CN115601855A - Vehicle driving condition construction method, electronic device and storage medium - Google Patents
Vehicle driving condition construction method, electronic device and storage medium Download PDFInfo
- Publication number
- CN115601855A CN115601855A CN202211504378.6A CN202211504378A CN115601855A CN 115601855 A CN115601855 A CN 115601855A CN 202211504378 A CN202211504378 A CN 202211504378A CN 115601855 A CN115601855 A CN 115601855A
- Authority
- CN
- China
- Prior art keywords
- kinematic
- vehicle
- clustering
- segments
- driving
- 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
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C5/00—Registering or indicating the working of vehicles
- G07C5/08—Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle or waiting time
- G07C5/0808—Diagnosing performance data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Feedback Control In General (AREA)
Abstract
The invention provides a vehicle running condition construction method, electronic equipment and a storage medium, wherein the method comprises the steps of firstly, obtaining and dividing running data of a plurality of vehicles in a target area to obtain a plurality of kinematic segments; then clustering is carried out according to a first clustering algorithm, and the type of each kinematic segment is determined; the initial clustering center of the first clustering algorithm is obtained through optimization of a group optimization algorithm; and finally, screening various kinematic fragments according to the Markov model, and combining the screened kinematic fragments to obtain the final driving condition of the vehicle in the target area. The vehicle driving characteristics can be classified through the first clustering algorithm, the classification result is prevented from falling into local optimum through the group optimization algorithm, and the randomness of vehicle driving is embodied through the Markov model, so that the actual driving characteristics of the electric vehicle are accurately embodied through a few kinematic segments, and the accuracy of the constructed driving condition is effectively improved.
Description
Technical Field
The application belongs to the technical field of vehicle testing, and particularly relates to a vehicle running condition construction method, electronic equipment and a storage medium.
Background
The automobile running working condition is a speed-time curve describing the running characteristics of a certain type of vehicles in a certain area under a specific traffic environment, can be used for performance evaluation of vehicle emission, oil consumption and the like, and provides reference for parameter matching and control strategy optimization of an automobile power system.
The common construction method of the driving condition is a short-stroke method, namely, data are divided into a plurality of short segments, and the corresponding driving condition is generated by analyzing and combining segment characteristic parameters. The method is characterized in that a common method in the process of establishing the working condition is to perform data dimension reduction through principal component analysis and then perform fragment screening after clustering, but the traditional clustering algorithm is easily trapped in local optimization, and the fragment screening process usually depends on expert experience, so that the influence of subjective factors is large. Therefore, the error of the driving condition constructed by the driving condition construction method in the prior art is larger.
Disclosure of Invention
In view of this, the invention provides a vehicle driving condition construction method, an electronic device and a storage medium, and aims to solve the problem that the error of the driving condition constructed in the prior art is large.
The first aspect of the embodiment of the invention provides a vehicle running condition construction method, which comprises the following steps:
acquiring driving data of a plurality of vehicles in a target area;
dividing the driving data to obtain a plurality of kinematic segments;
clustering the plurality of kinematic segments according to a first clustering algorithm to obtain the type of each kinematic segment; the initial clustering center of the first clustering algorithm is obtained through optimization of a group optimization algorithm;
and screening various kinematic segments according to the Markov model and the preset driving time, and combining the screened kinematic segments to obtain the final driving condition of the vehicle in the target area.
A second aspect of the embodiments of the present invention provides a vehicle driving condition constructing apparatus, including:
the system comprises an acquisition module, a display module and a control module, wherein the acquisition module is used for acquiring the driving data of a plurality of vehicles in a target area;
the division module is used for dividing the driving data to obtain a plurality of kinematic segments;
the clustering module is used for clustering the plurality of kinematic segments according to a first clustering algorithm to obtain the type of each kinematic segment; the initial clustering center of the first clustering algorithm is obtained by optimizing a population optimization algorithm;
and the screening module is used for screening various kinematic fragments according to the Markov model and the preset driving time length, and combining the screened kinematic fragments to obtain the final driving working condition of the vehicle in the target area.
A third aspect of the embodiments of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the vehicle driving condition construction method according to the first aspect when executing the computer program.
A fourth aspect of the embodiments of the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and the computer program, when executed by a processor, implements the steps of the vehicle running condition construction method according to the first aspect.
According to the vehicle running condition construction method, the electronic device and the storage medium provided by the embodiment of the invention, firstly, running data of a plurality of vehicles in a target area is obtained; dividing the driving data to obtain a plurality of kinematic segments; then clustering the plurality of kinematic segments according to a first clustering algorithm to obtain the type of each kinematic segment; the initial clustering center of the first clustering algorithm is obtained by optimizing a population optimization algorithm; and finally, screening various kinematic segments according to the Markov model and the preset driving time, and combining the screened kinematic segments to obtain the final driving working condition of the vehicle in the target area. The method comprises the steps of classifying according to the driving characteristics of a vehicle through a first clustering algorithm, avoiding classification results from falling into local optima by combining a group optimization algorithm, and screening the kinematics segments through a Markov model in order to reflect the randomness of vehicle speed change and avoid subjective influence caused by artificial screening because the vehicle speed in the actual driving process of the vehicle continuously changes along with the traffic condition, so that the screened kinematics segments can be closer to the actual running condition of the vehicle, reflect the influence of regional traffic conditions on the driving condition of the vehicle to a certain extent, realize that the actual driving characteristics of the electric vehicle are accurately reflected through a few kinematics segments, and effectively improve the accuracy of the constructed driving condition.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
FIG. 1 is an application scenario diagram of a vehicle driving condition construction method provided by an embodiment of the invention;
FIG. 2 is a flowchart illustrating an implementation of a method for constructing a driving condition of a vehicle according to an embodiment of the present invention;
FIG. 3 is a velocity profile corresponding to various operating conditions provided by the embodiment of the present invention;
FIG. 4 is a flow chart of an implementation of the vehicle driving condition construction method of the present invention and prior art;
FIG. 5 is a graph of speed of a plurality of vehicles in a target area as provided by an exemplary embodiment of the present invention;
FIG. 6 illustrates a latitude and longitude of a plurality of vehicle travel routes for a target area, in accordance with an exemplary embodiment of the present invention;
FIG. 7 is a graph of velocity comparison before and after filtering provided by an embodiment of the present invention;
FIG. 8 is a principal component eigenvalue lithograph provided by an embodiment of the present invention;
FIG. 9 is a diagram of an elbow rule provided by an exemplary embodiment of the present invention;
FIG. 10 is a graph of mean profile coefficient values provided by an embodiment of the present invention;
fig. 11 is a graph of distance traveled-average speed-hundred kilometers electricity consumption distribution for various types of kinematic segments provided by an exemplary embodiment of the present invention;
FIG. 12 is a graph of average speed versus average torque and speed of the motor for various kinematic segments provided by an exemplary embodiment of the present invention;
FIG. 13 is an acceleration-idle-uniform velocity ratio profile for various types of kinematic segments provided by an exemplary embodiment of the present invention;
FIG. 14 is a diagram illustrating a final driving condition of a vehicle in a target area according to an exemplary embodiment of the present invention;
FIG. 15 is a velocity-acceleration combined distribution diagram of original sample data provided by an embodiment of the present invention;
FIG. 16 is a combined final driving condition speed-acceleration profile provided by an exemplary embodiment of the present invention;
FIG. 17 is a schematic structural diagram of a vehicle driving condition construction device according to an embodiment of the present invention;
fig. 18 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
FIG. 1 is an application scenario diagram of a vehicle driving condition construction method provided by an embodiment of the invention. As shown in fig. 1, in some embodiments, the vehicle driving condition construction method provided by the embodiment of the present invention may include, but is not limited to, being applied to the application scenario. In an embodiment of the invention, the system comprises: a plurality of vehicles 11, and an electronic device 12.
