A new extension of particle swarm optimization and its application in electronic heat sink design

Abstract

Particle Swarm optimization (PSO) is a robust stochastic evolutionary computation technique which is based on the movement and intelligence of swarms. In this paper the PSO algorithm is modified to improve its performance in a class of design applications in heat transfer. The developed approach includes a new term called a chaotic acceleration factor (Ca) into the algorithm, which enhances its convergence rate and its accuracy. The modified PSO is empirically tested with well-known benchmark functions. Next it is applied in plate-fin design with the objective of dissipating the maximum heat generation from an electronic component by minimizing the entropy generation rate to obtain the highest heat transfer efficiency.