Abstract
Electrical machine is one of the major work forces in the industries and serve them in various capacity. Machine drives control involves many technical challenges for efficient and reliable operation. Over several decades, Induction Motor drives have been the most preferred motor drives for industrial applications, but recently, that momentum has shifted toward Permanent Magnet Synchronous Machine (PMSM) drives due to their high efficiency and power density.
In light of the fact that PMSM drives are considered as the drives of future, there lies a lot of technical challenges and issues related to PMSM control. Many PMSM control techniques have been proposed in the past, but there are still some open research areas that brings the worldwide researchers interest back to the PMSM drives control. One of the aspects, which relates to the cost effectiveness, robustness and reliability leads the sensorless control of a PMSM. Sensorless control is adopted due to its ability to estimate the rotor position inherently without any physical transducers for measuring the rotor position, hence reducing the overall cost and improving the reliability.
The work presented in this thesis is the result of studies carried out at Electric Drives Laboratory at Free University of Bozen-Bolzano and collaborative efforts at University of Udine. The topics explored are mainly concentrated toward PMSM control especially high frequency based sensorless control and machine parameters identification. Due to growing technologies and changing applications, industrial drives keeps posing different challenges and that leads the way for researchers interest to overcome these issues.
The objective of this thesis is to explore various technical aspects for the efficient use of electric drives with sensorless control. The focus is to highlight the challenges associated with a PMSM control and to investigate these issues thoroughly by providing a novel solution. Sensorless technique is usually implemented by HF injection-based methods for low speed range operation and their suitability for different varying conditions, choice of frequency, amplitude are few of the factors that changes under given application. The objective is to investigate the feasibility to design a self-adaptive HF injection and auto-tuning of the PLL gains to reduce the implementation burden and to make the PMSM drive control more user-friendly. Also reducing the associated losses due to undesired current resulting from the injected HF signal. Since motor control requires an accurate information about motor parameters to design a high performance controller, it becomes even more important to look for methods that can estimate the motor parameters accurately. Permanent Magnet flux is one of the parameters, which is required for tuning, and management in flux based sensorless algorithm, axes decoupling for current control, tuning of flux-weakening algorithms etc. Another objective is to devise a method that estimates the PM flux amplitude accurately with simple implementation algorithm to make the PMSM drive more robust, accurate and efficient.
In most part of the work carried out in this thesis, a broader perspective related to the practical implementation has been kept throughout the work using simpler implementation schemes with less dependence on the parameters to avoid any extensive off-line tests on a single machine.