Dynamic mode analysis of supersonic jet flows
Title: Dynamic mode analysis of supersonic jet flows
DNr: SNIC 2016/1-272
Project Type: SNIC Medium Compute
Principal Investigator: Markus Olander Burak <mabu@chalmers.se>
Affiliation: Chalmers tekniska högskola
Duration: 2016-06-01 – 2017-06-01
Classification: 20306 20302 20301
Keywords:

Abstract

As computations and experiments become more advanced, they generate an ever-increasing amount of data. In order to find more than the obvious trends requires a data-driven method able to characterise the data in meaningful ways with minimal user guidance. In the last ten years Dynamic Mode Decomposition (DMD) has been growing in popularity for identifying underlying time-dependent patterns in large datasets. It has been used for dynamic pattern identification for a wide range of applications such as spreading of infectious diseases, financial trading systems, global power systems and fluid mechanical systems. In the present project the DMD method will be applied for characterising the dynamic motions of supersonic jet flows. The identified dynamic motions, supported by acoustic measurements, will support the understanding noise generating mechanisms in a supersonic jet. When the emanating flow from the nozzle reach a supersonic condition, shocks are created as the flow adapts to the ambient condition. The motion of the shock-cell structure and its interaction with convective flow structures generate highly effective shock-associated noise components in addition to the noise sources found in subsonic jets. In order to mitigate the noise from supersonic jets, the underlying dynamic of the flow needs to be understood. The DMD method has proven to be a useful tool for characterising the underlying dynamics based on experimental Shadowgraph data of a jet. In this project the flow data will be generated using numerical prediction (Large-Eddy Simulation). This will allow a three-dimensional representation of the dynamic motions.