CN105160883B - A kind of energy-saving driving behavior analysis method based on big data - Google Patents
A kind of energy-saving driving behavior analysis method based on big data Download PDFInfo
- Publication number
- CN105160883B CN105160883B CN201510683076.3A CN201510683076A CN105160883B CN 105160883 B CN105160883 B CN 105160883B CN 201510683076 A CN201510683076 A CN 201510683076A CN 105160883 B CN105160883 B CN 105160883B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- user
- oil consumption
- energy
- data
- 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.)
- Active
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of energy-saving driving behavior analysis method based on big data, belong to intelligent transportation and automobile energy-saving technology field.This method collects vehicle-related information, environmental information and transport information, and the information collected is sent to car networking information service backstage through car-mounted terminal in vehicle traveling process by car-mounted terminal and road side system；Information service backstage carries out energy-saving driving analysis by same ID vehicle datas vertical analysis model, same types of vehicles data horizontal analysis model and with feature vehicle data comparative analysis model, and most energy-saving driving suggestion feeds back to user by way of car-mounted terminal voice broadcast and screen display in real time or at the end of a trip at last.This method can be that driver provides economic, reliable drive advice and looks forward to the design to vehicle for car and improve to provide theories integration in the process of moving, so as to reach the effect of energy-saving and emission-reduction.
Description
Technical field
The invention belongs to intelligent transportation and automobile energy-saving technology field, it is related to a kind of energy-saving driving behavior based on big data
Analysis method.
Background technology
At present, drive and study for vehicle energy saving both domestic and external, entered mainly or with some parameters for single unit vehicle
Based on row monitoring, such as vehicle oil consumption, technology status and driving behavior analysis, these analysis methods to a certain extent can
The purpose of energy-conservation is realized, but with the rise and development of car networking, increasing vehicle has been able to conveniently access
Network, further, since the development of field of traffic, can reflect that the road side system of traffic, navigation system have also been covered in real time
Comprehensively, it means that the analysis of energy-saving driving behavior is not limited solely to single unit vehicle, and can be based on whole big data and put down
Platform.
The content of the invention
In view of this, should it is an object of the invention to provide a kind of energy-saving driving behavior analysis method based on big data
Method can be that driver provides economic, reliable drive advice and is car enterprise to the design of vehicle and changes in the process of moving
Enter to provide theories integration, so as to reach the effect of energy-saving and emission-reduction.
To reach above-mentioned purpose, the present invention provides following technical scheme：
A kind of energy-saving driving behavior analysis method based on big data, this method is in vehicle traveling process, by vehicle-mounted
Terminal, satellite and road side system collect vehicle-related information, environmental information and transport information, and by the information collected through vehicle-mounted
Terminal is sent to car networking information service backstage；Information service backstage passes through same ID vehicle datas vertical analysis model, same type
Vehicle data horizontal analysis model and with feature vehicle data comparative analysis model carry out energy-saving driving analysis, most at last energy-conservation drive
Sail suggestion and feed back to user by way of car-mounted terminal voice broadcast and screen display in real time or at the end of a trip.
Further, the same ID vehicle datas vertical analysis model differentiates and recorded that the urgency of driver adds by car-mounted terminal
Speed, anxious deceleration, zig zag behavior, are analyzed to same ID vehicles real time data and historical data, driver are corrected in time
Driving bad habit and do not save behavior；
Concrete analysis process includes：
1) driving behavior to vehicle operator is differentiated and quantified, first the not power-save operation behavior to driver
It is divided into five classes：It is anxious to accelerate, it is anxious to slow down, take a sudden turn, anxious acceleration+zig zag and anxious deceleration+zig zag；Five class behaviors are carried out again
Classification and quantization；
2) Real-time Feedback is carried out to data：Driver's driving behavior is weighted after quantization, every kind of behavior is obtained different
The weights fraction of rank, during driver travels, when a certain behavior of driver reaches respective degrees, car-mounted terminal meeting
Sent to driver and warn and show the moment oil consumption；
3) number of times of car-mounted terminal statistics above-mentioned behavior of user in certain effective driving time, obtains corresponding section
Energy behavior point, and be compared with historical scores, understand whether oneself this period of above-mentioned behavior is reduced, for the energy-conservation of user
The calculating of behavior point, the weights for using each class behavior add up.
Further, it is described that five class behaviors are classified and quantify to include：Three class basic acts are classified and attached
Weighted value, the generation for intersecting behavior, the method for using weights multiplication：
When acceleration is in a m/s2To b m/s2In the range of belong to slight anxious and accelerate, grade is L, and weights are 1；In b m/s2
To c m/s2In the range of belong to moderate suddenly acceleration, grade is M, and weights are 2；When acceleration is more than c m/s2When belong to severe suddenly plus
Speed, grade is F, and weights are 3, wherein (a<b<c)；Similarly anxious deceleration and zig zag also refer to related data and carry out weights point
Match somebody with somebody；
Further, the same types of vehicles data horizontal analysis model is by collecting the average oil of vehicle over a period
Consumption, instantaneous fuel consumption information, each type of vehicle Actual Burnoff amount and theoretical fuel consumption are analyzed using the method for difference test
Oil consumption difference situation between difference and the same types of vehicles of comparative analysis different vendor production.
