Magnetizing current based improved rotor position and speed estimation of doubly-fed induction generator using model reference adaptive scheme

Abstract

An improved rotor position and speed estimation technique for doubly-fed induction generator (DFIG) has been described in this paper. DFIG is widely used as electrical generator in wind energy conversion system. For this wind generator, rotor position and speed information is extremely necessary for implementation of position/speed sensor-less control logics for attaining high dynamic performance. In this respect, magnetization current based rotor position and speed estimation technique is proposed. The estimation scheme adopts model reference adaptive system (MRAS). Various highly desirable features such as integrator- and differentiator-less operation, flux computation-less logic, least machine’s parameter sensitivity, start on fly and capability of operation on both sub- and super-synchronous modes have been achieved in proposed scheme. The improvement in proposed estimation model can be stated that it is free from magnetizing parameter mismatch irrespective of power level. On the other hand, existing excitation current based scheme does not provide accurate estimation due to deviation of magnetizing inductance in low output power condition i.e. limited operating range. Also, another traditional excitation current based technique includes direct flux estimation (subjected to flux dynamics directly) in addition to magnetizing parameter dependency. Both simulation and real-time validation of proposed estimation concept have been done to present superiority over traditional schemes.