Cascaded predictive controller design for speed control and load torque rejection of induction motor

Cascaded nonlinear predictive controller for induction motor drive is presented. The load torque, considered as an unknown disturbance, is rejected using a disturbance observer. First, a nonlinear multivariable predictive controller is applied to track electromagnetic torque and rotor flux norm trajectories. Then, a speed predictive control strategy is carried out from the electromechanical equation of the machine. Both controllers are applied in a cascade structure to induction motor. The prediction model is carried out via Taylor series expansion using Lie derivatives for nonlinear model. The derived predictive law minimizes a quadratic performance index of the predicted tracking error for multivariable system and a predicted error for speed control. The implementation of the cascaded control law does not need an on line optimization and the tracking of desired trajectories is achieved successfully. The load torque observer, derived from the speed predictive control, is simplified to a PI structure. This observer structure guarantees the disturbance rejection and the robustness to parameters variations. Simulation results show the high performance of the proposed control scheme.