The analysis vehicle Actual Burnoff amount includes with theoretical fuel consumption：Pass through the oil consumption number of collection vehicle multiple periods
According to trying to achieve the average fuel consumption of vehicle as the practical oil consumption of vehicle, and otherness point is carried out with the theoretical oil consumption of the type vehicle
Analysis, makes car enterprise understand the situation that the vehicle vehicle significantly exceeds theoretical oil consumption after putting on market with the presence or absence of practical oil consumption,
Look forward to providing theoretical and data supporting on vehicle oil consumption Theoretical Design afterwards for car, specifically include following steps：
1) oil consumption of the multiple periods measured during collection vehicle traveling, obtains one group of data and is designated as (x1, x2,
x3......xn), then the practical oil consumption of the vehicle is designated as
2) difference test：
1. because the average fuel consumption data of same types of vehicles are N (μ, the σ of Normal Distribution in theory2), wherein μ is
Desired value, is that car enterprise calculates obtained theoretical fuel consumption values, σ based on big data2It is designated as variance；
2. it is null hypothesis H0：μ=μ0
3. selection check statistic：
If 4. For significance, illustrate the vehicle practical oil consumption level and theoretical oil consumption
It is variant, and differ greatly；
3) by difference test, car enterprise can receive two useful informations：Practical oil consumption between same type of vehicle
Whether there is larger difference；Whether the vehicle that practical oil consumption exceeds theoretical oil consumption is more；
Further, the oil consumption difference situation between the same types of vehicles of the comparative analysis different vendor production is specifically wrapped
Include：
1) data acquisition and classification：The data of collection are made to classify by the distance travelled of vehicle；
2) otherness of vehicle oil consumption under different distance travelleds of different manufacturers production is compared；
Further, the same feature vehicle data comparative analysis model is by having the user of same characteristic features to enter to some
Oil consumption of any member all with optimum drive person in row classification, each class is compared, and being scored in corresponding Power Saving Class is
In system, user's scoring ranking is given, it is final that energy-saving driving suggestion is provided.
Further, the same feature vehicle data comparative analysis is specifically included：
1) user's classification of same characteristic features：The classification criterion of the user of same characteristic features includes average speed per hour, period, weather
With traffic events, route distance, speed per hour 60km/h driving times ratio, parking rate, the time scale of parking, vehicle feature, adopt
With PCA algorithm dimensionality reductions, then classified with k-means algorithms；
2) divided and score rank analysis with feature user Power Saving Class：After user's classification is completed, to same category
The energy-saving driving behavior of user is analyzed, with fuzzy control theory, by the acceleration in vehicle driving process and
Power Saving Class is estimated in the monitoring of acceleration change amount；Energy-saving driving behavior in user's each run is given a mark, and will
The ranking of user feeds back to car-mounted terminal；
3) optimum oil consumption compares advises feeding back with energy-saving driving：The optimum oil consumption is relatively by a trip of user
User's optimum oil consumption of oil consumption and same condition down train is compared, and the determination of calculated fuel consumption amount is based in this stroke
On the basis of the oil consumption of all users, the user of removal preceding 10% and rear 10% is used, takes the user's of centre 80%
Oil consumption average value；The energy-saving driving suggestion is one drawn according to Correlative data analysis such as points-scoring system and vehicle accelerations
Plant the strategy that can effectively improve user's energy-saving driving behavior.
Further, the strategy that can effectively improve user's energy-saving driving behavior includes：Be first by with identical conditions
The oil consumption of the optimum drive person of down train is compared, and show whether user's oil consumption is normal, if oil consumption is normal, informs that user protects
Hold and give certain virtual reward, if user's oil consumption is abnormal, further carry out average acceleration and acceleration change amount point
Analysis；Rational for average acceleration, user needs to adjust the dynamics stepped on the gas in driving procedure, to control the change of acceleration
Change amount；For average acceleration is unreasonable, acceleration change amount rational, user needs to reduce throttle in the process of moving；It is right
All irrational in average acceleration and acceleration change amount, then user should reduce throttle and reduce the dynamics stepped on the gas again.
The beneficial effects of the present invention are：Method of the present invention can analyze same vehicle over a period
The differentiation of anxious (anxious acceleration, the anxious deceleration, zig zag) behavior of energy-saving driving behavior, such as driver three, can correct driver's in time
Drive bad habit and do not save behavior；Analyze the information such as the average fuel consumption of same type vehicle over a period, be car enterprise with
Theoretical and data supporting is provided on vehicle oil consumption Theoretical Design afterwards；And the user of same condition down train is compared, and
Being graded with reference to the data compared for the energy-saving driving behavior of driver, there is provided energy-saving driving suggestion.
Brief description of the drawings
In order that the purpose of the present invention, technical scheme and beneficial effect are clearer, the present invention provides drawings described below and carried out
Explanation：
Fig. 1 is the analysis system structure chart of the energy-saving driving behavior based on big data；
Fig. 2 is with ID vehicle data vertical analysis flow charts；
Fig. 3 is same types of vehicles data horizontal analysis flow chart；
Fig. 4 is that energy-saving driving advises feedback mechanism flow chart；
Fig. 5 is with feature vehicle data comparative analysis flow chart.
Embodiment
Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are described in detail.
It is an object of the invention to provide a kind of analysis method of the energy-saving driving behavior based on big data, it can analyze same
The energy-saving driving behavior of individual vehicle over a period, such as driver three suddenly (suddenly accelerate, suddenly slow down, take a sudden turn) sentencing for behavior
Not, the driving bad habit of driver can be corrected in time and do not save behavior；Analysis same type vehicle being averaged over a period
The information such as oil consumption, are to provide theoretical and data supporting on the vehicle oil consumption Theoretical Design of car enterprise afterwards；And to under the conditions of
The user of driving is compared, and combines the data that compare and graded for the energy-saving driving behavior of driver and driven there is provided energy-conservation
Sail suggestion.
