@ 
ACTIVE CURRENT SHAPING FOR BETTER 
UTILITY INTERFACE 
A dissertation submitted to the 
Department of Electrical Engineering, University of Moratuwa 
in partial fulfillment of the requirements for the 
degree of Master of Science 
by 
MATHARAGE HASATH CHANDIKA PERERA 
SRf lanka 
M O R A T U W A 
Supervised by: Dr. J. P. Karunadasa 
Department of Electrical Engineering 
University of Moratuwa, Sri Lanka 
January 2009 
University of Moratuwa 
9 2 9 7 0 
9 2 9 7 0 
6x\< z(p«f3 
T W -
D E C L A R A T I O N 
T h e w o r k s u b m i t t e d in th is d isser ta t ion is the resul t of m y o w n inves t iga t ion , e x c e p t 
w h e r e o t h e r w i s e s ta ted. 
It has not a l ready b e e n a c c e p t e d for any deg ree , a n d is a lso not be ing c o n c u r r e n t l y 
s u b m i t t e d for any o ther degree . 
M.H. C h a n d i k a Pere ra 
I e n d o r s e the dec la ra t i on by the cand ida te . 
Dr. J .P. K a r u n a d a s a 
C O N T E N T S 
Dedica t ion i 
Abs t rac t v 
A c k n o w l e d g e m e n t vii 
List of F igu res vii i 
List of T a b l e s ix 
C H A P T E R I 
I N T R O D U C T I O N 
1.1. P o w e r s y s t e m h a r m o n i c s & e f fec ts of h a r m o n i c s 1 
1 . 2 . M e a s u r e s of h a r m o n i c s 3 
1 .3 . Impo r tance of h a r m o n i c mi t iga t ion 3 
1 . 4 . M e t h o d s of h a r m o n i c mi t iga t ion 4 
1 .5 .Ac t i ve P o w e r Fi l ters 6 
1 .6 .Ob jec t i ves 8 
1 .7 .Thes is O r g a n i z a t i o n 8 
C H A P T E R 2 
M A T E R I A L S A N D M E T H O D S 
2 .1 .Acqu is i t i on of vo l t age and cur ren t s igna ls of the p o w e r a p p l i a n c e 
2 . 1 . 1 . S a m p l e load 9 
2 .1 .2 .Da ta acqu is i t i on card '. 10 
2 . 1 . 3 . H a r d w a r e c i rcui t 11 
2.1.4. Acqu is i t i on of vo l t age and cur ren t s igna ls by M A T L A B 12 
2 . 1 . 5 . M A T L A B Data acquisition tool box 13 
2 .2 .Ana l ys i s of acqu i red s igna ls us ing M A T L A B 15 
2.3. D e t e r m i n a t i o n of the approp r ia te f i l ter cur ren t 18 
2 .4 .S imu la t i on to i m p l e m e n t act ive p o w e r f i l ter 
2 .4 .1 .C i rcu i t a r r a n g e m e n t of the s imu la t i on 19 
2 .4 .2 .S imu la t i on to ob ta in c i rcui t p a r a m e t e r s 21 
2.5. Va l i da t i on of the acqu i red c i rcui t p a r a m e t e r s 23 
C H A P T E R 3 
R E S U L T S 
3.1. E lect r ica l p a r a m e t e r s of the load 
3.1.1. V o l t a g e s igna l o f the s a m p l e load 24 
3 .1 .2 .Cu r ren t s igna l of the s a m p l e load 25 
3.1.3. E lect r ica l p a r a m e t e r s of the s a m p l e load 26 
3 .2 R e q u i r e d cu r ren t f r om the f i l ter 
3 .2 .1 .Cur ren t of the capac i to r cha rge r 26 
3.2.2. Fi l ter cu r ren t w i th capac i to r cha rge r 27 
3.3. S imu la t i on resul ts 
3 .3 .1 .Se lec t ion of the inductor 28 
3 .3 .2 .Se lec t i on of the s tep up t r ans fo rmer 31 
3 .3 .3Se lec t i on of the sw i t ch ing f r equency 33 
3 .4 Va l i da t i on 
3.4.1 .Va l ida t ion for load var ia t ions 36 
3 .4 .2Va l ida t ion for supp ly vo l tage var ia t ions 39 
i i i 
C H A P T E R 4 
D I S C U S S I O N 
R e f e r e n c e s 4 7 
A n n e x - 1 
A n n e x - 2 
A n n e x - 3 
4 8 
4 9 
50 
A B S T R A C T 
I nc reased use of non l i nea r e lec t r ica l l oads in jects h a r m o n i c cu r ren t s to p o w e r 
sys tems. H igh leve ls of powe r s y s t e m h a r m o n i c s c rea te vo l t age d i s to r t i on a n d 
en la rge p o w e r qua l i ty p rob lems . H a r m o n i c s resul t in poor p o w e r fac tor , l owe r 
e f f ic iency a n d in te r fe rence to ad jacen t c o m m u n i c a t i o n s y s t e m s . T h e h a r m o n i c 
cur rents f l ow into the uti l i ty supp l y l ines p r o d u c e s ex t ra losses. A n ac t ive p o w e r f i l ter 
uses a sw i t ch ing inver ter to p r o d u c e h a r m o n i c c o m p e n s a t i n g cu r ren ts . 
The m a j o r ob jec t i ve of th is p ro jec t w a s to e l im ina te e f fec ts of h a r m o n i c s a n d to 
improve p o w e r fac to r of a typ ica l non l i nea r load. A t t e m p t s w e r e m a d e to app l y ac t i ve 
power f i l ters for cu r ren t s h a p i n g of a spec i f i c load, con t ra ry to its c o m m o n 
app l i ca t ions of app ly ing at the po in t o f c o m m o n coup l ing . 
The Na t iona l I ns t rumen ts U S B - 6 0 0 8 mu l t i func t ion da ta acqu is i t i on ( D A Q ) m o d u l e 
w a s used to acqu i re da ta f r om the s a m p l e load viz. the c o m p u t e r p o w e r supp ly . A 
potent ia l d i v ide r w a s i nco rpo ra ted to the c i rcui t to acqu i re v o l t a ge s igna l . C u r r e n t 
s ignal w a s a c q u i r e d us ing a ser ies resistor . 
Filter cu r ren t w a s i m p l e m e n t e d by sw i t ch ing an induc to r us ing fou r i nsu la ted ga te 
bipolar t rans is to rs ( IGBT) a r r a n g e d in H b r idge con f igura t ion . T h e s imu la t i on c i rcu i t 
was i m p l e m e n t e d us ing M A T L A B S imu l i nk so f twa re tool . I n d u c t a n c e of the s w i t c h i n g 
inductor , vo l t a g e of the s tep up t r ans fo rmer and the sw i t ch ing f r e q u e n c y of t he 
sys tem w e r e ob ta i ned by s imu la t ion . S u b s e q u e n t l y the a b o v e c i rcu i t p a r a m e t e r s 
we re va l i da ted for va r iab le loads us ing total h a r m o n i c d is to r t ion as the d i s c e r n i n g 
cr i ter ion. 
V 
It w a s poss ib le to r educe T H D of the cu r ren t w a v e of c o m p u t e r p o w e r s u p p l y f r o m 
1 0 7 % to 12%. P o w e r fac to r w a s i m p r o v e d f r om 0.66 to uni ty. By i n c r e a s i n g t he 
power fac to r to uni ty, the cur ren t f l ow can be r e d u c e d by a p p r o x i m a t e l y 3 4 % . T h e 
obse rva t i ons m a d e here in are app l i cab le for h a r m o n i c e l im ina t ion in n o n l i n e a r l oads 
in gene ra l w i th n e c e s s a r y mod i f i ca t ions . 
vi 
A C K N O W L E D G E M E N T 
I a m grea t l y i ndeb t to m y supe rv i so r Dr. J. P. K a r u n a d a s a , H e a d of D e p a r t m e n t , 
D e p a r t m e n t of E lect r ica l Eng ineer ing , Un ivers i ty of M o r a t u w a , M o r a t u w a fo r his eve r 
p resen t w o r d s of w i s d o m . D e p a r t m e n t of E lect r ica l Eng ineer ing , a n d the D e p a r t m e n t 
of M e c h a n i c a l e n g i n e e r i n g of the Un ivers i ty of M ra tuwa , M o r a t u w a fo r p rov id ing m e 
this i m m e n s e l y v a l u a b l e oppor tun i t y to car ry ou t a p o s t g r a d u a t e p ro jec t w i t h all t he 
faci l i t ies a n d g u i d a n c e in a p leasan t env i r onmen t . 
