Hybrid Single Parent-Offspring MVMO for Solving CEC2018 Computationally Expensive Problems

Conference paper (2018)

Authors

José L. Rueda Intelligent Electrical Power Grids - EEMCS
I Erlich Universität Duisburg-Essen
Research Group
Intelligent Electrical Power Grids (EEMCS) (TU Delft)
Copyright: © 2018 José L. Rueda, Istvan Erlich
DOI: https://doi.org/10.1109/CEC.2018.8477807
More Info
expand_more

Published Date

2018

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Electrical Engineering, Mathematics and Computer Science

Department

Electrical Sustainable Energy

Research Group

Intelligent Electrical Power Grids

Abstract

Mean-Variance Mapping Optimization (MVMO) belongs to the family of evolutionary algorithms, and has proven to be competitive in solving computationally expensive problems proposed in the Icompetitions CEC2014, CEC2015, and CEC2016. MVMO can tackle such problems by evolving a set of solutions (population based approach) or a single solution (single parent-offspring approach). The evolutionary mechanism of MVMO performs within a normalized search space in the range [0, 1]. The power of MVMO stems from its ability - based on statistical analysis of the evolving solution based on a mapping function - to adaptively shift the search priority from exploration to exploitation. This paper introduces a newly defined mapping function as well as a new rule for using an embedded local search strategy, and presents several tests conducted by using the test bed of the CEC2018 competition. Numerical results indicate significant improvements on the results obtained in CEC2016 competition.