Fig. 1 is the analysis system structure chart of the energy-saving driving behavior based on big data, in vehicle traveling process, passes through car
Mounted terminal, satellite and road side system collect vehicle-related information, environmental information and transport information etc., are sent through car-mounted terminal to car
Networked information service background, backstage carries out energy-saving driving analysis by correspondence model, such as ID vehicle datas vertical analysis, similar
Type vehicle data horizontal analysis and with feature vehicle data comparative analysis etc., most energy-saving driving suggestion is in real time or one section at last
At the end of stroke user is fed back to by way of car-mounted terminal voice broadcast and screen display.
Based on this, the present invention sets up three kinds of analysis models：With ID vehicle data vertical analysis model, same types of vehicles data
Horizontal analysis model and with feature vehicle data comparative analysis model.
With ID vehicle data vertical analysis models：
The primary operational object of vehicle driver in the process of moving has throttle, brake, clutch and gear, and some are driven
The person of sailing is bad due to experience shortage or driving habit, takes and improperly operates in inappropriate road conditions, this not only can be to vehicle
Loss itself is caused, vehicle fuel consumption increase is also resulted in.Based on this, the present invention differentiates and recorded driver by car-mounted terminal
Three walk rapidly for (it is anxious accelerate, it is anxious slow down, zig zag), same ID vehicle datas are analyzed, in time the driving of correction driver
Bad habit and do not save behavior.Fig. 2 is with ID vehicle data vertical analysis flow charts.
1) to the differentiation and quantization of driving behavior：
Driver is as the principal factors in people-Che-road, due to the non-linear and time variation of itself, in driving procedure
Behavioral trait be often that perception is changeable, it is difficult to accurate quantificational description.Based on this, it is comprehensive that the present invention uses dynamic weighting
Quantitative model is closed, the not power-save operation behavior to driver first is classified, and is broadly divided into three classes：It is anxious to accelerate, it is anxious slow down with
And zig zag, this three class is driver's most common behavior in the process of moving, and is also cause vehicle oil consumption big main
Factor.In driver's actual mechanical process, this several class behavior is sometimes with generation.The present invention first to driver this three
Class basic act has made classification and additional weights.It is as follows：
Table 1：The basic driving behavior grade quantizing table of driver
When acceleration is in a m/s2To b m/s2In the range of belong to slight anxious and accelerate, grade is L, and weights are 1；In b m/s2
To c m/s2In the range of belong to moderate suddenly acceleration, grade is M, and weights are 2；When acceleration is more than c m/s2When belong to severe suddenly plus
Speed, grade is F, and weights are 3；Similarly anxious deceleration and zig zag also refer to related data and carry out weights distribution.
Generation for intersecting behavior, the method that the present invention uses weights multiplication, crossing lines are mainly to have two classes：It is anxious
Acceleration+zig zag, anxious deceleration+zig zag.
Table 2：Driver intersects driving behavior grade quantizing table
2) Real-time Feedback：
Driver's driving behavior is weighted after quantization, the weights fraction of every kind of behavior different stage has been obtained, has driven
During the person's of sailing traveling, when a certain behavior of driver reaches respective degrees, car-mounted terminal can send warning simultaneously to driver
The moment oil consumption is shown, such as vehicle accelerates suddenly when turning, and acceleration and turning speed exceed maximum safety limit
When spending, belong to the anxious acceleration of severe+severe zig zag, car-mounted terminal can be automatically too fast simultaneously with the acceleration of voice informing user turn inside diameter
The oil consumption of this moment is shown.
3) history compares：
Understanding of the user to oneself a period of time energy-saving behavior for convenience, car-mounted terminal can be counted certain effective
The number of times of these behaviors of user in driving time, obtains corresponding energy-conservation behavior point, and is compared with historical scores, understands
Whether oneself this period " three is anxious " behavior is reduced, and for the calculating of the energy-conservation behavior point of user, the present invention uses all kinds of
The weights of behavior add up.In addition reflection the most direct data of user's energy-saving efficiency be exactly fuel consumption number, by recording user
Fuel consumption in certain effectively driving time, and compared with historical data, understand whether oneself oil consumption situation makes moderate progress.
Same types of vehicles data horizontal analysis model：
Same types of vehicles data horizontal analysis is by collecting average fuel consumption, the instantaneous oil consumption of vehicle over a period etc.
Information, difference and contrast of each type of vehicle Actual Burnoff amount with theoretical fuel consumption are analyzed using the method for difference test
Oil consumption difference situation between the same types of vehicles of different vendor's production, Fig. 3 is same types of vehicles data horizontal analysis flow chart.
1) gross data is looked forward to car to be compared：
Car enterprise passes through the fuel consumption data of collection vehicle multiple periods, and the average fuel consumption for trying to achieve vehicle is used as the reality of vehicle
Oil consumption, and difference analysis is carried out with the theoretical oil consumption of the type vehicle, car enterprise is understood the vehicle vehicle after putting on market
It is to be provided on the vehicle oil consumption Theoretical Design of car enterprise afterwards with the presence or absence of practical oil consumption significantly beyond the situation of theoretical oil consumption
Theoretical and data supporting.In addition for the abnormal user of some fuel consumption, car enterprise is notified that the user takes adequate measures.Tool
Body is analyzed as follows：
(1) oil consumption of the multiple periods measured during collection vehicle traveling, obtains one group of data and is designated as (x1, x2,
x3......xn), then the practical oil consumption of the vehicle is designated as
(2) difference test：
1. because the average fuel consumption data of same types of vehicles are N (μ, the σ of Normal Distribution in theory2), wherein μ is
Desired value, is that car enterprise calculates obtained theoretical fuel consumption values, σ based on big data2It is designated as variance.
