SENSUS - Simulation-based design optimization of dynamic systems under uncertainties
SENSUS: "Simulationsbasierte Entwurfsoptimierung dynamischer Systeme unter Unsicherheiten"
The scientific objective of the joint research project is the development of innovative, digital methods for simulation-based optimization of complex parts and components. The focus lies on the enhancement of efficient mathematical optimization methods with the objective to take into account dynamic operating conditions and uncertain manufacturing processes. The project incorporates groups from mathematical and engineering sciences. It contributes to linking simulation-based engineering with applied mathematics through the theoretical description of practically relevant problems. Technical applications focus on part with mechanical and fluid dynamical functionalities from the industrial environment of Hamburg, which operate under dynamic conditions and uncertain parameters.
In the context of this project, our group targets to deal with problems arising in biomedical applications. Very often artificial materials and biomedical devices are used to restore and control a regular blood flow. Despite, however, all the advances in medicine, engineering and material science, the minimization of induced blood damage still remains a critical task so that good bio-compatibility is ensured. In this context, the derivation and implementation of CFD-based adjoint complements to blood damage equations as well as blood-related objective functionals are to be realized. The efficiently computed sensitivities are used by a gradient-based method to ultimately optimize the relevant shapes. The study is then extended in an FSI framework, so that wall elasticity is considered. Finally, the complete optimization framework is studied in the context of robust design. In this sense, the formulated optimization strategy is extended to minimize potential uncertainties of the objective functional subjected to inflow or/and material properties.
Displacement field for the shape optimization of an idealized 3D bypass-graft, as produced by a newly developed non-Newtonian adjoint method (left) and a classic Newtonian (right).
Adjoint-based shape optimization of a 3D idealized biomedical device, for the minimization of blood damage (hemolysis).
Metadata
Project Acronym
SENSUS
Project Title
Simulationsbasierte Entwurfsoptimierung dynamischer Systeme unter Unsicherheiten