The vehicle 11 is a vehicle selected in advance in a target area, the vehicle can be a passenger vehicle or a commercial vehicle, for example, drivers and driven vehicles can be recruited in a voluntary recruitment mode, drivers with different genders, ages and driving ages are selected for considering driving habits of different drivers, an electric vehicle for collecting data is selected based on the popular vehicle type and the vehicle retention rate of the target area, and the test time is selected to cover the peak period, the peak period and the low peak period of the work in one day. The selected vehicle runs autonomously, namely no specific regulation is made on the running route, and the driver can randomly select the running route according to the own normal route arrangement.
After the selection is completed, the vehicle 11 uploads a GPS signal to the electronic device 12 in real time at a preset frequency through the vehicle-mounted GPS, and the electronic device 12 constructs a vehicle driving condition after receiving the GPS signals of the plurality of vehicles 11.
The electronic device 12 may be a server or a terminal, the server may be a cloud server or a physical server, and the terminal may be a computer, a notebook, or the like, which is not limited herein.
FIG. 2 is a flowchart of an implementation of the method for constructing a driving condition of a vehicle according to the embodiment of the present invention. As shown in FIG. 2, in some embodiments, the vehicle driving condition construction method comprises the following steps:
s201, the traveling data of a plurality of vehicles in the target area is acquired.
In an embodiment of the present invention, the driving data may include, but is not limited to, at least one of the following: vehicle speed, longitude, latitude.
And S202, dividing the driving data to obtain a plurality of kinematic segments.
In the embodiment of the invention, the division of the kinematic segments can be performed through the speed v and the acceleration a. For example:
acceleration condition:a≥0.15m·s -2 and the deceleration working condition is as follows:a≤-0.15m·s -2 and idling condition: -0.15 m.s -2 ≤a≤0.15m·s -2 And is provided withv≤0.5km·h -1 And the constant speed working condition is as follows: -0.15 m.s -2 ≤a≤0.15m·s -2 And is provided withv≥0.5km·h -1 。
FIG. 3 is a velocity profile corresponding to various operating conditions provided by the embodiment of the present invention. As shown in fig. 3, the driving process of the automobile can be regarded as being formed by splicing a large number of kinematic segments, wherein the kinematic segments refer to a speed interval from the beginning of idling to the beginning of the next idling of the automobile and are formed by working conditions of acceleration, deceleration, constant speed and idling.
S203, clustering the plurality of kinematic segments according to a first clustering algorithm to obtain the type of each kinematic segment; and the initial clustering center of the first clustering algorithm is obtained by optimizing a population optimization algorithm.
In the embodiment of the invention, after division of the kinematic segment, the whole motion process cannot be completely embodied only by two parameters of speed and acceleration, and some characteristic parameters are required to be introduced to describe the actual working condition of the vehicle close to the real condition. In order to accurately describe the states and characteristics of the various kinematic segments, 16 characteristic parameters are selected as automobile driving characteristic evaluation indexes. In particular the run timeT/sDistance traveledSM, average velocityv m /(km·h -1 ) Average running speedv mr /(km·h -1 ) Maximum speed of the motorv max /(km·h -1 ) Standard deviation of velocityv std /(km·h -1 ) Average accelerationa am /(m·s -2 ) Average decelerationa dm /(m·s -2 ) Maximum accelerationa max /(m·s -2 ) Maximum decelerationa min /(m·s -2 ) Average absolute value of accelerationa ab_m /(m·s -2 ) Absolute value standard deviation of accelerationa ab_std /(m·s -2 ) Acceleration time ratioP a /%, deceleration timeRatio ofP d /%, ratio of idle timeP i /%, uniform velocity time ratioP c /%。
In the embodiment of the invention, the characteristics of the kinematic segments are reflected according to the characteristic parameters, the performance of the vehicle can be embodied, and then the kinematic segments are clustered according to the characteristic parameters and the first clustering algorithm, so that the types of the kinematic segments are determined. Types of kinematic fragments may include, but are not limited to, at least one of: low-speed segment, medium-speed segment, high-speed segment. The colony optimization algorithm may be a particle swarm algorithm, an ant colony algorithm, etc., and is not limited herein.
And S204, screening various kinematic segments according to the Markov model and the preset driving time, and combining the screened kinematic segments to obtain the final driving condition of the vehicle in the target area.
In the embodiment of the invention, the automobile driving process has strong randomness and no aftereffect and has a Markov characteristic, namely, the state of the automobile driving process at the time t +1 is only related to the time t and is not related to the time t-1. Therefore, the Markov model is used for calculating the state transition probability matrix by utilizing maximum likelihood estimation, and then the fragment screening is carried out according to the state transition probability matrix, so that the driving characteristics of the vehicle can be effectively embodied, and the kinematic fragment obtained by combination is more consistent with the vehicle condition of a target area.
In the embodiment of the invention, the vehicle running condition construction method, the electronic device and the storage medium provided by the embodiment of the invention firstly obtain the running data of a plurality of vehicles in a target area; dividing the driving data to obtain a plurality of kinematic segments; then clustering the plurality of kinematic segments according to a first clustering algorithm to obtain the type of each kinematic segment; the initial clustering center of the first clustering algorithm is obtained by optimizing a population optimization algorithm; and finally, screening various kinematic fragments according to the Markov model and the preset driving time, and combining the screened kinematic fragments to obtain the final driving condition of the vehicle in the target area. The classification can be carried out according to the driving characteristics of the vehicle through the first clustering algorithm, the classification result is prevented from falling into local optimization through the group optimization algorithm, the randomness of vehicle driving is reflected through the Markov model, the actual driving characteristics of the electric vehicle are accurately reflected through a few kinematic segments, and the accuracy of the constructed driving working condition is effectively improved.