2. it is null hypothesis H0：μ=μ0
3. selection check statistic：
If 4. For significance.Illustrate that the vehicle practical oil consumption level has with theoretical oil consumption
Difference, and differ greatly.
(3) by difference test, car enterprise can receive two useful informations：
1. whether the practical oil consumption between same type of vehicle has larger difference；
2. whether practical oil consumption is more beyond the vehicle of theoretical oil consumption.
This can help car enterprise is further to this type of vehicle to be improved, and reduce this otherness.In addition, for side
Just the improvement of user itself, this species diversity can feed back to car-mounted terminal in the form of ranking, and end user is according to actual conditions, choosing
Select and whether overhaul.
2) vehicle with other producers is compared：
With the rise of car networking, each producer to vehicle driving data it is shared be it is inevitable, for different manufacturers produce
Price, weight, power and other each factor identical vehicles, the size of fuel consumption will be passes that car enterprise gets the upper hand of in the market
Key.Based on this, Main Analysis of the present invention be same type different manufacturers production vehicle average fuel consumption, instantaneous oil consumption, tool
Body is analyzed as follows：
(1) data acquisition and classification：Because the difference of vehicle usage degree can cause the oil consumption situation of vehicle difference,
Therefore need that the data of collection are made to classify by the distance travelled of vehicle, such as the vehicle of 10000-20000 kilometers of traveling makes for one
Degree rank is used, average fuel consumption level of the same type vehicle in different usage degrees of each manufacturer production is obtained.
(2) analysis model：
The same vehicle of different manufacturers relatively mainly compares the same types of vehicles of different manufacturers production in different usage degrees
The otherness of lower oil consumption level.Illustrated by taking producer one and producer two as an example, following table is the producer one and producer two collected
The first type of vehicle of production average fuel consumption value under different usage degrees.
Table 3：The average fuel consumption of the vehicle of the different usage degrees of each manufacturer production
Wherein X1nRepresent the fuel consumption per hundred kilometers average water for all vehicles that a certain type usage degree that producer one produces is n
It is flat.Also analysis result can more intuitively be fed back to car enterprise by drawing line chart, facilitate car enterprise to understand the vehicle of oneself not
It is that car is looked forward in market orientation from now on and marketing strategy with the gap under usage degree between the vehicle of other manufacturer production
Data explanation is provided in terms of formulation.
With feature vehicle data comparative analysis model：
When driver drives a vehicle under the same conditions, because the running environment residing for vehicle is similar, its oil consumption situation is also managed
It should be more or less the same, if some drivers compare, other driver drives vehicle fuel consumption are excessive, and this is most likely due to operation not
With feature vehicle data comparative analysis it is by being divided the user of same characteristic features some information based on this when caused
Oil consumption of any member all with optimum drive person in class, each class is compared, in corresponding Power Saving Class points-scoring system,
User's scoring ranking is given, final to provide energy-saving driving suggestion, Fig. 4 is energy-saving driving suggestion feedback mechanism flow chart；Fig. 5 is with special
Levy vehicle data comparative analysis flow chart.
1) user's classification of same characteristic features：
The classification criterion of same characteristic features user include average speed per hour, period, weather and traffic events, route distance, when
Fast 60km/h driving times ratio, parking rate, time scale, the vehicle feature of parking etc..In view of respective between these indexs
Between there is correlation and dimension is too high, it is difficult to classified using traditional clustering method, therefore herein first using PCA principal components
Analytic approach carries out dimensionality reduction, and then user is classified by three-dimensional k-means clustering procedures.
(1) PCA dimensionality reductions：
Principal component analysis (PCA) is exactly that multiple variables are turned to a few overall target.Assuming that having n sample, each
Sample has m variable description, and the data of each variable are made into normalized, thus constitutes n × m rank data matrixes：
Step 1 calculates correlation matrix：
In formula, rij(i, j=1,2,3...m) represents primal variable xiWith xjCoefficient correlation, its calculation formula is：
Only need to calculate triangle element, r thereonijAnd rjiIt is equal.
Step 2 calculates characteristic value and characteristic vector
Solve characteristic equation | R- λ E |=0, obtain eigenvalue λi(i=1,2...m), and make its order arrangement, i.e. λ by size1
≥λ2≥...λm≥0；Then obtain respectively corresponding to λiCharacteristic vector qi(i=1,2...m).
It is contribution rate of accumulative total that step 3, which calculates principal component contributor rate,
Principal component ZiContribution rate：Contribution rate of accumulative total：Take
Three maximum eigenvalue λs of contribution rate1, λ2, λ3Corresponding principal component.
Step 4 calculates principal component load
Thus it can further calculate and obtain principal component scores：
(2) the K-means clusters based on three dimensions：
K-means algorithms are a kind of indirect clustering methods based on similarity measurement between sample, belong to unsupervised learning side
Method.N object is divided into k cluster by this algorithm using k as parameter, so as to there is higher similarity in cluster, and also it is similar between cluster
Degree is relatively low.The calculating of similarity is carried out according to the average value (center of gravity for being counted as cluster) of object in a cluster.This algorithm is first
K object is randomly choosed, each object represents the barycenter of a cluster.For remaining each object, according to the object with
The distance between each cluster barycenter, is assigned to it in cluster most like therewith.Then, the new barycenter each clustered is calculated.
Said process is repeated, until criterion function convergence.
