Causes of power system faults

Faults usually occur in a power system due to either insulation

EEE2416: POWER SYSTEMS II failure, flashover, physical damage or human error.
Some causes of system faults are:
 Equipment insulation fails
Power system faults  System over-voltages caused by lightning and switching
surges
 Insulation contamination

 Falling trees and tree branches shorting overhead lines

By
 Vehicles and aircraft hitting towers and poles

 Vandalism
K. Kaberere  Birds shorting overhead lines

 Winds and gales

 Ice storms
2

3 4

07 June 2016: This primate fell onto a transformer at the Gitaru

hydroelectric power station, and tripped it. This triggered a cascading
Wild fires in Australia, 2009 effect on the other generators which ended up disrupting power
generation and distribution in the whole country,"
6

1
 7 8

 When a fault occurs in the power system, large currents

and/or abnormal voltages are developed.
 Short circuit currents may cause thermal damage to
equipment
 Windings and busbars may suffer mechanical damage
due to high magnetic forces during faults
 Hence, faulty section should be disconnected
immediately so that normal operation of the rest of the
system is not affected.
 Relay detects the fault and circuit breaker isolates or
disconnects the faulty section (Fuse detects and isolates)
 For proper relay setting and CB interrupting capacity –
the values of expected fault currents and voltages should
be known – short circuit calculations
9

10

Short circuit calculations are broadly classified as:

 Three-phase / symmetrical / balanced faults – involve all
the 3 phases and sometimes even the ground – most severe
but not common
 Unsymmetrical / unbalanced faults – involve one or two
phases – voltages and currents become unbalanced –
UNSYMMETRICAL FAULTS
symmetrical components used for calculating fault
currents SHORT CIRCUITS
Short circuit current is determined by:
 The internal emfs of the machines in the network
 The machines internal impedances
 The impedance of the network between the machine
and the fault location
11 12

2
 Introduction Assumptions commonly made in 3- fault studies
Most power system faults are unsymmetrical.
 All sources are balanced and equal in magnitude & phase
They may be unsymmetrical short circuits or
 Sources are represented by the Thevenin’s voltage prior to
open conductors.
fault at the fault point
They may be classified as:  Large systems may be represented by an infinite bus-bar
1. Single line-to-ground (SLG or L-G)  Transformers are on nominal tap position
2. Line-to-line (L-L or DL)  Resistances are negligible compared to reactances
3. Double line-to-ground (DLG or L-L-G).  Transmission lines are assumed to be fully transposed and
all 3 phases have same impedance
Unsymmetrical faults result in the flow of
 Loads currents are negligible compared to fault currents
unbalanced currents in the system
 Line charging currents can be completely neglected
Symmetrical components are used to determine
post fault system currents and voltages
13 14

1. Single line-to-ground fault on an unloaded generator (S-L-G) The symmetrical components of the currents are

 I a 0  1 1 1 1   I a  1 1 1 1   I a 
 The fault  I a1   1 a a 2   I b   1 a a 2   0 
occurs on  I a 2  3 1 a 2 a   I c  3 1 a 2 a   0 
phase 'a'.
 The fault I a …………………………….(1)
conditions
I a 0  I a1  I a 2 
3
are: From the generator sequence network drawn earlier
1. Ib = Ic = 0
2. Va = 0 Va 0   0   Z 0 0 0   I a 0 
Va1    Ea    0 Z1 0   I a1 
Va 2   0   0 0 Z 2   I a 2 
Substitute Ia0 = Ia1 = Ia2 in the [Is] vector

15 16

 From (1) and (3), it can be

Va 0   0   Z 0 0 0   I a1  seen that the sequence
Ia0 = Ia1
Va1    Ea    0 Z1 0   I a1  ……………….(2)
networks are connected in
Va0
Va 2   0   0 0 Z 2   I a1  series after the occurrence of
the fault.
 Current flowing is Ia1 and the
But Va  Va 0  Va1  Va 2 voltage across the series
circuit is Ea. Ia1
From (2), and knowing Va = 0
Ea Va1

