CN108038538A - Multi-objective Evolutionary Algorithm based on intensified learning - Google Patents
Multi-objective Evolutionary Algorithm based on intensified learning Download PDFInfo
- Publication number
- CN108038538A CN108038538A CN201711279238.2A CN201711279238A CN108038538A CN 108038538 A CN108038538 A CN 108038538A CN 201711279238 A CN201711279238 A CN 201711279238A CN 108038538 A CN108038538 A CN 108038538A
- Authority
- CN
- China
- Prior art keywords
- mrow
- value
- population
- msup
- new
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Physiology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses the multi-objective Evolutionary Algorithm based on intensified learning, initial population is randomly generated from search space, obtained population is evaluated；Population to being unsatisfactory for end condition, new value is produced using the DEvariant operators and T operators of intensified learning selection, and the value of itself and neighborhood is intersected, and variation produces new explanation；The new explanation of generation, compared with the solution of original seed group, selection makes subproblem function meet the solution of optimal value, carrys out Population Regeneration；Using the new population of generation, 5 new dimension observation vector sum return value R are calculated, and then update the state of RL controllers；Judge whether to meet end condition, be unsatisfactory for, be constantly iterated calculating, until meeting end condition, terminate.The present invention solve thes problems, such as that MOEA/D is insensitive for T parameter regulations.
Description
Technical field
The present invention relates to science and field of engineering technology, more particularly to the multi-objective Evolutionary Algorithm based on intensified learning.
Background technology
In science and engineering field, there are substantial amounts of multi-objective optimization question (Multi-
ObjectiveOptimizationProblem, MOP), with single-object problem (Single-objective
Optimization Problem, SOP) different, the set that the optimal solution of MOP is made of so-called Pareto optimal solutions.Tradition
Multi-objective optimization algorithm includes weighting method, leash law, Objective Programming and Chebyshev approximation etc..These methods are all to turn MOP
SOP is changed to, the drawback is that needing sufficient priori, and is difficult to handle target noise, robustness is poor.Due to multiple target
The object function and constraint function of optimization problem are probably non-linear, non-differentiability or discontinuous, traditional mathematic programming methods
It is often less efficient, and the order that they give for weighted value or target is more sensitive.
2. evolution algorithm (EvolutionaryAlgorithm, EA) is a kind of random overall situation for simulating natural evolution process
Optimization method, EA search for the solution of problem by the way of information exchange between collective search and individual in population.Since EA is intrinsic
Concurrency, it is possible to find multiple Pareto optimal solutions in simulations.Compared with traditional algorithm, it the advantage is that：First, evolve
Search procedure has randomness, is not easy to be absorbed in local optimum；Secondly, EA has intrinsic concurrency, and can evolve searching at the same time
To multiple solutions, it is adapted to multi-objective optimization question；3rd, can handle it is discontinuous, the problems such as non-differentiability and non-convex Pareto forward positions,
Excessive priori is not required.
3. algorithm is dominant mechanism based on Pareto, using different fitness assignment strategies and selection mechanism；Take difference
Scheme keeps population diversity and avoids algorithm Premature Convergence so that the Pareto angle distribution of algorithm is uniform.
It is excellent due to multi-objective Evolutionary Algorithm 4. as multiple-objection optimization device a kind of efficient and with good robustness
Gesture, MOEA have been widely used in many fields of science and engineering, including control engineering, systems organization, production scheduling, number
According to excavation etc..
MOP is decomposed into the subproblem of N number of scalar by 5.MOEA/D.It by evolve one solution population come and meanwhile solve
All subproblems.For every generation population, population is the set of the optimal solution for each subproblem selected from all generations.Phase
The correlation degree of adjacent two subproblem keys is that the distance between the polymerizing factor vector by them is determined.For two adjacent sons
For problem, optimal solution should be closely similar.For each subproblem, simply with subproblem adjacent thereto
Information optimizes it.
MOEA/D has following characteristic：
MOEA/D provides a simple but effective method, that is, the method for decomposition is incorporated into multi-target evolution
In calculating.For usually in the decomposition method of Mathematical Planning field development, it can be really incorporated into EA, by using
MOEA/D frames solve the problems, such as MOP.Because MOEA/D algorithms are to optimize N scalars subproblem at the same time rather than directly ask MOP
Topic is solved as an entirety, then the fitness assignment and more for traditional MOEA algorithms for being not based on decomposing
The difficulty of sample control will be reduced in MOEA/D frames.
