EPE 2001 - Topic 06g: Estimation Techniques | ||
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2001 - Conference > EPE 2001 - Topic 06: ELECTRICAL MACHINES AND ADJUSTABLE SPEED DRIVES > EPE 2001 - Topic 06g: Estimation Techniques | ||
![]() | [return to parent folder] | |
![]() | A new automatic slip gain tuning procedure for IFOC motor drives
By A. Consoli; G. Scarcella; A. Testa | |
Abstract: The paper presents a straightforward technique to automatically tune the slip gain in Induction Motor
Drives based on the Indirect Field Oriented Control approach. A key feature of the proposed
technique is that it is effective even at very low speed with light loads, where standard tuning
procedures generally fail. Moreover it does not require the knowledge of any motor parameter.
According to the proposed technique the slip gain tuning is performed by processing the effects
exerted by an additional high frequency signal on the zero sequence stator voltage. As a result of the
interaction between the main air gap flux and a high frequency flux generated by the additional
voltage signal, a new zero sequence stator voltage component is generated at a frequency equal to the
difference between the frequency of the main air gap flux and the frequency of the additional high
frequency signal. By comparing the frequency of the air gap flux, estimated from the zero sequence
stator voltage and the frequency computed by the indirect field oriented control algorithm it is
possible to on line compensate rotor time errors. The proposed technique can be used either for the
initial drive commissioning, either for on-line tuning. Experimental results are presented to show that
the procedure is effective and does not interfere with normal operations of the drive.
| ||
![]() | A radial force estimation with search coil fluxes in a bearingless..
By K. Kiryu; A. Chiba; T. Fukao | |
Abstract: In this paper, a principle of radial force estimation based on the induced voltage in search coils
is shown. The proposed method is applied to a test machine. It is confirmed that the estimated radial
force is corresponding to the generated radial force in both static and dynamic tests.
| ||
![]() | A Reduced order Kalman Observer for Induction Motor Flux Estimation
By A. Bellini; S. Bifaretti; S. Costantini | |
Abstract: Many field oriented induction motor drives use a state observer to determine the components of the
rotor flux representative vector. To this aim, different kinds of observers, based on deterministic or
stochastic models, are used. The employment of a stochastic observer needs much more math
operations; therefore its implementation on a microcomputer-based control circuit requires suitable
simplifications. The paper presents a new reduced order stochastic observer, whose implementation
requires a number of math operations that can be easily on line performed by recent microprocessors.
| ||
![]() | A self-organising neuro-fuzzy speed estimator for DC motor drives
By D.Kukolj; E. Levi | |
Abstract: The paper deals with an application of a self-organising Takagi-Sugeno fuzzy rule-based model,
combined with a feed-forward neural network, as a speed estimator for separetely excited DC motor
drives. The proposed speed estimator is characterised with moderate computational requirements and
with a simple knowledge base. These features have enabled implementation of the speed estimator in
a low-cost 20-bit MASC 3500 digital signal processor. The implemented speed estimator was tested
for a variety of operating conditions and a very satisfactory accuracy was achieved.
| ||
![]() | Adaptation of VC of AC. Drive for Rotor Time Constant Variation
By P. Palacky; P. Brandstetter | |
Abstract: The range of controlled electrical drives is greatly expanded by the transition to A.C. motors. Control
in a moving reference frame, such as field orientation, presents a unifying principle for the design of
high dynamic performance drives. The main problem of the vector control in the field co-ordinates of
motor is the separation of torque and flux control circuits without mutually influencing. For this
purpose the knowledge of the value and position of the rotor flux vector is necessary. Parameters of
the machine are not constant. Especially the stator and rotor resistances are varied in dependence on
the changing temperature of the motor. In this paper is presented the rotor time constant adaptation
method.
| ||
![]() | Determination of Frequency Response Functions of Synchronous Machines
By N. Dedene; R. Pintelon; Ph. Lataire | |
Abstract: This paper presents a multiple-input multiple-output method for the measurement and identification of
the d-axis impedance functions of synchronous machines. The proposed procedures are tested at
different saturation levels on a 130kVA salient pole synchronous machine. The standstill frequency
response estimated main inductances are compared to the no-load open-circuit characteristic.
| ||
![]() | Estimating the Back-EMF of an Induction Motor
By V-M. Leppänen; J. Luomi | |
Abstract: Information about the back-emf of an induction motor can be used in various control and
estimation schemes. This paper proposes a simple method for fast tracking of the back-emf based on
the evaluation of the derivative of the stator current during a PWM switching period. No prior
knowledge of motor parameters is needed. Simulation results show that the estimated back-emf is in
close agreement with the actual one even under transient conditions.
| ||
![]() | Estimation of IM Variables based on ANN Utilising Apriori Information
By J. Žilková; J. Timko; V. Fedák | |
Abstract: Artificial Neural Networks (ANN) is utilised in estimator for system variables identification. Using
ANN approach the induction motor rotor flux observer has been developed. The method of estimation
of rotor flux is based on feedforward on-line learning ANN. Obtained results show good performance
both in steady and transient states of the developed observer. The estimator is robust against variances
of the rotor resistance.
| ||
![]() | Generalized Rotor-Flux-Oriented Control: Design and Analysis
By L. Harnefors | |
Abstract: We in this paper consider reduced-order observers for rotor flux estimation of induction motors. The
traditional ”current” and “voltage” models are shown to be special cases. It is also shown that the
flux dynamics form a nonlinear closed-loop system when the flux estimate is used for field orientation.
Proper observer gain selection is essential for good behavior of this system. A framework is developed,
in which the properties of any gain selection can be assessed with ease. Candidate gain selections
are considered, two of which yield schemes that do not use the rotor speed in their equations (socalled
inherently sensorless schemes). It is finally shown that for any gain selection, an equivalent
synchronous-frame implementation (i.e., indirect field orientation) always exists.
| ||
![]() | Impact of iron loss on direct torque control of induction machines
By E.Levi: T.Pham-Dinh | |
Abstract: General theory of electrical machines neglects the existence of the iron loss. As a consequence, all the
high performance control schemes are characterised with certain amount of detuning, caused by the
iron loss. Impact of iron loss on operation of field oriented induction machines has been recently
subjected to detailed scrutiny and appropriate methods of iron loss compensation have been
developed. As far as direct toque control (DTC) of induction machines is concerned, it appears that a
corresponding study has never been done. This paper discusses two aspects related to the problem of
iron loss in DTC of induction machines: detuning effects and means for compensation. A detailed
simulation study is at first performed and the influence of iron loss on accuracy of the DTC scheme is
evaluated. It is shown that the net consequence of the iron loss existence is a difference between the
torque reference and the average motor output torque, that may be substantial. On the other hand,
accuracy of stator flux magnitude and position estimation, being based on stator voltage and current
measurement, is not affected by the iron loss existence. On the basis of the results of the detuning
study, an iron loss compensation method is developed. The method can be used in an extremely
simple form for a variety of induction machines, that consists of deduction of a constant, iron loss
related, torque component from the initial torque estimate.
| ||
![]() | Luenberger Observer for High Speed Induction Machine Drives based on a New Pole Placement Method
By G.Griva; P.Ferraris; F.Profumo; R.Bojoi; R. Maceratini; G. Barba | |
Abstract: The Luenberger Observer is one of the most attractive
solutions to estimate the magnetic flux in Induction
Machine Drives. In fact, it is possible to obtain a speed
estimation for “sensorless” vector control with a simple modification of the algorithm. Nevertheless, when high operation speeds are needed, the
discrete-time implementation requires a considerable
computational effort because of the speed-dependent continuous-time model. This paper presents a new method to calculate the
feedback gain matrix which generalizes a well known
result and can guarantee stability of the observer over a
wide speed range. The accuracy can be further improved
by a simplified second-order approximation in the
discrete-time rotor equations. The overall computational
cost is very low and it is comparable to a first-order approximation.
| ||
![]() | Mechanical Parameter Identification for High Performance Drives
By A. Bellini; S. Bifaretti; S. Costantini | |
Abstract: In high performance drives it is important to have a rapid and precise control of the motor rotation
speed. To this aim a clean measure of the speed is necessary to have; it is useful to identify the load
parameters so to increase the sturdiness of the system and, eventually, proceed to a self-tuning of the
speed regulator. This paper proposes to employ a parameter estimator composed by a Steady State
Kalman Filter (SSKF) that furnishes dynamic variables (speed and acceleration), and by a Recursive
Least Square algorithm (RLS) that provides the load parameter.
| ||
![]() | Microcomputer-based control system for the drives with resilient coupl
By M. Tondos; W. Mysinski | |
Abstract: This paper is concerned with principles of the estimation of non-measurable state variables and
external torque load in drives with an elastic motor shaft (two mass system). In many drive systems,
the use of resilient couplings between the motor and the processing machine, due to elastic shafts and
gears, is often unavoidable. In this paper, speed control for torsional vibration suppression in a 2-mass
motor drive system is studied. The speed control method which uses a state and disturbance observer
with a PI controller has a better control response than a typical PI speed controller.
| ||
![]() | Observation strategy in a mean control structure for parallel...
By R. Peña-Eguiluz; M. Pietrzak-David; B. De Fornel | |
Abstract: This paper presents a comparison between two rotor flux observer strategies used in a mean drive
control structure for parallel connected dual induction motors, and compares the two flux observer
behaviour versus electrical and mechanical parameter variations. The studied system is an electrical
railway traction drive system. It has been modelled using the solver SABER.
| ||