POWER SYSTEM ANALYSIS - 2 (18EE71)

MODULE - 5: Symmetrical Fault Analysis and Power System Stability

Symmetrical Fault Analysis: ZBus Building Algorithm

It is a step-by-step programmable technique which proceeds branch by branch. It has the advantage that any
modification of the network does not require complete rebuilding of Zpus. Consider that Zpus has been
another branch is now added. Then
formulated upto a certain stage and
Z=branch impedance
Zpus(old) Zpus(new)

situations is presented.
Upon adding a new branch, one of the following
1. Z is added from a new bus to the reference bus (i.e. a new branch is added and the dimension of ZRus
This is type-1 modification.
goes up by one). and the dimension of Zzus goes up
bus to old bus (i.e., a new branch is add d
2. Z is added from a new an

by one). This is type-2 modification.

new loop is formed but the dinmension of ZgUs does not
Z, connects an old bus to the reference bus (i.e., a
change. This is type-3 modification.
formed but the dimension of Zgus does not change). This is
4. Z, connects two old buses (i.e., new loop is
type-4 modification.
5. Z,connects two buses (ZpUs remains unaffected in this case). This situation
new
can be avoided by suitable
numbering of buses and from now onwards will be ignored.
Notation: i, j ->old buses; r reference bus; k > new bus.
Type-1 Modification:
Fig. 1 shows a passive (linear) n-bus network in which branch with
impedance Z, is added to the newbus k and the reference bus r.
Now Passive
Ve Zplk linear
Zki Zik = 0, i = 1 , 2 , . . ,n
n-bus
Zkk= Zb network.
Hence.
01
Zpus(old)
Zpus(new) =
-(1) Fig. 1 Type-1 Modification

0 Z
n
Type-2 Modification:
Z is added from new bus k to the old bus j as in Fig. 2. It follows from his Passive
figure that . + linear
n-bus
VZh +V network
V
Vk =Zpl +Zj1l +Zj2l2 + +/2i5(; + Ik) ++Zjn'n k
r.
Rearranging, Vy = Zjal + Zj2l2 t +Zijl ++Zjn!n + (Z, + Zp) Ik
Fig. 2 Type-2 Modification
Consequently,

Zpus(old)
Zpus(new)= (2) (2

(2 +Z%)|
Basavarajappa S R, E & E, BIET, Davanagere
 Type-3 Modification:
Z connects an old bus (j) to the reference bus (r) as in Fig. 3.
This case follows from Fig. 2 by connecting bus k to the reference bus n
r,
i.e, by setting Vk = 0. Thus.
Passive
linear
V2 n-bus
ZBus(old) network
- (3) k-
=0
Z1 Z2 Zjn (Z t+ Zb)
Fig. 3 Type-3 Modification
Eliminate lk in the set of equations contained in the matrix
operation (3)
-0 =
Zj1l
t+ Zjal + +
***
Zjnln + (Zj +Z%) Ik
1

Now. V Zi1l +Zi2l2 t+Zin!n

=

Substituting Eq. (4) in Eq. (5) ++ Zijlk--

1)
k Z + Z4 tZ2la t +

(5)
Zjnl,) - - - (4)

Eq. (6), can be written in matrix form as

1 L2i \
Aaus(new) Zpus (old) =

(Z1 +Z») 71 Z2. Zg

-

(7)

Type-4 Modification:
Zryj
Zp connects two old buses as in Fig. 4. Equations can be written n
for all the network buses.
as follows
1
+Ik
Passive
VZi1l +Zi2l2 t+Z(li +k) +Zij(l-Ik) ..+Zinln- (8) linear
n-bus
Similar equations follow for other buses. network
The voltages of the buses i and j are,
V- V, Zplk =
*

V Zplk +V--(9)
Or

Zi +Zi2lz + Z,(1, + I) +Z,,(--) ++Zjn!n Fig. 4 Type-4 Modification

=
Zplk + Z,1'1 + Zi2l2 ++Zu(d, + lh) + Zij(I l) ..-+ Zin!n
-

Rearranging,
0 =
(Z1-Zj1) t+ (Zu-Z,)4, + (Z4-Z21), + + (Zin -Zjn)la + (2% +Zu +Z -Z

Zu)k -
(10)
Collecting equations similar to Eq. (8) and Eq. (10), we can write

(Z-Z)
Z21Za)
Zpus (old)
(11)
(Zn-Zp)
l(Z1-Z1) (Zi2-Zp) (in-Zn) (zh + Z +Z -22,)
Basavarajappa S R, E & E, BIET, Davanagere 2
Eliminate lk in Eq. (1 1). (similar to
type-2 modification) it follows that
(Z-Z1)|
ZBUS (new) = Zpus(old) -
(2a)l(z.- Z). (Zn -Zn)--~(12)
(Zh+Z 4+Z-2.24y}
the l(zZmZ)
use ot four relationships Eqs. (1), (2), (7) and (12), bus impedance matrix be built by a
SLcp-o-stepP procedure (bringing in one branch at a time). This procedure being a mechanical one can be easily
can

computerized.
When the network undergoes changes, the modification procedures can be employed to revise the bus
impedance matrix of the network. The opening a line (Zi) is equivalent to adding a branch in parallel to it with

impedance -Zj
0.1 0.
Problem 1: For the 3-bus system shown in Fig. 5, build Zgus matrix.
Solution: No. of buses, n = 3
Step-1: Add branch zr = 0,25= Zb (from bus 1 (new) to ref. bus r)
Type-1 modification, k= 1,Zkk Zb
0.25
Zpus [ 2 a l = [Z11l [0.25-(9. 0.25
Ref. bus-T
Step-2: Add branch z2 = 0.1 Z (from bus 2 (new) to bus I (old)
Fig. 5 Three-bus system.
[ZBUs(old) 1
Type-2 modification,kgj-! Zgus =

