CN110738754B - Vehicle kinematic segment extraction method, vehicle working condition analysis method and corresponding device - Google Patents

Vehicle kinematic segment extraction method, vehicle working condition analysis method and corresponding device Download PDF

Info

Publication number
CN110738754B
CN110738754B CN201911006306.7A CN201911006306A CN110738754B CN 110738754 B CN110738754 B CN 110738754B CN 201911006306 A CN201911006306 A CN 201911006306A CN 110738754 B CN110738754 B CN 110738754B
Authority
CN
China
Prior art keywords
segment
vehicle
complete
kinematic
parameters
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
Application number
CN201911006306.7A
Other languages
Chinese (zh)
Other versions
CN110738754A (en
Inventor
杨世春
张正杰
刘健
冯松
陈飞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Original Assignee
Beihang University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beihang University filed Critical Beihang University
Publication of CN110738754A publication Critical patent/CN110738754A/en
Application granted granted Critical
Publication of CN110738754B publication Critical patent/CN110738754B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06N3/02—Neural networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (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)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)

Abstract

The invention discloses a vehicle kinematics segment extraction method, which comprises the steps of firstly, obtaining parameters capable of describing vehicle kinematics characteristics in a complete kinematics segment of a vehicle, secondly, carrying out principal component analysis on the obtained parameters to obtain parameters capable of describing macroscopic characteristics of the complete kinematics segment, and finally, calculating to obtain parameters capable of describing transient characteristics of the complete kinematics segment; on the basis that the parameter obtained by principal component analysis mainly focuses on the macroscopic description of the kinematics segment, two parameters of an acceleration and deceleration mileage ratio and a segment power spectrum density are introduced as transient characteristic descriptions, the acceleration and deceleration mileage ratio represents the ratio of the mileage of a vehicle accelerated and decelerated running in one kinematics segment to the total mileage of the segment, namely the occurrence probability in the acceleration and deceleration process, the segment power spectrum density is the power in unit frequency, the intensity of vehicle acceleration and deceleration in one kinematics segment can be represented, and the method can realize accurate and effective extraction.

