooth degree on the road surface of different grades of road is different, different to the effect of attrition of tire;Rainy day,
Impact of the same road to the abrasion of tire is also different when snowy day and fine day;Generally northerner's personality is more anxious, during driving
Like anxious acceleration, anxious deceleration etc., southerner's personality compares Wen Wan, and driving style is also more stable, then the personality of different people
Driving style is also different to the degree of wear of the tire of vehicle.Show in record of examination parts carried out maintenance or more
Change, then can change in the situation of the abrasion of vehicle component, therefore the fault rate of parts also changes therewith,
Especially by the parts of part replacement Cheng Xin after, the fault rate of the parts need to re-start estimation,
In parts wear model, using the corresponding weight of record of examination, the fault rate of the parts can be adjusted
It is whole so as to more conform to the actual conditions of vehicle component abrasion.
In being embodied as, parts wear model can be trained using substantial amounts of training sample, obtain each zero
Weight corresponding to the abrasion relation factor of part.
In being embodied as, the parts wear model can be trained using various ways.
In an embodiment of the present invention, training sample is obtained, the training sample includes driving behavior data, parts
The degree value of abrasion relation factor data and corresponding parts wear.By the degree value of parts wear more than or equal to pre-
If the training sample mark positive sample of threshold value, the degree value of parts wear is labeled as bearing less than the training sample of the threshold value
Sample.Using logistic regression algorithm, according to the driving behavior data and parts wear relation factor data, to the positive sample
This and the negative sample carry out logistic regression training, obtain the parts wear model.
In an alternative embodiment of the invention, training sample is obtained, the training sample includes driving behavior data and right
The degree value of the parts wear answered, the degree value of parts wear is marked more than or equal to the training sample of predetermined threshold value
Positive sample, negative sample is labeled as by the degree value of parts wear less than the training sample of the threshold value.Calculated using logistic regression
Method, according to the driving behavior data, to the positive sample and the negative sample logistic regression training is carried out, and obtains described zero
Part wear model.
In being embodied as, during parts wear model training, in order to improve the accurate of parts wear model
Degree, to improve the estimation accuracy of the work probability of happening to parts.In an embodiment of the present invention, in parts wear mould
In type training process, the training sample of preset number can be at random selected from the training sample as test sample, be adopted
The test sample checking carries out accuracy validation to the parts wear model.
In being embodied as, a certain proportion of training sample can be chosen as test sample from the training sample.
For example, 20% is chosen from training sample as test sample.
In being embodied as, in the following way accuracy validation can be carried out to the parts wear model:By institute
State test sample and be randomly divided into N groups test subsample, N is natural number, and N >=3;Subsample is tested using wherein any N-1 groups
Accuracy validation is carried out to one group of test subsample outside the N-1 groups, until N groups test subsample is verified.
In an embodiment of the present invention, N values are 3, and the test sample is randomly divided into into three groups of test subsamples, are adopted
Two groups of test subsamples carry out accuracy validation to another set test subsample, until three groups of test subsamples are tested
Card.It is understood that in actual applications, N can also take 4, can also take 5 etc., it is also possible to take other values, concrete value can
To be set according to practical application scene and required precision, specifically do not limit.
For example, the test sample is divided into into tri- groups of A, B and C, two groups C groups is verified using A, B, inspection adopts institute
Parts wear model is stated to the fault rate of the test sample estimation in C groups and the physical fault probability of happening phase of C groups
Than whether meeting accuracy requirement.Equally, using A, C two groups B groups are verified, two groups A groups are verified using B, C.
In being embodied as, because the road conditions of the alternating in season, the change of parts quality, institute's travel change etc.,
Abrasion relation factor and driving behavior to vehicle component also constantly changing, in order that parts wear model with
The matching degree of the parts of the vehicle is higher.In an embodiment of the present invention, default model modification triggering thing is being detected
During part, the driving behavior data of the vehicle are obtained as more new samples, more new samples are to the parts wear using described in
Model is updated training, and the parts wear model that renewal training is obtained as the parts wear model trained.
By being updated to the parts wear model, the parts wear model and the vehicle can be caused
Parts matching degree it is higher, such that it is able to further improve the degree of accuracy to the estimation of component failure probability of happening.