(Zj +Z)|
bBUs(otd) = [0.25]
(from Eq.))
0.25 + 0.1 0.35
Zjj +Z, =Z11 t+Zb=
=

Z1j=Z11 =
0.25=Zj1
0.25 0.29
.
ZpUslo.25 0.35 l21 z-- (i)
Step-3: Add branch zZgi = 0.1= Zp (from bus 3 (new) to bus I (old)

ne2 |ZBUs(old) Ly
Type-2 modification, k = 3,j = 1, ZpUs
22

21Li1 2|(%+2%) 2 2 24
0.25 0.251
(from Eq.i))
ZBUs(otd) lo.25 0.35 Z2Z21
Zj = Z11= 0.25 = Zj1, = 0.25 =Zj2 Zij + 2 = Z11 + Z = 0.25+0.1 = 0.35

0.25 0.25 0.251 Z12 213

Z BUs 0.25 C35 0.25 Z21 Z22 Z23 - (iii)

lo.25 0.25 0.35 Z31 Z32 Z33

Step-4: Add branch 2jr = 0.25 = Z» (from bus 2 (old) to ref. bus r)

Type-3 nodification, j 3 ZsusZpus(old) - 2 2 | 1 Z 2p]

0.25 0.25 0.25]

BUS(otd) = 0.25 0.35 0.25|
(from Eq.(ii)
Lo.25 0.25 0.35
Z =Z12 0.25 = Zj1. Z2j = Z22 =0.35 = Zj2 Z3 Z32 = 0.25 = Zj3
i j t Zp = Z22 + Zb = 0.35 +0.25 = 0.6

I2
[0.25 0.25 0.251 1 [0.25 23 [0.1458 0.1042 0.1458] [Z1 Z12 Z131
:
Zgus =0.25 0.35 0.25 0.35| 0.25 0.35 0.25] = 0.1042 0.1458 0.1042 Z22 Z23-- (iv)
l0.25 0.25 0.35 6 Lo.25 Lo.1458 0.1042 0.24581 LZ31 Z32 33

Basavarajappa S R, E & E, BIET, Davanagere

3
 bus 2 (old) to bus 3 (old), Type-4 modification,:.
Step-5: Add branch zi = 0.1 =
Zh (from
(Z2 Za1 ( 4 1 -Zq1) (Zz - Zj2) (Zja -2p)]
ZBUS Zpus(old) (Z +Zu+ Z
0.1458 0.1042 0.1458
Z4))|(Z,-Za)
BUs(old) =
|0.1042 0.1458 0.1042| (from Eq.(iv))
Lo.1458 0.1042 0.2458
Z Z =Z12 -Z1 =0.1042 0.1458 = -0.0416 = Z1-Z1
Z2i-Z2j = Z22-Z23 = 0.1458-0.1042 = 0.0416 = Z2 - Zi2
Zi-Zgj Z32-Z33 = 0.1042 0.2458 = -0.1416 = Z3-Z3
Z +Zji +Z,j -2Zj = Zp + Z22 + Zg3 - 2 Zz3 = 0.1 +0.1458+ 0.2458 (2 x0.1042) =0.2832

0.1458 0.1042 0.1458] -0.0416]

ZBUS 0.1042 0.1458 0.10420.2832 .1 0.0416 -0.0416
2832 |-0.1416)
0.0416 0.1416]
Lo.1458 0.1042 0.2458
0.1397 0.1103 0.12501
Bus= |0.1103 0.1397 0.1250 - (v)
L0.1250 0.1250 0.1750J

Problem 2: 0.1397 0.1103 0.1250]

For the 3-bus system shown in Fig. 5, ZBUS matrix= 0.1103 0.1397
0.1250. The line between
the buses 3 and 2 with l0.1250 0.1250 0.17501
impedance 0.1 is removed by simultaneous opening of the breakers at both ends of the
lines, Determine the new Zpus matrix.

Solution: Opening a line (line 3-2). This is equivalent to connecting an

bus 2 (old) i.e. type-4 modification, i 3, j =2 =
impedance Zb =
-0.1 between bus 3 (old) and
(21-Z11)
ZguS Zgus (old)2p(ZbZy142-22,5aZa)||{Zi1 -2,1) (Ziz - Zj2) (13 -Zja)]
L(Z31-Z3)
0.1397 0.1103 0.1250
ZBUS(old) = 0.1103 0.1397 0.1250

Lo.12500.1250 0.1750
ZiZ1j Zi3- Z12 0.1250-0.1103 0.0147 Zi1 -Zi1
= =
= =

Z2i-Z2j = Z23- Z22 0.1250-0.1397 =-0.0147 Zi2 -

=

Zi2
Zgi-Z3 Z33- Z32 0.1750 --0.1250 0.0500 Z3-Za
= =
= =

Zp +Zi +Zj - 2Zij Zh +Z33

+Z22-2 Z32 -0.1 +0.1750 +0.1397
=
=

(2 x
0.1250) =
-0.0353
0.1397 0.1103 0.12501
ZBus 0.1103 0.1397 1 [0.0147 1
0.1250
lo.1250 0.1250 0.1750 -0.0353 -0.0147|
0.0500 J[0.0147 0.0147 0.0500]
0.1458 0.1042 0.1458]
2BUs= |0.1042 0.1458 0.1042 (same
Lo.1458 0.1042 0.2458
as
in step 4 in Pr. 1)
Problem 3: Construct the bus
shown in Fig. 6. impedance matrix for the network
j0.2 j0.8 j0.4

j0.4 j0.4

Fig. 6 Three-bus system

Basavarajappa S R, E &E, BIET, Davanagere
4
 olution: Ref. bus-r J0.4
Step-1: Add branch Zir = j0.2 = Z (from bus I (new) to ref. bus r) j0.2
Type-1 modification, k = 1, Zkk
ZBus = [Zl = [211] = j0.2] - - - ()
= Zb
O j0.8
J0.4
j0.4
Step-2: Add branch zZ2 = j0.8 = Zy (from bus 2 (new) to bus 1 (old))

Type-2 modification, k =2,j =

1, Zgus
[ZBUs(old) Z1 Linear graph
Z T(Z+Z)
ZBuSCold) = j0.2] (from Eq.i))

