DocumentCode
3433545
Title
The investigation of ANN space vector PWM and diagnostic controller for four switch three phase inverter fed induction motor drive
Author
Lee, Hong Hee ; Dzung, Phan Quoc ; Hoa, Truong Phuoc ; Phuong, Le Minh
Author_Institution
Sch. of Electr. Eng., Univ. of Ulsan, Ulsan
fYear
2009
fDate
10-13 Feb. 2009
Firstpage
1
Lastpage
6
Abstract
This paper is to present the investigation of neural-network-based controller of space vector modulation (ANN-SVPWM) for four switch three phase inverter (FSTPI) fed induction motor drive. This ANN-SVPWM controller completely covers the under modulation and over modulation modes with operation extended linearly and smoothly up to square wave (six-step). The ANN controller uses the individual training strategy with the fixed weight and supervised models. Furthermore, ANN diagnosis method for real-time fault detection of power switches is proposed in this paper. The complete ANN-SVPWM and diagnostic controller can be used in power applications such as motor drives. A computer simulation program is developed using Matlab/Simulink together with the neural network toolbox for training the ANN-controller.This method has been validated experimentally using kit ACE 1104 (DSPACE) .
Keywords
PWM invertors; induction motor drives; machine control; neurocontrollers; diagnostic controller; four switch three phase inverter; neural network toolbox; neural-network-based controller; real-time fault detection; space vector modulation; three phase inverter fed induction motor drive; Application software; Artificial neural networks; Fault detection; Fault diagnosis; Induction motor drives; Mathematical model; Phase modulation; Pulse width modulation inverters; Space vector pulse width modulation; Switches;
fLanguage
English
Publisher
ieee
Conference_Titel
Industrial Technology, 2009. ICIT 2009. IEEE International Conference on
Conference_Location
Gippsland, VIC
Print_ISBN
978-1-4244-3506-7
Electronic_ISBN
978-1-4244-3507-4
Type
conf
DOI
10.1109/ICIT.2009.4939636
Filename
4939636
Link To Document