Particle Swarm Optimization Using Projection Matrix for Behavior Constraints
Tomoyuki Miyashita · 11th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference · 2006
Particle Swam Optimization is one of useful optimization method that has a global search ability and is easy to implement. However there are some difficulties are observed for behavior constraints. This paper proposes a new approach to treat behavior constrains by projection of velocity vector of particles. By projection of velocity vector particles could approach feasible region formed by active constraints. Through several numerical examples, it is recognized better converge performance than original PSO. Nomenclature x = amplitude of oscillation f = cylinder diameter g = pressure coefficient σ = force coefficient in the x direction r = force coefficient in the y direction w = chord