After PCA dimensionality reductions, judge that the data of user characteristics are reduced to three-dimensional by multidimensional, so each sample point is in space
In there is only one coordinate to correspond to therewith, specific step is as follows：
1st, after a trip terminates, each index of the user collected obtains three principal components after being calculated by PCA, this
The feature of n sample can be described by these three principal component indexs, as follows：
It is z that k sample, which is randomly selected, as initial clustering center of mass point (cluster centroids)1, z2...zk∈R3。
2nd, the classification for user characteristics is built upon in this k initial clustering center of mass point, for each sample point, meter
Its Euclidean distance with initial sample point is calculated, the class that it should belong to is judged：
For each class j, such barycenter is recalculated：
K is the cluster numbers that we give in advance, c(i)Representative sample i and that class closest in k class, c(i)Value
It is 1 one into k.Barycenter zjUs are represented to belonging to the conjecture of center of a sample's point of same class.
The driving characteristics of user are divided into by k classes by the algorithm so that during similarity Datong District between each class user
The difference between all types of user has been distinguished again.
2) divided and score rank analysis with feature user Power Saving Class：
Power Saving Class points-scoring system is that after user's classification is completed, the energy-saving driving behavior to same category user is carried out
Analysis, the present invention uses fuzzy control theory, passes through the prison to the acceleration in vehicle driving process and acceleration change amount
Survey to estimate Power Saving Class.Analysis model is as follows：
The acceleration a and acceleration change amount Δ a of vehicle are that automobile energy-saving whether Important Parameters, wherein a mould judged
Paste collection is combined into S (Small), M (Medium) and L (large)；Δ a fuzzy set is N (Negative), M (Medium) and P
(positive), grade classification has VW (Very Wasting), W (Wasting) M (Medium), E (Economic) and VE
(Very Economic).Power Saving Class fuzzy matrix is as follows：
Table 4：Power Saving Class fuzzy matrix
Wherein acceleration a and acceleration change amount Δ a all be pick up the car an a trip terminate after average value, and with it is same
Under the conditions of other vehicles be compared divided rank, can use k-means algorithms, do not repeat herein.
In order to more intuitively reflect the degree of power conservation of its current driving behavior to user, the present invention is used in grade
Given a mark on the basis of division, scoring criterion is：VW=2 points, W=4 points, M=6 points, E=8 points, VE=10 points.Wherein most
High score is 10 points.
Table 5：Power Saving Class score value divides table
| Grade | VW | W | M | E | VE |
| Score | 2 | 4 | 6 | 8 | 10 |
By being given a mark to the energy-saving driving behavior in user's each run, and the ranking of user is fed back into vehicle-mounted end
End, can be very good to encourage user, allow the interesting participation of more users to come in.
3) optimum oil consumption compares advises feeding back with energy-saving driving：
Optimum oil consumption compares：
Optimum oil consumption is relatively by the oil consumption of a trip of user and user's optimum oil consumption with condition down train
It is compared, the determination of calculated fuel consumption amount is based on the basis of the oil consumption of all users in this stroke, it is contemplated that may
In the presence of some special statuss, the present invention uses the user of removal preceding 10% and rear 10%, takes the oil of the user of centre 80%
Consume average value.Because the running environment residing for vehicle is similar, its oil consumption situation ought to also be more or less the same, if some drivers compare
Other driver drives vehicle fuel consumption are excessive, and this is most likely due to caused by misoperation, by with optimum oil consumption phase
Than having promoted timely concern of the user to oneself energy-saving driving behavior.
Energy-saving driving suggestion feedback：
Energy-saving driving suggestion is one drawn according to Correlative data analysis such as the driving behaviors of points-scoring system and driver
Plant the strategy that can effectively improve user's energy-saving driving behavior.
The mechanism is compared by the oil consumption with the optimum drive person of same driving characteristics down train first, draws use
Whether family oil consumption is normal, if oil consumption is normal, informs that user keeps and gives certain virtual reward, if user's oil consumption is abnormal,
Then further driver's driving behavior is analyzed；By the way that the driving behavior of driver in a trip is driven with standard energy-conservation
The behavior of sailing is contrasted, and which behavior analysis driver there may be not energy factor, is then proposed according to these behaviors corresponding
Proposed projects, user can voluntarily select receive suggestion and do not receive suggestion.
Finally illustrate, preferred embodiment above is merely illustrative of the technical solution of the present invention and unrestricted, although logical
Cross above preferred embodiment the present invention is described in detail, it is to be understood by those skilled in the art that can be
Various changes are made to it in form and in details, without departing from claims of the present invention limited range.
