Robust control for high performance induction motor drives based on partial state-feedback linearization

This paper deals with a robust input-output feedback linearization control technique for induction motors. The idea arises from the observation that classic feedback linearization suffers primarily from two disadvantages: the accuracy of the dynamic model and the corresponding correct knowledge of the model parameters. To cope with the limit of the feedback linearization control, intrinsically connected to its model-based nature, the linear controller has been substituted with a suitably designed robust controller. Such robust controllers has been designed with the aim of copying any variation of the main parameters of the induction motor, particularly those depending on the heating/cooling of the machine like stator and rotor resistances and magnetic saturation of the iron core. The proposed controller has been tested experimentally on a suitably designed test set-up.