In some embodiments, S204 may include: carrying out maximum likelihood estimation on various kinematic fragments to determine a state transition probability matrix; and screening various kinematic fragments according to the state transition probability matrix and the preset driving time.
In the embodiment of the invention, the driving condition slave state of the electric automobile can be definediIs converted into a statejThe state transition probability matrix of (a) is:
wherein the content of the first and second substances, (ii) (p ij ) I I× In the form of a matrix of state transition probabilities,p ij is in the current state ofiThe next state isjThe kinematic segment of (a), wherein,p ij the value of (d) can be derived from a maximum likelihood estimate,Ithe number of types of the kinematic fragment.
In some embodiments, the screening of the various types of kinematic segments according to the state transition probability matrix and the preset driving duration includes: randomly selecting a first kinematic segment from the plurality of kinematic segments, determining the type of a second kinematic segment to be screened from the plurality of kinematic segments according to the type of the first kinematic segment and the state transition probability matrix, and screening the second kinematic segment from the plurality of kinematic segments according to the type of the second kinematic segment to be screened; determining the type of a third kinematic segment to be screened from the plurality of kinematic segments according to the type of the second kinematic segment and the state transition probability matrix, screening the third kinematic segment from the plurality of kinematic segments according to the type of the third kinematic segment to be screened until the ith kinematic segment is screened from the plurality of kinematic segments, wherein i is a positive integer larger than or equal to 4, and stopping screening; the screened ith kinematic segment and the total time length of all the kinematic segments with the serial numbers smaller than that of the ith kinematic segment are not smaller than the preset running time length; the state transition probability matrix includes a plurality of transition probabilities that represent probabilities of the kinematic segment transitioning between the respective types.
In an embodiment of the invention, a first kinematic segment, i.e. an initial state of the driving situation, is first determined. The first kinematic segment may be preset, or a kinematic segment with the highest feature correlation between the feature parameter and the feature of the initial state may be selected from a large number of kinematic segments, which is not limited herein.
In the embodiment of the invention, the type of the second kinematic segment can be determined according to the first kinematic segment and the state transition probability matrix, so that the second kinematic segment is selected, then the type of the third kinematic segment is determined according to the second kinematic segment and the state transition probability matrix, and the process is continuously executed until the total time length of the selected kinematic segments meets the preset time length, so that the screening of the kinematic segments is completed.
Since the state transition matrix can only determine the type of the selected kinematic segment, and the number of each type of kinematic segment is large, the selected kinematic segment is still randomly selected when the specific kinematic segment is selected. In some embodiments, after the type of the kinematics segments screened out according to the state transition matrix, the duration of each kinematics segment in the type may be further divided according to a preset division standard according to the duration of each kinematics segment in the type, so as to obtain a division result, and the kinematics segments in the type are determined to be screened according to the division result.
For example, if 3 types of kinematic segments are obtained by clustering, the segments are high-speed segments, medium-speed segments and low-speed segments, the number of the segments is 200, 668 and 619, the first segment screened is a high-speed segment, all the high-speed segments are divided into three high-speed segments a, B and C, the duration of which is less than t1, the duration of which is between t1 and t2, and the duration of which is greater than t2, and the number ratio is a: b: c =3:5:2, the first screened high-speed segment is a high-speed segment with the time length between t1 and t2, and when the high-speed segment is screened subsequently, the ratio of the screened three high-speed segments A, B and C is close to 3:5:2. and further screening is carried out according to the segment duration, so that the finally obtained running condition is more consistent with the actual running condition of the automobile in the region.
In some embodiments, S203 may include: and extracting characteristic parameters of the plurality of kinematic segments, and performing principal component analysis on the characteristic parameters to obtain a characteristic parameter matrix. And clustering the characteristic parameter matrix according to a first clustering algorithm, and determining the types of the plurality of kinematic segments.
In the embodiment of the invention, the principal component analysis PCA algorithm is adopted to perform the dimensionality reduction processing on the characteristic parameters of the kinematic fragments, and the information contained in the characteristic parameters of the original data is expressed through a plurality of unrelated principal components, so that the time for performing the clustering analysis operation on the characteristic parameter matrix can be reduced, and the principal components with the accumulated contribution rate of more than 85 percent or the characteristic value of more than 1 are usually selected.
Firstly, the original characteristic parameter matrix is standardized, so that the condition that the clustering analysis result is influenced due to the fact that the characteristic parameter values are dispersed to a large extent because characteristic parameter units are not uniform is avoided.
In the embodiment of the invention, the first four principal components can be selected after the principal components are analyzed and subjected to dimensionality reduction.
The characteristic parameters reflected by the first principal component are: average deceleration, maximum deceleration, idle time ratio;
the characteristic parameters reflected by the second principal component are: running time, running distance, average speed, average running speed, maximum speed and uniform speed-time ratio;
the third and fourth principal components mainly reflect an acceleration time ratio and a deceleration time ratio.
In some embodiments, S203 may include: and extracting characteristic parameters of the plurality of kinematic segments, and inputting the characteristic parameters into an automatic encoder to obtain a characteristic parameter matrix. And clustering the characteristic parameter matrix according to a fuzzy clustering algorithm, and determining the types of the plurality of kinematic segments.
In the embodiment of the invention, the automatic encoder is one of neural networks, and the basic idea is to directly use one or more layers of neural networks to map input data to obtain an output vector as a feature extracted from the input data. Conventional autoencoders are typically used for data dimensionality reduction or feature learning, similar to PCA, but are much more flexible than PCA in that they can characterize both linear and nonlinear transformations.
In some embodiments, the first clustering algorithm is a K-means clustering algorithm. Before S203, the vehicle driving condition construction method further includes: and determining an initial clustering center of the K-means clustering algorithm according to the particle swarm algorithm.
In the embodiment of the invention, compared with a density-based clustering algorithm, a fuzzy clustering algorithm and the like, the K-means clustering algorithm has a better clustering effect when clustering the kinematic segments, but the K-means clustering algorithm is more likely to fall into local optimization relative to other clustering algorithms, so that the initial clustering center of the K-means clustering algorithm can be determined according to the particle swarm algorithm before clustering by combining the global search capability of the particle swarm algorithm, thereby improving the accuracy of clustering.
Due to the fact that parameters such as coefficients and maximum speed are not properly selected, the convergence speed and accuracy of the particle swarm optimization are affected. The global search and the local search can be balanced by adjusting the inertia weight, and then the classification is carried out by using a K-means clustering algorithm to make the initial clustering centers as far as possible. The method comprises the following specific steps:
1. by initializing the population. Randomly generating initial cluster center, initial position of particlex i And velocityv i Learning factorb i . Calculating Euclidean distance from various types of internal data to clustering center by formula (2)f i The minimum fitness value is used as an individual extremum of the particle and is used as a global extremum.
