Commande de machine électrique en environnement Matlab/Simulink et temps réel .

Abstract

The electric machines had a great interest by the researchers, because their dvantages are seen in their aptitudes for adapting to any environments, and their efficiency, thus exceeding other nonelectric actuators. Considering, simplicity, reduced cost, and absence of maintenance, the asynchronous motor is undoubtedly the most used industrially, its range of use extend from micro motors until very great ones. The recent technological advances of power electronics and digital signal processing have opened to the researchers a ways to develop controls with good performances. The recent controls dominating in industry are, the control with constant V/f, the field oriented control FOC, and the direct torque control DTC. The first one is preferred in the applications with poor performances, whereas the FOC and DTC are much requested when they are high performances requirements. However, both present some disadvantages, like sensitivity to internals and/or externals uncertainties. What leaves the doors open to the researchers to develop new techniques to improve them. This is why this work presents the development of a structure of vector control, and in particular the indirect rotor field oriented control IRFOC. The first objective of this thesis is the introduction of the technique of space vector modulation SVPWM to control the inverter in order to attenuate the currents and torque ripples. The second objective is the use of the variable structure systems by sliding mode applied to the controllers for the robustness against uncertainties. Lastly, the objective is to remove the velocity and the load torque sensors by the introduction the method of model reference adaptive system MRAS, and using Luenberger observers type.