td> Below is a cache of http://www.imarest.org/estore/trans/v107/part2/ijauwl2w.pdf. It's a snapshot of the page taken as our search engine crawled the Web.
The web site itself may have changed. You can check the current page or check for previous versions at the Internet Archive. Yahoo! is not affiliated with the authors of this page or responsible for its content.
Cycloconverter configurations and controllers for marine propulsion applications
83
Trans IMarE, Vol 107, Part 2, pp 8399
Cycloconverter configurations
and controllers for marine propulsion
applications
L Ran,
BSc, PhD
, K S Smith,
BSc(Eng), PhD, AMIEE
and R Yacamini,
MSc, FIMarE, CEng
University of Aberdeen
Cycloconverter propulsion systems are becoming more common for cruise liners and
icebreakers and have been proposed for warship propulsion applications. An earlier paper
described the cycloconverter operating characteristics and explained how time domain
simulation can be used to analyse the steady state characteristics of a cycloconverter
supplying an induction motor load. The present paper will extend the earlier work to include
the transient performance of the cycloconverter drive for both induction and synchronous
motor drives. The principles of vector control, which is utilised in modern drive design, will
be reviewed and their practical application to high torque, low speed cycloconverter drives
will be discussed. The implementation of vector control into the existing computer model
will be illustrated using simulation results for cycloconverter circuits of differing topologies.
The relative merits of these topologies from an operational point of view will also be
analysed.
Authors biographies
Li Ran was born in WuSheng, Sichuan Province, China in 1963. He obtained his BSc and PhD degrees from Chongqing University in July
1984 and March 1989, respectively. Following the award of his PhD Dr Ran worked as a Lecturer in Electrical Engineering at Chongqing
University before joining the research group at Aberdeen in December 1992.
K S Smith received his BSc(Eng) and PhD degrees from the University of Aberdeen in 1988 and 1992, respectively. Since October 1991 he
has been a Lecturer in Electrical Engineering in the Department of Engineering, University of Aberdeen, with responsibility for the
teaching of heavy electrical power engineering. His main fields of interest are the interaction between electrical machines and power
electronic converters on closely coupled ac systems such as offshore oil and gas installations.
R Yacamini is currently a Reader in the Department of Engineering at the University of Aberdeen. His previous experience includes 10
years as a design engineer with English Electric and GEC in the rectifier and high voltage dc transmission fields. During this time he made
extensive use of physical simulations for controller design and system studies. This industrial period was followed by five years as a
lecturer at UMIST where he carried out research, using frequency domain computer programs, into HVDC and reactive compensators.
He took up his post in Aberdeen in 1982 and has been involved in consultancy work for the oil industry for most of this period. The main
thrust in his research has been to develop time domain CAD analysis packages for power electronic applications.
INTRODUCTION
The utilisation of electric drive technology for marine pro-
pulsion applications continues to generate considerable in-
terest from both a research and development viewpoint, as
well as for practical applications. Recent successful applica-
tions include the much publicised QE2 refit
1
and the British
Antarctic Research vessel the RRS James Clark Ross.
2
Ongo-
ing work at the Engineering Department of Aberdeen Uni-
versity is concentrating on the computer analysis, using
time domain simulation techniques, of the cycloconverter
propulsion systems proposed for frigate propulsion sys-
tems. An earlier paper
3
described the steady state analysis of
cycloconverter drives using the Saber simulator. During the
discussion which followed the presentation of this paper the
authors described how they hoped eventually to be able to
simulate the performance of a complete propulsion system.
This paper reports on the progress made to date towards
achieving this goal. The response of the propulsion drive
system under various transient operating conditions is in-
vestigated. For this purpose, it is necessary to implement
real time control in the computer simulation. This will also
allow the analysis of the drive performance during abnor-
mal or fault conditions. Another paper, soon to be pub-
lished, will describe the groups work on the vibration
characteristic of propulsion motors.
ELECTRICAL PROPULSION SYSTEMS
There is currently some debate concerning the likely ar-
rangement of a frigate propulsion system. It is clear, how-
ever, that such a system will contain all of the elements
shown in Fig 1. The ships power will probably be provided
by a combination of diesel generator and gas turbine genera-
tor sets, the number, and rating of these sets, chosen so as to
meet the power requirements of a wide range of operating
conditions. These will supply a high voltage switchboard
which can be divided into two sections, with front and rear
bars, to allow for maintenance and provide maximum flex-
ibility. The ships dedicated power supplies will be derived
Paper read on 13.12.1994 84
L Ran, K S Smith & R Yacamini
Fig 1
Possible electrical propulsion system
from this switchboard through transformers and motor-
generator sets, which will limit the level of distortion on
these critical systems.
The cycloconverter system will be supplied through step-
down transformers and will drive the propulsion motor,
which may be either an induction or synchronous machine.
This will be directly coupled to the propulsion shaft. This
system does not feature a gearbox (unlike the CODLAG
system on the present Type 23 frigate). A first step in develop-
ing a transient model of this system is clearly to develop a
computer model of the cycloconverter controller and its inter-
face with the propulsion motor. The operating principles of the
vector controller and its practical applications to cycloconverter
drives form the basis of the present paper. In order to appre-
ciate the operation of the vector controller, an understanding
of fundamental electrical machine characteristics is required.
CYCLOCONVERTER AND MOTOR CONNECTIONS
Many alternative arrangements of cycloconverter circuits,
having varying degrees of complexity are feasible.
4
Almost
all practical circuits utilise the six-pulse, symmetrical
cycloconverter as a basic building block. Individual six-
pulse cycloconverters can be combined to give the 12-pulse
circuits or pseudo 12-pulse circuits. The choice of converter
pulse number and circuit connection is very application
dependent. A number of common cycloconverter circuits
are shown in Figs 2a2c. Recently, consideration has been
given to eliminating the transformers for cost and space
reasons. However, the intercoupling between the commuta-
tion processes of different cycloconverters remains for such
arrangements. The models described in this paper can be
readily adapted to analyse this type of system if necessary.
The basic six-pulse circuit of Fig 2a is the most common
cycloconverter configuration. The output voltage of the
cycloconverter contains components which are of zero phase
sequence, however currents at these frequencies are pre-
vented from flowing due to the three wire connection used
to supply the motor. The true 12-pulse configuration is
shown in Fig 2b. The transformers at the inputs to the
thyristor bridges are arranged so that there is a 30 deg
(electrical) phase shift between the ac inputs to the two six-
pulse bridge groups. The double wound ac motor configu-
ration is also frequently considered for the purpose of either
higher power level or quieter operation during both normal
and fault conditions. Such an arrangement, referred to as a
pseudo 12-pulse one, is shown in Fig 2c. The two sets of three
phase stator windings are displaced by 30 deg (electrical) in
space. The voltages applied to these two sets of windings are
also shifted by 30 deg in time.
In addition to these, other configurations will also be
possible which in practice may suit various design and
operational purposes. For instance, when the input trans-
former is designed out of the configuration, it will be neces-
sary to isolate the supply to each motor winding at the
output side of the cycloconverter. This will result in a six
wire connection in the case of a three phase motor. For the
three wire connection, a fourth neutral line may also be
included in the configuration to permit an independent
supply between the motor phases. This is important for the
fault tolerant operation of the drive, which requires that
when one of the stator phases is lost the remaining phases
should still produce a forward rotating mmf of constant
amplitude.
5
This can be achieved by regulating the relative 85
Trans IMarE, Vol 107, Part 2, pp 8399
Fig 2a