Va   I a1 Z0  Ea  I a1 Z1  I a1 Z 2  0 The fault current

3 Ea
I a1 
Ea I f  I a  3 I a1 
Z 0  Z1  Z 2 …………………………….(3) Z 0  Z1  Z 2
Ia2 = Ia1
Va2

17 18

3
 If the neutral of the generator is ungrounded, the zero SLG faults on power systems
sequence network is open-circuited.
 Then, Ia1 and hence Ia = 0.  Prefault voltage is designated Vf and only appears
in the positive sequence network
After determining Ia1  The sequence networks for the system are first
Va1 = Ea – Ia1Z1 drawn
Va2 = – Ia1Z2  Then Thevenin equivalent for each network as
Va0 = – Ia1Z0 seen from the fault location is derived
 The value of equivalent impedance is applied in
Vb  Va 0  a 2Va1  aVa 2  the equations derived earlier for each type of
 Phase voltage of the healthy phases fault.
Vc  Va 0  aVa1  a 2Va 2 

19 20

Example 1 Table E1
Consider the power system whose one-line diagram is shown Item MVA Voltage rating X1 X2 X0
below. rating (kV) (pu) (pu) (pu)
G1 100 25 0.2 0.2 0.05
G2 100 13.8 0.2 0.2 0.05
T1 100 25/230 0.05 0.05 0.05
Bus 5 is at the midpoint of transmission line TL2. T2 100 13.8/230 0.05 0.05 0.05
a) Draw the sequence networks and clearly mark the buses. TL1 100 230 0.1 0.1 0.3
b) Reduce the networks in (a) to their Thevenin equivalents TL2 100 230 0.2 0.2 0.6
as seen at Bus 5.
c) Determine the fault current and the phase voltages for a
SLG fault at bus 5.
System data are as given in Table E1. 21 22

Positive sequence network

Negative sequence network
The negative sequence impedances of the power system
components are equal to the positive sequence impedances.
Hence, the Thevenin equivalents for the two sequence
ref networks are similar with the exception of the emf source, as
shown below.
j0.175

23 24

4
 Zero sequence network Single-line-to-ground fault at bus 5
10
I a 0  I a 2  I a1    j1.821
j 0.175  j 0.175  j 0.199
 I a  1 1 1    j1.821   j 5.46
 I b   1 a 2 a    j1.821   0  pu I n   j 5.46 pu
 I c  1 a a    j1.821  0 
2

Va 0   0   j 0.199 0 0    j1.82  0.362

Va1    1    0 j 0.175 0    j1.82    0.681 
Va 2   0   0 0 j 0.175    j1.82  0.319

Va  1 1 1   0.362   0 
Vb   1 a 2 a   0.681   1.022 238  pu
Vc  1 a a 2   0.319  1.022122 

25 26

2
Example 2 kVbase 1382
Zbase(line)    190.44
MVAbase 100
Zline1  Zline2  j 0.105 pu and Zline0  j 0.315 pu

1 j0.10 j0.105 j0.10 2

j0.15 j0.20

1.050 1.050

Positive sequence

a) Draw the sequence networks for the system Z1  j 0.20 // j 0.455  j 0.139
b)Reduce the sequence networks to their Thevenin equivalents as j0.139
seen from bus 2
c) Determine the fault current for a SLG fault at bus 2 V1
1.050
The pre-fault voltage Vf = 1.050 pu. Neglect the pre-fault load
current and the transformer phase shift. 27 Thevenin equivalent 28

1 j0.10 j0.105 j0.10 2 Single-line-to-ground fault at bus 2

j0.146
1.050
I a 0  I a 2  I a1    j1.963
j0.17 j0.21 V2 j 0.139  j 0.146  j 0.25
 I a  1 1 1    j1.963    j 5.888
Negative sequence Thevenin equivalent  I b   1 a 2 a    j1.963    0  pu
Z2  j 0.21// j 0.475  j 0.146  I c  1 a a 2    j1.963   0 

