IEEE Trans Ind Elect Final
IEEE Trans Ind Elect Final
IEEE Trans Ind Elect Final
NOMENCLATURE
U
V
U
V
K
ss
sk
s(t)
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
se
I. INTRODUCTION
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
magnet synchronous motor for the purpose of fault diagnosis
and in [22] to calculate additional losses due to rotor
eccentricity in an induction motor with PWM-voltage supply.
However, the calculation of forces in the FEM is sensitive to
discretization errors. In that regard the Maxwell stress tensor
approach is more sensitive than the virtual work method. In
[23] it has been shown that the value of torque calculated by
the virtual work method is very consistent irrespective of the
mesh refinement and changes marginally with a change in the
number of layers in the air-gap mesh. This can be important
for calculation of UMP because rotor displacement is usually
expressed in micrometers, which can be significantly smaller
than the size of the finite-element mesh in the air gap.
In [24] the dynamic eccentricity in a mechanical run prior
to excitation of the generator is modeled by observing the
whirling motion at a single frequency, which assumes that
shaft orbit has a circular shape. In reality, the shaft orbit
consists of multiple harmonic components which all together
affect the UMP in the machine. In linear models each
harmonic component can be treated separately and the total
UPM is equal to the sum of UMPs for individual harmonic
components. However, this approach is not valid under
nonlinear conditions in which case permeance networks or the
FEM are the most suitable approaches.
The contribution of this paper is a model of the measured
shaft orbit of a salient-pole synchronous generator and the
associated vibration displacements in two directions,
perpendicular to one another, with all their significant
harmonic components which have been used to calculate the
unbalanced magnetic pull by means of the FEM coupled with
circuit equations for the stator winding and the rotor damper
winding. For calculation of forces the virtual work method has
been applied. The no-load operation and the loaded condition
have both been included in the study and the damping effects
of the stator winding parallel paths and the rotor damper
winding have been taken into account.
II. THEORETICAL BACKGROUND
Shaft vibrations are defined as rapid motions of the shaft
which are related to its deflection in rotating motion. The
deflection of the shaft is a deformation of its elastic line which
connects all geometric centers of the shaft cross sections along
the axis of the bearings. This axis is an imaginary line which
connects geometric centers of the bearing bores and is aligned
with the z axis of the absolute reference frame. Since the shaft
bends under the action of static radial forces, its centers of
rotation form a static deflection line. In rotation the kinetic
displacement sk is added to the static displacement ss. (Fig. 1)
The motion of a shaft cross section is composed of a motion
of the shaft center C around the point S of the static deflection
line and also of a rotation of the cross section around the shaft
center C. This motion of the shaft is called precession. The
precession is defined by the motion of the shafts center of
gravity which can be given in the complex plane by equation
z Ue
j t U
Ve
j t V
ce jt
(1)
j t U
Ve
j t V
(2)
j t U
Ve
j t V
an e
j n t n
(3)
n 1
where an, n and n are the amplitude, the frequency and the
phase shift of the nth harmonic component and k is the
maximum number of higher order harmonic components.
Note that the higher order harmonic components do not affect
the time needed to go over one closed loop of the shaft orbit.
They only affect the shape of the orbit.
The size, shape and location of the shaft orbit vary with the
speed of rotation and load. In general, the kinetic shaft
displacements reach maxima at the critical speed. The most
common type of motion in synchronous machines is a
synchronous forward precession. The backward precession
can also occur with anisotropic bearings [25]. In the forward
precession the deflection line retains the sagged shape which
it has in the nonresonant motion, while in the backward
precession the deflection line alternates, i.e. the shaft flexes
twice for every turn around its center. This alternate flexion
requires high energy which makes the backward precession
less likely to appear.
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
III.
MEASURED ECCENTRICITY
5000 kVA
0.8
10500 V
10%
50 Hz
1500 rpm
1800 rpm
304 kgm2
800 mm
5.5 mm
8.8 mm
14.2 mm
820 mm
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
moves. In each time step the air-gap mesh along the boundary
between the two layers is modified so that nodes along the
boundary are mutual for both layers. Simultaneous rotary and
linear motions of the rotor are thus allowed. The stator
winding coils are objects in the FE model and their mutual
connections are modeled as an electric circuit which is solved
simultaneously with the rest of the model. The induced
voltages due to both rotary and linear motions are calculated
by FEM, and external current or voltage sources, resistors,
capacitors, inductors and other elements can be added to that
circuit as well. The end-winding leakage inductance has also
been added to the model. It has been calculated analytically
using a 3-D method based on closed form solution of
Neumann integrals [26].
