Advanced Power System Analysis

Elective -II
(Course Code: EE 785 )

Lecture No.4
( Conventional FACTS DEVICES)

1
1.1 FACTS devices

| VS | . | VR | 2
| VS | . | VR | |V |
P Sin  (1) and Q  Cos  - R (2)
|X| |X| |X|

Fig.1 Single line diagram of transmission line

From Fig.1 and eqn (1) and eqn (2), it is clear that active and reactive power flow through an ac
transmission line is the function of line reactance (X), magnitudes of sending end and receiving end
voltages and phase angle between sending end voltage and receiving end voltage.

If these parameters can be controlled, then

- The Power Flow Through The Line can be Controlled
- Voltage Level can be Maintained Within Permissible Limit
- Power Loss in the Line can Be Minimized
- Power Transfer capacity can Be Increased
- Stability can be Improved

FACTS devices uses Power Electronic Controlled Circuits to Control OR Change the system
parameters in a Fast and Effective Way to provide above control actions. The full form of FACTS is
“Flexible AC Transmission System”. FACTS devices helps to make the transmission system flexible.
That means a transmission system can be made to perform as per requirement.

2
Various FACTS devices so far developed are:
Conventional FACTS devices:
i) Thyristor Control Reactor (TCR)
ii) Thyristor switched capacitor (TSC)
iii) Fixed Capacitor Thyristor Controlled Reactor (FCTCR)
iv) Thyristor Controlled Series Capacitor (TCSC)
v) GTO controlled Series Capacitor (GCSC)

Switching Converter Type Advanced FACTS devices:

i) Static Synchronous Shunt Compensator (STATCOM)
ii) Static Synchronous Series Compensator (SSSC)
iii) Unified Power Flow Controller (UPFC)
iv) Interline Power Flow Controller (IPF)

3
1.2 Thyristor Controlled Rector (TCR):

Thyristor Controlled Reactor (TCR) can exhibit a

continuously variable shunt inductive reactance from a
fixed value of inductive reactance and generally
connected in shunt mode. Fig.2 shows the basic circuit
diagram of such a scheme.
Fig.2 Thyristor Controlled Reactor (TCR )

-TCR scheme has a fixed inductor ‘L’ in series with ac voltage controller. AC voltage
controller is a pair of thyristors (T1 and T2) connected in anti-parallel. T1 conducts the
positive half cycle of the reactor current IL and T2 conducts the negative half cycle of the
reactor current.

-The basic idea is to change the rms value of the inductor current IL by chopping a portion of
positive half cycle and negative half cycle which is possible with the control of firing angle
of thyristor. If magnitude of IL changes with constant system voltage VS, then the inductor
(L) in series with ACVCis equivalent to variable reactance XL(δ) as a function of firing
angle (δ)of the thyristor.

4
Let us assume that the thyristor T1 is fired during
positive half cycle of VS with some delay angle α. Then
the current through the inductor (L) builds up from ωt =
α as shown in Fig.3

Let R = internal resistance of the inductor

When VS becomes zero at ωt = π, the current through
inductor and T1 will not decreases to zero due to lagging
current drawn by the inductor. Let us say – the current
becomes zero at ωt = β. Since the inductor is connected
across the voltage VS, the voltage across the inductor is
given by:
  
V  I (R  jX )  V (3),
L L L S
Where X  reactance of the inductor
L
From π to β, the input voltage VS has gone to negative and T1
is still conducting as it is carrying positive current till ωt = β.
Therefore negative part of VS appears across the inductor and
Fig.3 Waveforms associated with TCR circuit
wave of VL will be as shown in Fig.3.

At ωt = β, iL becomes zero and T1 turns off naturally at ωt = β. The inductor currents remains zero until T 2 is
fired at ωt = π+α and voltage across the inductor VL = IL. (R +jXL) = 0 from β to π+α. From π+α to 2π, VL
will be negative and equal to VS and so on. Hence, the voltage across the inductor is a chopped sine wave.

5
 di
V L L  Ri (4)
S dt L

Solution of this differential equation gives and equation of iL

and it is given by:
R
V - t
iL  m Sin( t -  )  A e L (5)
Z
L
Where, Z R 2  (L)2 and   tan- 1( )
R
A = a constant whose value depends on the initial condition of the circuit.
The initial condition is: when ωt = α, iL= 0, Or when t=α/ω, iL= 0
R  R
V - V
 From eqn(5) 0  m Sin(  -  )  A e L   A  - m Sin(  -  ) e L
Z Z
R R
V V - t
Hence, i L  m Sin( t -  )  m Sin(  -  ) e . e L
L
Z Z
R
V V - (t   )
OR iL  m Sin( t -  )  m Sin(  -  ) e L (6)
Z Z
Part - I Part - II

Part-I = steady state part and Part-II = transient part

6
 R
V V - (t   )
OR iL  m Sin( t -  )  m Sin(  -  ) e L (6)
Z Z
Part - I Part - II
If α =ϕ, then part-II (transient part) = 0
V
Then i  m Sin( t -  ) i.e. i is contineous and sinusoidal
L Z L

• If ϕ < α , then iL will be discontinuous

• If ϕ > α , then iL will be continuous

In real practice, resistance of an inductor R<< XL

Therefore, assuming R= 0 (Approximately), ϕ =
900 , then eqn(6) becomes:

m Sin( t -  ) m Sin(  -  )
V V
iL   (7)
Z 2 Z 2

7
 m Sin( t -  ) m Sin(  -  )
V V
iL   (7)
Z 2 Z 2

When α = π/2, second part of eqn(7) = 0

That means iL Lags VS by 900, which is natural 900
lagging due to pure inductor as shown in Fig.4(a).
Therefore, α=π/2 is taken as firing angle δ = 0 to analyze
this circuit.

Therefore, from point a to b net current iLis negative,

hence T1 does not conduct from a to b. Therefore, the
waveform of iL( for firing angle=δ) will be as shown by
dotted curve in Fig.4( c) and the waveform of VL will be
as shown in Fig.4(d)
- It is clear that the waveform of iL and VL are not
sinusoidal.
- The fundamental component of VL is calculated as
follow:

8
 1 /2 -  3/2 -  2 
A    V Sint.Cost.dt   V Sint.Cost.dt   V Sint.Cost.dt 
1   m m m 
 0 /2   3/2  

1 /2 -  3/2 -  2 
B    V Sint.Sint.dt   V Sint.Sint.dt   V Sint. sin t.dt 
1   m m m 
 0 /2   3/2  

A 
Then V (1)  (A )2  (B )2 And   tan- 1 1 
L 1 1 1 B 
 1
Then fundamental component of inductor current can be
calculated and it is found to be:
m ( 1 - 2 - Sin2  )
V
I ( 1)   ....... (8)
L L  
V 1
In general current I  OR I  B.V, where B  susceptanc e
X X
Therefore eqn(8) can be written as : I ( 1)  B( ) V
L m
1 2 Sin2 
Where, B(  )   (1- - ) ....... (9)
L  
1   
And X (  )   ωL  ....... (10)
L B( )    2 - Sin2  

From the eqn (10), it is clear that the effective value of the inductive reactance X L(δ) can be varied by
changing the firing angle δ. Hence, the TCR branch is equivalent to a shunt variable inductive reactance.

Since, the inductor draws non-sinusoidal current, this scheme generates harmonics in the system.

9
1.3 Thyristor Switched Capacitor (TSC) :
In Thyristor Switched Capacitor scheme, a proper size
capacitor banks connected in parallel are switched in and out
with the help of thyristor switch pairs as shown in Fig.5
A small value of L is connected in series with C to reduce
possible transient current during switching ON of the
thyristor switch pair.
Fig.6 shows the waveforms of system voltage(VS), capacitor Fig.5 Thyristor switched capacitor scheme
current(Ic), capacitor voltage(Vc), and inductor voltage (VL)
Let the system voltage be given by : vs = VSmax Sint
V
Current th rough capacitor i  Smax Sin (t  90 0 )
c Z
V V
OR i  Smax Cos (t )  Smax Cos (t). ωC
c |X -X | | X - X | ωC
C L C L
X .V
OR i  C Smax Cos (t). ωC
c |X -X |
C L
 X /X 
OR i V  C L  Cos (t). ωC
c Smax  X /X  1
 C L 
X  n2 
Let n 2  c , Then ic  VSmax   Cos (t). ωC (11)
X  n 2  1 Fig.6 Waveforms of TSC scheme
L  
The instantaneous voltage across the capacitor is given:
 n2 
vc  ic .XC  VSmax   Cos (t - 90 0 )
 n 2  1
 
10
 Therefore, peak value of voltage across the capacitor is given:
 n2 
V  V (12)
C(max)  n 2  1 S(max)

 n2 
Here it shall be noted that   is always  1 . | V | |V |
 n2  1 C(max) Smax

n2 n2 X
vc  Vsmax Sin t (13) And ic  Vsmax C cost (13) Where, n2  c
n2 - 1 n2 - 1 XL

A TSC branch can be disconnected at any current zero instant (e.g. at t0 as shown in Fig.6). The capacitor
voltage is at its negative peak value of - VC(max) at this current zero instant. The disconnected capacitor stays
charged to this value. The charged TSC branch can be re-connected at a appropriate instant when the system
voltage is equal to the capacitor voltage (e.g. at tON as shown in Fig.6) so that the switching operation is
transient free.
The number of capacitor banks to be connected in parallel is determined by the maximum allowable step
change of reactive current. Small value of ‘L’ is connected in series with the each capacitor to limit high surge
current through the thyristor due to possible potential difference between the line and the capacitor at the
switching instant.

