Bookchapter Fke01 PDF

First Edition 2008
MAKBUL ANWARI 2008
Hak cipta terpelihara. Tiada dibenarkan mengeluar ulang mana-mana bahagian artikel,
ilustrasi, dan isi kandungan buku ini dalam apa juga bentuk dan cara apa jua sama ada
dengan cara elektronik, fotokopi, mekanik, atau cara lain sebelum mendapat izin bertulis
daripada Timbalan Naib Canselor (Penyelidikan dan Inovasi), Universiti Teknologi
Malaysia, 81310 Skudai, Johor Darul Tazim, Malaysia. Perundingan tertakluk kepada
perkiraan royalti atau honorarium.
All rights reserved. No part of this publication may be reproduced or transmitted in any
form or by any means, electronic or mechanical including photocopy, recording, or any
information storage and retrieval system, without permission in writing from Universiti
Teknologi Malaysia, 81310 Skudai, Johor Darul Tazim, Malaysia.
Perpustakaan Negara Malaysia Cataloguing-in-Publication Data
Recent trends in renewable energy and power electronics research / editor

Makbul Anwari.
Includes index
ISBN 978-983-52-0639-9
1. Renewable energy sources. 2. Power electronics. I. Makbul Anwari.
333.79
Editor: Makbul Anwari

Pereka Kulit: Mohd Nazir Md. Basri & Mohd Asmawidin Bidin
Diatur huruf oleh / Typeset by

Fakulti Kejuruteraan Elektrik
Diterbitkan di Malaysia oleh / Published in Malaysia by

PENERBIT
UNIVERSITI TEKNOLOGI MALAYSIA
34 38, Jln. Kebudayaan 1, Taman Universiti,
81300 Skudai,
Johor Darul Tazim, MALAYSIA.
(PENERBIT UTM anggota PERSATUAN PENERBIT BUKU MALAYSIA/
MALAYSIAN BOOK PUBLISHERS ASSOCIATION dengan no. keahlian 9101)
Dicetak di Malaysia oleh / Printed in Malaysia by

UNIVISION PRESS SDN. BHD.
Lot. 47 & 48, Jalan SR 1/9, Seksyen 9,
Jalan Serdang Raya, Taman Serdang Raya,
43300 Seri Kembangan,
Selangor Darul Ehsan, MALAYSIA.
Contents
CONTENTS
Preface v
Chapter 1 The Role Power Electronics in Future Energy 1

Systems and Green Industrialization
Malik Elbuluk and Nik Rumzi Nik Idris
Chapter 2 Design and Development of a Maximum 21

Power Point Tracking Converter for Building
Integrated Photovoltaics
Tan Chee Wei, Tim C. Green and Carlos A.
Hernandez-Aramburo
Chapter 3 Analysis of Perturbation Step Size for 40

Perturb and Observe Photovoltaic Maximum
Power Point Tracking Algorithm
Tan Chee Wei, Tim C. Green and Carlos A.
Hernandez-Aramburo
Chapter 4 Load Sharing Characteristic of Single Phase 58

PV Inverter Connected to Three Phase Grid
M. Imran Hamid, Makbul Anwari and Taufik
Chapter 5 Power Quality Behavior of Single Phase Fed 71

Adjustable Speed Drive from Grid of PV
Generation
Makbul Anwari, M. Imran Hamid and Taufik
Contents
Chapter 6 Development of a 5 kW Inverter for AC 84

Power Supply Applications
Awang Jusoh, Naziha Ahamd Azli and Zainal
Salam
Chapter 7 Cost Comparison for a Stand-Alone 97

Residential Photovoltaic System Between
Silicon and Cadmium in Malaysia
A Vigneswaran, Md Shah Majid, Hasimah
Abdul Rahman and Mohd Yusri Hassan
Chapter 8 A Modular Structured Multilevel Inverter 114

for Fuel Cell Applications
Naziha Ahmad Azli and Abdul Halim Mohd
Yatim
Index 129
vi Preface
PREFACE
Power electronics technology is used for efficient control of

motors in electric cars, industrial and consumer motor drives, and
the development of more reliable, lightweight switching power
supplies for sophisticated computer and communication
equipment. Recent advances in microprocessor technology are
leading to the integration of microelectronics and power
electronics technologies for smart and efficient control of motor
drives and power supplies.
Renewable energy systems that take advantage of energy
sources that wont diminish over time and are independent of
fluctuations in price and availability are playing an ever-increasing
role in modern power system. Solar-powered photovoltaic systems
are entering the marketplace; fuel cells that will generate
electricity without pollution are on the horizon.
Energy Conversion Department has emerged as a leading
department for power electronics and drive research and education
in Malaysia. Research programs undertaken by the department are
diversified and include the following disciplines: modeling,
simulation, and control of power electronics; grid-connected
inverters, inverters for photo-voltaic (PV) applications; and electric
drives.
A vast amount of technology in switching power supply and
electric drives has been generated at ENCON in the past three
years. The research results have been published in numerous
conference proceedings and technical journals. These papers serve
as invaluable teaching tools and reference guides for power
vi Preface
electronics engineers. To facilitate future referencing, ENCON has

organized in three books. Each book is grouped according to these
topical areas:
Part I : Modeling and Control of Power Converters and Drives

Part II : Recent Advances in Power Inverters
Part III : Recent Trends in Renewable Energy and Power
Electronics Research
ENCON will strive to maintain, update and expand the series.

Future plans for Energy Conversion Publication Series are
additional books with new topics of interest to industry and power
electronics community.
I specifically want to thank all contributors in this book, titled
Renewable Energy and Power Electronics Applications.
Makbul Anwari
Faculty of Electrical Engineering
Universiti Teknologi Malaysia
2008
1
THE ROLE POWER ELECTRONICS IN
FUTURE ENERGY SYSTEMS AND
GREEN INDUSTRIALIZATION
Malik Elbuluk
Nik Rumzi Nik Idris
1.1 FUTURE ENERGY SYSTEMS
Presently most of the worlds energy supply comes from fossil

and nuclear sources. In spite of the increasing issues of resource
limitations and environmental pollution, these sources continue to
be important in providing energy for the next few generations.
However, these sources are expected to be less available and are of
environmental concern. To meet the increasing global demand for
energy and to allow for the decrease of environmental pollution,
the uses of alternative or clean energy sources, which do not
depends on fossil fuels and has a tolerable environmental impact,
is increasing. Alternative energy sources (AES) like micro-
turbines, photovoltaic (PV), fuel cells, wave energy and wind
turbines seem to meet the requirement for clean energy. Such
resources have not yet become a major part of the electrical power
system grid. Integrating these generating sources into the existing
power system grid, known as distributed energy resources (DER),
can provide a more reliable and better quality power to consumers.
The electricity obtained from these energy sources is normally
unregulated and is not in a suitable form to be directly connected
to the power system grids or to non grid-connected loads.
Therefore, regardless of the power ratings of the alternative energy
2 Recent Trends in Renewable Energy and Power Electronics Research
generation unit, it has to be converted to a suitable form by

utilizing power electronic converters. Research on the issues
related to the interconnections of DERs with the utility grid is
currently being conducted worldwide [1],[2],[3]. The applications
of micro-turbines in distributed generation, including modeling,
performance analysis, and use of power electronic converters, are
addressed in [4][5][6]. Control techniques and inverter topologies
used for interfacing PV systems with the utility have been
investigated in [7][8][9]. Methods of preventing unintentional
islanding for PV systems are discussed in [8][10]. Distributed
generations based on wind technologies are also being investigated
[11]-[16].
Implementing DERs can be as simple as setting up backup
electricity generators not connected to the grid, or it can be a
sophisticated power generation station with energy storage, control
and management systems, which can receive or supply energy to
the grid. When connected to the grid, the role of power electronic
converters and controllers is not only to convert electricity to a
suitable form, but is also responsible for ensuring that the
distributed generation unit connected to the grid will not create
problems that can impair the quality of the power supply and the
safety of all other equipment connected to the point of common
coupling. The power ratings of the distributed generation units
(ranging from few kilowatts to tens of megawatts), the types of
converters and topologies used for interfacing depend on the
available resources and on consumers. In more recent papers
[17][18], problems related to the transition between modes of
operations of DERs are discussed; algorithms during transitions
are developed and implemented using digital signal processing
(DSP)-based systems. In most of the technologies used in DERs,
energy storage devices (ESD) have to be included whether or not
the systems are connected to the grid. The various configurations
of power electronic converters used in different type of energy
sources for DER systems are briefly discussed below followed by
a short description of types of ESD that can be used in DERs.
The role power electronics in future energy systems 3
1.1.1. Micro Turbines
Micro-turbines are small combustion turbines with outputs power

ranging from 25kW to 500kW with natural gas as their fuel. They
have several advantages including lower emission, high efficiency
and compact size. They can also be used for combined heat and
power (CHP) technology. They have single shaft that rotates at a
very high speed of over 40,000 rpm. A high-speed permanent
magnet generator, air compressor and turbines are all mounted on
the same shaft with air bearings [4][6]. Since micro-turbines have
single shaft with no gear-box, ac voltage generated from the high
speed generator will have a very high frequency therefore cannot
be directly used by the loads or directly connected to the grid,
which operate at lower frequency of 50-60 Hz. Figure 1 shows the
typical power electronic electronics used to convert electric energy
available from a micro turbine to a suitable voltage applied to the
loads. The ac voltage generated by a micro-turbine is rectified and
coupled to an inverter via a dc link voltage. The inverter will
convert the dc voltage to a regulated ac voltage suitable for the
load and compatible with the power system grid. Micro-turbines
respond slowly to a sudden change in power demand. The time
constants for the changes in power output of micro-turbines can be
from 5 ms to 5 s [4]. As such, it is necessary to install ESD
coupled to the dc link to compensate for transient power demand
that cannot be supplied by the micro-turbines. In Figure 1, the ESD
are coupled to the dc link via a bi-directional power flow dc-dc
converter. The bi-directional power flow of the dc-dc converter is
required to ensure that the ESD can be charged-up by the grid or
the micro-turbine.
Figure 1. Power electronic converters for micro-turbine

technology
1.1.2 Photovoltaic (PV) Systems

Photovoltaic (PV) systems are very flexible and modular. They can
be sized to give power ranging from 100 watts to few hundreds of
kilowatts; however they can be very expensive for large power
ratings. For DERs, applications can range from a roof-top PV
system for a household with power ratings between 1-5 kW or a
large PV system owned by utility, with power ratings of more than
100 kW [19]. The power available from PVs fluctuates depending
on the weather and is completely unavailable at night. Therefore,
in practice, PV systems are operated with ESD or connected to the
grid to ensure continuous power supply to the load. Normally, to
increase the efficiency, a maximum peak power tracker (MPPT) is
used to track an operating point such that maximum power can be
extracted from the PV arrays. Typical power electronic converters
arrangements for PV systems are shown in Figure 2. The bi-
directional power flow of the dc-dc converter is required to ensure
that the ESD can be charged-up by the grid or the PVs.
Figure 2. Power electronic converters for micro-turbine

technology
1.1.3 Fuel Cells

A fuel cell converts hydrogen and oxygen into electricity and heat
through electrochemical energy conversion process. Fuel cells are
suited for distributed generation due to their low emission and high
efficiency. Power rating of a fuel cells system can be from 100
kW to few MW. Heat produced from the electrochemical reaction
for electricity production can be used for CHP applications. The
dc voltage generated from the fuel cells may be too low for an
inverter to operate efficiently; therefore a dc-dc converter is
normally used to step up the voltage to a higher level. Typical
power electronic converters used for a fuel cells system are shown
in Figure 3. Due to a large time constant for changes in power
output of the fuel cells (much like the micro-turbines) [4], ESD
normally have to be installed to enable the generation unit to track
a sudden increase in load demand. Fuel cells are also slow to
respond to sudden load drop [20], therefore ESD will provide a
path for the fuel cells to send power while they adjust for a lower
power demand. Figure 3 shows inclusion of ESD coupled to the
dc link via a bi-directional converter.
Figure 3. Power electronic converters for fuel cells system
Figure 4. Singly fed induction generator wind turbine
1.1.4 Wind Turbines

Wind turbines are gaining popularity due to their clean energy and
lower cost. The power ratings of wind turbines can range from
100 kilowatts to a few megawatts. Wind turbines can be found
either in fixed speed or variable speed applications. In fixed speed
application, the generator, which is mechanically coupled to the
wind turbine, is directly connected to the utility grid, which means
that at a given operating point, the rotor has to rotate at a fixed
speed governed by the frequency of the utility. Fixed speed wind
turbines result in severe mechanical stresses caused by the
variation in wind energy and disturbances in the network
[21][12][13]. Variable speed wind turbines on the other hand,
operate at variable speed and therefore can capture maximum wind
energy compared to the fixed speed operation [11]. Instead of
being directly connected to the grid, variable speed configurations
are connected to the grid using power electronic converters. The
most common types of generators used in variable speed wind
turbines are the induction and synchronous generators [15]. Other
types of generators include the use of low speed machines to avoid
the use of a gear box, such as a switched reluctance generator [16]
or more generally, multi-pole machines [13][14]. Several possible
topologies are used in variable speed wind turbines [13] that can
be classified into to two groups: direct-in-line wind turbines and
doubly fed generator wind turbines. An example of the first group
that uses a squirrel cage induction generator is shown in Figure 4.
The induction generator is mechanically coupled to the wind
turbine via a mechanical gearbox. The variable frequency voltage
of the generator is rectified and fed to the inverter to produce a
constant frequency voltage that can be connected to the grid. An
example of a doubly fed machine configuration can be similar to
Figure 4 but it uses wound rotor induction machines. The doubly
fed configurations are gaining popularity due to their higher
efficiency and lower cost compared to the direct in line
configurations [21]. Connection to the grid may require a
transformer in order to match the generated and the grid voltages.
Due to the intermittent nature of the wind, for a continuous supply
of power to the load, connection to the grid or the use of ESD is
necessary.
1.2 ENERGY STORAGE DEVICES
DERs respond relatively slowly to a sudden increase in load

demand. It is reported that for micro-turbine and fuel cell systems,
the time constant for the power output can range from 5 ms to 50 s
[4]. If the DER system is not connected to the utility grid and no
ESD are installed, due to this slow response, a sudden increase in
load power demand will result in a voltage reduction to achieve
energy balance. There are a few types of ESD that are being used
and researched in power system applications: batteries, flywheels,
ultra-capacitors (or super-capacitors) and superconducting
magnetic energy storage systems (SMES) [22]. Battery
technologies, particularly the leadacid batteries, are widely used
in power system applications due to their low cost, high energy
density and capability, and their established and mature
technology. Flywheel technologies are gaining popularity [23] and
have been applied in wind turbine technologies [24][25]. At
present, ultra-capacitors or super-capacitors are mainly used for
low-energy, high peak power applications [22]. They have been
used in drive systems to improve ride-through capability during
voltage sags; their applications in DER have yet to be found.
SMES systems are known for their fast response and high
efficiency, however compared to other energy storage systems they
are presently expensive [22]. In power system applications, SMES
have been used for voltage sag compensation and to improve
stability performance [26][27].
1.3 POWER FLOW AND CONTROL
The DER system consists of power electronic converters and a

fast digital signal processor (DSP), as shown in Figure 5. It can be
operated as either in stand-alone or grid-connected modes. The
configuration of the power electronic converters depends on the
energy sources available as discussed in the previous section.

