CN110826145B - Automobile multi-parameter operation condition design method based on heuristic Markov chain evolution - Google Patents

Automobile multi-parameter operation condition design method based on heuristic Markov chain evolution Download PDF

Info

Publication number
CN110826145B
CN110826145B CN201910848560.5A CN201910848560A CN110826145B CN 110826145 B CN110826145 B CN 110826145B CN 201910848560 A CN201910848560 A CN 201910848560A CN 110826145 B CN110826145 B CN 110826145B
Authority
CN
China
Prior art keywords
sequence
strategy
working condition
population
max
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
CN201910848560.5A
Other languages
Chinese (zh)
Other versions
CN110826145A (en
Inventor
张曼
施树明
沈云波
程文冬
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.)
Xian Technological University
Original Assignee
Xian Technological 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 Xian Technological University filed Critical Xian Technological University
Priority to CN201910848560.5A priority Critical patent/CN110826145B/en
Publication of CN110826145A publication Critical patent/CN110826145A/en
Application granted granted Critical
Publication of CN110826145B publication Critical patent/CN110826145B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Biomedical Technology (AREA)
  • Genetics & Genomics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Physiology (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses a method for designing multi-parameter operation working conditions of an automobile based on a hyper-heuristic Markov chain evolution. Firstly, defining a strategy boundary variable and designing a plurality of strategy factors based on an operation operator meeting Markov property and by combining strategy functions and diversity; then, based on expected operation conditions, a strategy factor and working condition sequence distribution mechanism and a working condition population and optimal working condition sequence updating mechanism are formulated, and an evaluation function is designed; and finally, selecting a genetic algorithm as a high-level heuristic algorithm, taking the process of applying the multi-strategy factors to the working condition sequence population as a low-level heuristic algorithm, and establishing the high-efficiency multi-parameter working condition design method. The self-adaptability of the policy factor proportion in the design framework enables the operation efficiency to be further obviously improved compared with a Markov chain evolution method, and the method is strong in transportability and convenient for actual operation and use of automobile engineers.

