CN107908853A - Vehicle operational mode design method based on prior information and big data - Google Patents
Vehicle operational mode design method based on prior information and big data Download PDFInfo
- Publication number
- CN107908853A CN107908853A CN201711100950.1A CN201711100950A CN107908853A CN 107908853 A CN107908853 A CN 107908853A CN 201711100950 A CN201711100950 A CN 201711100950A CN 107908853 A CN107908853 A CN 107908853A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- mtd
- state
- msup
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Aviation & Aerospace Engineering (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Feedback Control In General (AREA)
Abstract
The present invention relates to the vehicle operational mode design method based on prior information and big data, including following three phases:First, the representational characteristic parameter of expression is determined;2nd, using typical condition database and resampling database as priori floor data storehouse, its state and state transinformation are counted, the prior information as design conditions;3rd, the new genetic algorithm design conditions based on prior information are utilized;The present invention is a kind of new method of the big data design vehicle operating condition using low sample frequency, solve thes problems, such as current vehicle operational mode design method to the low sample frequency data invalid less than 1Hz.
Description
Technical field
It is particularly a kind of to be transported based on the automobile of prior information and big data the present invention relates to vehicle operational mode design method
Row operating condition design method.
Background technology
As location technology, the development of mobile equipment and the popularization of intelligent terminal, urban transportation produce mass data daily,
And data increase rapidly with time increase.It is city under the big data background of representative in the traffic information data of magnanimity
The real-time running state of Floating Car acquisition technique collection taxi is used to bring new machine for the design of city operating mode in traffic system
Meet.Compared with the microcosmic adjacent change of the data even change of moment, traffic department more concerns the grand of floating car data
See phenomenon, such as the magnitude of traffic flow etc., so Floating Car frequency acquisition is mostly 0.1Hz even lower.
Vehicle operational mode is a large amount of gathered data concentrations of vehicle as a result, it is desirable to be obtained by rational operating condition design method
.In order to meet various demands in development of automobile, researchers have carried out vehicle operational mode design method deep grind
Study carefully.It is studied mainly around two methods.One kind is that the evaluation index of acquired original data is matched based on combination micro travel, raw
Into the method for target operating condition.Shi Shuming etc. are artificially overcome without considering the limitation of long period micro travel, introduce non-zero end
The micro travel of point, and micro travel definition is promoted, it have studied BeiJing, China city operating mode using micro travel combined optimization method;Separately
It is outer it is a kind of be based on Markov chain model generation operating mode method.Lin Jie et al. are based on Markov Chain stochastic model,
Traveling fragment is defined as state, using the initial data for building California Unified operating modes, devises new contrast
Operating mode.Due to considering the probability and transition probability of state generation, it is more representative that it contrasts the more original operating mode of operating mode.
Micro travel is the elementary cell of combination in micro travel combined optimization method, it represents continuous time of the start-stop as idling
Sequence, this method go out a representative operating mode by clustering micro travel unit Combinatorial Optimization;Based on Markov Chain mould
Using the continuous velocity of vehicle actual travel and acceleration time series as analysis object in type method, by discrete speed and add
Speed is defined as state, and statistics current time state is accorded with to the transfer matrix of subsequent time state by Method of Stochastic
The representative operating mode of conjunction condition.During operating condition design, no matter micro travel unit or continuous velocity, acceleration time series
It is using 1s as time interval, i.e., data frequency is 1Hz, and sample frequency is often higher than 1Hz in data acquisition.When original
When the frequency of data is less than 1Hz, micro travel unit or speed, acceleration time series can not just ensure it is continuous in time, namely
The acceleration information under true environment can not be obtained.Meanwhile the Shannon's sampling theorem recovered from signal, work as sample frequency
Less than signal 2 times of peak frequency when, original signal can not be rebuild.To sum up, current design conditions method will no longer be suitable for
Sample frequency is less than the initial data of 1Hz.
The content of the invention
The object of the present invention is to provide a kind of design method of the vehicle operational mode based on prior information and big data, profit
With Floating Car big data design operating conditions in traffic system, for the operating condition design of low frequency big data, efficiently solve and work as work
When condition sample frequency is less than 1Hz, current vehicle operational mode design method does not apply to problem.
