A Study on Nonlinear Model Predictive Control for Helicopter/Engine with Variable Rotor Speed Based on Linear Kalman Filter

A novel control method combining nonlinear model predictive control (NMPC) with linear kalman filter (LKF) and adaptive notch filter (ANF) for an integrated helicopter/engine system with variable rotor speed is proposed to enhance the response ability of turboshaft engine. Firstly, based on the integrated helicopter/engine model with variable rotor speed, an ANF combined with frequency estimation is introduced. Then, in order to estimate the significant and unmeasurable performance parameters utilized in model predictive control, such as temperature before turbine, a nonlinear model predictive controller based on LKF is presented. The simulation verifications demonstrate that the fundamental frequency of torsional vibration changes from 1.30 & x2006;Hz to 2.70 & x2006;Hz when the rotor speed varies continuously by 50 & x2006;%. In this case, compared with the notch filter, all torsional vibration amplitudes are damped below 0.1 & x2006;% by ANF. Meanwhile, in comparison with the PID controller, the NMPC can reduce the overshoot and droop of the power turbine speed to less than 1 & x2006;% with steady-state error no more than 0.5 & x2006;% at the off-design point of NMPC based on adaptive torsional vibration suppression. The applications of LKF and similar transformation improve the control accuracy and robustness performance of the NMPC designed at a single operating point.