DocumentCode :
2542635
Title :
A new high performance intelligent speed controller for induction motor based on supervisory loops
Author :
Kamalasadan, Sukumar
Author_Institution :
Univ. of West Florida, Pensacola, FL
fYear :
2008
fDate :
20-24 July 2008
Firstpage :
1
Lastpage :
7
Abstract :
A novel high performance intelligent controller for precise speed control of Induction Motor (IM) based on supervisory loop principle is proposed. The scheme consists of online growing Radial Basis Neural Network (RBFNN) controllers working in parallel with an implicit model adaptive controller based on system supervision. The proposed intelligent supervisory loop control architecture monitors system movement and compensate for drastic changes in system parameters and model errors. The main advantage of this unique architecture based on vector control is that, it is precise, feasible and more effective especially when the IM is subjected to unknown dynamics. Simulation results are presented in order to show that, under model errors and drastic changes, the tracking errors are reduced while using the proposed hybrid control scheme at various IM operating conditions.
Keywords :
adaptive control; induction motors; intelligent control; machine control; neurocontrollers; radial basis function networks; velocity control; adaptive controller; hybrid control; induction motor; intelligent speed controller; intelligent supervisory loop control architecture; radial basis neural network controller; system supervision; Adaptive control; Control system synthesis; Error correction; Induction motors; Intelligent control; Magnetic variables control; Neural networks; Programmable control; Sliding mode control; Velocity control; Implicit Model Adaptive Controller (IMAC); Induction Motor Speed control; Intelligent Controller; Intelligent Supervisory Loops (ISL); Radial Basis Neural Network (RBFNN);
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Power and Energy Society General Meeting - Conversion and Delivery of Electrical Energy in the 21st Century, 2008 IEEE
Conference_Location :
Pittsburgh, PA
ISSN :
1932-5517
Print_ISBN :
978-1-4244-1905-0
Electronic_ISBN :
1932-5517
Type :
conf
DOI :
10.1109/PES.2008.4596711
Filename :
4596711
Link To Document :
بازگشت