The purpose of the present invention is achieved through the following technical solutions:
Vehicle operational mode design method based on prior information and big data, including following three phases:
First, the representational characteristic parameter of expression is determined
According to markovian ergodic theory and the statistical analysis of big data resampling, determine that design conditions have
Representational characteristic parameter;Comprise the following steps that:
1) characteristic parameter of speed is determined according to Markov Chain ergodic theory:
Markov Chain n ∈ T at any one time1One-dimensional distribution:
P (n)=p (0) P (n) (1)
In formula, p (n) is Markov Chain any instant n ∈ T1When one-dimensional distribution, p (0) is initial distribution, and P (n) is
N-th step transition probability matrix;
From C-K equations, the n-th step transition probability matrix is
P (n)=Pn (2)
Separately from Markov Chain ergodic theory, there are Limit Distribution, sees formula (3)
π=π P (3)
In formula, π=(π1,π2,…,πN) it is Limit Distribution, P is a step transition probability matrix;
If by the use of π as the initial distribution of chain, i.e. p (0)=π, by formula (1), (2) and (3) formula, has
P (n)=p (0) P (n)=π Pn=π Pn-1=...=π P=π (4)
Then Markov Chain n ∈ T at any one time1One-dimensional distribution p (n) it is forever consistent with π;
From (4) formula, the transfer of the Markov Chain of limited a state Jing Guo any step is intended to the same limit
Distribution, namely Stationary Distribution;Strong Markov property is had according to vehicle operational mode, i.e., from a state to another state
Transfer, must have definite transition probability;Therefore using speed as state, the operating mode time series of low sample frequency big data
Identical Stationary Distribution is correspond to the operating mode time series of high sample frequency;Therefore, it is true according to the big data of low sample frequency
The representational velocity characteristic parameter of accepted opinion valency operating mode, including dead time ratio (%), average speed (km/h), average traveling
Speed (km/h) and travel speed standard deviation (km/h);
2) acceleration signature parameter is determined according to the statistical analysis of resampling data
Resampling technique is used to the big data of low sample frequency, obtains the resampling time series of approximate 1Hz data.Profit
With the higher uniformity of resampling sequence and original operating mode sequence, (speed and acceleration are general for the VA distributions of statistics resampling data
Rate is distributed) relevant common acceleration signature index, for evaluating the representativeness of design conditions;Finally definite resampling data
Acceleration time ratio (%), at the uniform velocity running time ratio (%), deceleration time ratio (%), PKE (m/s2), average accelerate
Spend (m/s2), average retardation rate (m/s2) and VA distribution related coefficients totally 7 acceleration relevant feature parameters;
Wherein, PKE is calculated as follows
PKE represents the positive acceleration kinetic energy in unit distance;VfAnd ViIt is to terminate and originate speed in single accelerator respectively
Degree, D is total kilometres;
2nd, statistical prior information
Using typical condition database and resampling database as priori floor data storehouse, its state and state transfer are counted
Information, the prior information as design conditions;Comprise the following steps that:
1) priori transition probability matrix storehouse is counted
The resampling operating mode of domestic and international typical condition and first stage step 2) is chosen, forms priori floor data storehouse;It is first
First, determine that velocity interval arrives maximum V for 0max, acceleration range is minimum value AminTo maximum Amax, speed step-length is set
For gapV, acceleration step-length is gapA;Then, idling section is defined, corresponding speed interval arrives for 0Acceleration model
Enclose forArriveFinally, in a space encoder, the state of all operating modes is defined;Joined using section representation
Value in some section of number, the speed V at the i-th momentiSection code miFor
mi=floor (Vi/gapV)+1
Acceleration AiSection code niFor
ni=floor ((Ai-Amin)/gapA)+1
Then by the speed interval code m at the i-th momentiWith acceleration section code niCalculate the state s at the i-th momentiFor
si=mi+(ni-1)×M
Wherein, M is speed interval code number, M=floor (Vmax/gapV)+1
The state transition probability matrix of each operating mode is calculated according to formula (5);
Wherein,pijFor one-dimensional space state siTo one-dimensional space state sjTransition probability, S is spatiality
Set, NijFor state siTo sjTransfer frequency, NiFor state siTo the transition frequency of free position;Finally obtain priori transfer
Probability matrix storehouse, is denoted as ETPMs (Empirical Transition Probability Matrix);
2) priori transfer matrix is counted
In ETPMs on the two dimensional surface of each transition probability matrix current state and NextState, multiple matrixes are counted
State and state transinformation;State corresponds the state in ETPMs and state transfer according to arranging from small to large
Onto a state-transition matrix;This matrix is referred to as priori transfer matrix, note ETM (Empirical Transition
Matrix), it whether there is the feasible zone of transfer relationship for expressing the state in original floor data storehouse and state;
3rd, the new genetic algorithm design conditions based on prior information are utilized
Using the velocity characteristic parameter of first stage step 1) and the acceleration signature parameter designing object function of step 2),
ETPMs construction populations and design mutation operator using second stage step 1), the EPM design crossover operators of step 2), most
Quadruple new genetic algorithm design operating conditions afterwards;Comprise the following steps that:
1) initial population constructs
First, a state transition probability matrix is randomly choosed in the ETPMs of second stage step 1);It is then based on horse
Er Kefu chains Method of Stochastic generates certain length status switch, specific as follows:Selection idling mode is original state X (1)
=k0, according to pseudo random number, random number r is taken between 0 to 11;Selection meets next shape of the state as sequence of formula (6)
State X (2)=k1, constantly repeat, until generation certain length and the operating mode sequence that final state is idling mode;Then use
Integer coding, is gene state encoding, and sequential coding is chromosome;Repeat all of above step and obtain a plurality of chromosome, can
The initial population of random composition arbitrary size;
2) object function designs
First, the characteristic parameter that setting first stage step 1) and step 2) obtain is with respect to tolerance absolute value;Then
According to formula (7) design of expression operating mode and the uniformity of original operating mode;
|Ioi-Iei(X)|≤di, (i=1,2 ..., n) (7)
Wherein X is design conditions, IoFor the characteristic parameter determined in first stage step 1) and step 2), IeTo set
The corresponding characteristic parameter of operating mode is counted, d is tolerance, and n is characterized the number of parameter;
Formula (7) is changed into formula (8)
Finally, by constructing side formula (9), object function (10) to the end is obtained
3) the crossover operator design of Markov property is met
First, any two individual X of equal length are randomly choosed(1), X(2);Using the ETM of one step 2) of stage, in X(1),
X(2)Randomly find while meet four neighboring genes of (11) to (14)
Wherein,WithWithIt is X(1)In two pairs of neighboring genes,WithWithIt is X(2)
In two pairs of neighboring genes, and i needs not be equal to j, and i' needs not be equal to j';
Then, X(1)And X(2)The non-equipotential transposition section being exchanged with each other between i to i' and j to j', generation meet Markov
Property intersection individual;
4) the mutation operator design of Markov property is met
First, a state transition probability matrix is randomly choosed in the ETPMs of one step 1) of stage;Then according to starting point
Requirement with terminal state for idling mode and certain length, one is generated using the Method of Stochastic of phase III step 1)
Bar state sequence;Using individual as variation after integer coding;
5) new genetic algorithm is implemented to evolve
Population Size and iterations are set, the new genetic algorithm being made of above four-stage, which is evolved, to be exported to optimal suitable
It should be worth, and decode its corresponding work condition state sequence, finally obtain speed and the time series of acceleration, i.e. design conditions.