Va 0   0   j 0.25 0 0    j1.963   0.491

1 j0.10 j0.315 j0.10 2 j0.25 Va1   1.05    0 j 0.139 0    j1.963    0.777 
j0.10
Va 2   0   0 0 j 0.146    j1.963   0.286
j0.05 V0
j0.15 Va  1 1 1   0.491  0 
Thevenin equivalent Vb   1 a 2 a   0.777   1.179 231.3  pu
Zero sequence Vc  1 a a 2   0.286  1.022128.7 
Z0  j 0.10  j 0.15  j 0.25
29 30

5
2. Line-to-line fault on an unloaded generator (L-L or DL) The symmetrical components of the voltage are

 Let the fault be Va 0  1 1 1 1  Va 

on phase ‘b‘ Va1   1 a a 2  Vb 
and ‘c’. Va 2  3 1 a 2 a  Vb 
 The fault
1
conditions are: Va 0  V  2Vb 
3 a
 
1. Ib = -Ic 1
Va1  Va  Vb  a  a 2 
2. Ia = 0 3
3. Vb = Vc 1

Va 2  Va  Vb  a 2  a 
3 
 Va1  Va 2 ………………….…………….(1)

31 32

The symmetrical components of current are  0   0   Z0 0 0   0 

 I a 0  1 1 1 1   0  Va1    Ea    0 Z1 0   I a1 
 I a1   1 a a 2    I c  Va1   0   0 0 Z 2    I a1 
3 
 I a 2  1 a a   I c 
2
Va1  Ea  I a1 Z1  I a1 Z 2
Ia0  0
Ea
I a1 
1

I a1  I c a 2  a
3  Z1  Z 2
…...………………………….…….(3)

1

Ia 2  Ic a  a 2
3  From (1) and (2), it can be concluded that the positive and
negative sequence networks must be connected in parallel
Ia1 = -Ia2 …………………………………….…….(2) without the zero sequence network since Z0 does not
appear in (3).
Va 0   0   Z 0 0 0   I a 0 
But Va1    Ea    0 Z1 0   I a1 
Va 2   0   0 0 Z 2   I a 2 
33 34

 The grounding of the

Ia1 Ia2 = -Ia1 generator neutral does not Va  Va 0  Va1  Va 2  2Va1  Healthy phase
affect the fault current.

Vb  Vc  Va 0  a 2Va1  aVa 2

The fault current

 
 a 2  a Va1  faulted phases
I f  Ib   Ic


I b  I a 0  a 2 I a1  aI a 2  I a1 a 2  a 

I f  a2  a 
Ea
Z1  Z 2
…...…………………….…….(4)

Va1  Va 2 and Va 0  0

35 36

6
 Example 3 Example 4
Repeat Example 1 for a L-L fault. Repeat Example 2 for a L-L fault.
10 1.050
I a1   I a 2    j 2.857 I a1   I a 2    j 3.684
j 0.175  j 0.175 j 0.139  j 0.146
Ia 0  0 Ia 0  0

 I a  1 1 1   0   0   I a  1 1 1   0   0 
 I b   1 a 2 a    j 2.857    4.95  pu In  0  I b   1 a 2 a    j 3.684    6.382  pu
 I c  1 a a 2   j 2.857   4.95 
 I c  1 a a   j 3.684   6.382 
2

Va1  Va 2   j 2.857  j 0.175  0.5 Va1  Va 2   j 3.684  j 0.146  0.538

Va 0  0 Va 0  0

Va  1 1 1   0   1.0  Va  1 1 1   0   1.0 

Vb   1 a 2 a   0.5    0.5  pu Vb   1 a 2 a   0.538    0.538  pu
Vc  1 a a 2   0.5   0.5  Vc  1 a a 2   0.538   0.538 
37 38

