The paper proposes a PLL-based model reference adaptive observer (PLL-MRAO) for rotor speed and position estimation in sensorless vector control of DFIG for stand-alone and grid-connected operations. Most of the existing MRAOs require integration of sensed voltages and currents for generation of the reference or estimated vectors in stator reference frame. The presence of small DC offsets in the sensed quantities can result in integrator drift, which can lead to malfunctioning of the control algorithm. The proposed PLL-MRAO eliminates the need for integration by generating the reference and estimated vectors in a synchronously revolving dq frame. The dynamics of the proposed PLL-MRAO is also superior to that of most of the existing non-linearized MRAOs. This is demonstrated experimentally on a 10 hp DFIG coupled to an induction motor. Experimental results pertaining to the utility of the proposed observer in synchronization of DFIG to the grid and sensorless vector control of stand-alone and grid-connected DFIG are provided in the paper.