Fig. 3. Recorded shaft orbit in mechanical run of the cold unit at 1500 rpm.
Field and armature winding currents are equal to zero.
K ss U x cos t U x jU y sin t U y
k
n2
(4)
(a)
where ss , Ux, Uy, Ux, Uy, axn, ayn, axn and ayn are given in
TABLE II. Note that Ux, Uy, Ux and Uy refer to the
fundamental component of the shaft orbit. Only the first 10
harmonic components of the Fourier series are given, i.e.
k = 10. According to Fig. 5, the first 10 components are
sufficient for good reconstruction of the shaft orbit.
The eccentric motion for one revolution of the rotor is
modeled using Infolytica MagNet 7.1.1 software by
simultaneously combining the rotation of the rotor about its
axis and the motion of the rotor axis with respect to the stator
center according to Fig. 5a for the no-load condition and
according to Fig. 5b for the loaded generator. The time needed
for the rotor axis to go through one full eccentric orbit is the
same as the time it takes the rotor to make one full revolution
about its axis at the speed of 1500 rpm. The time-stepping
method is used to model the motion of the rotor. The air-gap
is divided in two layers along its centerline. The layer along
the stator bore is stationary, while the other layer is attached to
the rotor and slides along the stationary one as the rotor
(b)
Fig. 4. Recorded shaft orbit in (a) no-load condition (1500 rpm, 10500 V),
(b) loaded condition (P = 1.84 MW, Q = 0.18 MVAr, 1500 rpm, 10500 V)
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
TABLE II.
HARMONIC CONTENT OF THE AVERAGE SHAFT ORBITS
n
2
3
4
5
6
7
8
9
10
225 rad/s
ss
-12.57+j78.26 m
Ux, m
22.415
axn, m
2.483
0.707
0.935
0.263
0.293
0.241
0.108
0.233
0.180
Uy, m
24.242
ayn, m
1.674
1.171
0.883
0.690
0.243
0.112
0.048
0.276
0.139
Ux, rad
Uy, rad
-0.763
-2.059
axn, rad
ayn,rad
2.708
3.087
-2.900
-2.644
-0.012
-3.042
0.286
0.878
-0.398
-0.825
1.420
0.302
-0.730
-2.109
0.288
-2.275
-0.332
0.899
120
Shaft displacement in y direction, m
NO-LOAD CONDITION
225 rad/s
9.49+j89.91 m
Ux, m
52.028
axn, m
2.735
1.683
0.761
1.232
0.447
0.220
0.079
0.209
0.192
Uy, m
43.073
ayn, m
1.515
0.493
0.505
1.813
0.765
0.257
0.145
0.294
0.197
Ux, rad
-1.860
axn, rad
-1.189
0.408
0.742
-0.032
-0.311
2.326
-2.394
-2.152
-1.478
100
90
80
70
60
-40 -30
-20
-10
0
10
Shaft displacement in x direction, m
20
(a)
Uy, rad
-0.240
160
ayn,rad
1.621
-1.102
0.146
-1.792
-1.098
-0.473
-0.504
0.029
-0.690
n
2
3
4
5
6
7
8
9
10
110
50
-50
LOADED CONDITION
ss
150
140
130
120
110
100
90
80
70
60
50
40
-50 -40 -30 -20 -10 0 10 20 30 40 50 60 70
Shaft displacement in x direction, m
(b)
Fig. 5. Measured (averaged) and reconstructed shaft orbit for (a) no-load and
(b) loaded condition
V. SIMULATION RESULTS
The FE simulations used for calculation of the UMP
resulting from the eccentric motion of the rotor according to
the recorded shaft orbits for no-load and loaded conditions
have been carried out for the following three cases:
1) Force calculation without stator winding parallel paths
and without rotor damper winding,
2) Force calculation with stator winding parallel paths and
without rotor damper winding,
3) Force calculation with stator winding parallel paths and
with rotor damper winding.
A. Case 1
This is the simplest case in which the damping effects of the
stator winding parallel paths and the rotor damper winding
have not been included. Therefore, the resulting UMP will
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
originate entirely from instantaneous displacements of the
rotor shaft center with respect to the center of the stator bore.
