Commande Neuronale Prédictive d’une machine asynchrone double étoile sans capteur mécanique
Loading...
Date
2015-10-07
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Résumé (Français et/ou Anglais) :
The work presented in this thesis deals with a subject entitled ' Sensorless Neural Predictive Control of a Double Star Induction Machine.' 'We performed first, the model of the DSIM fed by three-level voltage source inverter as well as direct and indirect vector control, allowing us to decouple the electromagnetic torque and flux. We adopted the classical PI controllers for the speed control, the flux, and thus for setting the stator currents. Then we are interested in the replacement of these regulators, by predictive controllers, this after developing a cascade predictive structure. In order to minimize the transient control and reduce the impact of measurement noise on the control signal, we used multivariable generalized predictive control instead the vector control, which must require a mechanical sensor that is considered fragile and expensive, to avoid these inconveniences, we used two different techniques. That is the MRAS structure and reduced order observer.
All predictive controllers introduced in the neural predictive control form. The results show, the effectiveness of the proposed method especially in the load disturbances and/or the change of the reference speed.
Description
Doctorat en Sciences
