Speaker: Prof. Jörn Sesterhenn (University of Bayreuth, Germany)
Title: Adjoint-based data assimilation for compressible and reactive turbulent flows
Date: Friday October 11th, 2019
Time: 10:40am
Place: EE-01
Abstract: The analysis of complex fluid mechanical phenomena is usually based on experimental and numerical analysis. Both approaches provide useful information. However, experimental data are usually incomplete, because not every state variable is accessible by measurements. On the other hand, numerical solutions are often affected by initial and boundary conditions, which do not fully match to the real flow conditions. To obtain a complete set of information of a real flow state, both
approaches can be combined by means of data assimilation.
Most commonly, a variational approach is used. Parameters of a numerical model are adapted until the numerical solution match the experimental data. The required adaptation (gradient) is determined by adjoint equations. The assimilated numerical state contains the unknown/unmeasured quantities under the constraint of the applied model. Furthermore, the quality of experimental data is enhanced by means of a model-based filtering, which can reduce measurement noise.
The application of an adjoint-based data assimilation for compressible flows will be presented by means of three different applications. The first application considers the identification of sound sources. The second application deals with the determination of instantaneous pressure distributions based on PIV data. The third application is the analysis of reactive flow configurations. The employed numerical framework and its requirements on the computational infrastructure are discussed in detail.
Bio: Jörn Sesterhenn received his Ph.D. 1997 in Computational Fluid Dynamics at the ETH Zürich. He primarily worked on the design of algorithms for compressible low Mach number flows. He spent two years at the University of Washington in Seattle and joined the TU Munich as Postdoc in 1999, where he received Habilitation in Fluid-Mechanics in 2004. Between 2006-2009 he was Professor for Numerical Methods in Aerospace Sciences at the University of the Armed Forces in Munich and became full professor for Numerical Fluid mechanics at the TU Berlin in 2009. In October 2019 he will change to the University of Bayreuth where he will hold the Chair for Mechanics and Fluid-Mechanics. His main research interest shifted form purely numerical methods to Data-Assimilation, which incorporates experimental techniques for technical and geophysical flows into numerical methods.