A Parallel Implementation of Particle Swarm Optimization Using Digital Pheromones
Vijay Kalivarapu, Jung Leng Foo, Eliot H. Winer · 11th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference · 2006
A parallel implementation of Particle Swarm Optimization (PSO) using digital pheromones to coordinate the movements of the swarm within an n-dimensional design space is presented in this paper. Digital pheromones are models simulating real pheromones emitted by insects for communication to indicate a source of food or a nesting location. This principle of communication and organization between each insect in a swarm offers substantial improvement when integrated into a Particle Swarm Optimization algorithm. Digital swarms are used to search a design space with digital pheromones aiding communication within the swarm to improve search efficiency. With statistical analysis, the pheromone strength in a region of the design space is determined. The swarm then reacts accordingly based on the probability that this region may contain an optimum. When implemented in a parallel computing architecture, significant performance increases were observed. This paper presents the method development and results from several test cases.