MALAVIYA NATIONAL INSTITUTE OF TECHNOLOGY JAIPUR

DEPARTMENT OF ELECTRICAL ENGINEERING

Digital Signal Processing Project

TOPIC:-Deblurring the image using fast fourier

transform and analysing the effect of noise on
deblurring

Submitted to
Dr. Hemant Kumar Meena

Prepared by
Shubham Kumar(2021UEE1418)
Prince Kumar(2021UEE1460)
Dinesh Kumar(2021UEE1504)
INTRODUCTION:-

The stability of an interconnected power system is its ability to return

to normal or stable operation after having been subjected to some
form of disturbance. Conversely, instability means a condition
denoting loss of synchronism or falling out of step. Stability
considerations have been recognized as an essential part of power
system planning for a long time. With interconnected systems
continually growing in size and extending over vast geographical
regions, it is becoming increasingly more difficult to maintain
synchronism between various parts of a power system. The
dynamics of a power system are characterised by its basic
features given below:-

1. Synchronous tie exhibits the typical behaviour that as power

transfer is gradually increased a maximum limit is reached beyond
which the system cannot stay in synchronism, i.e., it falls out of step.

2. The system is basically a spring-inertia oscillatory system with

inertia on the mechanical side and spring action provided by the
synchronous tie wherein power transfer is proportional to sin d or d
(for small E, 6 being the relative internal angle of machines).

3. Because of power transfer being proportional to sin d, the equation

determining system dynamics is nonlinear for disturbances causing
large variations in angle d, Stability phenomenon peculiar to non-
linear systems as distinguished from linear systems is therefore
exhibited by power systems (stable up to a certain magnitude of
disturbance and unstable for larger disturbances)

Types of Stability

1.Steady state stability:

*the ability of the power system to regain its original position after
subjecting to small disturbance.
*The study of steady stability is concerned with the determination of
the upper limit of the machine loading before loosing synchronism
provided the loading is gradually increased.

2.Transient stability:
*It is related to the large disturbance in the power system.
*when the large disturbance occurs the rotor speed ,rotor angle and
the power transfer undergoes fast change.

3.Dynamic stability:-
*When the small disturbance is continuously occurring in the power
system which are small enough not to loose synchronism but do
excite the system into the state of natural oscillations.
*The system is dynamically stable if the oscillation do not acquire
more than certain amplitude and die out quickly.
SWING EQUATION:-

Under normal operating conditions, the relative position of the rotor

axis and the resultant magnetic field axis is fixed. The angle between
the two is known as the power angle or torque angle. During any
disturbance, rotor will decelerate or accelerate with respect to the
synchronously rotating air gap mmf , and a relative motion begins. The
equation describing this relative motion is known as the swing
equation. If, after this oscillatory period, the rotor locks back into
synchronous speed, the generator will maintain its stability. If the
disturbance does not involve any net change in power, the rotor
returns to its original position. If the disturbance is created by a
change in generation, load, or in network conditions, the rotor comes
to a new operating power angle relative to the synchronously
revolving field.
In order to understand the significance of the power angle we refer to
the combined phasor/vector diagram of a two-pole cylindrical rotor
generator illustrated in Figure 3.2. From this figure we see that the
power angle δ r is the angle between the rotor mmf ⁡F r and the resultant
air gap mmf F sr, both rotating at synchronous speed. It is also the
angle between the no-load generated emf E and the resultant stator
voltage E sr. If the generator armature resistance and leakage flux are
neglected, the angle between E and the terminal voltage V , denoted by
δ , is considered as the power angle.

Consider a synchronous generator developing an electromagnetic

torque T e and running at the synchronous speed ω sm. If T m is the
driving mechanical torque, then under steady-state operation with
losses neglected we have
T m=T e

A departure from steady state due to a disturbance results in an

accelerating ¿ T e ) or decelerating ( T m <T e ) torque T a on the rotor.
T a=T m−T e
If J is the combined moment of inertia of the prime mover and
generator, neglecting frictional and damping torques, from law's of
rotation we have
2
d θm
J 2
=T a=T m−T e
dt

where θm is the angular displacement of the rotor with respect to the

stationary reference axis on the stator. Since we are interested in the
rotor speed relative to synchronous speed, the angular reference is
chosen relative to a synchronously rotating reference frame moving
with constant angular velocity ω sm, that is
θm =ω sm t +δ m

where δ m is the rotor position before disturbance at time t=0 ,

measured from the synchronously rotating reference frame.
Derivative of (11.4) gives the rotor angular velocity
d θm d δm
ω m= =ω ms+
dt dt