This method does not introduce the harmonics but it has the following disadvantages:
- Compensation is not continuous. The scheme gives a shunt capacitance equal to C when only one capacitor
is switched on and gives a shunt capacitance equal 2C when two capacitors are switched on in parallel.
Any intermediate values between C and 2C can not be obtained from this scheme.
- Each capacitor bank requires a separate thyristor switch pair.
- For smaller step control, larger number of capacitor banks and thyristor switch pair are required.

11
1.4 Fixed Capacitor Thyristor Controlled Reactor(FCTR) :

The TSC scheme described above cannot provide continuously

variable shunt capacitance. It provides step wise control only.
The FCTCR scheme can provide continuously variable shunt
capacitor effect as well as continuously variable shunt inductive
reactance effect. Fig.7 shows the basic circuit diagram of such a
scheme.
Basically, it is the parallel combination of a fixed capacitor and
TCR branch. The capacitive reactance is fixed and equal to -jXC
= -1/ωC. The effective inductive reactance of the TCR branch is
variable and it is given by: Fig.7 Basic circuit diagram of FCTCR scheme.
1    Therefore, the effective reactance of the parallel
X ()   ωL 
L B( )    2 - Sin2   combination also will be variable and it is given by:

jX ( )  (-jX )
X  L c (13)
eff jX ( )  (-jX )
L c

For capacitive compensation, the value of the fixed capacitor is chosen to compensate for maximum
inductive load that can come in the system. At that instant, the firing angle of the TCR branch fully retarded
(i.e. 900 ) giving zero current through the TCR branch.

If QL(max) = Maximum reactive power that can be consumed by the load, then the fixed capacitor has to
generate QC = QL(max) for unity power factor operation.

V2 Q (max)
L
S  Q (max) OR ωC V 2  Q (max) OR C (14)
X
C
L S L ω.V 2
S

12
When the inductive load decreases, the fixed capacitor will
over compensate. The overcompensation is neutralized
reducing the firing angle of TCR branch with the help of
closed loop control system.

13
For inductive compensation, the value of L and C are so chosen that X L < XC. Therefore IL> IC so that the
scheme displays inductive reactance effect at zero firing angle. The net inductive reactance can be reduce by
increasing the firing angle.

14
1.5 GTO Controlled Series Capacitor
It was described in the previous section that TCR scheme gives variable shunt inductive reactance as a
function of firing angle δ as shown below:

Fig.9 TCR Scheme Fig.10 Waveforms of TCR

2 Sin2  1   
B(  )  
1
(1- - ) X ()   ωL 
L   L B( )    2 - Sin2  
The operating principle of GCSC scheme can be understood with the help of operating principle of TCR
scheme. GCSC scheme gives variable series capacitance effect. It consists of a series capacitor in the
line shunted by a pair of thyristors connected in anti-parallel directions as shown in Fig.11.
When T1 and T2 are turned off (at  =/2)
for positive cycle and negative half cycle
of VC respectively, the line current ILine
continuously flows through the series
capacitor 'C' and the voltage across the
capacitor VC is pure sine wave and lags
the line current by 900. This corresponds
of operation of the scheme with turn off
angle  =/2.
Fig.11 GTO Controlled Series capacitor (GCSC)

15
Let line current i = Im Sinωt (ref pahsor)
Then voltage across the capacitor vc = Im.XC Sin(ωt – π/2 )
= VCmax Sin(ωt – π/2 )
Or vc = VCmax Sin(ωt – π/2 ) = - VCmax Cosωt (15)

It T1 and T2 are turned off lately (i.e. T1 and T2 are turn ON for fraction of positive half cycle and negative
half cycle of VC respectively), the line current diverts through T1 and T2 resulting zero current through the
series capacitor for fraction of positive half cycle and negative half cycle of VC respectively. This
corresponds to turn off angle  = /2 + .
If the T1 and T2 are turned off for remaining fraction of positive half cycle and negative half cycle of V C
respectively, the line current flows through the capacitor thus by charging the capacitor and the capacitor
voltage rises up to its peak value and becomes zero naturally after some time as the line current is AC
current.