Description

Vehicle kinematic segment extraction method, vehicle working condition analysis method and corresponding device
Technical Field
The invention relates to the technical field of vehicle working condition establishment and analysis, in particular to a vehicle kinematic fragment extraction method, a vehicle working condition analysis method and a corresponding device.
Background
The effective establishment and analysis of the vehicle kinematics segment have important significance for scientific management and control and reasonable prediction of vehicle energy consumption in the development process at the present stage, and can also provide a solution for solving the problem of mileage anxiety of the new energy vehicle. In the future, when the automobile industry enters the era of intelligent networking, the accurate kinematics segment extraction method can provide a solid foundation for path planning and decision schemes.
The traditional vehicle kinematics segment extraction method is limited to the simple extraction of information such as speed, acceleration, mileage, time and the like of a vehicle in the past period of time, on one hand, the extracted parameters may not truly and perfectly reflect the driving characteristics of the vehicle in the past period of time, and the concept of the "period of time" is not the analysis of numerical characteristics but is a general expression; on the other hand, the extracted information is only the simple expansion of the database, and the automatic acquisition and analysis of the data cannot be realized after the data volume is increased, so that the method has great limitation. In addition, in the prior art, a kinematic segment is often defined by taking fixed time intervals (for example, 10 minutes of data extracted as a segment), and the method cuts off a complete driving cycle, which may cause incomplete information and lack of relevance. In addition, for the parameters extracted from the kinematic segment, how to select proper characteristic parameters for analysis is also a problem to be solved urgently, because if too many characteristic parameters cause large calculation amount, and the describing capability of the parameters is different, the calculation amount is easy to waste; if the characteristic parameters are too few, the actual kinematic characteristics cannot be vividly depicted, and incomplete expression causes data shortage on subsequent energy consumption analysis.
Therefore, how to provide a vehicle kinematic segment extraction method capable of realizing accurate and effective extraction becomes a technical problem which needs to be solved urgently by the technical personnel in the field.
Disclosure of Invention
The invention aims to provide a vehicle kinematic fragment extraction method, a vehicle working condition analysis method and a corresponding device.
The invention provides a vehicle kinematics segment extraction method, which comprises the following steps:
acquiring parameters capable of describing vehicle kinematic characteristics in a complete kinematic segment of a vehicle, wherein the parameters at least comprise: driving distance s and average driving speed vavgMaximum traveling speed vmaxLength of operation tallAnd acceleration alpha, wherein the complete kinematics segment is an interval segment with zero vehicle speed of two adjacent times;
and performing principal component analysis on the obtained parameters to obtain parameters capable of describing the macroscopic features of the complete kinematic segment, wherein the parameters comprise the following steps: driving distance s and average driving speed vavgMaximum traveling speed vmaxAnd a running time period tall；
Calculating parameters capable of describing transient characteristics of the complete kinematic segment: the acceleration-deceleration mileage ratio W and the segment power spectral density S are as follows:
Figure BDA0002242877340000021
Figure BDA0002242877340000022
wherein N is the number of deceleration times in the complete kinematic segment, TiThe duration of the ith instant acceleration and deceleration process, s is the mileage of the complete kinematic segment, and alpha is TiAcceleration over a period of time-function, Ff(ω) is a function of acceleration obtained by Fourier transform, and f represents frequency.
Preferably, the method for obtaining parameters that can describe the kinematic characteristics of the vehicle in a complete kinematic segment of the vehicle comprises:
directly acquiring original parameters of a vehicle in the running process through a sensor, wherein the sensor at least comprises a wheel speed sensor, a timing sensor and an acceleration sensor;
and calculating the original parameters to obtain parameters capable of describing the vehicle kinematic characteristics.
Preferably, the parameters capable of describing the vehicle kinematic characteristics in the acquired complete kinematic segment of the vehicle further include: idle time tillAcceleration time taccTime of deceleration tdecMaximum acceleration amaxMinimum acceleration aminAverage acceleration a of deceleration sectionavgAnd the average acceleration a of the acceleration sectionavg+。
The method for extracting the vehicle kinematic segment provided by the invention has the following technical effects:
on the basis that the parameter obtained by principal component analysis mainly focuses on the macroscopic description of the kinematics segment, two parameters of an acceleration and deceleration mileage ratio and a segment power spectrum density are introduced as transient characteristic descriptions, the acceleration and deceleration mileage ratio represents the ratio of the mileage of a vehicle accelerated and decelerated running in one kinematics segment to the total mileage of the segment, namely the occurrence probability in the acceleration and deceleration process, the segment power spectrum density is the power in unit frequency, the intensity of vehicle acceleration and deceleration in one kinematics segment can be represented, and the method can realize accurate and effective extraction.
The invention also provides a vehicle working condition analysis method, which comprises the following steps:
by adopting the vehicle kinematic segment extraction method, a plurality of complete kinematic segments are respectively extracted, and the following six parameters which can describe the macro features and the transient features of the corresponding kinematic segments in each complete kinematic segment are extracted: driving distance s and average driving speed vavgMaximum traveling speed vmaxAnd a running time period tallAcceleration-deceleration mileage ratio W and segment power spectral density S;
selecting a typical complete kinematic segment from a plurality of complete kinematic segments, wherein the selection process comprises the following steps:
calculating the correlation coefficient eta XY of the complete kinematics segment according to the following formula, wherein the complete kinematics segment with the correlation coefficient more than or equal to 50 percent is a typical complete kinematics segment,
