PLL and Grid Synchronization

by
Subhash Joshi T G
Email: subhashj@cdac.in

1 of 63
 SUMMARY OF PRESENTATION
 Unit vectors and its significance
 Basics of transformations
 Session 1: Unit Vectors For 3 Balanced/Unbalanced Grid
 Selection of reference variable for control
 Selection of output parameter for PI-controller
 Design of PI-controller constants
 System bandwidth and method to select bandwidth
 Unit vector under unbalanced grid condition
 Test results
 A Simple method for unit vector construction for balanced grid
 Session 2: Unit Vectors For 1 Grid
 Constructing two unit magnitude 900 displaced components
 Mitigating the effect of grid frequency variation
 Approximation method
 Rigorous method
 Test results 2 of 63
 Unit Vectors
What is unit vector ?
 Two unity magnitude fundamental
sinusoidal quantities, which are displaced
by 900 from each other
 One of the unit vector is in phase with grid
voltage
 This should be free from harmonics
 Phase angle error from grid voltage should be minimum
Significance of unit vector ?
 STATCOM is an independent voltage source
 Unit vector helps to synchronize STATCOM voltage and Grid
voltage
 Also unit vector is used to extract the active and reactive power
component separately
 It is also used to separate the individual harmonics in case of
active power filters 3 of 63
 Transformations β axis
XB
 3 to - transformations θ

X  X R  X Y  X B 
1 120o α - axis
-120o XR
2

X 
3
X Y  X B  XY

2 β axis
- to D-Q transformations Q axis

 X d   U1 U 2   X   Xres
X   
Xβ
  X  D axis
 q  2 U U 1   
θ ρ
U1 and U2 unity magnitude Xα α axis
components, where U2 lag U1 by 900
For grid connected system U1 is
aligned along R-phase grid voltage
4 of 63
 SESSION 1

UNIT VECTORS FOR THREE PHASE

BALANCED/UNBALANCED GRID

5 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Let the grid voltages are,
VR
VR  Vg sin  g t V
VY 3 -
VY  Vg sin(  g t  120) VB
V

VB  Vg sin(  g t  120)
 Objective,
U1  sin g t
U 2   cos g t
 Apply three phase to two phase (-) transformation,
1 3
V  VR  (VY  VB )  Vg sin  g t
2 2
3 3
V  (VY  VB )   Vg cos  g t 6 of 63
2 2
 Unit vector for Balanced/Unbalanced
Grid condition
 If U1 and U2 are known then V and V can be transformed to
D-Q axis VR - V D-Q Vd
 Assume U1 and U2 are known, then V
Y 3 -
Vd   U1 U 2  V  VB V Vq
V     V 
 q  2  U U 1   
U1 U2
 Let U1 is not synchronized to VR and its frequency is g’, then,
 3  3  
Vd   sin  g t  cos  g t   2 g
V sin  t
g  V
2 g cos(( g   g )t ) 

V  cos   t sin   t   3 
 3 
  
q g g   Vg cos  g t   Vg sin((  g   g )t ) 
 2  2 
 When U1 gets synchronized with VR, then g=g’ and,
Vd   3 Vg 
V    2 
 q   0  7 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 If Vq = 0 is ensured, then U1 will be synchronized to VR
 To ensure Vq = 0, feedback with simple PI-controller can be
used

VR - V D-Q Vd
VY e(t) ki
3 - -+ kp 
VB V Vq s
U1 U2 0

 PI-controller will be sufficient,

 Since Vd and Vq are d.c. quantities and variation in e(t) with
time is minimal

8 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 What should be the output parameter of PI-controller ?
 PI-controller performs well when the selected output parameter magnitude
swing is minimal
 Our objective is to obtain U1 and U2, but they are sinusoidally varying
quantities and results in large swing in magnitude
 The angle of U1 and U2 are also will be varying with time in large extend
 Instead the frequency of U1 and U2 can be thought of selecting as output
parameter of PI-controller
U1  sin t , U 2   cos t

VR - V D-Q Vd
VY 3 - -+
e(t)
kp 
ki 
VB V Vq s
U1 U2 0
9 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 What should be the output parameter of PI-controller ?
 If the grid frequency variation is minimal, then
 PI-controller performance can be further improved by selecting the output
of PI-controller as ∆ (variation in grid frequency) instead of absolute grid
frequency, 

VR - V D-Q Vd
VY e(t) ki ∆ 1 t
3 - -+ kp  -+
VB V Vq s s

U1 U2 0 ref

U1 Lookup
ref is the nominal grid frequency U2 Table
10 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Design of PI-controller constants ?
 At time t=0, let grid and unit vectors (U1 and U2) frequency be g
 At time t=0+, grid frequency changed from  g to  g and unit vector
frequency remains at g
 3 to - transformation of grid voltage is,
3 3
V  Vg sin  g t , V   Vg cos  g t
2 2
 - to D-Q transformation gives,
 3   3  
Vd   sin  g t  cos  g t   2 g
V sin  t
g  V
2 g cos((  g   g )t ) 
V   cos  t sin  t   3  3 
 q  g g   V cos   t   Vg sin((  g   g )t ) 
 2  2 
g g

11 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Design of PI-controller constants ?
 Assuming grid frequency variation is minimal, then
sin((  g   g )t )  ( g   g )t 3
V cos((    )t )

Vd   2 g g g 

V   3 
 Hence Vq is,  q   Vg sin((  g   g )t ) 
2 
3
Vq  Vg ( g   g )t
2
VR(’g)
- V(’g) D-Q Vd
VY(’g)
e(t) ki ∆ 1 gt
VB(’g)
3 - - kp  -+
V(’g) Vq + s s
0 ref

U1(g) U2(g) U1(g) Lookup

U2(g) Table
12 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Design of PI-controller constants ?
 Assuming grid frequency variation is minimal, then
sin((  g   g )t )  ( g   g )t 3
V cos((    )t )

Vd   2 g g g 

V   3 
 Hence Vq is,  q   Vg sin((  g   g )t ) 
2 
3
Vq  Vg ( g   g )t
2
VR(’g)
- V(’g) D-Q Vd
VY(’g)
e(t) ki ∆ 1 gt
VB(’g)
3 - - kp  -+
V(’g) Vq + s s
0 ref

U1(g) U2(g) U1(g) Lookup

U2(g) Table
13 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Design of PI-controller constants ?
 Assuming grid frequency variation is minimal, then
sin((  g   g )t )  ( g   g )t 3
V cos((    )t )

Vd   2 g g g 

V   3 
 Hence Vq is,  q   Vg sin((  g   g )t ) 
2 
3
Vq  Vg ( g   g )t
2

e(t) ki ∆ 1 gt
- kp  -+
Vq + s s
0 ref
 g t
3 +
Vg -
2 14 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Design of PI-controller constants ?
 The block diagram can be further reduced by removing the quantities not
taking part in transients
 ref is constant and no effect on transients
 Once grid voltage is changed from  g to  g ,  g
can be assumed to be constant
 Simplified block diagram is,

Vqref=0 ki 1 e(t) ki ∆ 1 gt

+
- kp  - kp  -+
s s Vq + s s
0 ref
3
Vg
 g t
2 3 +
Vg -
2 15 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Design of PI-controller constants ?
 Open loop transfer function of the system is given by,

3  sk p  ki  1  Vqref=0 ki 1
G( s) H ( s)  Vg    +
- kp 
2  s  s  s s

 s 
 1 
3
Vg
3  ki k p  2
G ( s ) H ( s )  Vg  2

2   
  s  

 k i  

16 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Design of PI-controller constants ?
 Find asymptotic bode plot of the system,
 s 
 For zero:  1 
 Corner frequency,  z  ki k p 3  ki k p 
 Magnitude plot: +20dB/decade G ( s ) H ( s )  Vg  2

2   
 Phase plot: +450/decade
 For poles:
  s  

 k i  
 Corner frequency,  p  ki
 Magnitude plot: -40dB/decade
 Phase plot: -1800

17 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 s 
 Design of PI-controller constants ?  1 
3  ki k p 
 Closed loop system will be stable, G ( s ) H ( s )  Vg  2

2   
 If the open loop gain crosses 0dB (unity  s  
k 
gain) with -20dB/decade and ensuring  i  

atleast -1350 Phase margin

 If z > p, leads to a phase ki k p
close to -1800 makes system
prone to unstable
ki
 Hence to make system more
stable, keep z <= p
 Since phase plot of zero has a
slope of +450/decade, for z =
p, the phase is -1350 leads to
phase margin of 450

18 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 s 
 Design of PI-controller constants ?  1 
3  ki k p 
 z = p implies, G ( s ) H ( s )  Vg  2

2   
 s  
k 
ki  i  

 ki
kp
k p  ki
ki
 One more component present
in the open loop transfer
function is,
3 V
2 g
 This will not affect the phase
plot, but gain cross over
frequency will be pushed further
away increasing the bandwidth
of the system
19 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
k p  ki
 Find system bandwidth ?
Gain = 0dB
 Defined as the frequency at
which the closed-loop 20dB/dec
-40dB/dec
magnitude is equal to -3 dB g
 For phase of -1200, the open
loop bandwidth and closed loop 20dB/dec
bandwidth are found to be ki
closer
 Since gain at ki is 0dB, the
gain cross over frequency of
open loop system is,

 g  ki

-1350

20 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
k p  ki
 Find system bandwidth ?
Gain = 0dB
 The modified crossover
frequency considering 3/2Vg is, 20dB/dec
-40dB/dec
 Gain at  g  ki is, g

3 
20 log  Vg  20dB/dec
2  ki

 Gain at new crossover

frequency ( g ) is 0dB
 Slope during this time is
-20dB/decade

3 
0  20 log  Vg 
 2   20
 
-1350
log  g  log ki
21 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
k p  ki
 Find system bandwidth ?
Gain = 0dB
 The modified crossover
frequency considering 3/2Vg is, 20dB/dec
-40dB/dec
3 g

 g  Vg k i
2 20dB/dec
ki
 Closed loop bandwidth (appx.)
is,
3
BW  Vg k p
2
 Once the bandwidth is fixed
proportional constant (kp) can
be found

-1350

22 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
k p  ki
 How to fix the bandwidth ?
 Bandwidth is decided by the harmonics present in Vd and Vq components
as well as the response time requirement in transient conditions
 For a balanced three wire system the minimum harmonics expected on
Vd and Vq are 300Hz (transformation of 5th and 7th harmonics)
 Here bandwidth of 30Hz to 60Hz will be sufficient for proper
attenuation of harmonics in grid voltages

 In Summary:
 Based on the application and transient requirement fix the bandwidth
 If grid voltage peak is known, proportional constant (kp) can be found
BW
kp 
3 V
2 g
 Once kp is known integral constant ki can be computed

ki  k p
2
23 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Unit vector under unbalanced grid condition ?
 Under unbalanced grid conditions, the grid voltage contains fundamental
positive sequence as well as fundamental negative sequence components
 Let us construct the unit vector such that it is synchronized with
fundamental positive sequence component
 As earlier fundamental positive sequence present in the grid voltage when
transformed to D-Q reference frame reflected as d.c. component and
fundamental negative sequence present in the grid voltage reflected as
100Hz component
 Since we required only d.c. component in Vq, 100Hz component can be
easily removed by using simple low pass filter of corner frequency around
10Hz

24 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Unit vector under unbalanced grid condition ?
 C is the corner frequency of low pass filter
 Low pass filter will not affect the information of fundamental positive
sequence, as it is a d.c. quantity

VR
- V D-Q Vd
VY s e(t) ki ∆ 1 t
VB
3 -
s  C
-+ kp  -+
V Vq s s
0 ref

U1 U2 U1 Lookup
U2 Table

25 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Test results
 Transient performance analysis carried out by simulation using PSIM
simulation package
3 Grid
VR VY VB

400

200

-200

-400

Unit vectors
cosrow sinrow

0.5

-0.5

-1

0.06 0.08 0.1 0.12 0.14

Time (s)

50% step voltage sag Voltage restored

appeared in VR and VY 26 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Test results
 Transient performance analysis carried out by simulation using PSIM
simulation package
3 Grid
VR VY VB
400
200
0
-200
-400

Unit vectors
cosRow sinRow
1
0.5
0
-0.5
-1

t
theta
400
300
200
100
0
0.36 0.38 0.4 0.42 0.44 0.46
Time (s)

Step change in frequency (from 50Hz to 30Hz)

appeared in VR, VY and VB
27 of 63
 Unit vector for Balanced/Unbalanced
Grid condition
 Test results
 Unit vector implemented for unbalanced grid condition in TI’s DSP
TMS320F2812
Unit vector Grid voltages
Grid voltage

Unit vector

Balanced grid voltage VR and VB are nominal, VY=0

Unit vector t

Unit vector and angle t 28 of 63

 A Simple method for
Unit vector in Balanced Grid condition
 A simple method exists based on trigonometry to compute
unit vector for balanced grid condition
 Transform three phase grid voltage to - co-ordinate

V t   Vg sin  g t , V t    Vg cos  g t
3 3
2 2
 The output of Low Pass Filter (LPF) is given by,
V s       V s 
s s

V s , V s 
s  c s  c

VR V s V’
- s  C
VY
3 s
VB s  C V’
V
29 of 63
 A Simple method for
Unit vector in Balanced Grid condition
 Output of LPF in time domain is,
3 V 3 V
V t   2 sin  g t   , V t    2 cosg t   
g c g c

 g 2  c 2 g 2  c 2
g
where, tan  
c
 V (t ) and V (t ) are always 90 displaced
Unit vectors
0

irrespective of grid frequency and corner

frequency of LPF
 At the zero crossing of V (t ) , V (t ) will be
in peak and vice versa
 Unit magnitude of V (t ) and V (t ) can be
obtained by dividing the term by its own
magnitude,
30 of 63
 A Simple method for
Unit vector in Balanced Grid condition
 Block diagram of unit vector construction

X1 U1t   sin  g t   
X1 / Y1
Y1
VR V s V’(t)
- ZCD S&H ABS
s  C Trigger
VY
3 s Trigger
VB s  C V’(t) ZCD S&H ABS
V
Y2

U 2 (t )   cos g t   
X2 / Y2
X2

31 of 63
 A Simple method for
Unit vector in Balanced Grid condition
 Let us perform the following operations,
F1 (t )  V (t )  V (t )
3 Vg  c
 2 sin  g t   
c   g2 2

F2 (t )  V (t )  V (t )
3 Vg  c
 2 cos g t   
c   g2 2

c   g
where, tan  
c   g
 Phase angle () is minimum when c=250
 For c=250,  is zero when grid frequency is 50Hz 32 of 63
 A Simple method for
Unit vector in Balanced Grid condition
 Complete block diagram of unit vector construction
where, c=250
X1 U1 t   sin  g t   
X1 / Y1
V s V’ Y1
VR + ZCD S&H
- s  C - ABS
Trigger
VY
3 Trigger
s
VB ++ ZCD S&H ABS
V s  C V’ Y2


U 2 (t )   cos g t   
X2 / Y2
X2
 This method will introduce a small phase angle error when grid
frequency varies
2
 Each harmonics is reduced by 2 when compared to
fundamental 1 h
33 of 63
 A Simple method for
Unit vector in Balanced Grid condition
 Test results
Grid voltage