Wherein the content of the first and second substances,c ij is shown as the firstiClass II ofjThe number of the samples is one,m、c i are respectively the firstiThe number of the class data and the clustering center,kis the number of clusters to be formed,Tis the sign of the transposed matrix.
2. Inertial weight of pass equation (3)wAnd (3) realizing the coarse global search to the local fine search of the particles, updating the positions and the speeds of the particles of the whole particle swarm by the formulas (4) and (5), and calculating the fitness value of the updated particles.
Wherein the content of the first and second substances,w max 、w min the weight maximum and the weight minimum are respectively,f best 、f bad the best and worst fitness values for the particles, respectively.rand 1 、rand 2 Is a random number in the range of 0 to 1,P besti is the best position of the particles and is,G bestd in order to be a global extreme value,v i (t) Is a firsttThe velocity of the particles at the time of the sub-iteration,x i (t) Is as followstThe position of the particle at the time of the second iteration.
3. And judging whether the current particle reaches a convergence state or not according to the fitness variance of the particle swarm.f a The average fitness of the particle swarm is shown. When the fitness variance is less than the set threshold of 0.1, then the population tends to converge. And selecting 10 optimal particles to perform k-means clustering.
4. And calculating the Euclidean distance between each sample and the current clustering center, and performing particle clustering division according to the distance. And selecting the next clustering center on the basis of a wheel disc method, and updating the fitness value of the particle. And judging whether the fitness value is better or reaches the maximum iteration number, and ending the iteration if the fitness value is optimal or reaches the maximum iteration number.
In some embodiments, before S203, the vehicle driving condition construction method further includes: and determining an initial clustering center according to a random operator and a cross operator adaptive particle size regulation subgroup algorithm.
In the embodiment of the invention, the global characteristic of the particle swarm algorithm is increased, so that the clustering effect is further improved. The self-adaptive adjustment can be realized according to a random operator and a cross operator. In the iterative optimization process, the population diversity can be increased by adopting a crossover operator, and the optimal solution is calculated again by adopting a random operator after the optimal solution is obtained. And (3) carrying out self-adaptive mutation particle swarm algorithm of the crossover operator. Mainly by
And for the crossover operator, in each iteration, the first half of the particles with good fitness after sequencing are directly taken to enter the next generation, the second half of the particles are put into a particle selection pool to be pairwise paired, a random crossover position is generated to carry out genetic selection and crossover operation, filial generations with the same number as the parent generation are generated and then compared with the parent generation, and the half with good fitness is selected to enter the next generation so as to keep the number of the particles in the population unchanged. Therefore, not only can the population diversity be increased and the local optimum be jumped out, but also the convergence speed can be accelerated through the crossover operator.
For random operators, consider the current global extremum of the particlegBetter positions can be found under the action of Best, and the method meets the variation conditiongBest according to a certain probabilityp m And (5) carrying out mutation.p m The calculation formula of (c) is as follows:
wherein, the first and the second end of the pipe are connected with each other,zin order to preset the variation value, the variation value is set,σ 2 is the variance of the population fitness and is,σ 2 d is the maximum value of the variance of the fitness measure,f(gbest) is the current global extremum,f d is the desired optimal solution.
In some embodiments, before S203, the vehicle driving condition construction method further includes: and determining the optimal clustering number of the K-means clustering algorithm according to an elbow rule or a contour coefficient algorithm.
In the embodiment of the present invention, the contour coefficient function is:
wherein, the first and the second end of the pipe are connected with each other,α(i) In the same cluster, sampleiThe average distance from other samples, i.e. intra-cluster dissimilarity,b(i) Is a sampleiThe minimum of the average distances of all the points in the adjacent closest cluster, i.e. the dissimilarity between clusters.s(i)Averaging the contour coefficients of all the points to obtain the total contour coefficient of the clustering result,s(i)if the value is close to 1, the sample is illustratediThe clustering is reasonable; when the temperature is higher than the set temperatures(i)When the value is 0, the similarity of the samples in the two clusters is consistent, and the two clusters are the same cluster.
In the embodiment of the invention, the elbow rule is as follows:
as the clustering number k increases, the sample division becomes finer, the degree of clustering of each cluster gradually increases, and the sum of squares of errors becomes smaller. When k is smaller than the true number of clusters, the sum of squared errors will be large since an increase in k will increase the degree of aggregation per cluster. When k is close to the true clustering number and k is increased, the descending amplitude of the error square sum is suddenly reduced and becomes gentle as the k value is continuously increased.
Wherein, the first and the second end of the pipe are connected with each other,c ij is shown asiClass II ofjThe number of the samples is one,
is shown asiThe mean value of the class samples is calculated,kto be the number of clusters,m i is a firstiThe number of the class data is,SSE i is as followsiThe sum of the squared errors of the classes.
In some embodiments, after obtaining the driving data of the plurality of vehicles in the target area, the vehicle driving condition construction method further includes: and carrying out interpolation filling and abnormal value elimination on the running data of a plurality of vehicles in the target area.
In the embodiment of the invention, in the data acquisition process, under the influence of the accuracy of acquisition equipment, traffic environment and the like, the data has the phenomena of deletion, abnormality and the like, and the data quality is reduced to some extent. In order to ensure the reliability of the data, the data needs to be preprocessed and analyzed. The method comprises the following specific steps:
(1) GPS signals lack data. Because of the sheltering of high-rise buildings, tunnels and other road sections, the GPS signal positioning is inaccurate or discontinuous, and the vehicle speed data is lacked. And processing by adopting an interpolation method or a rejection method.
(2) And (4) processing idle speed data. Due to long-time traffic jam, the automobile is in a non-working interval for a long time, and the acquired data does not meet the requirements. The automobile is driven intermittently with the maximum speed less than 10 km.h -1 Considered as idle; and (3) directly removing the data after the vehicle speed is 0 and the duration time is less than 180 seconds as a screening principle.
(3) Maximum speed, acceleration limit. The automobile mainly runs in urban areas or suburban areas, and the automobile speed is limited to 120 km.h -1 Within, the acceleration is limited to-6 to 6 m.s -2 Within.
(4) And (4) speed filtering. Due to the influence of external factors, abnormal noise interference phenomenon exists in the driving data, and errors exist in the data. The original data is filtered by adopting a sliding average filtering algorithm, and the original data at the corresponding position is replaced by the average value of a plurality of data in the neighborhood of a sliding window with a fixed length to form an average value new sequence.