I w i s h to reco rd m y gra t i tude to the labora to ry s ta f f t he D e p a r t m e n t of E lec t r ica l 
Eng ineer ing , Un ivers i ty of M o r a t u w a , M o r a t u w a for the t echn i ca l s u p p o r t in ca r r y ing 
out m y r e s e a r c h wo rk . Fur ther to the s ta f f of the R e g i o n a l s u p p o r t C e n t e r 
( S o u t h e r n / U v a ) of Nat iona l W a t e r supp ly and D r a i n a g e Board , M a t a r a fo r the i r 
suppor t . 
Final ly, I s h o u l d t h a n k m a n y ind iv idua ls , f r iends and co l l eagues w h o h a v e no t b e e n 
m e n t i o n e d he re pe rsona l l y in m a k i n g th is educa t iona l p rocess a s u c c e s s . M a y be I 
cou ld not h a v e m a d e it w i thou t you r suppor ts . 
v i i 
LIST OF F I G U R E S 
Figure 1.1: 
F igure 1.2: 
F igure 2 .1 : 
F igure 2.2: 
F igure 2.3: 
F igure 2.4: 
F igure 2.5: 
F igure 2.6: 
F igure 3.1: 
F igure 3.2: 
F igure 3.3: 
F igure 3.4 
F igure 3.5: 
F igure 3.5: 
F igure 3.6: 
F igure 3.7: 
F igure 3.8: 
F igure 3.9: 
F igure 3 .10: 
F igure 3.11: 
F igure 3 .12 
Typ i ca l a r r a n g e m e n t of a pass i ve f i l ter 
C o m p e n s a t i o n charac te r is t i cs of a para l le l ac t ive p o w e r f i l ter 
H a r d w a r e c i rcui t to ob ta in vo l t age and cur ren t s igna ls 
S c h e m a t i c rep resen ta t i on of the da ta acqu is i t ion us ing M A T L A B 
M A T L A B c o m m a n d s used for da ta acqu is i t i on 
M A T L A B c o m m a n d s used for w a v e f o r m ana lys i s 
C i rcu i t d i a g r a m of the para l le l ac t ive fi l ter 
S imu la t i on c i rcui t 
V o u l t a g e s igna l of the s a m p l e load 
Cu r ren t s igna l of the s a m p l e load 
C u r r e n t w a v e of the capac i to r cha rge r 
C u r r e n t w a v e f o r m requ i red f r o m the para l le l ac t i ve f i l ter 
W a v e f o r m ob ta ined for i n d u c t a n c e of 1 0 m H , s tep up v o l t a g e of 
3 5 0 V a n d sw i t ch ing f r e q u e n c y of 1 0 0 k H z 
W a v e f o r m ob ta ined for i nduc tance of 1 0 m H , s tep up vo l t age of 3 5 0 V 
and sw i t ch ing f r equency of 1 0 0 k H z 
W a v e f o r m ob ta ined for i nduc tance of 5 0 m H , s tep up v o l t a g e of 3 0 0 V 
and sw i t ch ing f r equency of 5 0 k H z 
W a v e f o r m ob ta ined for i nduc tance of 5 0 m H , s tep up v o l t a g e of 3 5 0 V 
and sw i t ch ing f r equency of 1 0 k H z Resu l tan t tota l cu r ren t o b s e r v e d under i n d u c t a n c e of 5 0 m H , s tep up 
vo l t age of 3 5 0 V and sw i t ch ing f r equency of 1 0 0 k H z 
Resu l t an t to ta l cu r ren t o b s e r v e d unde r 5 0 % load u n d e r s e l e c t e d c i rcu i t 
p a r a m e t e r s 
Resu l t an t to ta l cu r ren t o b s e r v e d unde r 1 6 0 % load u n d e r s e l e c t e d 
c i rcui t p a r a m e t e r s 
Resu l t an t to ta l cu r ren t o b s e r v e d unde r 9 0 % of supp ly vo l t age u n d e r 
se lec ted c i rcui t p a r a m e t e r s 
Resu l tan t total cu r ren t o b s e r v e d unde r 1 1 0 % of supp l y v o l t a g e u n d e r 
se lec ted c i rcui t p a r a m e t e r s 
V I I I 
L I S T O F T A B L E S 
T a b l e 2 . 1 : T h e i n p u t r a n g e a n d t h e r e l e v a n t l e ve l o f a c c u r a c y 
T a b l e 3 .1 : E l e c t r i c a l p a r a m e t e r s o f t h e s a m p l e l o a d 
T a b l e 3 .2 : T o t a l h a r m o n i c d i s t o r t i o n v a r i a t i o n w i t h t h e i n d u c t a n c e 
T a b l e 3 .3 : T o t a l h a r m o n i c d i s t o r t i o n v a r i a t i o n w i t h t h e s t e p u p t r a n s f o r m e r v o l t a g e 
T a b l e 3 .4 : T o t a l h a r m o n i c d i s t o r t i o n v a r i a t i o n w i t h s w i t c h i n g f r e q u e n c y 
T a b l e 3 .5 : T o t a l H a r m o n i c D i s t o r t i o n v a r i a t i o n f o r d i f f e r e n t l o a d v a r i a t i o n 
ix 
CHAPTER I 
INTRODUCTION 
The increasing demand of power electronic devices, ranging from CFLs and switch 
mode power supplies to automated industrial assembly lines, has enlarged power 
quality problems in the electrical power utility. Those problems include a variety of 
electrical disturbances, which can originate in several ways and differently affects 
various kinds of sensitive loads. The proliferation of nonlinear loads increases 
contamination level in voltage waveform. Specifically it causes harmonic injection 
that subsequently result excessive neutral currents, and reactive power burden in the 
power system. 
1.1. Power system harmonics & effects of harmonics 
Harmonic is defined as "a sinusoidal component of a periodic wave or quantity 
having a frequency that is an integral multiple of the fundamental frequency" [1], 
Power system harmonics are integer multiples of the fundamental power system 
frequency. High levels of power system harmonics can create waveform distortion 
and thereby increase power quality problems. 
Major sources of waveform distortion include (1) Electronic switching power 
converters such as computers, uninterruptible power supplies (UPS), adjustable-
speed motor drives, solid-state rectifiers, electronic process control equipment and 
electronic lighting ballasts; (2) Arcing devices such as discharge lighting 
l 
(Fluorescent, Sodium & Mercury vapor), arc furnaces, welding equipment, electrical 
traction system; (3) Ferromagnetic devices such as transformers operating near 
saturation level, magnetic ballasts (Saturation Iron Core), induction heating 
equipment, chokes, motors; (4) Appliances such as television sets, air conditioners, 
washing machines, microwave ovens, vacuum cleaners, fax machines, 
photocopiers, printers. 
Harmonics result in poor power factor, lower efficiency and interfering with adjacent 
communication systems. Harmonic currents injected into the utility supply lines 
produces extra losses and cause voltage distortion. Besides interfering with data 
communications, telephone circuits, digital controls, they can also overheat 
transformers and supply apparatus. Harmonics flow into power factor capacitors, 
power cables and be magnified by resonance which even though not often can lead 
to disastrous results. Effects of harmonics on the power systems include frequent 
tripping of circuit breakers and protection relays, frequent fuse blowing, capacitor 
failures, degrading meters accuracy, overheating of motors and transformers, 
overloading of transformer neutral, telephone interference, malfunction of motor 
variable-speed drives, insulation failures, severe lamp flicker, voltage distortion and 
lagging in power factor. In summery, the major problem of harmonic currents include 
equipment heating, equipment malfunction, equipment failure, communications 
interference, fuse and breaker mis-operation, process problems and conductor 
heating. 
