Title :
Control Method of Sensorless Brushless DC Motor Based on Neural Network
Author :
Zhang Jinquan ; Gu Deying
Author_Institution :
Dept. of Autom. Eng., Northeastern Univ. at Qinhuangdao, Qinhuangdao, China
fDate :
May 31 2014-June 2 2014
Abstract :
Control Method of Sensorless Brushless DC motor Based Neural Network is proposed by analysis of indirect position detection of DC Brushless motors´ principles. Dynamical BP neural network model is built to identify the relation of phase voltage, phase current and rotor position of Brushless DC Motor. Network parameters gradient descent is employed to rectify error method in order to update the network parameters in stead of traditional position sensor, which the commutation signal of DC motor is achieved by terminal voltage and phase current mapping. The effectiveness of the control method is verified by simulations.
Keywords :
backpropagation; brushless DC motors; gradient methods; neurocontrollers; sensorless machine control; dynamical BP neural network model; error method; indirect position detection analysis; network parameter gradient descent method; phase current; phase current mapping; phase voltage; position sensor; rotor position; sensorless brushless DC motor control method; terminal voltage; Brushless DC motors; Commutation; Filtering; Neural networks; Rotors; BP Neural Network; Brushless DC Motor; Nonlinear system identification; Sensorless position detection;
Conference_Titel :
Control and Decision Conference (2014 CCDC), The 26th Chinese
Conference_Location :
Changsha
Print_ISBN :
978-1-4799-3707-3
DOI :
10.1109/CCDC.2014.6852987
