DocumentCode
1555119
Title
Torque ripple minimization in permanent magnet synchronous servodrive
Author
Colamartino, F. ; Marchand, C. ; Razek, A.
Author_Institution
Lab. de Genie Electr., Paris XI et VI Univ., Gif-sur-Yvette, France
Volume
14
Issue
3
fYear
1999
fDate
9/1/1999 12:00:00 AM
Firstpage
616
Lastpage
621
Abstract
In this paper the electromagnetic torque developed in permanent magnet synchronous motors is analyzed. Two strategies are discussed for torque ripple minimization. The first is based on numerical predetermination of the current waveform which is imposed by the control in machine phases to obtain a constant torque. The second is a torque regulation based on its on-line instantaneous estimation. To optimize the current waveform and minimize copper losses, two methods for current modulus minimization are compared. One is relative to the optimal current which leads to a perfect constant instantaneous torque with minimum instantaneous current modulus, and the second concerns only the mean current modulus. Relation between current components and torque reference can be expressed by functions after normalization. These functions optimizing the mean torque are easy to implement and can be used for torque ripple minimization
Keywords
electric current control; machine control; permanent magnet motors; servomotors; synchronous motor drives; torque control; constant torque; copper losses minimisation; current components; current modulus minimization; current waveform optimisation; electromagnetic torque; mean current modulus; minimum instantaneous current modulus; on-line instantaneous estimation; permanent magnet synchronous motors; permanent magnet synchronous servodrive; torque reference; torque regulation; torque ripple minimization; Artificial intelligence; Equations; Least squares methods; Minimization methods; Newton method; Optimization methods; Permanent magnets; Recursive estimation; Synchronous motors; Torque;
fLanguage
English
Journal_Title
Energy Conversion, IEEE Transactions on
Publisher
ieee
ISSN
0885-8969
Type
jour
DOI
10.1109/60.790924
Filename
790924
Link To Document