State space modeling and robust reduced-order control of combustion instabilities (Q5926854)

The authors discuss combustion processes occurring in steam and gas turbines for power generation and in jet and ramjet propulsion engines. In these processes, acoustic pressure and velocity oscillations can adversely affect operating conditions and system performance, produce unacceptable acoustic noise levels or even cause mechanical failure of system components. Traditionally, these oscillations have been mitigated by passive control, but this fails to take into account fluctuating operating conditions. NEWLINENEWLINENEWLINERecent advances in sensor and actuator construction and design have made feasible the application of active feedback control, although such problems as inaccuracy of the models and the need to approximate the (infinite dimensional) system by high order discretizations lead to increasing controller complexity and difficulties in the implementation of controllers designed by methods such as linear-quadratic Gaussian control and \(H_\infty\) control. NEWLINENEWLINENEWLINEThe authors develop a state space model and apply robust reduced-order feedback control theory under parametric uncertainty to the problem of suppressing the effects of combustion instabilities. Control actuation is achieved by modulating the pressure in the combustor using acoustic drivers; the sensors measure pressure fluctuations. NEWLINENEWLINENEWLINEComputer simulations show robust stability and robust performance.