There can be more than one energy source block; each type will
use a different dc-dc converter (or rectifier). The dc-dc converter
used for ESD must be of bi-directional power flow type to enable
charging and discharging of the storage devices. The inverter used
to interface with the load or grid can be operated in voltage- or
current-controlled modes. Alternatively, the power electronics can
also be used to supply power to the dc loads in which case, a dc-dc
converter (bi-directional power flow, if required) may be needed to
convert the dc voltage level appropriately. If the ESD have to be
re-charged using the grid, the inverter will be operated in
rectification mode and possibly with unity power factor. All of the
control signals for all the converters are generated by the fast
controller such as a DSP. The separating device and the sensors
block consists of solid-state switches and sensors, which are used
to isolate the unit from the grid and to re-connect it back to the grid
depending on the needs. The capabilities of the power electronics
are:
Regardless of whether it is connected to the grid or not, the
power electronics must be capable of providing a continuous
supply to the load with acceptable power quality. This means
that it must be capable of changing from a stand-alone mode to a
grid-connected mode (and vice versa) seamlessly. It must be
capable of avoiding unintentional islanding when an outage
occurs on the grid by disconnecting itself from the grid using the
separating device.
For maximum energy utilization, the power electronics must be
capable of managing the power flow between the energy sources,
ESD, the grid and the load. For example, it must be capable of
determining the optimum sources of energy (either from the grid,
energy sources or ESD) that can be used to supply the load at
any particular time, to give maximum efficiency and minimum
cost. It must know when to re-charge the ESD whether to use
the grid or the energy sources, or the combination of both, for
optimum cost and efficiency.
When connected to the grid, the power electronics must maintain

the power factor of the unit within the permissible value as seen
by the grid by controlling the active and reactive power flow.
The amount of harmonics injected to the grid must be within the
limits as imposed by the IEEE 519-1992 standard [33].
Figure 5. Power flow and control of the energy sources, ESDs,

the grid and loads
1.3.1. Stand-Alone Mode

Figure 6 illustrates a block diagram of the power controller
operating in stand-alone mode. The power processor in the figure
consists of power electronic converters, depending on the type of
available energy sources as discussed previously. The energy
sources can be from PV, fuel cells, micro-turbines, wind turbine or
the combinations of these sources. In stand-alone mode, the
inverter in the power processor block will be operated in voltage-
controlled mode supplying power to the loads. A power controller
operates in its stand-alone mode for two different reasons: (1) It is
supplying loads in remote area where connection to the grid
system is too expensive or impossible, for example, rural areas in

the developing countries, aerospace, military or automobile
applications (2) it is supplying critical loads and is disconnected
from the grid due to the failure on the grid side, such as hospitals,
banks or factories.
Figure 6. Power flow and control: stand-alone mode operation
1.3.2. Grid-Connected Mode

Interconnections of DER to the grid can be categorized into
two types: (1) interconnections with power only drawn from the
grid, (2) interconnections with power drawn and supplied back to
the grid. It is the second category of interconnections that concerns
the utility, especially with regard to the power quality and safety of
personnel or equipment connected to the grid. Figure 7 illustrates a
block diagram of power controller operating in grid-connected
mode with the load connected in parallel with the grid. Also shown
is the dc load that may be connected to the dc link via a dc-dc

converter. In this mode of operation, assuming that the grid
voltage is stiff, the inverter of the power processor will be operated
in current-controlled mode. The separating device and sensor
block is use to separate the power controller from the grid system
whenever a power failure is detected on the grid side, i.e. changing
from grid-connected to stand-alone modes. A control algorithm
will be developed to exploit the fast processing capability of the
DSP in order to ensure seamless operation as seen by the loads.
The grid-connected mode may or may not include ESD, with
several factors that determine this choice, as discussed below.
If the grid is intended as a backup supply to the load, ESD may
not be required. Power is obtained from the grid when power from
the energy sources is unavailable. With no ESD, power required
for a sudden load change or during peak power demand is obtained
from the grid and the energy sources. However, power outage on
the grid will result in the power supplied to the load being totally
dependent on the energy sources. Thus, due to slow response of the
energy sources to react to an instantaneous sudden load demand,
the quality of the power supplied to the load will be impaired. In
the worst case, when the energy sources are not generating power,
power outage on the grid will result in no power available on the
load. If the energy sources are sized to supply the load only during
average power usage, then during power outage on the grid, peak
power demand cannot be fulfilled. Excess energy generated by the
energy sources can be fed back only to the grid.
There are a few possible sources of energy from which the load
can obtain the power, i.e. from the energy sources, from the ESD,
from the grid or from the combinations of these sources. The
choice of source or combination of sources used to supply the
loads has to be made carefully to ensure maximum and efficient
energy utilization. This depends on several factors, such as the
instantaneous power available from the sources, energy on the
ESD, load characteristics, price of energy from the grid at that
instant and the desired power factor. This decision has to be made
by the controller for optimum energy utilization and cost. During

power outage on the grid, the energy sources and the ESD supply
power to the loads therefore instantaneous peak power demand can
be fulfilled. The ESD can be charged-up either by the grid or
energy sources depending on the instantaneous power available
and the load demand. The sizing of the ESD has to be determined
based on the load and energy sources characteristics. The ESD are
used to compensate the high transient power demanded by the
loads that cannot be delivered by the energy sources due to their
slow response.
Figure 7. Power flow & control: grid-connected mode
1.4. GREEN INDUSTRIALISATION
Green industrialization is a broader term concerned with

securing reliable, inexpensive supply of energy that is both
sustainable and environmentally friendly. The advantage is the

ability to incorporate advanced environmental technologies with
clean energy sources in manufacturing. These technologies can be
agricultural-related practices and food processing, recycling of
waste water and production of portable water, renewable energy
resources, biodegradable materials, industrial biotechnology and
pollution free engineering processes.
Besides the DERs and electric power systems, the beneficiaries of
green industrialization are agriculture, biotechnology chemicals,
pesticides, and Fertilizers, defense systems, food processing
Industries, health and family welfare, heavy industries, non-
conventional energy sources, paper & pulp industries, petroleum
and natural gas, pharmaceuticals, transportation including
railways, shipping, road transport and aerospace applications. In
all of the technologies that make or benefit from green
industrialization, power electronics plays an important role in
interfacing the energy systems needed to supply these systems.
1.5 SUMMARY
Power electronics is a key and enabling technology for many

industrial segments such as automotive, home and office,
communication, automation, energy supply and distribution.
There has been a very dynamic progress in power electronics and
its use due to advancement in materials, power semiconductor
technologies and system integration. The recent trends for the use
of alternative energy to replace fossil energy have increased the
demand for power electronics converter to interface these energy
sources to the electric power grid and to other non-grid-connected
loads. Therefore, as the search continues for clean energy and
green industrialization, power electronics is expected to play a
vital role in future energy systems.
REFERENCES
[1] http://www.eren.doe.gov/der/.
[2] R. B. Alderfer, M. M. Eldridge and T. J. Starrs, Making
Connections case studies of interconnection barriers and
their impact on distributed power projects, National
Renewable Energy Laboratory, May 2000.
[3] IEEE Distributed Resources and Electric Power Systems
Interconnection(P1547).
http://grouper.ieee.org/groups/scc21/1547/index.html
[4] R. Lasseter, Dynamic models for micro-turbines and fuel
cells Power Engineering Society Summer Meeting, Vol. 2,
pp. 761 766, 2001.
[5] M. Etezadi-Amoli and K. Choma, Electrical performance
characteristics of a new micro-turbine generator, Power
Engineering Society Winter Meeting, 2001. IEEE, Vol. 2,
pp. 736-740, 2001.
[6] M. Suter, Active filter for a micro-turbine, Twenty-Third
International Telecommunications Energy Conference
INTELEC 2001, pp. 162 165, 2001.
[7] P.G. Barbosa, L.G.B. Rolin, E.H. Watanabe and R. Hanitsch,
Control strategy for grid-connected DC-AC converters with
load power factor correction, IEE Proc- Generation,
Transmission and Distribution, Vol. 145, No. 5, pp. 487-491,
1998.
[8] G.A. Smith, P.A. Onions, D.G. Infield, Predicting islanding
operation of grid connected PV inverters, IEE Proceedings-
Electric Power Applications, Vol. 147, No. 1, pp. 16, Jan.
2000.
[9] S. Saha and V.P. Sundarsingh, Novel grid-connected
photovoltaic inverter, IEE Proceedings-Generation,
Transmission and Distribution, Vol. 143, No. 2, pp. 219
224, March 1996.
[10] Huang, S. J., Pa, F.S., Design and operation of grid-

connected photovoltaic system with power-factor control and
active islanding detection, IEE Proceedings Generation,
Transmission and Distribution, Vol. 148, No. 3, pp 243
250, May 2001.
[11] T. Thiringer and L. Linders, Control by variable rotor speed
of a fixed pitch wind turbines operating in a wide speed
range, IEEE Transaction on Energy Conversion, Vol. 8, No.
3, pp. 520-526, Sept. 1993.
[12] S.A. Papathanassiou and M.P. Papadopoulos, Mechanical
stresses in fixed-speed wind turbines due to network
disturbances IEEE Transaction on Energy Conversion, Vol.
16, No. 4, pp. 361367, Dec. 2001.
[13] L.H. Hansen, et al. Generators and power electronics
technology for wind turbines 27th IEEE industrial
electronics society annual conference (IECON),
[14] S.A. Papathanassiou, G.A. Vokas and M.P. Papadopoulos,
Use of power electronics converters in wind turbines and
photovoltaic generators, ISIE 95.
[15] J. Usaola and P. Ledesma, Dynamic incidence of wind
turbines in networks with high wind penetration, Power
Engineering Society Summer Meeting, Vol. 2, pp. 755-760,
2001.
[16] M.A. Mueller, Design of low speed switched reluctance
machines for wind energy converters
Ninth International Conference on Electrical Machines and
Drives, pp. 60-64, 1999.
[17] S. Sivakumar, T. Parsons, S.C. Sivakumar, Modeling,
analysis and control of bi-directional power flow in grid
connected inverter systems, Proceedings of the Power
Conversion Conf. PCC-Osaka, Vol. 3, 2002.
[18] R. Tirumala, N. Mohan, .C. Henze, Seamless transfer of
grid-connected PWM inverters between utility-interactive
and stand-alone modes, Seventeenth Annual IEEE Applied
Power Electronics Conf. and Exp., APEC 2002, Vol. 2, pp.

1081-1086, 2002.
[19] B. Kroposki and R. DeBlasio, Technologies for the new
millennium: Photovoltaic as a distributed resources, IEEE
Power Engineering Society Summer Meeting, Vol. 3, pp.
17981801, 2000.
[20] L. E. Lesster, Fuel cell power electronics,
www.satcon.com/news/pr/fuelcellarticle.pdf, SatCon
Technology Corporation, Cambridge, Massachusetts, USA.
[21] S. Muller, M. Deicke, R.W. De Doncker, Doubly fed
induction generator systems for wind turbines, IEEE
Industry Applications Magazine, Vol. 8, No. 3, pp. 26-33,
May-June 2002.
[22] P. F. Ribeiro, B.K. Johnson, M.L. Crow, A. Arsoy and Y.
Liu, Energy storage systems for advanced power
applications, proceeding of the IEEE, Vol. 89, No. 12, Dec
2001.
[23] R. Hebner, J. Beno and A. Walls, Flywheel batteries come
around again, IEEE Spectrum, Vol. 39, No. 4, pp. 46 51,
April 2002.
[24] T. Boutot, Liuchen Chang and D. Luke, A low speed
flywheel system for wind energy conversion, IEEE
Canadian Conference on Electrical and Computer
Engineering CCECE 2002. , Vol. 1, pp. 251-256, 2002.
[25] R. Cardenas, R. Pena, G. Asher and J. Clare, Control
strategies for enhanced power smoothing in wind energy
systems using a flywheel driven by a vector-controlled
induction machine, IEEE Trans. on Industrial Electronics,
Vol. 48, No. 3, pp. 625-635, June 2001.
[26] S. Polmai, T. Ise and S. Kumagai, Experiment on voltage
sag compensation with minimum energy injection by use of
a micro-SMES, Proceedings of the Power Conversion
Conference PCC-Osaka, Vol. 2, pp. 415-420, 2002.
[27] A. Abu-Siada, W.W.L. Keerthipala, and W.B. Lawrance,
Application of a super-conducting magnetic energy storage

unit to improve the stability performance of power systems,
IEEE Canadian Conference on Electrical and Computer
Engineering CCECE, Vol. 1, pp. 201-206, 2002.
[28] R.Solow, A Contribution to the Theory of Economic
Growth." Quarterly Journal of Economics 70 (February
1956), pp. 65-94.
[29] R.V. Ayres and A.V. Kneese, Production, Consumption and
Externalities, American Economic Review 59(1969): 282-
297.
[30] P. Roomer, "Endogenous Technological Change," Journal of
Political Economy 98, October 1990, S71-S102.
[31] V.Ruttan, Technology, Growth, and Development: An
Induced Innovation Approach, Oxford University Press,
2001.
[32] D. North, Institutions, Institutional Change and Economic
Performance, Cambridge University Press, 1991.
[33] IEEE Standard 519-1992 Recommended practice and
requirements for harmonics control in electrical power
systems.
[34] G. Joos, B.T. Ooi, D. McGillis, F.D. Galiana, R. Marceau,
The potential of distributed generation to provide ancillary
services, IEEE Power Engineering Society Summer
Meeting, Vol. 3, pp. 1762-1767, 2000.
[35] E. Goodstein, Economics and the Environment, John Wiley,
2002.
[36] B. W. Arthur, Increasing returns and path dependence in the
economy, University of Michigan Press, 1994.
[37] D. Bromley, Environment and Economy: Property Rights
and Public Policy, Blackwell, 1991.
[38] R. W. Kahn and G. L Hester, Marketable permits: lessons
for theory and practice, Ecology Law Quarterly 16, 1989:
380-391.
[39] B.A. Larson and M.K. Knudson, Whose price wins:
institutional and technical change in agriculture, Land

Economics 67, No.2, 1991: 213-224.
[40] R. M. Margolis and D.M. Kammen, Energy R&D and
Innovation: Challenges and opportunities for technology and
climate policy in S. Schneider, A Rosencranz and J-O Niles
(eds), A Reader in Climate Change Policy, Island Press,
2001.
[41] H. Lee and N. Darani, Electricity Restructuring and the
Environment, Harvard University Policy Group, December
1995.
[42] U.S. Department of Energy, Energy Information
Administration, The Changing Structure of the Electric
Power Industry 2000: an Update, U.S. Department of
Energy, Energy Information Administration, October 2000.
[43] P.L. Joskow and E. Kahn, A quantitative analysis of pricing
behavior in Californias wholesale electricity market during
Summer 2000, Energy Journal 23, No.4, 2002.
[44] J-M. Martin, Energy technologies: systemic aspects,
technological trajectories, and institutional frameworks,
Technological Forecasting and Social Change 53, 1996, pp.
81-95.
[45] S. Banzhaf, D. Burtraw, and K. Palmer, Efficient Emission
Fees in the U.S. Electricity Sector, Resources for the Future
Discussion Paper, October 2002.
[46] R. B. Alderfer, M. M. Eldridge and T. J. Starrs, Making
Connections case studies of interconnection barriers and
their impact on distributed power projects, National
Renewable Energy Laboratory, May 2000.
[47] R. Lesseter, et al., The CERTS microgrid concept, white
paper on integration of distributed energy resources prepared
for US department of Energy, Consortium for Electric
Reliability Technology Solutions, 2002.
[48] S. Stoft, Power System Economics, Wiley-IEEE Press, 2002.
[49] S. A. McCusker, B.F. Hobbs and Y.Ji, Distributed Utility
Planning Using Probabilistic Production Costing and

