A Multi-Objective Particle Swarm Optimization Incorporating Design Sensitivities
Masahiko Nakamura, Kazuhiro Izui, Shinji Nishiwaki, Masataka Yoshimura · 11th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference · 2006
Particle Swarm Optimization (PSO), which is one of a meta-heuristic optimization algorithm, has been successfully applied to single- and multi-objective optimization problems to obtain global or Pareto optimal solutions. However, PSO still has a difficulty in application to complicated structural and mechanical optimization problems since local optimization capability is inferior to general numerical optimization methods due to constraint handling problem. This paper discusses a new multiobjective optimization method that incorporates design sensitivities concerning objective and constraint functions to apply to structural and mechanical design optimization problems. Multi-objective PSO method are combined with a sequential linear programming (SLP) method. Numerical and structural optimization problems are solved by the proposed hybrid method to verify the optimization efficiency.