Unit vector

34 of 63
 SESSION 2

UNIT VECTORS FOR SINGLE

PHASE GRID

35 of 63
 Overview of the presentation
 Constructing two unit magnitude 900 displaced components
 Mitigating the effect of grid frequency variation
Approximation method
Rigorous method

36 of 63
 Overview of the presentation
 Constructing two unit magnitude 900 displaced components
 Mitigating the effect of grid frequency variation
 Approximation method
 Rigorous method

37 of 63
 Innovative method for constructing
unit vector
 Let R-phase grid voltage be, VR = Vg sin(St)
 Let the above voltage is passed through a LPF of corner
frequency, c
 Using Laplace analysis
Vg  C  S  1  C   S s 
 
FR(s)  2 2 
 
 
 2 2
 2 2 
 S  C  s  C   S  s   S s   S  +
-
c
VR FR
 The steady state output, FR(t) is given by s
Vg  C S
FR(t)|steady  sin  S t    and tan  
 S 2  C 2 C
 Transient term of FR(t) is given by,
V gcs ct
FR(t)|transient= e
 
 2   2 

c s 
38 of 63
 constructing unit vector (contd.)
Vg  C
 Let the FR(t) is again pass through a FR(t)|steady  sin  S t   
 S  C
2 2

LPF of same corner frequency, c VR = Vg sin(St)

+
-
c +
-
c
VR FR’
s FR s
 Using Laplace analysis

FR s   VgC  S
2 1 
 
1
 2
  
2 1
 
s
  
2
  2 1
 

 
2 
2 2 2 
S
2
 C
2

C
s  C
S C
s  C 2
s  S
C 2 2 C S
s  S 
2

 The steady state output, FR’(t) is given by


FR t   
VgC
2

cos S t    and cos 

2C S ,
 
C   S  2 2

 S
2
 C
2
  S 2  C 2 
sin

 S 2  C 2  
 Transient term of FR’(t) is given by,
V g c 2s  ct
FR’(t)|transient=  2
 c  c 2  s 2 t e
 
2    

c  s 
2 2 39 of 63

 constructing unit vector (contd.)
 Subtract FR’(t) from FR(t), let the 
FR t   
VgC
2

cos S t   
result be FR’’(t)  S
2
 C
2

VR = Vg sin(St)

+
-
c +
-
c
VR FR’
s FR s
-
+ FR’’

 The steady state output, FR’’(t) is given by


FR t  
Vg  C  S
sin  S t    and cos 
2C S ,
sin  
C   S  2 2

 S
2
 C
2
 
 S  C
2 2

 S 2  C 2  
 Transient term of FR’’(t) is given by,
  
V g cs   2   2  
 s c   t
 ct e c
FR’’(t)|transient=

c s 


 2   2   
 
c
2  2
s  

40 of 63
 constructing unit vector (contd.)
 Comparing steady state term of FR’(t) and FR’’(t)
VgC - c - c
2

FR t    cos S t   
VR + +

 
FR’
s FR s
 C
2 2
S
-
+ FR’’

 Vg  C  S
FR t   sin  S t   
 S
2
 C
2

 FR’(t) and FR’’(t) are always 900
displaced irrespective of grid
frequency and corner frequency of
LPF
 At the zero crossing of FR’’(t), FR’(t)
will be in peak and vice versa
 Unit magnitude of FR’(t) and FR’’(t)
can be obtained by dividing the term
by its own magnitude 41 of 63
 constructing unit vector (contd.)

VR +
-
c +
-
c FR’ X F1= -cos(St+)
X/Y
s FR s Y
ZCD S&H ABS
Trigger
Trigger
ZCD S&H ABS
-
Y
+
X/Y
FR’’ X F2 =sin(St+)

 Inference from the above result

VR = Vg sin(St)
 F2(t) is phase shifted from grid voltage
by an angle 

42 of 63
 constructing unit vector (contd.)
 Inference from the above result (contd.) VR = Vg sin(St)

F1(t) = -cos(St+)
 If corner frequency of LPF (c) is set F2(t) = sin(St+)
equal to grid frequency (s), i.e. c = s :   S
2 2

sin  
 
C

phase shift  = 0 S  C
2 2

2C S
F2(t) is in phase with grid voltage cos 

 S 2  C 2 
and F1(t) is lagging the grid voltage by 900
 Fix the corner frequency of LPF (c) is equal to s = 250
rad/sec, where grid frequency fs = 50Hz
 Let Grid frequency varies a maximum of 10% (45Hz to
55Hz)
 Phase shift,  = 60 for -10% of grid frequency variation
 Loose the proper synchronization

43 of 63
 constructing unit vector (contd.)
 Phase error with variation in grid frequency
1ph without comp
7

6
6.0
5

3
phase error

0
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10
-1

-2

-3

-4

-5

-6

-7
% ge of grid frequency variation

44 of 63
 Overview of the presentation
 Constructing two unit magnitude 900 displaced components
 Mitigating the effect of grid frequency variation
Approximation method
Rigorous method

45 of 63
 constructing unit vector (contd.)
 Mitigating the effect of grid frequency variation
 Give a  variation for the grid frequency s in the peak
of FR’’(t) V = V sin( t)R g S
 Substitute c= s and solving will give 
F t  
V 
sin  t   
 
g C S

 C
R 2 2 S

1     
   S

Vg     
 s 
F      
' ' 
R s s 2  2
c  s    1    
1      


      
 s  2 s  
 
Vg   
2
F ''       1
if   is negligible compare to 1
R s s 2 2  s 
c  s
 Peak of FR’’(t), F’’R(peak), is more or less independent of
grid frequency variation
F’’R(peak)  Vg/2 for c = s
46 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
 The following trignometric relation can be V = V sin( t) R g S

used to eliminate the phase angle  F (t) = -cos( t+) 1 S

sin  S t   Vg cos  cos S t   Vg sin   Vg sin  S t

F (t) = sin( t+) 2 S

  S
2 2

 cos S t   Vg cos  sin  S t   Vg sin   Vg cos  S t sin  
 
C

S  C
2 2

2C S
 Vgcos = 2F’’R(peak) cos 

 S 2  C 2 
 Vg  C  S
FR t   sin  S t   
 Solve the following mathematical relation 
 S 2  C 2 
V gcs

2 FR( peak )  Vg 
2Vg C
2

 Vg  Vg
  S
2 2
 F ''
R peak 

 
 2   2 

   
C c s
 
 C  
2 2 2 2
S S C F’’R(peak)  Vg/2 for c = s
VgC
2

2 FR( peak )  Vg  Vg sin  FR t   
 S
2
 C
2
 cos t   S

V gc 2
Vg sin   2 FR( peak )  2 FR( peak ) F'
R peak 

 
c 2  s 2 
 47 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
 Rewriting the relation V = V sin( t)
V gc 2
R g S

sin  S t   Vg cos  cos S t   Vg sin   Vg sin  S t Rpeak  c2  s2
F ' 
 
 
 
V gcs
F '' 
R peak   
 2   2 
 Substituting Vg  C  S

c s 

FR t   sin  S t   
 Vgcos = 2F’’R(peak) 
 S 2  C 2 
 Vg sin   2 FR( peak )  2 FR( peak ) VgC
2

FR t   
 S
2
 C
2
 cos t  
S

Vg sin  S t  U1  sin  S t   (2 FR( peak ) )  cos S t   2 FR( peak )  2 FR( peak ) 

cosst  
 
U  2FR '' (t )  2FR ' (t )  2 F 
1  R( peak ) 