Design method of the present invention is to carry out the mistake of operating condition design to the big data of low sample frequency based on prior information
Journey;From theoretical and statistical analysis, the expression representational characteristic parameter of operating condition is determined;Using typical condition database and
Resampling database empirically information, for expressing the state transinformation of 1Hz initial data;It is finally based on the new of design
Genetic algorithm Evolutionary Design goes out representative operating mode.
The present invention is a kind of new method of the big data design vehicle operating condition using low sample frequency, is efficiently solved
Current vehicle operational mode design method is under the low sample frequency data less than 1Hz, the problem of being no longer applicable in.Identical
Under design object, the representative of the original 1Hz data of Markov chain model design is currently based on original 1Hz data and utilization
Property operating mode contrast, three VA distribution there is high consistency, the power spectral density of three has similitude;With low sampling frequency
Rate data comparison, design conditions have been reappeared in vehicle actual motion by the travelling characteristic of the influences such as traffic flow, driving performance.
Brief description of the drawings
Fig. 1 is design method flow diagram of the present invention;
Fig. 2 is the Velocity Time sequence chart described in design method of the present invention;
Fig. 3 is the acceleration time series chart described in design method of the present invention;
Fig. 4 is the VA probability distribution graphs of the original operating modes of 1Hz described in design method of the present invention;
Fig. 5 is the VA probability distribution graphs of the representative operating mode described in design method of the present invention;
Fig. 6 is the VA probability distribution graphs of the design conditions described in design method of the present invention;
The power spectrum of the design conditions of Fig. 7 the method for the invention, representative operating mode and the original operating modes of 1Hz;
The design conditions of Fig. 8 the method for the invention and the power spectrum of 0.1Hz data operating modes.
Specific implementation method
Design method of the present invention is described in further detail by following embodiments and attached drawing.
With reference to Fig. 1, a kind of vehicle operational mode design method based on prior information and big data of the invention, including it is following
Three phases, details are as follows for embodiment:
Stage one:Determine the expression representational characteristic parameter of operating condition
Determine that the expression representational characteristic parameter of operating condition mainly includes following two steps:
1) characteristic parameter of speed is determined according to Markov Chain ergodic theory:
Markov Chain n ∈ T at any one time1One-dimensional distribution:
P (n)=p (0) P (n) (1)
In formula, p (n) is Markov Chain any instant n ∈ T1When one-dimensional distribution, p (0) is initial distribution, and P (n) is
N-th step transition probability matrix.
From C-K equations, the n-th step transition probability matrix is
P (n)=Pn (2)
Separately from Markov Chain ergodic theory, there are Limit Distribution, sees formula (3)
π=π P (3)
In formula, π=(π1,π2,...,πN) it is Limit Distribution, P is a step transition probability matrix.
If by the use of π as the initial distribution of chain, i.e. p (0)=π, by formula (1), (2) and (3) formula, has
P (n)=p (0) P (n)=π Pn=π Pn-1=...=π P=π (4)
Then Markov Chain n ∈ T at any one time1One-dimensional distribution p (n) it is forever consistent with π.
From (4) formula, the transfer of the Markov Chain of limited a state Jing Guo any step is intended to the same limit
Distribution, namely Stationary Distribution;Strong Markov property is had according to vehicle operational mode, i.e., from a state to another state
Transfer, must have definite transition probability;Therefore using speed as state, the operating mode time series of low sample frequency big data
Identical Stationary Distribution is correspond to the operating mode time series of high sample frequency;Therefore, it is true according to the big data of low sample frequency
The representational velocity characteristic parameter of accepted opinion valency operating mode, including dead time ratio (%), average speed (km/h), average traveling
Speed (km/h) and travel speed standard deviation (km/h).
2) acceleration signature parameter is determined according to the statistical analysis of resampling data
Resampling technique is used to the urban road big data of low sample frequency, when obtaining the resampling of approximate 1Hz data
Between sequence.Referring to table 1, the acceleration relevant feature parameters of the original operating modes of urban road 1Hz and resampling operating mode are compared for, its
Relative deviation absolute value is substantially remained within 10%, indicates the characteristic parameter similitude of resampling operating mode and original operating mode;
Thus, using the higher uniformity of resampling sequence and original operating mode sequence, the VA distributions of statistics resampling data (speed and
Acceleration probability distribution) relevant common acceleration signature index, for evaluating the representativeness of design conditions;It is final to determine weight
The acceleration time ratio (%) of sampled data, at the uniform velocity running time ratio (%), deceleration time ratio (%), PKE (m/s2)、
Average acceleration (m/s2), average retardation rate (m/s2) and VA distribution related coefficients totally 7 acceleration relevant feature parameters.
Table 1
Wherein, PKE is calculated as follows
PKE represents the positive acceleration kinetic energy in unit distance;VfAnd ViIt is to terminate and originate speed in single accelerator respectively
Degree, D is total kilometres.
Stage two:Statistical prior information
It is as follows that statistical prior information mainly includes following two steps:
1) priori transition probability matrix storehouse is counted
The resampling operating mode of domestic and international 81 typical conditions and first stage step 2) is chosen, forms priori floor data
Storehouse.First, determine that velocity interval arrives maximum V for 0max, VmaxFor 38m/s, acceleration range is minimum value AminTo maximum
Amax, AminFor -4m/s2, AmaxFor 4m/s2, it is 0.5m/s to set speed step-length gapV, and acceleration step-length gapA is 0.1m/s2。
Then, idling section is defined, corresponding speed interval arrives 0.1m/s for 0, and acceleration range arrives 0.02m/s for -0.022.Finally,
In a space encoder, the state of all operating modes is defined.Use the value in some section of representation parameter of section, the i-th moment
Speed ViSection code miFor
mi=floor (Vi/gapV)+1
Acceleration AiSection code niFor
ni=floor ((Ai-Amin)/gapA)+1
Then by the speed interval code m at the i-th momentiWith acceleration section code niCalculate the state s at the i-th momentiFor
si=mi+(ni-1)×M
Wherein, M is speed interval code number, M=floor (Vmax/gapV)+1
The state transition probability matrix of each operating mode is calculated according to formula (5).
Wherein,pijFor one-dimensional space state siTo one-dimensional space state sjTransition probability, S is spatiality
Set, NijFor state siTo sjTransfer frequency, NiFor state siTo the transition frequency of free position;Finally obtain priori transfer
Probability matrix storehouse, is denoted as ETPMs (Empirical Transition Probability Matrix).
2) priori transfer matrix is counted
In ETPMs on the two dimensional surface of each transition probability matrix current state and NextState, multiple matrixes are counted
State and state transinformation;State corresponds the state in ETPMs and state transfer according to arranging from small to large
Onto a state-transition matrix;This matrix is referred to as priori transfer matrix, note ETM (Empirical Transition
Matrix), it whether there is the feasible zone of transfer relationship for expressing the state in original floor data storehouse and state.
Stage three:Utilize the new genetic algorithm design conditions based on prior information
Mainly include following five steps using the new genetic algorithm design conditions based on prior information:
1) initial population constructs
First, a state transition probability matrix is randomly choosed in the ETPMs of second stage step 1);It is then based on horse
Er Kefu chains Method of Stochastic generates certain length status switch, and urban design operating mode length is 2400s, specific as follows:Choosing
Idling mode is taken, its state encoding is 3081, as original state X (1)=k0, according to pseudo random number, taken between 0 to 1 with
Machine number r1.Selection meets NextState X (2)=k of the state as sequence of formula (6)1, constantly repeat, until generation length
For 2400 and final state be 3081 operating mode sequence.Then integer coding is used, is gene state encoding, sequential coding
For chromosome.Repeat all of above step and obtain 100 chromosomes, it is random to form the initial population that size is 100.
2) object function designs
First, the velocity characteristic parameter of first stage step 1) is set with respect to tolerance absolute value as 5%, step 2)
The acceleration signature parameter arrived is 10% with respect to tolerance absolute value;Then according to formula (7) design of expression operating mode with it is original
The uniformity of operating mode;
|Ioi-Iei(X)|≤di, (i=1,2 ..., n) (7)
Wherein X is design conditions, IoFor the characteristic parameter determined in first stage step 1) and step 2), IeTo set
The corresponding characteristic parameter of operating mode is counted, d is tolerance, and n is characterized the number of parameter, totally 11.
Formula (7) is changed into formula (8)
Finally, by constructing side formula (9), object function (10) to the end is obtained
3) the crossover operator design of Markov property is met
First, any two individual X of equal length are randomly choosed(1), X(2);Using the ETM of one step 2) of stage, in X(1),
X(2)Randomly find while meet four neighboring genes of (11) to (14)
Wherein,WithWithIt is X(1)In two pairs of neighboring genes,WithWithIt is X(2)
In two pairs of neighboring genes, and i needs not be equal to j, and i' needs not be equal to j'.
Then, X(1)And X(2)The non-equipotential transposition section being exchanged with each other between i to i' and j to j', generation meet Markov
Property intersection individual.
4) the mutation operator design of Markov property is met
First, a state transition probability matrix is randomly choosed in the ETPMs of one step 1) of stage;Then according to starting point
The requirement that with terminal state be 3081 and length is 2400, one is generated using the Method of Stochastic of phase III step 1)
Status switch;Using individual as variation after integer coding.
