Numerical investigation of POD techniques in a turbulent channel flow

This study focuses on the identification of coherent structures in a turbulent channel flow using the proper orthogonal decomposition (POD) technique. Direct numerical simulation (DNS) data of turbulent channel flow at Re ۱۸۰   is employed for the analysis. The velocity fields obtained from the DNS serve as inputs for the POD algorithm, enabling the extraction of dominant coherent structures and modes inherent in turbulent flow. The POD modes, which are based on the specified Reynolds number, are utilized to examine the primary coherent structures of the flow. This analysis provides information on the size, shape, and strength of these structures. Understanding the effect of individual POD modes on the energy of the flow and how it spreads across mode numbers. By investigating how flow structures interact with each other and shape the overall flow, also we can gain important information about each mode's contribution to the whole field. Furthermore, this algorithm allows us to create a reduced-order model for flow control or investigation.

Keywords:

Turbulent channel flow , POD , DNS , Coherent structures