Hide/Show Apps

Development of a discrete adjoint-based aerodynamic shape optimization tool for natural laminar flows

2020
Kaya, Halil
An adjoint-based aerodynamic shape optimization framework for natural laminar flows is developed. A Reynolds-Averaged Navier-Stokes flow solver with the Spalart-Allmaras turbulence model is coupled with the recently developed Bas-Cakmakcioglu transition model in order to predict laminar to turbulent transition onset. In the gradient-based optimization process, the sensitivity derivatives required by the optimization algorithm is obtained by the discrete adjoint method, which is developed for the in-house flow solver and implemented for natural laminar flow airfoils and wings. In the development of the discrete adjoint method, an automatic differentiation tool is employed to take the discrete derivative of the modules in the in-house flow solver heavily modified. The parametrization of the aerodynamic surface is realized by the Free-Form Deformation technique. The sensitivity derivatives with respect to design parameters, which are computed by the adjoint method, are validated with the finite-difference method. The success of the adjoint-based aerodynamic shape optimization methodology developed in this study is then demonstrated by optimizing aerodynamic characteristics of several airfoils and wings for compressible turbulent and natural laminar flows.