Description

Automobile multi-parameter operation condition design method based on heuristic Markov chain evolution
Technical Field
The invention relates to a method for designing an automobile running condition, in particular to a method for designing an automobile multi-parameter running condition based on a hyper-heuristic Mahalanobis chain evolution.
Background
The generation of representative operating conditions of the automobile is a basic requirement of current automobile testing, evaluation, control, prediction and other researches. For the improvement of the adaptability, various factors are considered, for example, the size of the power element of the hybrid vehicle is selected to be sensitive to the road gradient, so that the design of the working condition including the road gradient is considered. For another example, when simulation verification is performed on the operation condition, only the representative condition design of two parameters of speed and acceleration is considered, and because the gear shifting strategy in the vehicle transmission system simulation software is fixed, and the difference exists between the gear shifting strategy and the actual gear shifting style of a driver, the design condition is difficult to meet the requirement of engine load consistency. Therefore, it is necessary to consider the design of the operating conditions including gear, speed and torque parameters. In summary, the need to design multi-parameter vehicle operating conditions has become strong.
The current multi-parameter working condition design method comprises two main flow methods based on a Markov chain method and an intelligent method derived from the Markov chain method. Although the prior literature is widely applied to the evaluation of rated power of an energy source of a hybrid electric vehicle, the optimization of a power train and the like by establishing a Markov chain model of three parameters of speed, acceleration and gradient. However, due to the nature of markov chain random generation, the advantages of this approach to designing high-dimensional operating conditions are not as significant. For the problem of too long simulation time of the multidimensional Markov chain model, although the number of states can be reduced by increasing the parameter step length, for example, the step lengths of speed, acceleration and gradient in the three-dimensional Markov chain model are properly increased relative to the step lengths of two-dimensional speed and gradient, thereby shortening the simulation time length, the simulation time length can cause the generation of a obviously fluctuating working condition sequence. In addition, as design accuracy demands higher and more parameters are demanded for the behavior design, the markov chain design method is far from coping with. The method for designing high-dimensional operation working conditions with comparative potential is a Markov chain evolution method, wherein the Markov chain is combined with the evolution method to design an operation strategy meeting Markov performance, so that the working condition design process has directivity to approach a target working condition; the method can significantly improve the operating efficiency of the design regime relative to the Markov chain method. However, the markov chain evolution method has two problems, one is that the operator proportion cannot be adjusted, i.e. the self-adaptive performance is lacked; secondly, the flexibility of the algorithm is poor, namely, when the engineering is applied, the designed new strategy factor needs to be transplanted into the traditional genetic algorithm or other functions. There is therefore a continuing need for improvements to existing markov chain evolution methods.
Disclosure of Invention
The invention aims to solve the technical problem of providing a method for designing the multi-parameter operation working condition of the automobile based on the hyper-heuristic Mahalanobis chain evolution, which can effectively solve the problems of self-adaptive adjustment of the proportion of strategy factors and the flexibility of engineering application, broaden the parameter dimension of the design working condition of the current method and improve the huge potential of the design efficiency.
The technical scheme adopted by the invention for solving the technical problems is to provide a method for designing the multi-parameter operation condition of the automobile based on the heuristic Markov chain evolution, which comprises the following steps: s1) selecting a plurality of strategy factors meeting Markov property by defining strategy boundary variables based on the operation strategy of the Markov chain evolution method; s2) designing a strategy factor and a distribution mechanism and an updating mechanism of the working condition sequence, and further generating an evaluation function based on the expected operation working condition; s3), a traditional genetic algorithm is adopted as a high-level control algorithm of a hyper-heuristic architecture, a strategy factor population consisting of strategy factor sequences is evolved and iterated, and finally the operation condition of the expected evaluation value is obtained.