and the rotor acceleration is

2 2
d θm d δm
2
= 2
dt dt

Substituting (11.6) into (11.3), we have

2
d δm
J 2
=T m−T e
dt

Multiplying (11.7) by ω m, results in

2
d δm
J ωm 2
=ω m T m −ωm T e
dt

Since angular velocity times torque is equal to the power, we write the
above equation in terms of power
2
d δm
J ωm 2
=Pm −P e
dt
The quantity J ωm is called the inertia constant and is denoted by M . It
is related to kinetic energy of the rotating masses, W k.
1 2 1
W k = J ωm= M ω m
2 2

or
2W k
M=
ωm

Although M is called inertia constant, it is not really constant when

the rotor speed deviates from the synchronous speed. However, since
ω m does not change by a large amount before stability is lost, M is
evaluated at the synchronous speed and is considered to remain
constant, i.e.,
2W k
M=
ω sm

The swing equation in terms of the inertia constant becomes

2
d δm
M 2
=P m−Pe
dt

It is more convenient to write the swing equation in terms of the

electrical power angle δ . If p is the number of poles of a synchronous
generator, the electrical power angle δ is related to the mechanical
power angle δ m by
p
δ= δ m
2

also,
p
ω= ω m
2

Swing equation in terms of electrical power angle is

2
2 d δ
M 2 =Pm −Pe
p dt
Since power system analysis is done in per unit system, the swing
equation is usually expressed in per unit. Dividing (11.16) by the base
power S B, and substituting for M from (11.12) results in
2 2 W K d δ Pm Pe
2
= −
p ωsm S B d t 2 S B S B

We now define the important quantity known as the H constant or

per unit inertia constant.
kinetic energy in MJ at rated speed WK
H= =
machine rating in MVA SB

The unit of H is seconds. The value of H ranges from 1 to 10 seconds,

depending on the size and type of machine. Substituting in (11.17), we
get
2
2 2H d δ
=P m( pu)−Pe( pu)
p ω sm d t 2

where Pm (pu) and Pe( pu) are the per unit mechanical power and electrical
power, respectively. The electrical angular velocity is related to the
mechanical angular velocity by ω sm=(2/ p)ω s. (11.19) in terms of
electrical angular velocity is
2
2H d δ
=P m( pu)−Pe (pu)
ωs d t2

The above equation is often expressed in terms of frequency f 0, and to

simplify the notation, the subscript pu is omitted and the powers are
understood to be in per unit.
2
H d δ
=Pm−P e
π f 0 d t2

where δ is in electrical radian. If δ is expressed in electrical degrees,

the swing equation becomes
2
H d δ
=Pm −Pe
180 f 0 d t 2

STEADY STATE STABILITY:-

We know that swing equation is :
2
H d δ
=Pm−P max sin ⁡δ
π f 0 d t2
(11.36)

The swing equation is a nonlinear function of the power angle.

However, for small disturbances, the swing equation may be
linearized with little loss of accuracy as follows. Consider a small
deviation Δ δ in power angle from the initial operating point δ 0, i.e.,
δ=δ 0 + Δ δ
(11.37)

Substituting in (11.36), we get

2
H d ( δ0 + Δ δ )
=P m−Pmax sin ⁡( δ 0 + Δ δ )
π f0 d t2

or
2 2
H d δ0 H d Δ δ
+ =Pm−P max ( sin ⁡δ 0 cos ⁡Δ δ+cos ⁡δ 0 sin ⁡Δ δ )
π f 0 d t2 π f 0 d t2

Since Δ δ is small, cos ⁡Δ δ ≅ 1 and sin ⁡Δ δ ≅ Δ δ , and we have

2 2
H d δ0 H d Δ δ
+ =Pm−P max sin ⁡δ 0−Pmax cos ⁡δ 0 Δ δ
π f 0 d t2 π f 0 d t2

Since at the initial operating state

2
H d δ0
=Pm−P max sin ⁡δ 0
π f 0 d t2

The above equation reduces to linearized equation in terms of

incremental changes in power angle, i.e.,
2
H d Δδ
+ Pmax cos ⁡δ 0 Δ δ=0
π f 0 dt2
(11.38)

The quantity Pmax cos ⁡δ 0 in (11.38) is the slope of the power-angle curve
at δ 0. It is known as the synchronizing coefficient, denoted by Ps . This
coefficient plays an important part in determining the system
stability, and is given by