3. Double line-to-ground fault on an unloaded generator (L-L-G or DLG) Va 0  1 1 1 1  Va  1 Va 
Va1   1 a a 2   0   Va 
 Let the fault be Va 2  3 1 a 2 a   0  3 Va 
on phase ‘b‘ V
and ‘c’.  Va 0  Va1  Va 2  a ………….………………….(1)
Ib 3
 The fault I a  0  I a 0  I a1  I a 2
conditions are: I a 0    I a1  I a 2  ………….………………….(2)
1. Ia = 0
Vb  0  Va 0  a 2Va1  aVa 2 
2. Vb = Vc = 0  ………….………….(3)
Vc  0  Va 0  aVa1  a 2Va 2 
If = Ib + Ic
But Va0 = – Ia0Z0
Substituting (2) for Ia0 gives
Ic Va0 = (Ia1 + Ia2)Z0
Va1 = Ea – Ia1Z1 ………….……………………….(4)
39
Va2 = – Ia2Z2 40

After making the necessary substitutions and simplifying,

we obtain Ia1 Ia2 Ia0
Ea
I a1 
 Z 2 Z0 
Z1   
 Z 2  Z0 
Z0 Va1 = Ea – Ia1Z1
I a 2   I a1
Z2  Z0 Va2 = – Ia2Z2
Z2 Va0 = – Ia0Z0
I a 0   I a1 Va  Va 0  Va1  Va 2  Healthy phase
Z2  Z0
I b  I a 0  a 2 I a1  aI a 2
It can be seen from (1) that the sequence voltages are equal.
Therefore, their networks must be connected in parallel. I c  I a 0  aI a1  a 2 I a 2
The fault current
I f  Ib  Ic
41 42

7
Case 4: Double-line-to-ground fault at bus 5
10
I a1    j 3.73 Va  1 1 1   0.348  1.044 
 j 0.175  j 0.199  Vb   1 a 2 a   0.348    0  pu
j 0.175  
 j 0.175  j 0.199  Vc  1 a a   0.348   0 
2

 j 0.199
Ia 2    j 3.73  j1.99
j 0.175  j 0.199
 j 0.175
Ia 0    j 3.73  j1.75
j 0.175  j 0.199

 I a  1 1 1   j1.75   0 
 I b   1 a 2 a    j 3.73    5.60152.1  pu
 I c  1 a a 2   j1.99   5.60 27.9 

I n  j 5.25 pu
43 44

Unsymmetrical faults with non-zero fault impedance ZF 2. L-L

1. SLG  If the fault occurs on phases ‘b’ and ‘c’ , the
 For bolted faults - ZF = 0 fault conditions are:
 For an arcing fault - ZF is arc impedance 1. Ia = 0
 Transmission line insulator flashover - ZF includes 2. Ib = -Ic
total impedance between line and ground 3. Vb - Vc = ZFIb
 If the fault occurs on phase 'a‘, the fault conditions VF
are: I a1   I a 2  Ia0 = 0
Z1  Z 2  Z F
1. Ib = Ic = 0
2. Va = ZFIa
VF
I a1  I a 2  I a 0 
Z 0  Z1  Z 2  3 Z F
45 46

3. DLG
 If the fault occurs on phases ‘b’ and ‘c’ and
ground, the fault conditions are:
1. Ia = 0
2. Vb = Vc SYMMETRICAL FAULTS
3. Vb = ZF(Ib + Ic)

I a1 
VF

VF SHORT CIRCUITS
Z1  Z 2 //  Z 0  3 Z F   Z2  Z0  3 Z F  
Z1   
 Z2  Z0  3 Z F 
 Z0  3 Z F 
I a 2   I a1 
 Z 2  Z 0  3 Z F 
 Z2 
I a 0   I a1 
 Z 2  Z 0  3 Z F 
47 48

8
Transients in RL series circuits The 1st term of (1) varies sinusoidally with time (ac
component) whereas the 2nd term is non-periodic and
decays exponentially with a time constant of L/R (dc
component)

e  Vmax sin( t   )
di Current i for  -  = 0 Current i for  -  = -/2
Vmax sin( t   )  L  Ri for t  0
dt
Vmax
i sin( t     )    Rt / L sin      .............(1)
Z  
where Z  R 2  ( L)2 and   tan 1   L / R 
49 50

Symmetrical short circuit faults (3- short circuit)

1. Unloaded synchronous machine
 The current that flows when a generator which is on no
load is shorted at the terminals is similar to that which
flows when the switch in the RL circuit is closed.
 However, due to armature reaction, the current flowing
in immediately after a fault, a few cycles later, and the
steady state value differ considerable
 Since the short circuit occurs at different points of the
voltage wave for the 3 phases (120° separation, 
different for each phase), the dc current is different in
each phase. Furthermore, if CB opens in reasonable
time,dc component will have decayed. Therefore neglect
Resulting plot of phase current Vs. is as follows.
51 52

The ac fault current in a synchronous machine can be

modeled by an RL circuit with time varying inductance L(t).
This is represented by
Xd" = direct-axis subtransient reactance
Xd' = direct-axis transient reactance
Xd = direct-axis synchronous reactance
53 Xd" < Xd' < Xd 54

9
 Initial fault current after fault occurrence is high due to reduced EXAMPLE 1
armature reaction but reduces to a lower value as armature
reaction increases. If the pre-fault rms phase voltage at the Two generators are connected in parallel to the L.V. side
generator terminal (on no load) is |Eg|, of a - transformer. G1 is rated 50 MVA, 13.8 kV and
G2 25 MVA, 13.8 kV. Each generator has a subtransient
| Eg | reactance of 25%. The transformer is rated 75 MVA, 13.8
| I | rms steady state current  / 69  kV with a reactance of 10%. Before the fault
Xd
| Eg | occurs, the H.V. side of the transformer is at 66 kV and
| I  | rms transient current the transformer is unloaded with no circulating current
X d
between the generators. Find the subtransient fault
| Eg |
| I  | rms subtransient current current in each generator when a short circuit occurs on
X d the H.V. side of the transformer.
The q-axis reactances do not significantly affect fault current
since armature resistance is small.
55 56

Solution 50 MVA
13.8 kV I1  2.734 pu
75 MVA
13.8 /69  kV
I2  1.367 pu

MVAbase  103
I base 
3  kVbase
25 MVA  2091.85 A
Let kVbase = 69 kV and
MVAbase = 50 MVA
13.8 kV I1  5.72kA
X eq  X 1 // X 2  X Tx I2  2.86kA
2
MVAbase ( new )  kVbase ( old ) 
X new  X old  
MVAbase ( old )  kVbase ( new ) 
 0.233 pu
Vf
X 1  0.25 pu If   4.101 pu
X 2  0.50 pu Pre-fault voltage X eq
X Tx  0.067 pu = 66/69 = 0.957 pu
57 58

2. Loaded synchronous machine

If the load is a motor, after the occurrence of the fault,
 Consider a generator that is loaded before a fault occurs the motor behaves like a generator and supplies a
at point P on the system as shown in the circuit below. current I″m = I″m1 - IL
 The pre-fault current and voltage at point P are IL and Vf E″m = Vt - j IL X″m
respectively, and the generator terminal voltage is Vt. The total fault current I″f = I″g + I″m = I″g1 + I″m1
Before the fault occurs and
immediately after, the internal
voltage of the generator is E"g.
E"g = Vt + j IL X"g

The generator feeds I"g to the

fault where
I"g = I"g1 + IL

59 60

10
 Solution
EXAMPLE 2
The synchronous motor in the power system shown below
is drawing 80 MW at 0.8 p.f. leading and a terminal
voltage of 13.4 kV, when a symmetrical 3 phase fault
occurs at Bus 2. Find the sub-transient current in the
generator, motor, and fault.

