Continuous adjoint approach to shape optimization with respect to 2D incompressible fluid flow

Brandner, Marek, Egermaier, Jiří, and Kopincová, Hana. "Continuous adjoint approach to shape optimization with respect to 2D incompressible fluid flow." Programs and Algorithms of Numerical Mathematics. 2025. 17-28. <http://eudml.org/doc/299970>.

@inProceedings{Brandner2025,
abstract = {The aim of this article is to briefly introduce the procedure for optimizing water turbine blades, which can lead to an innovative blade design and, consequently, an improvement in the desired properties of the water turbine, such as efficiency or the preferred pressure distribution on the blade. The computational method is based on formulating an objective function under certain constraint conditions, which are governed by the Navier-Stokes equations. This formulation enables the use of the Lagrange multiplier method, which incorporates the constraints into the augmented objective function. We derive the so-called adjoint problem, allowing us to simplify the gradient formulation for the chosen gradient-based optimization method.},
author = {Brandner, Marek, Egermaier, Jiří, Kopincová, Hana},
booktitle = {Programs and Algorithms of Numerical Mathematics},
keywords = {shape optimization; continuous adjoint},
pages = {17-28},
title = {Continuous adjoint approach to shape optimization with respect to 2D incompressible fluid flow},
url = {http://eudml.org/doc/299970},
year = {2025},
}

TY - CLSWK
AU - Brandner, Marek
AU - Egermaier, Jiří
AU - Kopincová, Hana
TI - Continuous adjoint approach to shape optimization with respect to 2D incompressible fluid flow
T2 - Programs and Algorithms of Numerical Mathematics
PY - 2025
SP - 17
EP - 28
AB - The aim of this article is to briefly introduce the procedure for optimizing water turbine blades, which can lead to an innovative blade design and, consequently, an improvement in the desired properties of the water turbine, such as efficiency or the preferred pressure distribution on the blade. The computational method is based on formulating an objective function under certain constraint conditions, which are governed by the Navier-Stokes equations. This formulation enables the use of the Lagrange multiplier method, which incorporates the constraints into the augmented objective function. We derive the so-called adjoint problem, allowing us to simplify the gradient formulation for the chosen gradient-based optimization method.
KW - shape optimization; continuous adjoint
UR - http://eudml.org/doc/299970
ER -