Ps =
dP
dδ δ |
=Pmax cos ⁡δ 0
0

(11.39)

Substituting in (11.38), we have

2
H d Δδ
+ Ps Δ δ=0
π f 0 dt2
(11.40)

The solution of the above second-order differential equation depends

on the roots of the characteristic equation given by
−π f 0
s2= Ps
H
(11.41)

When Ps is negative, we have one root in the right-half s-plane, and

the response is exponentially increasing and stability is lost. When Ps
is positive, we have two roots on the j−ω axis, and the motion is
oscillatory and undamped. The system is marginally stable with a
natural frequency of oscillation given by

ω n=
√
πf0
H s
(11.42)
P

It can be seen from Figure 11.3 that the range where Ps (i.e., the slope
dP /dδ ) is positive lies between 0 and 90∘ with a maximum value at no-
load ( δ 0=0 ).

As long as there is a difference in angular velocity between the rotor

and the resultant rotating air gap field, induction motor action will
take place between them, and a torque will be set up on the rotor
tending to minimize the difference between the two angular
velocities. This is called the damping torque. The damping power is
approximately proportional to the speed deviation.
 dδ
Pd =D
dt
(11.43)

The damping coefficient D may be determined either from design

data or by test. Additional damping torques are caused by the
speed/torque characteristic of the prime mover and the load dynamic,
which are not considered here. When the synchronizing power
coefficient Ps is positive, because of the damping power, oscillations
will damp out eventually, and the operation at the equilibrium angle
will be restored. No loss of synchronism occurs and the system is
stable.
If damping is accounted for, the linearized swing equation becomes
2
H d Δδ dΔδ
2
+D + Ps Δ δ=0
π f 0 dt dt
(11.44)

or
d Δδ π f 0 d Δ δ π f 0
2
+ D + P Δ δ=0
dt
2
H dt H s
(11.45)

or in terms of the standard second-order differential equation, we

have
d2 Δ δ d Δδ
2
+2 ζ ωn +ω 2n Δ δ=0
dt dt
(11.46 )

where ω n, the natural frequency of oscillation is given by (11.42), and ζ

is defined as the dimensionless damping ratio, given by

ζ=
√
D π f0
2 H Ps
(11.47)

The characteristic equation is

2 2
s +2 ζ ω n s+ω n=0
(11.48)
For normal operating conditions, ζ =D/2
characteristic equation are complex
√ πf0
H Ps
< 1, and roots of the

s1 , s 2=−ζ ω n ± j ω n √ 1−ζ 2
(11.49)
¿−ζ ωn + j ω d
(11.49)

where ω d is the damped frequency of oscillation given by

ω d=ωn √ 1−ζ 2
(11.50)

It is clear that for positive damping, roots of the characteristic

equation have negative real part if synchronizing power coefficient Ps
is positive. The response is bounded and the system is stable.
We now write (11.46) in state variable form. This makes it possible to
extend the analysis to multimachine systems. Let
x 1=Δ δ and x 2= Δ ω= Δ̇ δ then
ẋ 1=x 2 and ẋ2 =−ω2n x 1−2 ζ ωn x 2

Writing the above equations in matrix, we have

[][ ẋ1
=
0
2
1 x1
ẋ2 −ω n −2 ζ ω n x 2 ][ ]
(11.51)

or
ẋ (t )= Ax (t)
(11.52)
where

A=
[ 0
2
1
−ω n −2 ζ ω n ]
(11.53)

This is the unforced state variable equation or the homogeneous state

equation. If the state variables x 1 and x 2 are the desired response, we
define the output vector y (t ) as

y (t )= [ 10 01] [ xx ]
1

2
(11.54)

or
y (t )=Cx( t )
( 11.55)

Taking the Laplace transform, we have

s X (s)−x (0)=AX (s)

or
X (s)=¿
¿

where

(s I − A)=
[ s
2
−1
ωn s+ 2ζ ωn ]
(11.57)

Substituting for ¿, we have

X (s)=
[ s+ 2 ζ ωn 1
−ω2n
2
s ] x (0)

2
s +2 ζ ωn s +ω n

When the rotor is suddenly perturbed by a small angle Δ δ 0 , x1 (0)=Δδ 0

and x 2 (0)=Δ ω0=0, and we obtain
 ( s +2 ζ ωn ) Δ δ 0
Δ δ(s)= 2 2
s +2 ζ ω n s+ω n

and
2
−ωn Δ δ 0
Δ ω( s)= 2 2
s +2 ζ ωn s +ω n
Taking inverse Laplace transforms results in the zero-input response
Δδ0 −ζ ωn t
Δ δ= e sin ⁡( ω d t+θ )
√ 1−ζ 2

(11.58)

and
−ω n Δ δ 0 −ζ ωn t
Δ ω= e sin ⁡ω d t
√ 1−ζ 2

(11.59)

where ω d is the damped frequency of oscillation, and θ is given by

−1
θ=cos ⁡ζ
(11.60)

The motion of rotor relative to the synchronously revolving field is

Δδ0 −ζ ωn t
δ=δ 0 + e sin ⁡( ωd t +θ )
√1−ζ 2

(11.61)

and the rotor angular frequency is

ωn Δ δ 0 −ζ ωnt
ω=ω 0− e sin ⁡ωd t
√1−ζ 2

(11.62)

The response time constant is

1 2H
τ= =
ζ ωn π f 0 D
(11.63)

and the response settles in approximately four time constants, and

the settling time is
ts ≅4 τ
(11.64)
From (11.42) and (11.47), we note that as inertia constant H increases,
the natural frequency and the damping ratio decreases, resulting in a
longer settling time. An increase in the synchronizing power
coefficient Ps results in an increase in the natural frequency and a
decrease in the damping ratio.

INCLUDING CHANGE IN POWER

we will now investigate the system response to small power impacts.
Assume the power input is increased by a small amount Δ P . Then the
linearized swing equation becomes
2
H d Δδ dΔδ
2
+D + Ps Δ δ= Δ P
π f 0 dt dt
(11.66)

or
d2 Δ δ π f 0 d Δ δ π f 0 πf0
2
+ D + Ps Δ δ= ΔP
dt H dt H H
(11.67)

or in terms of the standard second-order differential equation, we

have
2
d Δδ d Δδ 2
2
+2 ζ ωn +ω n Δ δ=Δ u
dt dt
(11.68 )

where
π f0
Δ u= ΔP
H
(11.69)

and ω n and ζ are given by (11.42) and (11.47), respectively.

Transforming to the state variable form, we have
x 1=Δ δ and x 2= Δ ω= Δ̇ δ then
ẋ 1=x 2 and ẋ2 =−ω2n x 1−2 ζ ωn x 2

Writing the above equations in matrix, we have

 [][
ẋ1
=
0
2
1 x1 0
+ Δu
ẋ2 −ω n −2 ζ ω n x 2 1 ][ ] [ ]
(11.70)

or
ẋ (t )= Ax (t)+ B Δ u(t )
( 11.71)

This is the forced state variable equation or the zero-state equation,

and with x 1 and x 2 the desired response, the output vector y (t ) is given
by (11.55). Taking the Laplace transform of the state equation (11.71)
with zero initial states results in
s X (s)= AX (s)+B ΔU (s)

or
X (s)=¿
¿

where
Δu
Δ U (s )=
s

Substituting for ¿, we have

X (s)=
[ 2
s 1 s ][ ]
s+ 2 ζ ωn 1 0 Δ u
−ω2n
2
s +2 ζ ω n s+ω n

or
Δu
Δ δ(s)=
s ( s +2 ζ ω n s+ ω2n )
2

and
Δu
Δ ω( s)= 2 2
s +2 ζ ωn +ω n

Taking inverse Laplace transforms results in the step response

 Δ δ=
Δu
2
ωn [
1−
1
√ 1−ζ 2
e
−ζ ω t
sin ⁡( ω d t+θ )
n

]
(11.73)

where θ=cos−1 ⁡ζ and

Δu −ζ ωn t
Δ ω= e sin ⁡ω d t
ω n √1−ζ 2

(11.74)

Substituting for Δ u from (11.69), the motion of rotor relative to the

synchronously revolving field in electrical radian becomes

[ ]
π f0ΔP 1 −ζ ωn t
δ=δ 0 + 1− e sin ⁡( ω d t+ θ )
Hω
2
n √ 1−ζ 2

(11.75)

and the rotor angular frequency in radian per second is

πf0ΔP −ζ ωn t
ω=ω 0 + e sin ⁡ω d t
H ω n √ 1−ζ
2

(11.76)
Question:1
A 60 Hz synchronous generator having inertia constant H =
9.94 MJ/MVA and the transient reactance Xd’ = 0.3 pu is
connected to an infinite bus through a purely reactive
circuit as shown in the figure. The reactances are marked
on the diagram on a common system base. The generator is
delivering real power of 0.6 pu, 0.8 pf lagging to the infinite
bus at a voltage of 1pu. Assume the pu damping power
coefficient D = 0.138. Consider a small disturbance of ∆δ =
10o . For example, the breakers open and then quickly
closed. Obtain equations describing the motion of the rotor
angle and the generator frequency. Also develop, MATLAB
commands for above two parameters.

