CN106295083B - A kind of wheel based on NSGA-II algorithm repairs policy optimization method to rotation - Google Patents
A kind of wheel based on NSGA-II algorithm repairs policy optimization method to rotation Download PDFInfo
- Publication number
- CN106295083B CN106295083B CN201610866728.1A CN201610866728A CN106295083B CN 106295083 B CN106295083 B CN 106295083B CN 201610866728 A CN201610866728 A CN 201610866728A CN 106295083 B CN106295083 B CN 106295083B
- Authority
- CN
- China
- Prior art keywords
- rotation
- wheel
- population
- distance
- repaired
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000008439 repair process Effects 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000005457 optimization Methods 0.000 title claims abstract description 24
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 15
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 239000010410 layer Substances 0.000 claims description 16
- 239000002245 particle Substances 0.000 claims description 15
- 238000005299 abrasion Methods 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 239000002356 single layer Substances 0.000 claims description 3
- 230000002068 genetic effect Effects 0.000 abstract description 4
- 230000008859 change Effects 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012913 prioritisation Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Classifications
-
- 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
-
- 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)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Geometry (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Optimization (AREA)
- Artificial Intelligence (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Computer Hardware Design (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of wheels based on NSGA-II algorithm to repair policy optimization method to rotation, this method repairs strategy to rotation to take turns at present as object, it is that target establishes Model for Multi-Objective Optimization that it is most short, which to repair number, to service life longest and rotation with wheel, for wheel to rotation repair policy mandates rotation repair number as far as possible lack this feature, change local density's calculation method of traditional NSGA-II, genetic operator is had changed simultaneously, realizes effective optimization that wheel repairs strategy to rotation.
Description
Technical field
The present invention relates to a kind of wheels based on improved NSGA-II algorithm to repair policy optimization method to rotation, belongs to calculation
Method optimizes field.
Background technique
Since 21 century, China railways net rapidly develop, State Council, China annual data show 2012 with
China railways run 98000,103000 kilometers altogether within 2013.Plan furthermore according to China 13, national iron during 13
Road construction object is 2.9 ten thousand kilometers to 150,000 kilometers of newly built railway, wherein 1.1 ten thousand kilometers of high-speed rail.Although China railways are rapidly sent out
Exhibition, China railways maintenance be still within fixation repair the stage, wheel to rotation repair still by it is veteran wheel to Maintenance Engineer into
Row management.
Domestic and foreign scholars are for wheel to the research abundant of rotation repair row.Before 20th century, domestic and foreign scholars are relied primarily on
Many-body dynamics, friction mechanics, the physical characteristics such as car body track dynamics are taken turns analysis to wear characteristics and plan is repaired in rotation
Optimization slightly；After 20 worlds, with the continuous development of computer science and technology, sight is turned to data analysis etc. by a small number of scholars
Computer correlation technique has scholar to study the rotation problem of repairing using the methods of emulation, numerical analysis wheel, but these grind
Study carefully and all the service life of wheel pair and rotation are repaired number and treated in isolation, research quantity is few, still relies primarily on the correlative of wheel pair
Characteristic is managed, generated abundant data during train operation is not made full use of.
The Multiple factors such as state and temperature, running route, the humidity of wheel pair are related, and China region is wide, various regions topography
Style and features, weather form are all different, thus, it is difficult that China's Railway wheelset property is carried out to analyze from physical characteristic to define.
Wheel, as information content resource the most abundant, contains wheel and directly affects to locating environment to it to data, base of the present invention
In data resource abundant, strategy is repaired to rotation for wheel establishes while considering to take turns and number are repaired to revolving to service life and wheel more
Objective optimization model, meanwhile, number as far as possible less this requirement optimization NSGA-II is repaired in rotation during being repaired according to wheel to rotation, is realized
Effective wheel repairs policy optimization to rotation, and China is promoted to change from fixed repair to status maintenance.