The voltage across the capacitor (in general) can be written as follow:
1 t
VC (  )   i dt , Where  can be varied from /2 to 
C The second term is a constant for a given value
1 t I I
m  Cost t  m  Cost  Cos 
of ϒ and the first term is full waveform of vC
OR VC (  )   I m Sin t dt  
C ωC ωC or vC for ϒ= π/2. Therefore, for value of ϒ >
π/2 (say ϒ = π/2 + δ) the waveform of vC will
OR VC (  )  V
Cmax
 Cos t  Cos   get off set by δ as shown in Fig.1.5.
OR VC (  )  - V Cos t  VCmax Cos  (16)
Cmax

16
If this scheme is compared with TCR scheme, waveform
of inductor current (IL) is chopped in TCR scheme,
whereas waveform voltage across the capacitor (vC) is
chopped in the GCSC scheme. The shape of waveform of
chopped IL(δ) in TCR scheme is similar to the shape of
chopped vC(ϒ).
Therefore, the fourier analysis of waveform of vC(ϒ) to
calculate its fundamental component will be similar to
fourier analysis of waveform of IL(δ) to calculate its
fundamental component.
Hence, the fundamental component of vC(ϒ) can be written as:
I 2 Sin2 
V ( )  m ( 1 - - ) ....... (17)
C1 C  
In general,   /2   OR    - /2
 2 (   / 2) Sin(2    ) 
I  2
I Sin(2  ) 
 V ( )  m 1 - - 
OR V ( )  m 1 - 1 
C1 C    C1 C   
I  2  Sin(2  ) 
OR V ( )  m 2 -   (18)
C1 C   
In general, V ( )  Fundament al component Line current  effective capacitive reactance
C1
OR V ( )  I  X ( ) (19) Comparing eqn(18) and eqn(19):
C1 m C

1  2 Sin(2  )  Eqn (20) gives variable series capacitive reactance as the

X ( )   2-   (20) function of turn off angle ϒ.
C C   

17
1.6 Thyristor Controlled Series Capacitor (TCSC)
This is another scheme of controlling series capacitive
reactance. A basic scheme of TCSC is shown in Fig.12. It
consists of a fixed series capacitor shunted by a TCR branch.
A continuously variable series capacitance can be realized
when the firing angle of the TCR branch is varied.
Fig.12 Thyristor Controlled Series Capacitor
The net effective series reactance of the scheme is given by :
1 1 1 1 1 2 Sin2 
  (21) Where,  (1   )
X ( ) j X ( ) - jX X L ( ) ωL  
eff L C
πωL 1
OR X L ( )  (22) and XC   Constant (23)
2 - 2  Sin (2 ) ωC

(-jX ).(jX ( )
OR X ( )  C L (23)
eff jX ( )  jX
L C

It is clear from eqn (23) that the value of XL(α) can be varied by varying the firing angle α and
accordingly the value of net effective reactance (Xeff) will vary. Fig.13 shows the plots of XC, XL(α) and
Xeff with respect to α.

18
 πωL
When    , X L ( ) 
2 - 2  Sin (2 )

πωL
 
2 - 2  Sin (2 )

1 1 1 1 1 1
    
X ( ) j X ( ) - jX  - jX - jX
eff L C C C

This corresponds to the point ‘a’ on the curve of Xeff.

If α is decreased below π, XL(α) will have some

finite value and Xeff increases in negative. That
means Xeff increases in capacitive zone.
When α is decreased to αR, XL(αR) = XC, but
opposite in sign.
1 1 1
Then    0
X ( ) j X ( ) - jX
eff L R C
and X   (This is parallel resonance)
eff
This corresponds to the point ‘b’ on the curve of Xeff.
No current will flow in the line in such a condition.
The range α2 toα1 is to be avoided in the operation
of this scheme, where the line current will be very
low. Fig.13 Plot of XC, XL(α) and Xeff with respect to α

19
 When α is decreased below αR by very small
value, the value of Xeff shift up to + 
This corresponds to the point b’ on the curve of Xeff.

If α is further decreased below αR, Xeff will

have some finite value in inductive zone.
When  is decreased to  / 2,
πωL πωL
X L ( )    ωL  X L
    Sin ( )
2 - 2.  Sin (2. )
2 2
|X |. | X |
And X ( Min )  C L
eff |X X |
L C
This corresponds to the point ‘c’ on the curve of Xeff.

Hence, this scheme can be operated in

Capacitive Mode as well as Inductive
Mode.

However, Series Inductive reactance

compensation is not useful in the
transmission line.

20