Figure BDA0002242877340000023
wherein, X is parameter data which can describe macro features and transient features of each complete kinematic segment, Y is average parameter data of the complete kinematic segment as a whole, Cov (X, Y) is covariance of X and Y, and D (X) and D (Y) are variances of X and Y respectively;
the method comprises the following steps of training parameters which can describe macro features and transient features of corresponding kinematic segments by adopting typical complete kinematic segments through a neural network, and establishing identification from the parameters to vehicle appearance duration, driving scenes and driving styles, wherein:
based on the length of time t of movementallAnd a given duration t1、t2Wherein, t1＜t2Duration of motion t of a complete kinematic segmentall≤t1Then, it is the movement duration t of the complete kinematics segment for the short tripall＞t1And t isall≤t2If the motion duration is the middle-time trip, the motion duration t of the complete kinematics segmentall＞t2Then long trip, based on the average traveling speed vavgAnd a given velocity v1、v2Wherein v is1＜v2，vavg≤v1If the driving scene is congested, vavg＞v1And v isavg≤v2If the driving scene is normal, vavg＞v2The driving scene is relaxed and is based on the acceleration-deceleration mileage ratio W, the segment power spectral density S and the given W1、W2、S1、S2Wherein W is1＜W2、S1＜S2，W≤W1And S is less than or equal to S1The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S is less than or equal to S1The driving style of the driver is general, W is more than W2And S is less than or equal to S1The driving style of the driver is dynamic, W is less than or equal to W1And S is less than or equal to S2And S > S1Then, thenThe driving style of the driver is economical, and W is less than or equal to W2And W > W1And S is less than or equal to S2And S > S1The driving style of the driver is general, W is more than W2And S is less than or equal to S2And S > S1The driving style of the driver is dynamic, W is less than or equal to W1And S > S2The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S > S2The driving style of the driver is dynamic, W is more than W2And S > S2If the driver is in the power type, the driving style of the driver is changed into the power type;
and acquiring parameters capable of describing vehicle kinematic characteristics in real time, and inputting the parameters into the trained neural network to analyze the vehicle working condition.
The invention is also based on the vehicle working condition analysis method of the extraction method, so the same technical effects are achieved, further, a mapping relation of ' characteristic parameters ' -vehicle-road-person ' information is established through a neural network algorithm, the reliable identification of the time length from the establishment of the parameters to the occurrence of the vehicle, the driving scene and the driving style is completed, the parameters collected in real time are analyzed, and therefore, the identified vehicle travel time length, the scene where the vehicle is located and the driving style information can be provided for a decision control system of the vehicle to use.
Preferably, the neural network is an LVQ neural network.
Preferably, t is1Is 15min, t2Is 1h, v1Is 2m/s, v2Is 5 m/s.
The present invention also provides a vehicle kinematics segment extraction system, comprising:
a parameter acquisition module that acquires parameters of a complete kinematic segment of a vehicle that can describe kinematic characteristics of the vehicle, the parameters including at least: driving distance s and average driving speed vavgMaximum traveling speed vmaxLength of operation tallAnd acceleration alpha, wherein the complete kinematics segment is an interval segment with zero vehicle speed of two adjacent times;
a principal component analysis module for performing principal component analysis on the obtained parameters to obtain a macro capable of analyzing the complete kinematic segmentParameters described by the visual characteristics are as follows: driving distance s and average driving speed vavgMaximum traveling speed vmaxAnd a running time period tall；
The transient characteristic parameter calculation module is used for calculating parameters capable of describing transient characteristics of the complete kinematic segment: the acceleration-deceleration mileage ratio W and the segment power spectral density S are as follows:
Figure BDA0002242877340000031
Figure BDA0002242877340000032
wherein N is the number of deceleration times in the complete kinematic segment, TiThe duration of an instantaneous acceleration/deceleration process, s the mileage of the complete kinematic segment, and α TiAcceleration over a period of time-function, Ff(ω) is a function of acceleration obtained by Fourier transform, and f represents frequency.
Preferably, the parameter obtaining module includes:
the sensor submodule at least comprises a wheel speed sensor, a timing sensor and an acceleration sensor and is used for acquiring original parameters in the running process of the vehicle;
and the calculation submodule is used for calculating the original parameters to obtain parameters capable of describing the vehicle kinematic characteristics.
The invention also provides a vehicle working condition analysis system, which comprises the vehicle kinematic segment extraction system and further comprises:
the typical complete kinematics segment acquisition module selects a typical complete kinematics segment from a plurality of complete kinematics segments, and the selection process is as follows:
calculating the correlation coefficient eta of the complete kinematic segment according to the following formulaXYThe complete kinematics segment with the correlation coefficient more than or equal to 50 percent is a typical complete kinematics segment,
Figure BDA0002242877340000041
wherein, X is parameter data which can describe macro features and transient features of each complete kinematic segment, Y is average parameter data of the complete kinematic segment as a whole, Cov (X, Y) is covariance of X and Y, and D (X) and D (Y) are variances of X and Y respectively;
the neural network training module is used for carrying out neural network training on parameters which are described by macro features and transient features of corresponding kinematics segments and are typical complete kinematics segments, and establishing the recognition of the parameters to the occurrence time of the vehicle, the driving scene and the driving style, wherein:
based on the length of time t of movementallAnd a given duration t1、t2Wherein, t1＜t2Duration of motion t of a complete kinematic segmentall≤t1Then, it is the movement duration t of the complete kinematics segment for the short tripall＞t1And t isall≤t2If the motion duration is the middle-time trip, the motion duration t of the complete kinematics segmentall＞t2Then long trip, based on the average traveling speed vavgAnd a given velocity v1、v2Wherein v is1＜v2，vavg≤v1If the driving scene is congested, vavg＞v1And v isavg≤v2If the driving scene is normal, vavg＞v2The driving scene is relaxed and is based on the acceleration-deceleration mileage ratio W, the segment power spectral density S and the given W1、W2、S1、S2Wherein W is1＜W2、S1＜S2，W≤W1And S is less than or equal to S1The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S is less than or equal to S1The driving style of the driver is general, W is more than W2And S is less than or equal to S1The driving style of the driver is dynamic, W is less than or equal to W1And S is less than or equal to S2And S > S1The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S is less than or equal to S2And S > S1The driving style of the driver is general, W is more than W2And S is less than or equal to S2And S > S1The driving style of the driver is dynamic, W is less than or equal to W1And S > S2The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S > S2The driving style of the driver is dynamic, W is more than W2And S > S2If the driver is in the power type, the driving style of the driver is changed into the power type;
and the analysis module acquires parameters capable of describing vehicle kinematic characteristics in real time and inputs the parameters into the trained neural network to analyze the vehicle working condition.
The present invention also provides an electronic device comprising:
a memory for storing a computer program;
a processor for implementing the steps of the vehicle kinematics segment extraction method or the vehicle behavior analysis method as described above when executing the computer program.
The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program, when executed by a processor, implements the steps of the vehicle kinematics segment extraction method or the vehicle condition analysis method described above.
The invention provides a vehicle kinematic segment extraction system, a vehicle working condition analysis system, an electronic device and a computer readable storage medium, which all have the technical effects.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flowchart of an embodiment of a vehicle kinematics segment extraction method according to the present invention;
FIG. 2 is a flow chart of an embodiment of a vehicle condition analysis method provided by the present invention;
FIG. 3 is a schematic structural diagram of an embodiment of a vehicle kinematic fragment extraction system provided by the present invention;
fig. 4 is a schematic structural diagram of an embodiment of a vehicle condition analysis system provided in the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-4, fig. 1 is a flowchart illustrating an embodiment of a vehicle kinematic segment extraction method according to the present invention; FIG. 2 is a flow chart of an embodiment of a vehicle condition analysis method provided by the present invention; FIG. 3 is a schematic structural diagram of an embodiment of a vehicle kinematic fragment extraction system provided by the present invention; fig. 4 is a schematic structural diagram of an embodiment of a vehicle condition analysis system provided in the present invention.
With reference to fig. 1, the present invention discloses a method for extracting a vehicle kinematic segment, in one embodiment, the method comprises the following steps:
firstly, parameters capable of describing vehicle kinematic characteristics in a complete kinematic segment of a vehicle are obtained, and the parameters at least comprise: driving distance s and average driving speed vavgMaximum traveling speed vmaxLength of operation tallAnd acceleration alpha, wherein the complete kinematics segment is an interval segment with zero vehicle speed of two adjacent times;
in this embodiment, how to acquire the parameters that can describe the vehicle kinematic features in the complete kinematic segment of the vehicle is not limited, and for example, sensors may be provided to acquire some of the corresponding parameters.
In the invention, the complete kinematics segment is an interval segment with zero vehicle speed in two adjacent times, compared with the prior art, the complete kinematics segment can completely comprise working conditions of acceleration, deceleration, uniform speed and the like, and can better describe the motion characteristics.
Secondly, performing principal component analysis on the obtained parameters to obtain parameters capable of describing the macroscopic features of the complete kinematic segment, wherein the parameters are as follows: driving distance s and average driving speed vavgMaximum traveling speed vmaxAnd a running time period tall；
Principal component analysis is a common analysis method, which has been widely used in various technical fields, such as biotechnology. Mention is made in the prior art of: too many characteristic parameters cause large calculation amount, and the describing capability of the parameters is different, so that the calculation amount is easy to waste; if the characteristic parameters are too few, the actual kinematic characteristics cannot be vividly depicted, and incomplete expression causes data shortage on subsequent energy consumption analysis. In order to eliminate the correlation among the variables and finally obtain a plurality of comprehensive variables to reflect most of the information of the researched problem, the invention adopts a principal component analysis method to summarize and reflect the essence of things and simplify the subsequent calculation. Specifically, principal component analysis is performed on 30000 pieces of data and 12 characteristic parameters of single data to obtain the contribution rate of each parameter, and finally the driving mileage s and the average driving speed v are selectedavgMaximum traveling speed vmaxAnd a running time period tallThese 4 characteristic parameters serve as parameters for principal component analysis.
And finally, calculating parameters capable of describing transient characteristics of the complete kinematic segment: the acceleration-deceleration mileage ratio W and the segment power spectral density S are as follows:
Figure BDA0002242877340000061
Figure BDA0002242877340000062
wherein N is the number of deceleration times in the complete kinematic segment, TiThe duration of the ith instant acceleration and deceleration process, s is the mileage of the complete kinematic segment, and alpha is TiAcceleration in the time period is a function of time, Ff (omega) is a function obtained by Fourier transformation of the acceleration, and f represents frequency.