Summary of the invention
Present invention seek to address that the problem of existing wheel repairs dependence experience to rotation.According to wheel to using the time as long as possible and take turns
Number feature as few as possible is repaired to rotation, is analyzed for the characteristic of single wheel pair, is proposed the NSGA-II algorithm of optimization, have
Solve the problems, such as that wheel repairs rotation to effect, number is repaired in the rotation that wheel pair is reduced while increasing and taking turns to service life.
The present invention uses following technical scheme to solve above-mentioned technical problem:
A kind of wheel based on NSGA-II algorithm repairs policy optimization method to rotation, first with wheel to using time longest and rotation
Repairing number is at least target, revolves the wheel repaired according to specific needs to constraint condition is determined, establishes Model for Multi-Objective Optimization；Then it adopts
It solves to obtain to take turns with NSGA-II algorithm and strategy is repaired to rotation.
The bound for objective function is taken turns to determined by historical data during operation to abrasion according to wheel
Curve and different wheel are to determining.
The NSGA-II algorithm comprising steps of
Step (1): initial population is generated as parent population, and calculates the adaptive value of each particle in parent population；Parent
Each of population particle represents one group of rotation and repairs strategy；By abrasion model to each group of rotation repair strategy carry out rotation repair number and
Using the calculating of time, using wheel to using time longest, rotation to repair number at least as objective function, rotation is repaired according to specific needs
Wheel is to determining constraint condition；
Step (2): number is repaired according to rotation, parent population is divided into the classification that number level is repaired in different rotations；With 50% probability
It is repaired from rotation and selects particle in the least classification of number, and other particles of cross selection are constituted novel species in remaining classification with equiprobability
Group；And obtained new population is intersected, makes a variation to obtain progeny population；
Step (3): progeny population is merged with parent population, is carried out quick non-dominated ranking and is utilized gaussian kernel function meter
Calculate its crowding distance；The calculation method of specific gaussian kernel function is as follows:
Wherein, dcIt is value set by user, dijIt is the distance of j point-to-point i；
Step (4): according to step (3) quickly non-dominated ranking result and using gaussian kernel function calculate it is crowded away from
From the next-generation parent population of generation；
Step (5) judges whether to reach maximum number of iterations, if so, going to step (8)；Otherwise, return step (3),
The number of iterations adds 1；
Step (6): terminating, and the Pareto forward position of the parent population of new generation of generation is required disaggregation, to be taken turns
Strategy is repaired to rotation.
It is specially to repair number according to rotation entire population is divided into A, B, C three classes to parent kind heap sort in the step (2),
Then with 0.5 probability from A classification, and select to be formed new from B itself, C itself, A and B, B and C, A and C respectively with equiprobability
Population.
The method that parent population and progeny population after merging are calculated in the step (3) can also use Euclidean distance,
Chebyshev's distance, manhatton distance or Minkowski Distance are calculated to realize to calculate original single layer distance and be extended to entirely
The distance of office calculates.
The step (4) is specially to choose Pareto first layer first, then the second layer, and so on；Reach kth layer
When, if kth layer, which is all added, will be greater than open ended Population Size, the crowding distance according to population preferentially chooses crowding distance
Big particle is added in next-generation parent population.
It is that the present invention reaches that problem is repaired to rotation the utility model has the advantages that being 1. directed to take turns, while wheel is considered to service life and wheel pair
Two aspect of number is repaired in rotation, takes full advantage of wheel to the mass data generated during operation, efficiently solves wheel pair at present
The status of optimisation strategy missing is repaired in rotation；2. the improved method of crowding distance can be in improved NSGA-II proposed by the present invention
Expand the range of feasible zone, and expansion correlation disaggregation effectively to a certain extent；3. NSGA-II proposed by the present invention passes through needle
Disaggregation required for problem is effectively expanded by the improvement of the genetic operator of the problem, and solution is maintained at required as far as possible
In the region wanted.
Detailed description of the invention
Fig. 1 is selection operation process；
Fig. 2 is improved NSGA-II flow chart.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing, to the present invention into
Row is further described.The specific embodiments described herein are merely illustrative of the present invention, is not intended to limit the present invention.
