Volume 52, Number 4, 2018
|Page(s)||1501 - 1532|
|Published online||28 September 2018|
An adjoint-based method for the numerical approximation of shape optimization problems in presence of fluid-structure interaction
CMCS-MATHICSE-SB, Ecole Polytechnique Fédérale de Lausanne, Station 8, 1015 Lausanne, Switzerland
2 Dipartimento di Matematica “F. Brioschi”, Politecnico di Milano, via Bonardi 9, 20133 Milano, Italy
* Corresponding author: email@example.com
Accepted: 24 February 2017
In this work, we propose both a theoretical framework and a numerical method to tackle shape optimization problems related with fluid dynamics applications in presence of fluid-structure interactions. We present a general framework relying on the solution to a suitable adjoint problem and the characterization of the shape gradient of the cost functional to be minimized. We show how to derive a system of (first-order) optimality conditions combining several tools from shape analysis and how to exploit them in order to set a numerical iterative procedure to approximate the optimal solution. We also show how to deal efficiently with shape deformations (resulting from both the fluid-structure interaction and the optimization process). As benchmark case, we consider an unsteady Stokes flow in an elastic channel with compliant walls, whose motion under the effect of the flow is described through a linear Koiter shell model. Potential applications are related e.g. to design of cardiovascular prostheses in physiological flows or design of components in aerodynamics.
Mathematics Subject Classification: 49Q10 / 49J20 / 65K10 / 65N30 / 74F10 / 76D55
Key words: PDE-constrained optimization / shape optimization / fluid-structure interaction / adjoint problem
© EDP Sciences, SMAI 2018
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.