In being embodied as, default model modification trigger event can be default second duration, whenever reaching second
When long, the parts wear model is updated automatically.Second duration can be two months, or a season
Degree, can also be year.The concrete value of second duration can be set according to the value of first duration, or according to
Actual application scenarios are set.
Default model modification trigger condition can also be the change in region residing for vehicle, for example, live in before user
South, settled down to northeast later, due to northeast environment and it is southern differ greatly, can be main according to the vehicle for detecting
Zone of action is updated to parts wear model.
In order to those skilled in the art are better understood from and realize the present invention, in the embodiment of the present invention a kind of car is additionally provided
Component failure probability of happening estimation device.
With reference to Fig. 2, the structure that a kind of vehicle component fault rate in the embodiment of the present invention estimates device is given
Schematic diagram.The vehicle component fault rate estimation device 20 can include acquiring unit 21, processing unit 22, input
Unit 23, parts wear model 24, the output unit 26 of computing unit 25 and first, wherein:
The acquiring unit 21, is suitable to obtain driving behavior data of the vehicle in default first duration;
The processing unit 22, is suitable to process the driving behavior data in first duration;
The input block 23, is suitable to the processing unit 22 by driving behavior data input after process to having trained
The parts wear model 24;
The computing unit 25, is suitable for use with the parts wear model 24, according to the driving after the process for receiving
Behavioral data calculates finish time corresponding fault rate of the parts in first duration;
The |input paramete of the parts wear model 24 includes driving behavior, and intermediate parameters are the degree of wear, output ginseng
Number is the fault rate of parts, and the degree of wear has the first mapping relations with driving behavior, the parts
Fault rate has the second mapping relations with the degree of wear of the parts, and by training the parts wear is obtained
Each driving behavior corresponding weight and degree of wear in first mapping relations are closed in the described second mapping in model 24
Corresponding weight in system;
First output unit 26, is suitable to export the fault rate of the calculated parts.
In being embodied as, can be obtained by the vehicle diagnosing system (On-Board Diagnostic, OBD) of vehicle
Driving behavior data of the vehicle in first duration, it is also possible to carried out with the corresponding controllers in vehicle by vehicle-mounted TBOX
Communication, obtains driving behavior data of the vehicle in first duration, can also pass through the various biographies installed on vehicle
Sensor obtains in real time driving behavior data, and stores.It is understood that other intelligence that can also pass through to be installed on vehicle are hard
Part is obtaining driving behavior data of the vehicle in first duration.
In being embodied as, the driving behavior can include:Urgency acceleration, anxious deceleration, idling, hypervelocity, zig zag and urgency
One or more in brake.
From the foregoing, it will be observed that the driving behavior data in the vehicle in default first duration for collecting, using described in
The parts wear model of training, according to first reflecting between each driving behavior and the degree of wear in the parts wear model
Relation is penetrated, the degree of wear of the parts in default first duration is obtained, according to the fault rate and described zero of parts
The second mapping relations between the degree of wear of part, obtain the fault rate of parts.Different driving behaviors affects
The rate of depreciation of vehicle component, and the degree of wear of vehicle component affects fault rate, due to the driving in vehicle
Behavioral data is easier to obtain, and such that it is able to know the fault rate of vehicle component in time according to driving behavior, is
User provides more objective reference frame to car inspection and repair, improves vehicle safety and reliability.
In being embodied as, in order to improve the bad steering behavior of user, bad steering behavior is reduced as far as possible to vehicle
The impact of the fault rate of parts, improves traffic safety.In an embodiment of the present invention, the vehicle component failure
Probability of happening estimation device 20 can also include:The output unit 32 of analytic unit 31 and second.Fig. 3 is specifically referred to, is given
The embodiment of the present invention in another kind of vehicle component fault rate estimate the structural representation of device.
The analytic unit 31, is suitable in the finish time corresponding failure for calculating the parts in first duration
After probability of happening, according to the fault rate of calculated parts, with reference to the driving behavior in first duration
Data, using the parts wear model trained, analyze shadow of each driving behavior to the fault rate of parts
Ring, generate driving behavior Improving advice;
Second output unit 32, is suitable to export the driving behavior Improving advice of the generation of the analytic unit 31.
In being embodied as, the processing unit 22 is suitable to obtain in driving of the vehicle in first duration
Journey, counts the generation total degree of each driving behavior in first duration, and calculates each driving behavior generation in default mileage
Number of times as the driving behavior data after process.
In being embodied as, the vehicle component fault rate estimation device 20 can also include reminding unit
33.The reminding unit 33 is suitable to when the fault rate of calculated parts is more than preset failure probability of happening,
Send failure prompting message.
In being embodied as, the |input paramete of the parts wear model 24 also includes parts wear relation factor,
The degree of wear has the 3rd mapping relations with the driving behavior and parts wear relation factor, is obtained respectively by training
The each driving behavior and parts wear relation factor correspondence in the 3rd mapping relations in the parts wear model
Weight.
In being embodied as, the acquiring unit 21 is further adapted for obtaining the vehicle corresponding zero in first duration
Component wear relation factor data, and be input into the parts wear model 24 trained.
In being embodied as, the parts wear relation factor includes weather conditions, Road Factor, regional factor, property
One kind in lattice factor, record of examination and load-carrying, it is also possible to various in including above-mentioned factor.
In being embodied as, the vehicle component fault rate estimation device 20 can also include that the first training is single
Unit 34.First training unit 34 is suitable for use with following manner and trains the parts wear model:Obtain training sample, institute
Stating training sample includes the degree of driving behavior data, parts wear relation factor data and corresponding parts wear
Value;The degree value of parts wear is marked into positive sample more than or equal to the training sample of predetermined threshold value, by parts wear
Degree value be labeled as negative sample less than the training sample of the threshold value;Using logistic regression algorithm, according to the driving behavior
Data and parts wear relation factor data, to the positive sample and the negative sample logistic regression training is carried out, and obtains institute
State parts wear model 24.
In being embodied as, the vehicle component fault rate estimation device 20 can also include that the second training is single
Unit 35.Second training unit 35 is suitable for use with following manner and trains the parts wear model:Obtain training sample, institute
Stating training sample includes the degree value of driving behavior data and parts wear;By the degree value of parts wear be more than or
Positive sample is marked equal to the training sample of predetermined threshold value, the degree value of parts wear is less than into the training sample mark of the threshold value
It is designated as negative sample;Using logistic regression algorithm, according to the driving behavior data, the positive sample and the negative sample are carried out
Logistic regression is trained, and obtains the parts wear model 24.
In being embodied as, the vehicle component fault rate estimation device 20 can also include test cell
36.The test cell 36, is suitable to be selected at random from the training sample training sample of preset number as test sample,
Accuracy validation is carried out to the parts wear model 24 using test sample checking.
In being embodied as, the test cell 36 is suitable to for the test sample to be randomly divided into N groups test subsample, N
For natural number, and N >=3, subsample is tested using wherein any N-1 groups one group of test subsample outside the N-1 groups is entered
Row accuracy validation, until N groups test subsample is verified.
In being embodied as, the vehicle component fault rate estimation device 20 can also include updating block
37.The updating block 37, is suitable to obtain the driving behavior of the vehicle when default model modification trigger event is detected
Data are updated training to the parts wear model as more new samples using the more new samples, and will update instruction
The parts wear model for getting is used as the parts wear model trained.
In being embodied as, the vehicle component fault rate estimation device 20 can be an independent hardware
Device, or server, can also be application software or client, on corresponding equipment.May be appreciated
It is that the vehicle component fault rate estimation device 20 can also have other forms.
In being embodied as, the vehicle component fault rate estimates the concrete operating principle of device 20 and work
Flow process may refer to the description in the vehicle component fault rate evaluation method provided in the above embodiment of the present invention, this
Place repeats no more.
The embodiment of the present invention also provides a kind of vehicle component fault rate estimating system.The vehicle component event
Barrier probability of happening estimating system can include:Data acquisition unit and any one vehicle component such as offer in above-mentioned embodiment
Fault rate estimates device.
In being embodied as, the data acquisition unit can obtain driving row of the vehicle in default first duration
For data.
In being embodied as, the data acquisition unit can pass through the driving states sensor installed on the vehicle
Gather driving behavior data of the vehicle in default first duration, it is also possible to the vehicle is obtained from vehicle storage device and is existed
Driving behavior data in default first duration.
In being embodied as, the operation principle and workflow of vehicle component fault rate estimating system can join
The description seen in the vehicle component fault rate evaluation method and device provided in the above embodiment of the present invention, herein not
Repeat again.
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is can
Completed with instructing the hardware of correlation by program, the program can be stored in a computer-readable recording medium, storage
Medium can include:ROM, RAM, disk or CD etc..
Although present disclosure is as above, the present invention is not limited to this.Any those skilled in the art, without departing from this
In the spirit and scope of invention, can make various changes or modifications, therefore protection scope of the present invention should be with claim institute
The scope of restriction is defined.
Claims (28)
1. a kind of vehicle component fault rate evaluation method, it is characterised in that include:
Obtain driving behavior data of the vehicle in default first duration;
Driving behavior data in first duration are processed;
By the driving behavior data input after process to the parts wear model trained, using the parts wear mould
Type, calculates finish time corresponding fault rate of the parts in first duration, and exports, wherein, it is described
The |input paramete of parts wear model includes driving behavior, and intermediate parameters are the degree of wear, and output parameter is the event of parts
Barrier probability of happening, the degree of wear and driving behavior have the first mapping relations, the fault rate of the parts with
The degree of wear of the parts has the second mapping relations, obtains each driving in the parts wear model by training and goes
It is the corresponding weight and the degree of wear corresponding weight in second mapping relations in first mapping relations.
2. vehicle component fault rate evaluation method according to claim 1, it is characterised in that
The driving behavior includes following at least one:
It is anxious to accelerate, anxious deceleration, idling, hypervelocity, take a sudden turn and bring to a halt.
3. vehicle component fault rate evaluation method according to claim 2, it is characterised in that
In the calculating parts after the finish time corresponding fault rate of first duration,
Also include:
According to the fault rate of calculated parts, with reference to the driving behavior data in first duration, adopt
The parts wear model trained, analyzes impact of each driving behavior to the fault rate of parts;
For impact of each driving behavior to the component failure probability of happening, driving behavior Improving advice is generated, and exported.
4. vehicle component fault rate evaluation method according to claim 2, it is characterised in that
The driving behavior data in first duration are processed, including:
Obtain driving range of the vehicle in first duration;
The generation total degree of each driving behavior in first duration is counted, and calculates each driving behavior and occurred in default mileage
Number of times as the driving behavior data after process.
5. vehicle component fault rate evaluation method according to claim 1, it is characterised in that
Also include:When the fault rate of calculated parts is more than preset failure probability of happening,
Send failure prompting message.
6. vehicle component fault rate evaluation method according to claim 1, it is characterised in that
The |input paramete of the parts wear model also includes parts wear relation factor, and the degree of wear is driven with described
Sail behavior and parts wear relation factor has the 3rd mapping relations, the parts wear model is respectively obtained by training
In each driving behavior and parts wear relation factor corresponding weight in the 3rd mapping relations.
7. vehicle component fault rate evaluation method according to claim 6, it is characterised in that
Also include:
The corresponding parts wear relation factor data of the vehicle in first duration are obtained, and is input into described and is trained
Parts wear model.
8. vehicle component fault rate evaluation method according to claim 6, it is characterised in that
The parts wear relation factor includes following at least one:
Weather conditions, Road Factor, regional factor, nature factor, record of examination and load-carrying.
9. vehicle component fault rate evaluation method according to claim 6, it is characterised in that
The parts wear model is trained in the following way:
Training sample is obtained, the training sample includes driving behavior data, parts wear relation factor data and correspondence
Parts wear degree value;
The degree value of parts wear is marked into positive sample more than or equal to the training sample of predetermined threshold value, by parts wear
Degree value be labeled as negative sample less than the training sample of the threshold value;
Using logistic regression algorithm, according to the driving behavior data and parts wear relation factor data,
Logistic regression training is carried out to the positive sample and the negative sample, the parts wear model is obtained.
