Implementation of digital pheromones for use in particle swarm optimization

Foo, Jung
Kalivarapu, Vijay
Kalivarapu, Vijay
Winer, Eliot
Journal Title
Journal ISSN
Volume Title
Source URI
Research Projects
Organizational Units
Mechanical Engineering
Organizational Unit
Journal Issue

This paper presents a new approach to particle swarm optimization (PSO) using digital pheremones to coordinate the movements of the swarm within an n-dimensional design space. In traditional PSO, an initial randomly generated population swarm propagates towards the global optimum over a series of iterations. Each particle in the swarm explores the design space based on the information provided by previous best particles. This information is used to generate a velocity vector indicating a search direction towards a promising design point, and to update the particle positions. This paper presents how digital pheromones can be incorporated into the velocity vector update equation. Digital pheromones are models simulating the real pheromones produced 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 PSO. Particle swarms search the design space with digital pheromones aiding communication within the swarm to improve search efficiency. Through additional information from the pheromones, particles within the swarm exploring the design space and locate the solution more efficiently and accurately than traditional PSO. In this paper, the development of this method is described in detail along with the results from several optimization test problems.

<p>This is a conference proceeding from <em>Collection of Technical Papers - AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference</em>, (2006): AIAA 2006-1917, doi: <a href="" target="_blank">10.2514/6.2009-2192</a>. Posted with permission.</p>
Virtual Reality Applications Center, information analysis, interative methods, structural design, particle positions, particle swarm optimization