Further, the step S1 includes:
step S11: setting the step length of each parameter based on the actual data collected by the vehicle, dividing the states of each parameter, counting a multi-parameter state transition probability matrix P, and generating a target working condition length L by using a Markov chain random simulation method, wherein the starting and stopping states are candidate working conditions of idle speed;
step S12: based on the multi-parameter state transition probability matrix counted in the step S11, when any two working condition sequences are designed to have equal-length cross section exchange, a cross operator strategy that an individual sequence must meet the Markov chain state transition relation is generated, and a process of randomly selecting a working condition sequence to replace the working condition sequence to be mutated based on the candidate working condition in the step S11 is used as a mutation operator strategy;
step S13: defining the state position of any two working condition sequences as a strategy boundary variable R, and setting a strategy boundary variable vector R as [ R ═ R1,R2,...,Ri,...,Rj,...,Rn]，i,j∈[1,n]And R isi≠RjN is different R numbers;
step S14: according to the vector R of step S13, firstly dividing all equal length exchangeable segments of the two operator strategies satisfying Markov property in step S12 into a plurality of different n +1 groups, generating n +1 equal length cross operator strategies by the equal length exchangeable segments of the n +1 groups, namely n +1 equal length strategy factors capable of playing local search and global search capacities in the evolution process, secondly,keeping the length deviation at a certain threshold T according to the design working conditionrThe non-isometric exchangeable segments of the in-range sum vector R, n +1 group generate n +1 non-isometric strategy factors; finally, the mutation operator strategy including the working condition state updating capability has 2n +3 strategy factors;
step S15: based on the 2n +3 policy factors obtained in step S14, each policy factor ratio p is setrAnd randomly composing a strategy factor sequence Y (theta) with the length t [ theta ]1,θ2,...,θi,...,θt]Where t > (2n +3), θ ∈ [ θ [ ]e1,θe2,…,θe(n+1),θne1,θne2,…,θne(n+1),θm]，θe1,θe2,…,θe(n+1)Corresponding to an equal length cross policy factor, θne1,θne2,…,θne(n+1)Corresponding to a non-equal length cross policy factor, θmIs a complete mutation operator; simultaneously constructing a working condition sequence population Pop (a) with a population size of t and randomly composed of working condition state sequences2){X}＝[X1；X2；...；Xt]Wherein X is a working condition state sequence, and the initial population Pop (in the working condition sequence)2) Setting an evaluation index and a relative deviation threshold T of an operation condition under the condition of { X }rAnd calculating an initial function value F { X } - [ F (X) of the working condition sequence population by means of a satisfaction criterion1),F(X2),...,F(Xt)]And obtaining the optimal function value F (X) of the population of the working condition sequence by the formula (1)*) And optimum operating mode sequence X*；
F(X*)＝min(F{X})(1)。
Further, the parameters in step S11 include speed, acceleration, and road gradient, and the value of L is 1800S; in the step S13, n is 6, and R is [10,20,50,100,150,200 ]](ii) a T in the step S13rThe value of (a) is 10%; p in said step S13rThe value of (1:1:1:2:3:3:3:1:1: 2:3:3: 2) and the value of t is 30.
Further, the step S2 includes:
step S21: under the current k-th strategy factor sequence Y (theta),performing roulette operation on the function value F { X } of the sequence of instances of step S15 to generate a new sequence of instances of the ordered population
Figure BDA0002196130500000041
And correspondingly distributing the working condition sequences in the population to strategy factors in a strategy factor sequence Y (theta) one by one, performing strategy operation, and generating working condition sequence child population
Figure BDA0002196130500000042
Calculating working condition sequence population
Figure BDA0002196130500000043
Function value of
Figure BDA0002196130500000044
Step S22: updating population Pop(2)X and function values F X, to group the offspring
Figure BDA0002196130500000045
Sum function value
Figure BDA0002196130500000046
Respectively replacing parent population Pop(2){ X } and a population function value F { X }; acquiring a working condition sequence filial generation population according to a formula (2)
Figure BDA0002196130500000047
Best function value of
Figure BDA0002196130500000048
And optimal sequence
Figure BDA0002196130500000049
Figure BDA00021961305000000410
When in use
Figure BDA00021961305000000411
By mixing
Figure BDA00021961305000000412