Generalized Benders Decomposition, IEEE Transactions on
Power Systems, 17 (May 2002): 497-505.
[50] D. Burtraw, K. Palmer, R. Bharvikar, The Effect on Asset
Values of the Allocation of Carbon Dioxide Emission
Allowances, Resources for the Future Discussion Paper,
March 2002.
[51] http://www.lumina.com.
[52] H. Hashimoto, A spatial Nash equilibrium model in P.
Harker Ed. Spatial Price Equilibrium: Advances in Theory,
Computation and Application, Springer-Verlag, 1995.
[53] W. E. Becker and Michael Watts. Teaching economics at
the start of the 21st century: Still Chalk and Talk,
American Economic Review (May 2001), pp. 446-451.
[54] A. W. Chickering and Z. F. Gamson, "Seven Principles for
Good Practice in Undergraduate Education" AAHE Bulletin,
1987, 39, No.7, pp. 3-7.
Design and Development of a Maximum Power Point Tracking Converter 21
2
DESIGN AND DEVELOPMENT OF A
MAXIMUM POWER POINT TRACKING
CONVERTER FOR BUILDING
INTEGRATED PHOTOVOLTAICS
Tan Chee Wei
Tim C. Green
Carlos A. Hernandez-Aramburo
2.1 INTRODUCTION
In urban areas, the implementation of PV technology,

especially medium to large scale PV systems remain difficult due
to the lack of available land for PV array installation. Therefore,
the idea of integrating photovoltaic modules into buildings has
developed, which is termed as Building Integrated Photovoltaic
(BIPV). BIPVs not only reduce green energy for the buildings
consumption but also can add value to the buildings aesthetic
appearance. The implementation of BIPV projects has been
intensively pursued by many countries worldwide [1-3].
Although the rapid development of photovoltaic materials
technology and the increased production volumes for PV materials
have led to a reduction of the PV module costs [4], the capital
costs of BIPV systems remain very high. In order to reduce the
systems pay back period, it is a necessity to design a power
converter that is able to optimize the energy production for given
PV materials. A Maximum Power Point Tracking (MPPT)
algorithm is frequently used to maximize the power drawn from
PV modules under varying weather conditions. MPPT converters

help to reduce a PV system cost by decreasing the size of PV area
required for a given energy yield. Among MPPT techniques, the
Perturb and Observe (P&O) algorithm is popularly applied because
of the simplicity of its control structure [5, 6]. However, there have
been relatively few studies made of the P&O MPPT algorithm in
conditions of a rapidly changing environment such as sudden
shadow changes.
This chapter introduces building integrated photovoltaic systems
and several important factors in BIPV design are discussed. At the
lowest level of a BIPV system, a MPPT converter can help to
make best use of the PV array. For that reason, a current-mode
controlled MPPT converter is demonstrated. It incorporates an
improved P&O MPPT algorithm in order to track the maximum
available solar power under abrupt changes of atmospheric
conditions. The experimental setup of the converter using a digital
rapid prototype platform is described and the experimental results
are discussed.
2.2 BUILDING INTEGRATED PHOTOVOLTAIC
The term building integrated photovoltaics refers to

photovoltaic modules that have been designed to form an integral
part of a building material, as opposed to standard photovoltaic
modules which are generally mounted onto an existing roof [10]. It
represents the combination of proven renewable power technology
and building materials suitable for traditional building practices. It
means that solar panels are planned and built along with the
building structure. This combination offers a number of advantages
[7, 8]:
1. Financial appeal costs are combined for a building material

and a power source while installation costs are reduced for new
build buildings or when a building is refurbished
2. Increased potential area for PV installation building integrated
PV can be installed on the building roof, building wall or glazed
surfaces using semi-transparent PV products

3. Easy integration to standard construction practices can be
retrofitted to existing constructions
4. Improved appearance modules can be architecturally elegant
and provide the building owner with a highly visible public
expression of their environmental commitment.
The integration of PV into a building can be done by the following

approaches:
Curtain wall for vertical and inclined facades
Pitched and flat roof mounted, roof light (semi-transparent PV
module)
Integrated roof cladding, sheeting and tiling
Rain screen cladding and sunshades.
The realization of BIPV projects and applications has been

intensively pursued by many countries [1, 2, 9]. Case studies
which highlight the design and development of BIPV have been
presented in [10] and several BIPV facades installed in Battery
Park City, New York, have been reported in [11]. These projects
have been successfully installed and the systems have been
maintaining well.
2.3 BIPV DESIGN CONSIDERATIONS
Practically, BIPV design requires such things as the knowledge of

the weather conditions, careful consideration of orientation and tilt
angle, and incorporation of a MPPT converter to made best use of
PV materials.
2.4 SOLAR IRRADIANCE AND TEMPERATURE
Figure 1 presents the solar irradiance and ambient

temperature that measured on a typical autumn day at Imperial
College London, United Kingdom on 8th October 2006. It is
noticeable that the solar irradiance changes drastically, which is
most probably due to the cloud shading. The open-circuit voltage

of the PV panel decreases with the rise of panels temperature.
However, it has less significant changes over a day because the
temperature variation is not large. Due to the characteristic of the
PV panel, production of solar energy depends predominantly on
solar irradiance. In [12] it was shown that there is a linear
dependence between the current at maximum power point and the
short circuit current. This suggests that the PV current is strongly
influenced by the MPPT operation. Without a proper MPPT, a
BIPV installation may fail to achieve its objective as the system is
unable to track the maximum available energy effectively.
Figure 2 shows the derivative of data in Fig. 1. This derivative,
Solar Irradiance/Time, shows the speed of solar irradiance
changes in time (rate of change). It gives an identification of how
quick the solar irradiance changes, such as the speed of moving
clouds. From Fig. 2, it can be seen that the maximum rate of solar
irradiance change is about 8 W/m2/s. This rate of change means
that a drastic change of 100 W/m2 solar irradiance would take up
to 12.5 seconds. The plot gives a brief indication of how fast a
maximum power point tracking should response to drastic changes
of solar irradiance.
800 20
Solar Irradiance (W/m2)
600 15
Temperature ( C)
400 10
200 5
0 0
6 8 10 12 14 16 18 20
Time (hour)
Figure 1 Solar irradiance and ambient temperature measured on 8

October 2006 at Imperial College London
9 0.01
0.005
0
8
Solar Irradiance / Time (W/m 2/s)
Temperature / Time (C/s)

7
0
6 8 10 12 14 16 18 20
Time (hour)
Figure 2 Derivative of solar irradiance and temperature from Fig. 1
2.5 PV ORIENTATION AND TILT ANGLE
Amongst the BIPV design considerations, PV module

orientation and tilt angle are the two important aspects because
they have significant effect on the yearly energy yield. The
optimum orientation and tilt angle ensure that the production from
solar energy system is at its maximum and best use is made of the
PV array. These parameters have significant effect on the yearly
energy yield. Some BIPVs may be constrained by the shading
problem (by buildings or trees) or the building structure itself (not
suitable to install PV module at the optimum angle). In most cases,
PV modules are integrated as a part of the building wall (facade)
and the building roof. In order to understand better the effects of
these factors, the expected PV energy yield using the weather data
of London, United Kingdom has been investigated. The optimum
tilt angle for PV module in London is approximately 30 [13].
Figure 3 demonstrates the average PV energy yield
computed for a standard mono-crystalline PV module. The effect
of different PV tilt angle installation for four directions of
orientation throughout a year is compared in Fig. 3a and Fig. 3b.
The yearly PV energy yield is shown in Fig. 3c. From the plots, it
is shown that the south orientated PV module produces the largest
amount of energy while the north orientated produces the least. For
east and west orientated cases, both produce approximately the
same amount of energy over a year for optimum tilt angle and
facade. For faade design, the annual energy produced is
significantly reduced by about 19% for south orientated and 22%
for both east and west orientated as compared to the case of
optimum tilt angle. It is not worth installing PV modules on the
north orientated building walls in London because it is shaded by
the building itself.
Energy (kWh/m2/Day)
1
0.8
0.6
0.4
0.2
0
0 50 100 150 200 250 300 350
Day
Energy (kWh/m2/Day)
1
0.8 North South West East
0.6
0.4
0.2
0
0 50 100 150 200 250 300 350
Day
Yearly PV Energy Yield (kWh/m 2)
200
Optimum tilt angle
Facade
150
100
50
0
North East South West
Orientation
Figure 3 Solar energy yield for different orientation in London, (a) at
optimum tilt angle, (b) faade installation; (c) comparison of the
annual PV energy yield for different orientation and tilt angle
2.6 CURRENT-MODE CONTROLLED MPPT

CONVERTER
The current-mode controlled MPPT converter employs an

improved P&O algorithm with adaptive variable step-size. The
detail development of the modified P&O algorithm and the effect
of perturbation step-sizes have been reported in [14]. The process
of the improved P&O MPPT algorithm is illustrated in the flow
chart in Fig. 4. The change of power observed after the previous
perturbation is classified into one of three magnitude bands and an
acceleration factor is set as a result. There are also four operating
modes depending on whether the previous perturbation moved
toward or away from the MPP and whether the operation is above
or below the optimum voltage. The four modes have different gain
factors applied to the step-size which itself is dependent on the size
of the previous step. The perturbation step, I, is added to the
reference current, Iref at each iteration of the algorithm.
Iref(k) = Iref(k-1) + I (1)
Start
Sense VPV(k) and IPV(k)
Calculate
PPV(k) = VPV(k) IPV(k)
Calculate
Ppv_diff = |PPV(k) - PPV(k-1)|
No
Ppv_diff > ?
Yes
Yes
Ppv > PMPP/2 ?
No
Multiply with Bypass
Multiply with
= 0.015 = 0.003 =1
Yes No
Ppv_diff > 0 ?
Yes No Yes No
VPV(k) - VPV(k-1) > 0 ? VPV(k) - VPV(k-1) > 0 ?
kr = k1 kr = k2 kr = k3 kr = k4
Calculate
I = kr (Ppv_diff / Ppv)
Iref(k) = Iref(k-1) + (I)

Return
Figure 4 Flow chart of the improved Perturb and Observe MPPT

algorithm
The additional term, acceleration factor , is included to

enable the sensitivity to be enhanced or reduced to achieve both
rapid movement toward the MPP following a large change in
characteristic and to aid accurate convergence of the MPPT on the
true MPP. The size of the perturbation step is made dependent on
the sensitivity of the PV power to the previous perturbation. The
sign of the step is determined by kr, which is the opposite of the
sign of the slope in the P-V characteristic curve:
P PV _ diff
I = kr (2)
P PV
PPV _ diff = Ppv(k ) Ppv(k 1) (3)
where Ppv(k) is the present PV power and Ppv(k-1) is the previous

measured PV power and kr is a one of four constants for the four
possible combination of perturbation direction and Ppv/V slope
direction. The perturbation cycle is repeated for any changes of
environmental conditions to maintain the PV power at MPP.
The MPPT converter is a DC-DC step-up converter which
applies multi-loop control concept to track the maximum power
operating point. The configuration of the current-controlled boost
converter together with the MPPT controller is illustrated in Fig. 5.
It consists of inner and outer loops. The outer loop is the current-
reference-loop, which is the MPPT algorithm that specifies the
value Iref for the reference of inner loop. The inner loop is the
current-mode control loop, which regulates the inductor current in
the converter corresponding to the value of Iref by using a PI
controller. This arrangement removes the duty-cycle to current
dynamics from the MPPT and allows the MPPT to act directly on
one of the variables that contribute to the power.
Figure 5 Configuration of the Current-Mode Controlled DC-DC step-

up MPPT converter
2.7 MPPT SYSTEM OVERVIEW
The purpose of the experimental work was to verify the

simulated results of the studied P&O MPPT algorithm. The test rig
needed be able to emulate the sunlight intensity (a modest amount
of solar irradiance) used in the simulation. In brief, the PV module
is connected to the modified rapid prototyping platform and, in
turn, to a resistor load. The P&O MPPT algorithm was initially
implemented using the Matlab/Simulink software and then
converted into the C language by using Code Composer Studio
so that it can be downloaded to the DSP/FPGA processor boards.
The input signals from the rapid prototype are converted into
digital signals using analogue to digital converters (ADCs).
The voltage and the current of the PV array are sensed and
displayed on both the oscilloscope and the personal computer with
the help of MATLAB/Real Time Workshop. The advantage of
using the prototyping system is the ease of changing control
parameters online and the facility for saving the output data in the
MATLAB/Simulink work environment.
2.8 EXPERIMENTAL SETUP
A laboratory PV illumination test rig has been built to

verify the performance of MPPT converter. The PV module used is
the BP.365U, 65 Wp multi-crystalline which was placed in a 3D
steel enclosure. The PV illumination test rig setup is shown in Fig.
6. The solar irradiance was emulated using a set of halogen lamps
which was powered by a DC power supply. The inner surface of
the enclosure was covered with aluminium foil to ensure uniform
illumination on the PV module. Besides that, the enclosure was
equipped with two cooling fans and vents to avoid the rapid
temperature rise attributed to the heat produced by halogen lamps.
The full illumination by turning on all the nine strings halogen
lamps corresponds to approximately 700 W/m2 of solar irradiance.
Figure 6 The PV illumination test rig Illumination enclosure with

BP365U PV module; A are the halogen lamps, B are the vents, C is
the PV module, D are the cooling fans and E is the PV terminal
connection
The converter was implemented by adapting a rapid

prototyping platform for power electronics [15, 16], in which the
IGBT switch was controlled using pulse width modulation (PWM)
technique. The switching frequency was chosen to be 15 kHz to
avoid high switching losses while the sampling frequency for the
control algorithm was selected to be 10 kHz. An inductor of 2 mH
was selected to keep the converter operating in continuous
conduction mode. The input capacitance Cin and the output
capacitance C were 220 F and 1.1.mF respectively. The converter
load was an 80 resistor.
The experimental platform also contained of an FPGA (Altera
EP1K100) in combination with a DSP (Texas Instruments
TMS320C6711) [15] and current-voltage measurement modules
[16]. The daughter-card with the FPGA provides PWM output to
the converter containing isolated gate driver for the IGBT. This
digital rapid prototype platform uses a software framework, called