10. vehicle component fault rate evaluation method according to claim 1, it is characterised in that
The parts wear model is trained in the following way:
Training sample is obtained, the training sample includes the journey of driving behavior data and parts and corresponding parts wear
Angle value;
The degree value of parts wear is marked into positive sample more than or equal to the training sample of predetermined threshold value, by parts wear
Degree value be labeled as negative sample less than the training sample of the threshold value;
Using logistic regression algorithm, according to the driving behavior data, logic is carried out to the positive sample and the negative sample and is returned
Return training, obtain the parts wear model.
The 11. vehicle component fault rate evaluation methods according to claim 9 or 10, it is characterised in that right
During the parts wear model training, also include:
The training sample of preset number is selected at random from the training sample as test sample;
Accuracy validation is carried out to the parts wear model using test sample checking.
12. vehicle component fault rate evaluation methods according to claim 11, it is characterised in that
It is described that accuracy validation is carried out to the parts wear model using test sample checking, including:
The test sample is randomly divided into into N groups test subsample, N is natural number, and N >=3;
Accuracy validation is carried out to one group of test subsample outside the N-1 groups using wherein any N-1 groups test subsample,
Until N groups test subsample is verified.
13. vehicle component fault rate evaluation methods according to claim 1, it is characterised in that
Also include:
When default model modification trigger event is detected, the driving behavior data of the vehicle are obtained as more new samples,
Training is updated to the parts wear model using the more new samples, and the parts wear that training is obtained will be updated
Model is used as the parts wear model trained.
A kind of 14. vehicle component fault rates estimate device, it is characterised in that include:Acquiring unit, processing unit,
Input block, parts wear model, computing unit and the first output unit, wherein:
The acquiring unit, is suitable to obtain driving behavior data of the vehicle in default first duration;
The processing unit, is suitable to process the driving behavior data in first duration;
The input block, is suitable to the driving behavior data input after the processing unit processes to described zero for having trained
Part wear model;
The computing unit, is suitable for use with the parts wear model, calculates knot of the parts in first duration
Beam moment corresponding fault rate;
The |input paramete of the parts wear model includes driving behavior, and intermediate parameters are the degree of wear, and output parameter is zero
The fault rate of part, the failure that the degree of wear has the first mapping relations, the parts with driving behavior is sent out
Raw probability has the second mapping relations with the degree of wear of the parts, is obtained in the parts wear model by training
Each driving behavior corresponding weight and degree of wear correspondence in second mapping relations in first mapping relations
Weight;
First output unit, is suitable to export the fault rate of the calculated parts.
15. vehicle component fault rates according to claim 14 estimate device, it is characterised in that the driving
Behavior includes following at least one:
It is anxious to accelerate, anxious deceleration, idling, hypervelocity, take a sudden turn and bring to a halt.
16. vehicle component fault rates according to claim 15 estimate device, it is characterised in that also include:
Analytic unit and the second output unit, wherein:
The analytic unit, is suitable to general in the finish time corresponding failure generation of first duration in the calculating parts
After rate, according to the fault rate of calculated parts, with reference to the driving behavior data in first duration, adopt
With the parts wear model trained, impact of each driving behavior to the fault rate of parts is analyzed, for
Impact of each driving behavior to the fault rate of the parts, generates driving behavior Improving advice;
Second output unit, is suitable to export the driving behavior Improving advice that the analytic unit is generated.
17. vehicle component fault rates according to claim 15 estimate device, it is characterised in that the process
Unit, is suitable to obtain driving range of the vehicle in first duration, counts each driving behavior in first duration
Generation total degree, and calculate number of times that each driving behavior occurs in default mileage as the driving behavior data after process.
18. vehicle component fault rates according to claim 14 estimate device, it is characterised in that also include:
Reminding unit, is suitable to, when calculated component failure probability of happening is more than preset failure probability of happening, send failure and carry
Awake information.
19. vehicle component fault rates according to claim 14 estimate device, it is characterised in that described zero
The |input paramete of part wear model also includes parts wear relation factor, the degree of wear and the driving behavior and zero
Part abrasion relation factor has the 3rd mapping relations, and by training each driving behavior in the parts wear model is respectively obtained
And parts wear relation factor corresponding weight in the 3rd mapping relations.