100 MVA 100 MVA

13.8 /138  kV 138  / 13.8  kV Let kVbase at bus 1 = 13.8 kV and MVAbase = 100 MVA
XT1 = 0.10 XT2 = 0.10
2
kVbase
G
XTL= j20
M Zbase(line)  ZTh = j0.2//j0.455 pu
T1 MVAbase
100 MVA 1 T2 2 100 MVA = j0.139 pu
 190.44
13.8 kV 13.8 kV
X"g = 0.15 X″m = 0.2 XTL = j0.105 pu Vf = 0.9710 pu
61 62

Vf Fault Level Calculations

I f    j 6.986 pu
ZTh
 In a power system, the maximum the fault current (or
MVAbase  103 fault MVA) that can flow into a zero impedance fault is
Ibase   4183.7 A
3  kVbase necessary to be known for switch gear rating.
 This can either be the balanced three phase value or the
I f   j 29.2kA value at an asymmetrical condition.
 The fault level is usually expressed in MVA (or
I g1   j 8.9 kA
corresponding per-unit value), with the maximum fault
Im 1   j 20.3 kA current value being converted using the nominal voltage
P rating.
IL   4.309 36.87 kA MVAbase  3kVbase I base  103
3V cos  MVAbase
 3.447  j 2.585 kA MVA f   3kVbase I f  103
Z pu
I g  I g1  I L  7.212  61.45 kA
MVA f  103 I base
Im  I m  1  I L  23.14  98.57 kA If  
Z pu
3kVbase
63 64

Limiting short circuit current

 For a fault that occurs at the generator bus bar, the
impedance presented to the fault current is low and
hence the fault current is very high.
 This may cause thermal and mechanical damage to
equipment.
 To limit fault current, high impedance reactors are
connected between the source and the fault location.
 The Short circuit capacity (SCC) of a busbar is the fault level
of the busbar.  Disadvantages: they introduce a voltage drop, have active
 The strength of a busbar (or the ability to maintain its and reactive power losses, may adversely affect the optimum
voltage) is directly proportional to its SCC; an infinitely distribution of power flow in interconnected networks and
strong bus (or Infinite bus bar) has an infinite SCC, with a may also adversely affect the transient and dynamic stability
zero equivalent impedance and will maintain its voltage performance of power systems
under all conditions.  Possible locations of the reactors include:
65 66

11
1. Generator reactor – between generator and busbar 4. Earthing resistor or reactor connected to
Not common in modern design –generators transformer neutral
have inherently high reactance windings • Increase ZPS impedance hence limit SLG and DLG
and they are protected by very fast relays fault current
Neutral grounding
2. Busbar reactor – between sections of busbars
 A system whose neutral point is NOT connected to
Prevents total voltage collapse at the ground is said to have an isolated or free neutral
station in case of a fault on one bus bar.
 The connection of the generator and transformer neutrals
Limits circulating current in case of
influences fault currents and voltages during unbalanced
imbalance e.g. fault on one section of the
busbar. faults that involve the ground (S-L-G and D-L-G)
 Usually, grounding is done through the transformer or
3. Feeder reactor
generator neutral
Prevents voltage collapse in case of a fault  Transformers are normally solidly grounded (no
near the generating station. impedance connected in the neutral path)
67 68

 Generators are commonly grounded through an Read on:

impedance (resistance or inductive resistance) to Neutral grounding
limit the stator fault current  Advantages of neutral grounding
 In systems with isolated neutrals, supply can be  Advantages of isolated neutral
maintained with a S-L-G fault – the voltage of  Peterson coil
the healthy phases is above ground and there is
reduced interference with telephone lines since
there are no zero sequence currents

69 70

12