Development of empirical estimators for feedback control of high-speed axisymmetric jets

Citation metadata

Date: Sept. 2011
From: AIAA Journal(Vol. 49, Issue 9)
Publisher: American Institute of Aeronautics and Astronautics Inc.
Document Type: Author abstract; Report
Length: 231 words

Document controls

Main content

Abstract :

Localized arc filament plasma actuators have demonstrated significant potential in controlling high-speed and high-Reynolds-number axisymmetric jets in an open loop. As a first step in incorporating feedback for this control system, an empirical reduced-order model of the essential flow dynamics in the region surrounding the potential core of the unforced jet has been developed. An existing direct numerical simulation database, with a configuration similar to previous experiments, formed the testbed for this modeling phase. Real-time flow state estimation is a challenging problem in the implementation of feedback control for such complex flows of practical interest. Sensing the pressure in the irrotational near field close to the nozzle exit offers a suitable nonintrusive measurement that is driven by the jet's shear layer dynamics. Owing to convection, such a configuration naturally results in the measured pressure having a time lead compared with the state of the reduced-order model, which is very useful for feedback control. The sensing configuration used consists of an azimuthal ring array along with a linear array. Several estimation strategies are implemented and assessed using the numerical database. The time-invariant version of the linear Kalman filter is shown to have similar or better accuracy compared with a quadratic stochastic estimator, which in turn significantly outperforms a linear stochastic estimator. The filter is only as computationally complex as the linear stochastic estimator, thereby making it the strategy of choice. DOI: 10.2514/1.J050966

Source Citation

Source Citation   

Gale Document Number: GALE|A266940668