5) new genetic algorithm is implemented to evolve
Population Size 100 is set, iterations 100, the new genetic algorithm being made of above four-stage, which is evolved, to be exported
To optimal adaptation value, and its corresponding work condition state sequence is decoded, finally obtain speed and the time series of acceleration, that is, set
Count operating mode.
Design method of the present invention meets to check index and meets acceleration and deceleration frequent urban road operating modes, referring to Fig. 2 and
Fig. 3;The VA probability distribution of the original operating modes of 1Hz, the representative operating mode of conventional method design and design conditions three of the present invention has
There are local distribution concentration, the high feature of density, the similar result for indicating design method of the present invention of feature has reasonability, ginseng
See Fig. 4, Fig. 5 and Fig. 6;Design conditions, the representative operating mode of conventional method design and three kinds of the original operating mode of 1Hz of the present invention
Power spectrum has similitude on each frequency band, the validity of design method of the present invention is indicated, referring to Fig. 7;The present invention is set
The design conditions of meter method and the power spectrum of 0.1Hz data operating modes indicate design conditions and have reappeared vehicle actual travel process
In travelling characteristic affected by various factors, referring to Fig. 8;To sum up, operating condition design method of the present invention solves traditional operating condition design
Method can not use the problem of low-frequency data design liaison operating mode.
Claims (1)
1. the vehicle operational mode design method based on prior information and big data, it is characterised in that:Including following three phases:
First, the representational characteristic parameter of expression is determined
According to markovian ergodic theory and the statistical analysis of big data resampling, determine that design conditions have and represent
The characteristic parameter of property;Comprise the following steps that:
1) characteristic parameter of speed is determined according to Markov Chain ergodic theory:
Markov Chain n ∈ T at any one time1One-dimensional distribution:
P (n)=p (0) P (n) (1)
In formula, p (n) is Markov Chain any instant n ∈ T1When one-dimensional distribution, p (0) is initial distribution, and P (n) is the n-th step
Transition probability matrix;
From C-K equations, the n-th step transition probability matrix is
P (n)=Pn (2)
Separately from Markov Chain ergodic theory, there are Limit Distribution, sees formula (3)
π=π P (3)
In formula, π=(π1,π2,…,πN) it is Limit Distribution, P is a step transition probability matrix;
If by the use of π as the initial distribution of chain, i.e. p (0)=π, by formula (1), (2) and (3) formula, has
P (n)=p (0) P (n)=π Pn=π Pn-1=...=π P=π (4)
Then Markov Chain n ∈ T at any one time1One-dimensional distribution p (n) it is forever consistent with π;
From (4) formula, the transfer of the Markov Chain of limited a state Jing Guo any step is intended to same Limit Distribution,
Namely Stationary Distribution;Strong Markov property is had according to vehicle operational mode, i.e., from a state to the transfer of another state,
There must be definite transition probability;Therefore using speed as state, the operating mode time series of low sample frequency big data and high sampling
The operating mode time series of frequency correspond to identical Stationary Distribution;Therefore, evaluation work is determined according to the big data of low sample frequency
The representational velocity characteristic parameter of condition, including dead time ratio (%), average speed (km/h), average overall travel speed (km/h)
With travel speed standard deviation (km/h);
2) acceleration signature parameter is determined according to the statistical analysis of resampling data
Resampling technique is used to the big data of low sample frequency, obtains the resampling time series of approximate 1Hz data.Utilize weight
Sample sequence and the higher uniformity of original operating mode sequence, VA distributions (speed and the acceleration probability point of statistics resampling data
Cloth) relevant common acceleration signature index, for evaluating the representativeness of design conditions;The acceleration of final definite resampling data
Time scale (%), at the uniform velocity running time ratio (%), deceleration time ratio (%), PKE (m/s2), average acceleration (m/s2)、
Average retardation rate (m/s2) and VA distribution related coefficients totally 7 acceleration relevant feature parameters;
Wherein, PKE is calculated as follows
<mrow>
<mi>P</mi>
<mi>K</mi>
<mi>E</mi>
<mo>=</mo>
<mo>&Sigma;</mo>
<mfrac>
<mrow>
<msubsup>
<mi>V</mi>
<mi>f</mi>
<mn>2</mn>
</msubsup>
<mo>-</mo>
<msubsup>
<mi>V</mi>
<mi>i</mi>
<mn>2</mn>
</msubsup>
</mrow>
<mi>D</mi>
</mfrac>
<mo>,</mo>
<mfrac>
<mrow>
<mi>d</mi>
<mi>V</mi>
</mrow>
<mrow>
<mi>d</mi>
<mi>t</mi>
</mrow>
</mfrac>
<mo>></mo>
<mn>0</mn>
</mrow>
PKE represents the positive acceleration kinetic energy in unit distance;VfAnd ViIt is that termination and starting velocity, D are in single accelerator respectively
Total kilometres;
2nd, statistical prior information
Using typical condition database and resampling database as priori floor data storehouse, its state and state transfer letter are counted
Breath, the prior information as design conditions;Comprise the following steps that:
