2008 IEEE Region 10 Colloquium and the Third International Conference on Industrial and Information Systems, Kharagpur,
INDIA December 8 -10, 2008.
>497 <
978-1-4244-2806-9/08/$25.00© 2008 IEEE 1
UNIVERSAL PLL STRATEGY FOR SENSORLESS
SPEED AND POSITION ESTIMATION OF PMSM
Georges el Murr,Damian Giaouris, and J.W. Finch
School of Electrical, Electronic and Computer Engineering
Newcastle University
Newcastle upon Tyne, United Kingdom
Damian.Giaouris@ncl.ac.uk
Abstract— the implementation of the Phase Locked Loop (PLL)
structured observer in high frequency signal injection schemes
requires tuning of the PI controller to lock on the rotor position.
The PI controller is tuned depending on the Signal to Noise Ratio
(SNR) which is related to the amount of saliency/anisotropy
presented in the machine, the amplitude and phase of the
incoming modulated carrier signal. Thus, in case any of the
mentioned factors changes then the PI has to be retuned again,
otherwise it will loose track. In this paper a new PLL structure is
suggested to avoid any disturbance that affects the demodulation
and detection process. The new PLL scheme is totally
independent of any machine parameters except the
saliency/anisotropy, which is essential to obtain the rotor speed
and position information.
Keywords-component; High-frequency signal injection, Phase-
Locked Loop, Sensorless control, Surface-Mounted Permanent
Magnet Synchronous Motor
I. INTRODUCTION
The main objective of sensorless control is to achieve the
same control performance as vector control without using
speed or position sensor [1]. Such a sensor can be the most
expensive and fragile component in the whole electric drive.
Sensorless techniques Fig.1 are divided into two classes, those
using the fundamental properties or model of the machine [2-5]
and those exploiting subsidiary features often called
anisotropies based model [6-9].
Fig.1: Categories of sensorless control techniques
Sensorless control of a PMSM has become more and more
acceptable in industrial applications due to the large amount of
research effort that has been spent in developing reliable, low-
cost PMSM drives. The simplest fundamental model methods
are based on rotor flux position estimation, by integrating the
back-EMF [10]. Such an approach is very simple but fails at
low and zero speed. In general, low speed operation is critical
for such fundamental based methods, since the estimated rotor
flux can be very sensitive to stator resistance variations, and
measurement noises. In addition, sensorless schemes that are
based on the application of sophisticated identification
procedures (such as flux observers, Kalman filter [1-3] etc.)
allow, with some limitations very low speed operation, but can
be too complex and expensive to be used in practical systems.
Other estimation techniques can be based on stator phase
voltage third harmonics or spatial phenomena inherent within
the machine, such as slot harmonics, rotor/stator eccentricities,
and winding asymmetries. However, all these approaches can
fail at zero speed for lack of useful information.
Signal injection schemes [11-15] can be more efficient, at
low and zero speed, than any other sensorless estimation
scheme. Such methods have the capability to provide accurate
position and speed estimation without requiring information
about the motor parameters. This is due to the presence of
saliency/anisotropy in the machine. The saliency/anisotropy is
the inductance difference between direct-inductance l
d
and
quadrature-inductance l
q
and it is due to either the asymmetric
structure of the machine or the flux induced magnetic
saturation due to the fundamental excitation. The interaction
between the injected high frequency signal and the spatial
variations in the machine inductances produces an amplitude-
modulated signal containing useful information attached to the
carrier frequency. The demodulation and extraction of this
information constitute the main interest of this paper, which is
presented as follows:
In section II a theoretical background of the conventional
modulation, heterodyning demodulation, and the PLL
observers implemented in rotating and alternating signal
injection schemes. Section III covers the proposed
demodulation as well as the speed and position estimation
method. Finally, results of the estimated and actual rotor speed
and position are presented in section IV to validate the
proposed method and compare it with the conventional
schemes.
IEEE Kharagpur Section & IEEE Sri Lanka Section
Authorized licensed use limited to: Newcastle University. Downloaded on March 16,2010 at 10:17:08 EDT from IEEE Xplore. Restrictions apply.