Never standstill

Chapter Three
Power System Stability Study

Instructor: ADUGNA.E.
Electrical & Computer Engineering Department
Wollega University
School of Engineering and Technology

2018
 Stability Definition
Based on IEEE
Power system stability is the ability of an electric power system,
for a given initial operating condition, to regain a state of
operating equilibrium after being subjected to a physical
disturbance, with most system variables bounded so that
practically the entire system remains intact.
The stability of an interconnected power system means is the
ability of the power system is to return or regain to normal or
stable operating condition after having been subjected to some
form of disturbance. 2
Definition: Power system stability is a condition in which
the various synchronous machines of the system remain
in synchronism (“in step") with each other after having
been subjected some form of disturbance. Conversely,
instability means a condition involving loss of
synchronism, or falling “out of step”.
Stable operation of a power system requires a
continues match of energy (Pgen= Pload)
Stability problem is only concern of interconnected system

3
The Aim of Power system (PS) design and operation

 To have a quality power system

 supply at desired frequency and voltage

 To have reliable power system.

 To have secured power system
 The system should be normal even if emergency takes place

 To have stable system.

 To be operate in most economical way
 minimum operating cost and maintenance. 4
 Types of Power system

Depending on their configuration Power system

network can be classified in to two:

1. Stand Alone system

2. Inter connected system

 The question of stability Analysis arises when the
system is IC system
5
 Objective of PS stability study

Power is generated by synchronous generators that

operate in synchronism with the rest of the system.

i.e. have the same frequency, voltage and phase

sequence.

 We can thus define the power system stability as the

ability of the power system to return to steady state

without losing synchronism. 6

Stability depends on:
- Initial operating condition – in what mode operating
- Nature of disturbance – 3phase fault, line outage, Δload
- Amount of disturbance – small change, Sevier change

7
 Classification of Stability Problems

Generally classifying Stability problems can be classified in to two

1. Rotor Angle stability: can be classified as:

- small signal stability

- Transient stability

- Mid-term stability

- Long-term stability

2. Voltage stability:

- Large disturbance voltage stability

- Small disturbance voltage stability

8
Classification of Stability Problems

9
Rotor Angle stability:- Is the ability of interconnected synchronous

machines of power system remain in synchronism.

• This is the primary requirement for operation of power system.

• To maintain this we have to study torque balance of the system. i.e.

torque supplied by prime mover = electromagnetic torque developed

by generator

 to analyze stability analysis, we have to know dynamics of the

machines.
10
 Power system Stability Analysis
• Let us consider a rotating mass
If we apply two opposite force, T1 & T2 to the mass

If T1 = T2 the mass will not rotate (stable)

But if one of the force is greater/less than the other the
mass will move (rotate).
- It is called disturbance
11
The same is true for a machine (motor/generators)

If Tm = Te or Pm = Pe, the machine will be stable

But, if Tm > Te – the machine will accelerate (disturbance)

If Tm < Te – the machine will decelerate
12
The power - load angle characteristics and swing equation
The power - load angle characterises:
Consider the generator connected to infinite bus in fig. 1.
In finite bus
x
~ V E
IX
E I
a)_ 

V

Fig .1 I b)_

The power transferred to system is

EV sin δ
P (1)
X
● From eq.1, the generator output is directly proportional to
E & V , but inversely proportional to X.
● The  is known as load angle, and with V, E & X held constant
the power output is only a function of this angle  (see fig.2.)

Power
(p) Large E

Normal E

Low E
P1

0 1800 0
450 900 1350

Fig.2. Power /load angle relationship

 Swing Equation
(Dynamics of Synchronous Machines)
 The equation governs the motion of the rotor of the
synchronous machines
Derivation of the swing equation

15
 Swing equation Cont.
• Under normal condition the relative position of rotor axis and
resulting magnetic field axis is fixed.

Hence Tm = Te, Ta = 0

• The angle between rotor axis and field axis is called power
angle/torque angle (δ).

• During any disturbance the rotor may accelerate/decelerate

w.r.t. synchronously rotating machine.
Hence Tm ≠ Te, Ta ≠ 0
• the equation describe this is known as Swing equation.

