Advanced optimization by progressive mapping search method of PSO and neural network

Dong Hwa Kim, Jin Ill Park, X. Z. Gao

    Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

    Abstract

    This paper proposes a novel optimization approach by the fusion of the Progressive Mapping Search Method (PMSM) and the Neural Network (NN) aided Particle Swarm Optimization (PSO) that can obtain the global optimal solutions easily and speed up the overall search procedure. The PMSM merged with the NN and PSO has an important role as the navigation when the PSO is searching all the areas in order to acquire the optimum. It can help to improve the search capability of the original PSO method. That is, the PMSM together with the NN and PSO is trained to capture the PSO-searched solutions. To verify and demonstrate the effectiveness of our technique, we use a total of four test functions. The PMSM strategy employed in our paper is faster than the traditional PSO algorithm in all these four test functions. We also apply this new optimization scheme in the AVR (Automatic Voltage Regulator) system of the thermal power plant, which has resulted in faster and more stable responses.

    Original languageEnglish
    Title of host publicationSwarm, Evolutionary, and Memetic Computing - 4th International Conference, SEMCCO 2013, Proceedings
    Pages625-638
    Number of pages14
    Volume8298 LNCS
    EditionPART 2
    ISBN (Electronic)978-3-319-03756-1
    DOIs
    Publication statusPublished - 2013
    MoE publication typeA4 Article in a conference publication
    EventInternational Conference on Swarm, Evolutionary and Memetic Computing - Chennai, India
    Duration: 19 Dec 201321 Dec 2013
    Conference number: 4

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    NumberPART 2
    Volume8298 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Conference

    ConferenceInternational Conference on Swarm, Evolutionary and Memetic Computing
    Abbreviated titleSEMCCO
    CountryIndia
    CityChennai
    Period19/12/201321/12/2013

    Keywords

    • hybrid systems
    • learning systems
    • neural network
    • optimization
    • particle swarm optimization
    • Progressive mapping search method

    Fingerprint

    Dive into the research topics of 'Advanced optimization by progressive mapping search method of PSO and neural network'. Together they form a unique fingerprint.

    Cite this