CN107729431A - A kind of improved harmony chess game optimization method - Google Patents
A kind of improved harmony chess game optimization method Download PDFInfo
- Publication number
- CN107729431A CN107729431A CN201710906703.4A CN201710906703A CN107729431A CN 107729431 A CN107729431 A CN 107729431A CN 201710906703 A CN201710906703 A CN 201710906703A CN 107729431 A CN107729431 A CN 107729431A
- Authority
- CN
- China
- Prior art keywords
- mtd
- msubsup
- mrow
- harmony
- mtr
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2453—Query optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of improved harmony chess game optimization method, it is randomly assigned to operate by using distribution coefficient control, so as to increase the diversity in harmony storehouse, promote algorithm fast searching to being most hopeful to obtain the region of search of optimal solution, while by obtaining more outstanding solution to the appropriate fine setting of the region of search.
Description
Technical field
The present invention relates to intelligent algorithm field, particularly a kind of improved harmony chess game optimization method.
Background technology
As the expansion of environment for human survival, and the scope of transformation and the knowledge of natural environment are widened.The problem of being handled in reality
More complicate, using classical algorithm such as:Newton method, simple method, optimum gradient method, pattern search method etc. can not
Meet complicated optimization problem.Therefore, efficient optimized algorithm turns into one of goal in research of scientific worker.Count at a high speed simultaneously
The extensive use of calculation machine also provides favourable instrument guarantee for solving complexity problem.
Many highly difficult, high-dimensional optimization problem in Chinese national economy every field be present, such as:In transit
Optimal scheduling, the optimum allocation of resource, the optimal layout of production procedure, the optimum development on territory, rational deployment of crops etc.
All it is some dimensions height, the problem of being not easy to solve using classical algorithm.
Under such background, intelligent optimization algorithm is a very active field developed in recent years, is
Solves the effective way of high-dimensional complicated optimum problem.And it is widely used the field in practical problem, being related to and mainly includes agriculture
Industry, national defence, industry, engineering, traffic, chemical industry etc..Large quantities of intelligent optimization algorithms are emerged in recent years.Such as genetic algorithm, promptly search
Rope algorithm, simulated annealing, ant colony optimization algorithm and particle swarm optimization algorithm.
It is a kind of based on music principle that harmonic search algorithm is that South Korea scholar ZongWoo Geem in 2001 et al. are proposed
New intelligent search algorithm.In musical performance, sound of the musician by the memory of oneself by each musical instrument in regulation band repeatedly
Adjust, be finally reached a beautiful harmony state, Geem et al. and inspired by this phenomenon, it is proposed that harmonic search algorithm.Calculate
Method produces HM initial solution first, and is put into harmony data base;Then to each component of solution respectively with probability HMCR with
Sound memory scans in storehouse, is searched for 1-HMCR probability outside data base, it is expected that the respective components for obtaining new explanation is remembering
When being scanned in storehouse, after random search is to a certain component, then disturb finally by searching for the component with probability P AR
The each component obtained afterwards forms new explanation, if new explanation is better than the worst solution in data base, with worst solution in new explanation replacement storehouse such as
This circulation, untill meeting end condition.
The shortcomings that harmonic search algorithm of the prior art, is that its algorithm robustness is not high, and randomness is big, relatively
Initial harmony storehouse is relied on, is easily trapped into local optimum.
The content of the invention
It is an object of the invention to provide a kind of improved harmony chess game optimization method, to overcome present in prior art
Defect.
To achieve the above object, the technical scheme is that:A kind of improved harmony chess game optimization method, including it is as follows
Step:
Step S1:Determine the basic parameter of the object function of optimization problem, constraints and harmony search;
Step S2:Harmony data base HM is initialized;
Step S3:Produce new explanation;
Step S4:It is allocated operation at random;
Step S5:Harmony data base HM is updated, judges whether new explanation is better than the worst solutions of HM, if so, then updating HM;
Step S6:Repeat step S3 to step S5, until reaching maximum iteration or meeting criterion, then stop, and
Circulation output optimal solution.
In an embodiment of the present invention, in the step S1, the basic parameter includes:Musical instrument number m;Various musical instruments
Range of pitch;The harmony number M that can retain in harmony search library HM;Harmony data base retains probability HMCR, that is, is producing newly
Retain the x of solution component during solution from harmony data basei jProbability size;Harmony data base disturbs probability P AR, i.e., part is solved every time
Component disturb the probability size of fine setting;Maximum iteration, that is, the maximum times circulated, while be also end condition.
In an embodiment of the present invention, in the step S2, by the initial liberation of the M optimization problem randomly generated
Enter in harmony data base HM, be expressed as:
Wherein, XjFor j-th of solution vector,For i-th of component in j-th of solution vector;f(Xj) it is j-th of solution vector
Functional value, 1≤i≤m, 1≤j≤M.
In an embodiment of the present invention, in the step S3, a new explanation is produced every time
And new explanation is generated by the one or more in three kinds of modes:
1. the solution selected at random in harmony data base as probability using HMCR;
Generated 2. being randomly choosed with 1-HMCR probability;
3. it is probability to 1. and 2. middle component is finely adjusted to obtain using PAR.
In an embodiment of the present invention, the fine setting is carried out in the following manner:
Wherein, μ is pre-set bandwidths;Random numbers of the p between 0-1 and for adjusting amount trimmed,For in j-th of solution vector
I-th of component, 1≤i≤m, 1≤j≤M.
In an embodiment of the present invention, to new explanation caused by step 3, probability control is carried out by parameter preset Assign,
The solution of each dimension in solution is adjusted.
In an embodiment of the present invention, to, per one-dimensional solution, according to the ratio r of generating random number, obtaining the solution pair in solution
The ratio value part answered, this is partially increased to other dimension solutions, then is finely adjusted disturbance operation, carried out as follows:
Wherein, r is 0-1 random number,For i-th of component in j-th of solution vector,For in j-th of solution vector
A-th of component, 1≤i≤m, 1≤j≤M.
Compared to prior art, the invention has the advantages that:It is excellent that the present invention provides a kind of improved harmony search
Change method, what addition distribution coefficient controlled in basic harmonic search algorithm is randomly assigned to operate, so as to increase the more of harmony storehouse
Sample, so as to promote algorithm fast searching to be most hopeful to obtain the region of search of optimal solution, while by the region of search
It is appropriate to finely tune to obtain more outstanding solution.Operation harmony storehouse HM variations are randomly assigned by increase to be obviously improved, phase
It can reach the more preferable effect of optimizing faster compared with basic HS algorithms.
Brief description of the drawings
Fig. 1 is a kind of flow chart of improved harmony chess game optimization method in the present invention.
Fig. 2 is that optimizations of the RDHS and HS on 8 dimension Quartic functions shows schematic diagram in one embodiment of the invention.
Fig. 3 is property of the harmonic search algorithm on Rastrigin (8) under different Assign parameters in one embodiment of the invention
Can curve synoptic diagram.
Fig. 4 is the harmonic search algorithm of different allocation of parameters in one embodiment of the invention on 8 dimension Rastrigin functions
Optimization performance schematic diagram.
Fig. 5 is the harmonic search algorithm of different allocation of parameters in one embodiment of the invention on 30 dimension Rastrigin functions
Optimization performance schematic diagram.
Fig. 6 is the harmonic search algorithm of different allocation of parameters in one embodiment of the invention on 8 dimension Girewank functions
Optimization performance schematic diagram.
Embodiment
Below in conjunction with the accompanying drawings, technical scheme is specifically described.
A kind of improved harmony chess game optimization method of the present invention, as shown in figure 1, realizing in accordance with the following steps:
Step S1:Determine the basic parameter of the object function of optimization problem, constraints and harmony search.In this implementation
In example, basic parameter includes:
1. musical instrument (variable) number m;2. the range of pitch (variable-value scope) of various musical instruments;3. in harmony search library HM
The harmony number M that can retain, and M should be much smaller than feasible solution number;4. harmony data base retains probability HMCR, that is, producing newly
Retain the x of solution component during solution from harmony data basei jProbability size;5. data base disturbs probability P AR, i.e., every time to part solution point
Amount disturb the probability size of fine setting;6. maximum iteration, that is, the maximum times circulated, while be also that algorithm terminates bar
Part.
Step S2:Harmony remembers library initialization.The initial solution of the M optimization problem randomly generated is put into harmony data base
In HM, it can be expressed as:In formula, XjFor j-th of solution vector.For in j-th of solution vector
I-th of component;f(Xj) for the functional value of j-th solution vector.
Step S3:Produce new solution.A new explanation is produced every timeWherein, new explanation passes through three
Kind of mode is generated, including:
1. some solutions selected at random in harmony data base as probability using HMCR;
Generated 2. being randomly choosed with 1-HMCR probability;
3. some components in 1. and 2. are finely adjusted to obtain using PAR as probability.
Preferably, HMCR is 0.9, PAR 0.75.
Further, in the present embodiment, finely tune by below equation to carry out:
Wherein, μ is bandwidth, it is preferred that μ is used for the value for adjusting amount trimmed using the random number that 0.1, p is 0-1.
Step S4:It is allocated operation at random.To new explanation caused by step 3, by new arrange parameter Assign come probability
Control accordingly solves the solution of each dimension, particular by the ratio r to passing through generating random number in solution per one-dimensional solution, by its portion
Divide ratio value to increase on other dimension solutions, be finely adjusted disturbance operation afterwards.
Further, operation is randomly assigned to carry out using below equation:
Wherein r is 0-1 random number,It is the value that i dimensions are corresponded in certain group solution, the group solution is given for distributing one part value
Middle dimension is a component, i.e.,
Step S5:Update data base.Judge whether new explanation is better than the worst solutions of HM, if just renewal HM.
Step S6:Repeat step S3 to step S5, until reaching maximum iteration or meeting that criterion just stops circulating
Export optimal solution.
In order to allow those skilled in the art to further appreciate that method proposed by the invention, carried out with reference to instantiation
Explanation.
In this example, parameters are as follows:HMCR:0.9;PAR=0.75;μ=0.1;M=10;N=10000
Each test function result is as shown in the table:
1 each test function of table repeatedly calculates data result statistical form
As shown in Fig. 2 the optimization for RDHS in embodiment and HS on 8 dimension Quartic functions shows schematic diagram.
As shown in figure 3, it is property of the harmonic search algorithm on Rastrigin (8) under different Assign parameters in embodiment
Can curve synoptic diagram.
As shown in figure 4, tieed up for the harmonic search algorithm of different allocation of parameters in embodiment 8 on Rastrigin functions
Optimization performance schematic diagram.
As shown in figure 5, tieed up for the harmonic search algorithm of different allocation of parameters in embodiment 30 on Rastrigin functions
Optimization performance schematic diagram.
It is as shown in fig. 6, excellent on 8 dimension Girewank functions for the harmonic search algorithm of different allocation of parameters in embodiment
Change performance schematic diagram.
Above is presently preferred embodiments of the present invention, all changes made according to technical solution of the present invention, caused function are made
During with scope without departing from technical solution of the present invention, protection scope of the present invention is belonged to.
Claims (7)
- A kind of 1. improved harmony chess game optimization method, it is characterised in that comprise the following steps:Step S1:Determine the basic parameter of the object function of optimization problem, constraints and harmony search;Step S2:Harmony data base HM is initialized;Step S3:Produce new explanation;Step S4:It is allocated operation at random;Step S5:Harmony data base HM is updated, judges whether new explanation is better than the worst solutions of HM, if so, then updating HM;Step S6:Repeat step S3 to step S5, until reaching maximum iteration or meeting criterion, then stop, and circulate Export optimal solution.
- 2. a kind of improved harmony chess game optimization method according to claim 1, it is characterised in that in the step S1 In, the basic parameter includes:Musical instrument number m;The range of pitch of various musical instruments;The harmony that can retain in harmony search library HM Number M;Harmony data base retains probability HMCR, i.e., retains the x of solution component from harmony data base when producing new explanationi jProbability it is big It is small;Harmony data base disturbs probability P AR, i.e., to portion's decomposed component disturb the probability size of fine setting every time;Greatest iteration time Number, that is, the maximum times circulated, while be also end condition.
- 3. a kind of improved harmony chess game optimization method according to claim 1, it is characterised in that in the step S2 In,It is put into harmony data base HM, is expressed as by the initial solution of the M optimization problem randomly generated:<mrow> <mo>|</mo> <mrow> <mtable> <mtr> <mtd> <msubsup> <mi>x</mi> <mn>1</mn> <mn>1</mn> </msubsup> </mtd> <mtd> <msubsup> <mi>x</mi> <mn>2</mn> <mn>1</mn> </msubsup> </mtd> <mtd> <mn>...</mn> </mtd> <mtd> <msubsup> <mi>x</mi> <mi>m</mi> <mn>1</mn> </msubsup> </mtd> </mtr> <mtr> <mtd> <msubsup> <mi>x</mi> <mn>1</mn> <mn>2</mn> </msubsup> </mtd> <mtd> <msubsup> <mi>x</mi> <mn>2</mn> <mn>2</mn> </msubsup> </mtd> <mtd> <mn>...</mn> </mtd> <mtd> <msubsup> <mi>x</mi> <mi>m</mi> <mn>2</mn> </msubsup> </mtd> </mtr> <mtr> <mtd> <mn>...</mn> </mtd> <mtd> <mn>...</mn> </mtd> <mtd> <mn>...</mn> </mtd> <mtd> <mn>...</mn> </mtd> </mtr> <mtr> <mtd> <msubsup> <mi>x</mi> <mn>1</mn> <mi>M</mi> </msubsup> </mtd> <mtd> <msubsup> <mi>x</mi> <mn>2</mn> <mi>M</mi> </msubsup> </mtd> <mtd> <mn>...</mn> </mtd> <mtd> <msubsup> <mi>x</mi> <mi>m</mi> <mi>M</mi> </msubsup> </mtd> </mtr> </mtable> <mo>|</mo> <mtable> <mtr> <mtd> <mrow> <mi>f</mi> <mrow> <mo>(</mo> <msup> <mi>X</mi> <mn>1</mn> </msup> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>f</mi> <mrow> <mo>(</mo> <msup> <mi>X</mi> <mn>2</mn> </msup> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> <mtr> <mtd> <mn>...</mn> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>f</mi> <mrow> <mo>(</mo> <msup> <mi>X</mi> <mi>M</mi> </msup> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> </mtable> </mrow> <mo>|</mo> </mrow>Wherein, XjFor j-th of solution vector,For i-th of component in j-th of solution vector;f(Xj) for the function of j-th solution vector Value, 1≤i≤m, 1≤j≤M.
- 4. a kind of improved harmony chess game optimization method according to claim 1, it is characterised in that in the step S3 In,A new explanation is produced every timeAnd new explanation is entered by the one or more in three kinds of modes Row generation:1. the solution selected at random in harmony data base as probability using HMCR;Generated 2. being randomly choosed with 1-HMCR probability;3. it is probability to 1. and 2. middle component is finely adjusted to obtain using PAR.
- A kind of 5. improved harmony chess game optimization method according to claim 4, it is characterised in that it is described fine setting by with Under type is carried out:<mrow> <msubsup> <mi>x</mi> <mi>i</mi> <mi>j</mi> </msubsup> <mo>=</mo> <msubsup> <mi>x</mi> <mi>i</mi> <mi>j</mi> </msubsup> <mo>&PlusMinus;</mo> <mi>&mu;</mi> <mo>*</mo> <mi>p</mi> </mrow>Wherein, μ is pre-set bandwidths;Random numbers of the p between 0-1 and for adjusting amount trimmed,For in j-th of solution vector I component, 1≤i≤m, 1≤j≤M.
- 6. according to claim, a kind of improved harmony chess game optimization method described in 1, it is characterised in that to caused by step 3 New explanation, probability control is carried out by parameter preset Assign, the solution of each dimension in solution is adjusted.
- 7. a kind of improved harmony chess game optimization method according to claim 6, it is characterised in that to every one-dimensional in solution Solution, according to the ratio r of generating random number, obtains ratio value part corresponding to the solution, this is partially increased into other dimension solutions, then Disturbance operation is finely adjusted, is carried out as follows:<mrow> <msubsup> <mi>x</mi> <mi>i</mi> <mi>j</mi> </msubsup> <mo>=</mo> <msubsup> <mi>x</mi> <mi>i</mi> <mi>j</mi> </msubsup> <mo>*</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <mi>r</mi> <mo>)</mo> </mrow> </mrow><mrow> <msubsup> <mi>x</mi> <mi>a</mi> <mi>j</mi> </msubsup> <mo>=</mo> <msubsup> <mi>x</mi> <mi>a</mi> <mi>j</mi> </msubsup> <mo>+</mo> <msubsup> <mi>x</mi> <mi>i</mi> <mi>j</mi> </msubsup> <mo>*</mo> <mi>r</mi> </mrow>Wherein, r is 0-1 random number,For i-th of component in j-th of solution vector,For a-th in j-th of solution vector Component, 1≤i≤m, 1≤j≤M.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710906703.4A CN107729431A (en) | 2017-09-29 | 2017-09-29 | A kind of improved harmony chess game optimization method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710906703.4A CN107729431A (en) | 2017-09-29 | 2017-09-29 | A kind of improved harmony chess game optimization method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107729431A true CN107729431A (en) | 2018-02-23 |
Family
ID=61208273
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710906703.4A Pending CN107729431A (en) | 2017-09-29 | 2017-09-29 | A kind of improved harmony chess game optimization method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107729431A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108898853A (en) * | 2018-06-14 | 2018-11-27 | 广东工业大学 | A kind of control method of isolated traffic intersection signal adaptive |
CN109343479A (en) * | 2018-09-18 | 2019-02-15 | 合肥工业大学 | Flowshop production scheduling method and system based on the adaptive harmonic search algorithm of multiple target |
CN110554599A (en) * | 2019-02-27 | 2019-12-10 | 天津大学 | PI parameter optimization method based on adaptive harmony search algorithm |
-
2017
- 2017-09-29 CN CN201710906703.4A patent/CN107729431A/en active Pending
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108898853A (en) * | 2018-06-14 | 2018-11-27 | 广东工业大学 | A kind of control method of isolated traffic intersection signal adaptive |
CN109343479A (en) * | 2018-09-18 | 2019-02-15 | 合肥工业大学 | Flowshop production scheduling method and system based on the adaptive harmonic search algorithm of multiple target |
CN110554599A (en) * | 2019-02-27 | 2019-12-10 | 天津大学 | PI parameter optimization method based on adaptive harmony search algorithm |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Chen et al. | Harmony search algorithm with dynamic control parameters | |
Kong et al. | A simplified binary harmony search algorithm for large scale 0–1 knapsack problems | |
Li et al. | Improved tunicate swarm algorithm: Solving the dynamic economic emission dispatch problems | |
Wang et al. | Incorporating mutation scheme into krill herd algorithm for global numerical optimization | |
CN107729431A (en) | A kind of improved harmony chess game optimization method | |
Geem | State-of-the-art in the structure of harmony search algorithm | |
Resende et al. | GRASP and path relinking for the max–min diversity problem | |
Staudt et al. | Engineering high-performance community detection heuristics for massive graphs | |
Wang et al. | Self-adaptive harmony search algorithm for optimization | |
Pan et al. | A self-adaptive global best harmony search algorithm for continuous optimization problems | |
Brimberg et al. | Solving the maximally diverse grouping problem by skewed general variable neighborhood search | |
Mariotte et al. | Arbuscular mycorrhizal fungi reduce the differences in competitiveness between dominant and subordinate plant species | |
CN104936186B (en) | Cognitive radio network spectrum allocation method based on cuckoo searching algorithm | |
Zhao et al. | An improved particle swarm optimization with decline disturbance index (DDPSO) for multi-objective job-shop scheduling problem | |
CN102789493B (en) | Self-adaptive dual-harmony optimization method | |
CN103631758A (en) | Method for solving non-linear programming and absolute value equation through improved harmony search algorithm | |
Bansal et al. | Metrical task systems and the k-server problem on HSTs | |
CN106874555A (en) | A kind of Reed Muller logic circuits power consumption and area-optimized method | |
Pérez-Aracil et al. | A versatile multi-method ensemble for wind farm layout optimization | |
Brimberg et al. | Variable neighborhood search for the vertex weighted k-cardinality tree problem | |
Chen et al. | A hybrid dynamic harmony search algorithm for identical parallel machines scheduling | |
Zhao et al. | Semi-self-adaptive harmony search algorithm | |
Braun et al. | Multiple comparisons through orderly partitions: The maximum subrange procedure | |
Alomoush et al. | Solving 0/1 knapsack problem using hybrid HS and Jaya algorithms | |
Guan et al. | A unified model for Sierpinski networks with scale-free scaling and small-world effect |
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: 20180223 |
|
RJ01 | Rejection of invention patent application after publication |