Is given to X*Updating the optimal operating regime sequence X*By passing through
Figure BDA00021961305000000413
Is given to F (X)*) Updating the optimum function value F (X)*)；
When in use
Figure BDA00021961305000000414
When it is, X is*Assigned to population Pop(2){ X } in
Figure BDA00021961305000000415
In that
Figure BDA00021961305000000416
Position will correspond to function value
Figure BDA00021961305000000417
Is given to function value F (X)*) To achieve the optimal working condition sequence X*And the optimum function value F (X)*) Updating of (1);
step S23: outputting the optimal function value F (X) under the current k-th strategy sequence Y (theta)*)。
Further, the step S3 includes:
step S31: randomly forming a size K based on the strategy factor sequence of step S15maxInitial policy factor sequence population of
Figure BDA0002196130500000052
Calculating individual function values of strategy factor sequence
Figure BDA0002196130500000051
Step S32: according to the strategy factor sequence function value F { Y }, according to the default elite probability pdSelecting the retention timeBest K in the forepopulationmax×pdCarrying out traditional operator selection operation on the sequence individuals;
step S33: according to a default cross probability pcIn (K)max-Kmax×pd) Selection from one of the remaining individuals (K)max-Kmax×pd)×pcThe individual sequences are subjected to random two-point crossing operation to generate (K)max-Kmax×pd)×pcA plurality of crossing individuals;
step S34: for the rest Kmax-Kmax×pd-(Kmax-Kmax×pd)×pcPerforming single point mutation operation on the individual to generate Kmax-Kmax×pd-(Kmax-Kmax×pd)×pc(ii) individual variants;
step S35: forming a new strategy factor sequence population by the elite individuals in the step S32, the crossover individuals in the step S33 and the variant individuals in the step S34;
step S36: the best evaluation value F (Y) is judged and output*) Whether the desired threshold T has been reachedrRange, if F (Y)*)≤TrAt this time, F (Y)*)＝F(X*) Wherein Y is*Executing step S37 for the optimal strategy sequence combination, otherwise, returning to step S32;
step S37: the optimal working condition state sequence X*And decoding and outputting the multi-parameter expected operation condition time sequence.
Further, K in the step S31maxIs 24.
Compared with the prior art, the invention has the following beneficial effects: the invention combines the hyper-heuristic architecture with the Markov chain evolution method, effectively solves the problems of self-adaptive adjustment and program packaging of the proportion of strategy factors designed for the operating condition of the automobile, can widen the parameter dimension of the design condition of the current method, and can improve the huge potential of the design efficiency. In addition, the framework provided by the invention has strong portability and provides support for the operation and use of automobile engineers.
Drawings
FIG. 1 is a block diagram of a hyper-heuristic Markov chain evolution method of the present invention;
FIG. 2 is a graph of velocity over time for the results of the inventive design;
FIG. 3 is a graph of acceleration versus time for the results of the present invention;
FIG. 4 is a graph of slope versus time for the design results of the present invention;
FIG. 5 is a histogram of the average number distribution of the policy factors when the output of the 10 policy factor populations is expected to operate according to the present invention;
FIG. 6 is a histogram of the run time of 10 design tests of the hyper-heuristic Markov chain evolution method and Markov chain evolution method of the present invention.
Detailed Description
The invention is further described below with reference to the figures and examples.
FIG. 1 is a block diagram of a hyper-heuristic Markov chain evolution method of the present invention.
Referring to fig. 1, the method for designing the multi-parameter operating condition of the vehicle based on the hyperheuristic mahalanobis chain evolution provided by the present invention includes the following steps:
step S1: a plurality of strategy factors meeting Markov property are designed by defining strategy boundary variables based on the operation strategy of the Markov chain evolution method, and the specific process comprises steps S11 to S14.
Step S11: the method comprises the steps of setting speed, acceleration, road gradient step length and maximum value based on the actual collected data of the vehicle and including parameters such as speed, acceleration and road gradient, wherein delta v is 0.5m/s, and delta a is 0.1m/s2，Δg＝1％，vmin＝0m/s，vmax＝35m/s，amin＝-2m/s2，amax＝2m/s2，gmin-10% and gmaxDividing each parameter state, counting a three-parameter state transition probability matrix P according to a formula (3), and generating a target working condition length L by using a Markov chain stochastic simulation method, wherein the starting state and the stopping state are idle candidate working conditions; it is composed ofWherein L is 1800 s;