20. vehicle component fault rates according to claim 19 estimate device, it is characterised in that the acquisition
Unit, is further adapted for obtaining the corresponding parts wear relation factor data of the vehicle in first duration, and is input into institute
State the parts wear model trained.
21. vehicle component fault rates according to claim 19 estimate device, it is characterised in that described zero
Part abrasion relation factor includes following at least one:
Weather conditions, Road Factor, regional factor, nature factor, record of examination and load-carrying.
22. vehicle component fault rates according to claim 19 estimate devices, it is characterised in that also including the
One training unit, is suitable for use with following manner and trains the parts wear model:
Training sample is obtained, the training sample includes driving behavior data, parts wear relation factor data and correspondence
Parts wear degree value;
The degree value of parts wear is marked into positive sample more than or equal to the training sample of predetermined threshold value, by parts wear
Degree value be labeled as negative sample less than the training sample of the threshold value;
Using logistic regression algorithm, according to the driving behavior data and parts wear relation factor data, to the positive sample
This and the negative sample carry out logistic regression training, obtain the parts wear model.
23. vehicle component fault rates according to claim 14 estimate devices, it is characterised in that also including the
Two training units, are suitable for use with following manner and train the parts wear model:
Training sample is obtained, the training sample includes the degree value of driving behavior data and corresponding parts wear;
The degree value of parts wear is marked into positive sample more than or equal to the training sample of predetermined threshold value, by parts wear
Degree value be labeled as negative sample less than the training sample of the threshold value;
Using logistic regression algorithm, according to the driving behavior data, logic is carried out to the positive sample and the negative sample and is returned
Return training, obtain the parts wear model.
The 24. vehicle component fault rate estimation devices according to claim 22 or 23, it is characterised in that also wrap
Test cell is included, is suitable to be selected at random from the training sample training sample of preset number as test sample, using institute
State test sample checking carries out accuracy validation to the parts wear model.
25. vehicle component fault rates according to claim 24 estimate device, it is characterised in that the test
Unit, is suitable to for the test sample to be randomly divided into N groups test subsample, and N is natural number, and N >=3, adopts wherein any N-1
Group test subsample carries out accuracy validation to one group of test subsample outside the N-1 groups, until N groups test increment
This is verified.
26. vehicle component fault rates according to claim 14 estimate device, it is characterised in that also including more
New unit, is suitable to be obtained when default model modification trigger event is detected the driving behavior data of the vehicle as more
New samples, using the more new samples training is updated to the parts wear model, and renewal is trained into zero for obtaining
Component wear model is used as the parts wear model trained.
27. a kind of vehicle component fault rate estimating systems, it is characterised in that include:Data acquisition unit and right
The vehicle component fault rate estimation device described in 14 to 26 any one is required,
Wherein:
The data acquisition unit is suitable to obtain driving behavior data of the vehicle in default first duration.
28. vehicle component fault rate estimating systems according to claim 27, it is characterised in that the data
Harvester, is suitable to the driving states sensor collection vehicle by installing on the vehicle and is presetting in the first duration
Driving behavior data, or driving behavior data of the vehicle in default first duration are obtained from vehicle storage device.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611264872.4A CN106650157B (en) | 2016-12-30 | 2016-12-30 | Method, device and system for estimating fault occurrence probability of vehicle parts |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611264872.4A CN106650157B (en) | 2016-12-30 | 2016-12-30 | Method, device and system for estimating fault occurrence probability of vehicle parts |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106650157A true CN106650157A (en) | 2017-05-10 |
