DocumentCode :
3015342
Title :
Research of Maximum Ratio of Torque to Current Control Method for PMSM Based on Least Square Support Vector Machines
Author :
Zhao, Jun ; Liu, Weiguo ; Tan, Bo
Author_Institution :
Coll. of Autom., Northwestern Polytech. Univ., Xi´´an, China
fYear :
2010
fDate :
25-27 June 2010
Firstpage :
1623
Lastpage :
1628
Abstract :
Efficiency optimization control of salient pole permanent magnet synchronous motor (PMSM) based on maximum ratio of torque to current is presented. From basic equations of PMSM in reference frame, the method of maximum ratio of torque to current was derived using maximum principle. Then, based on the Least Squares Support Vector Machine (LSSVM), modeling of LSSVM controller of maximum ratio of torque to current by using simulation data. The simulation result shows that LSSVM has a good generalization ability and capability of modeling nonlinear process, and the original sampling data are very similar to the data predicted by LSSVM model. And using method of LSSVM maximum ratio of torque to current, the simulation of PMSM shows that the PMSM control system has a good dynamic ability and achieves efficiency optimization control.
Keywords :
electric current control; generalisation (artificial intelligence); least squares approximations; machine control; maximum principle; nonlinear control systems; optimisation; permanent magnet motors; support vector machines; synchronous motors; torque control; LSSVM controller; PMSM control system; current control method; dynamic ability; efficiency optimization control; generalization ability; least square support vector machines; maximum principle; maximum ratio; nonlinear process; reference frame; salient pole permanent magnet synchronous motor; torque control; Electromagnetics; Equations; Inspection; Mathematical model; Support vector machines; Torque; Training; least square support vector machine; maximum ratio of torque to current; permanent magnet synchronous motor;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electrical and Control Engineering (ICECE), 2010 International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-6880-5
Type :
conf
DOI :
10.1109/iCECE.2010.400
Filename :
5631678
Link To Document :
بازگشت