Zj =Z11 j 0 . 2 = Zja,
Zj + Z Z11 + Zp j0.2 + j0.8 =j1

ZBUs j0.2 0=lZas

j0.2 j1 Z22
Step-3: Add branch z3i = j0.4 Zh (from bus 3 (new) to bus I (old))
|ZBus(old)
Type-2 modification, k= 3,j = 1, Zpus

ZZj1 2 (Z+Z)
j0.2 j0.21 (from Eq.i))
ZpUs(old) j0.2 j1
Zj=Z11 j0.2 = Zj1. ZajZ21= j0.2 =Zj2. Zij +Z= Z11 +Z =j0.2 +j0.4 =j0.6
j0.2 j0.2 j0.2 Z11 Z12 Z13
Z21 Z22 Z23 - (iii)
ZBUS j0.2 j1 j0.2
Lj0.2 j0.2 j0.6 Z31 Z32 Z33

bus 2 (old) ref. bus r)

Step-4: Add branch zir =
j0.4 =
Z (from to

(L
Type-3 modification, j =2, Zgus Zpus (ola) (2 +Z6)7Z2Z11 Z2 Zja
j0.2 j0.2 j0.2
ZBUS(old) = |j0.2 j1 j0.2 (from [Link]))
Ljo.2 j0.2 j0.6 Z3 Z32 = j0.2 = Zj3
Z Z 2 = j0.2 = Z j . Z 2 Z 2 2 = j l = Zy2

Zyj +ZZ22 + Zp = j1 + j0.4 =j1.4

j0.2 j0.2 j0.2 2 j0.1714 j0.0571 j0.1714|

ZgUs=j0.2 1 j0.2 1 0 . 2 j1 jo.2] = |j0.0571 j0.2857 j0.0571=Z21 Zz2 Z2a - (iv)
j0.1714 j0.0571 j0.5714 l31 Zs2 Z33
Step-5: Add branch Zi = j0.4 = Z, (from bus 2 (old) to bus 3 (old). Type-4 modification, i = 2, j =3

[(Z1Z1)| 1
ZBUSZgus(ola)(2+
(z +Zu +Z- 22ij)| (ZZ) (71-Z1) (Za-Zq2) (Z-Za)]
Zq +2-22,a
jo.1714 j0.0571 j0.17147
ZBUS(old) = {j0.0571 j0.2857 j0.0571
(from Eq.(iv))
jo.1714 j0.0571 jo.5714]
Zi Zj =
Z12-Z13 =j0.0571 -j0.1714 -j0.1143 Zi1Z1 = =

Z2i-Z2j Z22- Z23 = j0.2857-j0.0S71 =j0.2286 Z2 - Zj2 =

Zgi Z3j =
Z32 - Z33 =j0.0571-j0.5714 -j0.5143 Z3 - Zj3 =

Zp + Zi + 2 - 2 2 = Z» + Z22 + Z 3 2 723 = j0.4 + j0.2857 + jo.5714 - (2 x j0.0571) = j1.1429

Basavarajappa S R, e & E, BIET, Davanagere

 jo.1714 j0.0571 j0.17141 1 0.1143
jo.0571 j0.2857 j0.2286 |[-j0.1143 j0.2286 - j0.5143
ZBus
jo.1714 j0.0571 jo.5714] 0.05711.1429-j0.s143)
jo.1600 jo.0800 j0.1200
.
Zgus =
j0.0800 j0.2400 j0.1600 -- (v)
ljo.1200 jo.1600 j0.3400
Problem4: Construct the bus impedance matrix for the network j0.5
shown in Fig. 7.
j0.3 j0.2
Solution:
Step-1: Add branch Zir = j0.3 Z (from bus I (new) to ref. bus r)
Type-1 modification, k=1, Zkk = Zb j0.4
Zpus [Zk) = [Z11l = [j0.3] - - ( ) j0.56 jO.35

Step-2: Add branch Z2j =

j0.5 =
Z (from bus 2 (new) to bus I (old)) Fig. 7 Three-bus system

Type-2 modification, k 2,j =

1, Zgus ZBUS(old) Z
j0.3 Ref. bus-r
(Z +Zb) j0.2
ZBUS(old)fo.3] (from Eq.(i) O
Z1 Z1 = j0.3 Zi1=
Zij +Z Z Z11 +Zp =j0.3 +j0.5 =j0.8 j0.5
jo.4 j0.35
j0.3 j0.31
ZBUs 0.3 j0.8 |-C 2--i ) j0.56
Linear graph
Step-3: Add branch Z3j =
j0.4 =
Zp (from bus 3 (new) to bus 1 (old))

Type-2 modification, k
ZBUS(otd)
=3,j=1, ZBUs
Z Zy2 + Z)
j0.3
BUSoia)o.3 j0.3 (from [Link])
j0.8) (from Eq.(ii))
Z1Z11 =j0.3 =Zj1, Z2j Z21 j0.3 = =
Zj2, Zij + Z= Z11 + Zb =
j0.3 + j0.4 =
j0.7
j0.3 j0.3 j0.31 [Z11 Z12 Z13
ZBUS=j0.3
Ljo.3 j0.3 j0.7
j0.8 j0.3=21
Z31
Z22
Z32
Z23 (iüi)
L33
Step-4: Add branch Zi| =j0.56 =
Zp (from bus 1 (old) to bus 3
(old)). Type-4 modification, i =

1,j=3
ZBUS ZBus(old)
[Z11-21) 1

j0.3 j0.3 j0.3

(Zh+Zi+Zjj- Z4ij)|(2Za)|
Z, +Z +Z-22. (Z3 -Zs) [(7Z11-Z) (2 -Zj2) (2 -Z)]
BUS(old) =|j0.3 j0.8 j0.3 (from [Link])
j0.3 j0.3 j0.7

