Robust Performance of Induction Motor Drives

This paper is an attempt to explore the possibility of using Extended Kalman Filter trained recurrent neural network for speed estimation of an induction motor drive. Also, the speed estimation is made robust by simultaneously adapting the rotor resistance and the rotor flux by using the same neural network. The training is very fast as it requires only one iteration. The proposed scheme is studied on an induction motor and it is seen that it gives better performance as compared to already existing algorithms in literature. Key words—Extended Kalman Filter, Recurrent Neural Network, Robust speed sensor less operation, Vector control

[1]  Antonio Mendoza,et al.  Diagnosis and Control in Induction Motor with Extended Kalman Filter , 2003, Modelling, Identification and Control.

[2]  H. Razik,et al.  Neural networks applied to the control of an induction motor , 2000, 7th IEEE International Power Electronics Congress. Technical Proceedings. CIEP 2000 (Cat. No.00TH8529).

[3]  Rubén Alejos Palomares,et al.  Speed estimation for an induction motor using the extended Kalman filter , 2004, 14th International Conference on Electronics, Communications and Computers, 2004. CONIELECOMP 2004..

[4]  Wang Zhenlei,et al.  Identification and control of induction motor using artificial neural networks , 2001, ICEMS'2001. Proceedings of the Fifth International Conference on Electrical Machines and Systems (IEEE Cat. No.01EX501).

[5]  T. Kanmachi,et al.  Sensor-less speed control of an induction motor , 1995 .

[6]  Adel Razek,et al.  Reduced-order observers for rotor flux, rotor resistance and speed estimation for vector controlled induction motor drives using the extended Kalman filter technique , 1999 .