Real-time Sensorless Vector Control of Induction Machines using an FPGA board

The aim of the work described in this paper is to present a real-time Sensorless Vector Control method (SVC) of Induction Machines (IM) using an FPGA board. It´s well known that the performance of sensorless control of induction motor drives is generally poor, especially at low speed due to measurement accuracy (offset, voltage distortions, current unbalances, machine parameter variation etc.). The approach used in this article refers to the real-time Extended Kalman Filter (EKF) to overcome the incertitude of the measurement by considering noise. From the measurement of stator voltages and currents, the rotor speed of the Induction Motor is estimated in real time. The results obtained from SVC are then compared to standard Vector Control (VC) which uses a speed sensor. Tests in different conditions demonstrate excellent steady-state operation and good dynamic performance.