CN106650157B CN106650157B (en) | 2020-07-28 |
Family
ID=58837994
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611264872.4A Active CN106650157B (en) | 2016-12-30 | 2016-12-30 | Method, device and system for estimating fault occurrence probability of vehicle parts |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106650157B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107506335A (en) * | 2017-07-28 | 2017-12-22 | 中车南京浦镇车辆有限公司 | The method that rail vehicle determines according to confidence level number of faults in the reliability growth phase |
CN110225123A (en) * | 2019-06-13 | 2019-09-10 | 广州小鹏汽车科技有限公司 | A kind of Vehicular data recording method and apparatus |
CN110455560A (en) * | 2019-07-19 | 2019-11-15 | 中车青岛四方机车车辆股份有限公司 | The repair method of vehicle, device and system |
CN110632909A (en) * | 2019-09-29 | 2019-12-31 | 潍柴动力股份有限公司 | Method and equipment for detecting reasonability of back-checking MAP data |
CN111597634A (en) * | 2020-05-12 | 2020-08-28 | 中车青岛四方机车车辆股份有限公司 | Method and device for determining performance parameters of motor train unit |
CN111762177A (en) * | 2020-06-30 | 2020-10-13 | 中国平安财产保险股份有限公司 | Vehicle loss monitoring method and device, computer equipment and storage medium |
CN113281059A (en) * | 2021-05-18 | 2021-08-20 | 潍柴动力股份有限公司 | SCR fault assessment method, SCR device and vehicle |
CN113284357A (en) * | 2020-02-20 | 2021-08-20 | 北汽福田汽车股份有限公司 | Information pushing method, device and system for vehicle |
WO2022196484A1 (en) * | 2021-03-15 | 2022-09-22 | 本田技研工業株式会社 | Order quantity calculating device, and order quantity calculating system |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030009271A1 (en) * | 2001-07-06 | 2003-01-09 | Susumu Akiyama | Vehicular relay device, in-vehicle communication system, failure diagnostic system, vehicle management device, server device and detection and diagnostic program |
US20100209881A1 (en) * | 2009-02-18 | 2010-08-19 | Gm Global Technology Operations, Inc. | Driving skill recognition based on behavioral diagnosis |
CN102120455A (en) * | 2010-01-08 | 2011-07-13 | 福特全球技术公司 | Vehicle maintenance advisor system and vehicle maintenance advisor method |
JP2012220378A (en) * | 2011-04-11 | 2012-11-12 | Mitsubishi Motors Corp | Vehicle behavior simulation device and method |
CN102874188A (en) * | 2012-09-01 | 2013-01-16 | 北京车网互联科技股份有限公司 | Driving behavior warning method based on vehicle bus data |
CN103493019A (en) * | 2011-04-29 | 2014-01-01 | 丰田自动车工程及制造北美公司 | Collaborative multi-agent vehicle fault diagnostic system & associated methodology |
CN104750964A (en) * | 2013-12-31 | 2015-07-01 | 中国移动通信集团公司 | Vehicle state determining method and vehicle state determining device |
CN105930907A (en) * | 2016-01-16 | 2016-09-07 | 平安科技(深圳)有限公司 | Vehicle maintenance analysis prompting method, prompting device and server |
-
2016
- 2016-12-30 CN CN201611264872.4A patent/CN106650157B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030009271A1 (en) * | 2001-07-06 | 2003-01-09 | Susumu Akiyama | Vehicular relay device, in-vehicle communication system, failure diagnostic system, vehicle management device, server device and detection and diagnostic program |
US20100209881A1 (en) * | 2009-02-18 | 2010-08-19 | Gm Global Technology Operations, Inc. | Driving skill recognition based on behavioral diagnosis |
CN102120455A (en) * | 2010-01-08 | 2011-07-13 | 福特全球技术公司 | Vehicle maintenance advisor system and vehicle maintenance advisor method |
JP2012220378A (en) * | 2011-04-11 | 2012-11-12 | Mitsubishi Motors Corp | Vehicle behavior simulation device and method |
CN103493019A (en) * | 2011-04-29 | 2014-01-01 | 丰田自动车工程及制造北美公司 | Collaborative multi-agent vehicle fault diagnostic system & associated methodology |
CN102874188A (en) * | 2012-09-01 | 2013-01-16 | 北京车网互联科技股份有限公司 | Driving behavior warning method based on vehicle bus data |
CN104750964A (en) * | 2013-12-31 | 2015-07-01 | 中国移动通信集团公司 | Vehicle state determining method and vehicle state determining device |