In the invention, the parameter obtained by principal component analysis mainly focuses on the macroscopic description of a kinematic segment, on the basis, two parameters of an acceleration and deceleration mileage ratio and a segment power spectrum density are introduced as transient characteristic description, the acceleration and deceleration mileage ratio represents the ratio of the mileage of a vehicle accelerated and decelerated in a kinematic segment to the total mileage of the segment, namely the occurrence probability of an acceleration and deceleration process, the segment power spectrum density is the power in unit frequency, the intensity of acceleration and deceleration of the vehicle in the kinematic segment can be represented, and the method can realize accurate and effective extraction.
In another embodiment, a method for obtaining parameters describing vehicle kinematic characteristics in a complete kinematic segment of a vehicle includes:
directly acquiring original parameters of a vehicle in the running process through a sensor, wherein the sensor at least comprises a wheel speed sensor, a timing sensor and an acceleration sensor;
and calculating the original parameters to obtain parameters capable of describing the vehicle kinematic characteristics.
Specifically, the wheel speed sensor can directly acquire the running speed v, and the timing sensor can directly acquire the running time tallThe acceleration sensor can directly acquire the acceleration. Further, the driving mileage is obtained according to the following formula
Figure BDA0002242877340000063
The average traveling speed can be calculated as follows
Figure BDA0002242877340000064
For maximum driving speed vmaxAnd the signal can be directly acquired by a wheel speed sensor.
Obviously, the method is not limited to this, and for example, a GPS sensor may be provided to directly collect the mileage.
Further, the parameters capable of describing the vehicle kinematic characteristics in the obtained complete kinematic segment of the vehicle further include: idle time tillAcceleration time taccTime of deceleration tdecMaximum acceleration amaxMinimum acceleration aminAverage acceleration a of deceleration sectionavgAnd the average acceleration a of the acceleration sectionavg+。
With reference to fig. 2, the present invention further provides a method for analyzing a vehicle condition, which in an embodiment includes:
firstly, the vehicle kinematic segment extraction method is adopted to respectively extract a plurality of complete kinematic segments, and the following six parameters which can describe the macro features and the transient features of the corresponding kinematic segments in each complete kinematic segment are extracted: driving distance s and average driving speed vavgMaximum traveling speed vmaxAnd a running time period tallAcceleration-deceleration mileage ratio W and segment power spectral density S;
secondly, selecting a typical complete kinematic segment from a plurality of complete kinematic segments, wherein the selection process is as follows:
calculating the correlation coefficient eta of the complete kinematic segment according to the following formulaXYThe complete kinematics segment with the correlation coefficient more than or equal to 50 percent is a typical complete kinematics segment,
Figure BDA0002242877340000065
wherein, X is parameter data which can describe macro features and transient features of each complete kinematic segment, Y is average parameter data of the complete kinematic segment as a whole, Cov (X, Y) is covariance of X and Y, and D (X) and D (Y) are variances of X and Y respectively;
furthermore, the parameters of the typical complete kinematics segment, which can describe the macro features and the transient features of the corresponding kinematics segment, are adopted to carry out neural network training, and the identification from the parameters to the occurrence time of the vehicle, the driving scene and the driving style is established, wherein:
based on the length of time t of movementallAnd a given duration t1、t2Wherein, t1＜t2Duration of motion t of a complete kinematic segmentall≤t1Then, it is the movement duration t of the complete kinematics segment for the short tripall＞t1And t isall≤t2If the motion duration is the middle-time trip, the motion duration t of the complete kinematics segmentall＞t2Then long trip, based on the average traveling speed vavgAnd a given velocity v1、v2Wherein v is1＜v2，vavg≤v1If the driving scene is congested, vavg＞v1And v isavg≤v2If the driving scene is normal, vavg＞v2The driving scene is relaxed and is based on the acceleration-deceleration mileage ratio W, the segment power spectral density S and the given W1、W2、S1、S2Wherein W is1＜W2、S1＜S2，W≤W1And S is less than or equal to S1The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S is less than or equal to S1The driving style of the driver is general, W is more than W2And S is less than or equal to S1The driving style of the driver is dynamic, W is less than or equal to W1And S is less than or equal to S2And S > S1The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S is less than or equal to S2And S > S1The driving style of the driver is general, W is more than W2And S is less than or equal to S2And S > S1The driving style of the driver is dynamic, W is less than or equal to W1And S > S2The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S > S2The driving style of the driver is dynamic, W is more than W2And S > S2If the driver is in the power type, the driving style of the driver is changed into the power type;
and finally, acquiring parameters capable of describing vehicle kinematic characteristics in real time, and inputting the parameters into the trained neural network to analyze the vehicle working condition.
Preferably, the neural network is an LVQ neural network.
Specifically, the LVQ neural network is divided into an offline training part and an online recognition part. The online identification part is used for acquiring parameters capable of describing vehicle kinematic characteristics in real time and inputting the parameters into the trained neural network to analyze the vehicle working condition. The off-line training part extracts 6 kinematic characteristic parameters by using the selected typical kinematic segment data, and then trains the LVQ neural network by using the parameters; the online identification part extracts characteristic parameters of vehicle running data collected in real time, the characteristic parameters are used as an input layer of the LVQ neural network identification model, the working condition type of the segment is further output, real-time identification and updating are carried out, and an identification period can be selected according to the real-time data processing capacity and the length of the collected segment. The trained LVQ neural network can realize on-line recognition of vehicle running data, can judge the vehicle travel time, the scene where the vehicle is located and the driving style information, and can be used by a decision control system of the vehicle.
