Fluid-dynamic Shape Optimization Using Sensitivity Analysis of Navier-Stokes Solutions

Abstract

The shape optimization using CFD (Computational Fluid Dynamics) is expected to be a powerful design tool, because CFD can simulate inviscid as well as viscous flow fields around a body of complex geometry. In the present paper, the gradient based optimization methods are applied to the shape optimization of a two-dimensional body in which the fluid dynamic objective function is minimized. Flow fields are computed by a finite-volume Navier-Stokes solver. Sensitivity analysis which gives the gradients of the objective function and the constraints is carried out by the adjoint variable method. The numerical procedure is described together with the results of several applications.