Figure BDA0002196130500000071
in the formula:
Figure BDA0002196130500000072
is the number of one-dimensional space states, Ni′j′Is a one-dimensional space state si′To a one-dimensional space state sjNumber of transfers of `, Ni′Is a one-dimensional space state si′The sum of the number of transitions to other states, pi′j′Is a one-dimensional space state si′To a one-dimensional space state sj′S is a space state set;
state s at time wwCalculated according to equation (4)
sw＝mw+(uw-1)×M+(hw-1)×M×N (4)
In the formula: m iswSpeed state at time w, uwAcceleration state at time w, hwThe gradient state at the moment w, the number of M-speed states and the number of N-acceleration states are calculated according to the formulas (5) to (9).
Figure BDA0002196130500000073
Figure BDA0002196130500000074
Figure BDA0002196130500000075
Figure BDA0002196130500000076
Figure BDA0002196130500000077
Step S12: based on the three-parameter state transition probability matrix P counted in the step S11, when any two working condition sequences are designed to have equal-length cross section exchange, a cross operator strategy that an individual sequence must meet the Markov chain state transition relation is generated, and a process of randomly selecting a working condition sequence to replace the working condition sequence to be mutated based on the candidate working condition in the step S11 is used as a mutation operator strategy;
step S13: defining the state position of any two working condition sequences as a strategy boundary variable R, and setting a strategy boundary variable vector R as [ R ═ R1,R2,...,Ri,...,Rj,...,Rn]，i,j∈[1,n]And R isi≠RjN is different R numbers; wherein R is [10,20,50,100,150,200 ]]N is 6;
step S14: dividing all equal-length exchangeable segments of the two operator strategies based on Markov property in the step S12 into a plurality of different n +1 groups according to the vector R in the step S13, wherein the equal-length exchangeable segments of the n +1 groups generate n +1 equal-length cross operator strategies, namely n +1 equal-length strategy factors capable of playing local search and global search capacities in the evolution process, and then keeping the length deviation at a certain threshold T according to the design working conditionrWithin-range and non-equal length exchangeable segments of the vector R, n +1 group generate n +1 non-equal length policy factors; finally, the mutation operator strategy including the working condition state updating capability has 2n +3 strategy factors; i.e. corresponding to 15 strategy factors, respectively, the cross section is [1:10 ]]s，[10:20]s，[20:50]s，[50:100]s，[100:150]s，[150:200]s，[>200]Equal length cross strategy factor of s interval, and cross section is [1:10 ]]s，[10:20]s，[20:50]s，[50:100]s，[100:150]s，[150:200]s，[>200]s interval non-equal length cross strategy factor and complete variation strategy factor, T r10 percent;
step S15: based on the 2n +3 policy factors obtained in step S14, each policy factor ratio p is setrAnd randomly composing a strategy factor sequence Y (theta) with the length t [ theta ]1,θ2,…,θi,...,θt]Where t > (2n +3), θ ∈ [ θ [ ]e1,θe2,...,θe(n+1),θne1,θne2,...,θne(n+1),θm]，θe1,θe2,...,θe(n+1)Corresponding to an equal length cross policy factor, θne1,θne2,...,θne(n+1)Corresponding to a non-equal length cross policy factor, θmIs a complete mutation operator; simultaneously constructing a working condition sequence population Pop with the population size of t and randomly composed of working condition state sequences(2){X}＝[X1；X2；...；Xt]Wherein X is a working condition state sequence, and the initial population Pop is in the working condition sequence(2)Setting an evaluation index and a relative deviation threshold T of an operation condition under the condition of { X }rAnd calculating an initial function value F { X } - [ F (X) of the working condition sequence population by means of a satisfaction function criterion1),F(X2),...,F(Xt)]And obtaining the optimal function value F (X) of the population of the working condition sequence by the formula (10)*) And optimum operating mode sequence X*(ii) a Wherein p isrIs (1:1:1:2:3:3:3: 2), t is 30;
F(X*)＝min(F{X}) (10)
wherein the set evaluation index includes: the number of the idle speed time proportion, the acceleration time proportion, the uniform speed time proportion, the deceleration time proportion, the average speed, the average running speed, the running speed standard deviation, and the probability distribution correlation coefficients of the positive acceleration kinetic energy, the average climbing gradient, the average descending gradient, the speed and the acceleration in unit distance are 11.
Step S2: and designing a strategy factor and working condition sequence distribution mechanism and an updating mechanism, and further designing an evaluation function based on expected operation working conditions, wherein the specific process comprises the steps from S21 to S23.
