PSO.zip_PSO_PSO+matlab_matlab pso_pso matlab_swarm

Particle Swarm Optimization Algorithms Library ----------------------------------------- This is a library in JAVA created by Pablo David Gutiérrez Pérez [email protected] for its own use and it is released as free software (license GPL) for being used by other researchers in Soft Computing. This library is created to give a framework for real coding optimizations using particle swarm optimization. Concretely, this library implements a total of nine methods with interesting approaches in PSO. * PSO: Basic particle swarm optimization with inertia weight. J. Kennedy, R. Eberhart Particle Swarm Optimization, IEEE International Conference on Neural Networks:1942-1948 (1995) * Selection PSO: P. Angelinne Using Selection to Improve Particle Swarm Optimization, IEEE icec:84-90 (1998) * Constriction PSO: M. Clerc, J. Kennedy , The particle swarm-explosion, stability and convergence in a multidimensional complex space, IEEE Transations on Evolutinary Computation: 58-73 (2002) * Stretching PSO: K.E. Parsopoulos, M.N. Vrahatis Recen approaches to global optimization problems through Particle Swarm Optimization Natural Computing 1 (2002) 235-306 * Composite PSO: K.E. Parsopoulos, M.N. Vrahatis Recen approaches to global optimization problems through Particle Swarm Optimization Natural Computing 1 (2002) 235-306 * Fully Informed PSO: R. Mendes, J. Kennedy, J. Neves, The Fully Informed Particle Swarm: Simpler, Maybe Better, IEEE Trans. Evol. Comput. vol 8:3 (2004) * Hierarchical PSO: S. Janson, M. Middendorf, A Hierarchical Particle Swarm Optimizer and Its Adaptative Variant, IEEE Trasn. on Systems, Man, And Cybernetics Vol 35:6 (2005) 1272-1282 * Frankenstein PSO: M.A. Montes de Oca, T. Stützle, M. Birattari, M. Dorigo, Frankenstein's PSO: A Composite Particle Swarm Optimization Algorithm IEEE Trans. Evol. Comput. Vol 13:5 (2009) pp. 1120-1132 * Orthogonal Learning PSO: Z-H Zhan, J. Zhang, Y. Li, Y-H. Shi, Orthogonal Learning Particle Swarm Optimization, IEEE Trans. Evol. Comput. (2011) The aim of this library is to allow the researcher to focus on its PSO, avoiding to make the same code several times for each PSO. This library has been implemented to be able to run over different plataforms. How to get it ------------- The last stable version are available from the research group of the author: Soft Computing and Intelligent Information Systems http://sci2s.ugr.es/EAMHCO/ This code is self-contained, it does not require any other non-standard library. Features of the library ----------------------- * All the algorithms have been implemented to allow the user to set the number of particles used, the function to minimize, the random seed and another parameters from every algorithm via command line. * Every method has its own parameters and shows a help if any are not properly set, showing how to use it. Install guide ------------- It is required for install the software: * Java. It has been tested with jde 1.6. Steps: * Compile the software, - ant build.xml or compiling directly the files in the package cec05 and any of the oder packages. * If the execution fails check the paths in cec05 package and recompile. * The code provided is ready to run with Netbeans and using the jar file in dist from the top folder. Examples -------- * PSO global: java -cp dist/PSO.jar Basic_PSO.basic_pso 1 10 12345678 40 0 * PSO social: java -cp dist/PSO.jar Basic_PSO.basic_pso 1 10 12345678 40 1 2 * Selection PSO: java -cp dist/PSO.jar Selection_PSO.pso 1 10 12345678 40 0 * Constriction PSO: java -cp dist/PSO.jar Constriction_PSO.pso 1 10 12345678 40 0 * Stretching PSO: java -cp dist/PSO.jar Stretching_PSO.pso 1 10 12345678 40 0 * Composite PSO: java -cp dist/PSO.jar Composite_PSO.pso 1 10 12345678 40 0 * Fully Informed PSO: java -cp dist/PSO.jar FIPSO.pso 1 10 12345678 40 0 1 1 * Hierarchical PSO: java -cp dist/PSO.jar -cp H_PSO.h_pso 1 10 12345678 40 4 5 0 * Frankenstein PSO: java -cp PSO.jar dist/Frankensteing_PSO.frankensteing_pso 1 10 12345678 40 2 2 * Orthogonal Learning PSO: java -cp dist/PSO.jar OLPSO.olpso 1 10 12345678 40 0 5 You can find more information about the parameters of these algorithms in their respectives papers.