It is exactly insensitive for T parameter regulations 6. still, MOEA/D is there is deficiency, T is small not to have a range, and big do not have
There is depth, adaptive adjustment capability is poor.
The content of the invention
The shortcomings that it is an object of the invention to overcome the prior art, there is provided the multi-objective Evolutionary Algorithm based on intensified learning,
To solve above-mentioned technical problem.
The present invention uses following technical scheme to achieve the above object：
Multi-objective Evolutionary Algorithm based on intensified learning, includes the following steps：Step 1, generate at random from search space
Initial population；
Step 2, evaluate obtained population according to assessment level；
The optimum value for the object function that step 3, renewal search；
Step 4, the approximate solution Z using generation*Judgement is compared with end condition, satisfaction just terminates；To being unsatisfactory for end
The only population of condition, new value is produced using the DEvariant operators and T operators of intensified learning RL controllers selection, and to it
Intersected with the value of neighborhood, variation produces new explanation；
Step 5, the new explanation produced select the solution for making subproblem function meet optimal value compared with the solution of original seed group,
Carry out Population Regeneration；
Step 6, the new population using generation, calculate 5 new dimension observation vector sum return value R, and then update RL controls
The state of device, judges whether to meet end condition, is unsatisfactory for, and is constantly iterated calculating, until meeting end condition, terminates.
Preferably, the step 1 concretely comprises the following steps：
Euclidean distance between step 1.1, calculating any two weight vectors, searches T nearest power of each weight vectors
Weight vector, wherein T is the number of the weight vectors in each neighborhood, for each i=1 ..., N, makes Bi={ i1,…,iT,
λi 1,…λi TIt is λiT nearest weight vectors；
Step 1.2, establish an exterior population EP, for storing the non-domination solution searched for during optimal solution and found, just
Beginningization EP is sky；
Step 1.3, uniformly random collection generation makes object function F (X)=(f from search space1(x),f2(x),…,fi
(x)) solution of optimal value is taken as initial population, wherein i=1,2 ..., m；X is one group of decision vector, and x is independent variable；
Step 1.4, using Chebyshev's method, object function F (X) is resolved into N straton problems：Wherein,I-th subproblem it is adjacent
Relation is by all subproblems on λiThe weight vectors of point represent, Z*Be the object function that can search at present it is optimal to
Value, also referred to as approximate solution, Z*=min { (f1(x),f2(x),…,fi(x))}。
Preferably, the value produced in the step 4 and the value of its neighborhood carry out following computing, produce new explanation：Step 4.1 is selected
Select computing：Two sequence numbers h, k are randomly selected from B (i), with genetic operator by xhAnd xkProduce a new value, wherein xhIt is
The current optimal solution of h-th of subproblem, and xkIt is the current optimal solution of k-th of subproblem；The value and its neighborhood of generation
Value be compared, carry out it is winning slightly eliminates operation, the high outstanding value of selection fitness stays, and is genetic to the next generation；
Step 4.2 crossing operation：Individual in population is matched, carries out the crossover operation of gene, produces new
Body；
Step 4.3 mutation operator：The variation that low probability is carried out to genic value operates.
Preferably, return value R is drawn by the following formula in the step 6：
The beneficial effects of the invention are as follows：Invention introduces intensified learning mechanism, is continued to optimize using RL controllers, can be with
Accomplish the adaptive of parameter；The operator specially selected using the RL controllers of intensified learning, R and five Wei Guan is rewarded according to maximum
Vector is examined, population is continued to optimize to produce optimal value, until meeting end condition, efficiently solves MOEA/D for T parameter tune
Save the problem of insensitive.
Brief description of the drawings
Fig. 1 is the method for the present invention flow diagram.
Fig. 2 is verification the verifying results figure of the present invention in test UF3 problems.
Fig. 3 is verification the verifying results figure of the present invention in test UF7 problems.
Embodiment
The present invention is further elaborated with specific embodiment below in conjunction with the accompanying drawings.
As shown in Figure 1, the multi-objective Evolutionary Algorithm based on intensified learning, includes the following steps：
Step 1, generate initial population at random from search space；
Euclidean distance between step 1.1, calculating any two weight vectors, searches T nearest power of each weight vectors
Weight vector, wherein T is the number of the weight vectors in each neighborhood, for each i=1 ..., N, makes Bi={ i1,…,iT,
λi 1,…λi TIt is λiT nearest weight vectors；
Step 1.2, establish an exterior population EP, for storing the non-domination solution searched for during optimal solution and found, just
Beginningization EP is sky；
Step 1.3, uniformly random collection generation makes object function F (X)=(f from search space1(x),f2(x),…,fi
(x)) solution of optimal value is taken as initial population, wherein i=1,2 ..., m；X is one group of decision vector, and x is independent variable；
Step 1.4, using Chebyshev's method, object function F (X) is resolved into N straton problems：Wherein,I-th subproblem it is adjacent
Relation is by all subproblems on λiThe weight vectors of point represent, Z*Be the object function that can search at present it is optimal to
Value, also referred to as approximate solution, Z*=min { (f1(x),f2(x),…,fi(x))}。
Step 2, evaluate obtained population according to assessment level；
The optimum value for the object function that step 3, renewal search；
Step 4, the approximate solution Z using generation*Judgement is compared with end condition, satisfaction just terminates；To being unsatisfactory for end
The only population of condition, new value is produced using the DEvariant operators and T operators of intensified learning RL controllers selection, and to it
Intersected with the value of neighborhood, variation produces new explanation；
The value of generation carries out following computing with the value of its neighborhood, produces new explanation：
Step 4.1 Selecting operation：Two sequence numbers h, k are randomly selected from B (i), with genetic operator by xhAnd xkProduce one
A new value, wherein xhIt is the current optimal solution of h-th of subproblem, and xkIt is the current optimal solution of k-th of subproblem；
The value of generation is compared with the value of its neighborhood, and progress is winning slightly to eliminate operation, and the outstanding value for selecting fitness high stays, and loses
Pass to the next generation；
Step 4.2 crossing operation：Individual in population is matched, carries out the crossover operation of gene, produces new
Body；
Step 4.3 mutation operator：The variation that low probability is carried out to genic value operates.
Step 5, the new explanation produced select the solution for making subproblem function meet optimal value compared with the solution of original seed group,
Carry out Population Regeneration；
Step 6, the new population using generation, calculate 5 new dimension observation vector sum return value R, and then update RL controls
The state of device, judges whether to meet end condition, is unsatisfactory for, and is constantly iterated calculating, until meeting end condition, terminates.
Return value R is drawn by the following formula：
As Figure 2-3, in order to show the validity of algorithm, two standard testing collection UF3, UF7 be have chosen to verify.Its
Middle UF3, UF7 are the optimization problem of 2 targets.Population size is set to 300.Test result indicates that more mesh based on intensified learning
The MOEA/D algorithms that mark optimization algorithm is better than in the adjusting to T parameters.
Invention introduces intensified learning mechanism, is continued to optimize using RL controllers, can accomplish the adaptive of parameter；Profit
The operator selected with the RL controllers of intensified learning, it is constantly excellent to produce optimal value according to the maximum dimension observation vectors of reward R and five
Change population, until meeting end condition, solve the problems, such as that MOEA/D is insensitive for T parameter regulations.
The above is present pre-ferred embodiments, for the ordinary skill in the art, according to the present invention
Teaching, in the case where not departing from the principle of the present invention and spirit, the changes, modifications, replacement and the change that are carried out to embodiment
Type is still fallen within protection scope of the present invention.
Claims (4)
1. the multi-objective Evolutionary Algorithm based on intensified learning, it is characterised in that include the following steps：
Step 1, generate initial population at random from search space；
Step 2, evaluate obtained population according to assessment level；
The optimum value for the object function that step 3, renewal search；
Step 4, the approximate solution Z using generation*Judgement is compared with end condition, satisfaction just terminates；To being unsatisfactory for end condition
Population, produce new value using the DEvariant operators and T operators of intensified learning RL controllers selection, and to itself and neighborhood
Value intersected, variation produce new explanation；
Compared with the solution of original seed group, selection makes subproblem function meet the solution of optimal value, comes more for step 5, the new explanation produced
New population；
Step 6, the new population using generation, calculate 5 new dimension observation vector sum return value R, and then update RL controllers
State, judges whether to meet end condition, is unsatisfactory for, and is constantly iterated calculating, until meeting end condition, terminates.
2. the multi-objective Evolutionary Algorithm according to claim 1 based on intensified learning, it is characterised in that the step 1
Concretely comprise the following steps：
Step 1.1, calculate any two weight vectors between Euclidean distance, search T nearest weight of each weight vectors to
Amount, wherein T is the number of the weight vectors in each neighborhood, for each i=1 ..., N, makes Bi={ i1,…,iT,
λi 1,…λi TIt is λiT nearest weight vectors；
Step 1.2, establish an exterior population EP, for storing the non-domination solution searched for during optimal solution and found, initialization
EP is sky；
Step 1.3, uniformly random collection generation makes object function F (X)=(f from search space1(x),f2(x),…,fi(x))
The solution of optimal value is taken as initial population, wherein i=1,2 ..., m；X is one group of decision vector, and x is independent variable；
Step 1.4, using Chebyshev's method, object function F (X) is resolved into N straton problems：Wherein,I-th subproblem it is adjacent
Relation is by all subproblems on λiThe weight vectors of point represent, Z*Be the object function that can search at present it is optimal to
Value, also referred to as approximate solution, Z*=min { (f1(x),f2(x),…,fi(x))}。
3. the multi-objective Evolutionary Algorithm according to claim 2 based on intensified learning, it is characterised in that in the step 4
The value of generation carries out following computing with the value of its neighborhood, produces new explanation：Step 4.1 Selecting operation：Two are randomly selected from B (i)
A sequence number h, k, with genetic operator by xhAnd xkProduce a new value, wherein xhIt is the current optimal of h-th subproblem
Solution, and xkIt is the current optimal solution of k-th of subproblem；The value of generation compared with the value of its neighborhood, winning summary is carried out
Operation is eliminated, the outstanding value for selecting fitness high stays, and is genetic to the next generation；
Step 4.2 crossing operation：Individual in population is matched, carries out the crossover operation of gene, produces new individual；
Step 4.3 mutation operator：The variation that low probability is carried out to genic value operates.
4. the multi-objective Evolutionary Algorithm according to claim 3 based on intensified learning, it is characterised in that in the step 6
Return value R is drawn by the following formula：
<mrow>
<mi>R</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>N</mi>
</munderover>
<mfrac>
<mrow>
<msup>
<mi>g</mi>
<mrow>
<mi>t</mi>
<mi>e</mi>
</mrow>
</msup>
<mrow>
<mo>(</mo>
<msubsup>
<mi>X</mi>
<mrow>
<mi>t</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mi>i</mi>
</msubsup>
<mo>|</mo>
<msup>
<mi>&lambda;</mi>
<mi>i</mi>
</msup>
<mo>,</mo>
<mi>z</mi>
<mo>*</mo>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msup>
<mi>g</mi>
<mrow>
<mi>e</mi>
<mi>t</mi>
</mrow>
</msup>
<mrow>
<mo>(</mo>
<msubsup>
<mi>X</mi>
<mi>t</mi>
<mi>i</mi>
</msubsup>
<mo>|</mo>
<msup>
<mi>&lambda;</mi>
<mi>i</mi>
</msup>
<mo>,</mo>
<mi>z</mi>
<mo>*</mo>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msup>
<mi>g</mi>
<mrow>
<mi>t</mi>
<mi>e</mi>
</mrow>
</msup>
<mrow>
<mo>(</mo>
<msubsup>
<mi>X</mi>
<mrow>
<mi>t</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mi>i</mi>
</msubsup>
<mo>|</mo>
<msup>
<mi>&lambda;</mi>
<mi>i</mi>
</msup>
<mo>,</mo>
<mi>z</mi>
<mo>*</mo>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>.</mo>
</mrow>
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711279238.2A CN108038538A (en) | 2017-12-06 | 2017-12-06 | Multi-objective Evolutionary Algorithm based on intensified learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711279238.2A CN108038538A (en) | 2017-12-06 | 2017-12-06 | Multi-objective Evolutionary Algorithm based on intensified learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108038538A true CN108038538A (en) | 2018-05-15 |
Family
ID=62095661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711279238.2A Pending CN108038538A (en) | 2017-12-06 | 2017-12-06 | Multi-objective Evolutionary Algorithm based on intensified learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108038538A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108805268A (en) * | 2018-06-08 | 2018-11-13 | 中国科学技术大学 | Deeply learning strategy network training method based on evolution algorithm |
| CN108830370A (en) * | 2018-05-24 | 2018-11-16 | 东北大学 | Based on the feature selection approach for enhancing learning-oriented flora foraging algorithm |
| CN110174118A (en) * | 2019-05-29 | 2019-08-27 | 北京洛必德科技有限公司 | Robot multiple-objective search-path layout method and apparatus based on intensified learning |
| CN110704959A (en) * | 2019-08-19 | 2020-01-17 | 南昌航空大学 | MOEAD (Metal oxide optical insulator deposition) optimization fixture layout method and device based on migration behavior |
| CN110782016A (en) * | 2019-10-25 | 2020-02-11 | 北京百度网讯科技有限公司 | Method and apparatus for optimizing neural network architecture search |
| CN111045325A (en) * | 2018-10-11 | 2020-04-21 | 富士通株式会社 | Optimization device and control method of optimization device |
| TWI741760B (en) * | 2020-08-27 | 2021-10-01 | 財團法人工業技術研究院 | Learning based resource allocation method, learning based resource allocation system and user interface |
-
2017
- 2017-12-06 CN CN201711279238.2A patent/CN108038538A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108830370A (en) * | 2018-05-24 | 2018-11-16 | 东北大学 | Based on the feature selection approach for enhancing learning-oriented flora foraging algorithm |
| CN108830370B (en) * | 2018-05-24 | 2020-11-10 | 东北大学 | Feature selection method based on reinforced learning type flora foraging algorithm |
| CN108805268A (en) * | 2018-06-08 | 2018-11-13 | 中国科学技术大学 | Deeply learning strategy network training method based on evolution algorithm |
| CN111045325A (en) * | 2018-10-11 | 2020-04-21 | 富士通株式会社 | Optimization device and control method of optimization device |
| CN110174118A (en) * | 2019-05-29 | 2019-08-27 | 北京洛必德科技有限公司 | Robot multiple-objective search-path layout method and apparatus based on intensified learning |
| CN110704959A (en) * | 2019-08-19 | 2020-01-17 | 南昌航空大学 | MOEAD (Metal oxide optical insulator deposition) optimization fixture layout method and device based on migration behavior |
| CN110704959B (en) * | 2019-08-19 | 2022-04-08 | 南昌航空大学 | MOEAD (Metal oxide optical insulator deposition) optimization fixture layout method and device based on migration behavior |
| CN110782016A (en) * | 2019-10-25 | 2020-02-11 | 北京百度网讯科技有限公司 | Method and apparatus for optimizing neural network architecture search |
| TWI741760B (en) * | 2020-08-27 | 2021-10-01 | 財團法人工業技術研究院 | Learning based resource allocation method, learning based resource allocation system and user interface |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108038538A (en) | Multi-objective Evolutionary Algorithm based on intensified learning | |
| Reagen et al. | A case for efficient accelerator design space exploration via bayesian optimization | |
| Oduguwa et al. | Bi-level optimisation using genetic algorithm | |
| CN109932903A (en) | The air-blower control Multipurpose Optimal Method of more parent optimization networks and genetic algorithm | |
| Yu et al. | Evolutionary fuzzy neural networks for hybrid financial prediction | |
| CN109214449A (en) | A kind of electric grid investment needing forecasting method | |
| He et al. | Optimising the job-shop scheduling problem using a multi-objective Jaya algorithm | |
| CN105808426A (en) | Path coverage test data generation method used for weak mutation test | |
| Sato et al. | Variable space diversity, crossover and mutation in MOEA solving many-objective knapsack problems | |
| CN104616062A (en) | Nonlinear system recognizing method based on multi-target genetic programming | |
| CN110163743A (en) | A kind of credit-graded approach based on hyperparameter optimization | |
| CN105117326A (en) | Test case set generation method based on combination chaotic sequence | |
| CN110111606A (en) | A kind of vessel traffic flow prediction technique based on EEMD-IAGA-BP neural network | |
| CN108563875A (en) | Analog circuit measuring point and frequency based on multiple-objection optimization combine preferred method | |
| CN109886448A (en) | Using learning rate changing BP neural network and the heat pump multiobjective optimization control method of NSGA-II algorithm | |
| Huang et al. | Multi-objective multi-generation Gaussian process optimizer for design optimization | |
| CN103473465B (en) | Land resource spatial configuration optimal method based on multiple target artificial immune system | |
| Guo et al. | Hybridizing cellular automata principles and NSGAII for multi-objective design of urban water networks | |
| Sun et al. | Solving interval multi-objective optimization problems using evolutionary algorithms with preference polyhedron | |
| Zheng et al. | Data-driven optimization based on random forest surrogate | |
| CN112132259B (en) | Neural network model input parameter dimension reduction method and computer readable storage medium | |
| CN114004065A (en) | Transformer substation engineering multi-objective optimization method based on intelligent algorithm and environmental constraints | |
| Roeva et al. | Generalized net model of selection operator of genetic algorithms | |
| Zhou et al. | Approximation model guided selection for evolutionary multiobjective optimization | |
| Yasin et al. | Optimal least squares support vector machines parameter selection in predicting the output of distributed generation |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180515 |
|
| RJ01 | Rejection of invention patent application after publication |