The results of simulations for both operating conditions are
unbalanced electromagnetic forces given as a function of time,
which are shown in Fig. 6. The maximum radial force for noload condition occurs at t = 12.4 ms and equals 2359 N. This
time instant corresponds to the maximum average shaft
displacement of 100.2 m. In addition, due to static
displacement of the shaft, the constant average force
F = 1750 N will appear.
Similarly, for the loaded generator the maximum radial
force occurs at t = 37.4 ms and equals 2518 N. This time
instant corresponds to the maximum average shaft
displacement of 133.8 m. The constant average force in the
case of the loaded generator equals F = 1612 N.
The maximum instantaneous force depends on the
maximum instantaneous displacement. On the other hand, the
size of the orbit reflects on the peak-to-peak variation of the
force. The peak-to-peak value of the total radial force in noload condition is Fp-p = 1294 N, while for the loaded generator
Fp-p = 1742 N.
Force, N
5000
3000
FEM: Fe=
FEM: with saturation
Analytical
Ephase
kd1
0.01
0.02
Time, s
0.03
0.04
7000
Force, N
6000
B 1
5000
Dl
p
(6)
sin q 2
q sin 2
sin 6 100 2
6sin 100 2
0.9561
(7)
(8)
Ephase 2 p
2 fwf d1 f p1 Dl
5747 2 2
2 50 96 0.9561 0.9397 0.8 0.82 (9)
0.9144 T
4000
3000
2000
kc
1000
0
0
2 fwkd1kp1 B 1
where q is the number of slots per pole and phase and y/p is
the coil pitch to pole pitch ratio. The air-gap flux density is
then
9000
8000
(a)
FEM: Fe=
(5)
2000
1000
0
Dr l s t 2
B 1
4 0 k c 0
6000
4000
0.01
0.02
Time, s
0.03
0.04
(b)
Fig. 6. Waveforms of the total unbalanced magnetic pull in (a) no-load and
(b) loaded condition calculated analytically and numerically using FEM
(simulation WITHOUT parallel paths and WITHOUT damper winding)
s
2
2d o
d o 0 d o
s
ln 1
atan
2 0 d o 2 0
1.1597
(10)
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
will be directly proportional to the shape of the shaft orbit,
which is determined by s(t). The calculation of force in (5)
assumes that iron is infinitely permeable, so comparison with
FE simulation is possible if the same assumption is made in
the FE model. This is accomplished by setting the relative
permeability of the stator and rotor core to 109 to simulate an
infinitely permeable material. However, in order to obtain the
same fundamental component of the induced phase voltage
and thus the same fundamental component of the air-gap flux
density as in the case when the actual B-H curves of the stator
and rotor core are taken into account, the field current had to
be reduced. Otherwise, with the same field current as in the
nonlinear simulation, due to increased overall permeance of
the magnetic circuit in the linear model, higher values of the
air-gap flux density and induced voltage would have been
obtained.
The waveforms of the total unbalanced force on the rotor in
no-load condition calculated analytically using (5) and
numerically using FEM with saturation and with infinitely
permeable iron are shown in Fig. 6a. The small oscillations
that are superimposed on the basic waveforms are due to
slotting effect and due to discretization errors which occur
because the shifts of the rotor are small relative to the size of
the FE mesh in the air gap.
The waveforms indicate that fundamental frequency is
25 Hz. The fundamental frequency of the force in the case of
whirling motion is equal to the frequency of precession [6].
Since in our case is equal to the angular speed of rotation of
the shaft around its center, in a four pole machine at 1500 rpm
this corresponds to the frequency of 25 Hz.
There is a very good agreement between analytically and
numerically calculated waveforms with infinitely permeable
iron. This result clearly indicates that the shape of the shaft
orbit is the dominant factor which determines the waveform of
the UMP.
The strong influence of saturation in the stator and rotor
core on the magnitude of force is also apparent. The peak
value of force is reduced by a factor of 2.36 when saturation is
taken into account. In a sense by adjusting the field current
and the value of induced voltage to be the same in nonlinear
and linear models, the saturation has been taken into account
to some extent. However, the equality of induced voltages
does not necessarily lead to the equality of calculated forces
with and without the presence of saturation in the FE model.
The correlation with the shaft orbit is also visible in Fig. 7a
which compares harmonic content of the shaft orbit and of the
total unbalanced force on the rotor (without the average
component). The magnitudes are normalized with respect to
their fundamental components (F1 and s1).
