IMPROVED PARTICLE SWARM OPTIMIZATION METHOD DIRECTED BY INDIRECT SURROGATE MODELING

Abstract

An improved particle swarm optimization algorithm is proposed and tested for two different test cases: surface fitting of a wing shape and an inverse design of an airfoil in subsonic flow. The new algorithm emphasizes the use of an indirect design prediction based on a local surrogate modeling as a part of update equations in particle swarm optimization algorithm structure. For all the demonstration problems considered herein, remarkable reductions in the computational times have been accomplished.

Keywords

• PSO
• Surrogate Modeling
• Inverse Design