The Design Of Brushless Dc Motor Servo System Based On Wavelet Ann - Machine Learning And Cybernetics, 2004 Proceedings Of 2004 International Conference On.pdf

Proceedings of the Third International Conference on Machme Learning and Cybemetics, Shanghai, 26-29 August 2004
THE DESIGN OF BRUSHLESS DC MOTOR SERVO SYSTEM BASED ON
WAVELET ANN
XIAN-XIANG LI', QIAN-JIN ZHANG', HONG-JUN XIAd
I Engineering School, Foshan University, Foshan, Guangdong 528000, China
2Department of Automatic Control Engineering, South China University of Technology, Guangzhou 510640,China
E-MAIL:******@fosu,
Abstract: Variational load and exterior disturbance affect performance
The paper presents a new approach to brushless DC of the servo system directly. Thrust ripples e into
motor =NO system, which is based on artificial neural being because of end-effect, slot effect, and PM flux
network (ANN) and wavelet transbrm. The approach is linkage harmonic wave of motor. Moreover. the servo
designed based on a three-layer forward artificial neural performance of system will lower because of parameters
network, which :trains and work parameten
variety of .system (active cell mass, viscous friction
in-line using a gradieot descending error algorithm. The
working and fault states of brushless DC motor are detected, coefficient and so on), non-linear variety of friction drag,
in which the time-frequency characterktics of diserete wavelet state observation noise, etc.
transform ***@Wand Id maximum .of modulus of From the analysis above,