A statistics-based genetic algorithm for quality improvements of power supplies

K. Y. Chan, K. W. Chan, Glory T.Y. Pong, M. E. Aydin, T. C. Fogarty, S. H. Ling

Research output: Journal article publicationJournal articleAcademic researchpeer-review

11 Citations (Scopus)

Abstract

This paper presents a new statistics-based evolutionary algorithm to improve the qualities of power supplies, in which operational costs and the stability of the power supply are optimised to provide a highly smooth but low-cost power supply service to customers. The proposed method is incorporated with the characteristics of the stochastic method, evolutionary algorithm and a more systematical statistical method, orthogonal design. It intends to compensate for the built-in randomness of the stochastic method and, at the same time, overcome the limitations of local search methods that are not suitable for handling multi-optima problems. Case studies on the WSCC 9-bus and New England 39-bus systems indicate that the proposed approach outperforms the existing method in terms of robustness in solution and convergence speed while the solution quality that can offer a more stable and cheaper power supply to customers is achieved.
Original languageEnglish
Pages (from-to)468-492
Number of pages25
JournalEuropean Journal of Industrial Engineering
Volume3
Issue number4
DOIs
Publication statusPublished - 29 Jul 2009

Keywords

  • Evolutionary algorithm
  • Orthogonal array
  • Power supplies
  • Power systems

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Cite this