Output details

Feedback control of flow resonances using balanced reduced-order models

This paper derives balanced reduced order models of resonant fluid systems and uses these to design stabilising feedback controllers. Specifically, the Eigensystem Realisation Algorithm is used, which captures input-output dynamics accurately ("balanced modelling") without requiring adjoint simulations, making it applicable experimentally. This paper is the first to use the ERA for model reduction and feedback control in fluids, and includes application to experimental data (verification contact: Prof. Cattafesta, cattafesta@eng.fsu.edu). It was the fruit of collaboration with Prof. Clancy Rowley's Princeton University Group (contact: cwrowley@Princeton.EDU), during which strategies from the combustion instability and flow modelling communities were merged (see also http://dx.doi.org/10.1017/jfm.2012.330).