Since domestic foreign steamer traditional at present repairs mainly with the physical characteristic of wheel pair rotation, many-body dynamics, car body are utilized
Based on dynamics of orbits and the methods of friction mechanics, numerical analysis, analogue simulation, it may appear that optimization object is excessively single, and does not have
The problems such as targeted.This algorithm relies on a large amount of wheel to data, in foundation with wheel to using time longest and rotation to repair number
On the basis of the minimum Model for Multi-Objective Optimization for target, in establishing the algorithm implementation procedure after model, mass data can quilt
For being emulated, and then is taken turns required for acquiring and strategy is repaired to rotation, and simulation process needs a large amount of historical data eventually
Determine the wear profile of wheel pair.Two objective functions of multi-objective Model are denoted as service life longest, and it is minimum that number is repaired in rotation, and its
Relevant constraint will then be limited by specific wheel pair, and objective function is as follows:
Wherein, Y is year, and F is that number is repaired in rotation, and D is that wheel footpath will be maintained between 1150~1250, and T is flange thickness,
It is maintained between 28 and 34, the value before the rotation that rotation is repaired each time is repaired is less than the value after rotation is repaired.
Moreover, the condition difference of the outer dimension of the wheel pair of different model, meanwhile, it is not the abrasion of the same wheel pair
Curve wear away parameter also can difference, thus specific model is needed according to specific wheel to establishing.
The present invention repairs problem to rotation for existing wheel, proposes the NSGA-II algorithm of optimization.On the one hand, due to traditional
NSGA-II only focuses on the range averaging of the point adjacent thereto on Pareto level locating for the point when calculating the crowding distance of solution
Value, can not effectively illustrate the global distribution character of the point.Thus the present invention is in the step of NSGA-II calculates crowding distance
The traditional calculation method changed, substitutes crowding distance using gaussian kernel function.The calculation method of gaussian kernel function
It is as follows:
Wherein, dcIt is threshold value set by user, dijIt is the distance of j point-to-point i, i, j indicate the particle in population；
Pervious calculating is limited to one layer, and the present invention calculates distance and is put into all individuals, other ratios
Such as Euclidean distance, Chebyshev's distance, manhatton distance, Minkowski Distance etc. also be may be implemented original single layer distance
It calculates and is extended to global distance and calculates, calculate the global density characteristic of any solution using the gaussian kernel function present invention, and with
The second characteristic when this is as NSGA-II non-dominated ranking.At the same time, during non-dominated ranking, all solutions can be divided
The layer etc. that is segmented into the 1st, 2,3 ..., wherein first layer is Pareto forward position.In traditional NSGA-II, all solutions can be according to its domination
Property be divided, thus, in NSGA-II algorithm proposed by the invention, each solution is gathered around there are two characteristic, the layer that NSGA-II is calculated
The crowding distance that secondary and gaussian kernel function calculates.On this basis, in non-dominated ranking of the invention, in front end level
Solution can be selected in next-generation parent first, and the gaussian kernel function crowding distance calculated on same layer is big preferentially to be entered compared with other solutions
Select parent.This method has measured each from whole angle and has solved crowding distance, made as one of prioritization scheme of the invention
Entire efficient solution set is effectively expanded, and avoids recessed area optimal.
On the other hand, policy mandates rotation is repaired to rotation due to wheel to repair number as few as possible, and simultaneous wheels to the greatest extent may be used to service life
Can be long, it is all to sacrifice wheel footpath to reply flange thickness, and scrap once wheel footpath reaches the limit of wheel, thus subtract that rotation each time, which is repaired,
Number is repaired in few upper rotation can reduce the wheel footpath lost because rotation is repaired to a certain extent, such wheel footpath have more spaces be by
It expends in the general wear of train operation.In addition, the service life of wheel pair is also main in the multi-objective Model that we are established
One of Consideration is wanted, thus, required solution must be one and obtain in taking as an elective course number and service life and preferably do effect
Combination solution.So the present invention is directed to the particularity of the application scenarios, NSGA-II genetic operator is improved.Firstly, of the invention
Feasible solution classification to obtaining, can will be set as the first kind less than 10 times with initial setting, because normal condition backspin repairs number
It is 3~5 times (repairing the limit every time), solution required for us should be fallen within 10 as far as possible, while excessive take as an elective course number
Certain economic loss is also brought along, rotation is found and repairs the few classification of number, selected during being selected with 50% probability
Select rotation repair the few feasible solution of number, and with equiprobability in remaining classification cross selection, by taking Fig. 1 as an example, parent population is divided into
A, B, C three classes, A, B, C repair number according to rotation and classify, thus take as an elective course in part of the number less than 10 and selecting in A class
It takes 50% in journey, while the diversity in order to guarantee solution, does not omit outstanding in inferior solution as a result, also combining that remaining is all kinds of,
Remaining equal proportion between any two is intersected, to keep the diversity of population.Then when carrying out selection operation, classification A will be with
50% probability is selected, and remaining part then has B itself, C itself, A and B, A and C, B and C to select with equiprobability respectively, this
Sample obtains new parent population.Optimization method of this method as genetic operator of the present invention, required for capable of effectively expanding
Feasible solution.
As shown in Fig. 2, the present invention is directed to while optimizing the multiple-objection optimization mould that wheel repairs number to service life and wheel to rotation
The key step of type are as follows:
(1) it initializes: generating initial population as parent population；Each of parent population particle represents one group of rotation and repairs
Strategy；By abrasion model to each group of rotation repair strategy carry out rotation repair number and using the time calculating, with use time longest,
Rotation repairs number and is at least used as objective function, and front wheel felloe thickness is repaired in relevant wheel footpath, the limits value of flange thickness and rotation and want small
Yu Xuanxiu rear rim thickness is as restrictive condition；
(2) adaptive value of each particle in parent population is calculated；
(3) according to according to adaptive value, i.e., repairing number according to rotation and entire population being divided into A, B, C three classes shown in Fig. 1, then with
0.5 probability forms novel species from B itself, C itself, A and B, B and C, A and C selection particle respectively from A classification, and with equiprobability
Group；
(4) intersected using the new population that selection obtains in (3), made a variation, obtain progeny population；
(5) progeny population is merged with parent population, carries out the first attribute of population after non-dominated ranking is merged；
(6) the crowding distance of population after merging, the second attribute of population after being merged are calculated using gaussian kernel function；
(7) comprehensively consider feasible zone the first attribute and the second attribute (selection Pareto first layer first, then second
Layer, and so on, when reaching kth layer, it will be greater than open ended parent if being all added, at this time according to the second attribute crowding distance
Carry out particle selection), next-generation parent is selected, the number of iterations adds 1；
(8) judge whether that reaching maximum number of iterations otherwise, goes to (3) if so, going to (9)；
(9) terminate, the Pareto forward position of the parent population of new generation of generation is denoted as required disaggregation；Pareto forward position is a system
Column repair the combination of number using time and rotation, oneself desired solution can be chosen and (time longest or take as an elective course number by concentrating in solution
At least), it is to repair number with rotation according to selected service life to determine (by phase each time when program is run that strategy is repaired in rotation
Strategy record is closed in an array), obtained disaggregation and strategy is correspondingly that determining solution just can determine that plan is repaired in rotation
Slightly.
