NOMENCLATURE

E
1
= Air gap voltage per phase
I
1
= Stator current per phase
I
1real
= Real part of I
1
I
1imag
= Imaginary part of I
1
I
2
= Rotor current per phase
I
2real
= Real part of I
2
I
2imag
= Imaginary part of I
2
I
m
= Magnetizing current per phase
pf = power factor
P
fw
= Friction and windage losses
P
g
= Air gap power
P
in
= Input Power
P
m
= Mechanical Power
P
out
= Output power
P
r
= Rotational losses (friction, windage and core
losses)
P
rcu
= Rotor copper losses
P
score
= Stator core losses
P
scu
= stator copper losses
R
1
= Stator resistance per phase
R
2
= Rotor resistance per phase referred to stator
s = Slip
V
1
= Terminal voltage per phase
X
1
= Stator leakage reactance per phase
X
2
= Rotor leakage reactance per phase referred to
stator
X
m
= Magnetizing reactance per phase
I. INTRODUCTION
Recently considerable attention is being focused on
environmentally clean and safe renewable energy sources
like wind, solar, hydro etc. The fast depletion of fossil fuels
and our over dependence on them is believed to soon bring
the wheels of our civilization to screeching halt. Henceforth
at this juncture the need is being felt for relevant
technological efforts in order to tap the vast potential energy
of hydro, wind etc. available in isolated locations. The use
of an induction generator [1] in general and squirrel cage
induction generator in particular for such applications in
remote, unattended and maintenance free sites is earning
more favors over that of the synchronous alternators. The
reason for the widespread popularity of induction generator
is owing to its capability to generate the power from
variable speed as well as constant speed prime movers, low
unit cost, reduced maintenance, rugged and brushless rotors,
absence of a separate d.c. source for excitation, absence of
moving contacts, inherent overload protection, improved
performance due to low transient impedance, natural
protection against short circuit etc.
Two modes of operation can be employed for an
induction generator. One is through self-excitation and other
is through external-excitation. In first mode, the induction
generator takes its excitation from VAR generating units,
generally realized in the form of capacitor banks. With
suitable capacitors connected across the terminals and with
rotor driven in either direction by a prime mover, voltage
builds up across the terminals of the generator due to self
excitation phenomenon leaving the generator operating under
magnetic saturation at some stable point. Such generator is
known as self-excited induction generator (SEIG) [2-6]. In
second mode it draws the reactive power for its operation
from the grid to which it is connected. Such generator is
known as externally excited generator or grid connected
induction generator (GCIG) [7-11].
In GCIG the grid regulates voltage and frequency
automatically. Thus GCIG has no voltage regulation problem
and henceforth eliminates the need of voltage regulators as
in case of SEIG. The synchronous machines already
connected to grid and running in parallel with an induction
generator provides the reactive power, while generator output
is decided by its operating slip. Therefore a bottleneck in
the application of GCIG remains the requirement of inductive
Estimation of saturation in grid connected induction generator
Shelly Vadhera and K.S. Sandhu
Deptt. of Electrical Engineering, National Institute of Technology, Kurukshetra, (HR) INDIA
(Received 15 Nov., 2009, Accepted 10 Jan., 2010)
ABSTRACT : Prediction of the saturation level and henceforth finding the value of magnetizing reactance X
m