The vehicle working condition analysis method based on the extraction method has the same technical effects, further, a mapping relation of ' characteristic parameters ' -vehicle-road-person ' information is established through a neural network algorithm, reliable identification of the time length from the establishment of the parameters to the occurrence time of the vehicle, the driving scene and the driving style is completed, the parameters collected in real time are analyzed, and therefore the identified travel time length of the vehicle, the scene and the driving style information can be provided for a decision control system of the vehicle to use.
In another embodiment, t1Is 15min, t2Is 1h, v1Is 2m/s, v2Is 5 m/s. The value is not unique and can be selected according to specific conditions, for example, congestion conditions among cities are different, and the difference of the values is further reflected.
With reference to fig. 3, the present invention provides a vehicle kinematic segment extraction system, including:
parameter acquisition module for acquiring a vehicle describable in a complete kinematic section of a vehicleParameters of the kinematic features, said parameters comprising at least: driving distance s and average driving speed vavgMaximum traveling speed vmaxLength of operation tallAnd acceleration alpha, wherein the complete kinematics segment is an interval segment with zero vehicle speed of two adjacent times;
the principal component analysis module is used for carrying out principal component analysis on the obtained parameters to obtain parameters capable of describing the macroscopic features of the complete kinematic segment, and comprises the following steps: driving distance s and average driving speed vavgMaximum traveling speed vmaxAnd a running time period tall；
The transient characteristic parameter calculation module is used for calculating parameters capable of describing transient characteristics of the complete kinematic segment: the acceleration-deceleration mileage ratio W and the segment power spectral density S are as follows:
Figure BDA0002242877340000081
Figure BDA0002242877340000082
wherein N is the number of deceleration times in the complete kinematic segment, TiThe duration of an instantaneous acceleration/deceleration process, s the mileage of the complete kinematic segment, and α TiAcceleration over a period of time-function, Ff(ω) is a function of acceleration obtained by Fourier transform, and f represents frequency.
Further, the parameter obtaining module comprises:
the sensor submodule at least comprises a wheel speed sensor, a timing sensor and an acceleration sensor and is used for acquiring original parameters in the running process of the vehicle;
and the calculation submodule is used for calculating the original parameters to obtain parameters capable of describing the vehicle kinematic characteristics.
Specifically, the wheel speed sensor can directly acquire the running speed v and the timing sensorThe running time t can be directly acquiredallThe acceleration sensor can directly acquire the acceleration. Further, the driving mileage is obtained according to the following formula
Figure BDA0002242877340000083
The average traveling speed can be calculated as follows
Figure BDA0002242877340000084
For maximum driving speed vmaxAnd the signal can be directly acquired by a wheel speed sensor.
With reference to fig. 4, the present invention further provides a vehicle condition analysis system, including the vehicle kinematic segment extraction system, further including:
the typical complete kinematics segment acquisition module selects a typical complete kinematics segment from a plurality of complete kinematics segments, and the selection process is as follows:
calculating the correlation coefficient eta of the complete kinematic segment according to the following formulaXYThe complete kinematics segment with the correlation coefficient more than or equal to 50 percent is a typical complete kinematics segment,
Figure BDA0002242877340000085
wherein, X is parameter data which can describe macro features and transient features of each complete kinematic segment, Y is average parameter data of the complete kinematic segment as a whole, Cov (X, Y) is covariance of X and Y, and D (X) and D (Y) are variances of X and Y respectively;
the neural network training module is used for carrying out neural network training on parameters which are described by macro features and transient features of corresponding kinematics segments and are typical complete kinematics segments, and establishing the recognition of the parameters to the occurrence time of the vehicle, the driving scene and the driving style, wherein:
based on the length of time t of movementallAnd a given duration t1、t2Wherein, t1＜t2Duration of motion t of a complete kinematic segmentall≤t1Then, it is the movement duration t of the complete kinematics segment for the short tripall＞t1And t isall≤t2If the motion duration is the middle-time trip, the motion duration t of the complete kinematics segmentall＞t2Then long trip, based on the average traveling speed vavgAnd a given velocity v1、v2Wherein v is1＜v2，vavg≤v1If the driving scene is congested, vavg＞v1And v isavg≤v2If the driving scene is normal, vavg＞v2The driving scene is relaxed and is based on the acceleration-deceleration mileage ratio W, the segment power spectral density S and the given W1、W2、S1、S2Wherein W is1＜W2、S1＜S2，W≤W1And S is less than or equal to S1The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S is less than or equal to S1The driving style of the driver is general, W is more than W2And S is less than or equal to S1The driving style of the driver is dynamic, W is less than or equal to W1And S is less than or equal to S2And S is more than S1, the driving style of the driver is economical, W is less than or equal to W2And W > W1And S is less than or equal to S2And S > S1The driving style of the driver is general, W is more than W2And S is less than or equal to S2And S > S1The driving style of the driver is dynamic, W is less than or equal to W1And S > S2The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S > S2The driving style of the driver is dynamic, W is more than W2And S > S2If the driver is in the power type, the driving style of the driver is changed into the power type;
and the analysis module acquires parameters capable of describing vehicle kinematic characteristics in real time and inputs the parameters into the trained neural network to analyze the vehicle working condition.
The present invention also provides an electronic device comprising:
a memory for storing a computer program;
a processor for implementing the steps of the vehicle kinematics segment extraction method or the vehicle behavior analysis method as described above when executing the computer program.
Further, the present invention also provides a computer readable storage medium, having a computer program stored thereon, where the computer program, when executed by a processor, implements the steps of the vehicle kinematics segment extraction method or the vehicle behavior analysis method as described above.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (10)