DSK/RTW to translate graphical processing algorithm into C
application code that can be run on the DSP Starter Kit (DSK). It
allows quick implementation of control algorithms developed in
MATLAB/Simulink using the Real Time Workshop and the Code
Composer Studio with minimal user intervention. The DSK/RTW
tool preserves strict software modularity and associates a C
application code unit with the MPPT algorithm that was developed
in MATLAB/Simulink. This framework relies on the TI
DSP/BIOS Real Time Operating System and supports Simulinks
External Mode feature, which allows on-line parameter changes
and scope data extraction.
2.9 EXPERIMENTAL RESULTS AND DISCUSSION
In order to verify the performance of the designed MPPT

algorithm, the converter was examined under static and dynamic
performance. The static performance was tested under step
changes of solar irradiance to observe how fast and accurate was
the convergence of the operating point on the MPP. The dynamic
MPPT tracking was based on variable changes of solar irradiance
and how effective the operating point moves along the locus of
true MPP at the different irradiance levels. The measured
temperature of the PV module was in the region of 25 C to 32 C
throughout the experimental examinations.
The start-up transient of the MPP tracking under full
illumination is shown in Fig. 7. The tracking process settles in less
than 200 ms with little initial hunting and negligible oscillation
during the steady-state. In the same diagram, IIrrad is the current
waveform which supplies the nine-string halogen lamps. The IIrrad
is used as an indicator to pinpoint the point at which the
illumination changed.
Figure 7 Experimental result during MPPT operation at initial start-

up
The cloud effect was emulated by switching off the control

switches of halogen lamp strings one at each stage, which
correspond to approximately 700, 500, 350, 225 and 145W/m2 of
solar irradiance. The response of PV power under step-down solar
irradiance examination is illustrated in Fig. 8a. The tracking
response shows that the controller is able to bring the power
operating point back to a stable point quickly under each step
change of solar irradiance. The MPPT process and operation can
be better appreciated by observing the I-V and P-V characteristic
curves as presented in Fig. 8b. It is noticeable that the tracking
point moves along the locus of the MPP as the solar irradiance
changes. In each step, the MPP is accurately identified and
therefore best use is made of the PV module.
(a)
I-V Characteristic
3
PV Current (A)
0
0 5 10 15 20 25
PV Voltage (V)
P-V Characteristic
50
PV Power (W)
40
30
20
10
0
0 5 10 15 20 25
PV Voltage (V)
(b)
Figure 8 (a) Experimental result for the response under a series of

irradiance step-down test, (b) the MPP tracking process and operation
Similarly, the solar irradiance step-up examination was

performed by switching on the sequence of the control switches.
The step-up MPP tracking response is presented in Fig. 9. The PV
power settles at steady-state quickly. Moreover, the power
extracted by the converter has little ripple and no oscillation at
steady-state. The high spike of IIrrad at each step-up test is due to
the inrush current when switching the control switches. In Fig. 9b,
it can be seen that the tracking process for step-up test is slower as
compare to the step-down test due to extra hunting points.
However, it does not affect the performance of the tracking in
terms of oscillations at steady-state.
(a)
I-V Characteristic
3
PV Current (A)
0
0 5 10 15 20 25
PV Voltage (V)
P-V Characteristic
50
PV Power (W)
40
30
20
10
0
0 5 10 15 20 25
PV Voltage (V)
(b)
Figure 9 (a) Experimental result for the response under a series of

irradiance step-up test, (b) the MPP tracking process and operation
Finally, the dynamic performance test was conducted by

gradually changing the IIrrad level to imitate the slow solar
irradiance changes. The tracking performance under variable solar
irradiance changes is demonstrated in Fig. 10. The test covers the
rise and decline of solar irradiance level. The results show that the
operating point moves along the MPP locus in both I-V and P-V
characteristic curves as the illumination changes.
(a)
I-V Characteristic
3
PV Current (A)
0
0 5 10 15 20 25
PV Voltage (V)
P-V Characteristic
50
PV Power (W)
40
30
20
10
0
0 5 10 15 20 25
PV Voltage (V)
(b)
Figure 10 (a) Experimental result for the response under variable

changes of irradiance using the improved P&O algorithm, (b) the MPP
tracking process and operation
2.10 CONCLUSION
The importance BIPV design considerations including the weather

conditions and PV orientation and tilt angle have been discussed.
The discussion identifies that the yearly energy yield is strongly
correlated with the PV orientation. An improved Perturb and
Observe MPPT algorithm with a current-mode controlled DC-DC
step-up converter has been demonstrated. It is effective in allowing
a BIPV system to harvest the maximum available solar energy. An
experimental PV illumination test rig and a digital rapid
prototyping platform using DSP and FPGA incorporating
MATLAB/Simulink Real Time Workshop have been reported.
Experimental results show that for both rapid and gradual changes
of solar irradiance, the designed current-mode controlled MPPT
converter is able to track the true MPP accurately. Further, it is
capable of reaching the steady-state condition quickly during a
transient with negligible oscillations. As a conclusion, the
installation of BIPV systems have to take into consideration not
only the location, PV orientation and tilt angle but also the power
conditioners that are able to effectively track the maximum
available solar energy and make best use of PV array.
REFERENCES
[1] J. de la Casa, P. J. Perez, J. Aguilera, G. Almonacid, and J.

M. Anguita, "Maintenance and evaluation during the design
of building integrated grid-connected photovoltaic systems,"
presented at Photovoltaic Energy Conversion, 2003.
Proceedings of 3rd World Conference on, 2003.
[2] H. Yang and Y. Fung, "Building-integrated photovoltaic
application - its practice in Hong Kong," Photovoltaic
Energy Conversion, 2003. Proceedings of 3rd World
Conference on, vol. 3, pp. 2490- 2493, 2003.
[3] I. Spanos and L. Duckers, "Expected cost benefits of
building-integrated PVs in UK, through a quantitative
economic analysis of PVs in connection with buildings,
focused on UK and Greece," Renewable Energy, vol. 29, pp.

1289-303, 2004.
[4] D. Poponi, "Analysis of diffusion paths for photovoltaic
technology based on experience curves," Solar Energy, vol.
74, pp. 331-40, 2003.
[5] P. Huynh and B. H. Cho, "Design and analysis of a
microprocessor-controlled peak-power-tracking system [for
solar cell arrays]," IEEE Transactions on Aerospace and
Electronic Systems, vol. 32, pp. 182-90, 1996.
[6] X. Liu and L. A. C. Lopes, "An improved perturbation and
observation maximum power point tracking algorithm for PV
arrays," presented at Power Electronics Specialists
Conference, 2004. PESC 04. 2004 IEEE 35th Annual, 2004.
[7] T. Kasabara, J. Plastow, and S. Ward, "Added values of
photovoltaic systems," Photovoltaic Energy Conversion,
2003. Proceedings of 3rd World Conference on, vol. 3, pp.
2419- 2422, 2003.
[8] M. Posnansky, S. Gnos, and S. Coonen, "The importance of
hybrid PV-building integration," presented at Photovoltaic
Energy Conversion, 1994., Conference Record of the Twenty
Fourth ; IEEE Photovoltaic Specialists Conference - 1994,
1994 IEEE First World Conference on, 1994.
[9] A. Hunter Fanney, B. P. Dougherty, and M. W. Davis,
"Performance and characterization of building integrated
photovoltaic panels," presented at Photovoltaic Specialists
Conference, 2002. Conference Record of the Twenty-Ninth
IEEE, 2002.
[10] D. Prasad and M. Snow, "Designing with Solar Power - A
source book for building integrated photovoltaic (BiPV),"
The Images Publishing Group Pty Ltd and Earthscan, 2005.
[11] J. Jurgens, J., "Building integrated facades in New York,
using high efficient back contact solar cells," 2006, pp. 3.
[12] M. A. S. Masoum, H. Dehbonei, and E. F. Fuchs,
"Theoretical and experimental analyses of photovoltaic
systems with voltageand current-based maximum power-
point tracking," Energy Conversion, IEEE Transactions on,
vol. 17, pp. 514-522, 2002.

[13] "Photovoltaics in Buildings A Design Guide," DTI, Report
No. ETSU S/P2/00282/REP, 1999.
[14] C. W. Tan, T. C. Green, and C. A. Hernandez-Aramburo,
"An Improved Maximum Power Point Tracking Algorithm
with Current-Mode Control for Photovoltaic Applications,"
Power Electronics and Drive Systems, Proceedings, 2005 6th
IEEE International Conference, 2005.
[15] J. Van den Keybus, B. Bolsens, K. De Brabandere, J.
Driesen, and R. Belmans, "DSP and FPGA based platform
for rapid prototyping of power electronic converters and its
application to a sampled-data three-phase dual-band
hysteresis current controller," presented at Power Electronics
Specialists Conference, 2002. pesc 02. 2002 IEEE 33rd
Annual, 2002.
[16] C. Gherasim, T. Croes, J. Van den Keybus, J. Driesen, and R.
J. M. Belmans, "Development of a flickermeter for grid-
connected wind turbines using a DSP-based prototyping
system," Instrumentation and Measurement, IEEE
Transactions on, vol. 55, pp. 550-556, 2006.
3
ANALYSIS OF PERTURBATION STEP
SIZE FOR PERTURB AND OBSERVE
PHOTOVOLTAIC MAXIMUM POWER
POINT TRACKING ALGORITHM
Tan Chee Wei
Tim C. Green
Carlos A. Hernandez-Aramburo
3.1 INTRODUCTION
In recent years, climate change has created an enormous concern

to everyone. One of the main concerns is the use of fossil fuels that
have been identified as one of the main contributors to the
greenhouse effect through the production of CO2 as a waste
product. In addition to that, the fast developments of industries
have made great inroads into the non-renewable fossil fuel. The
negative environmental impacts and fossil fuels depletion remain
as great challenges to energy production. Therefore, renewable
energies have been brought into use. Among them, solar or
photovoltaic (PV) energy is one of the most promising energy
resources because it is free of fuel costs, environmental friendly
and motionless.
There are several significant experiences within European
member countries with PV system. The Campaign for Take-Off
launched by European Commission seeks to install a total capacity
of 1 GWp PV power by 2010 [1] to promote decentralised
electrification in developing countries and in the EU-wide
Analysis of Perturbation Step-Size for Perturb and Observe Photovoltaic 41
domestic market. In addition to that, the largest PV power plant in

the world, which has the plant capacity of 5 MWp, went on stream
in August 2004 in Germany [2]. Although the rapid development
of photovoltaic materials technology and the increased demands
for PV materials have led to a reduction of the PV module costs [3]
the capital costs of PV systems are still very high. Thus, there is a
necessity to design a power converter that is not only high in
efficiency but also optimizes the energy production of the PV
material.
A maximum power point tracking (MPPT) algorithm is
commonly applied in power converters to maximize the power
drawn from PV modules under varying atmospheric conditions.
This is to ensure that the best use of the PV array in producing
clean electricity is made. Among the MPPT algorithms, the Perturb
and Observe MPPT method is the popularly used due to the
simplicity of its control structure [4]. However, there have been
relatively few studies [4, 5] made on the influence of applying
perturbation step-sizes in P&O algorithm. Therefore, this chapter
aims to investigate and evaluate the effect of applying different
perturbation step-sizes on the P&O MPPT algorithm.
This chapter begins with a review of PV characteristics and
proceeds with a review of available MPPT algorithms.
Subsequently, the P&O MPPT algorithm and perturbation step-
sizes are explained. The power converters which include a direct
duty-cycle controlled boost converter and a current-mode
controlled boost converter are briefly discussed. Then, the
simulation results of the effect using different perturbation step-
sizes are described analyzed.
3.2 ELECTRICAL CHARACTERISTIC OF PV CELLS
Figure 1 shows the equivalent circuit of a PV cell. The output

current of the equivalent circuit, I, can be expressed as a function
of the cells voltage, V, [6, 7]:
I = I L I D I sh (1)
V + I RS V + I R (2)
I = I L I o e nVT 1 S
R P
where IL, the photocurrent, is a function of cell temperature and

solar irradiance.
n Diode quality factor
k Boltzmanns constant (1.381 10-23)
q Electronic charge (1.602 10-19 C)
RS Lumped series resistance ( )
RP Lumped shunt resistance ( )
Rs I
+
ID Ish
+
IL RP V
_
Figure 1 The equivalent circuit of a photovoltaic cell
The PV module characteristic depends on the solar irradiance

and the temperature of the photovoltaic module. As the solar
irradiance increases, the photocurrent increases while the PV
voltage also increases slightly, hence, the power produced by the
PV module increases. The open circuit voltage of the PV module
decreases with a rise of the PV module temperature. It is,
therefore, desirable to keep the PV module temperature low in
order to extract the maximum power from the module. The effects
of solar irradiance and temperature on the I-V and P-V
characteristics of PV module are illustrated in Fig. 2 for the
example of a BP.365U, 65-watt multi-crystalline PV module the
characteristics of which are presented in Table 1.
(a) I-V characteristic
70
25 C
60 50 C
PV Power (W)
50 Solar Irradiance
40
Increases
30
20
10 Temperature
Increases
0
0 5 10 15 20 25
PV Voltage (V)
(b) P-V characteristic

Figure 2 The effect of solar irradiance and temperature on PV I-V and
P-V characteristics.
Table 1 Electrical characteristics of the 65 W multi-crystalline

photovoltaic module BP 365U at STCe
Parameter Value
Maximum power (Pmax) 65 W
Voltage at Pmax (Vmpp) 17.6 V
Current at Pmax (Impp) 3.69 A
Short circuit current, (ISC) 3.99 A
Open-circuit voltage (VOC) 22.1 V
(0.065 0.015)
Temperature coefficient of ISC
%/C
Temperature coefficient of VOC -(80 10) mV/C
NOCT (Nominal Operating Cell
47 2 C
Temperature)
Standard Test Condition (STC): 1 kW/m2 (1 Sun) at

spectral distribution of AM 1.5 and cell temperature of
25 C.
3.3 MPPT ALGORITHMS

An MPPT algorithm sets a reference value for one of the variables
(commonly the duty-cycle but also possibly a current or a voltage)
in the power converter that interfaces the PV array to a battery or
load. The reference is chosen so that the converter draws the
current or imposes the voltage that operates the PV panel at or near
the maximum power point (MPP) for the prevailing irradiance and
temperature. MPPT techniques can be categorized into off-line and
on-line methods. Off-line MPPT techniques require prior
information about the PV array and measurements of either the
solar irradiance, the short circuit current or the open circuit voltage
of the particular PV array [8, 9]. An example is the Constant
Voltage MPPT which keeps the operating point near the maximum
power point by matching the PV voltage to a fixed reference
voltage. This method starts from the assumption that the variations
of solar irradiance and temperature cause insignificant changes to
the voltage that defines the MPP and that a constant reference
voltage is an adequate approximation of the true MPP [10]. On-
line MPPT methods are more popular than the off-line methods.
This is because these methods are able to track the MPP quickly
and do not rely on the knowledge of the PV array characteristics.
The on-line MPPT methods include Hill Climbing, Perturb and
Observe, Incremental Conductance and hybrid methods.
3.3.1 P&O and MPPT Algorithm
There have been extensive applications of the P&O MPPT

algorithm in various types of PV system [4, 11, 12]. This is
because P&O algorithm has a simple control structure and few
measured parameters are required for the power tracking.
Moreover, it has an advantage of not relying on the PV module
characteristics in the MPPT process and so can be easily applied to
any PV panel. The name of algorithm itself reveals that it operates

by periodically perturbing the control variable and comparing the
instantaneous PV output power after perturbation with that before.
The outcome of the PV power comparison together with the PV
voltage condition determines the direction of the next perturbation
that should be used.
The P&O perturbation step-size used in any MPPT technique plays
a significant role in determining the accuracy and speed with
which the operating point moves towards the MPP. For a given
perturbation interval, the larger the perturbation step the faster the
operating point can be driven to the MPP. However, the larger the
perturbation step size, the larger are the intrinsic oscillations
around the MPP in steady-state. These oscillations would reduce
the effectiveness of the PV power conversion because of the large
error and lost opportunity to generate power. A smaller
perturbation step size reduces the magnitude of oscillation around
the MPP and increases the energy conversion effectiveness once
the MPP has been achieved. Nevertheless, this would only solve
the problem at steady-state; it would lead to a slow response under
rapidly changing environmental conditions and hunting of the
operating point remains during transients. A varying step-size may
help to overcome the drawbacks exhibited by fixed step-sizes.
3.3.2 Incremental Conductance MPPT Algorithm
The Incremental Conductance MPPT algorithm was developed in

[13], and is based on the fact that at the MPP (for any irradiance
and temperature), the derivative of the PV output power with
respect to the PV voltage is zero. Thus, the PV voltage can be
regulated relative to the voltage at the MPP by measuring the
incremental conductance, dI/dV and chord conductance, I/V.
dP d ( IV ) dI
= = I +V (3)
dV dV dV
dP
=0 at the MPP,
dV
dP to the left of the MPP,

>0
dV
dP to the right of the MPP
<0 (4)
dV
The result of the computation of (3) and (4) will determine the
direction of the required change in the control variable so as to
move the PV voltage towards the MPP. This algorithm has the
advantage of no oscillation occurs around the MPP in steady-state
unlike the continuous perturbations of the P&O algorithm.
However, the drawback of this algorithm is the complexity of the
design the controller.
3.3.3 Other MPPT Techniques
A novel MPPT technique for PV panels using switching

frequency modulation was developed by H. S.-H. Chung et al.
[14]. This method has the advantages of not requiring prior
knowledge of the PV panel and avoiding sophisticated
mathematical computation. However, it requires an external signal
to perturb the system and significant care has to be taken to ensure
that the small signal introduced into the modulation process is
correct.
An algorithm for rapid tacking of the MPP in a PV system was
presented in [15]. This technique achieves rapid movement toward
the MPP but must switch to a conventional MPPT algorithm when
the operating point approaches close to the MPP.
3.4 MPPT ALGORITHMS
For the analysis of P&O MPPT algorithm, a direct duty-cycle

control (DDC) and a current-mode control (CMC) were applied to
a power converter to examine the effect of different perturbation
step-sizes. The concept of the DDC converter is demonstrated in
Fig. 3.
Figure 3 The simplified block diagram of a direct-duty cycle control

MPPT converter
The MPPT converter with current-mode control aims to achieve

fast response and accurate holding of the MPP under rapidly
changing environmental conditions over the conventional P&O
MPPT.
As was observed in Fig. 2(a), the short-circuit current of a PV
module is proportional to the solar irradiance. In [8] it was further
shown that there is a linear dependence between the current at
MPP and short circuit current,
IMPP = MC ISC (5)
where MC is known as the current factor. This suggests that it is the

current that should be the controlled variable that is perturbed.
Perturbing the duty-cycle or the voltage can cause very large
changes in operating point but perturbing the current gives a more
useful indication of the sensitivity of the power to the PV
operating point. A fast-acting local current control loop adjusts the
duty-cycle of the converter to force the module current to follow
the reference value indicated by the MPPT algorithm. The
configuration of the current-controlled boost DC-DC that was
selected together with the MPPT is shown in Fig. 4.
Figure 4 Configuration of the Current-Mode-Controlled-Boost DC-

DC MPPT converter
Therefore, an improved P&O MPPT algorithm has been

developed and the process of the algorithm is illustrated in the
flow chart as presented in Fig. 5. The perturbation step, I, is
added to the reference current at each iteration of the algorithm:
Iref(k) = Iref(k-1) + I (6)

Figure 5 Flow chart of the improved Perturb and Observe MPPT

algorithm
The size of the step is made dependent on the sensitivity of the

PV power to the previous perturbation. The sign of the step is
determined by kr, which is the opposite of the sign of the slope in
the P-V characteristic curve:
PPV _ diff
I = kr
PPV (7)
PPV _ diff = P pv ( k ) P pv ( k 1 ) (8)
where Ppv(k) is the present PV power and Ppv(k-1) is the previous

measured PV power and kr is a one of four constants for the four
possible combination of perturbation direction and Ppv/V slope
direction. The values of kr were found by trial and error. The
additional term, , is included to enable the sensitivity to be
enhanced or reduced to achieve both rapid movement toward the
MPP following a large change in characteristic and to aid accurate
convergence of the MPPT on the true MPP. If the Ppv_diff is larger
than the value of , the perturbation step size in the control variable
will be reduced by multiplying an less than unity. This avoids
overshooting the MPP when approaching the MPP from a region
where the Ppv/V curve is steep. In the algorithm, is characterized
into two ranges: 0.003 is used for for Ppv between zero and PMPP
and 0.015 for PPV up to PMPP. If the Ppv_diff is less than the control
limit then is set to unity. The perturbation cycle is repeated for
any changes of environmental conditions to maintain the PV power
at MPP.
3.4.1 The Effect of Perturbation Step-Size
For a given perturbation interval, the larger the perturbation step

the faster the PV current can be driven to the MPP. However, the
larger the perturbation step-size, the larger are the intrinsic
oscillations around the MPP in steady-state. These oscillations can
reduce the effectiveness of the PV power conversion because of
the large error and lost opportunity to generate power. A smaller
perturbation step-size reduces the magnitude of oscillation around
the MPP and increases the energy conversion effectiveness once
the MPP has been achieved. However, this would only solve the
problem at steady-state; it would also lead to slow response under
rapidly changing environmental conditions. A small perturbation
step-size reduces the error caused by oscillation around the MPP
but greater deviation from the MPP occurs under rapid changes of
atmospheric conditions because of the slow response.
Consequently, there is a trade off between fast tracking and power
error in deciding a suitable fixed size of perturbation step.
3.4.2 Variable Perturbation Step-Size
In this improved P&O MPPT algorithm, a variable perturbation

step size was adopted. The variable step size could be implemented
in several ways. Here, the perturbation step size was based on the
PV power difference Ppv_diff due to the effect of a change in solar
irradiance and temperature. Equation (7) showed that the
perturbation step size is automatically tuned according to the PV
power difference. Therefore, a large change of solar irradiance or
temperature causes a large change in the power produced by PV
module and this automatically tunes the control variable I to a
larger size to respond faster to the atmospheric changes. During
steady-state, the PV power difference is approximately zero, thus
Iref is maintained close to the previous value with very small
perturbation steps. Although the variable perturbation step-size
gives better performance than a fixed step-size, it might hunt
wildly during a transient before reaching the steady-state
condition. Since the PV power waveform is mostly governed by
the PV current under changes of solar irradiance, controlling the
inductor current can help to reduce the oscillation in PV power
during the transient state.
3.5 SIMULATION AND RESULT ANALYSIS
A series of simulations were performed using the PLECS circuit

simulator as part of MATLAB/Simulink. The PV module
simulator was implemented based on (2). An MPPT with three
different fixed step sizes and a variable step-size were simulated
with a PV module simulator. The fixed step-sizes were applied to
DDC converter and the variable step-size was implemented with
CMC converter. Each perturbation step-size case was also
simulated under a series of solar irradiance step changes for 10
seconds. The simulations were carried out with the assumption that
the ambient temperature was at 28 C. The PV module simulator
was initially illuminated at 700 Wm2. It was then stepped down to
500 Wm2 after 2 s and further reduced to 350, 225 and 145 Wm2
at 2 s intervals.
Figure 6(a) shows that the large step-size (0.008) creates large
oscillations during full illumination and the magnitude of the
oscillations is less at lower illumination. It can be seen from the
figure that the PV power oscillated around the MPP and did not
settle. Figure 6(b) shows that the magnitude of power oscillation is
significantly reduced for a step-size of 0.001 compared to the
previous case of step-size. It can be seen in Fig. 6(b) that the
magnitude of the power oscillation has reduced to about 3 W
during the steady-state condition as compared to the 0.008 step-
size (Fig. 6(a)). In addition, the transient after each step change of
illumination settles quickly. However, the tracking operation
continues to hunt at steady-state conditions. The smallest fixed
step-size investigated in this chapter is 0.0001 as presented in Fig.
6(c). Since it is a very small step-size, it took longer to reach the
desired operating point. Note that the PV power response obtained
from the experimental set-up has longer initial transient compared
to the response obtained from the simulation model. This happened
in practical test because the 0.0001 step-size used in the MPPT
algorithm is too small and, in turn, the calculated duty-cycle
reference has not much different with the previous duty-cycle.
Therefore, the calculated reference produced little impact on the
next duty-cycle and thus it requires more steps to move the
operating point to the true MPP.
The simulated response of the variable perturbation step-size
incorporating the CMC is demonstrated in Fig. 6(d). The PV
tracking process is fast and the operating point accurately holds the
true MPP after each step change of illumination.
(a) simulated response applying 0.008 fixed step-size
(b) simulated response applying 0.001 fixed step-size

(c) simulated response applying 0.0001 fixed step-size
(d) simulated response applying variable step-size
Figure 6 Response of PV power under a series of step-down

irradiance test with various fixed step-sizes [(a) (c)] and an improved
variable P&O algorithm perturbation step-size [(d)]
3.6 CONCLUSION
An improved Perturb and Observe MPPT algorithm with a current-

mode controlled DC-DC converter has been described. The effect
of the perturbation step-size on the solar power production has
been examined. The fixed step-size algorithm exhibited
compromises in the choice of step-size. Too large a step-size led to
poor MPP accuracy (especially at high illumination) and
oscillatory behaviour. This would reduce the amount of cumulative
PV energy obtained. Too small a step-size led to slow response
and reduced energy harvest in changing illumination. The best
fixed step-size may improve both the accuracy of speed and
location of the MPP but it has to be tuned based on particular
model of PV panel used. Consequently, fixed-step MPPT requires
a compromise in the choice of step-size between tracking speed
and accuracy. Cloud and object shading mean that speed can be an
issue. Variable step-size is seen to overcome this compromise.
However, variable step-size should be treated carefully in order to
avoid MPPT operating point from hunting around the true MPP
during transients.
REFERENCES
[1] "White Paper for a Community Strategy and Action Plan.

Energy For The Future: Renewable Sources Of Energy,"
Communication from the Commission, 26 November 1997.
[2] "Sun & Wind Energy," Special International, pp. 16-20,
2/2004.
[3] D. Poponi, "Analysis of diffusion paths for photovoltaic
technology based on experience curves," Solar Energy, vol.
74, pp. 331-40, 2003.
[4] X. Liu and L. A. C. Lopes, "An improved perturbation and
observation maximum power point tracking algorithm for PV
arrays," presented at Power Electronics Specialists
Conference, 2004. PESC 04. 2004 IEEE 35th Annual, 2004.
[5] N. Femia, G. Petrone, G. Spagnuolo, and M. Vitelli,
"Optimization of Perturb and Observe Maximum Power
Point Tracking Method," Power Electronics, IEEE
Transactions on, vol. 20, pp. 963-973, 2005.
[6] M. A. de Blas, J. L. Torres, E. Prieto, and A. Garcia,
"Selecting a suitable model for characterizing photovoltaic
devices," Renewable Energy, vol. 25, pp. 371-80, 2002.
[7] G. Walker, "Evaluating MPPT Converter Topologies Using a
Matlab PV Model," Journal of Electrical & Electronic
Engineering, Australia, vol. 21, pp. 49-55, 2001.
[8] M. A. S. Masoum, H. Dehbonei, and E. F. Fuchs,
"Theoretical and experimental analyses of photovoltaic
systems with voltageand current-based maximum power-
point tracking," Energy Conversion, IEEE Transactions on,
vol. 17, pp. 514-522, 2002.
[9] M. A. S. Masoum, S. M. M. Badejani, and E. F. Fuchs,
"Microprocessor-controlled new class of optimal battery
chargers for photovoltaic applications," Energy Conversion,
IEEE Transactions on, vol. 19, pp. 599-606, 2004.
[10] G. J. Yu, Y. S. Jung, J. Y. Choi, I. Choy, J. H. Song, and G.
S. Kim, "A novel two-mode MPPT control algorithm based
on comparative study of existing algorithms," presented at

Conference Record of the Twenty-Ninth IEEE Photovoltaic
Specialists Conference 2002, 19-24 May 2002, New Orleans,
LA, USA, 2002.
[11] C. Hua, J. Lin, and C. Shen, "Implementation of a DSP-
controlled photovoltaic system with peak power tracking,"
Industrial Electronics, IEEE Transactions on, vol. 45, pp.
99-107, 1998.
[12] P. Huynh and B. H. Cho, "Design and analysis of a
microprocessor-controlled peak-power-tracking system [for
solar cell arrays]," IEEE Transactions on Aerospace and
Electronic Systems, vol. 32, pp. 182-90, 1996.
[13] K. H. Hussein, I. Muta, T. Hoshino, and M. Osakada,
"Maximum photovoltaic power tracking: an algorithm for
rapidly changing atmospheric conditions," Generation,
Transmission and Distribution, IEE Proceedings-, vol. 142,
pp. 59-64, 1995.
[14] H. S.-H. Chung, K. K. Tse, S. Y. R. Hui, C. M. Mok, and M.
T. Ho, "A novel maximum power point tracking technique
for solar panels using a SEPIC or Cuk converter," Power
Electronics, IEEE Transactions on, vol. 18, pp. 717-724,
2003.
[15] S. Jain and V. Agarwal, "A New Algorithm for Rapid
Tracking of Approximate Maximum Power Point in
Photovoltaic Systems," Power Electronics Letters, IEEE, vol.
2, pp. 16-19, 2004.
4
LOAD SHARING CHARACTERISTIC OF
SINGLE PHASE PV INVERTER
CONNECTED TO THREE PHASE GRID
M. Imran Hamid
Makbul Anwari
Taufik
4.1 INTRODUCTION
Application of photovoltaic as electrical energy source shows

increasing trend both in implementation on spread area over the
world and in capacity of plant. This trend is e triggered by many
factors such as the increasing of fossil fuel cost and declination of
production cost per kW electric from photovoltaic and also
technology development that cause the photovoltaic power
conversion more efficient [1].
Photovoltaic generation system can either be operated in

isolated system or be connected to the grid to form integrated
system, and with other electrical renewable energy source can
form distributed renewable energy generation. In an integrated
photovoltaic generation, one among aspects that take attention of
researches is the power flow and load sharing problem between the
grid and photovoltaic plant in order to supply electrical power for
connected load, it is intended that power flow mechanism ensure
that energy generated by the plant can be sent to the grid
optimally. In such a system, load flow and load sharing mechanism
are handled by the inverter as power interface between plant and
Load Sharing Characteristic Of Single Phase PV Inverter 59
grid, by using appropriate method, intended power flow and load

sharing control can be achieved.
Other aspect concerning to application of photovoltaic as a part

of distributed generation is the power quality resulted from their
operation, especially for voltage unbalance and harmonics. Trend
application of some single phase pv inverters and its pv array
connected together to supply three phase system as alteration of
high capacity centralized three phase pv inverter can be a factor
that effect to unbalance grid voltage due to diversity of irradiance
among array [2,3]. Harmonics distortion as a result of interaction
between pv inverter and other part of the grid components is also
one aspect that continuing explored [4].
This paper describes model and simulation of several single phase