Wherein the content of the first and second substances,y(t') Is the average value of the values,t'=1,2...n。nfor the purpose of the total data length,T 0 in order to be a step of time,l=1,2...T 0 ，x(t') Is the raw speed data.
In some embodiments, after S201, the vehicle driving condition construction method further includes: determining at least one similar vehicle of the target vehicle according to the driving data of the vehicles in the target area; the target vehicle is a vehicle with running data missing; the correlation between the running data of the similar vehicles and the running data of the target vehicle is greater than a preset threshold value; and filling the running data of the target vehicle according to the running data corresponding to at least one similar vehicle and a singular value threshold algorithm.
In some embodiments, after S204, the vehicle driving condition construction method further includes: extracting a first characteristic parameter of the driving data of the vehicle in the target area; extracting a second characteristic parameter of the final driving condition of the vehicle in the target area; verifying an error between the first characteristic parameter and the second characteristic parameter; and if the error is larger than the preset error, jumping to S204, and re-screening the plurality of kinematic segments. The first characteristic parameter represents the automobile performance characteristic corresponding to the driving data of the vehicle; the second characteristic parameter represents the automobile performance characteristic corresponding to the final running working condition of the vehicle.
The characteristic parameters reflecting the performance characteristics of the automobile can include, but are not limited to, at least one of the following: the running time, the running distance, the average speed, the average running speed, the maximum speed, the speed standard deviation, the average acceleration, the average deceleration, the maximum acceleration, the minimum acceleration, the acceleration time ratio, the deceleration time ratio, the idle time ratio and the uniform speed time ratio.
In the embodiment of the invention, the working condition construction and comparison are carried out by combining the particle swarm algorithm and the K-means clustering with the traditional K-means clustering, and the relative error between the constructed working condition and the characteristic parameters of the sample data is calculated by the following formula.
Wherein, the delta is a relative error,C k' 、U k' the first to construct working condition and original sample data respectivelyk'The characteristic parameters of the first and second groups are,n'the number of the characteristic parameters.
In the embodiment of the present invention, in each filtering, the markov model only determines the type of the next kinematic segment according to the type of the current kinematic segment, and does not determine the specifically selected kinematic segment, for example, when the type of the current kinematic segment is a high-speed segment, and the next kinematic segment is a medium-speed segment, the next kinematic segment may be a medium-speed segment B 1 Or a moderate speed segment B 2 Even if the type and order of each motion segment determined during the rescreening process is the same, different segments may be selected. Moreover, the type of the selected kinematics segment depends on not only the state transition probability matrix but also the selection of the first kinematics segment, wherein the first kinematics segment selects high-speed, medium-speed and low-speed segments, and the corresponding next kinematics segments are different.
The following provides an implementation example to explain the vehicle driving condition construction method of the present invention, but the method is not limited thereto. FIG. 4 is a flow chart of an implementation of the vehicle driving condition construction method of the present invention and the prior art. As shown in fig. 4, for comparative analysis of the advantages of the present invention compared to the prior art, the method of the present invention and the principal component + conventional K-means clustering analysis method are respectively used for construction of driving conditions. In this embodiment, the steps of the method of the present invention are specifically:
the method comprises the steps of firstly, acquiring vehicle speed data and longitude and latitude data. The data obtained are shown in fig. 5 and 6. Fig. 5 is a graph of the speed of a plurality of vehicles in a target area, as provided by an exemplary embodiment of the present invention. Where the horizontal axis is time and the vertical axis is speed. Fig. 6 illustrates the latitude and longitude of a plurality of vehicle driving routes in a target area according to an exemplary embodiment of the present invention. Wherein the horizontal axis is longitude and the vertical axis is latitude.
And secondly, preprocessing, namely performing filling, abnormal value removing, filtering and other processing on the acquired data. Thus, the speed-versus-speed curves before and after the treatment shown in fig. 7 were obtained. Fig. 7 is a graph of velocity versus time before and after filtering as provided by an embodiment of the present invention. Where the horizontal axis is time and the vertical axis is speed.
And thirdly, performing principal component analysis on the characteristic parameters of the preprocessed data to realize data dimension reduction and reduce clustering time. Fig. 8 is a principal component eigenvalue lithograph provided by an embodiment of the present invention. Wherein, the horizontal axis is the number of the principal components, and the vertical axis is the eigenvalue. As shown in fig. 8, the first 5 principal components may be selected for cluster analysis.
And fourthly, selecting the number of clusters. Fig. 9 is a diagram of elbow rules provided by an exemplary embodiment of the present invention. Wherein the horizontal axis is the cluster number and the vertical axis is the sum of the squares of the errors. As shown in fig. 9, when the number of clusters is less than 3, the sum of squared errors sharply decreases, and when the number of clusters is greater than 3, the sum of squared errors tends to be flat. The number of clusters can be chosen to be 3. Figure 10 is a graph of mean profile coefficient values provided by an example embodiment of the present invention. As shown in fig. 10, when the number of clusters is 3, the average contour coefficient is the highest, and the number of clusters can be selected to be 3.
And fifthly, optimizing the initial clustering center according to a particle swarm optimization algorithm.
And sixthly, performing clustering analysis according to a K-means clustering algorithm. Fig. 11 is a graph of distance traveled-average speed-hundred kilometers power consumption distribution for various types of kinematic segments provided by an exemplary embodiment of the present invention. Fig. 12 is a graph of average speed versus average torque and speed of the motor for various types of kinematic segments provided by an exemplary embodiment of the present invention. Fig. 13 is a graph of acceleration-idle-uniform velocity ratio profiles for various types of kinematic segments provided by an exemplary embodiment of the present invention. As shown in FIGS. 11-13, the obtained 3 types of kinematic segments have the shortest average driving distance of about 77m in type 1, the highest idle ratio, and the speed distribution of 0-10 km h -1 And the traffic characteristics when the traffic system runs on the urban congestion road section can be reflected. In category 2, the speed is distributed between 10 and 20 km.h -1 Acceleration, deceleration, etc,The uniform speed and the idle speed are relatively uniform, the running distance is moderate, and the traffic characteristics when the vehicle runs on the urban main road can be reflected. The class 3 speed is distributed between 20km and 40 km.h -1 The running time, the running distance, the maximum speed, the average running speed, the maximum acceleration and the maximum deceleration are higher than the other two types, and the traffic characteristics of running on the suburban road can be reflected.
And seventhly, screening the kinematic segments according to the Markov model, and splicing the screened kinematic segments to obtain the final driving working condition of the vehicle. FIG. 14 is a diagram illustrating a final driving condition of a vehicle in a target area according to an exemplary embodiment of the present invention. Wherein the horizontal axis is time, the vertical axis is speed, and the time length of the final running condition of the vehicle is 1500s.
