Neural networks and fuzzy controllers are considered as the most efficient approximators of different functions and have also proved their capability of controlling nonlinear dynamical systems. So, in this paper, the authors introduce a novel technique of control called 'hybrid control' which is Based on Feedback Linearization and Field Oriented Control of an Induction Motor, in order to replace the sliding mode controllers (speed and flux ones). In fact, the objectives required by the introduction of neural networks, 'RANNCs' is to perform the control which is shown by simulation results. Resume - Les reseaux de neurones et de controleurs flous sont consideres comme les plus efficaces approximateurs des differentes fonctions et ont egalement montre leur capacite de controler des systemes non lineaires. Ainsi, dans cet article, les auteurs ont introduit une nouvelle technique de controle appelee 'controle hybride', basee sur le feedback de la linearisation et sur la thematique axee sur le controle d'un moteur a induction, afin de remplacer les controleurs de mode de glissement (vitesse et flux). En fait, les objectifs requis par l'introduction de reseaux de neurones (RANNCs) consistent a effectuer le controle, qui est demontre par les resultats de simulation.
[1]
Manish Kumar,et al.
Neural Controllers
,
2008,
Wiley Encyclopedia of Computer Science and Engineering.
[2]
Vadim I. Utkin,et al.
Sliding mode control design principles and applications to electric drives
,
1993,
IEEE Trans. Ind. Electron..
[3]
Okyay Kaynak,et al.
A comparative study of neural network structures in identification of nonlinear systems
,
1999
.
[4]
Park Min-Ho,et al.
Chattering reduction in the position control of induction motor using the sliding mode
,
1989
.
[5]
F. Filippetti,et al.
Neural networks aided on-line diagnostics of induction motor rotor faults
,
1993,
Conference Record of the 1993 IEEE Industry Applications Conference Twenty-Eighth IAS Annual Meeting.