2 
1.2. Measures of Harmonics 
There are three methods of estimating harmonic load content: the Crest-factor (CF), 
Harmonic Factor or Total Harmonic Distortion (THD) and "K-Factor" [2], 
Crest Factor is a measure of the peak value of the waveform compared to the true 
RMS value. The mathematical definition, of the crest factor is the ratio of the peak 
value of the waveform divided by the rms value of the waveform: 
Total harmonic distortion (THD) of a signal is the ratio of the sum of the powers of 
all harmonic frequencies above the fundamental frequency. 
t-ttt-v Y, harmonic powers 
I H D = 
fundamental frequency power 
The K-factor is a number derived from a numerical calculation 
summation of harmonic currents generated by the non-linear load. 
K-factor, the more significant the harmonic current content [3], 
1.3. Importance of harmonic mitigation 
Modern technical applications are very vulnerable to power supply quality. 
International standards (IEC 1000-3-2 and IEEE 519-1992) allow a certain level of 
harmonic current depending on the source short circuit ratio. Short circuit ratio is the 
ratio of short circuit current to the load current. At a ratio of 20 or less, the norms 
allows only 5% total current harmonic distortion [4], Therefore reducing or even 
cancelling harmonics in the supply is essential in both economic and technological 
terms. 
3 
based on the 
The higher the 
The main advantage of elimination or reduction of harmonics in power systems is 
improvement in productivity and quality [4], That also minimizes equipment 
malfunction or failure. Further by minimizing current, it enables to reduce cable sizes 
especially in neutral conductor. Most importantly reduction of harmonics reduces 
energy wastage in power systems and reduces power consumption resulting less 
electrical power bills. 
1.4. Methods of harmonic mitigation 
There are two approaches to mitigate power quality problems. The first approach is 
known as load conditioning, which ensures that the equipment is less sensitive to 
power disturbances, allowing the equipment to operation even under significant 
voltage distortion. The other solution is to install line conditioning systems that 
suppress or counteracts with the power system disturbances. Both active filters and 
passive filters can be effectively used for harmonic mitigation under this category. 
Passive filters: Conventional solution for harmonic distortion has been the use of 
passive filters to minimize offending harmonics. Voltage harmonic distortion could 
often be controlled with a simple high pass filter, but reduction of current harmonic 
levels usually requires installation of multiple tuned traps, each of which reduces only 
one of the many harmonic frequencies produced by the non-linear load (Figure 1.1). 
4 
The other disadvantage is that the current of the filter is not controllable and it can 
also produce reactive power. Further, passive power filters have the problem of 
resonance. 
Active filters: A different solution has been recently introduced as a result of the 
advances in power semiconductors. Insulation Gate Bi-polar Transistors (IGBTs) 
made possible to have a power switch with nearly ideal performance and reliability. 
Consequently two new types of equipments were developed: the Pulse Width 
Modulated (PWM) -Unity Power Factor rectifiers and the Active Power Filters (APF). 
By construction, an active filter is similar to a PWM unity power factor rectifier, but 
different to it by the fact that the inverter for APF applications present a forth IGBT 
leg or a centre tap on the main filtering capacitor battery which is connected to the 
neutral [4], 
5 
1.5. Active Power Filters 
Active power filters were initially proposed by Sasaki and Machida in 1971, as a 
means of removing current harmonics [5], 
An active power filter uses a switching inverter to produce harmonic compensating 
currents. The basic principle of Active Power Filter (APF) is to utilize power 
electronics technologies to produce specified current components that cancel the 
harmonic current components caused by harmonic producing loads. 
Active power filters can perform one or more of the functions required to compensate 
power systems and to improve power quality. Their performance depends on the 
power rating and the speed of response. 
Active power line conditioners are typically equipped with IGBT or GTO (Gate Turn 
Off) thyristors, voltage source PWM converters and connected to low and medium 
voltage distribution systems arranged in shunt, series or both at the same time. In 
comparison to conventional passive LC filters, active power filters offer very fast 
control response and more flexibility in defining the required control tasks for a 
particular application. Smaller sizes of the component and been more economical 
solution are the major advantages of using active filters for harmonic mitigation. 
The two major types of active power filters are series active power filters and shunt 
or parallel active power filters. Shunt active power filters operate as a controllable 
current source and series active power filters operates as a controllable voltage 
source. The selection of the type of active power filter to improve power quality 
depends on the source of the problem. 
6 
Parallel active power filter compensate current harmonics by injecting equal-but-
opposite harmonic compensating current. In this case the shunt active power filter 
operates as a current source that injects harmonic components generated by the 
load but with a phase shifted at 180°. This principle is applicable to any type of load 
that is considered as a harmonic source. Moreover, with an appropriate control 
scheme, the active power filter can also compensate the load power factor. 
The concept of parallel or shunt active filter can be illustrated by the following figure. 
Line Current 
r 
JL 
v y 
Power Distr ibution 
Equivalent Circuit 
Load Current 
i J. I „ Non Linear Load 
/* Filter Current 
u 
Paral lel Active Power Filter 
Figure 1.2: Compensation characteristics of a parallel active power filter 
Numerous applications of active power filters are described in literature [6-10], 
7 
1.6. Objectives 
Therefore the major objective of this project was to eliminate effects of harmonics 
and to improve power factor by using a parallel active power filter, i.e. to connect an 
additional unit (the parallel filter) to the utility in order to inject the appropriate 
compensation current to eliminate harmonics and to improve power factor. The 
specific objectives include (1) study and understand the harmonic effects, (2) decide 
the APF suitable for current shaping and (3) simulating for results. 
1.7. Thesis Organization 
The thesis is composed of four main chapters. The first chapter brief the research 
background and the motivation of the work been produced. The same also 
introduces objectives of the research. 
The research methodology is described in the second chapter. The subject material 
is detailed under four main topics viz. (1) data acquisition, (2) data analysis, (3) 
simulating for results and (4) validation of results 
Results of the projects are given in the chapter three. The results are presented in 
four main sections, namely (1) acquired signals, (2) required filter current (3) 
simulation results and (4) validation of the results. 
The project is discussed in the chapter four. This includes a general discussion of 
the results, specific problems arising from the methodology, protective measures 
implemented, as well as a discussion of the results and suggestions for future. 
8 
CHAPTER 2 
METHODOLOGY 
The project basically focus on five main topics viz (1) acquisition of voltage and 
current signals of a power appliance, (2) analysis of the acquired signals, (3) 
determination of the appropriate filter (4) simulation to implement active power filter, 
(5) validation of the acquired circuit parameters. 
2.1. Acquisition of voltage and current signals of the power appliance 
Accuracy in acquiring voltage and current signals is the major factor that determines 
the research outcome as well as the value of the results. Type of load, properties of 
the data acquisition card and auxiliary hardware circuit has to be compatible for 
proper and efficient functioning of the system. 
2.1.1. Sample load 
Extremely high frequency components are not usually present in computer power 
suppliers since they are supported with passive low pass filters (Annex_1). 
According to the preliminary studies made on computer power supply the 
approximate current wave form maintain a high peak to rms ratio (or crest factor). 
Use of Rectifiers (to convert AC to DC voltage) is a known cause to produce such 
observations. 
9 
Power rating of computer power supply utilized as the sample load was 400 W. 
However since the power demand of a computer is determined by several factors, 
including weather different components such as cooling fans, optical drivers, hard 
disk drives and processor are activated or not, on average the power demand of the 
sample load was around 150 W (without monitor). Once the data is acquired at a 
particular instant, it was assumed that the power is constant for the period of 
operation. However for better performance data acquisition at regular intervals is 
needed. 
2.1.2. Data acquisition card 
The National Instruments USB-6008 multifunction data acquisition (DAQ) module 
(Annex_2) was used to acquire data from the sample load viz. the computer power 
supply. MATLAB software facilitates data communication with the Nl USB-6008. 
USB-6008 is supported in Data Acquisition Toolbox 2.8 (MATLAB R2006a) or newer. 
There were eight analog input channels of which the input resolution was 12 bit and 
the maximum sampling rate was up to 10 kS/s. The input impedance was 144 KQ. 
The input range and the relevant level of accuracy are given in the table 1. 
Table 2.1: The input range and the relevant level of accuracy. 
Input range (V) ±20 ±10 ±5 ±2 ±1 
Accuracy (mV) 14.7 7.75 4.28 2.21 1.53 
Details of the National Instruments USB 6008 DAQ card can be obtained online via 
http://sine.ni.eom/nips/cds/view/p/lanq/en/nid/14604. 
10 
2.1.3. Hardware circuit 
The circuit diagram of the hardware circuit that was utilized to obtain the voltage and 
current signal of the sample load is given below. 
230V @ 
2 . OA 
0 . 1 5 A 
5 . 6K 
l} 
2 7 0 K 
2 7 0 K 
5 . 6 K 
0 . 6 0 R / 7 W 
H M n 
L-AA/V-1 
0 . 6 0 R / 7 W 
Voltage Current . 
Signal Signal 
LOAD 
Figure 2.1: Hardware circuit to obtain voltage and current signals 
Procedure to acquire the current signal: Current signal was acquired using a 
series resistor. This configuration causes voltage reduction across the sample load. 
Therefore for higher accuracy the voltage across the resistor was maintained at 
minimum, approximately 0.2 - 0.3 V rms. 
Procedure to acquire the voltage signal: In order to obtain the voltage signal a 
potential divider technique was incorporated to the circuit such that the 
arrangement satisfy the following conditions (1) to limit power consumption at a 
11 
minimum; (2) to limit peak value of the voltage wave around 1 -2 V. Application of the 
above conditions enables the voltage signal to be acquired within the same range 
where the current signal falls. The same further minimizes power losses. 
Protective measures: Thermal fuses, 2.0 A fuse in the load path and a 0.15 A fuse 
in the potential divider path of the circuit, were used as over current protectors. 
An additional resister in parallel configuration was used as a protective measurement 
against open circuit faults when measuring voltage across a resister. 
Similarly two resisters in series were used as the non measuring resister of potential 
divider to prevent failures due to short circuiting of the particular resister. 
Maximum measuring limits were kept substantially lower than the maximum limits of 
the input range in order to minimize damages to the data acquisition card due to high 
voltage. 
2.1.4. Acquisition of voltage and current signals using MATLAB 
The device and two of the channels within were configured to obtain current and 
voltage signals by specific MATLAB commands. 
Sampling rate was set to a value of 5000 samples per second, which is half of the 
maximum sampling rate available. The sample rate was determined such that the 
data bulk does not overload the data acquisition card. By selecting the sample rate 
as 5000 samples per second, the system can identify up to 50th harmonic which is 
adequate for general applications. 
12 
The device was triggered by using the voltage signal. Data acquisition was triggered 
at zero and at the rising crossings of the voltage signal. Data was acquired 
throughout a single complete power cycle at a time. 
2.1.5. MATLAB Data acquisition tool box 
Data Acquisition Toolbox provides a complete set of tools for analog input, analog 
output, and digital I/O from a variety of PC-compatible data acquisition hardware. 
The toolbox permits to configure external hardware devices, read data into MATLAB 
and Simulink environments for immediate analysis, and send out data [11], 
Figure 2.2: Schematic representation of the data acquisition using MATLAB 
MATLAB Data acquisition tool box version 2.8 or above is required to communicate 
with Nl USB 6008. DAQ tool box 2.8 or above is available at MATLAB R14SP3+ or 
later versions. 
MATLAB commands used for data acquisition are illustrated on the following figure 
(Figure 2.3). Commands are described by bold letters. 
13 
a i = a n a l o g i n p u t ( ' n i d a q ' , ' D e v i 1 ) ; % Configure an National 
Instruments DAQ input Object 
V o l t a g e = a d d c h a n n e l ( a i , 1 ) ; % Add channel of input 
object 
C u r r e n t = a d d c h a n n e l ( a i , 0 ) ; 
ai.SampleRate = 5000; % Set sampling rate 
ai.SamplesPerTrigger =0.02*5000; % Set number of samples 
per one trigger 
set(ai, 'TriggerType', 1 software'); % software trigger is 
used 
set(ai, 'TriggerRepeat', 0); % Execute the trigger 
once when the trigger condition is 
met. 
set (ai, 'TriggerCondition','rising') ; % The trigger occurs 
when the signal has a positive 
slope when passing through the 
specified value 
set(ai, 'TriggerConditionValue' , 0) ; % set specified value 
for TriggerCondition 
set(ai, 'TimeOut', 2); % Specify the waiting 
time to complete a read or write 
operation 
set(ai,'TriggerChannel',ai.Channel(1)) % Specify channel 
serving as trigger source 
start(ai); % To start ai 
[d,t] = getdata(ai); % To acquire data for 
under conditions 
stop(ai); % To stop ai 
Figure 2.3: MATLAB commands used for data acquisition 
14 
2.1. Analysis of acquired signals using MATLAB 
The acquired voltage and current signals was analyzed subsequently in order to 
obtain parameters of the wave forms including instantaneous values, root mean 
square (rms) values, active power, effective power factor, total harmonic distortion 
(THD), harmonic spectrum, percentage harmonics, and distortion factor. Low 
magnitude higher order harmonics were shed for convenience in data analysis. 
Instantaneous values of voltage and current 
Instantaneous voltage was determined by multiplying the acquired voltage signal by 
a factor of 194. Factor 194 arises from the ratio of resistors of the potential divider. 
The instantaneous current values were obtained by dividing the voltage across the 
series resistor, by which the current signal is determined, by the resistance of the 
same. The resistance value is 0.3. 
RMS values of voltage and current 
Root mean square (rms) value of voltage (Vrms) and current (Irms) signals were 
calculated by the square roots of the average sum of square values of the sample 
points. 
rms value = V(Sum of square of instantaneous values/No of samples) 
Active power 
Active power (P) was obtained by integrating the product of instantaneous voltage 
and current values for a complete power cycle using following relationship. 
t+T 
Active Power (P) = J (V(t). I(t))dt 
t 
Where t indicated time, T was the period of the wave, and V(t) and l(t) were 
instantaneous voltage and current values 
15 
Effective power factor 
Effective power factor was calculated by dividing the active power calculated as 
above by both Vrms and Irms. 
, . , „ . . . Active power(P) 
Effective power factor ((b) = — 
Vrms x Irms 
Total harmonic distortion (THD) 
Magnitudes of harmonics were acquired by fast Fourier transform of instantaneous 
signals. The harmonic magnitudes were then used to calculate THD as follows. 
Total harmonic distortion(THD) = 
A22 + A2+A2 + - + A5, 
Where AN is the amplitude of N harmonic component. 
Harmonic spectrum and percentage harmonics 
Harmonic magnitudes obtained by fast Fourier transformation were used to obtain 
harmonic spectrum and the percentage harmonic magnitudes. 
Distortion factor 
Distortion factor was calculated using the following relationship. 
1 
Distortion factor = 
V l + THD2 
All MATLAB commands used for waveform analysis are illustrated in the following figure 
(Figure 2.6). 
16 
%Instantaneous voltage and current 
V = 19 4 * d(:,1); I = d(:,2)/30; 
%Calculating tirue power. 
P = trapz(t,V.*I)/T; 
%Calculating RMS values. 
Vrms = sqrt(sum(V.A2)/(N-l)); Irms = sqrt(sum(I.A2)/(N-l)); 
%Calculating effective power factor. 
EPF = P/(Vrms*Irms); 
%Calculating FFT of current signal. 