is the first need for accurate estimation of steady state performance of grid connected induction generator
(GCIG). Study reveals that most of the research persons adopt the unsaturated value for magnetizing reactance
for the steady state analysis of such machines. Whereas in this paper an iterative technique is proposed to predict
the saturated values of magnetizing reactance with change in operating conditions. The comparison between
experimental and simulated results proves the validity of technique proposed. Further efforts have been made to
predict the operating limits of machine using proposed modeling. Analysis has been extended to identify
effectiveness of the machine parameters to improve the operating performance of the generator. The results are
presented and inferences are drawn to suggest guidelines for real design problems in GCIG.
Keywords : Asynchronous Generator, Grid Connected Induction Generator, Iterative Technique, Steady State Analysis,
Wind Energy Conversion
International Journal on Emerging Technologies 1(1): 31-36(2010) ISSN : 0975-8364
e
t
32 Vadhera and Sandhu
VAR for both induction generator as well as the load, which
has to be fed by the generator.
The custom designed induction generator seems
unnecessary because an induction motor gives
corresponding performance as induction generator. The fact
that induction machines are readily available from several
manufacturers makes them very competitive for just in time
installation. However there is a scope to improve the
performance by proper designing of machine.
To incorporate design modifications, there is a need for
prior estimation of steady state performance of the machine.
Prediction of the saturation level and hence finding the value
of magnetizing reactance of induction generator is first step
in the steady state analysis of GCIG. A review of the available
literature reveals that although a lot of work has been
reported on analysis of GCIG using the philosophy of fixed
value of magnetizing reactance (X
m
) but no attempt seems
to have been made in using the saturated values of X
m
with
corresponding change in air gap voltage (E
1
). Computation
of the magnetic reactance is done using thevenin equivalent
circuit [7], while in [8] the value of X
m
is found by carrying
out variable voltage no load tests. [1], [9-11] uses the fixed
value of X
m
and [12] calculates the value of X
m
corresponding to induced electromotive force (EMF) by using
a software package of MathCAD.
In the present paper it has been shown that there is a
considerable variation in X
m
with load/slip and hence there
is need to compute saturated value of X
m
corresponding to
any operating load. Therefore, a new iterative technique has
been proposed to compute the corresponding values X
m
with change in operating conditions. Further the analysis is
extended to identify the effectiveness of various design
parameters in order to improve the performance
characteristics of a machine working as a generator.
II. STEADY STATE MODELING
R
1
j X
1 R
2
j X
2
V
1
I
1
I
2
R
2
1-s
s
I
m
E
1
j X
m
Fig.1(a). Equivalent circuit representation of an induction
machine.
Analysis of equivalent circuit representation of an
induction machine as given by Fig.1(a), with any value of
slip (negative for generator operation) results in to the
following mathematical expressions :
Z = R
1
+ jX
1
+
2
2
2
2
( )
( )
m
m
R
jX jX
s
R
j X X
s
_
+

,
_
+ +

,
...(1)
Unknown and saturated value of X
m
for generator
operation may be obtained using iterative technique, as
explained in section III.
Z = Z
real
+ jZ
imag
...(2)
Where
Z
real
= R
1

2 2
2
2 2
2
( )
m
m
R
X
s
R
X X
s
_
+ +

,
...(3)
Z
real
is negative for generator operation
Z
imag
= X
1
+
2
2
2 2
2
2 2
2
( )
( )
m m m
m
R
X X X X X
s
R
X X
s
_
+ +

,
_
+ +

,
...(4)
I
1
=
1
V
Z
...(5)
I
1
= I
1real
+ jI
1imag
...(6)
Where,
I
1real
=
( )
1 real
2 2
real imag
V Z
Z Z +
...(7)
I
1imag
=
( )
1 imag
2 2
real imag
V Z
Z Z +
...(8)
E
1
= V
1
I
1
(R
1
+ jX
1
) ...(9)
I
m
=
1
m
E
jX
...(10)
I
1
= I
2
I
m
...(11)
P
out
= 3V
1
I
1real
...(12)
This results in negative power for negative slip in case
of generator.
The input to rotor is
P
in
= P
out
+ 3I
1
2
R
1
+ 3I
2
2
R
2
+ P
r
...(13)
Equation (12) and (13) gives efficiency () of the
generator as;
=
out
in
P
P
...(14)
III. ITERATIVE TECHNIQUE
Iterative procedure for the computation of magnetizing
reactance X
m
in generating mode is summarized as
following :
Step 1. Assume
0
m
X corresponding to E
1
0
as 1.0 p.u.
from the relationship between E
1
and X
m
depicting the
magnetic characteristics of induction machine.
Step 2. Compute E
1
1
using (1) to (9)
Step 3. Find out the new value of magnetizing reactance
Xm
1
corresponding to air gap voltage computed in Step 2
Step 4. If |
1
m
X


0
m
X |
Then the value of
1
m
X may be used as the final
magnetizing reactance needed for further computation the
Vadhera and Sandhu 33
performance of the induction generator. Otherwise
0
m
X may
be replaced by the new value of
1
m
X and the procedure
may be repeated unless until the difference between
successive values of magnetizing reactance comes out as
desired.
IV. GENERATOR OPERATIVE LIMITS
It is found that for generator operation I
1real
in (7)
becomes negative in contrast to motor operation. This is
the indication for reversal of active power in generating
mode. Now real power flows from machine to grid system,
which is opposite in case of motoring mode. However as
indicated by I
1imag
in (8), direction of reactive power flow
remains same irrespective of operating mode. Such
observations leads to a new representation for the grid
connected induction generator as shown in Fig.1(b).
R
1
j X
1 R
2
j X
2
V
1
I
1imag
I
2imag
R
2
(1-s)
s
I
m
E
1
j X
m
I
1real
I
2real
Fig.1(b). Equivalent circuit representation of grid connected
induction generator.
Further for generator operation I
1real
is negative only if
Z
real
is negative quantity. This results in the generating
condition as given below;

2
2
2
2 2
2
( )
m
m
R X
R
s X X
s
1
_
+ + 1

,
1
]
> R
1
...(15)
Equation (15) with boundary conditions results in a
quadratic equation
as
2
+ bs + c = 0
Where, a = R
1
(X
2
+ X
m
)
2
, b = R
2
X
m
2
, c = R
1
R
2
2
Solution of quadratic equation gives the operating slip
of machine in generating zone as;
s =
2 2 4 2 2 2
2 2 1 2 2
2
1 2
4 ( )
2 ( )
m m m
m
R X R X R R X X
R X X
t +
+
...(16)
Out of the two solutions as obtained by (16), only one
feasible value can be selected as operating slip
corresponding to boundary conditions i.e., s
mg
. Therefore
s
mg
is the maximum value of slip up to which generator is
possible. However operating slip (s
mt
) corresponding to
maximum value of torque generated (ve in case of
generator) comes out to be;
s
mt
=
2
2 2
2
( )
e e
R
R X X + +
...(17)
Where,
R
e
=
[ ]
1 1 1
2 2
1 1
( )
( )
m m m
m
R X X X X X
R X X
+
+ +
X
e
=
2
1 1 1 2
2 2
1 1
( )
( )
m m
m
R X X X X X
R X X
+ +
+ +
Comparison of two expressions as given by (16) and
(17) yields the following deductions;
1. s
mt
s
mg
2. s
mt
> s
mg
3. s
mt
and s
mg
are dependent upon saturated values of
magnetizing reactance X
m
in addition to machine
parameters.
4. Operating range of the machine can be controlled
by proper handling of magnetic circuit i.e., X
m
and
machine parameters.
Power flow diagram for the induction generator has been
shown in Fig.2
-P
fw
-P
rcu
-(P
scu
+ P )
score
P
m
P
g
P
in
P
out
Stator Rotor Shaft
Fig.2. Power flow diagram of induction generator.
Fig.3 gives the flow chart for performance evaluation of
the generator and this has been programmed in MATLAB
Assume air gap voltage and compute
corresponding value of from given
relationship
E
X
1
K
K
m
Advance
iteration count
= + 1 K K
START
Read machine
parameters
Set initial value of slip
Set iteration count
= 0 K
Compute new value for E and find
now the corresponding value of X
1
K
K
+ 1
+ 1
m
Yes
No
Performance evaluation using generator
modeling
Set
= + 1 S S
END
Fig.3. Flow chart for performance evaluation.
34 Vadhera and Sandhu
V. RESULTS AND DISCUSSIONS
Proposed iterative technique is adopted to simulate the
results on Machine-1 [Appendix-I]
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.3 0.5 0.7 0.9
Unsaturated X
m
Saturated X
m
Experimental
I (pu)
1
P
(
p
u
)
o
u
t

Fig.4. Variation of output power with stator current.
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.3 0.5 0.7 0.9
I (pu)
1
Unsaturated X
m
Saturated X
m
Experimental
Fig.5. Variation of power factor with stator current.
Unsaturated X
m
Saturated X
m
Experimental
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 0.01 0.02 0.03
s
P
(
p
u
)
o
u
t

Fig.6. Variation of output power with slip.
0 0.01 0.02 0.03
s
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Unsaturated X
m
Saturated X
m
Experimental
p
f
Fig.7. Variation of power factor with slip.
The computed and experimental results for the variation
of output power and operating power factor with stator
current and slip are shown in Fig.4 to Fig.7. The simulated
results using the saturated value of Xm in contrast to
unsaturated value of Xm are far closer to experimental results.
This closeness of simulated results with experimental ones
establishes the validity of proposed iterative technique. This
paper lays an emphasis on computing saturated value of
Xm, accounting for saturation in magnetic circuits, which
was generally neglected by research persons earlier.
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0 0.2 0.6
s
1 0.4 0.8
X
m

(
p
u
)
Fig.8. Variation of magnetizing reactance with slip.
Fig.8 shows the simulated results for variation of
magnetizing reactance with operating slip. This variation in
Xm is due to the saturation effect.
4.0
3.0
2.0
1.0
0.0
s
0.4 0.3 0.2 0.1 0
Unsaturated X
m
Saturated X
m
T
o
r
q
u
e

(
p
u
)
Fig.9. Variation of torque with slip.
5.0
4.0
3.0
2.0
1.0
0.0
0 0.1 0.2 0.3 0.4
s
Unsaturated X
m
Saturated X
m
Q