Step S21: under the current k-th strategy factor sequence Y (theta), the function value F { X } of the condition sequence individuals in the step S15 is subjected to roulette operation, and a new condition sequence individual-ordered population is generated
Figure BDA0002196130500000091
And correspondingly distributing the working condition sequences in the population to strategy factors in a strategy factor sequence Y (theta) one by one, performing strategy operation, and generating working condition sequence filial generationsPopulation
Figure BDA0002196130500000092
Calculating working condition sequence population
Figure BDA0002196130500000093
Function value of
Figure BDA0002196130500000094
Step S22: updating population Pop(2)X and function values F X, i.e. the offspring population
Figure BDA0002196130500000095
Sum function value
Figure BDA0002196130500000096
Respectively replacing parent population Pop(2){ X } and a population function value F { X }; acquiring the filial population of the working condition sequence according to the formula (11)
Figure BDA0002196130500000097
Best function value of
Figure BDA0002196130500000098
And optimal sequence
Figure BDA0002196130500000099
Figure BDA00021961305000000910
When in use
Figure BDA00021961305000000911
By mixing
Figure BDA00021961305000000912
Is given to X*Updating the optimal operating regime sequence X*By passing through
Figure BDA00021961305000000913
Is given to F (X)*) Updating the optimum function value F (X)*) When is coming into contact with
Figure BDA00021961305000000914
When it is, X is*Assigned to population Pop(2){ X } in
Figure BDA00021961305000000915
In that
Figure BDA00021961305000000916
Will correspond to the function value
Figure BDA00021961305000000917
Is given to function value F (X)*) At the same position, to realize the optimum operating condition sequence X*And the optimum function value F (X)*) Updating of (1);
step S23: outputting the optimal function value F (X) under the current k strategy sequence Y (theta)*)。
Step S3: adopting a traditional genetic algorithm as a high-level control algorithm of a hyperheuristic framework, evolving and iterating a strategy factor population consisting of strategy factor sequences, and finally obtaining the operation condition of an expected evaluation value; the specific process includes steps S31 through S37.
Step S31: randomly forming a size K based on the strategy factor sequence of step S15maxInitial policy factor sequence population of
Figure BDA0002196130500000101
Calculating individual function values of policy factor sequence based on step S2
Figure BDA0002196130500000102
Wherein KmaxIs 24;
step S32: according to the strategy factor sequence function value F { Y }, according to the default elite probability pdSelecting the best K in the current populationmax×pdCarrying out traditional operator selection operation on the sequence individuals;
step S33: according toDefault cross probability pcIn (K)max-Kmax×pd) Selection from one of the remaining individuals (K)max-Kmax×pd)×pcThe individual sequences are subjected to random two-point crossing operation to generate (K)max-Kmax×pd)×pcA plurality of crossing individuals;
step S34: for the rest Kmax-Kmax×pd-(Kmax-Kmax×pd)×pcPerforming single point mutation operation on the individual to generate Kmax-Kmax×pd-(Kmax-Kmax×pd)×pc(ii) individual variants;
step S35: forming a new strategy factor sequence population by the elite individuals in the step S32, the crossover individuals in the step S33 and the variant individuals in the step S34;
step S36: the best evaluation value F (Y) is judged and output*) Whether the desired threshold T has been reachedrRange, if F (Y)*)≤TrAt this time, F (Y)*)＝F(X*) Wherein Y is*Executing step S37 for the optimal strategy sequence combination, otherwise, returning to step S32;
step S37: the optimal working condition state sequence X*And decoding and outputting the multi-parameter expected operation condition time sequence.
FIGS. 2-4 are time series of speed, acceleration and road grade for a test output expected operating condition of the present invention; FIG. 5 is a histogram of the average number distribution of policy factors in the population when the policy factors are evolved to randomly form 10 policy factor populations until the expected operating condition is output, showing the characteristic of the invention that adaptive combination of policy factors can be realized; FIG. 6 is a histogram of the run time of 10 design tests of the hyper-heuristic Markov chain evolution method and Markov chain evolution method of the present invention. Through statistics, compared with a working condition design method based on Markov chain evolution, the working efficiency of the method is improved by 63.52%. In addition, the functions of each part in the meta-heuristic architecture are easy to distinguish and package, the portability is strong, and support is provided for the operation and the use of automobile engineers.
Although the present invention has been described with respect to the preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (3)

1. A method for designing the multi-parameter operation condition of an automobile based on the meta-heuristic Mahalanobis chain evolution is characterized by comprising the following steps:
s1) selecting a plurality of strategy factors meeting Markov property by defining strategy boundary variables based on the operation strategy of the Markov chain evolution method;
s2) designing a strategy factor and a distribution mechanism and an updating mechanism of the working condition sequence, and further generating an evaluation function based on the expected operation working condition;
s3) adopting a traditional genetic algorithm as a high-level control algorithm of a hyper-heuristic architecture, evolving and iterating a strategy factor population consisting of strategy factor sequences, and finally obtaining the operation condition of the expected evaluation value;
the step S1 includes:
step S11: setting the step length of each parameter based on the actual data collected by the vehicle, dividing the states of each parameter, counting a multi-parameter state transition probability matrix P, and generating a target working condition length L by using a Markov chain random simulation method, wherein the starting and stopping states are candidate working conditions of idle speed;
step S12: based on the multi-parameter state transition probability matrix counted in the step S11, when any two working condition sequences are designed to have equal-length cross section exchange, a cross operator strategy that an individual sequence must meet the Markov chain state transition relation is generated, and a process of randomly selecting a working condition sequence to replace the working condition sequence to be mutated based on the candidate working condition in the step S11 is used as a mutation operator strategy;
step S13: defining the state position of any two working condition sequences as a strategy boundary variable R, and setting a strategy boundary variable vector R as [ R ═ R1,R2,...,Ri,...,Rj,...,Rn]，i,j∈[1,n]And R isi≠RjN is different R numbers;
step S14: according to the vector R of the step S13, firstly dividing all equal-length exchangeable segments of the two operator strategies based on Markov property in the step S12 into a plurality of different n +1 groups, generating n +1 equal-length cross operator strategies by the equal-length exchangeable segments of the n +1 groups, namely n +1 equal-length strategy factors capable of playing local search and global search capacities in the evolution process, and then keeping the length deviation at a certain threshold T according to the design working conditionrThe non-isometric exchangeable segments of the in-range sum vector R, n +1 group generate n +1 non-isometric strategy factors; finally, the mutation operator strategy including the working condition state updating capability has 2n +3 strategy factors;
step S15: based on the 2n +3 policy factors obtained in step S14, each policy factor ratio p is setrAnd randomly composing a strategy factor sequence Y (theta) with the length t [ theta ]1,θ2,...,θi,...,θt]Where t > (2n +3), θ ∈ [ θ [ ]e1,θe2,...,θe(n+1),θne1,θne2,...,θne(n+1),θm]，θe1,θe2,...,θe(n+1)Corresponding to an equal length cross policy factor, θne1,θne2,...,θne(n+1)Corresponding to a non-equal length cross policy factor, θmIs a complete mutation operator; simultaneously constructing a working condition sequence population Pop with the population size of t and randomly composed of working condition state sequences(2){X}＝[X1；X2；...；Xt]Wherein X is a working condition state sequence, and the initial population Pop is in the working condition sequence(2)Setting an evaluation index and a relative deviation threshold T of an operation condition under the condition of { X }rAnd calculating an initial function value F { X } - [ F (X) of the working condition sequence population by means of a satisfaction criterion1),F(X2),...,F(Xt)]And obtaining the optimal function value F (X) of the population of the working condition sequence by the formula (1)*) And optimum operating mode sequence X*；
F(X*)＝min(F{X}) (1)；
The step S2 includes:
step S21: under the current k-th strategy factor sequence Y (theta), the function value F { X } of the condition sequence individuals in the step S15 is subjected to roulette operation, and a new condition sequence individual-ordered population is generated
Figure FDA0002521441310000021
And correspondingly distributing the working condition sequences in the population to strategy factors in a strategy factor sequence Y (theta) one by one, performing strategy operation, and generating working condition sequence child population
Figure FDA0002521441310000022
Calculating working condition sequence population
Figure FDA0002521441310000023
Function value of
Figure FDA0002521441310000024
Step S22: updating population Pop(2)X and function values F X, to group the offspring
Figure FDA0002521441310000025
Sum function value
Figure FDA0002521441310000026
Respectively replacing parent population Pop(2){ X } and a population function value F { X }; acquiring a working condition sequence filial generation population according to a formula (2)
Figure FDA0002521441310000031
Best function value of
Figure FDA0002521441310000032
And optimal sequence
Figure FDA0002521441310000033
Figure FDA0002521441310000034
When in use
Figure FDA0002521441310000035
By mixing
Figure FDA0002521441310000036
Is given to X*Updating the optimal operating regime sequence X*By passing through
Figure FDA0002521441310000037
Is given to F (X)*) Updating the optimum function value F (X)*)；
When in use
Figure FDA0002521441310000038
When it is, X is*Assigned to population Pop(2){ X } in
Figure FDA0002521441310000039
In that
Figure FDA00025214413100000310
Position will correspond to function value
Figure FDA00025214413100000311
Is given to function value F (X)*) To achieve the optimal working condition sequence X*And the optimum function value F (X)*) Updating of (1);
step S23: outputting the optimal function value F (X) under the current k-th strategy sequence Y (theta)*)；
The step S3 includes:
step S31: randomly forming a size K based on the strategy factor sequence of step S15maxInitial policy factor sequence population of
Figure FDA00025214413100000312
Calculating individual function values of strategy factor sequence
Figure FDA00025214413100000313
Step S32: according to the strategy factor sequence function value F { Y }, according to the default elite probability pdSelecting the best K in the current populationmax×pdCarrying out traditional operator selection operation on the sequence individuals;
step S33: according to a default cross probability pcIn (K)max-Kmax×pd) Selection from one of the remaining individuals (K)max-Kmax×pd)×pcThe individual sequences are subjected to random two-point crossing operation to generate (K)max-Kmax×pd)×pcA plurality of crossing individuals;
step S34: for the rest Kmax-Kmax×pd-(Kmax-Kmax×pd)×pcPerforming single point mutation operation on the individual to generate Kmax-Kmax×pd-(Kmax-Kmax×pd)×pc(ii) individual variants;
step S35: forming a new strategy factor sequence population by the elite individuals in the step S32, the crossover individuals in the step S33 and the variant individuals in the step S34;
step S36: the best evaluation value F (Y) is judged and output*) Whether the desired threshold T has been reachedrRange, if F (Y)*)≤TrAt this time, F (Y)*)＝F(X*) Wherein Y is*Executing step S37 for the optimal strategy sequence combination, otherwise, returning to step S32;
step S37: the optimal working condition state sequence X*And decoding and outputting the multi-parameter expected operation condition time sequence.
2. The method for designing the multi-parameter operating condition of the automobile based on the heuristic mahalanobis link evolution of claim 1, wherein the parameters in the step S11 include speed, acceleration and road gradient, and the value of L is 1800S; the value of n in the step S13Is 6, R takes the value of [10,20,50,100,150,200](ii) a T in the step S13rThe value of (a) is 10%; p in said step S13rThe value of (1:1:1:2:3:3:3:1:1: 2:3:3: 2) and the value of t is 30.
3. The method for designing the multi-parameter operation condition of the automobile based on the meta-heuristic mahalanobis chain evolution as claimed in claim 1, wherein K in the step S31maxIs 24.
CN201910848560.5A 2019-09-09 2019-09-09 Automobile multi-parameter operation condition design method based on heuristic Markov chain evolution Active CN110826145B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910848560.5A CN110826145B (en) 2019-09-09 2019-09-09 Automobile multi-parameter operation condition design method based on heuristic Markov chain evolution

