Augmented Multistep Finite-Control-Set Model Predictive Control for Induction Motor-Drive System

This article develops an observer-augmented multistep model predictive control strategy with finite-control-set principle to improve the robustness of the control loop against disturbances, including external disturbances, parameter mismatches, and model uncertainties. The influence of the parameter mismatches on the multistep finite-control-set model predictive control is first discussed via simulations and quantified by analyzing the probability of suboptimality. Furthermore, in order to compensate for these effects, the disturbances are included in the system model of the control problem as an extended state and estimated with a disturbance observer. The estimated disturbances as well as the system states are then delivered to the optimization problem of the current control and incorporated for the computation of the solution. The proposed method is then implemented on a dSPACE system and tested under several scenarios. The effectiveness of the proposal is validated with experimental results.