Claims (3)
1. a kind of energy-saving driving behavior analysis method based on big data, it is characterised in that：This method in vehicle traveling process,
Vehicle-related information, environmental information and transport information are collected by car-mounted terminal and road side system, and the information collected is passed through
Car-mounted terminal is sent to car networking information service backstage；Information service backstage passes through same ID vehicle datas vertical analysis model, same
Type of vehicle data horizontal analysis model and with feature vehicle data comparative analysis model carry out energy-saving driving analysis, most save at last
Energy drive advice is fed back to by way of car-mounted terminal voice broadcast and screen display in real time or at the end of a trip
User；
The same ID vehicle datas vertical analysis model differentiated by car-mounted terminal and recorded driver it is anxious accelerate, it is anxious slow down,
Zig zag behavior, same ID vehicles real time data and historical data are analyzed, and the driving bad habit of driver are corrected in time
Do not save behavior；Concrete analysis process includes：
1) driving behavior to vehicle operator is differentiated and quantified, and the not power-save operation behavior to driver first is divided into
Five classes：It is anxious to accelerate, it is anxious to slow down, take a sudden turn, anxious acceleration+zig zag and anxious deceleration+zig zag；Five class behaviors are classified again
And quantization；
2) Real-time Feedback is carried out to data：Driver's driving behavior is weighted after quantization, every kind of behavior different stage is obtained
Weights fraction, during driver travels, when a certain behavior of driver reaches respective degrees, car-mounted terminal can be to driving
The person of sailing, which sends, warns and shows the moment oil consumption；
3) number of times of car-mounted terminal statistics above-mentioned behavior of user in certain effective driving time, obtains corresponding energy-conservation row
To divide, and it is compared with historical scores, understands whether oneself this period of above-mentioned behavior is reduced, for the energy-conservation behavior of user
The calculating divided, the weights for using each class behavior add up；
The same feature vehicle data comparative analysis model be by having the user of same characteristic features to classify to some, it is each
Oil consumption of any member all with optimum drive person in class is compared, in corresponding Power Saving Class points-scoring system, to user
Score ranking, final to provide energy-saving driving suggestion；
The same feature vehicle data comparative analysis is specifically included：
1) user's classification of same characteristic features：The classification criterion of the user of same characteristic features includes average speed per hour, period, weather and friendship
Interpreter's part, route distance, speed per hour 60km/h driving times ratio, parking rate, the time scale of parking, vehicle feature, using PCA
Algorithm dimensionality reduction, is then handled with k-means algorithms；
2) divided and score rank analysis with feature user Power Saving Class：After user's classification is completed, to same category user
Energy-saving driving behavior analyzed, with fuzzy control theory, by the acceleration in vehicle travel process and acceleration
The monitoring of variable quantity is spent to estimate Power Saving Class；Energy-saving driving behavior in user's each run is given a mark, and by user
Ranking feed back to car-mounted terminal；
3) optimum oil consumption compares advises feeding back with energy-saving driving：The optimum oil consumption is relatively by the oil consumption of a trip of user
The user's optimum oil consumption measured with same condition down train is compared, and the determination of calculated fuel consumption amount is based on all in this stroke
On the basis of the oil consumption of user, the user of removal preceding 10% and rear 10% is used, the oil consumption of the user of centre 80% is taken
Average value；The energy-saving driving suggestion is a kind of energy drawn according to Correlative data analysis such as points-scoring system and vehicle accelerations
It is effectively improved the strategy of user's energy-saving driving behavior；
The same types of vehicles data horizontal analysis model is by collecting vehicle average fuel consumption over a period, instantaneous oil consumption
Information, difference and contrast of each type of vehicle Actual Burnoff amount with theoretical fuel consumption are analyzed using the method for difference test
Analyze the oil consumption difference situation between the same types of vehicles of different vendor's production；
The analysis vehicle Actual Burnoff amount includes with theoretical fuel consumption：By the fuel consumption data of collection vehicle multiple periods,
The average fuel consumption of vehicle is tried to achieve as the practical oil consumption of vehicle, and difference analysis is carried out with the theoretical oil consumption of the type vehicle,
Car enterprise is understood the situation that the vehicle vehicle significantly exceeds theoretical oil consumption after putting on market with the presence or absence of practical oil consumption, be car
Theoretical and data supporting is provided on the vehicle oil consumption Theoretical Design of enterprise afterwards, following steps are specifically included：
1) oil consumption of multiple periods measured during collection vehicle traveling, obtain one group of data be designated as (x1, x2,
X3......xn), then the practical oil consumption of the vehicle is designated as
2) difference test：
1. because the average fuel consumption data of same types of vehicles are N (μ, the σ of Normal Distribution in theory2), wherein μ is expectation
Value, is that car enterprise calculates obtained theoretical fuel consumption values, σ based on big data2It is designated as variance；
2. it is null hypothesis H0：μ=μ0
3. selection check statistic：
If 4. significance, illustrates that the vehicle practical oil consumption level and theoretical oil consumption are variant, and differ greatly；
3) by difference test, car enterprise can receive two useful informations：Whether the practical oil consumption between same type of vehicle
With larger difference；Whether the vehicle that practical oil consumption exceeds theoretical oil consumption is more；
Oil consumption difference situation between the same types of vehicles of the comparative analysis different vendor production is specifically included：
1) data acquisition and classification：The data of collection are made to classify by the distance travelled of vehicle；2) different manufacturers production is compared
The otherness of vehicle oil consumption under different distance travelleds.
2. the energy-saving driving behavior analysis method according to claim 1 based on big data, it is characterised in that：Described pair
Five class behaviors are classified and quantify to include：Three class basic acts are classified and additional weights, the hair for intersecting behavior
Method that is raw, using weights multiplication：
When acceleration is in a m/s2To b m/s2In the range of belong to slight anxious and accelerate, grade is L, and weights are 1；In b m/s2To c
m/s2In the range of belong to moderate suddenly acceleration, grade is M, and weights are 2；When acceleration is more than c m/s2When belong to severe suddenly acceleration, etc.
Level is F, and weights are 3, wherein (a<b<c)；Similarly anxious deceleration and zig zag also refer to related data and carry out weights distribution.
3. the energy-saving driving behavior analysis method according to claim 1 based on big data, it is characterised in that：It is described to have
The strategy that effect improves user's energy-saving driving behavior includes：It is by the oil consumption with the optimum drive person of identical conditions down train first
It is compared, show whether user's oil consumption is normal, if oil consumption is normal, informs that user keeps and gives certain virtual reward, if
User's oil consumption is abnormal, then further carries out average acceleration and the analysis of acceleration change amount；It is rational for average acceleration,
User needs to adjust the dynamics stepped on the gas in driving procedure, to control the variable quantity of acceleration；Do not conform to for average acceleration
Reason, acceleration change amount are rational, and user needs to reduce throttle in the process of moving；For average acceleration and acceleration change
Amount is all irrational, then user should reduce throttle and reduce the dynamics stepped on the gas again.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510683076.3A CN105160883B (en) | 2015-10-20 | 2015-10-20 | A kind of energy-saving driving behavior analysis method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510683076.3A CN105160883B (en) | 2015-10-20 | 2015-10-20 | A kind of energy-saving driving behavior analysis method based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105160883A CN105160883A (en) | 2015-12-16 |
| CN105160883B true CN105160883B (en) | 2017-08-25 |
Family
ID=54801721
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510683076.3A Active CN105160883B (en) | 2015-10-20 | 2015-10-20 | A kind of energy-saving driving behavior analysis method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105160883B (en) |
Families Citing this family (40)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105865472B (en) * | 2016-04-06 | 2019-05-10 | 重庆邮电大学 | A kind of navigation method based on optimum oil consumption |
| CN105844913A (en) * | 2016-04-15 | 2016-08-10 | 苏州爱诺信信息科技有限公司 | Correlation analyzing method based on vehicle, road conditions and safe travel big data in network of vehicles |
| CN106096507B (en) * | 2016-05-27 | 2020-03-24 | 浩鲸云计算科技股份有限公司 | Intelligent traffic black car identification method |
| CN107784587B (en) * | 2016-08-25 | 2021-09-14 | 大连楼兰科技股份有限公司 | Driving behavior evaluation system |
| CN106227135A (en) * | 2016-09-21 | 2016-12-14 | 北京机械设备研究所 | There is the New energy automobile motor control device and method of teledata monitoring function |
| CN106740863A (en) * | 2016-11-30 | 2017-05-31 | 武汉长江通信智联技术有限公司 | Driving behavior analysis method |
| TWI591492B (en) * | 2016-12-19 | 2017-07-11 | Chunghwa Telecom Co Ltd | Energy consumption estimation system and method |
| CN108266278A (en) * | 2016-12-30 | 2018-07-10 | 中国移动通信集团公司 | Method, system, T platforms and the OBD terminals that a kind of automobile engine power is adjusted |
| CN108460057B (en) * | 2017-02-22 | 2023-09-29 | 深圳市赛格车圣科技有限公司 | User travel mining method and device based on unsupervised learning |
| CN107331179A (en) * | 2017-05-27 | 2017-11-07 | 东风商用车有限公司 | Economic driving auxiliary system based on big data cloud platform and implementation method |
| CN107180288B (en) * | 2017-07-21 | 2020-05-01 | 东软集团股份有限公司 | Driving behavior energy consumption measuring and calculating method and device, storage medium and electronic equipment |
| CN107563931A (en) * | 2017-08-16 | 2018-01-09 | 上海经达信息科技股份有限公司 | A kind of real-time driving behavior quality appraisal procedure of vehicle based on the Big Dipper or gps data |
| CN107585164B (en) * | 2017-09-04 | 2019-11-22 | 交通运输部公路科学研究所 | A kind of method and device for the driver that classifies |
| CN109598396B (en) * | 2017-10-18 | 2023-03-24 | 宁波轩悦行电动汽车服务有限公司 | Power-saving driving improvement real-time reminding method of electric vehicle rental system |
| CN109598960B (en) * | 2017-10-18 | 2021-07-27 | 宁波轩悦行电动汽车服务有限公司 | Driving suggestion method of electric vehicle rental system |
| CN109584459B (en) * | 2017-10-18 | 2021-02-26 | 宁波轩悦行电动汽车服务有限公司 | Power-saving driving suggestion method of electric vehicle rental system |
| CN109615154B (en) * | 2017-10-18 | 2022-05-17 | 宁波轩悦行电动汽车服务有限公司 | Driving data comparison method for electric vehicle rental system |
| CN110019376B (en) * | 2017-10-18 | 2022-05-17 | 宁波轩悦行电动汽车服务有限公司 | Driving comparison method for electric vehicle rental system |
| CN109598569B (en) * | 2017-10-18 | 2022-03-11 | 宁波轩悦行电动汽车服务有限公司 | Electric vehicle leasing comprehensive guiding method |
| CN110069512A (en) * | 2017-10-18 | 2019-07-30 | 宁波轩悦行电动汽车服务有限公司 | The power saving of electric car leasing system drives improvement idea querying method |
| CN108647836B (en) * | 2018-06-06 | 2021-11-12 | 哈尔滨思派科技有限公司 | Driver energy-saving evaluation method and system |
| CN108983599B (en) * | 2018-08-07 | 2021-07-16 | 重庆邮电大学 | Self-adaptive monitoring method for multi-parameter fusion in Internet of vehicles |
| CN111198379B (en) * | 2018-11-19 | 2024-02-13 | 陕西汽车集团股份有限公司 | Method for searching and positioning unloading position of unmanned transport vehicle |
| CN109686118A (en) * | 2019-01-24 | 2019-04-26 | 嘉兴行适安车联网信息科技有限公司 | A kind of economic control loop and its drive manner based on the collaboration of people's bus or train route |
| CN111311903B (en) * | 2019-02-25 | 2021-07-30 | 厦门雅迅网络股份有限公司 | Vehicle driving state monitoring method, terminal device and storage medium |
