Synchronous Machine

1 Armature
It is a three phase winding placed in the stator.
The three phase armature windings are placed 120◦ (electrical) apart in
space.
Stator iron is built up of thin laminations to reduce eddy current loss.
2 Field
Field winding is placed in the rotor and requires DC.
There are two types of rotor.
1 Cylindrical Poles
2 Salient Poles

Siva (IIT P) EE549 1 / 102

 Figure: Schematic diagram of a 3 − φ synchronous machine
Source: NPTEL
Siva (IIT P) EE549 2 / 102
The synchronous speed is
120f
Ns =
P
where Ns is in rpm, f is the frequency in Hz and P is the number of poles.

The relationship between the electrical angle (θ) and the mechanical angle
(θm ) is
P
θ = θm
2
In a 2-pole machine,
θ = θm

Siva (IIT P) EE549 3 / 102

Assumptions made in modelling.
1 Stator windings are distributed sinusoidally along the air gap.
2 Stator slots do not cause any variation in the rotor inductances.
3 Magnetic hysteresis is negligible.
4 Magnetic circuit is considered as linear.

The following convention is used.

1 q-axis is leading d-axis.
2 Two amortisseur in q-axis and one in d-axis are used. Hence 2
windings (including the filed winding) in d-axis and 2 windings in
q-axis.

Siva (IIT P) EE549 4 / 102

 Figure: Stator and Rotor windings of a synchronous machine
Source: NPTEL
Siva (IIT P) EE549 5 / 102
Voltage Equations
The stator voltage equations are
dλa
va = −ia rs + (1)
dt
dλb
vb = −ib rs + (2)
dt
dλc
vc = −ic rs + (3)
dt
where

va , vb , vc = instantaneous stator phase to neutral voltage

ia , ib , ic = instantaneous stator currents in phases a,b,c
rs = stator resistance per phase
λa , λb , λc = flux linkages in phases a,b,c

Siva (IIT P) EE549 6 / 102

The rotor circuit voltage equations are
dλfd
vfd = ifd rfd + (4)
dt
dλ1d
v1d = i1d r1d + (5)
dt
dλ1q
v1q = i1q r1q + (6)
dt
dλ2q
v2q = i2q r2q + (7)
dt

vfd , v1d , v1q , v2q = rotor field, 1d, 1q, 2q voltages

ifd , i1d , i1q , i2q = rotor field, 1d, 1q, 2q currents
rfd , r1d , r1q , r2q = rotor field, 1d, 1q, 2q resistances
λfd , λ1d , λ1q , λ2q = rotor field, 1d, 1q, 2q flux linkages

Siva (IIT P) EE549 7 / 102

Flux Linkage Equations

The flux linkages in the phase a, b and c windings are

λa = −laa ia − lab ib − lac ic + lafd ifd + la1d iid + la1q i1q + la2q i2q
λb = −lba ia − lbb ib − lbc ic + lbfd ifd + lb1d iid + lb1q i1q + lb2q i2q
λc = −lca ia − lcb ib − lcc ic + lcfd ifd + lc1d iid + lc1q i1q + lc2q i2q

The negative sign in the stator currents is due to their direction.

In matrix form,
 
 i
lafd la1d la1q la2q  fd 
     
λa laa lab lac ia
λb  = − lab lbb lbc  ib  + lbfd lb1d lb1q lb2q  i1d  (8)
i1q 
λc lac lbc lcc ic lcfd lc1d lc1q lc2q
i2q

λabc = −Lss iabc + Lsr ir

Siva (IIT P) EE549 8 / 102

The rotor circuit flux linkages can be written as
      
λfd lfda lfdb lfdc   lfdfd lfd1d 0 0 ifd
λ1d  l1da l1db l1dc  a i l1dfd l1d1d
  = − 0 0  i1d 
l1qa l1qb l1qc  ib +  0
    
λ1q  0l1q1q l1q2q  i1q 
ic
λ2q l2qa l2qb l2qc 0 0l2q1q l2q2q i2q
(9)
The mutual inductance between the d-axis windings and the q-axis
windings is zero as their fluxes are in quadrature.

λr = −Lrs iabc + Lrr ir

Lss , Lsr and Lrs depend on θ.

Siva (IIT P) EE549 9 / 102

The permeance varies with θ.

Pp
Pavg

0 90 180 270 360 θ (degree)

P = Pavg + Pp cos(2θ)

Siva (IIT P) EE549 10 / 102

 q-axis d-axis Phase a axis

MMFaq

MMFa

MMFad
θ

θ + 90◦

Figure: Phase a mmf

Siva (IIT P) EE549 11 / 102

Stator Self Inductances

peak MMFa = Na ia
peak MMFad = Na ia cos θ
peak MMFaq = Na ia cos(θ + 90◦ ) = −Na ia sin θ

Na ia cos θ
θ d
Na ia sin θ MMFa

Siva (IIT P) EE549 12 / 102

The fluxes along d and q axes are

φad = Na ia cos θPd

φaq = −Na ia sin θPq

The flux along the axis a is given by

φaa = φad cos θ − φaq sin θ

φaa = Na ia Pd cos2 θ + Na ia Pq sin2 θ

 
1 + cos 2θ 1 − cos 2θ
φaa = Na ia Pd ( ) + Pq ( )
2 2
 
Pd + Pq Pd − Pq
φaa = Na ia + cos 2θ
2 2

Siva (IIT P) EE549 13 / 102

The self inductance laa is
   
Na φaa Pd + Pq Pd − Pq
laa = = Na2 + Na2 cos 2θ
ia 2 2

This can be written as

laa = laa0 + laap cos 2θ (10)

   
2 Pd + Pq 2 Pd − Pq
where laa0 = Na and laap = Na Since the
2 2
windings of phases b and c are identical to a and are displaced from it by
120◦ and 240◦ , respectively,

lbb = laa0 + laap cos 2(θ − 120◦ ) = laa0 + laap cos(2θ + 120◦ ) (11)
◦ ◦
lcc = laa0 + laap cos 2(θ − 240 ) = laa0 + laap cos(2θ − 120 ) (12)

Siva (IIT P) EE549 14 / 102

Stator Mutual Inductances

The flux linking phase a due to mmf of phase b is

φba = φbd cos θ − φbq sin θ

φba = Nb ib Pd cos(θ − 120◦ ) cos θ + Nb ib Pq sin(θ − 120◦ ) sin θ

Pd
φba = Nb ib [( (cos(2θ − 120◦ ) + cos(120◦ ))+
2
Pq
(cos(120◦ ) − cos(2θ − 120◦ ))]
2
 
Pd + Pq Pd − Pq ◦
φba = Nb ib − + (cos(2θ − 120 ))
4 2

Siva (IIT P) EE549 15 / 102

The flux linkage of phase a due to current ib is
 
Pd + Pq Pd − Pq
λba = Na φba = Na Nb ib − + (cos(2θ − 120◦ ))
4 2

Since Na = Nb , the mutual inductance between a and b is

 
Na φba 2 Pd + Pq Pd − Pq ◦
lba = = Na − + (cos(2θ − 120 ))
ib 4 2

1
lba = lab = − laa0 + laap cos(2θ − 120◦ )
2 (13)
1
= − laa0 − laap cos(2θ + 60◦ )
2
Similarly,
1
lca = lac = − laa0 − laap cos(2θ − 60◦ ) (14)
2
1
lbc = lcb = − laa0 − laap cos(2θ − 180◦ ) (15)
2

Siva (IIT P) EE549 16 / 102

 − 12 laa0 − laap cos(2θ + 60◦ ) − 21 laa0 − laap cos(2θ − 60◦ )
 
laa0 + laap cos 2θ
Lss = − 1 laa0 − laap cos(2θ + 60◦ )
2
laa0 + laap cos(2θ + 120◦ ) 1 ◦
− 2 laa0 − laap cos(2θ − 180 )
− 21 laa0 − laap cos(2θ − 60◦ ) − 21 laa0 − laap cos(2θ − 180◦ ) laa0 + laap cos(2θ − 120◦ )
(16)

Siva (IIT P) EE549 17 / 102

Mutual inductances between stator and rotor
Since the rotor windings see a constant permeance,
 
lafd cos θ la1d cos θ −la1q sin θ −la2q sin θ
Lsr = lbfd cos(θ − 120◦ ) lb1d cos(θ − 120◦ ) −lb1q sin(θ − 120◦ ) −lb2q sin(θ − 120 )◦ 

lcfd cos(θ + 120◦ ) lc1d cos(θ + 120◦ ) −lc1q sin(θ + 120◦ ) −lc2q sin(θ + 120◦ )
(17)
Let
lafd = lbfd = lcfd = lsfd
la1d = lb1d = lc1d = ls1d
la1q = lb1q = lc1q = ls2q
la2q = lb2q = lc2q = ls2q

 
lsfd cos θ ls1d cos θ −ls1q sin θ −ls2q sin θ
Lsr = lsfd cos(θ − 120◦ ) ls1d cos(θ − 120◦ ) −ls1q sin(θ − 120◦ ) −ls2q sin(θ − 120◦ )
lsfd cos(θ + 120◦ ) ls1d cos(θ + 120◦ ) −ls1q sin(θ + 120◦ ) −ls2q sin(θ + 120◦ )
(18)

Siva (IIT P) EE549 18 / 102

Lss , Lsr and Lrs vary with angle θ which in turn varies with time.
This introduces complexity in solving problems.
Therefore, a suitable transformation is required.
dq0 transformation is used to transform the stator variables into new
variables.

Siva (IIT P) EE549 19 / 102

dq0 Transformation

The transformation matrix is

cos(θ − 120◦ ) cos(θ + 120◦ )
 
cos θ
Tdq0 = k1 − sin θ − sin(θ − 120◦ ) − sin(θ + 120◦ ) (19)
k2 k2 k2

It is also called Park’s Transformation.

The constants k1 and k2 are arbitrary.
In most of the literature, k1 = 32 and k2 = 21 are used.
q
k1 = 23 and k2 = √12 are also used in some literature.
2 1
In this course, k1 = 3 and k2 = 2 are used.

Siva (IIT P) EE549 20 / 102

Let us apply the dq0 transformation to stator currents.

ia = Im cos(ωt)
ib = Im cos(ωt − 120◦ )
ic = Im cos(ωt + 120◦ )
   
id ia
iq  = Tdq0 ib  (20)
i0 ic
cos(θ − 120◦ ) cos(θ + 120◦ )
    
id cos θ Im cos ωt
iq  = 2 − sin θ − sin(θ − 120◦ ) − sin(θ + 120◦ ) Im cos(ωt − 120◦ )
3
i0 1
2
1
2
1
2 Im cos(ωt + 120◦ )
   
id Im cos(ωt − θ)
iq  =  Im sin(ωt − θ)  (21)
i0 0

Siva (IIT P) EE549 21 / 102

If θ is measured with respect to the synchronously rotating reference,

θ = ωt + δ
   
id Im cos δ
iq  = −Im sin δ  (22)
i0 0
For balanced steady state operation, id and iq are constant.

1
i0 = (ia + ib + ic ) = 0
3

Since k1 = 32 and k2 = 12 , the peak value of id and iq is the same as

the peak value of the stator current.
Two fictitious windings carrying current id and iq rotating at
synchronous speed produce the same effect as that of three phase
stator windings.

Siva (IIT P) EE549 22 / 102

The inverse transformation is
 
cos θ − sin θ 1
T−1 ◦ ◦
dq0 = cos(θ − 120 ) − sin(θ − 120 ) 1
  (23)
◦ ◦
cos(θ + 120 ) − sin(θ + 120 ) 1

vdq0 = Tdq0 vabc ; idq0 = Tdq0 iabc

vabc = T−1 −1
dq0 vdq0 ; iabc = Tdq0 idq0

Let us apply the dq0 transformation to the instantaneous power.

 T     T   
va ia vd id
s = vb  ib  = T−1 vq  T−1 iq  (24)
dq0 dq0
vc ic v0 i0
 T  
vd id
s = vq  (T−1
dq0 ) T −1  
Tdq0 iq
v0 i0

Siva (IIT P) EE549 23 / 102

  T  3  
vd 2 0 0 id
3
s =  vq   0 2 0
  iq 
v0 0 0 3 i0
3
s = va ia + vb ib + vc ic = (vd id + vq iq + 2v0 i0 ) (25)
2
The transformation is not power invariant.

va ia + vb ib + vc ic 6= vd id + vq iq + v0 i0

q
2 √1 ,
If k1 = 3 and k2 = 2
the transformation will be power invariant.

Siva (IIT P) EE549 24 / 102

Let us apply the dq0 transformation to stator voltages.
     
vd rs 0 0 id
vq  = −Tdq0  0 rs 0  T−1 iq 
dq0
v0 0 0 rs i0
  
λd
d
+Tdq0 T−1 dq0
λq 
dt
λ0
     
vd rs 0 0 id
vq  = −Tdq0  0 rs 0  T−1
dq0
iq 
v0 0 0 rs i0
 
λ
d  d 
+Tdq0 T−1
dq0 dt λq
λ0
 
λd
d  
+Tdq0 T−1
dq0
 λq 
dt
λ0
Siva (IIT P) EE549 25 / 102
          
vd rs 0 0 id 0 −ω 0 λd λd
vq  = −  0 d
rs 0  iq  + ω 0 0 λq  + λq 
dt
v0 0 0 rs i0 0 0 0 λ0 λ0
The stator and rotor voltage equations are as follows:
dλd
vd = −rs id − ωλq + (26)
dt
dλq
vq = −rs iq + ωλd + (27)
dt
dλ0
v0 = −rs i0 + (28)
dt
dλfd
vfd = rfd ifd + (29)
dt
dλ1d
v1d = r1d i1d + (30)
dt
dλ1q
v1q = r1q i1q + (31)
dt
dλ2q
v2q = r2q i2q + (32)
dt
Siva (IIT P) EE549 26 / 102
Applying the dq0 transformation to the stator flux equations.
 
    ifd
λd id
λq  = −Tdq0 Lss T−1 iq  + Tdq0 Lsr i1d 
 
dq0 i1q  (34)
λ0 i0
i2q

After simplification,
  3  
λd 2 (laa0 + laap ) 0 0 id
 λq  = −  3
0 2 (laa0 − laap ) 0 iq 
 
λ0 0 0 laa0 i0
(35)
 
 i
0  fd 

lsfd ls1d 0
i1d 
+ 0 0 ls1q ls2q  i1q 
0 0 0 0
i2q

Siva (IIT P) EE549 27 / 102

Let us define the following.

3 3
ld = (laa0 + laap ); lq = (laa0 − laap )
2 2
where ld , lq are the inductances along d, q-axis corresponding to the total
flux linkage.
ld = lmd + lls ; lq = lmq + lls
where

lmd = the mutual inductance along d

lmq = the mutual inductance along q
lls = leakage inductance

Siva (IIT P) EE549 28 / 102

 λd = −(lmd + lls )id + lsfd ifd + ls1d i1d (36)
λq = −(lmq + lls )iq + ls1q i1q + ls2q i2q (37)
λ0 = −lls i0 (38)
3
λfd = − lsfd id + lfdfd ifd + lfd1d i1d (39)
2
3
λ1d = − ls1d id + lfd1d ifd + l1d1d i1d (40)
2
3
λ1q = − ls1q iq + l1q1q i1q + l1q2q i2q (41)
2
3
λ2q = − ls2q iq + l1q2q i1q + l2q2q i2q (42)
2
(43)

Siva (IIT P) EE549 29 / 102

Advantages of dq0 transformation

1 The dynamic performance equations have constant inductances.

2 For balanced conditions, zero sequence quantities disappear.
3 For balanced steady state operation, the stator quantities have
constant values. For other modes of operation, they vary with time.
Stability studies involve slow variations having frequencies below 2 to
3 Hz.
4 The d and q-axed parameters may be directly measured from tests.

Siva (IIT P) EE549 30 / 102

Per unit representation

The dq0 transformation converts inductances into constants

irrespective of angle θ.
However, the mutual inductances between the stator windings and
the rotor windings are not equal.
To make them equal, a proper per unit system should be chosen.
This will minimize computational complexity.

Siva (IIT P) EE549 31 / 102

Base Quantities for Stator Variables
Let us define the following.

Vbase = Vrms (l − n) kV
Sbase = 3Vbase Ibase = three phase rated MVA
Sbase
Ibase = kA
3Vbase
Vbase
Zbase = Ω
Ibase
ωbase = ωs = 2πf elect. rad/sec (44)
2
ωm base = ωbase mech. rad/sec
P
Zbase
Lbase = H
ωbase
Vbase 1
λbase = Wb-turn, tbase = s/rad
ωbase ωbase

Siva (IIT P) EE549 32 / 102

 √
Vdq0,base = 2Vbase kV
3
Sbase = Vdq0,base Idq0,base MVA
2 √
2 Sbase 2 Sbase 2Sbase √
Idq0,base = = √ = = 2Ibase kA
3 Vdq0,base 3 2Vbase 3Vbase
Vdq0,base (45)
Zdq0,base = Ω
[Link]
Zdq0,base
Ldq0,base = H
ωbase
Vdq0,base
λdq0,base = Wb-turn
ωbase

Siva (IIT P) EE549 33 / 102

Base Quantities for Rotor Variables

Sbase = Vfd,base Ifd,base

lmd
Ifd,base = Idq0,base
lsfd
Sbase 3 lsfd
Vfd,base = = Vdq0,base
Ifd,base 2 lmd
Vfd,base (46)
λfd,base =
ωbase
Vfd,base
Zfd,base =
Ifd,base
Vfd,base Zfd,base
Lfd,base = =
ωbase Ifd,base ωbase

Siva (IIT P) EE549 34 / 102

Let us define them for damper windings.
1 d-axis winding

Sbase = V1d,base I1d,base

lmd
I1d,base = Idq0,base
ls1d
Sbase 3 ls1d
V1d,base = = Vdq0,base
I1d,base 2 lmd
V1d,base (47)
λ1d,base =
ωbase
V1d,base
Z1d,base =
I1d,base
V1d,base Z1d,base
L1d,base = =
ωbase I1d,base ωbase

Siva (IIT P) EE549 35 / 102

2 q-axis winding

Sbase = V1q,base I1q,base

lmq
I1q,base = Idq0,base
ls1q
Sbase 3 ls1q
V1q,base = = Vdq0,base
I1q,base 2 lmq
V1q,base (48)
λ1q,base =
ωbase
V1q,base
Z1q,base =
I1q,base
V1q,base Z1q,base
L1q,base = =
ωbase I1q,base ωbase

Siva (IIT P) EE549 36 / 102

 Sbase = V2q,base I2q,base
lmq
I2q,base = Idq0,base
ls2q
Sbase 3 ls2q
V2q,base = = Vdq0,base
I2q,base 2 lmq
V2q,base (49)
λ2q,base =
ωbase
V2q,base
Z2q,base =
I2q,base
V2q,base Z2q,base
L2q,base = =
ωbase I2q,base ωbase

Siva (IIT P) EE549 37 / 102

Let us define the following.
vd vq v0
Vd = , Vq = , V0 =
Vdq0,base Vdq0,base Vdq0,base
id iq i0
Id = , Iq = , I0 =
Idq0,base Idq0,base Idq0,base
vfd v1d v1q v2q
Vfd = , V1d = , V1q = , V2q =
Vfd,base V1d,base V1q,base V2q,base
ifd i1d i1q i2q
Ifd = , I1d = , I1q = , I2q =
Ifd,base I1d,base I1q,base I2q,base
(50)
λd λq
ψd = , ψq =
λdq0,base λdq0,base
λfd λ1d λ1q λ2q
ψfd = , ψ1d = , ψ1q = , ψ2q =
λfd,base λ1d,base λ1q,base λ2q,base
rs
Rs =
Zdq0,base
rfd r1d r1q r2q
Rfd = , R1d = , R1q = , R2q = ,
Zfd,base Z1d,base Z1q,base Z2q,base

Siva (IIT P) EE549 38 / 102

Voltage Equations in per unit
The stator and rotor voltage equations given in equations (26)-(32) are
converted to per unit form as follows:
ω 1 dψd
Vd = −Rs Id − ψq + (51)
ωbase ωbase dt
ω 1 dψq
Vq = −Rs Iq + ψd + (52)
ωbase ωbase dt
1 dψ0
V0 = −Rs I0 + (53)
ωbase dt
1 dψfd
Vfd = Rfd Ifd + (54)
ωbase dt
1 dψ1d
V1d = R1d I1d + (55)
ωbase dt
1 dψ1q
V1q = R1q I1q + (56)
ωbase dt
1 dψ2q
V2q = R2q I2q + (57)
ωbase dt
Siva (IIT P) EE549 39 / 102
Instantaneous Power

Before converting it into per unit, let us substitute voltage equations given
in (26)-(28) in (25).
 
3 dλd dλq dλ0
s= (−rs id − ωλq + )id + (−rs iq + ωλd + )iq + 2(−rs i0 + )i0
2 dt dt dt
   
3 dλd dλq dλ0 3 3
rs id2 + rs iq2 + 2rs i02


= id + iq + i0 + ω (λd iq − λq id ) −
2 dt dt dt 2 2
(59)

For a balanced system, v0 and i0 are zero.

 

 


 3  dλ  
d dλ q 3 3 2 2


s= id + iq + ω (λd iq − λq id ) − rs id + rs iq (60)

 2 dt dt 2
| {z } 2
| {z } 


| {z } 

I II III

Siva (IIT P) EE549 40 / 102

 I = rate of change of stored magnetic energy in the stator windings
II = electrical power crossing the air gap
III = stator copper losses

The electrical torque Te in N-m can be found

3
Pe Pe ω (λd iq − λq id )
Te = = 2 = 2 2
ωm Pω Pω (61)
3P
Te = (λd iq − λq id ) N-m
22
To convert it into per unit, let us define the following.
3
Sbase Vdq0,base Idq0,base
Tbase = = 2 2
ωm,base P ωbase (62)
3P
= λdq0,base Idq0,base
22
Siva (IIT P) EE549 41 / 102
The electrical torque (Te ) in per unit is given by
3P
2 2 (λd iq − λq id )
Te = 3P
2 2 λdq0,base Idq0,base

Te = ψd Iq − ψq Id (63)
In per unit calculations, the torque and power are the same.

Siva (IIT P) EE549 42 / 102

Flux Linkage Equations in per unit
1 Flux linkage equations of windings in d-axis

λd = −(lmd + lls )id + lsfd ifd + ls1d i1d (64)

3
λfd = − lsfd id + lfdfd ifd + lfd1d i1d (65)
2
3
λ1d = − ls1d id + lfd1d ifd + l1d1d i1d (66)
2
From (51),

id = Id Idq0,base ; ifd = Ifd Ifd,base ; i1d = I1d I1d,base

Equations (64) -(66) are divided by their respective bases to convert

them into per unit form.

λd (lmd + lls )Id Idq0,base lsfd Ifd Ifd,base ls1d I1d I1d,base
=− + +
λdq0,base λdq0,base λdq0,base λdq0,base
(67)

Siva (IIT P) EE549 43 / 102

 λfd 3 lsfd Id Idq0,base lfdfd Ifd Ifd,base lfd1d I1d I1d,base
=− + + (68)
λfd,base 2 λfd,base λfd,base λfd,base
λ1d 3 ls1d Id Idq0,base lfd1d Ifd Ifd,base l1d1d I1d I1d,base
=− + + (69)
λ1d,base 2 λ1d,base λ1d,base λ1d,base
From (46) - (48),

[Link] Vfd,base V1d,base

λdq0,base = ; λfd,base = ; λ1d,base =
ωbase ωbase ωbase

By using the above terms, the equations (67) - (69) can be written as
follows:
ωbase (lmd + lls )Id Idq0,base ωbase lsfd Ifd Ifd,base ωbase ls1d I1d I1d,base
ψd = − + +
Vdq0,base Vdq0,base Vdq0,base
(70)

Siva (IIT P) EE549 44 / 102

 3 ωbase lsfd Id Idq0,base ωbase lfdfd Ifd Ifd,base ωbase lfd1d I1d I1d,base
ψfd = − + +
2 Vfd,base Vfd,base Vfd,base
(71)
3 ωbase ls1d Id Idq0,base ωbase lfd1d Ifd Ifd,base ωbase l1d1d I1d I1d,base
ψ1d = − + +
2 V1d,base V1d,base V1d,base
(72)
From (46) and (47),

lmd 3 lsfd
Ifd,base = Idq0,base ; Vfd,base = Vdq0,base
lsfd 2 lmd
lmd 3 ls1d
I1d,base = Idq0,base ; V1d,base = Vdq0,base
ls1d 2 lmd

By using the above terms, the equations (70) - (72) can be written as
follows:

Siva (IIT P) EE549 45 / 102

 ωbase (lmd + lls )Id Idq0,base ωbase lmd Ifd Idq0,base ωbase lmd I1d Idq0,b
ψd = − + +
Vdq0,base Vdq0,base Vdq0,base
(73)
ωbase lmd Id Idq0,base ωbase lfdfd Ifd Ifd,base ωbase lfd1d I1d I1d,base
ψfd = − + +
Vdq0,base Vfd,base Vfd,base
(74)
ωbase lmd Id Idq0,base ωbase lfd1d Ifd Ifd,base ωbase l1d1d I1d I1d,base
ψ1d = − + +
Vdq0,base V1d,base V1d,base
(75)

The mutual inductances have been made equal.

Siva (IIT P) EE549 46 / 102

2 Flux linkage equations of windings in q-axis

After following the same steps used in d-axis,

ωbase (lmq + lls )Iq Idq0,base ωbase lmq I1q Idq0,base ωbase lmq I2q Idq0,ba
ψq = − + +
Vdq0,base Vdq0,base Vdq0,base
(76)
ωbase lmq Iq Idq0,base ωbase l1q1q I1q I1q,base ωbase l1q2q I2q I2q,base
ψ1q =− + +
Vdq0,base V1q,base V1q,base
(77)
ωbase lmq Iq Idq0,base ωbase l1q2q I1q I1q,base ωbase l2q2q I2q I2q,base
ψ2q =− + +
Vdq0,base V2q,base V2q,base
(78)

Siva (IIT P) EE549 47 / 102

Let us define the following parameters to simplify the equations (73) -(75)
and (76) - (78).

xls ωbase lls ωbase lls Idq0,base

Xls = = =
Zdq0,base Zdq0,base Vdq0,base
xmd ωbase lmd ωbase lmd Idq0,base
Xmd = = =
Zdq0,base Zdq0,base Vdq0,base
xmq ωbase lmq ωbase lmq Idq0,base
Xmq = = =
Zdq0,base Zdq0,base Vdq0,base
xfdfd ωbase lfdfd ωbase lfdfd Ifd,base
Xfd = = = (79)
Zfd,base Zfd,base Vfd,base
x1d1d ωbase l1d1d ωbase l1d1d I1d,base
X1d = = =
Z1d,base Z1d,base V1d,base
x1q1q ωbase l1q1q ωbase l1q1q I1q,base
X1q = = =
Z1q,base Z1q,base V1q,base
x2q2q ωbase l2q2q ωbase l2q2q I2q,base
X2q = = =
Z2q,base Z2q,base V2q,base
Siva (IIT P) EE549 48 / 102
 ωbase lfd1d I1d,base ωbase lfd1d lsfd
Xfd1d = = Ifd,base
Vfd,base Vfd,base ls1d
ωbase lfd1d lsfd
=
Zfd,base ls1d
(80)
ωbase l1q2q I2q,base ωbase l1q2q ls1q
X1q2q = = I1q,base
V1q,base V1q,base ls2q
ωbase l1q2q ls1q
=
Z1q,base ls2q

Xd = Xls + Xmd , Xq = Xls + Xmq , Xlfd = Xfd − Xmd

(81)
Xl1d = X1d − Xmd , Xl1q = X1q − Xmq , Xl2q = X2q − Xmq

Siva (IIT P) EE549 49 / 102

 Xfd1d X1q2q
By using (79), (80), (81) and defining Kd = Xmd and Kq = Xmq , the flux
linkage equations can be written as follows:

ψd = Xd (−Id ) + Xmd Ifd + Xmd I1d (82)

ψfd = Xmd (−Id ) + Xfd Ifd + Kd Xmd I1d (83)
ψ1d = Xmd (−Id ) + Kd Xmd Ifd + X1d I1d (84)
ψq = Xq (−Iq ) + Xmq I1q + Xmq I2q (85)
ψ1q = Xmq (−Iq ) + X1q I1q + Kq Xmq I2q (86)
ψ2q = Xmq (−Iq ) + Kq Xmq I1q + X2q I2q (87)

Siva (IIT P) EE549 50 / 102

If Kd and Kq are assumed to be unity,
1 d-axis flux linkage equations

ψd = Xd (−Id ) + Xmd Ifd + Xmd I1d (88)

ψfd = Xmd (−Id ) + Xfd Ifd + Xmd I1d (89)
ψ1d = Xmd (−Id ) + Xmd Ifd + X1d I1d (90)

2 q-axis flux linkage equations

ψq = Xq (−Iq ) + Xmq I1q + Xmq I2q (91)

ψ1q = Xmq (−Iq ) + X1q I1q + Xmq I2q (92)
ψ2q = Xmq (−Iq ) + Xmq I1q + X2q I2q (93)

Siva (IIT P) EE549 51 / 102

Synchronous Machine Parameters
Synchronous machine parameters are normally specified in terms of
sub-transient and transient reactances, open circuit sub-transient and
transient time constants.
Those parameters need to be defined.
1 Sub-transient reactance

1
Xd00 = Xls + 1 1 1
(94)
Xmd + Xlfd + Xl1d

This can be found by applying a voltage at stator terminals such that

only Id flows with all rotor circuits shorted.
At t = 0+ , ψfd and ψ1d will be zero.

Siva (IIT P) EE549 52 / 102

Hence,

ψd = Xd (−Id ) + Xmd Ifd + Xmd I1d (95)

0 = Xmd (−Id ) + Xfd Ifd + Xmd I1d (96)
0 = Xmd (−Id ) + Xmd Ifd + X1d I1d (97)

Solving (96) and (97) for Ifd and I1d and considering (81),

Xmd Xl1d
Ifd = Id (98)
Xmd Xlfd + Xmd Xl1d + Xlfd Xl1d
Xmd Xlfd
I1d = Id (99)
Xmd Xlfd + Xmd Xl1d + Xlfd Xl1d

Substituting (98) and (99) in (95),

 2 (X
Xmd

lfd + Xl1d )
ψd = − Xd − Id (100)
Xmd Xlfd + Xmd Xl1d + Xlfd Xl1d

Siva (IIT P) EE549 53 / 102

Since Xd = Xls + Xmd ,
 
Xmd Xlfd Xl1d
ψd = − Xls + Id
Xmd Xlfd + Xmd Xl1d + Xlfd Xl1d
! (101)
1
= − Xls + 1 1 1
Id = −Xd00 Id
Xmd + Xlfd + Xl1d

Similarly, the quadrature axis sub-transient reactance can be

expressed as,
1
Xq00 = Xls + 1 1 1
(102)
Xmq + Xl1q + Xl2q

Siva (IIT P) EE549 54 / 102

2 Transient reactance
1
Xd0 = Xls + 1 1
(103)
Xmd + Xlfd

It is found by the same approach followed for Xd00 with an additional

assumption that the damper winding transients settle down faster
compared to the field winding transients. Hence I1d =0. With this
assumption, by solving (95) and (96) and using (81),
 2
Xmd
  
Xmd Xlfd
ψd = − Xd − Id = − Xls + Id
Xmd + Xlfd Xmd + Xlfd
! (104)
1 0
= − Xls + 1 1
Id = −Xd Id
Xmd + Xlfd

Similarly, the quadrature axis transient reactance can be defined as

1
Xq0 = Xls + 1 1
(105)
Xmq + Xl1q
Siva (IIT P) EE549 55 / 102
3 Open circuit sub-transient time constant
A voltage is applied to the field winding with stator terminals open so
that Id = 0.
At t = 0+ , ψ1d = 0.
Substituting these assumptions in (90) and (55),

X1d
Ifd = − I1d (106)
Xmd
 
1 dIfd dI1d
Xmd + X1d + R1d I1d = V1d = 0 (107)
ωbase dt dt
 
1 dIfd 1 X1d dI1d R1d
= − − I1d (108)
ωbase dt ωbase Xmd dt Xmd
Substituting (89) in (54), we get
 
1 dIfd dI1d
Xfd + Xmd + Rfd Ifd = Vfd (109)
ωbase dt dt

Siva (IIT P) EE549 56 / 102

To get a first order differential equation in terms of I1d with a forcing
fucntion Vfd , let us substitute (106) and (108) in (109).
 
1 Xfd X1d dI1d Xfd R1d dI1d Rfd X1d
− − I1d + Xmd − I1d = Vfd
ωbase Xmd dt Xmd dt Xmd
(110)
On simplification, we get
 2 
1 Xmd − Xfd X1d dI1d Xmd
− I1d = Vfd (111)
ωbase Xfd R1d + X1d Rfd dt Xfd R1d + X1d Rfd

In a practical generator, R1d >> Rfd . Therefore, X1d Rfd can be

neglected. Simplifying (111), we get
 
1 Xmd Xlfd dI1d Xmd
− Xl1d + − I1d = Vfd (112)
ωbase R1d Xmd + Xlfd dt Xfd R1d

Siva (IIT P) EE549 57 / 102

Hence, the open circuit d-axis time sub-transient time constant is
 
00 1 Xmd Xlfd
Td0 = Xl1d +
ωbase R1d Xmd + Xlfd
! (113)
1 1
= Xl1d + 1 1
ωbase R1d Xmd + Xlfd

Similarly, the open circuit q-axis time sub-transient time constant can
be expressed as
!
00 1 1
Tq0 = Xl2q + 1 1
(114)
ωbase R2q Xmq + Xl1q

Siva (IIT P) EE549 58 / 102

4 Open circuit transient time constant
00 , the current in
In addition to the assumptions made in calculating Td0
the damper winding I1d is zero. From (89) and (54), we get

ψfd = Xfd Ifd (115)

Xfd dIfd Vfd

+ Ifd = (116)
ωbase Rfd dt Rfd
Hence, the open circuit d-axis transient time constant, in seconds, is
defined as
0 Xfd
Td0 = (117)
ωbase Rfd
Similarly, the open circuit q-axis transient time constant can be
expressed as
0 X1q
Tq0 = (118)
ωbase R1q

Siva (IIT P) EE549 59 / 102

Let us also define the following new variables.

Xmd Xmd Xmq

Eq0 = ψfd , Efd = Vfd , Ed0 = − ψ1q (119)
Xfd Rfd X1q

Our aim is to express the voltage equations and flux linkage equations
using the sub-transient and transient reactances and open circuit time
constants along with (119).

From (89) and (90),

   −1    −1  
Ifd Xfd Xmd ψfd Xfd Xmd Xmd
= + I (120)
I1d Xmd X1d ψ1d Xmd X1d Xmd d
2 =X X
Let ∆ = Xfd X1d − Xmd lfd md + Xl1d Xmd + Xlfd Xl1d Therefore,

Siva (IIT P) EE549 60 / 102

 1 1 2


Ifd = (X1d ψfd − Xmd ψ1d ) + X1d Xmd − Xmd Id
∆ ∆ (121)
1 1 2


I1d = (−Xmd ψfd + Xfd ψ1d ) + Xfd Xmd − Xmd Id
∆ ∆
Equation (121) can be expressed using Xd00 , Xd0 , Xd and (121) as given
below.
X2
 
1 Xfd X1d Xmd 1 Xmd
X1d ψfd = ψfd = 1 + md ψfd
∆ Xmd ∆ Xfd Xmd ∆ Xfd
2
Xmd (Xd −Xd0 )(Xd0 −Xd00 ) Xmd
Since ∆ = (Xd0 −Xls )2
and Eq0 = Xfd ψfd ,

(Xd − Xd0 )(Xd0 − Xd00 )

 
1 1
X1d ψfd = 1+ Eq0
∆ Xmd (Xd0 − Xls )2

Siva (IIT P) EE549 61 / 102

 2 (Xd − Xd0 )(Xd0 − Xd00 )
 
1 1 Xmd 1
Xmd ψ1d = ψ1d = ψ1d
∆ Xmd ∆ Xmd (Xd0 − Xls )2

1 2

1 Xlfd Xl1d Xmd
X1d Xmd − Xmd Id =
∆ Xlfd ∆
(Xd0 −Xls ) Xlfd Xl1d Xmd
Since 1
Xlfd = X
(Xd −Xd0 ) md
and ∆ = Xd00 − Xls ,

(Xd − Xd0 )(Xd00 − Xls )

 
1 2

1
X1d Xmd − Xmd Id = Id
∆ Xmd (Xd0 − Xls )

Siva (IIT P) EE549 62 / 102

 1 Xfd Xmd
Xmd ψfd = ψfd
∆ ∆ Xfd
Xfd (Xd0 −Xd00 ) Xmd
Since ∆ = (Xd0 −Xls )2
and Eq0 = Xfd ψfd ,

1 (Xd0 − Xd00 ) 0
Xmd ψfd = E
∆ (Xd0 − Xls )2 q

Xfd (Xd0 − Xd00 )

ψ1d = ψ1d
∆ (Xd0 − Xls )2
1 2 Xlfd Xmd (X 0 − Xd00 )
(Xfd Xmd − Xmd )Id = Id = d0 Id
∆ ∆ (Xd − Xls )

Siva (IIT P) EE549 63 / 102

By using the above terms, the equation (121) can be written as
(Xd − Xd0 )(Xd0 − Xd00 ) (Xd0 − Xls )(Xd00 − Xls )
  
1 0 0
Ifd = Eq + Eq + Id + ψ1d
Xmd (Xd0 − Xls )2 (Xd0 − Xd00 )
1
Eq0 + (Xd − Xd0 )(Id − I1d )


=
Xmd
(122)

(Xd0 − Xd00 )
I1d = (−Eq0 + ψ1d + Id (Xd0 − Xls )) (123)
(Xd0 − Xls )2
Substituting (122) and (123) in (88),

ψd = Xd (−Id ) + Xmd Ifd + Xmd I1d

= Eq0 − Xd0 Id + (Xd0 − Xls )I1d
(124)
(X 00 − Xls ) 0 (Xd0 − Xd00 )
= −Xd00 Id + d0 E + ψ1d
(Xd − Xls ) q (Xd0 − Xls )

Siva (IIT P) EE549 64 / 102

Similarly, the q-axis flux linkage and currents can be expressed as

(Xq00 − Xls ) 0 (Xq0 − Xq00 )

ψq = −Xq00 Iq − E d + ψ2q (125)
(Xq0 − Xls ) (Xq0 − Xls )
1
−Ed0 + (Xq − Xq0 )(Iq − I2q )


I1q = (126)
Xmq
(Xq0 − Xq00 ) 0
I2q = (E + ψ2q + Iq (Xq0 − Xls )) (127)
(Xq0 − Xls )2 d

Siva (IIT P) EE549 65 / 102

    
00 1 Xmd Xlfd 1 Xl1d Xmd + Xl1d Xlfd + Xmd Xlfd
Td0 = Xl1d + =
ωbase R1d Xmd + Xlfd ωbase R1d Xfd
 0
(Xd − Xls )2
  
1 ∆ 1
= =
ωbase R1d Xfd ωbase R1d (Xd0 − Xd00 )

Similarly, !
00 1 (Xq0 − Xls )2
Tq0 =
ωbase R2q (Xq0 − Xq00 )

By substituting the equations (122) - (127) in the voltage equations given

in (54) - (57), we get

1 dψd ω
= Vd + Rs Id + ψq (128)
ωbase dt ωbase
1 dψq ω
= Vq + Rs Iq − ψd (129)
ωbase dt ωbase
1 dψ0
= V0 + Rs I0 (130)
ωbase dt
Siva (IIT P) EE549 66 / 102
 0
dEq0 (Xd0 − Xd00 )
Td0 = −Eq0 − (Xd − Xd0 )[Id − (−Eq0
dt (Xd0 − Xls )2 (131)
+ (Xd0 − Xls )Id + ψ1d ] + Efd

00 dψ1d
Td0 = Eq0 − (Xd0 − Xls )Id − ψ1d (132)
dt
0 (Xq0 − Xq00 ) 0

0 dEd 0 0
Tq0 = −Ed + (Xq − Xq ) Iq − (E + (Xq0 − Xls )Iq + ψ2q )
dt (Xq0 − Xls )2 d
(133)
00 dψ2q
Tq0 = −Ed0 − (Xq0 − Xls )Iq − ψ2q (134)
dt
(X 00 − Xls ) 0 (Xd0 − Xd00 )
ψd = −Xd00 Id + d0 E + ψ1d (135)
(Xd − Xls ) q (Xd0 − Xls )
(Xq00 − Xls ) 0 (Xq0 − Xq00 )
ψq = −Xq00 Iq − E d + ψ2q (136)
(Xq0 − Xls ) (Xq0 − Xls )

Siva (IIT P) EE549 67 / 102

Apart from the above equations, the swing equation also needs to be
considered.

dδ
= ω − ωs (137)
dt
2H dω
= Tm − Te − D(ω − ωbase ) (138)
ωs dt
where Te = ψd Iq − ψq Id

Siva (IIT P) EE549 68 / 102

Example 1

A 555 MVA, 24 kV, 0.9 pf, 60 Hz, 3 phase, 2 pole synchronous generator
has the following inductances and resistances associated with the stator
and rotor field windings: (P. Kundur Example 3.1)

laa = 3.2758 + 0.0458 cos(2θ) mH

lab = −1.6379 − 0.0458 cos(2θ + 60◦ ) mH
lls = 0.4129 mH
lafd = 40 cos θ mH; lfd = 576.92 mH
rs = 0.0031 Ω; rfd = 0.0715 Ω

1 Determine ld and lq in Henrys.

2 Determine xmd and xmq in Ohms.
3 Determine the per unit values all the parameters in dq0 frame.

Siva (IIT P) EE549 69 / 102

Example 1 (contd.)

1 We know that
3 3
ld = (laa0 + laap ); lq = (laa0 − laap )
2 2
(refer to slide 28).

3
ld = (3.2758 + 0.0458) = 4.9824 mH
2
3
lq = (3.2758 − 0.0458) = 4.845 mH
2

Siva (IIT P) EE549 70 / 102

Example 1 (contd.)

xmd = xd − xls xmq = xq − xls

(refer to slide 49 and equation (81) ) Let us first find inductances

lmd = ld − lls = 4.9824 − 0.4129 = 4.5695 mH

lmq = lq − lls = 4.845 − 0.4129 = 4.4321 mH

xmd = ωbase lmd = 2 × π × 60 × 4.5695 = 1.7227 Ω

xmq = ωbase lmq = 2 × π × 60 × 4.4321 = 1.6709 Ω

Siva (IIT P) EE549 71 / 102

Example 1 (contd.)

3 Let us find the base quantities (refer to slides 32, 33, & 34).

24
Vbase = √ = 13.8564 kV Sbase = 555 MVA
3
Sbase 555
Ibase = = = 13.3512 kA
3Vbase 3 × 13.8564
√ √
Vdq0,base = 2Vbase = 2 × 13.8564 = 19.5959 kV
√
Idq0,base = 2Ibase = 18.8814 kA
Vdq0,base 19.56
Zdq0,base = = = 1.0378 Ω
Idq0,base 18.8814
Zdq0,base 1.0378
Ldq0,base = = = 2.753 mH
ωbase 2 × π × 60

Siva (IIT P) EE549 72 / 102

Example 1 (contd.)

lmd 4.5695
Ifd,base = Idq0,base = × 18.8814 = 2.158 kA
lsfd 40
(lsfd = lafd ).