CN105930907A (en) * | 2016-01-16 | 2016-09-07 | 平安科技(深圳)有限公司 | Vehicle maintenance analysis prompting method, prompting device and server |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107506335A (en) * | 2017-07-28 | 2017-12-22 | 中车南京浦镇车辆有限公司 | The method that rail vehicle determines according to confidence level number of faults in the reliability growth phase |
CN110225123A (en) * | 2019-06-13 | 2019-09-10 | 广州小鹏汽车科技有限公司 | A kind of Vehicular data recording method and apparatus |
CN110455560A (en) * | 2019-07-19 | 2019-11-15 | 中车青岛四方机车车辆股份有限公司 | The repair method of vehicle, device and system |
CN110632909A (en) * | 2019-09-29 | 2019-12-31 | 潍柴动力股份有限公司 | Method and equipment for detecting reasonability of back-checking MAP data |
CN113284357A (en) * | 2020-02-20 | 2021-08-20 | 北汽福田汽车股份有限公司 | Information pushing method, device and system for vehicle |
CN111597634A (en) * | 2020-05-12 | 2020-08-28 | 中车青岛四方机车车辆股份有限公司 | Method and device for determining performance parameters of motor train unit |
CN111597634B (en) * | 2020-05-12 | 2023-10-10 | 中车青岛四方机车车辆股份有限公司 | Method and device for determining performance parameters of motor train unit |
CN111762177A (en) * | 2020-06-30 | 2020-10-13 | 中国平安财产保险股份有限公司 | Vehicle loss monitoring method and device, computer equipment and storage medium |
WO2022196484A1 (en) * | 2021-03-15 | 2022-09-22 | 本田技研工業株式会社 | Order quantity calculating device, and order quantity calculating system |
JP7492648B2 (en) | 2021-03-15 | 2024-05-29 | 本田技研工業株式会社 | Order quantity calculation device and order quantity calculation system |
CN113281059A (en) * | 2021-05-18 | 2021-08-20 | 潍柴动力股份有限公司 | SCR fault assessment method, SCR device and vehicle |
CN113281059B (en) * | 2021-05-18 | 2022-09-23 | 潍柴动力股份有限公司 | SCR fault assessment method, SCR device and vehicle |
Also Published As
Publication number | Publication date |
---|---|
CN106650157B (en) | 2020-07-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106650157A (en) | Method, device and system for vehicle part fault probability estimation | |
CN110376003B (en) | Intelligent train service life prediction method and system based on BIM | |
CN108053075B (en) | Scrapped vehicle prediction method and system | |
CN108475358B (en) | Method and system for evaluating driver's trip performance | |
CN102717765A (en) | Fatigue driving detection method and anti-fatigue driving auxiliary device | |
US20130184928A1 (en) | Driver behavior diagnostic method and system | |
CN103871122A (en) | Driving behavior analysis method and driving behavior analysis system | |
CN104092736A (en) | Vehicle networking device, server and system, scoring method and data collection method | |
CN108369682B (en) | Method and system for evaluating driver's trip performance | |
CN104809878A (en) | Method for detecting abnormal condition of urban road traffic by utilizing GPS (Global Positioning System) data of public buses | |
CN112287602B (en) | Motor train axle temperature fault early warning method based on machine learning and isolated forest | |
CN108431837B (en) | Method and system for evaluating driver's trip performance | |
KR102141988B1 (en) | The realtime bus safety and reliability integrated indicator monitoring method for autonomous public bus and the system of the same | |
CN111762096A (en) | New energy automobile safety early warning method and system based on artificial intelligence | |
CN108475359B (en) | Method and system for evaluating driver's trip performance | |
EP3960576A1 (en) | Method and system for analysing the control of a vehicle | |
CN102981497A (en) | Fault diagnosis device and diagnosis method for high-speed train braking system | |
CN111144485A (en) | Vehicle accident judgment method and system based on xgboost classification algorithm | |
CN107918826A (en) | The driver's evaluation and dispatching method that a kind of driving environment perceives | |
CN108369681B (en) | Method and system for evaluating driver's trip performance | |
CN115586023A (en) | Fault diagnosis method and system for rail vehicle transmission system | |
CN109784586B (en) | Prediction method and system for danger emergence condition of vehicle danger | |
CN110060370B (en) | Equivalent statistical method for times of rapid acceleration and rapid deceleration of vehicle | |
CN108268678A (en) | Driving behavior analysis method, apparatus and system | |
KR102204199B1 (en) | Prediction method and diagonoisis method of vehicle for autonomous public bus |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |