CN110163373B - Mixed gene operation method - Google Patents
Mixed gene operation method Download PDFInfo
- Publication number
- CN110163373B CN110163373B CN201810148173.6A CN201810148173A CN110163373B CN 110163373 B CN110163373 B CN 110163373B CN 201810148173 A CN201810148173 A CN 201810148173A CN 110163373 B CN110163373 B CN 110163373B
- Authority
- CN
- China
- Prior art keywords
- target
- adaptation
- adaptive
- pursuit
- targets
- 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
- 238000000034 method Methods 0.000 title claims abstract description 14
- 108090000623 proteins and genes Proteins 0.000 title abstract description 13
- 230000006978 adaptation Effects 0.000 claims abstract description 65
- 230000003044 adaptive effect Effects 0.000 claims abstract description 42
- 238000011156 evaluation Methods 0.000 claims abstract description 30
- 230000002068 genetic effect Effects 0.000 claims abstract description 19
- 238000004519 manufacturing process Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 238000010845 search algorithm Methods 0.000 claims description 4
- 238000002922 simulated annealing Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 abstract description 6
- 230000013011 mating Effects 0.000 description 9
- 230000035772 mutation Effects 0.000 description 4
- 210000000349 chromosome Anatomy 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000010076 replication Effects 0.000 description 3
- 238000004904 shortening Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
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
Abstract
The invention provides a mixed gene operation method, which comprises the following steps: receiving a pursuit target to randomly generate a plurality of solving paths; converting each solution path into a first adaptation target; comparing the first adaptive target with the pursuit target, and performing a genetic operation on the first adaptive target to generate a plurality of second adaptive targets when an infeasible solution is generated; performing a single-point search operation and a comparison operation on the second adaptive target to generate a plurality of first evaluation paths and second evaluation paths; converting each first evaluation path and each second evaluation path into a mixed adaptation target; when the mixed adaptation target meets the pursuit target, the mixed adaptation target is regarded as an optimal solution, so that the aims of rapid convergence and reduction of calculation time are achieved.
Description
Technical Field
The invention relates to an algorithm, in particular to a hybrid gene algorithm.
Background
The gene algorithm (Genetic Algorithm, GA for short) is a search algorithm for solving optimization in computational mathematics, and the gene algorithm retains excellent offspring through evolution and iteration of an initial parent.
The gene algorithm flow is as follows: randomly generating n chromosomes at first; calculating the fitness values of all chromosomes by using the fitness function; repeating the steps 2 to 4 of the foregoing "evaluating adaptation function", "selecting", "copying", "mating", "mutating", and the like, and performing the steps 2 to 4 times of the foregoing is referred to as 1 iteration until convergence, wherein the convergence is provided that the number of iterations reaches a certain number or all chromosomes are very similar.
The genetic algorithm has global searching capability, so that the whole solution in the solution space can be searched out without the problem of rapid decline of the local solution; however, the local searching capability of the genetic algorithm is poor, so that the simple genetic algorithm is time-consuming, the searching efficiency is low in the later period of evolution, and in practical application, the genetic algorithm is easy to generate the problem of premature convergence, and the problem of miscalculation is caused because the optimal solution is not necessarily found to converge.
Disclosure of Invention
In order to solve the above problems, the present invention provides a hybrid genetic algorithm, which combines a pursuit target with a single-point search operation and a comparison operation through genetic algorithm, so as to achieve the purpose of rapid convergence and optimization, and further reduce the calculation time.
An embodiment of the present invention provides a hybrid gene operation method, which includes the steps of: a random step of receiving a pursuit target and randomly generating a plurality of solving paths aiming at the pursuit target; a first conversion step of converting each solution path into a corresponding first adaptation target; a first searching step, comparing the first adaptation target with the pursuit target, and performing a genetic operation on the first adaptation target to generate a plurality of second adaptation targets when an infeasible solution is generated; a second searching step, performing a single-point searching operation on the second adaptive target to generate a plurality of first evaluation paths; a third searching step, performing a comparison operation on the second adaptive target to generate a plurality of second evaluation paths; a second conversion step of converting each of the first evaluation paths and each of the second evaluation paths into a hybrid adaptation target; and a solution obtaining step, wherein the mixed adaptation target is compared with the pursuit target, and when the mixed adaptation target accords with the pursuit target, the mixed adaptation target is regarded as an optimal solution.
The single-point searching operation compares each second adaptive target with the pursuit target one by one to generate each first evaluation path conforming to the pursuit target.
Wherein, the single-point search operation is simulated annealing.
The comparison operation stores the second adaptation target in a database.
The comparison operation converts each second adaptation target stored in the information base into a comparison adaptation target, and when the comparison operation receives each new second adaptation target again, the comparison operation compares each new second adaptation target with each comparison adaptation target so as to generate each second evaluation path.
Wherein the comparison operation is a tabu search algorithm.
Wherein the genetic algorithm selectively replicates, mates and mutates each first adaptation objective to produce said second adaptation objective.
By the method, the optimal solution is rapidly calculated through gene operation and single-point search operation and comparison operation according to the pursuit target, and excellent individuals can be reserved and the diversity of groups can be maintained, so that the instruction period is improved.
Meanwhile, the comparison operation can store the excellent second evaluation path in the information base so as to be convenient for operation by the excellent second evaluation path in the next operation, thereby effectively shortening the operation time and improving the operation efficiency and the operation result.
Drawings
FIG. 1 is a flow chart of the steps of the present invention.
FIG. 2 is a schematic flow chart of the present invention.
FIG. 3 is a comparison flow chart of the present invention.
Description of the reference numerals
Tracking target 10 second evaluation path 71
Solution path 20 information repository 72
First adaptation target 30 compares adaptation target 73
Gene operation 40 hybrid adaptation target 80
Random step S1 of selective replication 41
Mating 42 first conversion step S2
Mutation 43 first search step S3
Second adaptive target 50 second searching step S4
Third search step S5 of the single point search operation 60
First evaluation path 61 second conversion step S6
The comparison operation 70 yields a solution step S7.
Detailed Description
For the purpose of illustrating the central concepts of the present invention as presented in the summary of the invention above, specific embodiments are presented. The various objects in the embodiments are drawn to scale, size, deformation or displacement as appropriate for the description, and not to scale for the actual components, as previously described.
Referring to fig. 1 to 3, the present invention provides a hybrid gene operation method, which includes the following steps:
randomizing step S1: a pursuit target 10 is received, and a plurality of solving paths 20 are randomly generated for the pursuit target 10. Further, in the embodiment of the present invention, for the pursuit target 10 to be solved, a feasible solution that may be in line with the pursuit target 10 is randomly generated, and each solution path 20 is a feasible solution.
A first switching step S2: each solution path 20 is translated into a corresponding first adaptation target 30. Further, in the embodiment of the present invention, each solution path 20 needs to be converted into an adaptive function that can be used in conjunction with different calculation modes that will be used by different applications, so each first adaptation target 30 is equivalent to the adaptive function.
The first searching step S3: the first adaptive targets 30 are compared with the pursuit targets 10, wherein when one of the first adaptive targets 30 meets the pursuit target 10, the condition is satisfied, and the met first adaptive target 30 is regarded as the best solution.
However, when any one of the first adaptation targets 30 cannot meet the pursuit target 10, i.e. an infeasible solution is generated, a genetic operation 40 is performed on the first adaptation targets 30, and the genetic operation 40 performs selective replication 41, mating 42 and mutation 43 operations on each of the first adaptation targets 30 to generate a plurality of second adaptation targets 50. Wherein, the first adaptive targets 30 are subjected to selective replication 41 of genetic operation 40, so as to reserve the first adaptive targets 30 which are more suitable for the pursuit target 10 and exclude the first adaptive targets 30 which are not suitable for the pursuit target 10, thereby performing preliminary screening; performing the mating 42 of the genetic algorithm 4 on each first adaptive target 30 which is more suitable for the pursuit target 10, and performing the mating 42 two by two to generate new first adaptive targets 30 by taking each first adaptive target 30 which is more suitable for the pursuit target 10 as a group, wherein whether the mating 42 is performed is determined by the mating probability, and the positions after the mating 42 are also randomly determined, so that the diversity of each first adaptive target 30 is maintained; the mutation 43 of the gene operation 40 is performed on each first adaptation target 30 generated by the mating 42, and whether each first adaptation target 30 is mutated 43 is determined in a random manner, and a plurality of second adaptation targets 50 are generated after the mutation 43.
The second searching step S4: in the embodiment of the present invention, the single-point search operation 60 is a simulated annealing method (Simulated Annealing), and the single-point search operation 60 is an approximation method for searching a large search range in a fixed time, and the single-point search operation 60 compares each second adaptive target 50 with each pursuit target 10 one by one, so as to have the capability of jumping off the minimum value of the region, thereby quickly finding the best solution.
Third searching step S5: the second adaptation targets 50 are subjected to a comparison operation 70 to generate a plurality of second evaluation paths 71, and during the calculation of the comparison operation 70, the comparison operation 70 stores the second adaptation targets 50 in an information base 72, converts the second adaptation targets 50 stored in the information base 72 into comparison adaptation targets 73, and when the calculation is continued, the comparison operation 70 continuously receives new second adaptation targets 50 again, and the comparison operation 70 compares the new second adaptation targets 50 with the comparison adaptation targets 73 to generate second evaluation paths 71, wherein in the embodiment of the invention, the information base 72 stores 10 comparison adaptation targets 73.
Further description: when the second adaptive targets 50 are compared 70 for the first time, the second adaptive targets 50 are converted into comparison adaptive targets 73 as matching the pursuit targets 10; when the operations of the second and third times are continued and the second adaptation targets 50 generated in the first searching step S3 enter the comparison operation 70 again, the new second adaptation targets 50 are compared with the comparison adaptation targets 73 in the information base 72 to find the second adaptation targets 50 more preferably matching the pursuit targets 10, and then the more suitable second adaptation targets 50 are converted into the comparison adaptation targets 73, and the original comparison adaptation targets 73 are replaced. In the embodiment of the present invention, the comparison operation 70 is a Tabu Search algorithm (tab Search), and the comparison operation 70 is an auxiliary heuristic, and the previous Search result is recorded to avoid sinking into the local optimal solution, so as to avoid detour Search, ensure effective exploration of diversity, and finally realize global optimization.
Meanwhile, the comparison operation 70 can store the excellent second evaluation path 71 in the information base 72, so that the calculation is performed by the excellent second evaluation path 71 in the next operation, thereby effectively shortening the operation time and improving the operation efficiency and the operation result.
In the embodiment of the present invention, the second searching step S4 and the third searching step S5 are performed simultaneously, and the second adaptive targets 50 and the pursuing targets 10 are compared one by one through the single-point searching operation 60, and meanwhile, the comparison operation 70 is used to avoid sinking into the local optimal solution, so that the operation time can be effectively shortened, the convergence is rapid, and the purpose of rapid operation is achieved.
A second conversion step S6: each first evaluation path 61 and each second evaluation path 71 are converted into a hybrid adaptation target 80. Further, the first evaluation paths 61 that are calculated by the single-point search operation 60 and the second evaluation paths 71 that are calculated by the comparison operation 70 and that are in line with the pursuit target 10 are combined to find the paths that are most in line with the pursuit target 10, and then the most in line paths are converted into the hybrid adaptation target 80.
And (7) obtaining a solution step S7: the hybrid adaptation target 80 is compared with the pursuit target 10, and when the hybrid adaptation target 80 meets the pursuit target 10, the hybrid adaptation target 80 is regarded as an optimal solution.
Illustrating: during the manufacturing process of the machine factory, a machine failure will generate a variant manufacturing schedule, and at this time, the pursuit goal 10 of the line schedule is to minimize the difference between the initial processing time and the completion processing time of the original manufacturing schedule and the variant manufacturing schedule.
First, a random step S1 is performed to randomly generate each solution path 20 for solving the pursuit target 10 for the pursuit target 10, and how each solution path 20 is configured to process the machine. Next, in the first conversion step S2, each solution path 20 is converted into a corresponding first adaptation target 30 for the scheduling system of the factory to operate.
Then, the first searching step S3 compares the converted first adaptive targets 30 with the pursuit targets 10, and finds out whether there is a pursuit target 10 that is consistent with minimizing the difference between the initial processing time and the completion processing time of the original manufacturing schedule and the variant manufacturing schedule in the first adaptive targets 30, if yes, the satisfied first adaptive targets 30 are used as the plan of the variant manufacturing schedule to solve the problem of machine failure; if this is not the case, the genetic algorithm 40 is performed to generate different second fitness objectives 50 to seek an optimal solution that would otherwise meet the pursuit objective 10.
Then, each second adaptive target 50 performs the second searching step S4 and the third searching step S5 simultaneously, and compares each second adaptive target 50 with the following target 10 one by one through the single-point searching operation 60, and simultaneously uses the comparing operation 70 to avoid sinking into the local optimal solution, so as to find out the manufacturing schedule capable of meeting the following target 10.
Then, the second conversion step S6 is used to comprehensively determine and convert each first evaluation path 61 and each second evaluation path 71 into the hybrid adaptation target 80, and the solution step S7 is used to find the manufacturing schedule that best meets the pursuit target 10, so as to minimize the variation between the start processing time and the end processing time of the variant manufacturing schedule and the original manufacturing schedule.
Therefore, the invention can rapidly calculate the optimal solution according to the pursuit target 10 through the gene operation 40 and the single-point search operation 60 and the comparison operation 70, and can keep excellent individuals and maintain the diversity of groups, thereby improving the instruction period.
In summary, the above-mentioned method can be designed to overcome the drawbacks of the prior art, and further has many advantages and practical values, while the present invention has been described in terms of the preferred embodiments, it is not intended to limit the scope of the present invention, and any person skilled in the art will be able to make various changes and modifications to the above-mentioned embodiments without departing from the spirit and scope of the present invention, so that the scope of the present invention is defined by the appended claims.
Claims (5)
1. A hybrid genetic algorithm suitable for use in a manufacturing process of a mechanical factory, the hybrid genetic algorithm comprising the steps of:
and (3) a random step: receiving a pursuit target, and randomly generating a plurality of solving paths aiming at the pursuit target, wherein when a machine fault occurs in the production process, a variant manufacturing schedule is generated, and the pursuit target is used for minimizing the difference between an original manufacturing schedule and the starting processing time and the finishing processing time of the variant manufacturing schedule;
a first conversion step: converting each solution path into a corresponding first adaptive target for the scheduling system of the mechanical factory to operate;
a first searching step: comparing the first adaptive target with the pursuit target, if the first adaptive target accords with the pursuit target, taking the first adaptive target which accords with the pursuit target as a plan of the variant manufacturing schedule so as to solve the problem of machine fault, and if the first adaptive target does not accord with the pursuit target, carrying out a genetic operation on the first adaptive target so as to generate a plurality of second adaptive targets;
and a second searching step: performing a single-point search operation on the second adaptive target to generate a plurality of first evaluation paths;
and a third searching step: performing a comparison operation on the second adaptation targets to generate a plurality of second evaluation paths, wherein the comparison operation stores the second adaptation targets in an information base, the comparison operation converts each second adaptation target stored in the information base into a comparison adaptation target, and when the comparison operation receives each new second adaptation target again, the comparison operation compares each new second adaptation target with each comparison adaptation target to generate each second evaluation path;
a second conversion step: converting each first evaluation path and each second evaluation path into a mixed adaptation target; and
the solution step: and comparing the mixed adaptive target with the pursuit target, wherein when the mixed adaptive target accords with the pursuit target, the mixed adaptive target is regarded as an optimal solution, so that the starting processing time and the ending processing time of the variant manufacturing schedule can be minimized with the variation among the original manufacturing schedules.
2. The method of claim 1, wherein the single point search compares each of the second adaptive targets with the pursuit target one by one to generate each of the first evaluation paths corresponding to the pursuit target.
3. The method of claim 2, wherein the single point search operation is simulated annealing.
4. The method of claim 1, wherein the comparison is a tabu search algorithm.
5. The method of claim 1, wherein the genetic algorithm selectively replicates, mates and mutates each first adaptation objective to produce said second adaptation objective.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810148173.6A CN110163373B (en) | 2018-02-13 | 2018-02-13 | Mixed gene operation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810148173.6A CN110163373B (en) | 2018-02-13 | 2018-02-13 | Mixed gene operation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110163373A CN110163373A (en) | 2019-08-23 |
| CN110163373B true CN110163373B (en) | 2023-11-03 |
Family
ID=67635302
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810148173.6A Active CN110163373B (en) | 2018-02-13 | 2018-02-13 | Mixed gene operation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110163373B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101534141B1 (en) * | 2014-08-05 | 2015-07-07 | 성균관대학교산학협력단 | Rationale word extraction method and apparatus using genetic algorithm, and sentiment classification method and apparatus using said rationale word |
| JP2017010544A (en) * | 2015-06-16 | 2017-01-12 | 富士通株式会社 | Multiple goal optimization method and device |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8131656B2 (en) * | 2006-01-31 | 2012-03-06 | The Board Of Trustees Of The University Of Illinois | Adaptive optimization methods |
| US20100080298A1 (en) * | 2008-09-30 | 2010-04-01 | Hsueh-Ming Hang | Refined Weighting Function and Momentum-Directed Genetic search pattern algorithm |
| US9189733B2 (en) * | 2013-03-13 | 2015-11-17 | The Aerospace Corporation | Systems and methods for vector scalability of evolutionary algorithms |
-
2018
- 2018-02-13 CN CN201810148173.6A patent/CN110163373B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101534141B1 (en) * | 2014-08-05 | 2015-07-07 | 성균관대학교산학협력단 | Rationale word extraction method and apparatus using genetic algorithm, and sentiment classification method and apparatus using said rationale word |
| JP2017010544A (en) * | 2015-06-16 | 2017-01-12 | 富士通株式会社 | Multiple goal optimization method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110163373A (en) | 2019-08-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2018407695B2 (en) | Constrained stable matching strategy-based flexible job-shop scheduling method | |
| Mousa et al. | Local search based hybrid particle swarm optimization algorithm for multiobjective optimization | |
| Hadidi et al. | Structural optimization using artificial bee colony algorithm | |
| US10931027B2 (en) | Method for array elements arrangement of l-shaped array antenna based on inheritance of acquired character | |
| Cao et al. | Assembly planning using a novel immune approach | |
| CN113204417A (en) | Multi-satellite multi-point target observation task planning method based on improved genetic and firefly combined algorithm | |
| CN110163373B (en) | Mixed gene operation method | |
| CN110222816B (en) | Deep learning model establishing method, image processing method and device | |
| An et al. | MoDNA: motif-oriented pre-training for DNA language model | |
| CN112257872B (en) | Target planning method for reinforcement learning | |
| Yaguchi et al. | Basic study of proximate optimality principle based combinatorial optimization method | |
| JP2020061012A (en) | Optimization device, optimization device control method, sampling device, and sampling device control method | |
| CN116933485A (en) | Assembly sequence planning method of genetic greedy combination algorithm | |
| Arai et al. | Fuzzy genetic algorithm for prioritization determination with technique for order preference by similarity to ideal solution | |
| Sun et al. | A novel hybrid estimation of distribution algorithm for solving hybrid flowshop scheduling problem with unrelated parallel machine | |
| CN111126707A (en) | Energy consumption equation construction and energy consumption prediction method and device | |
| CN104570759B (en) | The quick Binomial Trees of control system midpoint orientation problem | |
| TWI652629B (en) | Mixed genetic algorithm | |
| CN116915311B (en) | Satellite task scheduling method based on co-evolution algorithm | |
| Lo et al. | Applying a chaos-based firefly algorithm to the permutation flow shop scheduling problem | |
| Zapotecas Martínez et al. | A multi-objective meta-model assisted memetic algorithm with non gradient-based local search | |
| CN115439020B (en) | Multi-system emergency linkage method and system in data center construction process | |
| Suganthan et al. | Adaptation methods in differential evolution: A review | |
| Wang et al. | An improved LHS approach for constrained design space based on successive local enumeration algorithm | |
| Hedar et al. | Genetic algorithm with population partitioning and space reduction for high dimensional problems |
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 |