Sbase 3 lmd
Vfd,base = = Vdq0,base = 257.183 kV
Ifd,base 2 lsfd

Vfd,base
Zfd,base = = 119.18 Ω
Ifd,base
Zfd,base
Lfd,base = = 316.12 mH
ωbase

Siva (IIT P) EE549 73 / 102

Example 1 (contd.)

The per unit values are

1.7227
Xmd = = 1.66 p.u.
1.0378
1.6709
Xmq = = 1.61 p.u.
1.0378
4.5695
Lmd = = 1.66 p.u.
2.753
4.4321
Lmq = = 1.61 p.u.
2.753
In the per unit system, L = X .

Siva (IIT P) EE549 74 / 102

Example 1 (contd.)

lfd 576.92
Xfd = Lfd = = = 1.825 p.u
Lfd,base 316.12
0.0031
Rs = = 0.003 p.u.
1.0378
0.0715
Rfd = = 0.0006 p.u.
119.18

Siva (IIT P) EE549 75 / 102

Steady State Analysis

1 In the steady state, the rotor rotates at the synchronous speed

(ω = ωbase ).
2 Damper winding currents (I1d , I1q , I2q ) are zero.
3 In steady state, all the differential components vanish.
4 In steady state, the zero sequence component is zero.

Siva (IIT P) EE549 76 / 102

In steady state, the equations (128) - (137) become,

Vd = −Rs Id − ψq (139)
Vq = −Rs Iq + ψd (140)
Efd = Eq0 + (Xd − Xd0 )Id (141)
ψ1d = Eq0− (Xd0 − Xls )Id (142)
Ed0 = (Xq − Xq0 )Iq (143)
ψ2q = −Ed0 − (Xq0 − Xls )Iq (144)

The d-axis flux linkage equations given in (88) - (90) in the steady state
are

ψd = Xd (−Id ) + Xmd Ifd (145)

ψfd = Xmd (−Id ) + Xfd Ifd (146)
ψ1d = Xmd (−Id ) + Xmd Ifd (147)

Siva (IIT P) EE549 77 / 102

The q-axis flux linkage equations given in (91) - (93) in the steady state
are

ψq = Xq (−Iq ) (148)
ψ1q = Xmq (−Iq ) (149)
ψ2q = Xmq (−Iq ) (150)

Substituting Ifd from (146) in (145), we get

 
ψfd + Xmd Id
ψd = Xd (−Id ) + Xmd
Xfd

X 2 
Xmd (151)
= −Id Xd − md + ψfd
Xfd Xfd
= −Id Xd0 + Eq0

Siva (IIT P) EE549 78 / 102

From (144) and (150),

−Xmq Iq = −Ed0 − (Xq0 − Xls )Iq

(Xmq + Xls )(−Iq ) = −Ed0 − Xq0 Iq
(152)
Xq (−Iq ) = −Ed0 − Xq0 Iq
ψq = −Ed0 − Xq0 Iq

Substituting (151) and (152) in (139) and (140),

Vd = −Rs Id + Xq0 Iq + Ed0 (153)

Vq = −Rs Iq − Xd0 Id + Eq0 (154)

Siva (IIT P) EE549 79 / 102

Let the three phase terminal voltage in per unit be

vta = Vm cos(ωt + α)
vtb = Vm cos(ωt + α − 120◦ ) (155)
◦
vtc = Vm cos(ωt + α + 120 )

Applying the Park’s transformation, we get

Vd = Vm cos(ωt + α − θ)
(156)
Vq = Vm sin(ωt + α − θ)

θ is the angle between d-axis and the stationary axis a. Let

θ = ωt + θ0

where θ0 is rotor position at t = 0.

Siva (IIT P) EE549 80 / 102

In per unit system, peak value and RMS value are the same.

Vd = Vt cos(α − θ0 )
(157)
Vq = Vt sin(α − θ0 )

Though Vd adn Vq are scalar quantities, they can be represented as

phasors due to their trigonometric relations.

Vt ∠α − θ0 = Vd + Vq (158)

q-axis

Vt
δ
α − θ0
d-axis
Siva (IIT P) EE549 81 / 102
Let us define δ which is the angle between q-axis and Vt .

δ = 90◦ − (α − θ0 )

Vd + Vq = Vt (sin δ +  cos δ) = Vt (159)

Similarly, the stator current It lagging Vt by φ can also be expressed as

Id + Iq = It (sin(δ + φ) +  cos(δ + φ)) = It (160)

Substituting (153) and (154) in (159), we get

Vt = −Rs Id + Xq0 Iq + Ed0 + (−Rs Iq − Xd0 Id + Eq0 ) (161)

Siva (IIT P) EE549 82 / 102

Rearranging (161),

Vt = −Rs Id + Xq0 Iq + Ed0 + (−Rs Iq − Xd0 Id + Eq0 )

= −Rs (Id + Iq ) − Xq (Id + Iq ) + Xq (Id + Iq ) + Xq0 Iq + Ed0 + (−Xd0 Id + Eq0 )
= −(Rs + Xq )It + Ed0 − (Xq − Xq0 )Iq + ((Xq − Xd0 )Id + Eq0 + Xd Id − Xd Id )
= −(Rs + Xq )It + Ed0 − (Xq − Xq0 )Iq + ((Xq − Xd )Id + (Xd − Xd0 )Id + Eq0 )

From (143), Ed0 − (Xq − Xq0 )Iq = 0. From (141), Efd = Eq0 + (Xd − Xd0 )Id .

Vt = −(Rs + Xq )It + ((Xq − Xd )Id + Efd ) (162)

Xmd
Since Efd = Rfd Vfd = Xmd Ifd ,

((Xq − Xd )Id + Xmd Ifd ) = Vt + (Rs + Xq )It (163)

Siva (IIT P) EE549 83 / 102

Let the internal voltage of the generator be Eq = (Xq − Xd )Id + Xmd Ifd .

Eq = Vt + (Rs + Xq )It

(164)
Eq = Vt + (Rs + Xq )It

Eq is defined as the internal voltage behind the synchronous reactance Xq .

Equation (164) can be represented as

Xq Rs It
+

+
Eq δ Vt
−

Figure: Equivalent circuit of steady state synchronous generator

Siva (IIT P) EE549 84 / 102

 q-axis
Eq It Xq

It Rs
δ Vt
φ
It
d-axis
Figure: Phasor diagram

Siva (IIT P) EE549 85 / 102

Under no-load or open-circuit conditions, Id = Iq = 0. From (143), (140),
(141) , (143), (145), and (148) we get

Vq = Eq = Eq0 = Efd = Vt = ψd = Xmd Ifd (165)

Vd = Ed0 = ψq = 0 (166)
Therefore
Eq = Xmd Ifd (167)
The terminal voltage

Vt = Vd + Vq = Vq = Xmd Ifd (168)

Siva (IIT P) EE549 86 / 102

Steady state - No Saliency
If saliency is neglected,
Xd = Xq = Xs
where Xs is the synchronous reactance. Therefore,
Eq = Vt + (Rs + Xs )It (169)
with Xd = Xq , the magnitude of Eq is
Eq = Xmd Ifd (170)

Xs Rs It
+
+
Eq δ Vt
−
−
Figure: Steady state equivalent circuit with saliency neglected

Siva (IIT P) EE549 87 / 102

Active and Reactive Power
The terminal complex power is
∗
S = Vt It
= (Vd + Vq )(Id − Iq )
= (Vd Id + Vq Iq ) + (Vq Id − Vd Iq )

Pt = Vd Id + Vq Iq (171)
Qt = Vq Id − Vd Iq (172)
The steady state torque is given by
Te = ψd Iq − ψq Id
From (139) and (140)
Te = (Vq + Rs Iq )Iq + (Vd + Rs Id )Id
= Vd Id + Vq Iq + (Id2 + Iq2 )Rs (173)
∗
= Pt + It2 Rs = Real(Eq It )
Siva (IIT P) EE549 88 / 102
Steady state values - Procedure
For stability studies, it is necessary to find the initial steady-state values.
1 Normally, P , Q and V are specified. From the terminal conditions,
t t t
p
Pt2 + Qt2
 
−1 Pt
It = , φ = cos
Vt Vt It
Eq = Vt ∠0◦ + It ∠−φ(Rs + Xq ) (174)
2 Find δ.
q-axis
Eq It Xq

 
−1 It Xq cos φ − It Rs sin φ
δ = tan
Vt + It Rs cos φ + It Xq sin φ
δ Vt (175)
φ It Rs
It
Siva (IIT P) EE549 89 / 102
Steady state values - Procedure (contd.)

3 With δ known, the dq components of stator voltage and current are

given by

Vd = Vt sin δ (176)
Vq = Vt cos δ (177)
Id = It sin(δ + φ) (178)
Iq = It cos(δ + φ) (179)

Siva (IIT P) EE549 90 / 102

Steady state values - Procedure (contd.)

4 The remaining machine quantities are computed as follows:

ψd = Vq + Rs Iq
ψq = −(Vd + Rs Id )
ψd + Xd Id
Ifd =
Xmd
Efd = Ifd Rfd
ψfd = Xmd (−Id ) + Xfd Ifd
ψ1d = Xmd (Ifd − Id )
ψ1q = ψ2q = −Xmq Iq
Te = Pt + It2 Rs

Siva (IIT P) EE549 91 / 102

Example 2

Consider a synchronous machine serving a load with

Vt = 1.0 10◦ It = 0.5 −20◦

It has Xd = 1.2, Xq = 1.0, Xmd = 1.1 , Xd0 = 0.232 and Rs = 0.

Find δ, Vd , Vq , Id , Iq , ψd , ψq , Eq , Efd , Te , Pt .
1 Let us find δ.
  √ !
−1 It Xq cos φ − It Rs sin φ −1 0.5 × 1 × 3/2
δ = tan = tan
Vt + It Rs cos φ + It Xq sin φ 1 + 0.5 × 1 × 1/2
δ = 19.1066◦

Siva (IIT P) EE549 92 / 102

Example 2 (contd.)

Vd = Vt sin δ = 1 × sin(19.1◦ ) = 0.3273

Vq = Vt cos δ = 1 × cos(19.1◦ ) = 0.9449
Id = It sin(δ + φ) = 0.5 × sin(19.1◦ + 30◦ ) = 0.3780
Iq = It cos(δ + φ) = 0.5 × cos(19.1◦ + 30◦ ) = 0.3273
3

ψd = Vq + Rs Iq = 0.9449
ψq = −(Vd + Rs Id ) = −0.3273
ψd + Xd Id 0.9449 + 1.2 × 0.3780
Ifd = = = 1.2713
Xmd 1.1
Eq = (Xq −Xd )Id +Xmd Ifd = (1−1.2)×0.3780+1.1×1.2161 = 1.3229

Siva (IIT P) EE549 93 / 102

Example 2 (contd.)

Te = ψd Iq − ψq Id = 0.9449 × 0.3273 + 0.3273 × 0.3780 = 0.4330

∗
Pt = Real(Vt It ) = Real(1.0 10◦ × 0.5 20◦ ) = 0.4330

Te = Pt + It2 Rs
Since Rs = 0, Te = Pt .

Siva (IIT P) EE549 94 / 102

Saturation Effect

Assumptions
1 Saturation does not affect leakage inductances as the path of leakage
fluxes is mainly air.
2 Saturation affects only mutual inductances.
3 The effect of saturation on mutual inductances can be computed
through the open circuit characteristics of machines.
4 There is no saturation effect on mutual coupling between d and q
axes.
Therefore, the effects of saturation may be represented as

Xmd,sat = Ksd Xmd,unsat , Xmq,sat = Ksq Xmq,unsat

where Ksd and Ksq are the degrees of saturation in the d and q-axis,
respectively.

Siva (IIT P) EE549 95 / 102

At no load, Id = Iq = 0. From (145) and (148),

ψd = Xmd Ifd , ψq = 0 (180)

From stator voltage equations,

q
Vt = Vd2 + Vq2 = ψd (181)

It shows that in per unit the terminal voltage and the stator flux linkage
are the same.
At no load, the stator flux linkage is
q
ψt = ψd2 + ψq2 = ψd = Vt (182)

Siva (IIT P) EE549 96 / 102

 Vt (p.u.) Air gap line

1.0

Isc (p.u.)

1.0

If ,ag If ,nl If ,sc If

Figure: Open circuit characteristics of a synchronous generator

Siva (IIT P) EE549 97 / 102

if we neglect Ra ,
KIf ,sc = 1.0Xs,unsat (183)
where K is the slope of the air gap line.
To produce 1.0 p.u. terminal voltage,

KIf ,ag = 1.0 (184)

Hence,
If ,sc
Xs,unsat = (185)
If ,ag
Similarly, the saturated synchronous reactance is
If ,sc
Xs,sat = (186)
If ,nl

Let us define the short circuit ratio (SCR).

If ,nl 1
SCR = = (187)
If ,sc Xs,sat

Siva (IIT P) EE549 98 / 102

 ψ
ψt,ag

ψJ

ψt

ψlin

If ,ag If ,nl If

Figure: Open circuit characteristics

Siva (IIT P) EE549 99 / 102

 ψt ψt,ag
Xmd,unsat = = (188)
If ,ag If ,nl
ψt
Xmd,sat = (189)
If ,nl
From (188) and (189),

ψt ψt
Xmd,sat = Xmd,unsat = Xmd,unsat (190)
ψt,ag ψt + ψJ

If ψt ≤ ψlin ,
ψJ = 0; Xmd,sat = Xmd,unsat
If ψt > ψlin ,
ψJ = Asat e Bsat (ψt −ψlin )

Siva (IIT P) EE549 100 / 102

 1 Cylindrical pole rotor
Xmq,sat = Xmd,sat
2 Salient pole rotor: The effect of saturation on q-axis is neglected as
the air gap is more along q-axis.
It can be extended to loaded conditions as given below.

ψd = (Xls + Xmd )(−Id ) + Xmd Ifd + Xmd I1d = −Xls Id + ψmd

(191)
ψq = (Xls + Xmq )(−Iq ) + Xmq I1q + Xmq I2q = −Xls Iq + ψmq

Substituting the stator voltage equations in steady state given in (139)

and (140) in (191),

ψmd = ψd + Xls Id = Vq + Rs Iq + Xls Id

(192)
ψmq = ψq + Xls Iq = −Vd − Rs Id + Xls Iq

ψmd + ψmq = Vq + Rs Iq + Xls Id + (−Vd − Rs Id + Xls Iq ) (193)

Siva (IIT P) EE549 101 / 102

Let El be the voltage behind leakage reactance.

El = Vt + (Rs + Xls )It (194)

From (193) and (194),

q
ψt = 2 + ψ 2 = |E |
ψmd (195)
mq l

ψt computed from (195) is used to calculate Xmd,sat and Xmq,sat .

They have to be used for calculating sub-transient and transient
reactances and time constants.

Siva (IIT P) EE549 102 / 102