16
•   Ta = Tm – Te ............................. [1]
multiply the above equation by ω
ωTa = ωTm – ωTe ...................... [2]

Pa = ωTa, Pm = ωTm Pe = ωTe

Pa = Pm – Pe............................. [3]
from Pa = ωTa, but Ta = Jε

= Jωε, but M = J ω
= M ε, but ε =

Hence Pm – Pe = Pa = M

17
Where
Tm- mechanical Torque [N-m]
Te- Electrical Torque [N-m]
Ta- Acceleration Torque of the machine [N-m]
Pm- mechanical power [hp or kw]
Pe- Electrical Power [kw]
Pa- Acceleration power of the machine [kw]
J – is inertia of prime mover and generator [kgm2]
M – is angular momentum
ω - Angular speed [rad]
ε – Angular acceleration [rad/s2]
δ Angular displacement of the rotor w.r.t. reference axis [rad]
18
 Per unit Inertia constant
Inertia constant (H) - is the ratio of kinetic energy stored in the
machine and rating of the machine.
 
H= but KE = ½Jω2
H= but M = Jω

H=

M = but ω = 2
M=
Hence Pa =

19
H of the machine varies from 1 to 10s of seconds depending on
the type and size of the machine
Type of the machine Inertia constant [MJ/MVA]

Steam Turbine generator 3-10

Hydro generators 2-4
Synchronous motors 2
Induction motors 0.5
Synchronous condenser (reactive 1-1.25
power generator)

 Large H means, the machine is large and small H means the

machine is small.
- In steam turbine since the generator is connected with several
stages of turbines it takes much time constant (H) 20
Now the swing equation is expressed as:

21
Swing equation for two Machine

H1 H 2
where, H12 
H1  H 2
pm1 H 2  pm 2 H 1
Pm12 
H1  H 2
pe1 H 2  pe 2 H 1
Pe12 
H1  H 2
12  1   2
22
• Example1: A four-pole, 60-Hz synchronous generator has
a rating of 200 MVA, 0.8 power factor lagging. The
moment of inertia of the rotor is 45,100 kgm2.
determine M and H.
Solution:

1
KE  J 2
2
1
 (45100)(188.459) 2  801.2 MJ
2

23
Example2: a 50Hz 4 pole turbo generator rated 20MVA, 13.2KV has an
inertia constant of 9 sec. determine the energy stored in the rotor at
synchronous speed and acceleration if 25000hp mechanical power and
elect cal power developed is 15000kw determine the change in angle at
the end of 15 cycles
HG 20 MVA *9 sec
M    1.146kw.sec/ Hz
f 50* 
1
KE  M 
2
1
= *1.146 * 2 *50
2
= 90kw

24
Pa  Pm  Pe
= 25000*0.735-15000 kw
= 3375kw
2
 
pa = M 2 ,
t
 2  Pa 3375kw
   2.945rad / sec
t 2
M 1.146
d
 2.945t  A
dt
25
 Steady State Stability (Small signal stability)
• The ability of the power system to remain in synchronism when
subject to small disturbances
• Stability is assured if the system returns to its original operating
state (voltage magnitude and angle profile)
• Let us consider a small disturbance (perturbation)
From swing equation,

From the network equation we have

Pe = Pmax sin(δ)
26
 If we consider damping effect

H  2 e d
D  Pm (p.u)  Pmax sin .............[16]
 f o t 2
dt

Let us consider a small change in power curve

δ = δ0 + Δδ
where, δ - is rotor angle after perturbation
δ0 - is initial rotor angle and
Δ δ – is change in rotor angle due to perturbation

Now this equation is non linear equation we shall lineaze the

equation as follows:

27
  2 0   d  0  
M D
t 2
dt
 Pm  Pmax sin  0  

but , sin 0  sin0 cos   cos 0sin

since, Δδ is very low,
sin Δδ  0 and cos Δδ  Δδ
 2 0 )  2  d  0  d 
M M D D
t 2
t 2
dt dt
 Pm  Pmax sinδ 0 cosΔδ - Pmax cosδ 0sinΔδ.............[18]

28
  2 0 )  
2
d 0  d 
M M D D
t 2
t 2
dt dt
 Pm  Pmax sinδ 0 - Pmax cosδ 0 Δδ...........................[19]

=

if we remove the steady state component and consider
Δδ components, the equation will become:
Ms 2  Ds  M  D 2  0
 D  D2  4
s12 

This is called dynamics linearized equation of

synchronous generator.
29
the quantity Pmax cosδ0  , w/c is called

synchronization coefficient or stiffness coefficient

dPe
It comes is obined from  slop of Pe -  graph
d
Thus, the dyn. linearized equation will be:
 2  d 
M D    0....................................[20]
t 2
dt
this equation can be analyzed by finding of roots of
the equation.

30
 The Laplace transform of the above equation will becomes:

Ms  Ds    0......................[21]
2

Thus the roots are:

 D  D 2  4M 
S12 
2M
For normal operation,
M   D 2

Always the value of S is obtained in polar form of αjω

Where,

=  frequency of oscillation

where,  

31
For lightly damped system, i.e. small D, δ = δo + Δδ → δo

How ever, for the negative values of ψ or D one of the root will be >0
For positive root, the oscillation of the machine will increase exponentially
This is called “Run away”. (lose of synchronous characteristics)

 The aim of small signal stability is to analysis weather the system

Back to initial condition /goes to run away after small perturbation.

32
 Damping Response of steady state stability
from the equation below,
H  2  d 
D    
 f t 2
dt
 2   fD d   fD
     
t 2
H dt H
 2  d  
  2n  n   
t 2
dt
 fo D  fo
where, n  and  
H 2 H
Damded frequency of oscillation (d ) will be
d  n 1   33
For small perturbation, the dynamic response (solution of equation{24})
Will be:


  e n t sin(d t )

 
and
n
  e n t sin(d t)

 
time constant of the system will be:
1 2H
= 
n  fD
34
 Effect of Damping and synchronization constant

For +ve D
(stable)

For -ve D
(unstable)

35
 Transient stability Analysis

 Transient stability is defined as the ability of the power

system to bring it to a stable condition or remain in
synchronism after a large disturbance.
 The transient stability limit is the maximum power that
can be transferred by a machine to a fault or a receiving
system during a transient state without loss of
synchronism. Transient stability limit is always less than
steady state stability limit.
36
Classical Approach
- Since we consider a very short time (5 – 10 sec ) we
assume:
1. Mechanical input to generator remain constant
2. Mechanical damping and AVR action are neglected
3. The transient model of network is neglected and static
model of the network is considered
4. Loads are represented as a constant
impedance/Admittance
5. The mechanical angle of each machine coincides with
electrical phase of voltage behind transient reactance

37
To analysis transient stability we utilizes:
- The dynamic equation of the machine and
- The static equation of the network

 Basically the dynamic equation of the machine is the

swing equation

38
 Network Equation
Let us consider the network below

From the above network

39
P2 =Re[V2 I*2 ]
v1 <δ - V2 <00 *
= V2 [ ]
x total <θ
V1V2 V22
= cos(θ - δ) - cos(θ)
x total x total
let, the line is purely inductive, θ = 900

 the max. transferd power from one bus

V1V2
to the other will be sin(δ)
x12
V1V2
where, Pmax = it is called steady state limit
x12

40
graphically

For normal operation the maximum value of δ is 900

41
 Single machine infinite bus system

Amount of power delivered to infinite bus is:

P2 =Re[V2 I*2 ]
0
v1 <δ - V2 <0 *
= V2 [ ]
x total <θ
0 V1V2
for θ = 90 the max.power deliverd = sin(δ)
x total 42
 Equal area criteria

One common method to study transient stability is:

Equal area criteria.
It is quick prediction method of power system stability

 2
from swing equation, M  Pm  Pe
t 2

d
Multplying both sides by 2 ,
dt
 2 d d
M 2 2 ( Pm  Pe )
t 2
dt dt

43
  2
from swing equation, M  Pm  Pe
t 2

d
multply both side by 2
dt
 2 d d
 M 2  2 ( Pm  Pe )
t 2
dt dt
 2 d d d 2
Recal - 2  ( )
t 2 dt dt dt
d d 2 d
thus, M ( ) 2 ( Pm  Pe )
dt dt dt
Integrate both side
d 2 d
 M( )  2 ( Pm  Pe )
dt dt
d 2 d

dt

M dt ( Pm  Pe )

d
= relative speed w.r.t synchronously rotating generator 
dt
d
thus, for stable system =0
dt 44
Let us Consider

45
 1. Equal area criteria during Sudden increase of Pm in
power system
 Initially operating at point a – (Pmo ,Peo) steady state operation
 During sudden increase of Pm, Pm > Pe.

The machine will accelerate (rotate more than synchronous speed)

energy will be stored in the form of KE. as a result δ keep increasing until
energy dissipate equal with energy stored (δmax) to decelerate.
 Post increment the system will operate @ point d,e at this point, the
machine will run with decelerate and energy dissipate finally, rotor
speed = synchronous speed but Pe>Pm .

To maintain Pe = Pm, the machine will oscillate between point a to e

Due to the damping of the machine the oscillation settle around point c
46
Due to the damping of the machine the oscillation settle around
point c

47
To be stable energy stored should be = energy dissipated
i.e. Area1-Area2 =0 [using energy conservation]
If Area1 =Area 2, system will be stable
If Area1 
Area 2, system will be unstable
This is called equal area criteria.

Area1   ( Pm1  Pe)d 

max
Area 2 

 ( Pe  Pm1)d 

for stable sytem,

 max

 ( Pm1  Pe)d   
 
( Pe  Pm1)d   0
48
 Max. Power to be Increased to Machine

Consider

49
To be stable Area2 should be at least equal to Area1
The limit occurs when δmax is at the intersection of line Pm1 and
power angle curve

Area1   ( Pm1  Pe)d 


but Pe =Pmax sin


thus, Area1= Pm1( )   ( P max sind

max
Area 2 

 ( Pe  Pm1)d 

substituting , Pe =Pmax sin

max
Area 2 

 ( P max sin d  Pm(max-)
50
for stable sytem,
 max
Pm1( )   ( P max sind   ( P max sin d  Pm(max-)
 

since, Pe = pm =Pmaxsin
max -Pmax sinmax = Pmax (cos cosmax )
this equation can be solved by itterative technique like NR, GS ...
the aim is to determine max once max is obtained we can determine Pm
sin ce, Pm  Pmax sin max
and 1  180  max

51
 2. Equal area criteria during fault
Consider

During steady state, V1 =V1<δo

During fault Pe = 0, no power delivered from source side.

52
53
During fault the machine accelerate and ωr will increase
as a function of time
 During steady state, the system operate @point a
Pe=Pm
 During fault, the system operate @point b Pe=0
 After fault is cleared the system operate @point e and
stored energy is dissipate @ point f
 Machine oscillates around point a and settled by
damping of machine.
• To be stable A2 should at least equal to A1 otherwise
can be larger.

54
If area 1 = area2

 max

 ( P )d  

m

( Pe  Pm )d
55
Suppose if the fault is cleared at δ = δ’

Now, A1>A2 so the system will goes to disturbance.

Thus δ has to be limited up to some δcr called critical angle.
56
 Critical clearing time

Critical clearing time - is a time at which the fault has to

be cleared for stable operation.
For stable operation
 max

 (P

m )d  

( Pe  Pm )d

 max
  (P

m )d  

( Pmax sin   Pm )d

 Pm (c - o) = Pmax (cos c - cos max )  Pm( max  c)

solving for c
-1 Pm
c = cos [ ( max  )  cos  max ]
Pmax
The system will be stable If the fault is cleared before δc. Otherwise...
57
Most of the time it is inconvenient to interpret angle. Therefore we
have to express δc in terms of time.
From swing equation,
H  2
 Pm  Pe
 f t 2