PV inverters that work independently to supply three phase
system. Model and simulation of load flow and load sharing
control between pv generations will be presented. The current of
PWM voltage source PV inverter is controlled by current control
strategy (CCVSI-PWM) using grid parameter as current reference
[5] and then implement to ramp comparison and hysteresis control
method. Further, effect of irradiance level and current control
performance to load flow and load sharing will be analyzed.
Power quality aspect due to unbalance supply between single
phase pv inverter and grid power quality behavior is also simulated
and evaluated. Model and simulation are implemented using
Matlab/Simulink.
4.2 PV MODUL-INVERTER CONFIGURATION
In pv generation system, pv inverter holds the role as interface

between photovoltaic module and ac power grid. In this function,
pv inverter and associated generation system equipment should
have ability to maximize power extracting from the array, matches
DC voltage output from pv array and produces sinusoidal ac
voltage with minimum distortion in output side and controls the
power flow. If the pv inverter is from grid tie type, it is must be

completed with synchronizing mechanism and protection from
islanding condition or from others negative impact of grid
phenomena. In certain application, there is also need for data
communication and monitoring, nowadays many commercial pv
inverter have completed by advance data communication and
monitoring system facilities.
From overall basic abilities listed above, pv inverter is hoped to

operate in optimum conversion efficiency. Photovoltaic material
exploration, converter topologies and its control mechanism have
and being continuously developed. Development of power
conversion affectivity is also can be reached by optimizing
configuration between pv module/array and associated pv inverter.
In application, according to connection and configuration between
pv module and pv inverter, photovoltaic generation can be
identified as:
Photovoltaic generation using central pv inverter, this

configuration is reached by paralleling some pv module strings
(some pv modules in serial connection) and connected to dc side of
a relatively large capacity pv inverter. High conversion efficiency
can be reached from this configuration but faces weakness if there
are diversity of pv module and diversity of irradiance or partial
shading. This configuration is also susceptible to reliability
problem because all pv generation depend on only one equipment.
This configuration is often implemented on high capacity
generation plant to optimize the cost, but unfortunately this is also
a drawback because it is limiting the flexibility to adjust plant
capacity.
Photovoltaic generation using string pv inverter, to overcome the

weakness of central inverter configuration, the string pv inverter
configuration was introduced. In this topology, each pv module
string is connected to its pv inverter with own MPPT respectively.
By this way, optimum power extracting from each string can be
achieved, overall efficiency is better and pv generation reliability

is increased because the system is not depend to only one
equipment anymore. Plant capacity is also easy to adjust to follow
demand dynamic.
Photovoltaic generation using multi-string pv inverter, this

configuration is developed to adopt the advantageous (cost and
technical) of both previous configurations. Some pv module
strings with dc-dc converter and their own MPPT mechanism are
connected to a pv inverter. Optimum power extracting from each
module string can be achieved and in other side reduced cost for
dc-ac converter can be done.
Photovoltaic generation using team system, even though optimum

power extraction can be achieved using multi-string pv inverter
configuration, but implementation of a single dc-ac converter in
second stage of electrical conversion in other hand decrease the
reliability of generation system. Team system configuration was
introduced to solve this problem, this configuration contains
several pv module strings and its own pv inverters set in parallel.
When irradiance high enough, each pv module strings and its own
pv inverters work independently to supply power for ac side, on
the contrary, if the irradiance go down, module string
configuration is changed in such a way so that not all of dc-ac
converter operate. This scheme ensures that dc-ac converter
always operates around their rated power.
The above configuration of pv generation systems is described

on Fig.1. String pv inverter, influenced by cost consideration, plant
capacity, and flexibility to enlarge the plant have shown
widely implemented and have evolved to become standard pv
system technology for grid connected pv generation plant [6].
String pv inverter for capacity under 5 kW generally produced in
single phase. To form 3 phase supply to the grid it is done by
connecting at least three inverters, by this system, flexibility to
adjust plant capacity is easier.
Central Inverter String Inverter Multi-string Inverter Team concept

Configuration Configuration Configuration Configuration
dc dc dc dc dc dc dc dc dc dc
ac ac ac ac dc dc dc ac ac ac
dc
ac
Grid Grid Grid Grid
Figure 1 PV module configurations in a PV plant [3,6]
4.3 LOAD SHARING METHOD OF PV INVERTER
Load flow control in pv generation is done by power converter

using appropriate load flow control method. In this case, current
control and voltage control method are widely used. Voltage
control method uses error of grid voltage and distributed
generation terminal voltage at both side of coupling inductor as
control signal for the controller, meanwhile current control method
uses error between actual current sent and desired (setting) current
as control signal [5].
Current control method has some advantageous compared to

voltage control method that cause this method is preferred to use in
many converter application [7]. In current control method, to
determine switching instant, two, out of many strategies can be
used: hysteresis current control and ramp time comparison. Here
the two methods describe as follows [8]:
Hysteresis current control

In this method, inverter output current is forced to follow the
current reference in each phase. Deviation between these two
quantities is limited by upper and lower band in a hysteresis loop.

If actual current reach upper limit of hysteresis band, the inverter
leg is switched off so that the current decrease till reach lower
band of hysteresis loop where the inverter leg is switch on again
and actual current increase to upper band. The sinusoidal-wave
shape of reference signal causes inverter switching frequency vary
and gives different current ripple in one fundamental inverter
period. Gap width between upper and lower band of hysteresis
loop determine the magnitude of current ripple. The hysteresis
control method and inverter current shape is shown in Fig. 2.
30
urrent (A)
Upper Band
C
20
10
Lower Band
Inverter Current
-10
-20
-30
0.04 0.042 0.044 0.046 0.048 0.05 0.052 0.054 0.056 0.058 0.06
Time (s)
Figure 2 Hysteresis controller
Ramp comparison current control

In this mode of controller, a sinusoidal-wave signal is added to a triangle
signal for creating a sinusoidal-triangle reference. This reference then compare
with actual current. The point where sinusoidal - triangle wave and actual
current crosses become time when the inverter leg is switched. If the current
error greater than sinusoidal-triangle, the inverter leg is switch off and in
contrary if the current error less than sinusoidal-triangle then the inverter leg is
switch on. As the hysteresis methods, this method also causes the inverter
switching frequency and PWM pulse width vary and give different current
ripple in one fundamental inverter period. The Ramp comparison current control
method and inverter current shape created are shown in Fig.3.
40
Current (A)
30
Sinusoidal triangle
reference
20
10 Inverter current
-10
-20
-30
-40
0.02 0.022 0.024 0.026 0.028 0.03 0.032 0.034 0.036 0.038 0.04
Time (s)
Figure 3 Ramp comparison controller
4.4 LOAD FLOW AND LOAD SHARING

CHARACTERISTIC OF PV INVERTER
Fig. 4 shows simplified simulation circuit diagram to show

load flow characteristic of several single phase inverter connected
to grid. Photovoltaic dc power is generated from pv array with
input variable are temperature and irradiance, pv inverter then
convert dc to ac voltage with frequency and phase according to
detected grid voltage using phase locked loop (PLL) scheme [9].
Power flow to grid is controlled by current controller based on set
up current reference, two current control methods described
previously are implemented.
aA
A
N B bB
C
cC
Three Phase Grid
Iref
Plant A
Iref
Current Control Product2 PLL

T(oC)
Sinus +
50 Out1 In1 v
Gate -
Ga In3
800
i
Ga g + Breaker3
+ - A A
T
A 2 1
DC Out ( + )
-
DC Out ( - ) B B B
PV Array A Inverter Transformer

Filter
Iref1
Iref
Plant B Current Control1 Product1 PLL1

T(oC)1
Sinus +
50 Out1 In1 v
Gate -
Ga1 In3
800
Ga g i Breaker1
+ +
- A A
T
2 1
A
DC Out ( + )
DC Out ( - ) -
B B B
PV Array B
LCL Filter1
Iref2
Iref
Plant C Current Control2 Product3 PLL2

T(oC)2
Sinus +
50 Out1 In1 v
Gate -
Ga2 In3
800
Ga g i Breaker2
+
+ - A A
T
A
A 2 1
DC Out ( + )
DC Out ( - ) -
B B B
PV Array C
LCL Filter2
Load
Figure 4 Three single phase inverters are connected to grid
Fig.5 and 6 shows application of both current control method

described above that use to control pv inverter current. For 10
Amp setting current sent to the grid, hysteresis control cause
distortion to the line parameter of 3.35 % THD for voltage
distortion and 9.58 % THD for current distortion, more ripple
appear both on line voltage and current during pv inverters are
switch on. In the other hand, ramp comparison control method
gives less distortion both line voltage and current. For the same
sent current set up, this control method gives 0.44 THD for voltage
distortion and 1.25 % THD for current distortion. Higher
distortions appear on inverter switching event.
Figure 5 Grid voltage and current when PV inverters are controlled

using hysteresis control method
Figure 6 Grid voltage and current when PV inverters are controlled

using hysteresis control method
The current drawn by connected load is supplied from both grid

and pv plant. It is different with paralleling two sources such as
transformers or generators in supplying a connected load, where
capacity and internal impedance are parameter that determine load
sharing between both equipments, load current drawn by pv
inverter in a grid is not depend on capacity of pv plant. Optimum
power can be sent to grid as long as there are sufficient power
generated by the plant. Fig. 7 shows simulation result on how a
certain load current is supplied from both grid and pv inverter.
Initially, whole load drawn by constant load is supplied from the
grid, at point when pv plant switch on, pv plant current increase
and the current from grid decrease and then go to steady condition
till the pv plant switch off. A switching surge appears as distortion
during switch on.
25
Grid Current PV Current Load Current

20
15
10
Current ( Amp )
-5
-10
-15
-20
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
Time (s)
Figure 7 Load current sharing between grid and plant on phase a

when plant switched
Grid parameter condition when a dynamic load such as electric

motor is operated is shown in Fig. 8. Initially in time range t1, a
resistive load has connected and supplied fully by the grid. At the
beginning of t2, single phase pv inverter is switch on so that load
current supplied from the grid and pv plant. Further, at beginning
of t3 the motor is switch on and draw starting current. It is shown
that cause of starting condition, grid voltage experiences voltage
drop. Because PLL in pv inverter operates based on grid voltage,
which reference current for current control follows the grid voltage
shape, so that during motor starting, the current created by pv
inverter also imitate the grid voltage shape. In this condition, high
current for starting is compensated by grid. More current flows
from the grid.
In a pv plant that use a number of single phase pv inverter,

unbalance supply between phase is un avoided. Unbalance supply
is resulted from diversification of irradiance between pv array or
partial shading in a part of pv array. Variation of irradiance
causes the operation point of pv inverters MPPT shift to different
level of current output on v-i characteristic of the array. In load
flow control mechanism of pv inverter, variation of irradiance is

responded by adjust the current reference of the current control.
Cause of each pv inverter works independently, then current level
sent by each pv inverter to three phase grid are different.
1500
S p e e d (R p m )
1000
500
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

Time (Sec)
400
200
V o lt a g e (V o lt )
-200
-400
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Time (sec)
80
Current (Am p)
60
IL=Ipv +Igr
Ipv Igr
40
20
-20
-40
-60
-80
t1 t2 t3 t4
-100
-120
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
Figure 8 Load current sharing between grid and plant on phase a

when plant switched
Simulation result base on software of several single phase PV

inverters that work independently to supply three phase system
was presented. Load sharing characteristic between pv inverters
and grid using two current control method show that the ramp time
comparison control gives lower power distortion than hysteresis
current control. Simulation also shown that in synchronizing
connecting pv inverter to grid, application of grid voltage as PLL
reference has a weakness when the grid voltage distorted as shown
when the grid operates with dynamic and non linear loads.
4.5 CONCLUSION
Simulation result base on software of several single phase PV

inverters that work independently to supply three phase system
was presented. Load sharing characteristic between pv inverters
and grid using two current control method show that the ramp time
comparison control gives lower power distortion than hysteresis
current control. Simulation also shown that in synchronizing
connecting pv inverter to grid, application of grid voltage as PLL
reference has a weakness when the grid voltage distorted as shown
when the grid operates with dynamic and non linear loads.
REFERENCES
[1] Trends in Photovoltaic Applications. Survey Report of

Selected IEA Countries Between 1992 and 2006. International
Energy Agency Photovoltaic Power Systems, IEA-PVPS-T1-
16:2007. [Online]. Available: www.iea-pvps.org
[2] P.A.B. James, A.S. Bahaj, R.M. Braid, PV array <5 kWp +
single inverter = grid connected PV system: Are multiple
inverter alternatives economic?, Solar Energy 80 (2006)
11791188
[3] M. Calais, J.M.A. Myrzik, T. Spooner, V.G. Agelidis,
Inverters for Single-Phase Grid connected Photovoltaic - An
overview, Power Electronics Specialists Conference, vol 2,

page 1173-1178, 18-23 June 2001, Bologna Italy.
[4] J.H.R. Enslin, Peter J. M. Heskes, Harmonic Interaction
Between a Large Number of Distributed Power Inverters and
the Distribution Network, IEEE Trans. On Power
Electronics, vol. 19, no. 6, November 2004.
[5] S. H. Ko, S. R. Lee, and H. Dehbonei, Application of
Voltage- and Current-Controlled Voltage Source Inverters for
Distributed Generation System, IEEE Trans. Energy
Conversion, vol. 21, no. 3, pp. 782-792, September 2006.
[6] F. Blaabjerg, R. Teodorescu, Z. Chen and M. Liserre, Power
converter and control of renewable energy systems
[7] L. Borle, M. Dymond, and C.V. Nayar, Development and
testing of a 20 kW grid interactive photovoltaic power
conditioning system in Western Australia, 1996.
[8] D. M. Brod and D. W. Novotny, Current Control of VSI-
PWM Inverters, IEEE Trans. on Industry App. vol. IA-21.
no. 4, May/Jun 1985.
[9] Ciobotaru, M., Teodorescu, R., and Blaabjerg, F., "A New
Single-Phase PLL Structure Based on Second Order Generalized
Integrator," Power Electronics Specialists Conference, 2006.
PESC '06. 37th IEEE , vol., no., pp. 1-6, 18-22 June 2006.
Power Quality Behavior of Single Phase Fed Adjustable Speed Drive 71
5
POWER QUALITY BEHAVIOR OF
SINGLE PHASE FED ADJUSTABLE
SPEED DRIVE SUPPLIED FROM
GRID OF PV GENERATION
Makbul Anwari
M. Imran Hamid
Taufik
5.1 INTRODUCTION
The increasing of energy consumption, electricity demand and

environmental issue have pushed more effort to produce more
electricity. The renewable energy source such as wind, solar, fuel
cells, small hydro and wave energy tend to be exploited
significantly [1]. Solar energy that converted via photovoltaic
cells conversion is one among promising renewable energy source
and with wind energy become dominant renewable source being
exploited [2-3]. Simplicity for build in any place, time for
realization and flexibility for built in wide range capacity cause
this technology more and more used.
PV generation system can be operated in grid connected mode

and with other generation system, can perform distributed
generation system. In isolated area, where electrical consumers can
not reached by the utility grid, PV generation are built and
operated in isolated mode [4]. The problem facing by application
of PV generation and other renewable energy source generation is
quality of electrical power quantity. Power density of renewable

energy resource (i.e. irradiance level and temperature in PV
conversion) that vary over the time potentially cause voltage and
frequency variation or sag/swell pattern on the grid. Also,
application of power converter as interface between energy source
and grid and their interaction with other system components may
cause high harmonics distortion [5].
Beside from source side, power quality problem can also be

created from load side, power quality behaviors in this case are
combination from both side of the system. Trend application of
electrical equipment based on power electronic devices and mode
operation of equipment can cause significantly power quality
problem in the grid [6], their connection to the grid both single or
three phase are also one of factors that affects power quality
parameter on the grid. Adjustable Speed Drive (ASD) is one
among such equipment. Along with fast growing application and
demand of ASDs that have proven has capability in energy
consumed reduction and ability to fulfill torque and speed pattern
needed by motor load, it causes increasing production of various
type of ASDs (topology and control method) [7]. This various
production offer many features in operation and installation to the
main.
This paper presents study of power quality behavior in

operation three phase induction motor and its ASD fed from grid
of PV generation trough single phase connection. Such connection
is found on residential connection or home industrial that supplied
from the main trough single phase line. The study is focused on
observation of line harmonics on supplying phase created during
steady state operation of ASDS and their effect on other phase of
the grid. Power quality behavior of system component during
power quality event when motor is starting and
accelerating/decelerating are observed and analyzed. The study is
done by modeling and simulation the system using
Matlab/Simulink.
5.2 SYSTEM DESCRIPTION
Photovoltaic array and pv inverter

