An Analysis of Particle Properties on a Multi-swarm PSO for Dynamic Optimization Problems

The particle swarm optimization (PSO) algorithm has successfully been applied to dynamic optimization problems with very competitive results. One of its best performing variants is the one based on the atomic model, with quantum and trajectory particles. However, there is no precise knowledge on how these particles contribute to the global behavior of the swarms during the optimization process. This work analyzes several aspects of each type of particle, including the best combination of them for different scenarios, and how many times do they contribute to the swarm's best. Results show that, for the Moving Peaks Benchmark (MPB), a higher number of trajectory particles than quantum particles is the best strategy. Quantum particles are most helpful immediately after a change in the environment has occurred, while trajectory particles lead the optimization in the final stages. Suggestions on how to use this knowledge for future developments are also provided.

[1]  Hartmut Schmeck,et al.  Designing evolutionary algorithms for dynamic optimization problems , 2003 .

[2]  Jürgen Branke,et al.  Memory enhanced evolutionary algorithms for changing optimization problems , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[3]  Michael N. Vrahatis,et al.  Recent approaches to global optimization problems through Particle Swarm Optimization , 2002, Natural Computing.

[4]  Xiaodong Li,et al.  A particle swarm model for tracking multiple peaks in a dynamic environment using speciation , 2004, Proceedings of the 2004 Congress on Evolutionary Computation (IEEE Cat. No.04TH8753).

[5]  James Kennedy,et al.  Particle swarm optimization , 2002, Proceedings of ICNN'95 - International Conference on Neural Networks.

[6]  Rolf Drechsler,et al.  Applications of Evolutionary Computing, EvoWorkshops 2008: EvoCOMNET, EvoFIN, EvoHOT, EvoIASP, EvoMUSART, EvoNUM, EvoSTOC, and EvoTransLog, Naples, Italy, March 26-28, 2008. Proceedings , 2008, EvoWorkshops.

[7]  R. Eberhart,et al.  Comparing inertia weights and constriction factors in particle swarm optimization , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).

[8]  T. Blackwell,et al.  Particle swarms and population diversity , 2005, Soft Comput..

[9]  P. John Clarkson,et al.  Erratum: A Species Conserving Genetic Algorithm for Multimodal Function Optimization , 2003, Evolutionary Computation.

[10]  Jürgen Branke,et al.  Multi-swarm Optimization in Dynamic Environments , 2004, EvoWorkshops.

[11]  Shengxiang Yang,et al.  Evolutionary Computation in Dynamic and Uncertain Environments , 2007, Studies in Computational Intelligence.

[12]  Julian F. Miller,et al.  Genetic and Evolutionary Computation — GECCO 2003 , 2003, Lecture Notes in Computer Science.

[13]  Martin Middendorf,et al.  A Hierarchical Particle Swarm Optimizer for Dynamic Optimization Problems , 2004, EvoWorkshops.

[14]  Maurice Clerc,et al.  The particle swarm - explosion, stability, and convergence in a multidimensional complex space , 2002, IEEE Trans. Evol. Comput..

[15]  Jürgen Branke,et al.  Multiswarms, exclusion, and anti-convergence in dynamic environments , 2006, IEEE Transactions on Evolutionary Computation.

[16]  Tim M. Blackwell,et al.  Swarms in Dynamic Environments , 2003, GECCO.

[17]  Tim Blackwell,et al.  Particle Swarm Optimization in Dynamic Environments , 2007, Evolutionary Computation in Dynamic and Uncertain Environments.