Sensorless Speed Estimation of Brushless Doubly-Fed Reluctance Generator Using Active Power Based MRAS

This paper proposes an active power based model reference adaptive system for sensorless speed estimation of primary field oriented brushless doubly-fed reluctance generator (BDFRG). In this regard, the active power associated with secondary winding of BDFRG is used for the process of speed estimation. BDFRG is more reliable and requires less maintenance. Its structural advantages demonstrate higher working efficiency and lower losses as compared to equivalent doubly-fed induction machines. Moreover, active power of BDFRG can be controlled by suitable speed regulation, which in turn can be achieved by a low rating bidirectional converter connected to the stator secondary (i.e., control) winding. These inherent features make BDFRG as a prospective candidate in the applications like wind and tidal power generation. The efficacy of the proposed control technique is established by a 1.6 kW BDFRG in Matlab/Simulink. A dSPACE-1103 based laboratory prototype is used for the hardware validation.