Zi Zi Z11- Z13 =j0.3-j0.3 0 Zi1-Z1

=
= =

Z2i-Z2j =Z21 - Z23 =j0.3-j0.3 0 Z2-Zi2 = =

Z3i Z3 Z31 - Z3 =j0.3- j0.7 -j0.4

=
=
Zi3-Zi3
Z + Zii tZj- 2Zij= Z + Z11 + Z33-2 Z13 =j0.S6
=

+j0.3 +j0.7 (2 xj0.3) =j0.96

j0.3 j0.3 j0.3
ZBUS j0.3 j0.8 j0.3.
1
[O 0-j0.4]
j0.3 j0.3 j0.7]
j0.96-j0.4
Basavarajappa S R, E & E, BIET, Davanagere
6
 j0.3000 j0.3000 j0.30001
BUs = |j0.3000 j0.8000 j0.3000| .- (iv)
j0.3000 j0.3000 j0.5333]
Step-5: Add branch zir =
j0.2 Zh (from bus 2 (old) to ref. busr)
[21
Type-3 modification. j 2, Zaus =
Zovs(old) -tt 4 Zp
j0.3000 j0.3000 j0.3000
Z3
BUS(old |j0.3000 j0.8000 j0.3000 (from Eq.(iv))
j0.3000 j0.3000 j0.5333
Z1=Z12 j0.3 = Zj1. Z2 Z3Z32
z2 = j0.8 = Zj2. = j0.3 = Zj3

Zij+ Z Z22 + Z = j0.8 + j0.2 =j1

j0.3000 j0.3000 j0.3000 jo.3

BUS |j0.3000 j0.8000 j0.3000 j0.8 0.3 j0.8 j0.3]
ljo.3000 j0.3000 j0.5333
jo.2100 j0.0600 j0.2100 11 Z2 3
jo.0600 j0.1600 j0.0600 21 Z22 223 (v)
Lj0.2100 j0.0600 j0.4433 Z31 Z32 Z33
i 2, j 3
(old) to bus 3 (old)), Type-4 modification,
= =

Step-6: Add branch zii =j0.35 Z (from bus 2

1
(ZZ1)
Zgus =Zpus(old) --(Z a-Za) -41) (Za 22) (4s -2,)]
-

j0.2100
+Z+ZjZi)|(Zgi-Zsi)
jo.0600 j0.21001
ZBUS(old) =
|j0.0600 j0.1600 j0.0600 (from Eq.(v))
Ljo.2100 j0.0600 j0.4433
ZiZ1 = Z12 - Z 1 3 = j0.06 -j0.21 = -j0.15 = Zi1- Z1

Z2i-Z2j = Z22 - Z23 = j0.16-j0.06 = j 0 . 1 =- Zi2 - Zi2

Z3i-Z3j = Z 3 2 - Z33= j 0 . 0 6- j 0 . 4 4 3 3 = -j0.3833 = Zi3-Zi3

Zp + Zii + Zij - 27ij = Zh + Z22 + Zg3-2Z23 = j0.35 + j0.16 + j0.4433 ( 2 xj0.06) =j0.8333

jo.2100 j0.0600 j0.2100 1 0.15 1.

Zpus=j0.0600 j0.1600 j0.0600 j0.1|j0.15 j0.1 -j0.3833]
jo.2100 jo.0600 j0.4433 j0.8333j0.3833
j0.1830 jo.0780 j0.14101
BUS j0.0780 j0.1480 j0.1060 - (vi)
Ljo.1410 j0.1060 j0.2670
j0.1830 j0.0780 j0.1410
Problem 5: For the 3-bus system shown in Fig. 7, ZBus matrix = jo.0780 jo.1480 jo.1060. The line between
ljo.1410 j0.1060 j0.2670]
the buses 1 and 3 with impedance j0.56 is removed by simultaneous opening of the breakers at both ends of
the lines, Determine the new ZBUs matrix.

Solution: Opening a line (line 1-3). This is equivalent to connecting an impedance Z = -j0.56 between bus 1 (old)
and bus 3 (old) i.e. type-4 modification, i = 1,j =3

(Z1-Z1)
1
ZRus(old) (o+Zu*Zij
(zZ-22, ai)%-21) (Zi2 -Zj2) (Zis -Zi3)]
(Zat -Zsj)
j0.1830 j0.0780 j0.14101
ZBUStotd) = |j0.0780 j0.1480 j0.1060|
jo.1410 j0.1060 j0.2670
Basavarajappa S R, E &E, BIET, Davanagere
 Zi-Z1 = Z1 - Z13 = j0.1830-j0.1410 =j0.0420 Z,1-Z1
Z2i - Z2j = Z21 - Z23 = j0.0780 j0.1060 = -j0.0280 Z,2 - Z2

Z3i-Z3 Z31-Z33 =j0.1410- j0.2670

=
=
-j0.1260 =
Z3 -Z3
Z+Z + Z - 27 = Zp + Z1 +Z33- 2 Z13 = -j0.56 + j0.1830 + j0.2670 - (2 x j o . 1 4 1 0 ) = -j0.3920

jo.1830 j0.0780 j0.14101 j0.0420

ZBUS 0.0780 j0.1480 j0.1060| j0.0280 o.0420 j0.0280 j0.1260]
jo.1410 jo.1060 jo.2670] -j0.3920-jo.1260
j0.1875 j0.0750 j0.1275
ZBUS =j0.0750 j0.1500 j0.1150
j0.1275 j0.1150 j0.3075

ZBUS Algorithm for Short Circuit Studies:

A balanced three-phase fault is defined as the simultaneous
short circuit across the three phases. It occurs infrequently,
but it is the most severe type of fault. The short circuit
calculations carried out for simple systems whose passive System Gen n
networks can be easily reduced. The transient reactance of
en
synchronous generator under short circuit is time varying Gen r
quantity and is used in fault calculations. In order to apply the Gen 2
four
steps short circuit computation developed to large
of
systems, it is necessary to evolve a systematie general
algorithm so that a digital computer can be used. Consider an
Fig. 8 n-bus system under steady load
n-bus system shown schematically in Fig. 8 operating at
steady load.
Step-1: Short cireuit computation is to obtain prefault voltages at all buses and currents inall lines through a
load flow study. The prefault bus voltage vector is

v
VBUS --(13)

Let us assume that the rt" bus is faulted through a fault impedance Z. The postfault bus voltage vector will
be given by

VBUs =VSus +AV - (14)

where AV is the vector of changes in bus voltages
caused by the fault.
n

Step-2: Draw the passive Thevenin network of the system System

with generators replaced by transient reactances with their
emfsshorted, as shown in F1g. 9.
2 Adn
Xa1
Step-3: Excite the passive Thevenin network with -V° in
series with Z (Fig.9). The vector AV comprises the bus
voltages of this network. Now, AV ZBUS ] = -
-
(15)
Fig. 9 Network for computing changes
where in bus voltages caused by the fault
Lin
BUS =
bus impedance matrix of the passive Thevenin network-- (16)
|4n1 Lnn
J = bus current injection vector

Basavarajappa S R, E & E, BIET, Davanagere 8

Since the network is injected with current -1 only at the rt bus, then

=- -(17)

Substituting Eq. (17) in Eq. (15), we have for the rth bus, AV, =-2,rr 1 - - -(18)
Step-4: Voltage at the rt" bus under fault is
---(19)
= +A" =V-Zrr
However, this voltage must equal, V = Z I - - (20)

From Eqs.(18) and (19). z1f = v - Z or 1 = -(21)

At the ith bus (from Eqs. (15) and (17)

AV=-Zir 1
= Vi-Zir , i= 1,2,., n --- (22)

Substitutingfor from Eq. (21), we have =V Zrrt2 - - (23)

Fori r in Eq.(23),

+Z V---(24)
In the above relationship V° s, the prefaultbus voltages are assumed to be known from a load flow study.
Zpus matrix of the short-circuit study network of Fig. 9 can be obtained by the inversion ofits Yeus matrix or
Zpus building algorithm. It should be observed here that the SC study network of Fig. 9 is diferent from the
coesponding load flow study network by the fact that the shunt branches corresponding to the generator
reactances do not appear in the load flow study network. Further, in formulating the SC study network, the
load impedances are ignored, these being very much larger than the impedances of lines and generators. Of
course synchronous motors must be included in Zgus formulation for the SC study.
Postfault currents in lines are given by

000 Bus-i
Y ( v ) - - (25) 00 Bus-i

For calculation of postfault generator current,

examine Figs. 10(a) and 10(b).
From the load flow study (Fig. 10(a))
Prefault generator output = Pgi t jQGi

(a) (b)
Poi-jQGi - (26) Fig. 10 System networks

E =V+jXGil->(27)
From the SC study i s obtained. It then follows from Fig. 10(b) that

jXG
- -- (28)
Basavarajappa S R, E& E, BIET, Davanagere
 Power System Stability:
In most practical systems. there are still more than two machines to be considered from the point of
vie
machine by a numerical technique on the d:
System stability. Therefore, solve the swing equation of each
computer. Even in the case of a single machine connected to infinite bus bar. the critical clearing time ca
be obtained from equal area criterion and to make this calculation numerically through swing [Link]
The
are two methods available for the solution of the swing equation: Point-by-Point method and Runge-Kuta

method.

Mumerical Solution of Swing Equation by Point-by-Point method

The point-by-point method of solution of swing equation is a conventional and approximatemethod. First
illustrate this method for one machine connected to infinite bus bar. The procedure is, however. general and
can be applied to every machine of a multimachine system.
Consider the swing equation

ddt2 Pm-Pmax sin6) =a --(29) M

GH H
where M = m o r in pusystem, M = = angular momentum or inertia constant.

P
The solution 6(t) is obtained at
discrete interyals
of time with interval spread of At uniform Pein-2) Disete soution

throughout. Accelerating _power and change in Poo-1) Continuous solution

speed which are continuous functions of time are
discretized as below:
Poti)
1. The accelerating power Pa computed at the
beginning of an interval is assumed to remain n-2 n-1
constant from the middle of the
preceding Continuous solution
interval to the middle of the interval being Discrete sokution
considered as shown in Fig. 11. Wa-1/2
2. The angular rotor
velocity w =dQ/dt (over Wn-3/2
and above synchronous
velocity s) is wn-i/7n-v2
assumed constant throughout any interval, at
t h e value computed for middle of the interval n-2 -3/2 n-1 n-1/2 n

as shown in Fig. 11.

In Fig. 11, the numbering on t/At axis pertains
to the end of intervals. At the end of (n -1)th ***(*************

interva! the acceleration power is Abn

****

Patn-1) PmPmax sinðn-1---(30) En-2

where -1 has been previously calculated. The -2 n-1
change in velocity (w =d6/dt) caused by the
Pa(n-1) assumed constant over At from (n- Fig.11 Point-by-point solution of swing equation
3/2)to (n-1/2) is
n-1/2) -dn-3/2)= {At/M)Pa(n-1)---(31)
The change in ô during (n
1)h interval is A6,-1 8n-1- On-2
-

and during the n" interval

A6, ôn ôn-1 At wn-1/2 - (33) on-3/2
=
At
-
=
=
(32)
Subtracting Eq. (32) from Ey. (33) and using Eq. (31), we
get,

A6, (At)2
A8,n-1 + M Pa(n-1) - -
(34)
Using this we can write, ô, =
ôn-1 t A8,n- (35)
Basavarajappa S R, E & E, BIET, Davanagere
10
 The process of computation is now repeated to obtain Patn), 48n+1 and 6,41. The time solution in
diserete form is thus carried out over the desired length of time, normally 0.5 sec. Continuous form of solution
is obtained by drawing a smooth curve through discrete values as shown in Fig. 11. Greater accuracy of
solution can be achieved by reducing the time duration of intervals.
The occuTence or removal
of a fault or initiation of any switching event causes a discontinuity in
accelerating power Pa. If such a discontinuity occurs at the beginning of an interval, then the average ot the
values of Pa before and after the discontinuity must be used. Thus, in computing the increment of angle
occurring during the first interval after a fault is applied at t = 0, Eq. (34) becomes

(At)Pal(ot)
2
-- (36)
M
where Pa(o*) Is the accelerating power immediately after occurrence of fault. Immediately before the fault the
system is in steady state, so that Pa(o-) = 0 and do is a known value. If the fault is cleared at the beginning ol
the nh interval. in calculation för this interval one should use for Pafn-) thevaluelPan-1) + Pan-1*|
where Pa(n-1)- is the accelerating power immediately before clearing and Pa(n-1)+ is that immediately after
needed. The
If the discontinuity occurs at the middle of an interval, no special procedure is
clearing the fault.
increment of angle during such an interval is calculated, as usual, from the value of Pa at the beginning of
the interval.

