No Detail Found!

Neural learning algorithm based mras rotor resistance estimator using reactive power technique for vector controlled induction motor drive.

Author(s)
M. Nandhini Gayathri Department/SEEE, SASTRA University, Thanjavur, India
R. Sankaran Department/SEEE, SASTRA University, Thanjavur, India
S. Himavathi EEE Department, Pondicherry Engineering College, Pondicherry, India
Abstract
This paper presents a new method for estimating the rotor resistance of an induction motor. The rotor resistance changes significantly with temperature and frequency. This variation has a major influence on the field oriented control performance of an induction motor due to the deviation of slip frequency from the set value. In conventional MRAS, the adaptation is done using a PI-controller. The MRAS approach using reactive power as a functional candidate for rotor resistance estimation makes MRAS computationally simpler and easy to design. This paper proposes the role of neural learning algorithm for adaptation in a MRAS based rotor resistance estimator. The proposed scheme combines the advantages of reactive power technique and the capability of neural network to form a scheme named “Neural Learning-Reactive Power Model Reference (NL-RPMR) based rotor resistance estimator” for Induction Motor Drives. In the NLRPMR scheme, the error between neural network model and reference model is back propagated to adjust the weights of the neural network of the estimator. The performance of NL-RPMR based rotor resistance estimator is extensively simulated and compared with the conventional MRAS method. The promising results obtained are presented.
Publication Details
Page(s) 356-364
DOI DOI not available
Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 45, Issue: 1, Year: 2012

Request the full text from PASTIC

More from Journal of Theoretical and Applied Information Technology
0

Citations

0

Downloads

26

Views

Copyright ©  PASTIC National Center, Islamabad www.pastic.gov.pk