1) priori transition probability matrix storehouse is counted
The resampling operating mode of domestic and international typical condition and first stage step 2) is chosen, forms priori floor data storehouse;First, really
Determine velocity interval and arrive maximum V for 0max, acceleration range is minimum value AminTo maximum Amax, it is gapV to set speed step-length,
Acceleration step-length is gapA;Then, idling section is defined, corresponding speed interval arrives for 0Acceleration range isArriveFinally, in a space encoder, the state of all operating modes is defined;Using section representation parameter some
Value in section, the speed V at the i-th momentiSection code miFor
mi=floor (Vi/gapV)+1
Acceleration AiSection code niFor
ni=floor ((Ai-Amin)/gapA)+1
Then by the speed interval code m at the i-th momentiWith acceleration section code niCalculate the state s at the i-th momentiFor
si=mi+(ni-1)×M
Wherein, M is speed interval code number, M=floor (Vmax/gapV)+1
The state transition probability matrix of each operating mode is calculated according to formula (5);
<mrow>
<mi>P</mi>
<mo>=</mo>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msub>
<mi>p</mi>
<mn>11</mn>
</msub>
</mtd>
<mtd>
<msub>
<mi>p</mi>
<mn>12</mn>
</msub>
</mtd>
<mtd>
<mn>...</mn>
</mtd>
<mtd>
<msub>
<mi>p</mi>
<mrow>
<mn>1</mn>
<mi>m</mi>
</mrow>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>p</mi>
<mn>21</mn>
</msub>
</mtd>
<mtd>
<msub>
<mi>p</mi>
<mn>22</mn>
</msub>
</mtd>
<mtd>
<mn>...</mn>
</mtd>
<mtd>
<msub>
<mi>p</mi>
<mrow>
<mn>2</mn>
<mi>m</mi>
</mrow>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mrow></mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mrow></mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mo>.</mo>
</mtd>
<mtd>
<mrow></mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>p</mi>
<mrow>
<mi>m</mi>
<mn>1</mn>
</mrow>
</msub>
</mtd>
<mtd>
<msub>
<mi>p</mi>
<mrow>
<mi>m</mi>
<mn>2</mn>
</mrow>
</msub>
</mtd>
<mtd>
<mn>...</mn>
</mtd>
<mtd>
<msub>
<mi>p</mi>
<mrow>
<mi>m</mi>
<mi>m</mi>
</mrow>
</msub>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>,</mo>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>&Element;</mo>
<mi>S</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>5</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein,pijFor one-dimensional space state siTo one-dimensional space state sjTransition probability, S is spatiality set,
NijFor state siTo sjTransfer frequency, NiFor state siTo the transition frequency of free position;Finally obtain priori transition probability square
Battle array storehouse, is denoted as ETPMs (Empirical Transition Probability Matrix);
2) priori transfer matrix is counted
In ETPMs on the two dimensional surface of each transition probability matrix current state and NextState, the shape of multiple matrixes is counted
State and state transinformation;State corresponds the state in ETPMs and state transfer to one according to arranging from small to large
On state-transition matrix;This matrix is referred to as priori transfer matrix, and note ETM (Empirical Transition Matrix), is used
It whether there is the feasible zone of transfer relationship to express the state in original floor data storehouse and state;
3rd, the new genetic algorithm design conditions based on prior information are utilized
Using the velocity characteristic parameter of first stage step 1) and the acceleration signature parameter designing object function of step 2), utilize
The ETPMs construction populations of second stage step 1) and design mutation operator, the EPM design crossover operators of step 2), finally by four
The new genetic algorithm design operating conditions of part composition;Comprise the following steps that:
1) initial population constructs
First, a state transition probability matrix is randomly choosed in the ETPMs of second stage step 1);It is then based on Ma Erke
Husband's chain Method of Stochastic generates certain length status switch, specific as follows:Selection idling mode is original state X (1)=k0,
According to pseudo random number, random number r is taken between 0 to 11;Selection meets NextState X (2) of the state as sequence of formula (6)
=k1, constantly repeat, until generation certain length and the operating mode sequence that final state is idling mode;Then compiled using integer
Code, is gene state encoding, sequential coding is chromosome;Repeat all of above step and obtain a plurality of chromosome, can random groups
Into the initial population of arbitrary size;
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<msub>
<mi>k</mi>
<mn>1</mn>
</msub>
<mo>-</mo>
<mn>1</mn>
</mrow>
</munderover>
<msub>
<mi>p</mi>
<mrow>
<msub>
<mi>k</mi>
<mn>0</mn>
</msub>
<mi>j</mi>
</mrow>
</msub>
<mo>&le;</mo>
<msub>
<mi>r</mi>
<mn>1</mn>
</msub>
<mo>&le;</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msub>
<mi>k</mi>
<mn>1</mn>
</msub>
</munderover>
<msub>
<mi>p</mi>
<mrow>
<msub>
<mi>k</mi>
<mn>0</mn>
</msub>
<mi>j</mi>
</mrow>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>6</mn>
<mo>)</mo>
</mrow>
</mrow>
2) object function designs
First, the characteristic parameter that setting first stage step 1) and step 2) obtain is with respect to tolerance absolute value;Then basis
The uniformity of formula (7) design of expression operating mode and original operating mode;
|Ioi-Iei(X)|≤di, (i=1,2 ..., n) (7)
Wherein X is design conditions, IoFor the characteristic parameter determined in first stage step 1) and step 2), IeTo design work