X = f f t ( I ' ) ; Ampl i tudes = [X(l)/N, 2*abs(X(2:N))/N]; 
%Error correction 
K= floor(N/2); Amplitudes = Amplitudes(1:K); 
%Calculating Total Harmonic Distortion. 
THD = sqrt(Amplitudes(3:K)*(Amplitudes(3:K))')/Amplitudes(2) 
%Calculating Distortion Factor 
DF = (1+THDA2)A(-0.5); 
%Calculating Precentage Harmonic Levels. 
H_Percent = 100*Amp l i t udes /Amp l i t udes (2 ) ; 
%Shedding Negligable Percentages. 
k = K; 
while H_Percent(k)<2 
if H_Percent(k)<2 
H_Percent = H_Percent(1:k-1); 
end; 
k=k-l; 
end; 
Figure 2.4: MATLAB commands used for waveform analysis 
17 
2-3. Determination of the appropriate filter current 
Load current was obtained directly using the data acquisition card as described in 
section 2.1.1.The ideal sinusoidal total current was determined using the following 
equation. 
P = VI cos <fi » Equation--1 
Where; P = active power, V = rms voltage, I = rms current and cos 4> = power factor. 
P and V were obtained as described in section 2.1.4. In order to keep the ideal 
current in the same phase with the voltage, power factor was considered as unity. 
Then, 
T _ P 1 — » Equation--2 
The total current obtained from the utility for the whole system, consisting of the 
sample load (computer power supply) and the filter, is equal to the sum of load 
current and the filter current as the filter is configured in parallel with the load. 
Therefore filter current is equal to the subtraction of actual load current from the ideal 
sinusoidal total current. 
18 
2.4. 
UNIVERSITY 0irMORATUWA,SRllANKA 
NiGHATUWA 
Simulation to implement active power filter 
Once the appropriate current was determined as described above, a circuit was 
constructed in simulation, by which the correct current waveform could be generated. 
2.4.1 Circuit arrangement of the simulation 
Filter current was implemented by switching an inductor using four insulated gate 
bipolar transistors (IGBT) arranged in H bridge configuration. A charged capacitor 
was employed to provide the appropriate voltage to the inductor and thereby the 
required current was obtained. To charge the capacitor a step up transformer with a 
full bridge rectifier convertor was used. The simulation circuit is given in Figure 2.5. 
S1-S4 are IGBT switches. 
Figure 2.5: Circuit diagram of the parallel active filter 
The simulation circuit was implemented using MATLAB Simulink software tool. 
19 
9 2 9 7 a 
„ 
TH
D
 
Fi
gu
re
 
2.
6:
 S
im
ul
at
io
n 
ci
rc
ui
t 
As the sampling rate available in the Nl USB 6008 data acquisition card was not 
sufficient for precise implementation of the simulation, the number of sample points 
was increased by generating additional sample points using the MATLAB software. 
For-loop technique was made use of assuming linearity between two adjacent 
sample points for the purpose (Annex_3). 
Current signal and voltage signal acquired from Data Acquisition Card was saved as 
"V" and "I" in MATLAB workspace. As described in section 2.3, filter current was 
determined and saved in MATLAB workspace as "Fil_current". Subsequently the 
acquired load current was saved as "Load_current" in MATLAB workspace, for 
convenience in identification. 
The current signal need to be obtained by switching circuit (ir) was determined by 
subtracting capacitor charging current from the filter current. Deference between the 
current described above (ir) and the actual current through the inductor was fed to 
the switching signal generator via a PI controller. Subsequently the signals 
generated by the switching signal generator were fed to IGBTs. 
The total current of the system was obtained by adding acquired load current, 
charging circuit current and current through the inductor. 
2.4.2 Simulation to obtain circuit parameters of the active filter 
THD was used as the measure of harmonic content. Hence, the optimal circuit 
parameters of the active filter were obtained by minimizing THD. 
The capacitance of the capacitor need to be sufficient in order to deliver the 
appropriate electrical charge to the system. Yet, installing a large capacitor is not 
21 
feasible because of physical size of it and the cost. A capacitor with the capacity of 
200 pF was used in the simulation. 
Inductance of the switching inductor, voltage of the step up transformer and the 
switching frequency of the system were worked out by simulation. However, THD of 
the simulation circuit varies with all three factors mentioned above i.e. (1) inductance 
of the Inductor, (2) output voltage of the step up transformer, (3) switching frequency 
of IGBTs. The simulation program was highly complex since all three factors were to 
be optimized simultaneously. Therefore, the first parameter was optimized while the 
other two were kept under minimum sensitivity. Once one of the parameters was 
optimized as above the next was optimized for the optimum value of the first 
parameter. 
Determination of the inductance of the switching inductor: Application of the 
appropriate inductance is essential as in a given voltage; large inductance has less 
potential to remove higher order harmonics. On the other hand the wave signals 
tend to deviate more from the reference wave when the inductance was small. 
A range of inductance values starting from 10 mH up to 100 mH in intervals of 10 
mH were studied under four different' combinations of two different switching 
frequencies viz. (1) 50 kHz, (2) 100 kHz, and transformer voltages viz. (1) 300V and 
(2) 400 V. The resulting waveforms and Total harmonic distortions (THDs) were 
observed. 
Physical parameters (size) of the inductor were important as when Inductors of high 
inductance were bulky and hence not suitable for low power applications as the one 
22 
being studied. Therefore Inductors of inductance higher than 100 mH was not 
considered in the simulation. 
• 
Voltage of the step up transformer: The appropriate inductance was identified as 
50mH. Higher voltages increase handling difficulties, insulation requirements and 
require equipments with high rated voltage. In contrast low voltages were incapable 
to supply the required filter current. Therefore four different transformer voltages; 
250 V, 300 V, 350 V, 400 V were studied at three switching frequencies 20 kHz, 50 
kHz, 100 kHz. The resulting waveforms and THD were observed. 
Switching frequency of the system: As described above the proper inductance 
and Transformer voltage were identified as 50 mH and 350 V respectively. The 
waveforms and THDs were studied at a range of switching frequencies starting from 
5 kHz up to 200 kHz at 50 mH inductance and 350V step up transformer voltage. 
Even though high switching frequencies were considered in the simulation such is 
extremely difficult to achieve in practical applications. 
2.5. Validation of the acquired circuit parameters 
At the best possible inductance of 50mH, transformer voltage of 350V, and switching 
frequency of 100 kHz which were chosen by the simulation, the behavior of the 
active power filter was studied under different (1) load variations (50% to 200%) and 
(2) system voltage variations (90% to 110%). 
Total harmonic distortion and the current wave form were observed. 
23 
CHAPTER 3 
RESULTS 
3.1. Electrical parameters of the sample load 
3.1.1. Voltage signal of the sample load 
With the exception of a slight distortion observed towards the peaks, the voltage 
wave of the sample load was nearly perfect sinusoidal wave. Higher order 
harmonics were negligibly low (Figure 3.1). 
4 0 0 
A 
300 
200 
100 
> — 0 
oi 03 
± : ' 6 
5 -100 
-200 
-300 
-400 
Time (s) 
r 
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 
Harmonic Order 
Figure 3.1: Voltage signal of the sample load. (A) voltage waveform, (B) voltage 
harmonic distribution 
24 
B 
o 6 0 
C o 
E 
3.1.2. Current signal of the sample load 
Unlike the voltage waveform of the load current was highly distorted, and in addition 
to the fundermental higher order harmonics were quite prominent. Among them the 
3rd, 5th,7th and 9th order harmonics were foremost. 
0 . 0 6 
Time (s) 
B 120 
100 --_• 
80 
o 60 'c 
o E 
<u 
X 4 0 
20 
H n n a XL XL XL J 
1 2 3 4 5 6 7 8 9 10 11 12 13 
C [ ] .n .XLnJ] n n 
14 15 16 17 18 19 20 21 22 23 24 25 
Harmonic Order 
Figure 3.2: Current signal of the sample load. (A) current waveform, 
(B)current harmonic distribution 
25 
3.1.3. Electrical parameters of the sample load 
Electrical parameters of the acquired voltage and the current signal 
table 3.1. 