Fig. 15 is a velocity-acceleration combined distribution diagram of original sample data provided by an embodiment of the present invention. FIG. 16 is a combined final driving condition speed-acceleration profile provided by an exemplary embodiment of the present invention. As shown in FIGS. 15 and 16, the constructed driving conditions of the vehicle and the velocity-acceleration combined distribution of the total sample data are distributed uniformly as a whole, and the multiple positions are located at the velocity of 0-40 km.h -1 The region of lower acceleration indicates that the constructed operating condition is consistent with the overall sample data.
In conclusion, the beneficial effects of the invention are as follows: the classification can be carried out according to the driving characteristics of the vehicle through the first clustering algorithm, the classification result is prevented from falling into local optimization through the group optimization algorithm, the randomness of vehicle driving is reflected through the Markov model, the actual driving characteristics of the electric vehicle are accurately reflected through a few kinematic segments, and the accuracy of the constructed driving working condition is effectively improved.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
Fig. 17 is a schematic structural diagram of a vehicle driving condition construction device according to an embodiment of the present invention. As shown in fig. 17, in some embodiments, the vehicle driving condition constructing device 17 includes:
an obtaining module 1710, configured to obtain driving data of a plurality of vehicles in a target area.
The dividing module 1720 is configured to divide the driving data to obtain a plurality of kinematic segments.
A clustering module 1730, configured to cluster the plurality of kinematic segments according to a first clustering algorithm to obtain a type of each kinematic segment; and the initial clustering center of the first clustering algorithm is obtained by optimizing a population optimization algorithm.
And the screening module 1740 is configured to screen various kinematic segments according to the markov model and the preset driving duration, and combine the screened kinematic segments to obtain a final driving condition of the vehicle in the target area.
Optionally, the screening module 1740 is specifically configured to perform maximum likelihood estimation on the various kinematic segments, and determine a state transition probability matrix; and screening various kinematic fragments according to the state transition probability matrix and the preset driving time.
Optionally, correspondingly, the screening module 1740 is specifically configured to randomly select a first kinematic segment from the multiple kinematic segments, determine a type of a second kinematic segment to be screened from the multiple kinematic segments according to the type and the state transition probability matrix of the first kinematic segment, and screen the second kinematic segment from the multiple kinematic segments according to the type of the second kinematic segment to be screened; determining the type of a third kinematic fragment to be screened from the plurality of kinematic fragments according to the type of the second kinematic fragment and the state transition probability matrix, screening the third kinematic fragment from the plurality of kinematic fragments according to the type of the third kinematic fragment to be screened until the ith kinematic fragment is screened from the plurality of kinematic fragments, and stopping screening when i is a positive integer more than or equal to 4; the screened ith kinematic segment and the total duration of all the kinematic segments with the serial numbers smaller than the ith kinematic segment are not smaller than the preset running duration; the state transition probability matrix includes a plurality of transition probabilities that represent probabilities of the kinematic segment transitioning between the respective types.
Optionally, the clustering module 1730 is specifically configured to extract feature parameters of a plurality of kinematic segments, and perform principal component analysis on the feature parameters to obtain a feature parameter matrix. And clustering the characteristic parameter matrix according to a first clustering algorithm, and determining the types of the plurality of kinematic segments.
Optionally, the first clustering algorithm is a K-means clustering algorithm. The vehicle driving condition constructing apparatus 17 further includes: and the optimization module is used for determining an initial clustering center of the K-means clustering algorithm according to the particle swarm algorithm.
Optionally, the optimization module is further configured to determine an initial clustering center according to a random operator and a cross operator adaptive particle size regulation subgroup algorithm.
Optionally, the vehicle driving condition constructing device 17 further includes: and the selecting module is used for determining the optimal clustering number of the K-means clustering algorithm according to an elbow rule or a contour coefficient algorithm.
Optionally, the vehicle driving condition constructing device 17 further includes: and the preprocessing module is used for carrying out interpolation filling and abnormal value elimination on the driving data of the vehicle in the target area.
Optionally, the preprocessing module may be further configured to determine at least one similar vehicle of the target vehicle according to the driving data of the vehicles in the target area; the target vehicle is a vehicle with running data missing; the correlation between the running data of the similar vehicles and the running data of the target vehicle is greater than a preset threshold value; and filling the running data of the target vehicle according to the running data corresponding to at least one similar vehicle and a singular value threshold algorithm.
Optionally, the vehicle driving condition constructing device further includes: the judging module is used for extracting a first characteristic parameter of the driving data of the vehicle in the target area; extracting a second characteristic parameter of the final driving condition of the vehicle in the target area; verifying an error between the first characteristic parameter and the second characteristic parameter; and if the error is larger than the preset error, skipping to the step of screening various kinematic segments according to the Markov model and the preset driving time length, and re-screening the plurality of kinematic segments.
The first characteristic parameter represents automobile performance characteristics corresponding to the running data of the vehicle; the second characteristic parameter represents the automobile performance characteristic corresponding to the final running condition of the vehicle;
the vehicle driving condition constructing device provided by the embodiment can be used for executing the method embodiment, the implementation principle and the technical effect are similar, and the embodiment is not described again.
Fig. 18 is a schematic diagram of an electronic device provided in an embodiment of the present invention. As shown in fig. 18, an embodiment of the present invention provides an electronic device 18, where the electronic device 18 of the embodiment includes: a processor 1800, a memory 1810, and computer programs 1820 stored in the memory 1810 and executable on the processor 1800. The processor 1800, when executing the computer program 1820, implements the steps of the various vehicle driving condition construction method embodiments described above, such as S210 to S240 shown in fig. 2. Alternatively, the processor 1800, when executing the computer program 1820, implements the functions of the modules/units in the system embodiments described above, such as the functions of the modules 1710 to 1740 shown in fig. 17.
Illustratively, the computer program 1820 may be partitioned into one or more modules/units, which are stored in the memory 1810 and executed by the processor 1800, to implement the invention. One or more of the modules/units may be a series of computer program instruction segments capable of performing certain functions that are used to describe the execution of the computer program 1820 in the electronic device 18.
The electronic device 18 may be a server, a terminal, or the like, and is not limited herein. The terminal can include, but is not limited to, a processor 1800, a memory 1810. Those skilled in the art will appreciate that fig. 18 is merely an example of an electronic device 18 and is not intended to limit the electronic device 18 and may include more or fewer components than shown, or some components may be combined, or different components, e.g., the terminal may also include input-output devices, network access devices, buses, etc.
The Processor 1800 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The storage 1810 may be an internal storage unit of the electronic device 18, such as a hard disk or a memory of the electronic device 18. The memory 1810 may also be an external storage device of the electronic device 18, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), etc. provided on the electronic device 18. Further, memory 1810 may also include both internal and external storage units of electronic device 18. The memory 1810 is used for storing computer programs and other programs and data required by the terminal. The memory 1810 may also be used to temporarily store data that has been output or is to be output.