Problem 1: A 20 MVA, 50 Hz generator delivers 18 MW over a double circuit line to an infinite bus. The
transient reactance is Xa
generator has kinetic energy of 2.52 MJ/MVA at rated speed. The generator
=

0.35 pu. Each transmission circuit has R = 0 and a reactance of 0.2 pu on a 20 MVA base. |E"| = 1.1 pu and

infinite bus voltage V = 1.020°. A three-phase short circuit occurs at the mid-point of one of the transmission
lines. Plot swing curves with fault cleared by simultaneous opening of breakers at both ends of the line at 2.5
cycles and 6.25cycles after the occurrence of fault. Also plot the swing curve over the period of 0.5 sec if the
fault is sustained.

Solution: Before we can apply the step-by-step method. we need to calculate the inertia constant
M and the power angle equations under pre-fault and post-fault conditions.
Base MVA =20, f= 50 Hz, H =2.52 MJ/MVA, X4 = 0.35 pu, Xi = 0.2 pu, |E'| = 1.1 pu, |VI = 1.0 pu
H 2.52
Inertia constant, M(pu) =-
180f 180 x 50
= 2.8 x
10 s2/ele. deg
I. Pre-fault: X, =
Xa +=0.35+ 0.45 pu
PelPmax1 Sinð =
E|VI sind = 1.1x 1
-sind = 2.44 sind - - + (i)
X 18
0.45
Prefault powertransfer = 2 0 . 9 pu = Pm = mechanical input

Initial power angle is given by 2.44 sindo = 0.9 00= 21.64°

II. During fault: A positive sequence j0.2
reactance diagram is shown in Fig. 12(a). j0.35
Converting star to delta, we obtain the (Y to A)
network of Fig. 12(b), in which E') j0.1 j0.1

0.35x 0.1+0.2 x 0.1 +0.35x 0.2

X 0.1 (a)
= 1.25 pu
(b)
Fig. 12 Positive sequence diagrams

Sins=.I x 1
Pe11Pmax 11 Sinð =
2 s i n ð = 0.88 sind
1.2 -(ii)

Basavarajappa S R, E & E, BIET, Davanagere

11
I Post-fault: With the faulted line switched off
XuI = K+X = 0.35+0.2 = 0.55 pu

Pe 111Pmar iil Sinð =. ET nh = sinó = 2 sinó - -

(ii)
0.55

Let us choose At 0.05 The recursive

=
sec.
relationships for step-by-step Swing curve calculation are
follows: Pa(n-1) Pm-Prmaz Sinó,-i---(iv)
.

46, =
A5,-:+(a0) Pan-1) - (v)
6,= 6n-1 + A6,-- (vi)
Since there is a
discontinuity in Pe and hence in [Link] average value of Pa must be used for the first
Pa(0)0pu and Pa(o") = 0.9 0.88 sin 21.64* = 0.576 pu
interval

Sustained Fault:
PalOmg) 0.288 pu
Calculations are carried out in Table 1 in accordance with the
The second column of the table shows
recursive relationship (iv). (v) and (vi) above.
first column. Pmar in the case
Pmar the maximum power that can be transferred át time t
given in the
of a sustained fault undergoes à sudden change at t 0 and remains constant
thereafter. The procedure of
sec.
calculations is illustrated below by calculating the row to t 0.1 coresponding =

(0.1 sec) =31.59, Prmax= 0.88 pu, sinô (0.1sec) = 0.524

P(0.1sec) =
PmazSinó (0.1sec) 0.88 x 0.524
=
=
0.461 pu
Pa(0.1sec) =
0.9- 0.461
0.439 pu =

P(0.1sec) =280.050.439 =3.92

A6(0.1 sec) 46(0.05 sec) +

=

P(0.1sec) - 7.38° + 3.92

6(0.15 sec) ô(0.1 sec) + A6(0.1 sec) 31.59° + 11.30°
=
=
11.30°
=

Table for sustained fault =

42.89
Pmaz PPmarSinQ| P 0.9-P
sEc
pu
sino
pu
P A6
pu
2.44 0.368 0.9 0.0
8.929 P deg deg deg
0.88 0.368 0.324 21.6
0.576
0.368 0.288 21.64
0.05 0.88 0.410 0.361
2.57
21.64
0.539 4.81
0.10 0.88 0.524 0.461 0.439 7.38 24.21
0.15 0.88 0.680
0.598 0.301
3.92
2.68
11.30 31.59
0.20 0.88 0.837
0.736
13.98 42.89
0.25 0.88 0.163 1.45 15.43
0.953 0.838 0.060 56.87
0.30 0.88 0.999 0.55 15.98 .30
0.35 0.879 0.021 0.18
0.88
0.968 0.852 16.16 88.28
0.40 0.88 0.048 0.426 16.58 104.44
0.856 0.754 0.145 1.30
0.45 0.88 0.657 17.88 121.02
0.50 0.578 0.321 2.87
0.88 20.75 138.90
159.65
6(t) for sustained fault as calculated in Table 1 is
is unstable. plotted in Fig. 13 from which it is obvious that
the system

Fauli Cleared in 2.5

cycles:
Time to
clear fault =
0.05 sec
Basavarajappa S R, E & E, BIET, Davanagere
12
 suddenly changes from 0.88 at t = 0.05 to 2.0 at
05. Since the discontinuity occurs at the beginning 160
of an interval, the average value of the Pa will be
140
assumed to remain constant from 0.025 sec to 0.075
sec. The rest of the procedure is the same and 120
Substained fault

complete calculations are shown in Table 2. The

swing curve is plotted in Fig. 13 from which we find O0

that the generator undergoes a maximum swing of

37.5° but is stable as ô finally begins to decrease.
Fault cleared
at 6.25 cycdes

Fault Cleared in 6.25 cycles:

.25 40
Time to clearfault =- = 0.125 sec
50
Since the discontinuity now lie in the middle of an 20 . auf cleared
at 2.5 cycles
interval, no special procedure is necessary, as in
deriving Eqs. (iv) - (vi) discontinuity is assumed to .10.2 o.3 o.4 0.5 06
t(s)
occur in the middle of the time interval. The swing 12.10 for a sustained fault and
curve as calculated in Table 3 is also plotted in Fig. Fig.13 Swing curves for Example
2.5 and 6.25 cycls
for clearing in
13. It is observed that the system is stable with a
maximum swing of 52.5° which is much larger than that in the case of 2.5 cycle
for progressively greater
(To find the critical clearing time, swing curves can be obtained, similarly, first find the
bound. In this example, however, we can
clearing time till the torque angle 8 increases without
critical clearing angle (6cr) using Eq.
Pmmax-Ôo) +Pmaxi COSOo +Pmaxin COSÖmax
cosocr maxill Pmaxi1
o = 21.64° = 0.3777 rad,
max 180°-sin Pm 180°-sin =153.2563° = 2.6748 rad
maxll
_0.9 (2.6748-0.3777)-(0.88xcos (21.64))+(2 xcos (153.2563)) -0.47907
cosocr 2-0.88

Ocr=Cos-1(-0.47907) 118.6246° =

and then read the critical clearing time from the swing curve corresponding to the sustained fault case.
At Ser 118.6246°, the critical clearing time ter = 0.38 sec). .

Table 2for fault cleared at 2.5 cycles

Pmax sinð
PePmaxsinð | Pa = 0.9-Pe (ALP=
sec pu pu pu 8.929 Pa deg deg deg
0 2.44 0.368 0.9 0.0 21.64
0.88 0.368 0.324 0.576 21.64
Oavg 0.368 0.288 2.570 2.57 21.64
0.05 .88 0.410 0.360 0.540 ---

24.21
0.05 2.00 0.410 0.820 0.080 24.21
| 0.05 avg 0.310 2..767 5.33 24.21
0.10 2.00 0.493 0.986 -0.086 -0.767 4.5 29.54
15 2.00 0.560 .120 -0.220 -1.960 2.60 34.10
0.20 2.00 0.597 .190 -0.290 -2.580 0.02 36.70
0.25 2.00 0.597 1.190 -0.290 -2.580 2.56 37.72
0.30 2.00 0.561 1.120 -0.220 -1.960 -4.52 34.16
0.35 2.00 0.494 0.989 -0.089 -0.790 5.31 29.64
0.40 2.00 0.410 0.820 0.080 0.710 -4.60 24.33
0.45 2.00 0.337 0.675 0.225 2.000 -2.60 19.73
0.50 17.13
Basavarajappa S R, E &E, BIET, Davanagere 13
Table 3 for fault cleared at 6.25 cycles
Pe=Pmaxsinð Pa 0.9-Pe
= Cae Pa =

Pmax sino
pu pu 8.929 Pa
M
deg deg deg
sec pu
0 2.44 0.368 0.9 0.0 21.64
0.324 0.576 21.64
0.88 0.368
Oave 0.368 0.288 2.570 2.57 21.6
0.88 0.410 0.361 0.539 4.810 7.38 24.21
0.05
3.920 11.30 31.59
0.10 0.88 0.524 0.461 0.439
0.15 2.00 0.680 1.360 -4.460 -4.100
-5.660
7.20
1.54
-42.89
50.09
0.20 2.00 0.767 1.530 -0.630
2.00 0.780 1.560 -0.660 -5.890 -4.35 51.63
0.25
0.30 2.00 0.734 1.460 -0.560 -5.080 -9.43 47.28
0.35 2.00 0.613 1.220 -0.327 -2.920 -12.35 37.85
0.40 2.00 0.430 0.860 0.040 0.350 -12.00 25.50
0.45 2.00 0.233 0.466 0.434 3.870 -8.13 13.50
0.50 2.00 5.37

Problem 2: A 20 MVA. 50 Hz generator delivers 50 MW over a double jo.5

circuit line to an infinite bus. The generator has an inertia constant H = 2.7
j0.3
MJMVA at rated speed. The generator transient reactance is
Xd 0.3 pu. P
Each transmission circuit has a
reactance of 0.5 pu on a 50 MVA base.
E' = 1.4pu and infinite bus voltage V 1.040°. A =
solid three-phase F) j0.15 j0.35s (V)
ground occurs on line -2. The system reactances during fault are shown in
Fig. 14. Plot swing curve for sustained fault up to a time of 0.4 sec.

Fig. 14 Positive sequence diagram

Solution: Before apply the step-by-step methqd, we need to calculate the inertia constant
we can
M and the power angle equations
under pre-fault and pesfault conditions.
Base MVA =50, f= 50 Hz, H 2.7 MJ/MVA,
X 0.3 pu, X, 0.5 pu, |E'| 1.4 pu, |V| 1.0 pu
= = = =

Inertia constant, M(pu) :

H 2.7
180 = 3x 104 s2/[Link]
180 f x 50
L. Pre-fault: X =
Xa+ =
0.3 += 0.55 pu
*********** **** ***** *****

IE'TVT 1.4 x 1
PelPmax1 Sinð ==
sind sind =
2.5454 sind-- (i)
X 0.55
50
Prefault power transfer =

5== 1pu
=
Pm =
mechanical input
Initial power angle is given by 2.5454 sinðo =
1
6 23.13
II. During fault: A jO.5
positive
reactance diagram is shown in sequence X
Fig. 15(a). j0.3
Converting star to delta, we obtain the P
network of Fig. 15(b), in which (Y to A)
E j0.15 j0.35 (V) (E')
jo.35
(a) (b)
Fig. 15 Positive sequence diagrams
0.3 x 0.15 +0.5 x
0.15 +0.15 x 0.3
= 1.8 pu
0.15

Pe llPmax I1 Slno = sin= 1.4 x 1 sind

1.8
=
0.7778 sind-- (ii)
Basavarajappa S R, E& E, BIET, Davanagere
14
 ehoose At = 0.05 sec. The recursive relationships for step-by-step swing curve calculation are as
us
Pan-1) = Pm- Pmax sinô,-1 - - (iv)
lows

(0.05)2
6, =
Að-1tMPa(n-1) Aô,-1tv10- Patn-1)
= =
Aô,-1 + 8.33 Paln-1)
*

(v)
, = 8-1 +A6,-- (vi)
discontinuity in P, and hence in Pa, the average value of P, must be used for the first interval.
Since there is a
Pa(0) 0 pu and Pafot) = E i E 4 ° = t76pu
0)0 pu ana
Fafot)= 1773 u 23.13 O 674%
.615 63472
= B 8 pu

Sustained Fault:
Pa(Oavg) 2

and (vi) above.

