Model-free Predictive Torque Control of an Induction Machine Based on Parameter Estimation

The uncertainty or variation of the electric machine parameters in predictive torque control (PTC) has a noticeable impact on the controller's performance. This paper proposes a model-free PTC strategy based on the estimation of the prediction model parameters using input and output data of the controlled system, applied to an induction machine. This approach has the advantage of not requiring a detailed previous knowledge of the system, with a high robustness to mismatch in the inductance parameters of the machine. The stator resistance is identified as a critical parameter for PTC, therefore an adaptation mechanism based on support vector regression is proposed to increase the robustness of the system. Simulation tests are carried out to validate the effectiveness of the proposed strategy.