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910848560.5A CN110826145B (en) 2019-09-09 2019-09-09 Automobile multi-parameter operation condition design method based on heuristic Markov chain evolution

Publications (2)

Publication Number Publication Date
CN110826145A CN110826145A (en) 2020-02-21
CN110826145B true CN110826145B (en) 2020-08-28

Family

ID=69547975

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910848560.5A Active CN110826145B (en) 2019-09-09 2019-09-09 Automobile multi-parameter operation condition design method based on heuristic Markov chain evolution

Country Status (1)

Country Link
CN (1) CN110826145B (en)

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4873825B2 (en) * 2000-08-04 2012-02-08 リサーチ ディベロップメント ファンデーション Urocortin protein and its use
KR101318250B1 (en) * 2012-02-10 2013-10-15 (주)프람트테크놀로지 Inference query processing using hyper cube
CN103744733B (en) * 2013-07-03 2017-03-22 中国人民解放军国防科学技术大学 Method for calling and configuring imaging satellite resources
US9621275B2 (en) * 2014-07-01 2017-04-11 Mitsubishi Electric Research Laboratories, Inc. Method for generating constant modulus multi-dimensional modulations for coherent optical communications
CN104657221B (en) * 2015-03-12 2019-03-22 广东石油化工学院 The more queue flood peak staggered regulation models and method of task based access control classification in a kind of cloud computing
KR102599073B1 (en) * 2016-05-09 2023-11-10 스트롱 포스 아이오티 포트폴리오 2016, 엘엘씨 Methods and systems for the industrial internet of things
CN107122573B (en) * 2017-06-26 2020-04-17 吉林大学 Automobile operation condition design method based on Markov chain evolution
CN107908853B (en) * 2017-11-10 2020-07-31 吉林大学 Automobile operation condition design method based on prior information and big data
CN109118023B (en) * 2018-09-21 2022-03-01 北京交通大学 Public transport network optimization method
CN109460862B (en) * 2018-10-22 2021-04-27 郑州大学 Method for solving multi-objective optimization problem based on MAB (multi-object-based) hyperheuristic algorithm
CN109948781A (en) * 2019-03-21 2019-06-28 中国人民解放军国防科技大学 Continuous action online learning control method and system for automatic driving vehicle

Also Published As

Publication number Publication date
CN110826145A (en) 2020-02-21

Similar Documents

Publication Publication Date Title
Wu et al. Multiobjective optimization of HEV fuel economy and emissions using the self-adaptive differential evolution algorithm
CN108846472A (en) A kind of optimization method of Adaptive Genetic Particle Swarm Mixed Algorithm
CN106447024A (en) Particle swarm improved algorithm based on chaotic backward learning
CN108053119A (en) A kind of Modified particle swarm optimization algorithm for solving zero-waiting Flow Shop Scheduling
Malikopoulos A multiobjective optimization framework for online stochastic optimal control in hybrid electric vehicles
Zhang et al. High-efficiency driving cycle generation using a Markov chain evolution algorithm
CN101907869B (en) Method for controlling a vehicle
CN107122573B (en) Automobile operation condition design method based on Markov chain evolution
Malikopoulos Stochastic optimal control for series hybrid electric vehicles
CN111907342A (en) Working condition identification control method of pure electric vehicle
Guo et al. Predictive energy management strategy of dual-mode hybrid electric vehicles combining dynamic coordination control and simultaneous power distribution
CN109784497A (en) Based on the method for calculating the AI auto-building model that figure is evolved
Montazeri-Gh et al. Driving condition recognition for genetic-fuzzy HEV control
CN110826145B (en) Automobile multi-parameter operation condition design method based on heuristic Markov chain evolution
Liessner et al. Simultaneous electric powertrain hardware and energy management optimization of a hybrid electric vehicle using deep reinforcement learning and Bayesian optimization
Zhang et al. Self-adaptive hyper-heuristic Markov chain evolution for generating vehicle multi-parameter driving cycles
CN106548301A (en) Power consumer clustering method and device
CN104598760B (en) A kind of energy-storage battery typical condition formulating method
CN104200271A (en) Multi-objective optimization algorithm for engine
CN116843500A (en) Charging station planning method, neural network model training method, device and equipment
CN114745288B (en) Complex network survivability model quantization method based on block chain and dynamic weighting
CN110807290B (en) Road gradient-containing automobile operation condition design method based on self-adaptive Markov chain evolution
Pedram Pourhashemi et al. Application of the fuel-optimal energy management in design study of a parallel hybrid electric vehicle
Boehme et al. Multi-objective optimal design of parallel plug-in hybrid powertrain configurations with respect to fuel consumption and driving performance
Gonçalves et al. Solving economic load dispatch problem by natural computing intelligent systems

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