Online position error correction technique for sensorless control of multipole permanent magnet machines

Abstract

In order to improve the reliability of electric machine drive systems, the position encoder is often replaced with an estimator, such as the extended Kalman filter. However, estimation errors can still occur, especially in machines with high pole number, commonly used in renewable energy systems. The high number of pole pairs amplifies the effect of estimation errors, leading to a substantial divergence between actual and controlled currents, potentially causing harm to the machine through the excessive heat generation or demagnetization of permanent magnets. To address this issue, an error compensation method has been proposed and tested in a control scheme for a tidal stream system based on a multipole dual-star permanent magnet synchronous generator. The method estimates the position error by determining the q-axis permanent magnet flux and correcting it through a PI regulator. Simulation results demonstrate the effectiveness of the proposed method, even with a non-null initial rotor position.