| CN110060370B (en) * | 2019-04-15 | 2021-04-27 | 上海经达信息科技股份有限公司 | Equivalent statistical method for times of rapid acceleration and rapid deceleration of vehicle |
| CN111812306A (en) * | 2019-04-29 | 2020-10-23 | 北京嘀嘀无限科技发展有限公司 | Oil product monitoring method, electronic equipment and storage medium |
| CN111907440A (en) * | 2019-05-10 | 2020-11-10 | 联合汽车电子有限公司 | Internet automobile energy-saving assistant system and control method thereof |
| CN112445969A (en) * | 2019-09-05 | 2021-03-05 | 中车时代电动汽车股份有限公司 | Man-machine interaction method and device based on Internet of vehicles |
| CN110979341A (en) * | 2019-10-08 | 2020-04-10 | 复变时空(武汉)数据科技有限公司 | Driver driving behavior analysis method and analysis system |
| CN112757898A (en) * | 2019-11-06 | 2021-05-07 | 宝马股份公司 | Method, device, vehicle and system for providing a trip summary of a vehicle |
| EP3865966B1 (en) * | 2020-02-11 | 2023-11-08 | Volkswagen Aktiengesellschaft | Method, computer program, apparatus, vehicle, and network component for controlling a maneuver within a platoon |
| CN111348039B (en) * | 2020-03-31 | 2021-06-18 | 长安大学 | Driving active intervention system and method based on acceleration change rate optimization |
| CN111626510A (en) * | 2020-05-27 | 2020-09-04 | 广西职业技术学院 | Theoretical calculation and analysis method for fuel consumption of hybrid electric vehicle |
| CN111931810A (en) * | 2020-06-29 | 2020-11-13 | 东风商用车有限公司 | Energy-saving driving behavior analysis system based on multiple vehicles |
| CN114764662A (en) * | 2021-01-12 | 2022-07-19 | 上海汽车集团股份有限公司 | Method and system for ranking vehicle information in region |
| CN112866373A (en) * | 2021-01-14 | 2021-05-28 | 浙江吉利控股集团有限公司 | Vehicle energy-saving driving guiding method and system |
| US11798412B2 (en) * | 2021-06-02 | 2023-10-24 | Guangzhou Automobile Group Co., Ltd. | Method and device for generating driving suggestion, and computer-readable storage medium |
| CN114290908A (en) * | 2021-12-06 | 2022-04-08 | 武汉展为物联科技有限公司 | Method and system for reducing energy consumption of vehicle |
| CN113928333B (en) * | 2021-12-16 | 2022-04-01 | 广东机电职业技术学院 | Energy consumption reduction method and system based on auxiliary driving |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855599B (en) * | 2011-06-28 | 2017-11-07 | 王健 | Intelligent bus system |
| TWI496709B (en) * | 2012-10-09 | 2015-08-21 | Univ Lunghwa Sci & Technology | Car driving behavior analysis system and its device |
| CN103106794B (en) * | 2013-01-11 | 2015-01-21 | 福建工程学院 | Road traffic energy-saving method based on probe vehicle technology |
| CN103489322B (en) * | 2013-08-21 | 2016-04-27 | 黑龙江工程学院 | Ecological drive assist system and the method utilizing this system supplymentary to drive |
-
2015
- 2015-10-20 CN CN201510683076.3A patent/CN105160883B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN105160883A (en) | 2015-12-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105160883B (en) | A kind of energy-saving driving behavior analysis method based on big data | |
| CN109359389B (en) | Urban electric vehicle charging decision method based on typical load dynamic game | |
| EP3850551A1 (en) | Systems and methods for managing energy storage systems | |
| CN109927709A (en) | A kind of route or travel by vehicle working condition determining method, energy management method and system | |
| CN104331953A (en) | Car behavior data identification and management method based on internet of things | |
| DE102015113701A1 (en) | Route-to-empty calculation based on route for a vehicle | |
| CN110717218B (en) | Electric drive vehicle distributed power drive system reconstruction control method and vehicle | |
| CN104092736A (en) | Vehicle networking device, server and system, scoring method and data collection method | |
| CN109552338A (en) | A kind of pure electric automobile ecology driving behavior appraisal procedure and system | |
| WO2021188647A1 (en) | Systems and methods for managing velocity profiles | |
| CN112550059B (en) | Intelligent electricity changing method and system based on big data and sharing bicycle cloud platform | |
| CN112200932A (en) | Method and equipment for evaluating energy consumption of heavy-duty diesel vehicle | |
| CN115587678A (en) | Intelligent map traveling method and system for commercial vehicle | |
| CN117409583A (en) | Vehicle state evaluation and early warning method based on intelligent network-connected automobile control center | |
| CN115221234A (en) | Method and system for portraying user based on power assembly data | |
| CN111806443A (en) | Pure electric heavy truck unmanned energy consumption optimization method | |
| CN110019376B (en) | Driving comparison method for electric vehicle rental system | |
| CN116701870A (en) | Energy consumption detection method, device, equipment and readable storage medium | |
| CN112508288A (en) | Ordered charging scheduling system and method based on temperature control load prediction | |
| CN116702996A (en) | Electric bus energy prediction method based on typical working condition recognition | |
| CN114074660A (en) | Predictive cruise fuel-saving control method and device and storage medium | |
| CN114122466B (en) | Method, system, equipment and medium for optimizing management of fuel cell vehicle battery | |
| SE538248C2 (en) | Determination of energy consumption | |
| CN109615154B (en) | Driving data comparison method for electric vehicle rental system | |
| CN109598569B (en) | Electric vehicle leasing comprehensive guiding method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |