Torque ripple minimization in PM synchronous motors - an iterative learning control approach

Abstract

Permanent magnet synchronous motor (PMSM) drives are widely used for high-performance applications where torque smoothness is an essential requirement. Parasitic torque pulsations in PMSM are generated due to harmonics in the stator flux linkage, cogging, current offsets etc. This paper describes a new instantaneous torque control scheme - iterative learning control (ILC), for torque ripple minimization. The proposed scheme has the advantage that torque ripples produced by the non-sinusoidal flux distribution is periodic in nature and therefore can be eliminated. This scheme is based on a dynamic torque controller and the conventional current controllers to minimize torque ripples. The dynamic torque controller compares the desired motor torque with the instantaneous motor torque, and generates the reference current (iq(-)ref) iteratively from cycle to cycle so as to reduce the error. The effectiveness of the proposed ILC scheme for torque ripple minimization is evaluated through simulation studies. Results obtained show that torque ripples can be significantly reduced using the proposed ILC scheme.