Since we want to obtain to obtain time to be cleared during fault,

Pe = 0 – during fault.
Thus the above equation can be re written as:
H  2
 Pm double integrating
 f t 2

f
the equation yields  Pmt 2  
2H
2H
thus, t c = (c -)
 fPm
58
 3. Equal area criteria for pre loaded system

Pre – unloaded system Pre – Loaded system

Let Pm2 = 1.5*Pm1,

Area 1 of pre-loaded system will be increased by 1.5.
59
Increasing the pre fault load by 50% increases the acceleration
power by 1.5.
Acceleration area will be >> deceleration area as a result the
system goes to disturbance.
Therefore if your pre fault load is higher than rated value,
the fault has to be cleared much earlier than critical clearing time
for stable operation.

60
 Stability Margin
Stability Margin – is also one of an indicator of stability of the
system

The Larger the Margin area the more stable system

System with zero margin will easily goes to disturbance
61
Equal area criteria during three phase fault with out
reclosed

pre  fault ,
X Tot =X d +X Txn +(X L1//X L2 )

During fault ,
X Tot =X d +X Txn )

post  fault ,
X Tot =X d +X Txn + X L1

Compare Xtotal of the three cases

62
  𝛿1  𝛿 2
𝑃𝑚−∫ 𝑃𝑜− 𝑃𝑒 2 = ∫ ( 𝑃𝑒 2 ) − 𝑃𝑚 63
𝛿𝑜 𝛿1
•Determining
  δ1 the critical clearing time will be:

Equal area criteria during three phase fault with Line

Reclosing

pre  fault , post  fault ,

X Tot =X d +X Txn + X L1
X Tot =X d +X Txn +(X L1 //X L2 )

During fault , pre  fault ,

X Tot =X d +X Txn ) X Tot =X d +X Txn +(X L1//X L2 )
64
 Factors affecting Transient stability
1. Post fault reactance
2. Duration of fault

3. Inertia of the machine

4. Generator internal voltage
5. Loading status of generator before disturbance

6. Internal reactance of generator

7. Generator output power during disturbance

65
 Methods of Improving Transient stability
To enhance transient stability we may control Pe or Pm

1. Based on Pe:

To improve transient stability A2 should be higher than A1.

Hence to increase A2 we have to increase Pe

Pe can be increased by

- Increasing induced emf

- decreasing reactance

- Based on Pm:
66
 Numerical solution of swing equation
● To obtain δ(t) from swing equation numerical method should
be used. Several methods available, but here we discuss only
method called step-by-step (point-by-point) method.
● The solution  (t) is obtained at discrete interval spread of t
uniform thorough out. An accelerating power and change in
speed, which are continuous function of time, are discretized as
follow:
 The accelerating power Pa computed at the beginning of an
internal is assumed to remain constant from the middle of the
preceding interval to the middle of the interval being considered
as shown in fig.below
 The angular rotor velocity  is assume constant through out
any interval, at the value computed for the middle of the interval
as shown in fig below
Fig.
The acceleration power at the end of the (n-1) the interval
Pa (n-1) = pm - pmax sin n –1.

3  1
 - in the interval (n  )to n   is
2  2
Pa  n 1 t
n 1  n  3 
2 2 M
The change in  during the (n-1)th interval is
 n 1   n 1   n  2  t n 3 / 2
th
and during the n int erval
 n   n   n 1  t  n  1
2
 From the above
p a ( n 1)
 n   n 1 ` t   n 1 / 2   n 3 / 2   t .
2

M
p a  n 1
 n   n 1  t 2

M
Finally,
 n   n 1   n
 The process will be repeated to obtain δn+1 & δn+1.
t = 0.05 is usually taken.
 Any change in operational condition causes an abrupt change
in the value of pa when two value of Pa apply, the mean is used.
To do so:
- We have to use high speed excitation system
- Having breaking resistance
- Using series/parallel compensation
- Shading load
- Using FACT devices
2. Based on Pm control:
Use fast valve technique on governor

71
 ? Any Question ?
?
? ?
? ?
? ?
? ? ? ?
? ? 7