The corresponding characteristic parameter of condition, d are tolerance, and n is characterized the number of parameter;
Formula (7) is changed into formula (8)
<mrow>
<msub>
<mi>f</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mo>|</mo>
<mrow>
<mn>1</mn>
<mo>-</mo>
<mfrac>
<mrow>
<msub>
<mi>I</mi>
<mrow>
<mi>e</mi>
<mi>i</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>)</mo>
</mrow>
</mrow>
<msub>
<mi>I</mi>
<mrow>
<mi>o</mi>
<mi>i</mi>
</mrow>
</msub>
</mfrac>
</mrow>
<mo>|</mo>
<mo>,</mo>
<msub>
<mover>
<mi>f</mi>
<mo>&OverBar;</mo>
</mover>
<mi>i</mi>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>d</mi>
<mi>i</mi>
</msub>
<msub>
<mi>I</mi>
<mrow>
<mi>o</mi>
<mi>i</mi>
</mrow>
</msub>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>8</mn>
<mo>)</mo>
</mrow>
</mrow>
Finally, by constructing side formula (9), object function (10) to the end is obtained
<mrow>
<msub>
<mi>T</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>f</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>X</mi>
<mo>)</mo>
</mrow>
<mo>+</mo>
<msub>
<mover>
<mi>f</mi>
<mo>&OverBar;</mo>
</mover>
<mi>i</mi>
</msub>
<mo>-</mo>
<mi>min</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mi>i</mi>
</msub>
<mo>(</mo>
<mi>X</mi>
<mo>)</mo>
<mo>,</mo>
<msub>
<mover>
<mi>f</mi>
<mo>&OverBar;</mo>
</mover>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>9</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>F</mi>
<mo>=</mo>
<munder>
<mi>min</mi>
<mrow>
<mi>x</mi>
<mo>&Element;</mo>
<mi>S</mi>
</mrow>
</munder>
<mrow>
<mo>(</mo>
<mfrac>
<mn>1</mn>
<mi>n</mi>
</mfrac>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<msub>
<mi>T</mi>
<mi>j</mi>
</msub>
<mo>(</mo>
<mi>X</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>10</mn>
<mo>)</mo>
</mrow>
</mrow>
3) the crossover operator design of Markov property is met
First, any two individual X of equal length are randomly choosed(1), X(2);Using the ETM of one step 2) of stage, in X(1), X(2)
Randomly find while meet four neighboring genes of (11) to (14)
<mrow>
<mi>E</mi>
<mi>T</mi>
<mi>M</mi>
<mo>{</mo>
<msup>
<mi>X</mi>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<msup>
<mi>X</mi>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<mi>i</mi>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>}</mo>
<mo>></mo>
<mn>0</mn>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>11</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>E</mi>
<mi>T</mi>
<mi>M</mi>
<mo>{</mo>
<msup>
<mi>X</mi>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<msup>
<mi>X</mi>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<mi>j</mi>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>}</mo>
<mo>></mo>
<mn>0</mn>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>12</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>E</mi>
<mi>T</mi>
<mi>M</mi>
<mo>{</mo>
<msup>
<mi>X</mi>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<mrow>
<msup>
<mi>j</mi>
<mo>&prime;</mo>
</msup>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<msup>
<mi>X</mi>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<msup>
<mi>i</mi>
<mo>&prime;</mo>
</msup>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>}</mo>
<mo>></mo>
<mn>0</mn>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>13</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>E</mi>
<mi>T</mi>
<mi>M</mi>
<mo>{</mo>
<msup>
<mi>X</mi>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<mrow>
<msup>
<mi>i</mi>
<mo>&prime;</mo>
</msup>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>|</mo>
<msup>
<mi>X</mi>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<msubsup>
<mi>x</mi>
<msup>
<mi>j</mi>
<mo>&prime;</mo>
</msup>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</msubsup>
<mo>}</mo>
<mo>></mo>
<mn>0</mn>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>14</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein,With WithIt is X(1)In two pairs of neighboring genes,With WithIt is X(2)In two pairs
Neighboring gene, and i needs not be equal to j, i' needs not be equal to j';
Then, X(1)And X(2)The non-equipotential transposition section being exchanged with each other between i to i' and j to j', generation meet Markov property
Intersect individual;
4) the mutation operator design of Markov property is met
First, a state transition probability matrix is randomly choosed in the ETPMs of one step 1) of stage;Then according to starting point and end
Dotted state is the requirement of idling mode and certain length, and a bar state is generated using the Method of Stochastic of phase III step 1)
Sequence;Using individual as variation after integer coding;
5) new genetic algorithm is implemented to evolve
Population Size and iterations are set, the new genetic algorithm being made of above four-stage, which is evolved, to be exported to optimal adaptation
Value, and its corresponding work condition state sequence is decoded, finally obtain speed and the time series of acceleration, i.e. design conditions.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711100950.1A CN107908853B (en) | 2017-11-10 | 2017-11-10 | Automobile operation condition design method based on prior information and big data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711100950.1A CN107908853B (en) | 2017-11-10 | 2017-11-10 | Automobile operation condition design method based on prior information and big data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107908853A true CN107908853A (en) | 2018-04-13 |
CN107908853B CN107908853B (en) | 2020-07-31 |
Family
ID=61843897
