Experimental Study of Adaptive Control of High-Speed Flow-Induced Cavity Oscillations

Abstract

This paper describes progress towards the development of an active suppression system for open cavity oscillations. A leading-edge array of piezoelectric zero-net mass-flux (ZNMF) actuators is used to suppress cavity flow oscillations determined from measured pressure fluctuations near the rear wall of cavities with length-to-depth ratios (L/D) of 5 and 6 at Mach 0.3 and 0.4. The waveform parameters of the actuator array excitation signal are systematically investigated using open-loop control. Closed-loop control using the quasi-static downhill simplex and the dynamic ARMARKOV adaptive disturbance rejection and generalized predictive control (GPC) algorithms are applied and compared to open-loop control. Up to 30% broadband reduction in rms pressure fluctuations over a 4 kHz bandwidth and reduction in multiple Rossiter modes are obtained and compare favorably with optimized open-loop sinusoidal control. Through the investigation of different actuator slot geometries, larger slots which generate larger momentum with a forcing frequency near the resonant frequency of the actuator showed the best performance for suppressing the cavity oscillation.

Keywords

• cavity oscillations
• adaptive control
• zero-net mass-flux actuator
• armarkov
• gpc
• downhill simplex