Abstract
This paper employs the on-line parameter estimation to improve the position sensorless performance of a Permanent-Magnet Synchronous Generator (PMSG) drive. The generator resistance will vary with the operating current and temperature, which causes the estimating error of the rotor angle. Therefore, the accurate resistance estimation of the PMSG will improve the position sensorless performance. This PMSG drive system consists of feedback circuits, the current control loop, position sensorless algorithm and on-line parameter estimation scheme. The dynamic model of a PMSG is established for the development of current control, position sensorless and on-line parameter estimation. The current control and extended-EMF position sensorless control are demonstrated. The recursive least squares method is implemented to estimate the resistance of the PMSG. The implementation of PMSG resistance estimation will improve the accuracy of the rotor position estimation. All the control and estimation algorithms are digitally fulfilled by a microcontroller. The experimental results verify the accuracy of the estimated parameter and the improvement of the position sensorless algorithm.