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711100950.1A Active CN107908853B (en) | 2017-11-10 | 2017-11-10 | Automobile operation condition design method based on prior information and big data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107908853B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110807290A (en) * | 2019-09-09 | 2020-02-18 | 西安工业大学 | Road gradient-containing automobile operation condition design method based on self-adaptive Markov chain evolution |
CN110826145A (en) * | 2019-09-09 | 2020-02-21 | 西安工业大学 | Automobile multi-parameter operation condition design method based on heuristic Markov chain evolution |
CN111401573A (en) * | 2018-12-17 | 2020-07-10 | 中国科学院沈阳自动化研究所 | Working condition state modeling and model correcting method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103246943A (en) * | 2013-05-31 | 2013-08-14 | 吉林大学 | Vehicle operating condition multi-scale predicting method based on Markov chain |
US20170074964A1 (en) * | 2015-09-16 | 2017-03-16 | Here Global B.V. | Method And System Of Location Estimation And Navigation Of Autonomous Vehicles |
CN107122573A (en) * | 2017-06-26 | 2017-09-01 | 吉林大学 | The vehicle operational mode design method evolved based on Markov Chain |
-
2017
- 2017-11-10 CN CN201711100950.1A patent/CN107908853B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103246943A (en) * | 2013-05-31 | 2013-08-14 | 吉林大学 | Vehicle operating condition multi-scale predicting method based on Markov chain |
US20170074964A1 (en) * | 2015-09-16 | 2017-03-16 | Here Global B.V. | Method And System Of Location Estimation And Navigation Of Autonomous Vehicles |
CN107122573A (en) * | 2017-06-26 | 2017-09-01 | 吉林大学 | The vehicle operational mode design method evolved based on Markov Chain |
Non-Patent Citations (3)
Title |
---|
SHUMING SHI等: "Research on Markov property analysis of driving cycles and its application", 《TRANSPORTATION RESEARCH PART D》 * |
石琴等: "基于组合聚类法的行驶工况构建与精度分析", 《汽车工程》 * |
邢志国等: "基于VP分布的乘用车运行工况设计方法研究", 《交通信息与安全》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111401573A (en) * | 2018-12-17 | 2020-07-10 | 中国科学院沈阳自动化研究所 | Working condition state modeling and model correcting method |
CN111401573B (en) * | 2018-12-17 | 2023-10-27 | 中国科学院沈阳自动化研究所 | Working condition state modeling and model correcting method |
CN110807290A (en) * | 2019-09-09 | 2020-02-18 | 西安工业大学 | Road gradient-containing automobile operation condition design method based on self-adaptive Markov chain evolution |
CN110826145A (en) * | 2019-09-09 | 2020-02-21 | 西安工业大学 | Automobile multi-parameter operation condition design method based on heuristic Markov chain evolution |
Also Published As
Publication number | Publication date |
---|---|
CN107908853B (en) | 2020-07-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104239961B (en) | A kind of based on the Economic Dispatch optimization method of crossover algorithm in length and breadth | |
CN107908853A (en) | Vehicle operational mode design method based on prior information and big data | |
CN103530689A (en) | Deep learning-based clustering method | |
CN102737356B (en) | Intelligent public transportation dispatching calculation method | |
CN103208027B (en) | Method for genetic algorithm with local modularity for community detecting | |
CN104635494B (en) | A kind of vehicle rear-end collision collision fuzzy control method based on genetic algorithm optimization | |
CN101807224B (en) | Mesoscopic-microcosmic integrated traffic simulation vehicle flow loading method | |
CN102768701B (en) | High-voltage switch cabinet insulator electric field optimization method based on quantum genetic algorithm | |
CN109815523A (en) | Train operation multiple target differential evolution algorithm based on decomposition | |
CN107845261A (en) | A kind of automobile overall situation operating mode dynamic reconfiguration method based on tensor fusion transport information | |
CN107464005A (en) | Expanded path planning method for vehicle reservation | |
CN102594909A (en) | Multi-objective community detection method based on spectrum information of common neighbour matrix | |
CN110517482A (en) | A kind of Short-time Traffic Flow Forecasting Methods based on 3D convolutional neural networks | |
CN104200272A (en) | Complex network community mining method based on improved genetic algorithm | |
CN110516870A (en) | A kind of more recycling website garbage collecting and transportation methods based on coevolution | |
CN103605734A (en) | Characteristic vector based data transmission compression method and system | |
CN107122573B (en) | Automobile operation condition design method based on Markov chain evolution | |
CN103324821A (en) | GM (1, 1) model prediction method based on combined interpolation | |
CN105022852A (en) | Method for solving product assembly sequence planning problem on the basis of immune particle swarm algorithm | |
CN105868443A (en) | Construction method for microtopographical micrometeorological element field | |
Zhang et al. | Self-adaptive hyper-heuristic Markov chain evolution for generating vehicle multi-parameter driving cycles | |
CN116541644A (en) | Big piece transportation monitoring point layout system based on improved genetic algorithm | |
CN116468469A (en) | Network taxi demand prediction method based on improved EMD-LSTM | |
CN103489040B (en) | A kind of flight collision solution desorption method comprising Local Search | |
CN105043786B (en) | A kind of vehicle operational mode Markov chain design method containing road grade |
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 |