SENSORLESS CONTROL OF GRID CONNECTED DOUBLY FED INDUCTION MOTOR DRIVE

Abstract

A simple and implicit sensorless algorithm for estimating the rotor position and speed of the doubly fed induction machine is proposed in this thesis. Instead of computing the stator magnetic flux directly or indirectly, the stator flux vector components in the stationary reference frame are substituted with analytical equivalents in terms of measurable stator and rotor quantities. Thus, it is an implicit estimation process devoid of computing any flux terms and requires only the measured stator voltage and currents in stator and rotor windings apart from knowledge of the machine parameters. it is an open-loop scheme. 1-lowever, it does not involve integration, recursive techniques, re-computations or programmable low-pass filters etc. Under this scheme, accurate speed measurement, immunity against variation in stator resistance and rotor resistances or in the magnetising inductance are the major advantages of this thesis. Reduced complexity and the computational burden enable the easy implementation of the algorithm on a low-cost fixed point processor. In such high-precision drives, sensor-less control is preferred because of the reduced cabling and hardware complexity, increased reliability, and low maintenance requirements On the whole, the estimation algorithm is a new vision that holds ability in the area of sensorless control of the DFlM, whether used as a motor or as a generator. 1-lere the discussed electrical system is modeled and simulated in MATLAB/SIMULINK software.