The embodiment of the invention provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the steps in the embodiment of the vehicle running condition construction method are realized.
The computer-readable storage medium stores a computer program 1820, the computer program 1820 includes program instructions, which when executed by the processor 1800 implement all or part of the processes of the method of the embodiments, and can also be implemented by the computer program 1820 instructing associated hardware, and the computer program 1820 can be stored in a computer-readable storage medium, and the computer program 1820 can implement the steps of the method embodiments when executed by the processor 1800. The computer program 1820 includes, among other things, computer program code, which can be in the form of source code, object code, an executable file or some intermediate form. The computer readable medium may include: any entity or device capable of carrying computer program code, recording medium, U.S. disk, removable hard disk, magnetic disk, optical disk, computer Memory, read-Only Memory (ROM), random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution media, and the like.
The computer readable storage medium may be an internal storage unit of the terminal of any of the foregoing embodiments, for example, a hard disk or a memory of the terminal. The computer readable storage medium may also be an external storage device of the terminal, such as a plug-in hard disk provided on the terminal, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like. Further, the computer-readable storage medium may also include both an internal storage unit and an external storage device of the terminal. The computer-readable storage medium is used for storing a computer program and other programs and data required by the terminal. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
It should be clear to those skilled in the art that, for convenience and simplicity of description, the foregoing division of the functional units and modules is only used for illustration, and in practical applications, the above function distribution may be performed by different functional units and modules as needed, that is, the internal structure of the device is divided into different functional units or modules, so as to perform all or part of the above described functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the technical solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal and method may be implemented in other ways. For example, the above-described apparatus/terminal embodiments are merely illustrative, and for example, a module or a unit may be divided into only one type of logical function, and may be implemented in another manner, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in the form of hardware, or may also be implemented in the form of a software functional unit.
The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, all or part of the flow in the method according to the embodiments of the present invention may also be implemented by a computer program instructing related hardware, and the computer program may be stored in a computer readable storage medium, and when executed by a processor, the computer program may implement the steps of the above-described embodiments of the method. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying computer program code, recording medium, U.S. disk, removable hard disk, magnetic disk, optical disk, computer Memory, read-Only Memory (ROM), random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution media, and the like.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. A vehicle driving condition construction method is characterized by comprising the following steps:
acquiring running data of a plurality of vehicles in a target area;
dividing the driving data to obtain a plurality of kinematic segments;
clustering the plurality of kinematic segments according to a first clustering algorithm to obtain the type of each kinematic segment; the initial clustering center of the first clustering algorithm is obtained by optimizing a population optimization algorithm;
and screening various kinematic segments according to the Markov model and the preset driving time, and combining the screened kinematic segments to obtain the final driving condition of the vehicle in the target area.
2. The vehicle driving condition construction method according to claim 1, wherein the screening of the various types of kinematic segments according to the markov model and the preset driving duration comprises:
carrying out maximum likelihood estimation on various kinematic fragments to determine a state transition probability matrix;
and screening various kinematic fragments according to the state transition probability matrix and the preset driving time.
3. The vehicle driving condition construction method according to claim 2, wherein the screening of various kinematic segments according to the state transition probability matrix and the preset driving duration comprises:
randomly selecting a first kinematic segment from the plurality of kinematic segments, determining the type of a second kinematic segment to be screened from the plurality of kinematic segments according to the type of the first kinematic segment and the state transition probability matrix, and screening the second kinematic segment from the plurality of kinematic segments according to the type of the second kinematic segment to be screened;
determining the type of a third kinematic segment to be screened from the plurality of kinematic segments according to the type of the second kinematic segment and the state transition probability matrix, screening the third kinematic segment from the plurality of kinematic segments according to the type of the third kinematic segment to be screened until the ith kinematic segment is screened from the plurality of kinematic segments, and stopping screening when i is a positive integer more than or equal to 4;
the screened ith kinematic segment and the total duration of all the kinematic segments with the serial numbers smaller than that of the ith kinematic segment are not smaller than the preset running duration; the state transition probability matrix includes a plurality of transition probabilities representing probabilities of the kinematic segment transitioning between the respective types.
4. The vehicle driving condition construction method according to claim 1, wherein the clustering the plurality of kinematic segments according to a first clustering algorithm to determine the types of the plurality of kinematic segments comprises:
extracting characteristic parameters of the plurality of kinematic segments, and performing principal component analysis on the characteristic parameters to obtain a characteristic parameter matrix;
and clustering the characteristic parameter matrix according to a first clustering algorithm, and determining the types of the plurality of kinematic segments.
5. The vehicle driving condition construction method according to claim 1, characterized in that the first clustering algorithm is a K-means clustering algorithm;
before clustering the plurality of kinematic segments according to a first clustering algorithm, determining the types of the plurality of kinematic segments, the method further comprises:
determining an initial clustering center of the K-means clustering algorithm according to a particle swarm algorithm;
adaptively adjusting the initial clustering center determined by the particle swarm algorithm according to a random operator and a cross operator;
and determining the optimal clustering number of the K-means clustering algorithm according to an elbow rule or a contour coefficient algorithm.
6. The vehicle running condition construction method according to claim 1, wherein after acquiring the running data of the plurality of vehicles in the target area, the method further comprises:
performing a data filling process on the travel data of a plurality of vehicles of the target area; wherein the data filling process is an interpolation filling process or a singular value filling process;
the singular value filling process comprises the following steps:
determining at least one similar vehicle of the target vehicle according to the driving data of the vehicles in the target area; wherein the target vehicle is a vehicle with running data missing; the correlation between the driving data of the similar vehicle and the driving data of the target vehicle is greater than a preset threshold value;
and filling the running data of the target vehicle according to the running data corresponding to the at least one similar vehicle and a singular value threshold algorithm.
7. The vehicle driving pattern construction method according to any one of claims 1 to 6, wherein after obtaining the final driving pattern of the vehicle in the target region, the method further comprises:
extracting a first characteristic parameter of the driving data of the vehicle in the target area;
extracting a second characteristic parameter of the final driving condition of the vehicle in the target area; the first characteristic parameter represents automobile performance characteristics corresponding to the running data of the vehicle; the second characteristic parameter represents the automobile performance characteristic corresponding to the final running working condition of the automobile;
verifying an error between the first characteristic parameter and the second characteristic parameter;
and if the error is larger than the preset error, skipping to the step of screening various kinematic segments according to the Markov model and the preset driving time length, and re-screening the plurality of kinematic segments.
