Simple and Adaptive Particle Swarms

Doctoral thesis English OPEN
Bratton, Daniel
  • Subject: G490
    acm: MathematicsofComputing_NUMERICALANALYSIS

The substantial advances that have been made to both the theoretical and practical aspects of particle\ud swarm optimization over the past 10 years have taken it far beyond its original intent as a biological\ud swarm simulation. This thesis details and explains these advances in the context of what has been\ud achieved to this point, as well as what has yet to be understood or solidified within the research community.\ud Taking into account the state of the modern field, a standardized PSO algorithm is defined for\ud benchmarking and comparative purposes both within the work, and for the community as a whole.\ud \ud This standard is refined and simplified over several iterations into a form that does away with potentially\ud undesirable properties of the standard algorithm while retaining equivalent or superior performance\ud on the common set of benchmarks. This refinement, referred to as a discrete recombinant swarm (PSODRS)\ud requires only a single user-defined parameter in the positional update equation, and uses minimal\ud additive stochasticity, rather than the multiplicative stochasticity inherent in the standard PSO. After a\ud mathematical analysis of the PSO-DRS algorithm, an adaptive framework is developed and rigorously\ud tested, demonstrating the effects of the tunable particle- and swarm-level parameters. This adaptability\ud shows practical benefit by broadening the range of problems which the PSO-DRS algorithm is wellsuited\ud to optimize.
  • References (42)
    42 references, page 1 of 5

    4 A Simplified, Recombinant PSO Algorithm 68 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 4.2 PSO with Discrete Recombination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 4.3 Simplifying Recombinant PSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 4.3.1 PSO-DR Model 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4.3.2 PSO-DR Model 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4.4 Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 4.4.1 Dip Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 4.5 Velocity Bursts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 4.5.1 Bursting under PSO-DR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 4.6 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 4.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

    [11] M. Clerc. Tribes, a parameter free particle swarm optimizer. In Proc of the OEP, Paris, France, 2003.

    [12] R. Mendes. Population Topologies and their Influence in Particle Swarm Performance. PhD thesis, Escola de Engenharia, Universidade do Minho, 2004.

    [13] G.B. Dantzig. Programming of interdependent activities II. Mathematical model. Econometrica, 17:200-211, 1949.

    [14] N. Karmarkar. A new polynomial-time algorithm for linear programming. Combinatorica, 4:373- 395, 1984.

    [25] C. Schumacher, M. D. Vose, and L. D. Whitley. The No Free Lunch and problem description length. In Proc of the Genetic and Evolutionary Computation Conference (GECCO 2001), pages 565-570, San Francisco, CA, USA, 2001.

    [26] T. Ba┬Ęck. Evolutionary Algorithms in Theory and Practice. Oxford University Press, USA, 1995.

    [27] A.S. Fraser. Simulation of genetic systems by automatic digital computers. Aust. J. Biol. Sci., 10, 1957.

    [28] A. Fraser and D. Burnell. Computer Models in Genetics. New York, McGraw-Hill, 1970.

    [29] J.L. Crosby. Computer Simulation in Genetics. London, John Wiley & Sons, 1973.

  • Metrics
    0
    views in OpenAIRE
    0
    views in local repository
    43
    downloads in local repository

    The information is available from the following content providers:

    From Number Of Views Number Of Downloads
    Goldsmiths Research Online - IRUS-UK 0 43
Share - Bookmark