Sensorless control of PMSG for wind turbines based on the on-line parameter identification

This paper proposes a position-sensorless control strategy combined with the on-line parameter identification for a permanent magnet synchronous generator (PMSG) which is applied in the wind power generation system. The accuracy of the position estimation based on the phase-locked loop (PLL) technique can be improved by the proposed method, in which the motor parameters including the d- and q-axes inductances used for the position estimation and the current close-loop control, are identified using the recursive least-squares (RLS) method. Simulation study demonstrates that the proposed self-sensing method is reliable and effective for PMSG-based wind power generation system and is capable of achieving more accurate position estimation than the generation system without parameter identification.