8. A vehicle running condition constructing apparatus characterized by comprising:
the system comprises an acquisition module, a display module and a control module, wherein the acquisition module is used for acquiring the driving data of a plurality of vehicles in a target area;
the dividing module is used for dividing the driving data to obtain a plurality of kinematic segments;
the clustering module is used for clustering the plurality of kinematic segments according to a first clustering algorithm to obtain the type of each kinematic segment; the initial clustering center of the first clustering algorithm is obtained by optimizing a population optimization algorithm;
and the screening module is used for screening various kinematic segments according to the Markov model and the preset driving time length, and combining the screened kinematic segments to obtain the final driving working condition of the vehicle in the target area.
9. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the vehicle driving condition construction method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when executed by a processor, implements the steps of the vehicle behavior construction method as recited in any one of claims 1 to 7 above.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211504378.6A CN115601855B (en) | 2022-11-29 | 2022-11-29 | Vehicle driving condition construction method, electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211504378.6A CN115601855B (en) | 2022-11-29 | 2022-11-29 | Vehicle driving condition construction method, electronic device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115601855A true CN115601855A (en) | 2023-01-13 |
| CN115601855B CN115601855B (en) | 2023-04-07 |
Family
ID=84853058
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211504378.6A Active CN115601855B (en) | 2022-11-29 | 2022-11-29 | Vehicle driving condition construction method, electronic device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115601855B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108198425A (en) * | 2018-02-10 | 2018-06-22 | 长安大学 | A kind of construction method of Electric Vehicles Driving Cycle |
| CN108280484A (en) * | 2018-01-30 | 2018-07-13 | 辽宁工业大学 | A kind of driver's accelerating performance online classification and discrimination method |
| CN109932191A (en) * | 2019-03-15 | 2019-06-25 | 中国汽车技术研究中心有限公司 | A kind of real road driving cycle library and construction method |
| CN111310280A (en) * | 2020-02-27 | 2020-06-19 | 杭州电子科技大学 | Method for extracting automobile kinematic segments and synthesizing working condition diagram |
| CN112683543A (en) * | 2020-12-21 | 2021-04-20 | 国网上海市电力公司 | Gas-steam combined cycle unit phase-feed capacity cooperative test method and device |
| CN113032898A (en) * | 2021-03-10 | 2021-06-25 | 中国重汽集团济南动力有限公司 | Construction method of semi-trailer tractor working condition |
| CN113688558A (en) * | 2021-06-18 | 2021-11-23 | 长安大学 | Automobile driving condition construction method and system based on large database samples |
| CN114328660A (en) * | 2021-12-24 | 2022-04-12 | 联合汽车电子有限公司 | Screening method, computing device and storage medium for engine similar working conditions |
| CN114861833A (en) * | 2022-06-15 | 2022-08-05 | 山东大学 | Driving typical working condition construction method and system for improving clustering by grey wolf algorithm |
-
2022
- 2022-11-29 CN CN202211504378.6A patent/CN115601855B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108280484A (en) * | 2018-01-30 | 2018-07-13 | 辽宁工业大学 | A kind of driver's accelerating performance online classification and discrimination method |
| CN108198425A (en) * | 2018-02-10 | 2018-06-22 | 长安大学 | A kind of construction method of Electric Vehicles Driving Cycle |
| CN109932191A (en) * | 2019-03-15 | 2019-06-25 | 中国汽车技术研究中心有限公司 | A kind of real road driving cycle library and construction method |
| CN111310280A (en) * | 2020-02-27 | 2020-06-19 | 杭州电子科技大学 | Method for extracting automobile kinematic segments and synthesizing working condition diagram |
| CN112683543A (en) * | 2020-12-21 | 2021-04-20 | 国网上海市电力公司 | Gas-steam combined cycle unit phase-feed capacity cooperative test method and device |
| CN113032898A (en) * | 2021-03-10 | 2021-06-25 | 中国重汽集团济南动力有限公司 | Construction method of semi-trailer tractor working condition |
| CN113688558A (en) * | 2021-06-18 | 2021-11-23 | 长安大学 | Automobile driving condition construction method and system based on large database samples |
| CN114328660A (en) * | 2021-12-24 | 2022-04-12 | 联合汽车电子有限公司 | Screening method, computing device and storage medium for engine similar working conditions |
| CN114861833A (en) * | 2022-06-15 | 2022-08-05 | 山东大学 | Driving typical working condition construction method and system for improving clustering by grey wolf algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 陈宝，黄春，谢光毅，付江华，黄泽好: "基于大样本的电动汽车行驶工况构建方法研究", 《重庆理工大学学报（自然科学）》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115601855B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110097755B (en) | Highway traffic flow state identification method based on deep neural network | |
| CN111539454B (en) | Vehicle track clustering method and system based on meta-learning | |
| CN106095963B (en) | Vehicle driving behavior analysis big data public service platform under internet + era | |
| Förster et al. | Data-driven identification of characteristic real-driving cycles based on k-means clustering and mixed-integer optimization | |
| CN110414584B (en) | Automobile road spectrum clustering synthesis method and system | |
| CN104464304A (en) | Urban road vehicle running speed forecasting method based on road network characteristics | |
| CN112037539B (en) | Method and system for recommending signal control scheme for saturated urban traffic network | |
| CN114299742B (en) | Speed limit information dynamic identification and update recommendation method for expressway | |
| CN110910641A (en) | Neural network-based real-time risk assessment method for highway traffic operation | |
| CN113362600B (en) | Traffic state estimation method and system | |
| CN115092165A (en) | Clustering model-based driver style identification method under different cycle conditions | |
| CN115145900A (en) | Big data cleaning method for power battery of electric vehicle | |
| CN112435356A (en) | ETC interference signal identification method and detection system | |
| Taylor et al. | Road type classification through data mining | |
| CN113255620B (en) | Abnormal event detection method based on vehicle group track analysis in video | |
| CN115601855B (en) | Vehicle driving condition construction method, electronic device and storage medium | |
| CN113657432A (en) | Commercial vehicle driving behavior risk level identification method based on Internet of vehicles data | |
| CN114639238B (en) | Urban expressway traffic state estimation method based on normalized spectral clustering algorithm | |
| CN113420706B (en) | Vehicle detection method based on multi-layer feature fusion | |
| CN113947904A (en) | Multi-scale short-term traffic flow prediction method based on S-G filtering and deep belief network | |
| CN115271001B (en) | Vehicle driving condition identification method and device, vehicle and storage medium | |
| CN116884225B (en) | Method and device for evaluating influence of proportion of automatic driving vehicles on traffic efficiency | |
| CN113380032B (en) | Hierarchical clustering method-based highway congestion judgment method and device | |
| CN116913097B (en) | Traffic state prediction method and system | |
| CN112883020B (en) | Big data application-based analysis and management system |
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 |