Skip to Main Content
This paper presents a novel speed estimation algorithm for sensorless speed control in a direct stator flux orientation of an induction motor. The speed observation uses the stator currents and voltages as inputs and, contrary to what is proposed this last decade, is achieved in a reference frame linked to the stator current space vector. This leads to a simple equation for speed calculation which is a function of the stator flux. In direct stator flux oriented control (DSFOC), the stator flux is available for measurement or estimated. In the case presented in this paper, the stator flux is estimated using a reduced order Kalman Filter (KF) algorithm to reduce the computational requirements. The proposed speed estimation algorithm is tested numerically on an FPGA real-time simulation platform used as a tool for rapid control prototyping. Real-time simulation results show that accurate and fast speed estimation is achieved even at standstill.