Adjoint based design optimization of subsonic airfoils with a panel code

Download

Berk_Sarıkaya_Thesis.pdf

Date

2022-7-25

Author

Sarıkaya, Berk

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

306
views

215
downloads

An in-house panel code written in Fortran is automatically differentiated and developed into an adjoint-based aerodynamic shape optimization tool for airfoil profiles. The automatic differentiation tool, FDOT, is employed in reverse mode to obtain the discrete-adjoint solver. The adjoint-based sensitivity derivatives (i.e. gradient vector) are validated against the finite difference approach. The computed sensitivity derivatives are then employed for the gradient based aerodynamic shape optimization of subsonic airfoil profiles for given target lift and moment coefficients. In addition, an adverse pressure gradient minimization term is added to the objective function for milder stall characteristics. A multi-point design optimization method is also implemented for an improved off-design performance. Airfoils are parametrized by the Class Shape Transformation. Single and multi-point optimizations are driven by the open-source optimizer, DAKOTA, using a quasi-Newton method. Case studies for cambered airfoils at low angles of attack are presented for single and multi-point design optimizations. The surface pressure distributions and the aerodynamic loads for the optimum airfoil profiles are further verified with the open-source RANS solver, SU2. It is shown that the adjoint based design optimization methodology developed is efficient and robust.

Subject Keywords

panel method, discrete adjoint, aerodynamic optimization, multi-point optimization, adverse pressure gradient

URI

https://hdl.handle.net/11511/98565

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

Aerodynamic design optimization of three dimensional rocket nozzles using adjoint method Eyi, Sinan (null; 2013-09-13) A design optimization method based on three dimensional Euler equations is developed. A finite volume method is implemented to discretize the Euler equations. Newton's method is used to solve the discretized form of Euler equations. Newton's method requires the calculation of the Jacobian matrix which is the derivative of the residual vector with respect to the flux vector. Different upwind methods are used in the calculation of flux vectors. Numerical and analytical methods are utilized in the evaluation o...
AIRFOIL DESIGN OPTIMIZATION USING THE NAVIER-STOKES EQUATIONS Eyi, Sinan; LEE, KD (Springer Science and Business Media LLC, 1994-12-01) A design optimization technique is presented which couples a computationally efficient Navier-Stokes code with a numerical optimization algorithm. The design method improves the aerodynamic performance of an airfoil subject to specified design objectives and constraints. Recent advances in computers and computational fluid dynamics have permitted the use of the Navier-Stokes equations in the design procedure to include the nonlinear, rotational, viscous physics of transonic flows. Using numerical optimizati...
Automatic image annotation by ensemble of visual descriptors Akbaş, Emre (2007-06-22) Automatic image annotation systems available in the literature concatenate color, texture and/or shape features in a single feature vector to learn a set of high level semantic categories using a single learning machine. This approach is quite naive to map the visual features to high level semantic information concerning the categories. Concatenation of many features with different visual properties and wide dynamical ranges may result in curse of dimensionality and redundancy problems. Additionally, it usu...
Interactive evolutionary multi-objective optimization for quasi-concave preference functions Fowler, John W.; Gel, Esma S.; Köksalan, Mustafa Murat; Korhonen, Pekka; Marquis, Jon L.; Wallenius, Jyrki (2010-10-16) We present a new hybrid approach to interactive evolutionary multi-objective optimization that uses a partial preference order to act as the fitness function in a customized genetic algorithm. We periodically send solutions to the decision maker (DM) for her evaluation and use the resulting preference information to form preference cones consisting of inferior solutions. The cones allow its to implicitly rank solutions that the DM has not considered. This technique avoids assuming an exact form for the pref...
Accurate equivalent models of sandwich laminates with honeycomb core and composite face sheets via optimization involving modal behavior Cinar, Okan; Erdal Erdoğmuş, Merve; Kayran, Altan (2017-03-01) An approach is introduced for determining accurate two-dimensional equivalent laminated models of sandwich laminates with honeycomb core and composite facesheets by optimization involving modal behavior. The approach relies on minimizing the objective function which is defined as the sum of the square of the differences between the natural frequencies of the honeycomb sandwich laminate estimated by the finite element analysis of the 3D detailed model with the actual honeycomb core geometry and by the 2D equ...

Citation Formats

B. Sarıkaya, “Adjoint based design optimization of subsonic airfoils with a panel code,” M.S. - Master of Science, Middle East Technical University, 2022.