Table 3.1: Electrical parameters of the sample load 
Active power of the load = 144 W 
Power factor = 0.66 
Voltage signal 
1. Rms voltage 
2. Total Harmonic Distortion 
3. Displacement Factor 
= 229.7 V 
= 2.68% 
= 0.9996 
Current signal 
1. Rms current 
2. Total Harmonic Distortion 
3. Displacement Factor 
= 0.945 A 
= 107% 
= 0.683 
3-2 Required current from the filter 
3.2.1. Current of the capacitor charger • 
The current of the capacitor charger was obtained under the following condition viz. (1) 
Inductance of 50 mH, (2) Step up transformer secondary voltage of 350 V, (3) 
Switching frequency of 100 kHz. 
The current was measured once the capacitor was stabilized. The current of the 
capacitor charger was significantly low compared to the filter current. 
26 
0.08 
-0.08 J ! 
Time (s) 
Figure 3.3: Current wave of the capacitor charger 
3.2.2. Filter current with capacitor charger 
The required filter current was determined as described in the section 2.2. 
Incorporation of the capacitor charging circuit into the filter circuit did not significantly 
affect the filter current. 
Following figure (Figure 3.4) describes the required current from the active power 
filter in order to eliminate harmonic effects. The current variation resulted after 
incorporating capacitor charging circuit with' the active filter was insignificant. 
27 
T i m e (s) 
B 
4 
- 3 
T i m e (s) 
Figure 3.4: Current waveform required from the parallel active filter, (A) The 
capacitor charger incorporated, (B) In the absence of the capacitor charger 
3.3. Simulation results 
3.3.1. Selection of the inductor 
Percentage THD for different inductance values were obtained in the simulation of 
filter current. The table 3.2 provides THD of the total current. 
28 
Table 3.2: Total harmonic distortion variation with the inductance 
Inductance (mH) 
Step up transformer 
voltage (V) 
Switching frequency 
(kHz) 
Total Harmonic 
Distortion (%) 
400 100 76 
10 50 335 
300 100 59 
50 251 
400 100 29 
20 50 81 
300 100 24 
50 60 
400 100 18 
30 50 45 
300 100 14 
50 34 
400 100 14 
40 50 31 
300 
100 16 
50 24 
400 100 11 
50 50 24 
300 100 22 
50 32 
400 100 9 
60 50 21 
300 
100 48 
50 62 
400 
100 8 
70 50 19 
300 
100 87 
50 89 
400 
100 8 
80 50 15 
300 
100 125 
50 143 
400 100 10 
90 50 17 
300 
100 175 
50 217 
400 
100 16 
100 50 24 
300 
100 232 
50 260 
29 
Total current is the summation of the load current and the filter current. Inductance 
value of 50 mH which resulted acceptable THD percentages was selected for further 
studies. Along with the THD percentages waveforms and other parameters were 
taken into consideration when the inductor was selected. 
Rapid waveform variations observed under lower inductance values are shown in the 
following figure (Figure 3.5). 
- 1 . 5 
Time (s) 
Figure 3.5: Waveform obtained for inductance of 10mH, Step up voltage of 400V and 
switching frequency of 100kHz, (A) active filter current, (B) resultant total current 
30 
3.3.2 Selection of the step up transformer 
Inductance value was selected to be 50 mH, The simulation results of the step up 
transformer secondary voltage at three different switching frequencies; 20 kHz, 50 
kHz, and 100 kHz were observed, at the selected inductance of 50 mH. Results are 
given in the table 3.3. 
Table 3.3: Total harmonic distortion variation with the step up transformer voltage 
Step up transformer 
voltage (V) Switching frequency 
Total Harmonic 
Distort ion (%) 
250 
100 335 
50 372 
20 375 
300 
100 23 
50 32 
20 50 
350 
100 11 
50 21 
20 60 
400 
100 11 
50 24 
20 67 
The most appropriate step up transformer secondary voltage was selected as 350 V 
that produced minimum or acceptable THD percentage. Secondary voltage of 350 V 
was used in further studies to determine the suitable switching frequency. 
31 
The filter is unable to produce required current shape for lower step up voltages. 
Following figure (Figure 3.6) indicates the incapability of the filter in providing 
required current shape, for lower voltages. 
Figure 3.6: Waveform obtained for inductance of 50mH, Step up voltage of 
300V and switching frequency of 50kHz, (A) active filter current, (B) resultant 
total current 
3.3.3 Selection of the switching frequency 
The simulation results of the switching frequency of the APF at the selected 
inductance of 50 mH and the transformer secondary voltage of 350 V are given in 
the table 3.4. 
Table 3.4: Total harmonic distortion variation with switching frequency 
Switching frequency (kHz) 
Total Harmonic 
Distortion (%) 
5 305 
10 225 
20 57 
25 42 
40 23 
50 21 
100 11 
200 5 
With increasing switching frequency THD decreases. However higher switching 
frequencies cause practical problems such as high switching losses, difficulties to 
implement. Therefore 100 kHz switching frequency was selected to proceed with the 
simulation. Following figure (Figure 3.7) represents current waveform variations 
observed under lower switching frequency, under transformer secondary voltage of 
350 V and inductance of 50 mH. 
33 
T i m e (s) 
Figure 3.7: Waveform obtained for inductance of50mH, Step up voltage of 
350V and switching frequency of 10kHz, (A) active filter current, (B) resultant 
total current 
34 
Resultant total current for selected parameters 
The resultant total current observed applying the selected parameters viz. 
inductance of 50 mH, switching frequency of 100 kHz, and step up transformer 
secondary voltage of 350 V are depicted in the figure 3.8. The figure depicts higher 
order harmonics up to 40th order even though miniature higher order harmonics 
were present even above 40th order. 
Figure 3.8: Resultant total current observed under inductance of 50 mH, switching 
frequency of 100 kHz, and step up transformer secondary voltage of 350 V, 
(A) Waveform, (B) harmonic distribution 
35 
3.4. Validation 
The parameter values determined by the simulation were 50 mH for the inductance 
of the inductor, 350 V for the transformer voltage and 100 kHz for the switching 
frequency. Validity of those values was further confirmed using a range of values of 
sample loads and supply voltages 
3.4.1. Validation for load variations 
The parameters were validated at different values of load currents assuming to be of 
the same waveform shape. The results are given in table 3.5. 
Table 3.5: Total Harmonic Distortion variation for different load variation 
Load variation (%) 
Total Harmonic Distortion 
(96) 
50 21 
60 17 
70 14 
80 13 
90 11 
100 11 
120 12 
140 14 
160 21 
180 32 
200 55 
36 
Following figures represent waveform shape and harmonic distribution, for 50% of 
load and for 160% of load. For the range of 50%of load to 160% of load, the 
maximum observed THD was only 21%. 
H a r m o n i c o r d e r 
Figure 3.9: Resultant total current observed under 50% load under selected circuit 
parameters, (A) waveform, (B) harmonic distribution 
37 
T i m e ( s ) 
H a r m o n i c o rde r 
Figure 3.10: Resultant total current observed under 160% load under selected 
circuit parameters, (A) waveform, (B) harmonic distribution 
38 
3.4.2. Valioation for supply voltage variations 
The parameters were validated for 10 % supply voltage variations assuming the 
same active power consumption. Following figures represent resultant current wave 
shape and harmonic distribution for two different supply voltages. Higher order 
harmonics were not significant. 
B 
E 4 0 
•E,B.rgTIl.,0.,l3.TPr^T«. fl, n, fl ,n,f].r,fi,r. ,11 ,n J*Jb*f 
H a r m o n i c o rde r 
Figure 3.11: Resultant total current observed under 90% of supply voltage under 
selected circuit parameters, (A) waveform, (B) harmonic distribution 
39 
1 - r 
H a r m o n i c o r d e r 
Figure 3.12: Resultant total current observed under 110% of supply voltage under 
selected circuit parameters, (A) waveform, (B) harmonic distribution 
40 
CHAPTER 4 
DISCUSSION 
Active power filters have wide applications for controlling harmonic currents injected 
from nonlinear loads. In this project, parallel active power filters were effectively used 
to minimize Total Harmonic Distortion (THD), which used to measure harmonic 
contents, as well as to improve power factor using a computer power supply as a 
sample load. It was possible to reduce THD of the current wave of the computer 
power supply from 107 % to 11 %. Power factor was improved from 0.66 to unity. By 
increasing the power factor to unity, the current flow is possible to reduce by 
approximately 34% while minimizing power loss of the system. 
