Reduced order modeling for the numerical optimization of fluid-structure interaction problems

Abstract eng:
An approach for the optimization of Fluid-Structure Interaction (FSI) problems employing reduced order models is presented. The basic components of the method are: an implicit partitioned solution approach with a finite-volume flow solver and a finite-element structural solver for the FSI, a sequential quadratic programming approach for the optimization as well as a direct coupling via the Message-Passing-Interface (MPI) for the integration of the optimization algorithm and the fluid and structural solvers. Reduced order models based on Proper Orthogonal Decomposition (POD) with snapshots and Kriging are used within the optimization process for the approximation of the coupled FSI solution. Numerical investigations concern the efficiency of the MPI coupling, the verification of the optimization procedure, the quality of the approximations achieved with the POD models and the influence of relevant model parameters. By comparing the proposed POD optimization with a full order FSI optimization it is demonstrated that the usage of the POD model can yield considerable savings in computing time without significant losses in the quality of the optimization results.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket