PSO
所属分类:Java编程
开发工具:Java
文件大小:76KB
下载次数:41
上传日期:2010-12-03 15:20:09
上 传 者:
hcsleep
说明: 改进的差分PSO算法,Java实行的,有兴趣的可以参考下
(PSO algorithm to improve the differential, Java implementation, and are interested can refer to the following)
文件列表:
deps\.DS_Store (6148, 2009-06-27)
deps\release\.DS_Store (6148, 2009-06-20)
deps\release\DEPSAgent.class (3984, 2009-06-20)
deps\release\DEPSGroup.class (3549, 2009-06-20)
deps\release\DEPSO.class (5894, 2009-06-20)
deps\release\encode\EvalElement.class (927, 2009-06-20)
deps\release\encode\EvalStruct.class (974, 2009-06-20)
deps\release\encode\IEncodeEngine.class (145, 2009-06-20)
deps\release\Global\BasicBound.class (769, 2009-06-20)
deps\release\Global\GlobalString.class (1893, 2009-06-20)
deps\release\Global\OutputMethods.class (1232, 2009-06-20)
deps\release\Global\RandomGenerator.class (2144, 2009-06-20)
deps\release\goodness\ACRComparator.class (1655, 2009-06-20)
deps\release\goodness\BCHComparator.class (615, 2009-06-20)
deps\release\goodness\IGoodnessCompareEngine.class (282, 2009-06-20)
deps\release\goodness\IUpdateCycleEngine.class (155, 2009-06-20)
deps\release\ICycleOutputEngine.class (146, 2009-06-20)
deps\release\knowledge\IStateSetEngine.class (201, 2009-06-20)
deps\release\knowledge\SearchPoint.class (1631, 2009-06-20)
deps\release\knowledge\SearchPointSet.class (1012, 2009-06-20)
deps\release\knowledge\StateInfoHandler.class (1886, 2009-06-20)
deps\release\problem\.DS_Store (6148, 2009-05-19)
deps\release\problem\constrained\.DS_Store (6148, 2009-04-30)
deps\release\problem\constrained\Michalewicz_G1.class (1331, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G10.class (1224, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G11.class (898, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G12.class (1197, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G13.class (1090, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G2.class (1308, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G3.class (1071, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G4.class (1288, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G5.class (1384, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G6.class (1108, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G7.class (1806, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G8.class (1039, 2009-06-20)
deps\release\problem\constrained\Michalewicz_G9.class (1455, 2009-06-20)
deps\release\problem\constrained\PressureVessel.class (1301, 2009-06-20)
deps\release\problem\constrained\WeldedBeam.class (1650, 2009-06-20)
deps\release\problem\ProblemEncoder.class (2661, 2009-06-20)
... ...
#Software description:
The DEPSO algorithm V1.0.001 for (constrained) numerical optimization
#Command-line examples
$ cd release
$ java DEPSO [NAME=VALUE] ...
//See release\RUNExample.bat for more examples.
$ cd release
$ sh RUNExample.bat
#Problems to be solved
For existing examples, please refers to the files located in the directories:
problem/constrained and problem/unconstrained
For creating new instances, please refers to the source code files:
1) problem/constrained/Michalewicz_G1.java, for constrained optimization;
2) problem/constrained/Michalewicz_G3.java, for equality constraints;
3) problem/unconstrained/GoldsteinPrice.java, for unconstrained optimization.
Please see http://www.adaptivebox.net/doi/DEPSO for downloading the source code
of the up-to-date list of implemented benchmark instances.
#Setting parameters:
NAME VALUE_type Range DefaultV Description
Problem String *
The problem to be solved
//For example: problem.constrained.Michalewicz_G2 is the default value
----------------------------------------------------------------------------------
N integer >5 70 The number of agents
T integer >1 2000 The maximum learning cycles
isACR boolean false Constraint-handling: BCH(false), ACR(true)
//Basic constraint-handling (BCH) rule or adaptive constraints relaxing (ACR) rule
Tout integer >0 100 The output interval (not important)
//The total evalution times is about N*T
FACTOR real (0, 1.2] 0.5 DE: scale constant
CR real [0, 1] 0.9 DE: crossover constant
//The parameters of DE operator
c1 real [0, 2] 1.494 PSO: learning factor for pbest
c2 real [0, 2] 1.494 PSO: learning factor for gbest
weight real [0, 1] 0.729 PSO: inertia weight
//The parameters of PSO operator
#Output Information:
The program outputs information of the best solution every "Tout" cycles.
At the end, it outputs the location and optimum values of the best point.
//Vcon: the weighted constraint violation value (>=0): if Vcon==0, then no violation
//Vopt: the value of objective function
#Version History
-> Version: V1.0.001 (http://www.adaptivebox.net/research/download/maos/depso/depsV1.0.001.zip)
* For boundary-handling, the cycled version [3] instead of the periodic version [1] is considered,
so that all new solutions are generated within the original search space, as well as the agents
are searching within a virtually infinite space.
* The adaptive constraints relaxing (ACR) rule [2] might tackle the problem with equality constraints
more efficiently than the basic constraint-handling (BCH) rule does.
* Newly introduced parameters: isACR
-> Version: V1.0.000 (http://www.adaptivebox.net/research/download/maos/depso/depsV1.0.000.zip)
* It implements the original DEPSO algorithm in [1].
* Setting parameters: Problem, N, T, Tout, FACTOR, CR, c1, c2, weight
#Reference:
[1] W.-J. Zhang, X.-F Xie. DEPSO: Hybrid particle swarm with differential evolution
operator. IEEE International Conference on Systems, Man & Cybernetics, Washington,
DC, USA, 2003: 3816-3821.
-> For the original DEPSO algorithm
[2] X.-F. Xie, W.-J. Zhang, D.-C. Bi. Handling equality constraints by adaptive
relaxing rule for swarm algorithms. Congress on Evolutionary Computation (CEC),
Portland, OR, USA, 2004: 2012-2016.
-> For handling equality constraints: the adaptive constraint-relaxing rule (ACR)
[3] X.-F. Xie, J. Liu. A compact multiagent system based on autonomy oriented
computing, IEEE/WIC/ACM International Conference on Intelligent Agent Technology
(IAT), Compigne, France, 2005: 38-44.
-> For boundary-handling: the cycled version for PSO, the random version for DE
# License description:
*******************************************************************
* DEPSO is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 3.0 of the License, or (at your option) any later version.
*
* DEPSO is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License 3.0 for more details.
*
* Please acknowledge the author(s) if you use this code in any way.
*******************************************************************
#Contact Information:
Portal: http://www.adaptivebox.net/doi/DEPSO
EMAIL: xiexiaofeng@tsinghua.org.cn
近期下载者:
相关文件:
收藏者: