Multi-variable Model Predictive Control Design for a Turbofan Engine and Performance Comparison with Min-Max Controller

The turbofan engine controller is responsible for providing the thrust requested by the pilot, while maintaining structural and operational constraints. Therefore, the strategies used to control the engine of an aircraft should be able to consider the constraints of the system in their structure. In this research, a multivariable model predictive controller based on a linear state space model for a turbofan engine is designed. This controller has the ability to accommodate various input and output constraints during supplying of required thrust. Due to the lack of matching between the linearized model for the controller and the nonlinear engine model, the feedback correction method is used in the control structure to improve the performance of the MPC controller. In this method, in addition to fuel as the main control input, bleed is also considered as an auxiliary control input for closed loop to reduce the possibility of compressor surge. In simulation, using a non-linear thermodynamic model, the controller performance is compared with the Min-Max method, our which is typically used to control aircraft engines. Results of this simulation show the effective performance of the proposed controller.