User-preference Particle Swarm Algorithm for Airfoil Design Architecture

Airfoil optimization frameworks rely on extensive computational resources to provide the most feasible design for the required flight condition. Extending the framework to encompass multiple design criteria adds further complexity due to the conflicting nature of the design objectives. In this paper, we propose the integration of designerdriven preferences into a multi-objective particle swarm heuristic to efficiently navigate the design space. This results in a highly accurate and proficient rate of convergence, since only preferred regions of the Pareto landscape are explored. The algorithm is introduced and applied to a typical transonic airfoil design scenario where the objective value is provided by a Reynolds-averaged Navier-Stokes solver. Experimental case-studies are presented, which highlight the computational efficiency of the user-preference algorithm.