This project focus on the application of active power filters for current shaping of a 
specific load, contrary to its common applications at the point of common coupling. 
Applications of active power filters for individual loads have several advantages. 
Firstly, application of active filter to specific loads simplify the generation of reference 
current compared to the arrangement where the active filter is at the point of 
common coupling. Further when the active filter is at the point of common coupling it 
does not mitigate internal harmonic effects, whereas with application of active filters 
for specific loads the same problem does not arise. 
Application of active filters for specific loads is undeniably costly. Naturally wave 
form distortion is more in individual specific loads compared to that at the point of 
41 
common coupling. Consequently, the specific loads require more sensitive and 
highly efficient active filters. 
One of the main advantages of the prototype developed is its ability to display and 
analyze voltage and current waveforms of power appliances. Therefore, optimum 
filter circuit parameters for specific loads can be determined with much convenience. 
The same feature enables to use the prototype as a power quality monitoring unit in 
addition to its usage for current shaping. Certain modifications, such as changing the 
series resistor being used for acquiring the current signal, are required for the 
purpose. 
Despite computers consume low power, with the escalating usage; computers 
emerge as a potential source for utility system waveform distortions. Therefore as a 
typical nonlinear load computer power supply was studied in detail. Yet, the 
conclusions derived in the study are applicable to nonlinear loads in general. The 
conclusions were derived for the sample load, computer power supply which require 
specific load with narrow limits. Further the results were validated within narrow 
limits. Therefore it is necessary to study applicability of the results on alternative 
loads such as variable speed drives, CFL bulbs, switch mode power supplies and 
uninterrupted power supplies to implement a general purpose APF. 
IGBT switching causes additional power loss. The loss of power is proportional to the 
switching frequency [11], The fact need to be considered when such switching 
systems are implemented in low power appliances. 
42 
The system was run using a computer. Yet it is possible to design electronic circuits 
or programmed micro-controller systems to achieve the same. This would increase 
efficiency as well as the expediency of the system. 
THD was used as the discerning criteria in this study. THD can be effectively used to 
improve power quality and to economize power loss. However rather than the total 
harmonics studying specific harmonics are more important in some situations such 
as radio interference and other instances where resonance takes place. THD 
exclusively does not provide all the details of the waveform distortion. Therefore in 
order to make better judgment, the nature of the harmonics needs to be studied 
further. 
In order to develop the prototype worked out in this project to the level of 
implementation further studies are needed in the areas of IGBT switching specifically 
on the application of IGBT driving circuits. 
Excessive presence of harmonics in power systems is harmful not only to the 
consumer who is responsible for the non-linear load but also to other consumers 
and utility supplier. Therefore with the increasing harmonics in the power systems 
regulatory measures such as introducing tariff barriers against THD need to be 
implemented in future. 
Technical problems and concerns are discussed in detail in the following 
paragraphs. 
43 
a C q U i S i , i ° n ^ A — was used i n o r d e r t 0 o b t a i n ( h e 
current signal. T h e r e f o r e f o r a c c u r a t e ^ ( ^ ^ ^ ^ ^ 
resistor a, a minimum. W h e n t h e r m s v o | ( a g e ^ ^ ^ ^ ^ 
Peak voltage w o o , b e approximately „ , In order to obtain high , e v e l o f a c c u r a c y 
« — e n , of b o , h c u r r e n t a n d v o | ( a g e s i g n a | s j ( ^ _ ( h a ( ( h e ^ ^ 
Of the voltage signal is retain at 1 -2 V. 
because of the conduction drop of the dind» th a . 
op the diode that causes waveform distortions As an 
operating iimits substant ia,, lower (approximate, 2 0 % ) than the maximum iimits of 
the input range. 
voltage of the opera,iona, a m p l e r was considerably high. This wouid possibly he 
was high, „ was possibie to obtain an appropriate current signal b y 0.3 V voltage 
d r 0 P W h i C h W a S » V amplifiers were no, necessary Input 
not arise. 
Determination of ap p r o p r i a ,e ffl«er „ In determining the fl„er current the 
* multiplying the instantaneous voltage values from the ratio of rms curren, to rms 
voltage. This assump,ion is valid as ,he THD was very low in the voltage signal 
44 
Simulation to implement active power filter: Waveform of the supply voltage was 
assumed to be of pure sinusoidal. However, practically certain level of distortion 
specifically in the maximum and minimum points of the waveform was observed 
violating those assumptions. Yet, since the distortion effect was not significant it did 
not cause noteworthy alternations in the results. 
Physical size of the inductor was a considerable issue when selecting the inductor. 
Selecting inductors larger than 50 mH was not practical since the purpose was to 
eliminate harmonics in a computer which is comparatively small. Also the current 
injection inductor must be small enough so that the injected current di/dt will be 
greater than that of the reference current (the compensating current signal in the 
active power filter) for the injected current to track the reference. 
One step up transformer was used with a rectifier bridge to deliver the required 
voltage to charge the dc capacitor. Capacitive voltage multipliers are a better option 
for the purpose since the space requirement could be reduced when compared to 
use of a step up transformer. 
In order to minimize the drive voltage, one capacitor with four IGBTs was configured 
as a H bridge. The same can be achieved by using two IGBTs with two capacitors 
instead of one. However this cause the drive voltage of the circuit would be doubled, 
which is not advisable when safety is considered. 
Even though extremely high switching frequencies could be applied in simulations 
such could not be used practically. Switching frequencies higher than 100 kHz s are 
practically difficult to achieve. 
45 
The resultant total current had least amount of higher order harmonics other than the 
fundamental. Higher frequencies are extremely dangerous to some components 
such as capacitor banks, power transformers. These higher order harmonics can be 
easily eliminated using simple passive low pass filters. But using a low pass filter in 
the filter circuit reduces its capability to eliminate higher order harmonics. Therefore, 
as simulation was carried out only for the filter circuit, the effect of low pass filter can 
not be simulated. 
46 
References: 
1. Ming-Yin Chan, Ken K.F. Lee, Michael W.K. "A case study survey of harmonic 
currents generated from a computer centre in an office building" Architectural 
Science Review, Sept, 2007 
2. Daut, H.S. Syafruddin, Rosnazri AN, M. Samila and H. Haziah, "The Effects of 
Harmonic Components on Transformer Losses of Sinusoidal Source 
Supplying Non-Linear Loads" American Journal of Applied Sciences 3 (12): 2131-
2133, 2006 
3. Application Note, AN102 "K-Factor Defined", Xitron Technologies, Manufacturers 
of Engineering and Production Test Equipment. 
4. R. Magureanu, S. Ambrosii, D. Creanga, L. Bratosin, A. Draghici," Active Power 
Filters, advanced Control", ATEE - 2002, 
5. Sasaki, H. and Machida, T."A New Method to Eliminate AC Harmonic Currents 
by Magnetic Flux Compensation", IEEE Trans. Pwr App. Sys., PAS-90, 1971, pp 
2009-2019 
6. Richard M. Duke and Simon D. Round "The steady-state performance of a 
controlled current active filter", IEEE Trans. Power Electronics, Vol. 8, No. 3, April 
1993, pp 140. 
7. V.B. Bhavaraju and Prasad N. Enjeti, "Analysis and design of an active power 
filter for balancing unbalanced loads", IEEE Trans. Power Electronics, Vol. 8, No. 
4 October 1993, pp 640. 
8. Janko Nastran, Rafael Cajhen, Matija Seliger, and Peter Jereb "Active power 
filter for nonlinear ac loads", IEEE Trans. PE, Vol. 9, No. 1, January 1994, pp 92. 