(
p
u
)
Fig.10. Variation of reactive power with slip.
The variation of torque and reactive power with slip for
saturated and unsaturated values of Xm are shown in Fig.9
and Fig.10 respectively. It is observed that accounting of
saturation for analysis purpose affects the generating torque
and reactive power consumption of the machine.
Vadhera and Sandhu 35
s
0 0.01 0.02 0.03
K = 1
K = 0.75
K = 1.25
E
f
f
i
c
i
e
n
c
y
1.0
0.8
0.6
0.4
0.2
0.0
Fig.11. Variation of efficiency with slip, stator resistance = KR
1
.
0
Fig.12. Variation of efficiency with slip, rotor resistance = KR
2
.
0
Fig.13. Variation of efficiency with slip, stator reactance = KX
1
.
0
Fig.14. Variation of efficiency with slip, rotor reactance = KX
2
.
Fig.15. Variation of power factor with slip, stator resistance = KR
1
.
Fig.16. Variation of power factor with slip, rotor resistance = KR
2
.
Fig.11 to Fig.14 shows the effects of variations in
machine parameters on the operating efficiency of the
generator. It is observed;
1. Variations in stator and rotor resistances will affect
the operating efficiency of the generator. This effect
is more pronounced at low operating slips.
2. The stator or rotor reactance doesnt play much role
in the enhancement of efficiency of induction
generator.
3. Rotor resistance is more effective to control
efficiency in comparison to stator resistance.
Similarly Fig.15 and Fig.16 indicate that any change in
stator and rotor resistance affects the operating power factor
of the machine. However this effect is large, for same
variation of resistance, in case of rotor as compared to stator.
These discussions leads to the conclusion that stator and
rotor resistance may be selected as main design parameters
for the induction machines to be operated as generators.
VI. CONCLUSION
For the analysis of grid connected induction generators,
estimation of saturated magnetizing reactance is very
important in contrast to motor operation (may be considered
36 Vadhera and Sandhu
as unsaturated value), due to the fact that loading of
generator shifts the operation in to saturated region of
magnetization curve whereas it is not so for motor operation.
Study reveals that most of the research persons adopt the
unsaturated value for magnetizing reactance for the analysis
of such machines, which may lead to inaccurate performance
estimation.
In this paper an iterative technique is proposed to
predict the saturated values of magnetizing reactance with
change in operating conditions. The comparison between
experimental and simulated results proves the validity of
technique proposed. Further efforts have been made to
predict the operating limits of machine using proposed
modeling. Analysis has been extended to identify
effectiveness of the machine parameters to improve the
operating performance of the generator. It is found that
operating performance of the machine may be improved by
proper design of stator and rotor parameters.
APPENDIX I
Specifications;
Machine I
Three Phase, 2.2KW\3HP, 230V, 8.6A, 50Hz, Delta
connected, Squirrel Cage Induction Machine.
V
base
= 230 V
I
base
= 4.96 A
N
base
= 1500 RPM
The Machine parameters are:
R
1
= 3.35, R
2
= 1.76, X
1
= 4.85, X
2
= 4.85
Variation of magnetizing reactance Xm with air gap
voltage E
1
is
0 E
1
117.87 Xm = 108,
117.87 E
1
171.052 Xm = 135.553-0.2337 E
1
,
171.052 E
1
211.919 Xm = 151.160-0.325 E
1
,
211.919 E
1
344.411 Xm = 213.919-0.621 E
1
,
344.411 < E
1
Xm = 0,
REFERENCES
[1] J.E. Barkle, R.W. Ferguson Induction Generator-Theory and
Application AIEE Transaction, 73: 12-19(1954).
[2] S.S. Murthy, O.P. Malik, and A.K. Tandon, Analysis of
self-excited induction generators, Proc. IEE, 129c(6):
260-265(1982).
[3] L. Quazene and G. McPherson, Analysis of the isolated
induction generator, IEEE Trans. Power Apparatus and
Systems, 102(8): 2793-2798(1983).
[4] T.F. Chan, Analysis of self-excited induction generators
using an iterative method, IEEE Trans. Energy Conversion,
10(3): 502-507(1995).
[5] K.S. Sandhu and S.K. Jain, Operational aspects of self-
excited induction generator using a new model, Electric
Machines and Power Systems, 27(2): 169-180(1999).
[6] K.S. Sandhu, Iterative model for the analysis of self-excited
induction generators, Electric Power Components and
Systems, 31(10): 925-939(2003).
[7] S.S. Murthy, C.S. Jha, P.S. Nagendra Rao, Analysis of Grid
Connected Induction Generator Driven By Hydro/Wind
Turbine Under Realistic System Constraints, IEEE Trans.
Energy Conversion, 15(1): 1-7(1990).
[8] A.H. Gorashi, S.S. Murthy, B.P. Singh, Bhim Singh, Analysis
of Wind Driven Grid Connected Induction Generator Under
Unbalanced Grid Connections, IEEE Trans. Energy
Conversion, 9(2): 217-223(1994).
[9] S.S.Y. Narayanan, B.K. Murthy, G.S. Rao, Dynamic Analysis
Of A Grid Connected Induction Generator Driven By A
Wave-Energy Turbine Through Hunting Networks, IEEE
Trans. Energy Conversion, 14(1): 115-121(1999).
[10] M.A. Abdel-Halem, Solid State Control of a Grid Connected
Induction Generator, Electrical Power Components and
Systems, 29(4): 163-178(2001).
[11] T. Fukami, K. Nakagawa, Y. Kanamaru and T. Miyamoto,
A Technique For The Steady State Analysis Of a Grid
Connected Permanent Magnet Induction Generator, IEEE
Trans. Energy Conversion, 19(2): 318-324(2004).
[12] M.V.A. Nunes, J.A.P. Lopes, H.H. Zurn, V.H. Bezerra,
R.G.Almeida, Influence of The Variable Speed Wind
Generators In Transient Stability Margin of The
Conventional Generators Integrated In Electrical Grids,
IEEE Trans. Energy Conversion, 19(4): 692-701, December
(2004).