Similar analysis can be done for the loaded condition as
well. However, in this case the fundamental component of the
air-gap flux density depends on the resulting actions of the
field winding and the armature winding MMFs. In the FE
model an iterative procedure is used to determine the field
current and the position of the armature current vector which
yield the required active and reactive power output
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
origin of this particular harmonic component can be
determined by simulating the eccentric motion using one
particular harmonic component of the shaft orbit at a time.
Unfortunately, this approach can only be used in a linear
model, since it employs the principle of superposition which
cannot be used in a nonlinear model. Nevertheless, the linear
approach can also give a valid insight into the origin of the
100 Hz component. For this purpose a linear finite-element
model has been used for the cases of no-load operation with
and without stator winding parallel paths. In both cases the
rotor damper winding has not been included. For each case
three simulations have been performed. The static deflection
ss , the first and the second harmonic components of the shaft
orbit defined in TABLE II for no-load operation have been
simulated respectively. From each simulation the x and y
components of the UMP have been calculated (Fig. 10 and
Fig. 11). In addition, for the case when stator winding parallel
paths are included, the induced stator currents which circulate
within parallel branches of the phase windings have been
calculated (Fig. 12). The total radial force resulting from all
three simulations is equal to
2
0.6
0.4
0.2
0
25
50
(a)
1
Force: FEM with saturation (F1=704.1 N)
Force: FEM with Fe= (F1=2367.8 N)
0.8
0.6
0.4
0.2
0
(11)
where Fx0, Fx1, Fx2, Fy0, Fy1 and Fy2 are the x and y components
of the UMP originating from the static deflection, the first and
the second harmonic components of the shaft orbit
respectively. The same total radial force can be calculated
from a single FE simulation in which the static deflection and
the first two harmonic components of the shaft orbit are used
simultaneously to define the eccentric motion of the rotor. Of
course, the same principle can be expanded to simulate
independently an arbitrary number of shaft orbit harmonic
components. However, in that case the simulations would
require very dense FE mesh and small time steps in order to
correctly account for individual higher harmonic components
of the shaft orbit due to their small magnitude and high
frequency.
Note that the total radial force in Fig. 10b is comparable to
the force in Fig. 6a. It is basically the same waveform, except
that in Fig. 10b the static deflection and two harmonic
components of the shaft orbit are used, while in Fig. 6a all 10
harmonic components are used to simulate the eccentric
motion of the rotor. Similarly, the total radial force in Fig. 11b
can be compared with the force in Fig. 8a calculated for Case
2. The difference is in the magnitude of the force since the
result in Fig. 11b has been obtained using linear model with
static deflection and two harmonic components of the shaft
orbit, while in Fig. 8a all 10 harmonic components are used in
a nonlinear model.
0.8
25
50
(b)
Fig. 7. Comparison of normalized harmonic content of the shaft orbit and of
the total unbalanced magnetic pull in (a) no-load and (b) loaded condition
calculated numerically and analytically (simulation WITHOUT parallel paths
and WITHOUT damper winding)
3000
2500
2000
Force, N
Fx 0 Fx1 Fx 2 Fy 0 Fy1 Fy 2
1500
1000
500
0
0
0.01
0.02
Time, s
0.03
0.04
(a)
3000
2500
Force, N
Fradial
2000
1500
1000
500
0
0
0.01
0.02
Time, s
0.03
0.04
(b)
Fig. 8. Waveforms of the total unbalanced magnetic pull in (a) no-load and
(b) loaded condition for all three cases
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
5500
600
WITHOUT stator parallel paths and WITHOUT rotor damper winding
WITH stator parallel paths and WITHOUT rotor damper winding
WITH stator parallel paths and WITH rotor damper winding
5000
Total radial force, N
500
400
Force, N
10
300
200
100
4500
4000
3500
3000
25
50
2500
0
(a)
0.01
0.02
Time, s
900
700
600
Force, N
0.04
(b)
800
0.03
500
400
4000
300
3000
200
100
2000
25
50
Force, N
Fy0
Force, N
3000
Fy1
Fy2
Fx2
1000
0.01
0.015
Fx2
0.02 0.025
Time, s
0.03
0.035
0.04
(a)
5000
4000
3000
2000
0
0
-1000
Fx0
0.005
0.005
1000
-2000
0
-2000
0
4000
Fx1
Fx1
Fx0
Loaded
Average
Peak force
force
(% attenuation)
1736 N
2518 N (0 %)
1133 N
2087 N (17.1 %)
654 N
1233 N (51.0 %)
5000
2000
Fy2
1000
-1000
TABLE III.