Calculations are carried out in Table 1 in accordance with the recursive relationship (iv), (v)
time t given in the
The second column of the table shows Pax the maximum power that can be transferred
at
0t and remains constant
first column. Pnax in the case of a sustained fault undergoes a sudden change at
t =

to t 0.l =

thereafter. The procedure of calculations is illustrated below by calculating the row corresponding
sec.
(0.1sec) 34.40°, Pmax 0.7778 pu, = sins (0.1sec) = 0.5649
x 0.5649 = 0.4394 pu
Pe(0.1sec) Pnaxsino (0.1sec) 0.7778
= =

Pa(0.1sec) = 1 - 0.4394 = 0.5606 pu

0.052
(At)
P(0.1sec)=
M 3 x 10-4
x 0.5606 = 4.66970

A6(0.1 sec) = 46(0.05 sec) + (Ac) P(0.1sec) = 8.38°+ 4.6697° = 13.05

M
S(0.15 sec) = 6(0.1 sec) + 46(0.1 sec) = 34.40°+ 13.05° = 47.45

Table 1 for sustained fault

Pmax sino Pe=Pmaxsinð P=1 - P aP
sec pu pu pu 8.33 P deg deg deg
2.5454 0.3928 1.0 0.0 ---

0.3928 0.3055 0.6945 23.13

0.7778
Oavg 0.3928 0.3472 2.89 2.89 23.13
0.05 0.7778 0.4386 0.3412 0.6588 5.4878 8.38 26.02
0.10 0.7778 0.5649 0.4394 0.5606 4.6697 13.05 34.40
0.15 0.7778 0.7367 0.5730 0.4270 3.5569 T6.61" >47.45
|0.20 0.7778 0.8992 0.6994 0.3006 2.5039 19.11 64.0
0.25 0.7778 0.9929 0.7723 0.2277 1.8967 21.01 83.17
0.30 0.7778| 0.9695 0.7541 0.2459 2.0483 23.06 104.18
0.35 0.7778 0.7961 0.6192 0.3808 3.1720 26.23 127.24
0.40 0.7778 0.4466 0.3474 0.6526 5.4361 31.66 153.47

180 Swing Curve

160

140

120
6(t) for sustained fault as calculated i n 0
Table 1 is plotted in Fig. 16 from 0

which it is obvious that the system is

unstable. O

20

005 0.15 02 .25 0.3

ime in sec

Fig. 16 Swing curve with sustained faul

Basavarajappa S R, E & E, BIET, Davanagere
5
 Kutta Method:
Numerical Solution of Swing Equation by Runge
is
The swing equation for i" generator
Pmi- Pmaxi sinð,) for i
=
1,2, m--(i)
state variable form;
For multimachine system swing equation is written in

Then,

Pmaxi sinôj) =f2(ô,w) for

i =1,2,...m - -^
(ti)
d =(Pmi
-

2i3
Order-2 method:
Algorithm for Runge-Kutta voltage and power.
1. Carry out the load flow study
(before transient) using specified
transient reactances of generator (E).
2. Find the voltage behind
line reclosed).
3. Compute YBus (during fault, postfault,
size At.
Set time count r=0 and step
set
4.
5. Compute *ji and åzi from equation (ii).
of equation (ii).
the two constants for each differential equation
6. Compute
and h{ = f2(8{,aw). At

+h}).At and h =f287 + k}, »f h{).At

+
k
A.
=
aw
fA(67 +kZ.
Compute next state estimation such that
where A6{ = ; (k + k2) and,

+ = w + Awi where Aw=;(h{+ h)

is to be plotted. Ifr < rfinal thenr = r + 1 and repeat from
8. Test for time linmit for which swing curve
step 5 above, otherwise print results and stop.
9. Plot the curve between rotor angle (8) and time (t).

Problem 3: A 20 MVA, 50 Hz generator delivers 50 MW

over a double j0.5
circuit line to an infinite bus. The generator [Link] inertia constant H 2.7 =
jo.3
is Xa 0.3 P
MJ/MVA at rated speed. The generator transient pu. =
reactance
base.
Each transmission circuit has a reactance of 0.5 pu on a 50 MVA
j0.15 j0.35
E'l 1.4 pu and infinite bus voltage V =1.020°. A solid three-phase
=

ground occurs on line -2. The system reactances during fault are shown in
Fig. 17. Plot swing curve for sustained fault up to a time of 0.4 sec by
Runge Kutta method. Fig. 17 Positive sequence diagram

Solution: Before we can apply the e p ep [Link] need to calculate the inertia constant
M and the power angle equations under pre-fault and pofault conditions.
Base MVA = 50, f= 50 Hz, H 2.7 MJ/MVA, X = 0.3 pu, X = 0.5 pu, |E'| = 1.4 pu, |VI = 1.0 pu
H 2.7
Inertia constant, M (pu) =
180
f180x 50
= 3 x
104s2/ele. deg
I. Pre-fault: X, = Xa+= 0.3 + 0.55 pu

E'|VI s i n s =1.4x 1
E E sins = 2.5454 sind - - > ()
PePmax!Sinð = 55

Prefault power transfer = E 1 pu =

50 Pm = mechanicai input
50
Initial powerangle is given by 2.5454 sindo= 1 do 23.13° = 0.4036 rad

Basav Wo ddo0
dt
asavarajappa S R, E & E, BIET, Davanagere 16