چكيده به لاتين
Nowadays, modern commercial aircrafts employ turbofan engines. Turbofan engine control is required to satisfy various operational and structural constraints such as shaft overspeed and acceleration, turbine inlet temperature, compressor surge margin and etc. In other words, the turbofan engine controller has the task of providing the requested pilot's thrust while satisfying all kinds of physical and operational constrains. The main control parameter for turbofan engines is the fuel injected into the combustion chamber. Also, to prevent destructive phenomena in compressors such as surge and stall, these engines are equipped with a bleed system. Thus, in order to control the turbofan engine, in addition to the fuel flow variable, other variables such as bleed are used.
In this thesis, multivariable model predictive control design for turbofan engine is presented. For this purpose, the process of extracting the governing equations on the controller is presented comprehensively and various constraints of inputs and outputs are specified. A multivariable controller (here two variables: the fuel flow and bleed air) is designed based on the nonlinear thermodynamic model. The performance of the controller is then studied, and the effect of bleed on the performance of the turbofan engine is evaluated. In addition, due to the mismatching between the linearized and nonlinear models, feedback correction method has been used to improve the performance of the MPC controller in tracking of optimal output and maintaining constraints. Finally, the effect of key parameters of the MPC controller such as prediction horizon, control horizon and control penalty factor has been investigated. The simulation results reveal that the multivariable control approach is effective as the lack of bleed causes engine surge problem. Moreover this study shows that the speed response of the MPC controller outperforms the traditional Min-Max controller.