2VgC  S  C   S 
2 2
 C   S 
U1  sin  S t      tan  
 S
2
 C
2
 and
S

48 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
 Rewriting the relation V =V R g sin(St)
V gc 2
F' 
 cos S t   Vg cos  sin  S t   Vg sin   Vg cos  S t R peak  


c 2  s 2 

V gcs
F '' 
R peak   
 2   2 
 Substituting  Vg  C  S

c s 

FR t   sin  S t   
 Vgcos = 2F’’R(peak) 
 S 2  C 2 
 Vg sin   2 FR( peak )  2 FR( peak )
VgC
2

FR t   
 S
2
 C
2
 cos t  
S

 Vg cos  S t  U 2   cos S t   (2 FR( peak ) )  sin  S t   2 FR( peak )  2 FR( peak ) 

U 2  2 FR (t )  2FR( peak ) cos S t    2FR( peak ) sin  S t  



2VgC  S  C   S 
2 2
 C   S 
U2   cos S t      tan  
 S
2
 C
2
 and
S

49 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
 Comparing steady state term of U1(t) and U2(t) V = V R g sin(St)

2VgC  S  C   S 
2 2

U1  sin  S t     
 S
2
 C
2

2VgC  S  C   S 
2 2

U2   cos S t     
 S
2
 C
2

 Inference from the above result
 U1(t) and U2(t) are 900 displaced
 At the zero crossing of U1(t), U2(t) will be in peak and vice
versa
 Unit magnitude of U1(t) and U2(t) can be obtained by
dividing the term by its own magnitude
 U1(t) is phase shifted from grid voltage by angle (-)
50 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
VR = Vg sin(St)
 Inference from the above result (contd.)  C   S 
tan  
 Fix the corner frequency of LPF (c) is equal S
to s = 250 rad/sec, where grid frequency fs sin  
  S
2 2

 
C

S  C
2 2
= 50Hz
2C S
 Let Grid frequency varies a maximum of cos 

 S 2  C 2 
10% (45Hz to 55Hz)
 Give a  variation and substitute c= s in
the phase angle (-)
tan   tan 
tan     
1  tan  tan 
2
1   
 
2  S    
2   
3
tan      S  S    
if 2  1
and   is negligible compare to 1
 C  S      
 s 
 2 
 s 

1    
   
  S  51 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
 Phase error with variation in grid frequency
Phase shift, (-) = 0.320 for -10% grid frequency variation
1ph with comp (appx)

1
phase error

0
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10
0.32
-1

-2

-3
52 of 63
% ge of grid frequency variation
 constructing unit vector (contd.)

FR’(peak)
VR +
-
c +
-
c FR’(t)
F1(t)

s FR s X
Y + U1(t)
ZCD S&H ABS X/Y -
+
Trigger
Trigger
+
ZCD S&H ABS X/Y -
+
- Y
X
U2(t)
+
FR’’(t)
FR’’(peak) F2(t)

cosst  
U1(t)  
X U  2FR '' (t )  2FR ' (t )  2 F 
Y cos 1  R( peak ) 
ZCD S&H ABS X/Y
U 2  2 FR (t )  2FR( peak ) cos S t    2FR( peak ) sin  S t  

Trigger
Trigger F1(t) = -cos(St+)
ZCD S&H ABS X/Y F2(t) = sin(St+)
U2(t) Y
X
sin
53 of 63
 Overview of the presentation
 Constructing two unit magnitude 900 displaced components
 Mitigating the effect of grid frequency variation
Approximation method
Rigorous method

54 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
VR = Vg sin(St)
 The approximation made in the earlier   S
2 2

sin  
 
C
derivation is: S  C
2 2

2 FR( peak )  Vg  Vg sin  V gcs

F '' 
Vg   
2 R peak   
 2   2 
F '' c  s  1
if   is negligible compare to 1 
c s 
R 2 
2  s  V gc 2
s  s   F'
R peak 

 
c 2  s 2 
 
 The above method is Approximation method
 Another method of finding out Vgsin(st) is:
Vg  C Vg  C
FR peak   FR peak   C   S  & FR peak   FR peak   C   S 
 S
2
 C
2
  S
2
 C
2

Vg  C
 
2 2

 S
 
2 2

FR peak   FR peak  FR peak   FR peak    2

 C 
2 2 C

 Vg C  S
2 2
 Vg sin 

 
S

S  C
FR peak   Vg C 2 
2 2

 2

  S   C
2
 

55 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
Rewriting the relation VR = Vg sin(St)
V gc 2
sin  S t   Vg cos  cos S t   Vg sin   Vg sin  S t
F' 
R peak   
c 2  s 2 
 
V gcs
 cos S t   Vg cos  sin  S t   Vg sin   Vg cos  S t F ''
R peak  


 2   2 

c s 
 Substituting 
U  2FR '' (t )  2FR ' (t )  2 F  cos st  

 
1  R ( peak ) 

 Vgcos = 2F’’R(peak) U 2  2 FR (t )  2FR( peak ) cos S t    2FR( peak ) sin  S t  


 2  Vg  C  S
 F 2  F  
FR t   sin  S t   
 V g sin
 R( peak )

R( peak)  
 S 2  C 2 
F  VgC
2
R( peak) FR t   
  cos t  
  cos t    C
2 2 S


 FR( peak )  FR( peak )
2 2 S

U1  Vg sin  S t  2 FRt    
FR( peak )
S

 

 
 FR( peak )  FR( peak )
U 2  Vg cos  S t  2 FR( peak ) cos S t     
2 2
 sin  t  

FR( peak )
S

 

56 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
 Phase error with variation in grid frequency
Phase shift, (-) = 00 for any grid frequency
1ph with comp (rigorous)

1
phase error

0
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10
0

-1

-2

-3 57 of 63
% ge of grid frequency variation
 constructing unit vector (contd.)
- FR’(peak)
VR +
-
c + c FR’(t) X F1(t)
X/Y + U1(t)
s FR s +
Y1
ZCD S&H ABS
Trigger Y12
+ X
-
Y22 X Y
Trigger
ZCD S&H +
ABS Y2 2
- X/Y - U2(t)
+ X F2(t)
2
FR’’(t)
FR’’(peak)

U1(t) X 
U1  Vg sin  S t  2 FRt   

 FR( peak )  FR( peak )
2 2
 cos t  
Y cos 
FR( peak )
S
ZCD S&H ABS X/Y 
 

Trigger  
 FR( peak )  FR( peak )
U 2  Vg cos  S t  2 FR( peak ) cos S t     
2 2
 sin  t  

FR( peak )
S
Trigger 
 

ZCD S&H ABS X/Y F1(t) = -cos(St+)
U2(t) Y
X
sin
F2(t) = sin(St+) 58 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
 Test results
 Grid frequency (simulated using function generator)
is varied at t=t1 from 50Hz to 45Hz

Grid voltage and unit vector without Grid voltage and unit vector with
compensation for grid frequency variation compensation for grid frequency variation

59 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
 Test results
 Steady state waveform at 25Hz and 75Hz with
approximation method

Frequency = 25Hz Frequency = 75Hz

60 of 63
 constructing unit vector (contd.)
Mitigating the effect of grid frequency variation (contd.)
 Test results
 Step change of frequency from 25Hz and 75Hz as
well as from 75Hz to 25Hz

Grid voltage and unit vector without Grid voltage and unit vector with
compensation for grid frequency variation compensation for grid frequency variation
61 of 63
 Conclusions
 Single grid voltage is considered for the construction of unit
vector
 Two unity magnitude fundamental sinusoidal quantities,
which are displaced by 900 from each other
 One of the unit vector is in phase with grid voltage
irrespective of the grid frequency variation

62 of 63
Thank you

63 of 63