COMPARISON OF PEAK RADIAL FORCE FOR ALL THREE CASES
Case 1
Case 2
Case 3
Fy0
Fig. 9. Comparison of harmonic contents of the UMP in (a) no-load and (b)
loaded condition
No-load
Average
Peak force
force
(% attenuation)
1790 N
2359 N (0 %)
1220 N
2022 N (14.3 %)
645 N
1005 N (57.4 %)
Fy1
0.01
0.02
Time, s
0.03
0.04
(b)
0.01
0.015
0.02 0.025
Time, s
(a)
0.03
0.035
0.04
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
Ff Ff 1 cos ps e t
Stator current, A
0
-0.25
0.01
0.02
Time, s
0.03
0.04
(a)
0.4
iU1
0.3
iV1
iW1
0.2
0.1
0
-0.1
-0.2
-0.3
-0.4
0
0.01
0.02
Time, s
0.03
0.04
0.03
0.04
(b)
0.04
iU2
0.03
iV2
iW2
0.02
0.01
0
-0.01
-0.02
(13)
(14)
0.25
-1
0
-0.03
0
(15)
0.01
0.02
Time, s
(c)
1.5
iU0+iU1+iU2
1.25
iV0+iV1+iV2
1
Stator current, A
cos s e t se cos s se0
p
0,2 p ,4 p
iW0
-0.5
iV0
-0.75
(12)
iU0
0.5
Stator current, A
1
0.75
Stator current, A
11
iW0+iW1+iW2
0.75
0.5
0.25
0
-0.25
-0.5
-0.75
-1
0
0.01
0.02
Time, s
0.03
0.04
(d)
Fig. 12. Induced stator winding currents in no-load condition calculated using
linear FEM which circulate within the branches connected in parallel
(WITHOUT rotor damper winding taken into account) resulting from the
following harmonic components of the shaft orbit: (a) static deflection
(constant term), (b) 1st harmonic, (c) 2nd harmonic, (d) static, 1st and 2nd
together
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
12
(18)
where
(a)
(b)
Fig. 13. Principle cross-section of the four-pole generator with double layer
short pitched stator winding arrangement, referent current directions and
positions of the phase winding axes for the cases of (a) normal operation, and
(b) operation with open-circuited stator winding and induced circulating
currents in the parallel branches
f1
160
r Br2 B2 2 0 , Br B 0
(16)
Fx
Fy
Br2 D
l cos s d s
20 2
Br2 D
l sin s d s
2 0 2
(17)
160
se2 Ff 1 Fai1 Dl .
(19)
Fig. 14. Flux lines of the armature winding field due to induced circulating
currents in the winding parallel paths with field winding current set to zero
without rotor damper winding and with only static deflection of the shaft orbit
taken into account
f2
CONCLUSION
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
correlations correctly. This principle can be applied to all
three cases studied in the paper.
The results shown in the paper indicate that for this
generator the stator winding parallel paths alone attenuate the
UMP by 14.3 % in no-load and 17.1 % in loaded condition.
The rotor damper winding and the stator winding parallel
paths together yield the total attenuation of 57.4 % in no-load
and 51.0 % in loaded condition.
[16]
[17]
[18]
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
13
Fault Diagnosis, IEEE Trans. Ind. Electronics, Vol. 55, No. 2, pp. 610619, February 2008
A. Bellini, F. Filippetti, C. Tassoni and G.A. Capolino, Advances in
Diagnostic Techniques for Induction Machines, IEEE Trans. Ind.
Electronics, Vol.55, No.12, pp. 4109-4126, December 2008
J. T. Li, Z. J. Liu and L. H. A. Nay, Effect of Radial Magnetic Forces in
Permanent Magnet Motors With Rotor Eccentricity, IEEE Trans. Mag,
Vol. 43, No. 6, pp. 2525-2527, June 2007
A., Mahyob, M.Y.O. Elmoctar, P. Reghem, G. Barakat, Induction
Machine Modelling Using Permeance Network Method For Dynamic
Simulation of Air-Gap Eccentricity, 12th European Conference on
Power Electronics and Applications (EPE 2007), pp. 1-9, Aalborg,
Denmark, 2-5 September 2007
R. Perers, U. Lundin and M. Leijon, Saturation Effects on Unbalanced
Magnetic Pull in a Hydroelectric Generator With an Eccentric Rotor,
IEEE Trans. Mag., Vol. 43, No. 10, pp. 3884-3890, October 2007
Lin Wang, R.W. Cheung, Zhiyun Ma, Jiangjun Ruan and Ying Peng,
Finite-Element Analysis of Unbalanced Magnetic Pull in a Large
Hydro-Generator Under Practical Operations, IEEE Trans. Mag, Vol.