In a photovoltaic generation, sun energy in form of irradiance
and is convert to DC electrical energy trough cells in photovoltaic
module or array. As other renewable energy sources, the power
from the sun that indicates by irradiation and temperature are time
varying. In other hand, photovoltaic cells as conversion equipment
shows non linearity in electrical parameters: current and voltage,
due to irradiation and temperature. Photovoltaic cell characteristic
shows that the higher the irradiation, the higher current output
created and the higher temperature the lower voltage developed.
This characteristic follows the well known voltage-current
relationship of photovoltaic cells one diode model as [8]:
(1)
Where ISC is short circuit current, Io is dark saturation current, q is

electric charge = 1.6022x10-19 C, k = 1.3807x10-23 JK-1 is
Boltzmanns constant, Ao is ideality factor of photovoltaic cell and
T is the absolute temperature. Rs and Rsh are photovoltaic cell serial
and shunt resistance respectively. The V-I photovoltaic cells
characteristic dependence due temperature and irradiance is
illustrated on Fig. 1.
Power electronics converter, in this case, pv inverter is used to

interface the varying power density generated by photovoltaic to
utility level of electrical power. In a pv inverter, some function
are employed: adjusting the voltage level of photovoltaic output
to meet voltage operation of inverter; tracking the voltage
/current point where maximum power can be extracted and
creating sinusoidal ac power. If the pv inverter is used as grid tied
pv inverter, it must be completed by synchronizing and power

flow control mechanism when the inverter to be connected to the
grid. Reliable protection, such as anti islanding to protect the
inverter from over load must be added. Also, cause of it is
considered as power source for utility, grid tied inverter must be
fulfil tight standard of power quality like IEEE 1547 [9].
In the market, pv inverter are met in three phase and single

phase types, in both isolated and grid tied connection. Three
phase pv inverters is produced in relatively high capacity and used
for large application such central inverter in a pv generation plant.
Meanwhile single phase inverters that produced generally as string
inverter in lower capacity is used in small application, such as in
building photovoltaic system or in individually residential
photovoltaic electricity. In a pv generation plant using single phase
inverter, enlarging capacity is done by implementing a number of
pv inverters and connecting them to form three phase connection.
Figure 1 V-I characteristic of PV module
As the point of power flow control, power quality aspects of pv

generation are highly depend on performance of pv inverter. The
problem arise from connecting pv inverter as power source to grid
can be in form of voltage and current distortion caused by
harmonics produced, voltage variation in form of over or under

voltage, unbalance voltage between phase and degradation of
power factor.
Adjustable speed drives

An adjustable speed drive (ASD) is an electrical device that
used to provide continuous range process speed control of
electrical motor. Other term concerning to this equipment are
variable speed drives, adjustable frequency drives or variable
frequency inverters. The latter two term only refers to a certain ac
system.
Generic ASD contains rectifier and inverter. The rectifier is a
circuit used to converts fixed ac voltage power from the line to
either fixed or adjustable voltage dc. The dc power than convert to
variable ac voltage and frequency by the inverter. Conversion
process in both rectifier and inverter accomplished by electronic
switch that work under control schema so that the desired dc
voltage level, ac voltage and frequency can be reached. Three
basic types of inverter widely used in ASDs application are
voltage source inverter, current source inverter and the most
commonly used due to power quality: pulse-width modulated
inverter.
According to input voltage, in market, for three phase motor

ASD is met also in single phase and three phase input. Three
phase input ASD generally used in industrial application
meanwhile single phase input ASD met in other relatively small
industry and household application (refrigeration, washing
machines, etc) which their electricity supplied from utility using
single phase connection.
Even though in capacity, single phase ASD individually seen

not significant, but because its application spread in large number
of user, then their operation may cause significant effect to power
system behavior especially in power quality. Single phase input
mean that the ASD employ single phase rectifier which known
creates rich triplen harmonics current. Magnification of this

current resulting from algebraic superposition between phases as
nature of this type of harmonics current, may causes over heating
in neutral line conductor and shifting the neutral point. Excessive
unbalance loading is also one among aspects that may appear in
application of single phase load. Unbalance loading in three phase
system may further cause unbalance voltage between phase in
three phase system, one phase may experiences over voltage and
the other experiences under voltage condition.
As generally drive characteristic, dynamic operation such as

starting, acceleration and deceleration of single phase fed ASD is
also give effect in form of voltage sag and swell on the grid.
5.3 SIMULATION
In order to observe the power quality behavior of connecting

single phase fed ASDs, a simulation was built and run with
Matlab/Simulink. A pv generation plant contains pv array and
three phase pv inverter supply the single phase fed ASD and three
phase resistive load. A delta-wye transformer is used as
distribution transformer, secondary neutral connection is set up to
allow parameter measurement. Varying input dc power for pv
inverter is done by adjusting irradiance of pv array. Circuit
breaker are employed in each single phase ASD in order to observe
grid voltage and current during connection and disconnection. The
ASD is operated in several speed level by adjusting the ac side
ASDs inverter frequency, the motor load is from a type of load
that its torque is function of square of speed. The simplified
simulation diagram is shown in Fig. 2. Using the simulation
diagram, some power quality behavior during ASD operation is
reviewed and analyzed, they are:
Ta
Ga Gate
Ga
g
A
+ aA
a
T A
A A A
b bB
B
DC Out ( + ) B B B B
- c
cC C
C C C
DC Out ( - )
C
n
Grid Load
node 10
Solar Array PV Inverter Filter Transformer
B
A
+
-
C
Speed Control
-
A
C
g
Load
C
A
Tm
Induction Motor
Figure 2 Simulation diagram
Connecting the single phase ASD to the grid.

Initially the pv generation runs in normal condition supplying a
resistive load. At time t = 1, the single phase fed ASD is
connected, the motor start till reach its nominal speed. Motor and
grid parameter are shown on Fig. 3.
During starting, the ASD input current change rapidly for
supplying starting current of the motor until reach its nominal
speed. Current shape shows typical high distortion single phase
rectifier waveform used in ASD. In the same time, grid phase
voltage where single phase fed ASD is connected experiences
voltage drop meanwhile other phase shown over voltage resulting
unbalance voltage between phase. This condition may affect the
surrounding customers connected to same coupling point, three
phase customers equipment will experience an unbalance supply
on their terminal as well as single phase customers load will
supplied by under or over voltage.
Motor Speed
1500
Rpm
1000
500
1 1.2 1.4 1.6 1.8 2
ASD input Current

30
Current (A)
20
10
-10
-20
-30
0.9 1 1.1 1.2 1.3 1.4 1.5 1.6
Time (sec)
Grid Voltage During ASD Starting
2000
1000
Voltage (Volt)
-1000
-2000
0.95 1 1.05 1.1 1.15 1.2
Grid Current During ASD Startin
20
10
Current (A)
-10
-20
0.95 1 1.05 1.1 1.15 1.2
Figure 3 ASD input and grid parameter during ASD operation
Fig. 4 shows appearance of neutral current during operation of

ASD, this figure is taken by connecting three single phase fed
ASDs on each grid phase in the same capacity so that the grid is
experienced balance load. Cause of triplen harmonics created by
each ASD, even in balance loading, the neutral current still
appears. Neutral current become more important in a grid with
large number single phase load in unbalance connection and the
loads are from non linear type such as rectifier.

0.1
0.05
Current (Amp)
-0.05
-0.1
0.95 1 1.05 1.1 1.15 1.2
Time (Sec)
Figure 4 Neutral current
Loading effect of ASD to THD

Speed variation of ASD is done by varying operation frequency
of input voltage to the motor from ASDs inverter. In this
simulation, frequency is taken in several frequency, total
harmonics distortion (THD) on grid voltage during ASD frequency
50 Hz is shown in Fig.5. Trend of voltage distortion in some
frequency and loading level of ASD are shown in Fig. 6.
Figure 5 Grid distortion during single phase ASD operation

25
THD V
20 THD I
THD (% of Fund)
15
10
0
0.5 0.8 1 1.2 1.5
Speed (pu)
30
THD V
25
THD I
THD (% of Fund)
20
15
10
0
0.25 0.5 0.75 1 1.25
Loa d (pu)
Figure 6 Trend of voltage distortion in varying operation speed and

loading level of ASD
Dependence of power output of pv array to irradiance condition

are simulated and the result are shown in Fig. 7. Three irradiance
level are applied with initially 800 W.m-2 and than at certain time it
is changed to 500 and 900 W.m-2, during these condition, ASD is
switch on and effect to ASD is observed. Fig.8 shows input
voltage and speed of ASD at the events.
800
750
DC Voltage (Volt)
700
650
600
G= 800 W.m-2 G= 500 W.m-2 G= 900 W.m-2
550
500
0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (Sec)
80
70
60
DC Current (Amp)
50
40
30
20
10
0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Time (Sec)
Figure 7 DC output of pv array when irradiance is adjusted
2000
1000
V oltage (V olt)
-1000
-2000
0.5 1 1.5 2 2.5 3 3.5
1500
Speed (Rpm )
1000
500
0
0.5 1 1.5 2 2.5 3 3.5
Time (Sec)
Figure 8 ASD Input voltage and motor speed when irradiance is

adjusted as Fig. 7.
1.5 CONCLUSION
Application of single phase fed ASDs leads to severe power

quality problems such as high contents of triplen harmonics
especially harmonics and voltage dip during dynamic operation of
ASD. Also, it was shown effect of varying speed and loading level
of ASD to harmonics content fed to the grid and effect of variation
input power to pv array that cause parameter changing both in grid
and ASD performance.
REFERENCES
[1] IEA Report 2007, Trend In Photovoltaic Application

[2] M. Arrouf, S. Ghabrourb, Modelling and Simulation of a
Pumping System Fed by Photovoltaic Generator within the
Matlab/Simulink Programming Environment, Desalination
209 (2007) 2330
[3] F. Blaabjerg, Z. Chen and S.B. Kjaer, Power Electronics as
Efficient Interface in Dispersed Power Generation Systems,
IEEE Trans. on Power Electronics Vol. 19, No.5 September
2004
[4] J.S.G. Ehnberg and M.H.J. Bollen, Generation Reliability for
Small Isolated Power Systems entirely based on Renewable
Sources, Power Engineering Society General meeting, 2004.
(IEEE 2004), pp. 2322 2327 Vol.2
[5] J.H.R. Enslin, P. J. M. Heskes, Harmonic Interaction
Between a Large Number of Distributed Power Inverters and
the Distribution Network, IEEE Trans. On Power
Electronics, Vol. 19, No. 6, November 2004.
[6] D.0. Koval, J. Leonard, and Z.J. Licsko, Power Quality of
Small Rural Industries, IEEE Trans. on Ind. App., Vol. 29,
No. 4, July August 1993
[7] J.J. Cathey, A MATLAB-Based Graphical Technique for
ASD Study, IEEE Trans on Ed. Vol. 45 No.2 May 2002
[8] A. Luque, Handbook of photovoltaic science and engineering,
John Wiley and Sons, 2003.
[9] IEEE Standard for Interconnecting Distributed Resources

With Electric Power Systems, IEEE Std. 1547, 2003.
[10] P. Caramia, A. Russo and R. Carbone, Attenuation of
Harmonic Pollution due to the Adjustable Speed Drives in the
Electric Circuits of the Power Plant Auxiliary Services, 0-7
803- 59 35 - 6/00 (0c) 2000 IEEE.-
[11] M. Grtzbach and R. Redmann, Line Current Harmonics of
VSI-Fed Adjustable-Speed Drives, IEEE Trans. On Ind. app
Vol. 36, No. 2, March/April 2000
[12] B.K. Bose, "Modern Power Electronics and AC Drives,"
Prentice-Hall, New Jersey, 2002.
[13] N. Mohan, T.M. Undeland, W.P. Robbins, Power
Electronics, Converters, Applications and Design, John
Wiley and Sons Inc. 1995.
6
DEVELOPMENT OF A 5 kW INVERTER
FOR AC POWER SUPPLY
APPLICATIONS
Awang Jusoh
Naziha Ahamd Azli
Zainal Salam
6.1 INTRODUCTION
Inverter is an electrical device that converts DC input power

(from alternate or renewable energy outputs such as PV arrays and
fuel-cells) into sinusoidal output AC power supply (used by the
consumers). To harness the renewable energy, energy conversion
technologies are required. Besides, the conversion process is vital
especially when the power source is different to the load power
requirement, for example in an AC motor drive system, having
supplied from the PV arrays. Cost-reduced and simple locally
designed and manufactured inverter is a promising proposition for
various AC power supply applications. With the depletion of the
fossil fuel and its unabated price hike, the use of inverter is likely
to grow steadily. Besides, the recent trend in inverter system not
only emphasize on lower cost but also on its ability to operate with
higher reliability, flexibility and efficiency.
A simple inverter circuit may consist of a bridge circuit
(power switches), a low-pass filter and a low frequency
transformer that normally operates in a unidirectional power
Development of a 5 kW Inverter for AC Power Supply Applications 85
flow. Where as, in much advanced inverter topology where the

inverter size much smaller and lighter, such as in the DC-DC
converter type high frequency link and the cyclo-converter
type high frequency link, the flow of power is bidirectional. By
applying proper switching signals to the power devices (normally
signals generated by a controller with certain control algorithm),
an AC signal is generated across the bridge circuit output.
Applying good switching technique such as Sinusoidal Pulse
Width Modulation (SPWM) will improve the quality of the
inverter AC output voltage. However the output voltage signal is
normally distorted and consists of harmonics. The low-pass filter is
often used to smoothen the inverter output voltage by reducing the
total harmonics distortion (THD). The size and complexity of the
filter mostly depends on the wave shape generated by the
inverter, either the quasi-square, SPWM or stepped waveform,
depending on the technology of modulation used [1]. On the other
hand, the use of a transformer is vital in an inverter system. The
function of the transformer is first to step up the output voltage to
the required mains voltage and second is to provide electrical
isolation for the inverter. The transformer itself can be of two
types, which is the bulky and big size 50 Hz line frequency
transformer or the reduced in size and weight high frequency
transformer [2]. In an Uninterruptible Power Supply (UPS), the
use of a transformer is mandatory to step up to the required
mains voltage since the UPS is normally supplied by a low battery
backup voltage.
This topic basically presents the hardware development of a 5
kW inverter circuit for AC power supply applications. In general,
it uses the Pulse Width Modulation (PWM) technology on a
standard H-bridge. The inverter topology which is used to produce
a stable low voltage 240 VAC incorporated with a low pass filter
and a 50 Hz isolation transformer is described. Detail explanation
on the construction of the gate drive circuit is given. A simple
design is proposed for the purpose of increasing reliability and
limiting cost. The algorithm chosen in controlling the inverter
output is also elaborated. The success of the developed 5 kW

inverter in operating at normal as well as transient conditions is
demonstrated and verified by the experimental results presented at
the end of the topic.
6.2 CIRCUIT TOPOLOGY
Overall the circuit topology consists of two parts, which

are power circuit and control circuit. The 5 kW inverter circuit
and its controller are shown in Figure 1. There are three
stages involved in the power stage construction, which are a
single-phase PWM bridge (inverter leg), a low-pass L-C filter, and
a low frequency 50 Hz isolation transformer. At the first stage
of conversion, the DC supply from PV or other DC sources
are converted to the AC waveform by applying high
switching PWM technique to the bridge (S1, S2, S3, S4)
switches. Then at the second stage, the generated PWM AC
waveform is filtered by a low-pass L-C filter so that a pure low
voltage AC is produced. In the third stage, a low frequency
isolation transformer acts as an isolation device to isolate the
system from the load, as well as to step up the voltage to a 240 VAC
as required by the load.
The proper control signal to drive the power switches in the