9. Mukul Rastogi, Ned Mohan and Abdel-Aty Edris "Hybrid-active filtering of 
harmonic currents in power systems", IEEE 95 WM 258-4 PWRD. 
10. W.K. Chang, W.M.Grady and M.J.Samotyj "Controlling harmonic voltage and 
voltage distortion in a power system with multiple active power line 
conditioners", IEEE 95 WM 257-6 PWRD. 
11. http://en.wikipedia.org/wiki/IGBT transistor on 30th November 2008 
47 
as»r,r ST/SV tfe M' 
—I. *•< i 
'rrsj}?-'
1 
iSCTTSSvj CX«| I f 
I —Hr^' 
jg | jJ^ 
1,-xauuv 
617 
siNawnmsNi 
1VNOIJLVNi? 
•XI ss8*bg|m«s AM.W1 m * L soosiw aprag 
JO 'LAD^MCPU.WN >A3W1 uoneaifdde |N jo FJ9)E| JO) 
suGiaia-. Buwopj aqi IRIM 0|qip>duico w swwap mas W *®Mp 
9.I8MPJB.I OVQ uooounjitjnui >o sp»jpuni| uo sj8)uno3 pue 'Q/| pai&p 
indirai Bo|W6 'sndui fc|«.ie jo> aaejisiui BuioiaiKiBwd s^Buis • 
sjsdtun^saipiws ai»\pjsi| 
mm S3WI 0)1 l»l!&P I|C fo uoijeinftjuoa ajewjoj . 
i»fcud jnoAuejs-diunf « OUM/UIOOIH 
\e 9|QE|IE.« speqimop ojewijos eaj) QPQ'J UEI)J SJO ^ . 
ASM SUQ»A|£WE PWUSMPE SQEUJ OI 
Buiyioj ueijHiuaitpuAs pu» 'BuusfiBui '6wa»i cmeiuwny • 
Sl»U»ACjdtUI 83UBlUJQ)J3d 
sawnoooi JO)A6a|outpaifiuuieaaspapwjqifiinjsj . 
'Ojpms iu8wamse8f,j 6uisn oiseg k '++3/3 igy^ 
13N' MpniS l«*!A *#0 pu» SMBd*3|eu6s AIQIAW lAO/sAMpUAfflq . 
'MiiA«t«1 "! uotiwydde JioA him a oj ticuejeuaf. apna sijeuiojny » 
•juejassy am fasn Dumitueifad 
ou i|iiM sweuiainsesui aiarose pue isej ajeaio 01 spinS y * 
fiuppu) S8JI11B8) 
intv, iuetudo(8A»p jnoA dn spaads aieyjjos s«w8s utauBinseau in 
» »OI siM/bftpfuo DDM/*ia »* MI 
MSW 'AUDWN JO U0KJ8A B«IS| ET|J UCIQO OX 
sjewjos oipnig wauMimw puc Ki/wopw^q^ sssjdjgieuSis 
M3IW1 'M3IAW IN U itwudop.isp UOTIBMJDCJE JO) ai«AI)TS JSWP 
XMDVQ-IN t® wtsJ8Ajs9)e| eq) Suisn spuawuKmi siuatunnsuj lajotjejg 
»uiuu9tudo|aA9p pue uoiem&juo:) 'dmas uiajais aonpai 01 paiftsap 
spoi jaijio pue 'sjueiasse q/\ siooj uoiiejiiDijuea 5Me)j8iui 
fiunuuiejtoid uofjKsi|dde EAIJINIUI sspnpui 'SJBMJJQS jSAup xuipygiN 
pUlWJB )|N1C| '8JI!M))0S S03W8S U!8UI3JIKesSUJ SlUaUJIUlSU| IStKJIlBN 
8J(?Mj}os pspuaiiiuiooey 
sitwisAs papp9C|ae JOJ Q/| se su«neaidde . 
fs|i8i8p »t amiap8»B/i*oom IBI/, 
. 6«3i*l aawpwtf t»stm psseq-qe) «Aie«awi A|«a|diue» 
») weMpiei) ova jo dHisuewm wspms - #sn qei OMUspec*? . 
Bisffioi eiep ?>l3e>|0,<>.10 lejtoiuunjL'.usAsea pm ^ort -&ii£«6«| eiea . 
:spii|w.i saiduueig 
|Hii»ss3 we Aii3i|dujis pue '.IOOTI HUOJ neuis isoo MO| 8J9t|« 
SU0II93FJDD& )C .isqtunu E JOJ W TO 6009-aSfl pue 8009*380 
'suowaidde wsuisjnsesiu 
xqduios AIOTU JOJ q6nou« 8|!)TSJ8A SITMMJNSRAUI SIOINB 
JO) 10TIOU8 8|du»s m ssfifcai m^i 7a woauuoo gsn Ac|d-Pue-6n|d 
MMM s9sw«p (DVQi waiKtntoe ejep dRjsj^ IN oisiuod Anus jsoa 
-MO( me 6009-asft pue 8009-8Srt IN suouoidde iuatuwseaui 
JOJ 33101)3 jo snq aieo e oiu seq gsn 'siuaiuruBtii (euoietg 
UK)J) SUOIlBACUUf M3U ptB SltHUMAQKltU ifiptMpUeq JU958J 
suoijBai|tJdy pue MSIAJSAQ 
J9»na»3 
SOS(t 
S'4C 
tu 
»J»0 
mi 21 
mi u 
mi mm 
8M)li|0S«a 
U«t»0 BlSO 
01 n 
OJfqp a n 
MB »Nt *lmr.it t »HTj Iiij|it8«:s 
Ifidui »t)U| 
• H®p 'sc'mM 
10 MS 8 3sn 9»}'8Sn 
K FC^S ? 9CT MTO'0SRI 
5«8 CTpoy 
f J 
# 
* 
• 
• 
• 
• 
• 
I « 
« • 
• 
# 
# 
'sui8iȣ Gtijeiwio io) wg 
xwova-IN pe«|u.*Bp oi p8« nr^, 
31 ssajd*3iKiO{$ M31W1 • 
{pa|>R|3«i| aj«MU»s 
sa,-iiAi»5 Hiaiiiajiisesj^ 
++3/3ISNV « 
13N'»)S83iensiA'#0 • 
aiewjos sHfiieduioa tatjio 
Ojpms iusiuamseet-N • 
sswigjetifsis mmn • 
M3IA«H • 
t,«wr)» 
|XS0»N • 
C«0S'd)C'tW-f9 puc -ztl 6KIA Mwpatu • 
smwAS KHIIWIO 
fwvmmtfri » » 
IJUIBO&il » 
JewnoozHMS'JW-iK * 
S8H10/1 l«liJ!P SOTO/lli Z\ * 
mq?! wondiixtfepuBz . 
s/SWldn 
SJKJ 11 * zi W sjndui TEFJUE A « 
smdu| 6O|EUV 8 S/S>| 8T? O\ DN NG-N JO -U 
asn m ova uoiiou^ii|n|/\| PAIOAAOJ-SNG 'ISOQ-MO-| 
2-xeuuv 
Annex-3 
MATLAB commands used for increasing sampling rate 
I(101,:)=I(1,:); 
N=100; 
for i = 0:99 
for j = 1 :N 
11 (i*N+j,:)=l(j+1,:)+((I(i+2,:)- I(i+l,:))*j/N); 
end 
end 
for k = 1:10000 
T(k,:)=(k-1)*0.000002; 
end 
Tl=[T;T+0.02;T+0.04;T+0.06;T+0.08]; 
I2=[I1;I1;I1;I1;I1]; 
Load _current = [T1 12]; 
V(101,:)=V(1,:); 
for 1 = 0:99 
for m = 1 :N 
V I (l*N+m,:)=V(l+l,:)+(( V(l+2,:)- V(l+l,:))*m /N); 
end 
end 
V2=[V1;V1;V1;V1;V1]; 
Load _voltage =[T1 V2];