Torque ripple minimization in PM synchronous motors using iterative learning control

Parasitic torque pulsations exist in permanent magnet synchronous motors (PMSMs) due to nonsinusoidal flux density distribution around the air-gap, errors in current measurements, and variable magnetic reluctance of the air-gap due to stator slots. These torque pulsations vary periodically with rotor position and are reflected as speed ripple, which degrades the PMSM drive performance, particularly at low speeds. Because of the periodic nature of torque ripple, iterative learning control (ILC) is intuitively an excellent choice for torque ripple minimization. In this paper, first we propose an ILC scheme implemented in time domain to reduce periodic torque pulsations. A forgetting factor is introduced in this scheme to increase the robustness of the algorithm against disturbance. However, this limits the extent to which torque pulsations can be suppressed. In order to eliminate this limitation, a modified ILC scheme implemented in frequency domain by means of Fourier series expansion is presented. Experimental evaluations of both proposed schemes are carried out on a DSP-controlled PMSM drive platform. Test results obtained demonstrate the effectiveness of the proposed control schemes in reducing torque ripple by a factor of approximately three under various operating conditions.