44, No. 6, pp. 1558-1561, June 2008
B.M. Ebrahimi, J. Faiz, and M.J. Roshtkhari, Static-, Dynamic-, and
Mixed-Eccentricity Fault Diagnoses in Permanent-Magnet Synchronous
Motors, IEEE Trans. Ind. Electronics, Vol. 56, No. 11, pp. 4727-4739,
November 2009
A. Belahcen, and A. Arkkio, Computation of additional losses due to
rotor eccentricity in electrical machines, IET Electr. Power Appl.,
Vol. 4, Iss. 4, pp. 259266, 2010
S. Salon, S. Bhatia and D. Burow, Some Aspects of Torque
Calculations in Electrical Machines, IEEE Trans. Mag, Vol. 33, No. 2,
pp. 2018-2021, March 1997
D. Guo, F. Chu, D. Chen, The Unbalanced Magnetic Pull And Its
Effects On Vibration In A Three-Phase Generator With Eccentric
Rotor, Journal of Sound and Vibration 254 (2), pp. 297312, 2002
Agnieszka Muszyska, Rotordynamics, Taylor & Francis, 2005
D. Ban, D. arko, I. Mandi, Turbogenerator End-Winding Leakage
Inductance Calculation Using a 3-D Analytical Approach Based on the
Solution of Neumann Integrals, IEEE Trans. Mag, Vol. 20, No. 1,
pp. 98-105, March 2005
T. Schneider and A. Binder, Design and Evaluation of a 60 000 rpm
Permanent Magnet Bearingless High Speed Motor, 7th International
Conference on Power Electronics and Drive Systems (PEDS '07), pp. 18, Bangkok, Thailand, November 27-30, 2007
G. Joksimovic, C. Bruzzese, and E. Santini, Static Eccentricity
Detection in Synchronous Generators by Field Current and Stator
Voltage Signature Analysis Part I: Theory, ICEM 2010 19th
International Conference on Electrical Machines, 6 pages, Rome, Italy,
September 6-8, 2010
G. Joksimovic, C. Bruzzese, and E. Santini, Static Eccentricity
Detection in Synchronous Generators by Field Current and Stator
Voltage Signature Analysis Part II: Measurements, ICEM 2010 19th
International Conference on Electrical Machines, 5 pages, Rome, Italy,
September 6-8, 2010
L. Frosini and P. Pennacchi, The effect of rotor eccentricity on the
radial and tangential electromagnetic stresses in synchronous machines,
Proc. of IECON 06 - 32nd Annual Conf. of the IEEE Industrial
Electronics Society, pp. 1287-1292, Paris, France, November 7-10, 2006
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) <
Drago Ban (M01) was born in 1939 in VrpoljeSibenik, Croatia. He received the B.Sc., M.Sc.
and Ph.D. degrees in electrical engineering from
the Faculty of Electrical Engineering, University
of Zagreb, Zagreb, Croatia in 1965, 1975 and
1987 respectively.
From 1968 till 1989 he worked as a
development and project engineer in KONARElectrical Industry, Zagreb, Croatia. He is
currently a full Professor at the Department of
Electrical Machines Drives and Automation,
Faculty of Electrical Engineering and
Computing, University of Zagreb, Croatia. His
scientific and professional activities have been related to the field of electrical
machinery, electrical drives and technical diagnostics.
Professor Ban has a long experience in organization of international
conferences, seminars and lectures. He was the chairman of several
international conferences on electrical drives and power electronics: EDPE
1996, EDPE 98 and EDPE 2000. In 2002 he was the general chairman of the
10th International Power Electronics and Motion Control Conference, EPEPEMC 2002, 9-11. Sept. Dubrovnik, Croatia. He is a member of Editorial
Board of the Journal of Electrical Engineering, several international
professional associations, and a member of Croatian Academy of Engineering.
14