1. A vehicle kinematic segment extraction method is characterized by comprising the following steps:
acquiring parameters capable of describing vehicle kinematic characteristics in a complete kinematic segment of a vehicle, wherein the parameters at least comprise: driving distance s and average driving speed vavgMaximum traveling speed vmaxLength of operation tallAnd acceleration alpha, wherein the complete kinematics segment is an interval segment with zero vehicle speed of two adjacent times;
and performing principal component analysis on the obtained parameters to obtain parameters capable of describing the macroscopic features of the complete kinematic segment, wherein the parameters comprise the following steps: driving distance s and average driving speed vavgMaximum traveling speed vmaxAnd a running time period tall；
Calculating parameters capable of describing transient characteristics of the complete kinematic segment: the acceleration-deceleration mileage ratio W and the segment power spectral density S are as follows:
Figure FDA0002724800390000011
Figure FDA0002724800390000012
wherein N is the number of deceleration times in the complete kinematic segment, TiThe duration of the ith instant acceleration and deceleration process, s is the mileage of the complete kinematic segment, and alpha is TiAcceleration over a period of time-function, Ff(ω) is a function of acceleration obtained by Fourier transform, and f represents frequency.
2. The vehicle kinematic segment extraction method according to claim 1, characterized in that the method of obtaining parameters that can describe vehicle kinematic features in a complete kinematic segment of a vehicle comprises:
directly acquiring original parameters of a vehicle in the running process through a sensor, wherein the sensor at least comprises a wheel speed sensor, a timing sensor and an acceleration sensor;
and calculating the original parameters to obtain parameters capable of describing the vehicle kinematic characteristics.
3. The vehicle kinematic segment extraction method according to claim 1, wherein the parameters that can describe the vehicle kinematic features in the acquired complete kinematic segment of the vehicle further comprise: idle time tillAcceleration time taccTime of deceleration tdecMaximum acceleration amaxMinimum acceleration aminAverage acceleration a of deceleration sectionavgAnd the average acceleration a of the acceleration sectionavg+。
4. A method of analyzing a vehicle behavior, comprising:
the vehicle kinematic segment extraction method according to any one of claims 1 to 3, wherein a plurality of complete kinematic segments are extracted respectively, and the following six parameters which can describe macro features and transient features of the corresponding kinematic segment in each complete kinematic segment are extracted: the running mileage s,Average running speed vavgMaximum traveling speed vmaxAnd a running time period tallAcceleration-deceleration mileage ratio W and segment power spectral density S;
selecting a typical complete kinematic segment from a plurality of complete kinematic segments, wherein the selection process comprises the following steps:
calculating the correlation coefficient eta of the complete kinematic segment according to the following formulaXYThe complete kinematics segment with the correlation coefficient more than or equal to 50 percent is a typical complete kinematics segment,
Figure FDA0002724800390000013
wherein, X is parameter data which can describe macro features and transient features of each complete kinematic segment, Y is average parameter data of the complete kinematic segment as a whole, Cov (X, Y) is covariance of X and Y, and D (X) and D (Y) are variances of X and Y respectively;
the method comprises the following steps of training parameters which can describe macro features and transient features of corresponding kinematic segments by adopting typical complete kinematic segments through a neural network, and establishing identification from the parameters to vehicle appearance duration, driving scenes and driving styles, wherein:
based on the length of time t of operationallAnd a given duration t1、t2Wherein, t1＜t2Duration of operation t of a complete kinematic segmentall≤t1The running time t of the complete kinematics segment is a short tripall＞t1And t isall≤t2And the running time is the running time t of the complete kinematics segment for the middle-time tripall＞t2Then long trip, based on the average traveling speed vavgAnd a given velocity v1、v2Wherein v is1＜v2，vavg≤v1If the driving scene is congested, vavg＞v1And v isavg≤v2If the driving scene is normal, vavg＞v2If the driving scene is loose, based on the acceleration-deceleration mileage ratio W and the segment power spectrumDensity S and given W1、W2、S1、S2Wherein W is1＜W2、S1＜S2，W≤W1And S is less than or equal to S1The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S is less than or equal to S1The driving style of the driver is general, W is more than W2And S is less than or equal to S1The driving style of the driver is dynamic, W is less than or equal to W1And S is less than or equal to S2And S > S1The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S is less than or equal to S2And S > S1The driving style of the driver is general, W is more than W2And S is less than or equal to S2And S > S1The driving style of the driver is dynamic, W is less than or equal to W1And S > S2The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S > S2The driving style of the driver is dynamic, W is more than W2And S > S2If the driver is in the power type, the driving style of the driver is changed into the power type;
and acquiring parameters capable of describing vehicle kinematic characteristics in real time, and inputting the parameters into the trained neural network to analyze the vehicle working condition.
5. The vehicle condition analysis method according to claim 4, wherein the neural network employs an LVQ neural network.
6. The vehicle condition analysis method according to claim 4, characterized in that t1Is 15min, t2Is 1h, v1Is 2m/s, v2Is 5 m/s.
7. A vehicle kinematics segment extraction system, comprising:
a parameter acquisition module that acquires parameters of a complete kinematic segment of a vehicle that can describe kinematic characteristics of the vehicle, the parameters including at least: driving distance s and average driving speed vavgMaximum traveling speed vmaxLength of operation tallAnd the acceleration a is set to be,the complete kinematic segment is an interval segment with zero vehicle speed of two adjacent times;
the principal component analysis module is used for carrying out principal component analysis on the obtained parameters to obtain parameters capable of describing the macroscopic features of the complete kinematic segment, and comprises the following steps: driving distance s and average driving speed vavgMaximum traveling speed vmaxAnd a running time period tall；
The transient characteristic parameter calculation module is used for calculating parameters capable of describing transient characteristics of the complete kinematic segment: the acceleration-deceleration mileage ratio W and the segment power spectral density S are as follows:
Figure FDA0002724800390000021
Figure FDA0002724800390000022
wherein N is the number of deceleration times in the complete kinematic segment, TiThe duration of an instantaneous acceleration/deceleration process, s the mileage of the complete kinematic segment, and α TiAcceleration over a period of time-function, Ff(ω) is a function of acceleration obtained by Fourier transform, and f represents frequency.
8. The vehicle kinematic fragment extraction system of claim 7, wherein the parameter acquisition module comprises:
the sensor submodule at least comprises a wheel speed sensor, a timing sensor and an acceleration sensor and is used for acquiring original parameters in the running process of the vehicle;
and the calculation submodule is used for calculating the original parameters to obtain parameters capable of describing the vehicle kinematic characteristics.
9. A vehicle behavior analysis system comprising the vehicle kinematic fragment extraction system of claim 7 or 8, further comprising:
the typical complete kinematics segment acquisition module selects a typical complete kinematics segment from a plurality of complete kinematics segments, and the selection process is as follows:
calculating the correlation coefficient eta of the complete kinematic segment according to the following formulaXYThe complete kinematics segment with the correlation coefficient more than or equal to 50 percent is a typical complete kinematics segment,
Figure FDA0002724800390000031
wherein, X is parameter data which can describe macro features and transient features of each complete kinematic segment, Y is average parameter data of the complete kinematic segment as a whole, Cov (X, Y) is covariance of X and Y, and D (X) and D (Y) are variances of X and Y respectively;
the neural network training module is used for carrying out neural network training on parameters which are described by macro features and transient features of corresponding kinematics segments and are typical complete kinematics segments, and establishing the recognition of the parameters to the occurrence time of the vehicle, the driving scene and the driving style, wherein:
based on the length of time t of operationallAnd a given duration t1、t2Wherein, t1＜t2Duration of operation t of a complete kinematic segmentall≤t1The running time t of the complete kinematics segment is a short tripall＞t1And t isall≤t2And the running time is the running time t of the complete kinematics segment for the middle-time tripall＞t2Then long trip, based on the average traveling speed vavgAnd a given velocity v1、v2Wherein v is1＜v2，vavg≤v1If the driving scene is congested, vavg＞v1And v isavg≤v2If the driving scene is normal, vavg＞v2If the driving scene is loose, based on the acceleration-deceleration mileage ratio W and the segment power spectral densityS and given W1、W2、S1、S2Wherein W is1＜W2、S1＜S2，W≤W1And S is less than or equal to S1The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S is less than or equal to S1The driving style of the driver is general, W is more than W2And S is less than or equal to S1The driving style of the driver is dynamic, W is less than or equal to W1And S is less than or equal to S2And S > S1The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S is less than or equal to S2And S > S1The driving style of the driver is general, W is more than W2And S is less than or equal to S2And S > S1The driving style of the driver is dynamic, W is less than or equal to W1And S > S2The driving style of the driver is economic, and W is less than or equal to W2And W > W1And S > S2The driving style of the driver is dynamic, W is more than W2And S > S2If the driver is in the power type, the driving style of the driver is changed into the power type;
and the analysis module acquires parameters capable of describing vehicle kinematic characteristics in real time and inputs the parameters into the trained neural network to analyze the vehicle working condition.
10. An electronic device, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the method according to any one of claims 1 to 6 when executing said computer program.
CN201911006306.7A 2019-07-16 2019-10-22 Vehicle kinematic segment extraction method, vehicle working condition analysis method and corresponding device Active CN110738754B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910640222 2019-07-16
CN2019106402222 2019-07-16

