Indirect Rotor Field Oriented Control of Induction Motor With Rotor Time Constant Estimation
This thesis presents an estimation technique of the inverse rotor time constant for Indirect Rotor Field Oriented Control (IRFOC) induction motor application. In this estimation technique two different equations are used to estimate the rotor flux in the stator reference frame. One of the equatio...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English |
Published: |
2004
|
Subjects: | |
Online Access: | http://psasir.upm.edu.my/id/eprint/5966/1/FK_2004_79%20IR.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | This thesis presents an estimation technique of the inverse rotor time constant for
Indirect Rotor Field Oriented Control (IRFOC) induction motor application. In this
estimation technique two different equations are used to estimate the rotor flux in the
stator reference frame. One of the equations is a function of the rotor time constant,
rotor angular velocity and the stator currents, and the other equation is a function of
measured stator currents and voltages. The equation that uses the voltage and the
current signals of the stator serves as reference model, while the other equation
works as an adjustable model with respect to the variation of the rotor time constant.
Measurements of two phases of the current, and speed using an optical encoder are
required in this estimation technique. The stator phase voltages are estimated from
the DC bus voltage and the switching commands signals with compensation of the
dead time effect.
Field oriented control of induction motor is gaining wides acceptance in high
performance AC motor drive applications. Field oriented control, in its both forms as a direct or indirect, gives the AC motor dynamics that are equivalent to that of a
DC motor. However, direct and indirect field oriented control suffer from specific
theoretical and practical problems. The approach of direct field oriented control with
Hall sensors for flux sensing has limitations governed by the physical structure of the
machine itself. On the other hand, the approach of indirect field oriented control of
induction machines is highly dependent on the rotor parameters, which are not easily
accessible for measurements except for the rotor speed.
In a DC motor, spatial relationship of the torque and flux is maintained by the
physical construction of the motor armature and field circuits. However, in an
induction motor such spatial relationship does not maintain as such machine has
usually a single terminal where electric power is supplied. Therefore, such
relationship is maintained by external control methods. In a basic IRFOC of an
induction motor, speed and phase currents are sensed in order to control the stator
current vector such a way so it can be resolved into two components, one is to
control the rotor flux and the other to control the motor torque. Successful
decomposition of stator current vector into these two components requires the
knowledge of the instantaneous position, of the rotor flux vector. Since the position of
the rotor flux vector is estimated in an IRFOC scheme, and is dependent on the
motor model (more specifically the rotor parameters), these parameters must be
obtained accurately and match the motor parameters at all times. Unfortunately, rotor
parameters vary and are not easily accessible for measurements. Therefore, this
uncertainty about the rotor flux vector position degrades the dynamic operation of
the drive.Enormous efforts have been made to improve IRFOC
complicated hardware and software in order to coixpensate for such imperfection.
Hence, this work focuses on the Indirect Rotor Field Oriented Control of induction
motors with estimation of the rotor time constant. A simple yet effective rotor time
constant identification method is presented and used for updating the slip calculator
used by the IRFOC algorithms.
A complete simulation model of an induction motor and IRFOC scheme is presented
and tested using SIMULINWMATLAB, and experimentally implemented on a DSP
Board (MCK243j without any need for voltage phase sensors. Simulation and
experimental results were presented and compared to verify the validity of the
proposed estimator for different operating conditions. |
---|