差分粒子群算法DEPSO Java源代码

#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. #Setting parameters: NAME VALUE_type Range DefaultV Description Problem String * Michalewicz_G2 The problem to be solved N integer >5 70 The number of agents T integer >1 2000 The maximum learning cycles 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 c1 real [0, 2] 1.494 PSO: learning factor for pbest c2 real [0, 2] 1.494 PSO: learning factor for gbest w real [0, 1] 0.729 PSO: inertia weight #Output Informaton: 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 #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), Compi�gne, France, 2005: 38-44. * -> For boundary-handling: the cycled version for PSO, the random version for DE * * Summary: * The basic difference from the original DEPSO algorithm [1] is that using the cycled * version [3] instead of the periodic version for boundary-handling, so that all new * solutions are generated within the original search space, as well as allows the * agents searching within a virtually infinite space. * Another improvement is that the adaptive constraints relaxing (ACR) rule [2], which * might tackle the problem with equality constraints more efficiently than the basic * constraint-handling (BCH) rule does. # 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: URL: http://www.adaptivebox.net/research EMAIL: xiexiaofeng@tsinghua.org.cn