Publications (2)

Publication Number Publication Date
CN110738754A CN110738754A (en) 2020-01-31
CN110738754B true CN110738754B (en) 2021-02-23

Family

ID=69270781

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911006306.7A Active CN110738754B (en) 2019-07-16 2019-10-22 Vehicle kinematic segment extraction method, vehicle working condition analysis method and corresponding device

Country Status (1)

Country Link
CN (1) CN110738754B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114332825B (en) * 2022-03-10 2022-06-17 中汽信息科技(天津)有限公司 Road terrain distribution identification method and device based on deep learning and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080093848A (en) * 2007-10-16 2008-10-22 (주)케이시엠 Device for improving fuel efficiency using engine rpm computing device of vehicle and method thereof
JP2017009415A (en) * 2015-06-22 2017-01-12 東電設計株式会社 Deformed-state monitoring method of road bridge floor slab
CN108647836A (en) * 2018-06-06 2018-10-12 哈尔滨思派科技有限公司 A kind of driver's Energy efficiency evaluation method and system
CN109278736A (en) * 2017-07-20 2019-01-29 丰田自动车株式会社 The damping control device of vehicle

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080093848A (en) * 2007-10-16 2008-10-22 (주)케이시엠 Device for improving fuel efficiency using engine rpm computing device of vehicle and method thereof
JP2017009415A (en) * 2015-06-22 2017-01-12 東電設計株式会社 Deformed-state monitoring method of road bridge floor slab
CN109278736A (en) * 2017-07-20 2019-01-29 丰田自动车株式会社 The damping control device of vehicle
CN108647836A (en) * 2018-06-06 2018-10-12 哈尔滨思派科技有限公司 A kind of driver's Energy efficiency evaluation method and system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
analysis of kinematics and dynamics of the vehicle drive system with a parallel power flow transmission;ZBIGNIEW PAWELSKI;《The Archives of Automotive Engineering》;20161231;第71卷(第1期);全文 *
大型营运客车用户关联试验场可靠性理论分析及试验研究;张禄;《中国博士学位论文全文数据库（电子期刊）》;20150831;全文 *

Also Published As

Publication number Publication date
CN110738754A (en) 2020-01-31

Similar Documents

Publication Publication Date Title
Silvas et al. Synthesis of realistic driving cycles with high accuracy and computational speed, including slope information
CN110377984A (en) A kind of industrial equipment remaining useful life prediction technique, system and electronic equipment
CN110610260A (en) Driving energy consumption prediction system, method, storage medium and equipment
CN107025274A (en) Equipment health status intelligent perception system and method based on Hadoop
Zylius Investigation of route-independent aggressive and safe driving features obtained from accelerometer signals
CN110705774A (en) Vehicle energy consumption analysis prediction method and system
CN108556682A (en) A kind of continual mileage prediction technique, device and equipment
Han et al. Short-term vehicle speed prediction based on convolutional bidirectional lstm networks
JP2000181526A (en) Device and method for estimating/predicting plant state
Fan et al. Predicting vacant parking space availability: a long short-term memory approach
CN113689594A (en) Novel automobile driving condition construction method
CN110738754B (en) Vehicle kinematic segment extraction method, vehicle working condition analysis method and corresponding device
EP3726434B1 (en) Method for reducing exhaust emissions of a vehicle drive system with combustion engine
CN106972959A (en) A kind of online service method and system of sensing algorithm
CN107423761B (en) Rail locomotive energy-saving optimization operation method based on feature selection and machine learning
CN107305626A (en) Pedestrian movement's track detection and analysis system violating the regulations
CN116244647A (en) Unmanned aerial vehicle cluster running state estimation method
CN104933471A (en) Phase space reconstruction method for use in short-term wind speed prediction method
Fafoutellis et al. Dilated LSTM networks for short-term traffic forecasting using network-wide vehicle trajectory data
CN106875681A (en) A kind of traffic congestion tendency Forecasting Methodology and system
CN109145175A (en) Space-time data prediction technique based on stacking Ensemble Learning Algorithms
CN113095479A (en) Method for extracting ice-below-layer structure based on multi-scale attention mechanism
Heydarian et al. Automated benchmarking and monitoring of an earthmoving operation's carbon footprint using video cameras and a greenhouse gas estimation model
Pan et al. Traffic speed prediction based on time classification in combination with spatial graph convolutional network
CN108520311B (en) Haze prediction model establishing method and system combining SOFM (software on a programmable) network and BP (back propagation) neural network

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