Evaluation of differences between DMD and OPD in detection of coherent structures

0565932 - ÚT 2023 CZ eng A - Abstract
Kluková, Zuzana - Nosek, Štěpán - Jaňour, Zbyněk
Evaluation of differences between DMD and OPD in detection of coherent structures.
International Workshop on Physical Modeling of Flow and Dispersion Phenomena - Book of Abstracts. Prague: Institute of Thermomechanics of the Czech Academy of Sciences, 2022 - (Jurčáková, K.; Kellnerová, R.). s. 110-110. ISBN 978-80-87012-81-9.
[PHYSMOD 2022 - The International Workshop on Physical Modelling of Flow and Dispersion Phenomena. 29.08.2022-31.08.2022, Praha]
Institutional support: RVO:61388998
Keywords : coherent structures * dynamic mode decomposition * oscillating pattern decomposition * sensitivity analysis
OECD category: Meteorology and atmospheric sciences

It is well known that flow within the Atmospheric boundary layer is not absolutely chaotic, but it contains some structures (so called coherent structures). Using mathematical methods, it is possible
to reveal these structures in time-resolved data obtained for instance from PIV measurements or numerical modelling. Next to the widely used proper orthogonal decomposition (POD) arising from
the singular value decomposition, there are methods producing complex eigenvalues and eigenvectors containing information about the cyclic propagation or pulsation of the coherent structures, frequency of their excitation, and decay time of the amplitude oscillation. As an example, we can mention dynamic mode decomposition (DMD) and oscillating pattern decomposition (OPD). Both the methods (DMD and OPD) are based on computing the POD modes, but following algorithm is different. Within this contribution, I will evaluate the differences between these two methods on the PIV data and I will show how the paremeters choice within the algorithm influence the results.
Permanent Link: https://hdl.handle.net/11104/0337691