Identification of Effective Control Parameters for a Supersonic Dual-Stream Jet Flow using DMD-ROM*

A supersonic dual-stream jet flow comprised of Mach 1.6 core and Mach 1.0 bypass streams is investigated. The mixing of the two flows downstream of a splitter plate results in a thick shear layer which produces instabilities that propagate and dominate the flow field. The shedding also manifests as a signature high-frequency tone in the far-field acoustics. A shock train originates from within the nozzle near the splitter plate trailing edge (SPTE), and a shock-induced flow separation is observed. To control the flow, micro-jet actuators are placed near the SPTE region with the goal of suppressing the resonant tone and influencing the development of the prominent shock wave. Active control of complex turbulent flows can be difficult to design when the parameter space is large, therefore, snapshots of the uncontrolled flow are used in a data-driven dynamic mode decomposition reduced-order model (DMD-ROM) approach to identify effective control configurations. The flow and forcing snapshots are projected onto a reduced subspace and the dynamics are advanced in time. This facilitates fast investigation of various tunable actuator-specific parameters. The optimal parameters identified by the DMD-ROM approach are implemented in simulation to achieve the control objectives.