Robust tracking control of aero-engine rotor speed based on switched LPV model

Thrust control in aero-engines is achieved indirectly due to the lack of thrust sensing technologies. A related variable must be chosen for the control, typically high or low pressure rotor speeds. In this paper, an H∞ controller is designed for the rotor speed tracking in aero-engines. First, a small deviation linear model at steady state is obtained from the experimental data. Then, sets of small-deviation linear models are used to construct a linear parameter-varying (LPV) model with gain-scheduled parameters that capture the nonlinearity of the aero-engine dynamics. The LPV model is then converted to a switched convex polytopic form with hysteresis switching logic. Next, a theoretical sufficiency criterion is provided to guarantee H∞ performance based on linear matrix inequalities (LMIs). Relevant theoretical results are applied to prove stability when switching between subsystems. Simulation results are given to show the validity of the proposed design method, where the proposed the strategy with hysteresis switching logic can reduce the computational cost and avoid false switching due to disturbances.