SOLUTION:-
The transfer reactance between the generated voltage and the infinite
bus is
0.3
X =0.3+0.2+ =0.65
2

The per unit apparent power is

0.6 −1 ∘
S= ∠ cos ⁡0.8=0.75 ∠ 36.87
0.8

The current is
¿ ∘
S 0.75 ∠−36.87 ∘
I= ¿= ∘
=0.75∠−36.87
V 1.0∠ 0

The excitation voltage is

 E' =V + jXI=1.0 ∠ 0 ∘+( j 0.65) ( 0.75∠−36.87∘ ) =1.35∠ 16.79 ∘

Thus, the initial operating power angle is 16.79∘=0.2931 radian. The

synchronizing power coefficient given by (11.39) is
(1.35)(1)
Ps =Pmax cos ⁡δ 0 = cos ⁡16.79∘=1.9884
0.65

The undamped angular frequency of oscillation and damping ratio

are

ωn ¿
√ πf0
H s
P=
√
(π )(60)
9.94
1.9884=6.1405 rad / sec

ζ ¿
√
D π f 0 0.138
2 H Ps
=
2 √
(π )(60)
(9.94 )(1.9884)
=0.2131

The linearized force-free equation which determines the mode of

oscillation given by (11.46) with δ in radian is
2
d Δδ dΔδ
2
+2.62 +37.7 Δ δ =0
dt dt

From (11.50), the damped angular frequency of oscillation is

ω d=ωn √ 1−ζ 2=6.1405 √ 1−¿ ¿

corresponding to a damped oscillation frequency of

6.0
f d= =0.9549 Hz
2π

From (11.61) and (11.62), the motion of rotor relative to the

synchronously revolving field in electrical degrees and the frequency
excursion in Hz are given by equations

δ=16.79∘ +10.234 e−1.3 t sin ⁡( 6.0 t+77.6966 ∘)

f =60−0.1746 e−1.3 t sin ⁡6.0 t

The above equations are written in MATLAB commands as follows

Question:2
The generator of Example 11.2 is operating in the steady
state at δ 0=16 . 79∘ when the input power is increased by a
small amount Δ P=0 . 2 per unit. The generator excitation and
the infinite bus bar voltage are the same as before, i.e.,
E =1 . 35 per unit and V =1 . 0 per unit.
'

(a) Using (11.75) and (11.76), obtain the step response for
the rotor angle and the generator frequency.
(b) Obtain the response using the MATLAB step function.

SOLUTION:-
(a) Substituting for H , δ 0 , ζ , and ω n evaluated in Example 11.2 in
(11.75), and expressing the power angle in degree, we get
(180)(60)(0.2)
δ=16.79∘ + or
(9.94)¿ ¿

δ=16.79 +5.7631 [ 1−1.0235 e sin ⁡( 6 t +77.6966 ) ]

∘ −1.3 t ∘

Also, substituting the values in (11.76) and expressing the frequency

in Hz , we get
(60)(0.2)
f =60+
2(9.94 )(6.1405) √ 1−¿ ¿ ¿

or
−1.3 t
f =60+ 0.10 e sin ⁡6 t

The above functions are plotted over a range of 0 to 3 seconds and the
result is shown in Figure 11.9.

(b) The step response of the state equation can be obtained

conveniently, using the MATLAB Control Toolbox
[ y , x ]=lsim ⁡(A , B , C , D ,u , t) function or [ y , x ] ¿ step ⁡( A , B ,C , D , iu, t) function.
These functions are particularly useful when dealing with
multimachine systems. [ y , x ]=step ⁡( A , B , C , D ,iu ,t ) returns the output
and step responses of the system. The index iu specifies which input
to be used for the step response. With only one input iu=1. The
matrices y and x contain the output and state response of the system
at the regularly spaced time vector t .
From (11.70), the state equation for is

[][
ẋ1
=
0 1 x1 0
+ Δu
ẋ2 −37.705 −2.617 x 2 1 ][ ] [ ]
and

y= [ 10 01][ xx ]
1

From (11.69), Δ u=(60 π /9.94)(0.2)=3.79. The following MATLAB

commands are used to obtain the step response.