An Improved Rotor Speed Observer for Standalone Brushless Doubly-Fed Induction Generator Under Unbalanced and Nonlinear Loads

The conventional control methods for brushless doubly-fed induction generator (BDFIG) normally employ mechanical sensors to acquire the information of rotor speed, which brings many disadvantages in the cost, complexity, reliability, etc. This paper presents an improved rotor speed observer (RSO) for the sensorless operation of standalone BDFIGs, which is based on the power winding (PW) voltage and control winding (CW) current. In order to eliminate the impact of unbalanced and nonlinear loads on the RSO, second-order generalized integrators (SOGIs) and low-pass filters (LPFs) are introduced to pre-filter the PW voltage and CW current, respectively. Through comprehensive parameter design, the response speed of the improved RSO will be not lower than that of the basic RSO with ensuring the filtering effect of these additional filters. In addition, the proposed RSO is independent to machine parameters except the pole pairs. Comprehensive experiments are conducted and results verify the proposed improved RSO applied to the standalone BDFIG. Also, the applicability of the proposed RSO on another dual-electrical-port machine, doubly-fed induction generator, is confirmed by simulation results.