Claims (5)
1. a kind of wheel based on NSGA-II algorithm repairs policy optimization method to rotation, which is characterized in that first with wheel to using the time
It is at least target that number is repaired in longest and rotation, revolves the wheel repaired according to specific needs to constraint condition is determined, establishes multiple-objection optimization mould
Type；Then it solves to obtain to take turns using NSGA-II algorithm and strategy is repaired to rotation；The NSGA-II algorithm comprising steps of
Step (1): initial population is generated as parent population, and calculates the adaptive value of each particle in parent population；Parent population
Each of particle represent one group of rotation and repair strategy；By abrasion model to each group of rotation repair strategy carry out rotation repair number and use
The calculating of time revolves the wheel pair repaired using wheel to using time longest, rotation to repair number at least as objective function according to specific needs
Determine constraint condition；
Step (2): number is repaired according to rotation, parent population is divided into the classification that number level is repaired in different rotations；With 50% probability from rotation
It repairs and selects particle in the least classification of number, and other particles of cross selection are constituted new population in remaining classification with equiprobability；
And obtained new population is intersected, makes a variation to obtain progeny population；
Step (3): progeny population is merged with parent population, quick non-dominated ranking is carried out and calculates it using gaussian kernel function
Crowding distance；The calculation method of specific gaussian kernel function is as follows:
Wherein, dcIt is value set by user, dijIt is the distance of j point-to-point i；
Step (4): according to step (3) the quickly result of non-dominated ranking and the crowding distance life calculated using gaussian kernel function
At next-generation parent population；
Step (5): judge whether to reach maximum number of iterations, if so, going to step (8)；Otherwise, return step (3), iteration
Number adds 1；
Step (6): terminating, and the Pareto forward position of the parent population of new generation of generation is required disaggregation, to obtain taking turns to rotation
Repair strategy.
2. wheel according to claim 1 repairs policy optimization method to rotation, which is characterized in that the constraint item of the objective function
Part according to wheel to wheel determined by historical data during operation to wear profile and different wheel to determining.
3. wheel according to claim 1 repairs policy optimization method to rotation, which is characterized in that parent in the step (2)
Kind heap sort is specially to repair number according to rotation entire population is divided into A, B, C three classes, then with 0.5 probability from A classification, and
It selects to form new population from B itself, C itself, A and B, B and C, A and C respectively with equiprobability.
4. wheel according to claim 1 repairs policy optimization method to rotation, which is characterized in that calculate and close in the step (3)
The method of parent population and progeny population after and uses Euclidean distance, and Chebyshev's distance, manhatton distance or Min can husbands Si
Cardinal distance, which is realized to calculate with a distance from original single layer from calculating, is extended to global distance calculating.
5. wheel according to claim 1 repairs policy optimization method to rotation, which is characterized in that headed by the step (4) is specific
Pareto first layer is first chosen, then the second layer, and so on；When reaching kth layer, it will be greater than to hold if kth layer is all added
The Population Size received then preferentially is chosen the big particle of crowding distance according to the crowding distance of population and is added in next-generation parent population.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610866728.1A CN106295083B (en) | 2016-09-29 | 2016-09-29 | A kind of wheel based on NSGA-II algorithm repairs policy optimization method to rotation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610866728.1A CN106295083B (en) | 2016-09-29 | 2016-09-29 | A kind of wheel based on NSGA-II algorithm repairs policy optimization method to rotation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106295083A CN106295083A (en) | 2017-01-04 |
| CN106295083B true CN106295083B (en) | 2019-10-11 |
Family
ID=57716237
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610866728.1A Active CN106295083B (en) | 2016-09-29 | 2016-09-29 | A kind of wheel based on NSGA-II algorithm repairs policy optimization method to rotation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106295083B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107909194B (en) * | 2017-11-07 | 2021-07-13 | 电子科技大学 | System-level testability design multi-objective optimization method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708406A (en) * | 2012-05-10 | 2012-10-03 | 湖北省电力公司 | Scheduling graph optimizing method based on multi-target genetic algorithm |
