Title :
A neural networks based field oriented control scheme for induction motors
Author :
Ba-Razzouk, A. ; Cheriti, A. ; Olivier, G.
Author_Institution :
Section Electrotech., Ecole Polytech. de Montreal, Que., Canada
Abstract :
This paper attempts to introduce the artificial neural network (ANN) principles in the development of an ANN based field oriented control (FOC) of an induction motor. A two-layer ANN has been trained offline to estimate the slip, the direct and the quadrature current commands that are to be used in implementing an indirect FOC scheme. Simulation work indicates the feasibility of this approach as an alternative to the conventional approach, and the practical considerations for implementing this network on a DSP development system are discussed. This paper also proposes a method for estimation and adaptation of the rotor time constant of indirect FOCs. Simulation results show excellent potentials of the proposed scheme
Keywords :
control system analysis; control system synthesis; induction motors; learning (artificial intelligence); machine control; machine theory; neurocontrollers; rotors; slip (asynchronous machines); artificial neural network; control design; control simulation; direct current command; field oriented control; induction motors; offline training; quadrature current command; rotor time constant; slip; two-layer ANN; Artificial neural networks; Frequency; Inductance; Induction motors; Mathematical model; Neural networks; Rotors; Stators; Synchronous motors; Torque;
Conference_Titel :
Industry Applications Conference, 1997. Thirty-Second IAS Annual Meeting, IAS '97., Conference Record of the 1997 IEEE
Conference_Location :
New Orleans, LA
Print_ISBN :
0-7803-4067-1
DOI :
10.1109/IAS.1997.628954
