Commandes adaptative et prédictive de la machine asynchrone

Abstract

The induction motor given a new technology perspective in many industries, due to their simple mechanical structure, easy maintenance and low cost. However its dynamic structure is very complex (system non-linear multi variable strongly coupled) makes his control is complicated requiring complex control algorithms. With the progress in power electronics associated with the appearance of fast components switches, caused significant changes in the design of systems control/regulators. This development has encouraged the emergence of various commands: Vector Control, non-linear and predictive control... etc The vector control has been in recent years towards the most important research and best suited to industrial requirements. However However, this structure requires the establishment of a sensor in rotor and is very sensitive in to parameters machine variations that induce loss of decoupling. The aim of this thesis in the first place is to present a multitude of alternative techniques to vector control, namely (sensorless control and adaptive control). Then the development of a high performance non-linear, predictive and adaptive predictive control applied to the induction motor, with the objective of this work is directed towards improving trajectory tracking capability, stability guarantee, robustness to parameters variations and disturbance rejection.Some numerical simulations tests are realised and the obtained results demonstrate the efficiency and the dynamic performances of the proposed strategy