| CN102831479A (en) * | 2012-07-29 | 2012-12-19 | 江苏大学 | NSGA-II (non-domination sequencing genetic algorithm) multi-objective optimization searching method |
| CN103489335A (en) * | 2013-09-16 | 2014-01-01 | 北京航空航天大学 | Flight conflict resolution method based on multi-objective optimization |
| CN103838929A (en) * | 2014-02-25 | 2014-06-04 | 中国计量学院 | Turning repair decision optimization method for rail transit vehicle wheel sets |
| CN104155931A (en) * | 2014-07-04 | 2014-11-19 | 广东工业大学 | NSGA-II-based integrated optimization method for tire mould processing and assembling |
| CN105550771A (en) * | 2015-12-09 | 2016-05-04 | 中冶南方工程技术有限公司 | Multi-objective optimization method of steelmaking-continuous casting production scheduling based on NSGA-II |
-
2016
- 2016-09-29 CN CN201610866728.1A patent/CN106295083B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708406A (en) * | 2012-05-10 | 2012-10-03 | 湖北省电力公司 | Scheduling graph optimizing method based on multi-target genetic algorithm |
| CN102831479A (en) * | 2012-07-29 | 2012-12-19 | 江苏大学 | NSGA-II (non-domination sequencing genetic algorithm) multi-objective optimization searching method |
| CN103489335A (en) * | 2013-09-16 | 2014-01-01 | 北京航空航天大学 | Flight conflict resolution method based on multi-objective optimization |
| CN103838929A (en) * | 2014-02-25 | 2014-06-04 | 中国计量学院 | Turning repair decision optimization method for rail transit vehicle wheel sets |
| CN104155931A (en) * | 2014-07-04 | 2014-11-19 | 广东工业大学 | NSGA-II-based integrated optimization method for tire mould processing and assembling |
| CN105550771A (en) * | 2015-12-09 | 2016-05-04 | 中冶南方工程技术有限公司 | Multi-objective optimization method of steelmaking-continuous casting production scheduling based on NSGA-II |
Non-Patent Citations (1)
| Title |
|---|
| 高速动车组车轮型面多目标优化镟修;张宝安 等;《同济大学学报（自然科学版）》;20130331;第41卷(第3期);第438-440页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106295083A (en) | 2017-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107168054B (en) | Multi-robot task allocation and path planning method | |
| CN102663100B (en) | Two-stage hybrid particle swarm optimization clustering method | |
| CN108090624B (en) | Urban ecological safety simulation and prediction method for improving cellular automaton | |
| CN106529674A (en) | Multiple-unmanned-aerial-vehicle cooperated multi-target distribution method | |
| CN104200387B (en) | A kind of land use zoning method for taking geographical entity semantic similarity into account | |
| Zhao et al. | Improved multiple-objective dynamic programming model for reservoir operation optimization | |
| CN110728421B (en) | Road network charging optimization method based on charging demand big data | |
| CN103294928A (en) | Combination forecasting method of carbon emission | |
| CN110533304B (en) | Power system load uncertainty analysis method | |
| CN110427690A (en) | A kind of method and device generating ATO rate curve based on global particle swarm algorithm | |
| CN104218571B (en) | A kind of running status appraisal procedure of wind power plant | |
| CN110826244A (en) | Conjugate gradient cellular automata method for simulating influence of rail transit on urban growth | |
| CN106295083B (en) | A kind of wheel based on NSGA-II algorithm repairs policy optimization method to rotation | |
| Zhu et al. | Green routing fuel saving opportunity assessment: A case study using large-scale real-world travel data | |
| CN106600055A (en) | Wind speed prediction method the basis of self excitation threshold autoregression model | |
| CN103353895A (en) | Pre-processing method of power distribution network line loss data | |
| CN101877034A (en) | Land utilization automatic division method | |
| Dai et al. | RETRACTED ARTICLE: Spatial interaction and network structure evolvement of cities in terms of China’s rail passenger flows | |
| CN109272151A (en) | A kind of vehicle path planning algorithm optimization method based on Spark | |
| CN111832599B (en) | Gas station prediction method based on machine learning random forest | |
| Bao et al. | Iterative modeling of wind turbine power curve based on least‐square B‐spline approximation | |
| CN110070208B (en) | Railway daily passenger capacity acquisition method based on data correction | |
| Wang et al. | Forecast of Electric Vehicle Ownership Based on MIFS-AdaBoost Model | |
| CN102043870A (en) | Method for evaluating and selecting optimal wiring route of electric locomotive | |
| Wu et al. | Economic Dependence Relationship and Spatial Stratified Heterogeneity in the Eastern Coastal Economic Belt of China |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |