Catalog Home Page

A Java-based parallel platform for the implementation of evolutionary computation for engineering applications

Fung, C.C., Li, J.B., Wong, K.W. and Wong, K.P. (2004) A Java-based parallel platform for the implementation of evolutionary computation for engineering applications. International Journal of Systems Science, 35 (13/14). pp. 741-750.

Link to Published Version: http://dx.doi.org/10.1080/00207720412331303651
*Subscription may be required

Abstract

As computers continually improve in performance and decrease in manufacturing cost, distributed systems consisting of multiple computers implemented as parallel computation platforms have become viable for engineering applications which demand intensive computation power. This paper proposes an extended version of a previously developed low cost parallel computation platform called para worker. The system is based on a cluster structure which is a form of a distributed system. The new system is termed para worker 2 which differentiates it from the earlier system. The new proposed system adds enhanced features of improved dynamic object reallocation, adaptive consistency protocols, and location transparency. Performance of the para worker 2 has proven to be superior to the para worker. Testing was based on an execution of Genetic Algorithm to solve the Economic Dispatch problem in Power Engineering. The proposal is particularly useful for the implementation and execution of computational intelligence techniques such as evolutionary computing for engineering applications.

Publication Type: Journal Article
Murdoch Affiliation: School of Information Technology
Publisher: Taylor & Francis
Copyright: (c) 2004 Taylor & Francis
URI: http://researchrepository.murdoch.edu.au/id/eprint/563
Item Control Page