CN104866514A - Argon atom cluster structure optimization technology based on Memetic algorithm - Google Patents
Argon atom cluster structure optimization technology based on Memetic algorithm Download PDFInfo
- Publication number
- CN104866514A CN104866514A CN201410070222.0A CN201410070222A CN104866514A CN 104866514 A CN104866514 A CN 104866514A CN 201410070222 A CN201410070222 A CN 201410070222A CN 104866514 A CN104866514 A CN 104866514A
- Authority
- CN
- China
- Prior art keywords
- search
- atmo
- cluster
- individuality
- atom
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Feedback Control In General (AREA)
- Complex Calculations (AREA)
Abstract
The invention discloses an argon atom cluster structure optimization technology based on a Memetic algorithm. The breadth advantage of group algorithm search and the depth advantage of a local search algorithm are combined, Lennard-Jones potential energy is taken as an evaluation function of an optimization algorithm, the diversity of particles is increased through cross and variation of individuals, the optimization search efficiency is improved by using preferential selection and local search of the individuals, performance in aspects of computation complexity, rapid convergence, the global situation and the like are considered comprehensively, and the global optimum stable structure of an atom cluster is obtained.
Description
Technical field
The present invention relates to molecular structure and intelligent optimization field, specifically a kind of ar atmo Cluster Structures optimisation technique based on Memetic algorithm.
Background technology
Ar atmo cluster be several so that thousands of atoms, molecule or ions by the metastable microcosmic of physics or chemical binding force composition and submicroscopic system.Many character of cluster all depend on its rock-steady structure, and the Stable structure research of elementide is the advanced subject that Present Domestic is studied outward.A most stable structure of elementide is all generally the geometry with lowest potential energy.Thus, the key solving the Stable structure problem of elementide is how by finding lowest potential energy state to the search of potential energy surface.At present, genetic algorithm, improve PSO algorithm etc. are used in the optimization problem of clustering architecture by people.But there is speed of convergence and slowly, be easily absorbed in local minimum and for shortcomings such as larger cluster system optimization weak effects in these algorithms.
Summary of the invention
The object of the invention is to the deficiency overcoming existing ar atmo Cluster Structures optimization method, a kind of ar atmo Cluster Structures optimisation technique based on Memetic algorithm is proposed, with the evaluation function that Lennard-Jones potential energy is optimized algorithm in searching process, make full use of the global search based on population and the partial heuristic search based on individuality, by optimizing Species structure, reject bad individuality early, accelerate the solving speed of algorithm, ensureing on the basis compared with better utility energy, improve ability of searching optimum, obtain high-quality solution, to reach the effect of optimization realizing most stable cluster structure.
The technical solution used in the present invention comprises following steps:
Step 1: coding.According to the solution space of elementide Atom location variable, feasible solution tables of data is shown as the floating type string structure data of search volume, these string structure data various combinations form different feasible solution.
Step 2: produce initial population.Determine the atom number n of ar atmo cluster, search population scale M, solution space dimension size D=3n, probability of crossover p
c, mutation probability p
v, random generation M initial individuals, evolutionary generation variable k=1, maximum evolutionary generation K
max.
Step 3: intersect.According to probability of crossover p
cin M individuality, choose arbitrarily two carry out hybridization computing, two that produce colony of new generation new individual.
Step 4: variation.Mix in raw new colony, according to mutation probability p in hybridization computing
vtherefrom choose several body, carry out mutation operation.
Step 5: calculate fitness function.New colony after variation is calculated respectively to the fitness function of each individuality according to Lennard-Jones potential energy, its formula is:
Wherein, r
ijrepresent the Euclidean distance between the atom i of m individuality and atom j; ε represents potential well depth, usually can be taken as 1; σ representative collision (potential energy is zero) time nuclear pitch from, get
Step 6: select.From current group, select the individuality of M excellent (fitness is high), select probability is directly proportional to its fitness, gives up the individuality that fitness is low.
Step 7: Local Search.Quasi-Newton method is adopted to carry out Local Search to all individualities in population.
Step 8: if meet stop condition or reach maximum iteration time (k=K
max), then optimizing terminates, the global optimum obtained, and is the optimum structure distribution of ar atmo cluster Atom; Otherwise k:=k+1, goes to step 3.
Wherein,
In described step 2, search population scale M and maximum evolutionary generation Kmax needs to set according to concrete problem scale, search population scale general span be [20 40], and the atom number in ar atmo cluster is more, search for population scale and required maximum evolutionary generation larger.
In described step 7, different local searching strategy can being adopted for different optimization problems, as climbing method, simplex optimization method, method of conjugate gradient, Newton method etc., here in order to improve local convergence speed, adopting quasi-Newton method.
Compared with the prior art the present invention has the following advantages: ar atmo Cluster Structures optimisation technique of the present invention combines the range advantage of colony's algorithm search and the degree of depth advantage of local search algorithm, with the evaluation function that Lennard-Jones potential energy is optimized algorithm, the diversity of particle is added by the crossover and mutation between individuality, utilize individual preferentially choosing with Local Search to improve Optimizing Search efficiency, consider computation complexity, fast convergence, effective search ability, the aspect performance such as of overall importance, to obtain global optimum's rock-steady structure of elementide.
Accompanying drawing explanation
Fig. 1 is the Structure of Atomic Clusters optimisation technique process flow diagram based on the present invention program.
Fig. 2 is the result optimized different Structure of Atomic Clusters based on the present invention program.
Embodiment
In order to understand technical scheme of the present invention better, below embodiment is described in further detail, and in conjunction with an application example, embodiment is described, but be not limited thereto.
Embodiment: with ar atmo cluster Ar
nthe structure optimization of (n=3 ~ 30) is example, adopts population scale M=20, maximum evolutionary generation K
max=7.
As shown in Figure 1, embodiment can be divided into the following steps to the inventive method workflow:
(1) encode.By elementide Atom variable solution space [-2 2]
dfeasible solution tables of data be shown as the floating type string structure data of search volume, wherein D represents solution space dimension size, and these string structure data various combinations form different feasible solution.
(2) initial population is produced.According to the atom number n (n=3 ~ 30) of different ar atmo cluster, determine solution space dimension size D=3n, search population scale M=20 is set, gets probability of crossover p
c=0.95, mutation probability p
v=0.1, random generation M initial individuals, determines evolutionary generation variable k=1, maximum evolutionary generation K
max.
(3) intersect.According to probability of crossover p
cin M individuality, choose arbitrarily two carry out hybridization computing, two that produce colony of new generation new individual.
(4) make a variation.Mix in raw new colony, according to mutation probability p in hybridization computing
vtherefrom choose several body, carry out mutation operation.
(5) fitness function is calculated.New colony after variation is calculated respectively to the fitness function of each individuality according to Lennard-Jones potential energy, its formula is:
Wherein, r
ijrepresent the Euclidean distance between the atom i of m individuality and atom j; ε=1 represents potential well depth; σ representative collision (potential energy is zero) time nuclear pitch from, get
(6) select.From current group, select the individuality of M excellent (fitness is high), select probability is taken as
Give up the individuality that fitness is low.
(7) Local Search.Quasi-Newton method is adopted to carry out Local Search to all individualities in population.
(8) if meet stop condition or reach maximum iteration time (k=K
max), then optimizing terminates, the global optimum obtained, and is the optimum structure distribution of ar atmo cluster Atom; Otherwise k:=k+1, turns (3).
Fig. 2 shows the inventive method and conventional simulation annealing algorithm (SA), the comparative result of genetic algorithm (GA) in part ar atmo Cluster Structures is optimized.Algorithm SA and GA all adopts population scale M=20 in optimizing process, maximum evolutionary generation K
max=1000, and random repetitive cycling 50 times, choose optimal result in 50 results and compare.As seen from the figure, when ar atmo cluster is less, i.e. n=3,4, when 5, three kinds of algorithms can obtain optimum structure, but when ar atmo cluster number increases, the inventive method still can be found out close to optimum structure, show good Optimal performance, SA and GA is then along with the increase of ar atmo cluster number, although evolutionary generation is abundant, optimum results is far from reaching optimum.As can be seen here, although the inventive method is compared on computing time, SA, GA not too large advantage, convergence efficiency and Optimal performance is far superior to above-mentioned two kinds of algorithms, and has better robustness.
Above the ar atmo Cluster Structures optimisation technique based on Memetic algorithm of the present invention has been described in detail, but specific implementation form of the present invention is not limited thereto.Concerning the those skilled in the art of the art, the various apparent change carried out it when not deviating from spirit and the right of the method for the invention is all within protection scope of the present invention.
Claims (3)
1., based on an ar atmo Cluster Structures optimisation technique for Memetic algorithm, it is characterized in that described method comprises the steps:
Step 1: coding.According to the solution space of elementide Atom location variable, feasible solution tables of data is shown as the floating type string structure data of search volume, these string structure data various combinations form different feasible solution.
Step 2: produce initial population.Determine the atom number n of ar atmo cluster, search population scale M, solution space dimension size D=3n, probability of crossover p
c, mutation probability p
v, random generation M initial individuals, evolutionary generation variable k=1, maximum evolutionary generation K
max.
Step 3: intersect.According to probability of crossover p
cin M individuality, choose arbitrarily two carry out hybridization computing, two that produce colony of new generation new individual.
Step 4: variation.Mix in raw new colony, according to mutation probability p in hybridization computing
vtherefrom choose several body, carry out mutation operation.
Step 5: calculate fitness function.New colony after variation is calculated respectively to the fitness function of each individuality according to Lennard-Jones potential energy, its formula is:
Wherein, r
ijrepresent the Euclidean distance between the atom i of m individuality and atom j; ε represents potential well depth, usually can be taken as 1; σ representative collision (potential energy is zero) time nuclear pitch from, get
Step 6: select.From current group, select the individuality of M excellent (fitness is high), select probability is directly proportional to its fitness, gives up the individuality that fitness is low.
Step 7: Local Search.Quasi-Newton method is adopted to carry out Local Search to all individualities in population.
Step 8: if meet stop condition or reach maximum iteration time (k=K
max), then optimizing terminates, the global optimum obtained, and is the optimum structure distribution of ar atmo cluster Atom; Otherwise k:=k+1, goes to step 3.
2. a kind of ar atmo Cluster Structures optimisation technique based on Memetic algorithm according to claim 1, is characterized in that in described step 2, search population scale M and maximum evolutionary generation K
maxneed to set according to concrete problem scale, search population scale general span be [20 40], and the atom number in ar atmo cluster is more, search for population scale and required maximum evolutionary generation larger.
3. a kind of ar atmo Cluster Structures optimisation technique based on Memetic algorithm according to claim 1, it is characterized in that in described step 7, local searching strategy can adopt climbing method, simplex optimization method, method of conjugate gradient, Newton method etc., the present invention, in order to improve local convergence speed, adopts quasi-Newton method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410070222.0A CN104866514A (en) | 2014-02-26 | 2014-02-26 | Argon atom cluster structure optimization technology based on Memetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410070222.0A CN104866514A (en) | 2014-02-26 | 2014-02-26 | Argon atom cluster structure optimization technology based on Memetic algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104866514A true CN104866514A (en) | 2015-08-26 |
Family
ID=53912348
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410070222.0A Pending CN104866514A (en) | 2014-02-26 | 2014-02-26 | Argon atom cluster structure optimization technology based on Memetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104866514A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330302A (en) * | 2017-07-10 | 2017-11-07 | 无锡职业技术学院 | The biological die body reconstructing method of joint Memetic algorithms and S system |
| CN115579050A (en) * | 2022-12-08 | 2023-01-06 | 香港中文大学（深圳） | Method and system for searching key original set in biomolecule functional dynamics |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020099929A1 (en) * | 2000-11-14 | 2002-07-25 | Yaochu Jin | Multi-objective optimization |
| CN103116703A (en) * | 2013-02-04 | 2013-05-22 | 西安交通大学 | Synergic variation differential evolutionary algorithm for high-dimensional parameter space wave form inversion |
-
2014
- 2014-02-26 CN CN201410070222.0A patent/CN104866514A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020099929A1 (en) * | 2000-11-14 | 2002-07-25 | Yaochu Jin | Multi-objective optimization |
| CN103116703A (en) * | 2013-02-04 | 2013-05-22 | 西安交通大学 | Synergic variation differential evolutionary algorithm for high-dimensional parameter space wave form inversion |
Non-Patent Citations (2)
| Title |
|---|
| 曾思琴: "基于智能优化算法的原子团簇结构研究", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
| 赖向京: "原子团簇结构预测的现实途径——高性能启发式算法", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330302A (en) * | 2017-07-10 | 2017-11-07 | 无锡职业技术学院 | The biological die body reconstructing method of joint Memetic algorithms and S system |
| CN115579050A (en) * | 2022-12-08 | 2023-01-06 | 香港中文大学（深圳） | Method and system for searching key original set in biomolecule functional dynamics |
| CN115579050B (en) * | 2022-12-08 | 2023-03-14 | 香港中文大学（深圳） | Method and system for searching key original set in biomolecule functional dynamics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Wind power forecasting based on singular spectrum analysis and a new hybrid Laguerre neural network | |
| CN103971174B (en) | Hydropower station group optimized dispatching method based on improved quantum-behaved particle swarm algorithm | |
| CN103276686A (en) | Method of determining downstream design flood of cascade reservoirs | |
| Wang et al. | A novel control strategy approach to optimally design a wind farm layout | |
| CN103972908A (en) | Multi-target reactive power optimization method based on adaptive chaos particle swarm algorithm | |
| CN104598565B (en) | A kind of K mean value large-scale data clustering methods based on stochastic gradient descent algorithm | |
| CN105356465B (en) | Transmission Expansion Planning in Electric platform and application that a kind of economy is coordinated with security | |
| Zhang et al. | Fuzzy time series forecasting based on k-means clustering | |
| CN105974799A (en) | Fuzzy control system optimization method based on differential evolution-local unimodal sampling algorithm | |
| CN104764980A (en) | Positioning method for distribution circuit fault section based on BPSO and GA | |
| CN104268077A (en) | Chaos genetic algorithm based test case intensive simple algorithm | |
| CN105373856A (en) | Wind electricity power short-term combined prediction method considering run detection method reconstruction | |
| CN115238993A (en) | Reservoir group multi-objective optimization scheduling scheme decision system based on NSGA-II algorithm | |
| CN104182806A (en) | Optimal operation method of hydropower station group on the basis of orthogonal dimensionality reduction search algorithm | |
| CN103903072A (en) | High-dimensional multi-target set evolutionary optimization method based on preference of decision maker | |
| CN103530531B (en) | A kind of wind power lasting characteristic describing method of estimating based on maximum likelihood | |
| CN106372739A (en) | Demand response effect evaluation method based on demand response baseline | |
| CN104866514A (en) | Argon atom cluster structure optimization technology based on Memetic algorithm | |
| CN102937946B (en) | A kind of complicated function minimal value searching method based on constraint regular pattern | |
| CN104598657B (en) | A kind of gene die body reconstructing method based on memetic algorithms | |
| CN101719194B (en) | Artificial gene regulatory network simulation method | |
| CN104318008B (en) | A kind of condenser Optimization Design based on loose constraint heredity simplex algorithm | |
| CN106815656B (en) | Method for acquiring cascade reservoir energy storage dispatching diagram | |
| CN102768735B (en) | Network community partitioning method based on immune clone multi-objective optimization | |
| CN104392317A (en) | Project scheduling method based on genetic culture gene algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150826 |