Multiobective optimization using compromise programming and an immune algorithm

F Campelo, FG Guimaraes, H Igarashi

    Research output: Contribution to journalArticlepeer-review

    Abstract

    The m-AINet is a modified version of the artificial immune network algorithm for single-objective and multimodal optimization (opt-AINet), with constraint-handling capability and improvements aiming to reduce the computational effort required by the original algorithm. In this paper we extend this algorithm for multiobjective problems by using the compromise programming approach to aggregate the objectives in the evaluation step. We adopt a compromise programming formulation that is theoretically able to map the whole Pareto front. The proposed multiobjective m-AINet is tested on the design of a loudspeaker with two objectives, showing promising results.
    Original languageEnglish
    Pages (from-to)982-985
    JournalIEEE Transactions on Magnetics
    Volume44
    Issue number6
    DOIs
    Publication statusPublished - 20 May 2008

    Fingerprint

    Dive into the research topics of 'Multiobective optimization using compromise programming and an immune algorithm'. Together they form a unique fingerprint.

    Cite this