PWM power bridge is obtained from a C167 microcontroller, as
shown in Figure 2. The microcontroller C167 generates two
signals, that is high frequency PWM signal and low frequency 50
Hz square wave signal. The PWM signals are generated by
comparing the high frequency triangular waveform and the
sinusoidal waveform, where the negative cycle of the sinusoidal is
unfolded. The signals to each inverter leg, such as signals to S1
and S2 in leg A, has to be complementary to each other and a
suitable dead time is compulsory to avoid shoot through problem.
These signals are then applied to the logic gate circuit so as to
generate proper signal via logic comparison. Figure 3 shows the
PWM signals generated by the microcontroller, as well as the

signals for the top device at each leg obtained by implementing the
AND operator on both the PWM and 50 Hz square wave signals.
These PWM signals are then applied to the IGBT power devices
through the gate driver circuit, which is discussed later.
S1 S3
vPWM C vo
V dc
S2 S4
G a t e D ri ve L a G a te D r iv e Lb
C1 6 7
M i c ro c o n tro lle r
Figure 1 The basic diagram of 5 kW inverter and controller
C167 Microcontroller PWM LEG 1
PWM Gate Drive

S1
50Hz Square S2
PWM LEG 2
Gate Drive
S3
S4
Figure 2 Control signals and gates
Figure 3 Control and PWM signals

6.3 GATE DRIVE CIRCUIT
In the gate drive circuit implementation, IC SG3526 is used as

an isolation power to the main gate driver IC HCPL3150. IC
SG3526 generates gate two signals, one to the top Mosfet M1 and
second to drive bottom Mosfet M2, as shown in Figure 4. These
signals are complementary each others. The converter topology
used is push-pull with a high frequency transformer, operating at
the switching frequency of 200 kHz. The input voltage supply for
this circuit is 12 VDC. The generated AC waveform of the output
transformer is rectified by the diodes connected to the voltage
regulator IC LM7815 and LM7915 as shown in Figure 5. These
regulators generate regulated output voltage of +15 V (VCC) and
15V (VEE) for HCPL3150 DC power supply. To generate four
sets DC power supply to the HCPL3150, another push-pull
converter is required, however the signal to the Mosfet can be
obtained from similar IC SG3526.
Figure 4 Isolation power to HCPL3150 gate driver

Figure 5 Regulated output voltage 15V to HCPL3150 gate driver
The dead time is required to ensure that both top and bottom
Mosfets never turn on simultaneously in the standard H-bridge.
The dead time circuit topology is shown in Figure 6. Two input
signals come from the microcontroller C167, that is high frequency
PWM signal and low frequency 50 Hz square wave signal. These
signals are then applied to the logic gate circuit (AND operator IC
7408) so as to generate proper signal via logic comparison. The
timing diagram for the dead time circuit is shown in Figure 7. The
control PWM signal (pin 3) of IC 7408 is injected to the RCD
circuit network to produce signal of pin 9. When the PWM signal
is positive edge, diode D13 is in reverse biased and the voltage at
pin 9 increases exponentially due to the capacitor C34 charging.
As this voltage exceeds the thrash-hold voltage Vt of the Schmitt-
trigger IC 7414, the logic output of IC 7414 changes to logic low.
Likewise as the PWM signal is in negative edge, D13 is in forward
biased due to the capacitor discharges which cause pin 9 voltage
drops abruptly. Logic output of IC 7414 changes to logic high.
Clearly it is shows that there is a zero state at the initial positive
edge, which known as a dead time period. The voltage waveform
VGS is the required driver voltage to drive the Mosfet or IGBT
power device.
Figure 6 Dead time circuit
Input
Pin 3
7408
Vt
Pin 9
7414
Pin 8
7414
+15V
Output VGS
Deat time,td
-15V
Figure 7 Dead Time timing diagram for one gate drive
The circuit connection between isolation power supply and the

Mosfet driver HCPL3150 is shown in Figure 8. The output pin 6
and 7 are connected the Mosfet gate through a resistance Rg. 15
VDC is connected to pin 8 , whereas -15 V is connected to pin 5.
The Mosfet source or IBGT emitter needs to be connected to point
0V power supply.
Figure 8 Power Supply to HCPL3150
6.4 CONTROL AND MODULATION TECHNIQUE
The PWM signal can be generated by the microcontroller by

comparing the high frequency triangular waveform (carrier signal)
and the modulating waveform (sinusoidal signal). The negative
part of the modulating waveform is unfolded. The carrier
frequency signal is the switching frequency of the power devices
while the modulating frequency is the fundamental inverter output
frequency. The PWM scheme of the high frequency PWM bridge
used is based on symmetrical regular sampling method [5]. The
modulating technique is vital so that the Total Harmonic Distortion
(THD) produced by the inverter is at acceptable limit. The
switching angle is derived based on the volt-second equalization
method, shown in Figure 9. For a given modulation ratio, mr, and
modulation index, mI, the pulse width of the kth pulse, k , can be
calculated using (1).
k = 2 o m I sin k (1)
mf
where k = 1 ..
2
By using (1), the switching angle, i.e. the rising and falling
edge of the kth pulse can be calculated as:
k
Rising edge, 1k = k (2)
2
k
Falling edge, 2 k = k + (3)
2
Vm(t)
Vc(t)
Figure 9 Volt-second modulation technique
6.5 IMPLEMENTATION AND RESULTS
The high frequency PWM bridge is constructed using the

power IGBT Powerex CM50DY-12H 50A, 600V. This device has
a low-loss switch with good switching capability. The low
frequency power transformer is wound on a high power iron
powdered toroid. The hardware is shown in Figure 10. The
integrated circuit HCPL-3150, 2A is used as the IGBT gate driver
to drive all switches in the PWM bridge. Siemens SAK-C167CS-
LM (16 bit fixed-point micro controller) is used to generate the
PWM signals for the PWM bridge. Besides, to ensure the
reliability of PWM generation, there are three main relays used in
the system, which is DC relay, charger relay and AC relay. DC
relay is used to control the DC input voltage to the inverter,
whereas AC relay is used to monitor output AC voltage from any
faulty voltage, and the charger relay is used in the process of
charging gradually the DC link capacitor during start-up. This is

important to ensure that there is no inrush current flows due to the
used of big size of DC link capacitor.
(a) Low frequency transformer (b) Control Circuit
(c) Power Circuit
Figure 10 Hardware construction of the 5 kW inverter
Figure 11 shows the experimental results on the signals

generated by the microcontroller and the logic gates. The plot
indicates that the microcontroller successfully generates the PWM
signals applied to both inverter legs. The generated PWM signal
(top trace) is compared to the low frequency 50 Hz signal (bottom
trace). It produces the proper switching signal to drive leg A and B
of the PWM bridge (middle trace).
Figure 11 Microcontroller and gate logic signals
To examine the performance and operation of the designed and

constructed 5 kW inverter, i.e. on the power and control side, some
tests are carried out. Figure 12 shows the plot of output voltage,
load current and load power when the inverter operates in
continuous condition. The figure clearly indicates the capability of
the inverter to produce constant AC output voltage of 240 VAC. To
see the effectiveness of the constructed inverter in transient
condition, a load change is imposed. This is shown in Figure 13,
where the load power is stepped from 500 W to 5 kW. The figure
shows that the inverter is able to work effectively to produce a
stable output voltage of 240 VAC even though the disturbance is
imposed.
Power
Voltage
Current
Figure 12 Inverter operating at continuous 5 kW load

Load step
changes
Power
Voltage
Current
Figure 13 Inverter operating during load change from 500 W to 5 kW
6.6 CONCLUSIONS
A hardware development of a 5 kW inverter for AC power supply

applications has been discussed. Experimental results have shown that
the inverter operates well in producing a stable output voltage of 240 VAC
in normal operating condition with a 5 kW load. The robustness of the
constructed inverter in transient condition is further proven as the output
voltage is still steadily maintained even though a large load disturbance
is imposed.
REFERENCES
[1] Saha, S and Sundarsingh, V. P, Novel Grid Connected Photovoltaic

Inverter, IEE Proc. Generation, Transmission and Distribution. Vol.
143, No.2, pp. 219-224, March 1996.
[2] Watanabe H. et al. A Novel Utility Interactive Photovoltaic Inverter

with Generation Control Circuit, IEEE Transaction on Industry
Application, pp.721-725, 1998.
[3] Energy Research Program for Malaysia in The Next Five Years And
Beyond, MOSTE, August 1995.

[4] Zainal Salam, Toh Leong Soon and Mohd Zulkifli Ramli, Hardware
Implementation of the High Frequency Link Inverter Using the
dSPACE DS1104 DSP Board, IEEE International Power
Engineering Conference, PECON06, Putra Jaya, Selangor,
Malaysia, pp. 348 352, Nov. 2006.
[5] Mohd Zulkifli Ramli, Zainal Salam, Leong Soon Toh, and Chee Lim
Nge, A Bidirectional High-Frequency Link Inverter Using
Center-Tap Transformer ", IEEE 35th Annual Power Electronics
Specialist Conference, Auchen, Germany, pp. 3883 3888, 2004.
Cost Comparison For A Stand-Alone Residential Photovoltaic System 97
7
COST COMPARISON FOR A STAND-
ALONE RESIDENTIAL
PHOTOVOLTAIC SYSTEM BETWEEN
AMORPHOUS SILICON AND CADMIUM
TELLURIDE IN MALAYSIA
A Vigneswaran
Md Shah Majid
Hasimah Abdul Rahman
Mohd Yusri Hassan
7.1 INTRODUCTION
The price of crude oil has risen from a low of USD 2 per barrel to
a high of USD 126 (as at 4th June 2008). The present system
cannot be maintained for more than one or two generations. This
resulted in searching for an alternative energy source. Since
Malaysia lies entirely in the equatorial region with an average
2
daily solar radiation of 4,500kWh/m , with sunshine duration of
about 12 hours, solar photovoltaic system is applicable to be used
as an alternative primary energy to replace the conventional
system[1]. Stand-alone domestic PV systems provide electricity to
households in remote areas. The system provides basic electricity
for lighting, refrigeration and other domestic electrical equipments.
These application have been installed almost everywhere in the
world, especially in rural areas of developing countries. The PV is

often the most appropriate technology to meet the energy demand
of isolated communities. Stand-alone PV systems generally offer
an economic alternative to the extension of electricity distribution
grid at distance of more than 1 or 2 km from existing power lines.
Over the past decades alone, the Government has been
continuously subsidizing the populations fossil fuels need. The
fuel subsidy in 2005 alone was about RM25 billion, where RM9
billion were for power generation and RM16 billion were for the
transport sector. Therefore, new and effective ways to reduce the
overwhelming dependence on natural gas, as well as the energy
subsidy, have to be found. One natural resource that Malaysia has
in abundance and which is totally free is the sunlight [2].
PV cells are made of semiconductor materials. The major types
of materials are crystalline and thin films, which vary from each
other in terms of light absorption efficiency, energy conversion
efficiency, manufacturing technology and cost of production. Thin
films constitute about 10% of total PV module shipments, the PV
industry sees great potentials of thin-film technology to achieve
low-cost PV electricity uses either a-Si or CdTe[3].
An approach to economic analysis that looks at the total life cost
of this different system is therefore required. The evaluation of the
economic analysis is carried out using the Life-cycle Costing
methodology (LCC). LCC analysis is a tool used to compare the
ultimate delivered costs of technologies with different cost
structures [4]. Rather than comparing only the initial capital costs or
operating costs, LCC analysis seeks to calculate the cost of
delivering a service over the life of the project. The final cost per
kWh is estimated independent of the technology used to deliver the
electricity.
The LCC method considers the initial costs and all other future
costs or cash benefits and discounts them to their present value.
Moreover, this system allows the breaking down of costs, which
enables one to carry out a sensitivity analysis study that analyzes
the impact of each of the parameters involved in the LCC on the
cost of electricity.
7.2 COST AND ANALYSIS OF PV CELLS AND BALANCE

OF SYSTEM
7.2.1 Solar Radiation and Temperature Data
Meteorological data was obtained from the Meteorology

Headquarters Petaling Jaya, which comprised solar radiation, mean
relative humidity and maximum temperature in Ipoh for 4 years
beginning year 2003. A pyranometer is used to measure the solar
radiation. These data together with other information are used to
evaluate the PV energy production and cost analysis using
RETScreen
7.2.2 Load Characteristics
A typical urban house was chosen as a stand-alone model for the

simulation where the loads that were chosen are an average
household to power all the different loads in the house on a typical
day. For complex loads like a household, it is sometimes difficult
to anticipate every electrical load. For example, electric clocks,
lighting, TVs, stereos and other appliances sometimes draw small
amounts of power even when they are turned off. For this reason,
the duration of hours used per day for every load is multiplied by a
fudge factor of 1.5. Figure 4.1 provides an example load for a
household using this method. These values are keyed in directly
into the load characteristics section in RETScreen.
Figure 7.1: Typical household electrical appliances and run times
7.2.3 PV Module Cost/kWh
As of February 2008, there are currently 196 solar module prices

below USD 4.75 per watt or 12.2 % of the total sample. This
compares with 185 prices below USD 4.75 per watt in January
2008 [5]. Amorphous silicon photovoltaic module is the rapidly
growing alternative technology. The use of lower cost materials
and thin film technology results in modules in the USD 4 per watt
peak range and also offers the ability to integrate with standard
building materials such as roofing, windows, etc. Cell efficiency is
in the range of 8 10% [6]. CdTe solar companies manufacture
Cadmium Telluride thin films and have the potential advantage of
higher efficiency, lower complexity and lower cost than CIGS at
USD 2.80. First Solar is an established vendor that sells CdTe
solar modules at an efficiency of 9.3% [7].
Table 7.1: Minimum price per watt peak for various photovoltaic
modules
Type of PV Cell Price per Watt Peak Price per Watt Peak
(USD) (RM)
Amorphous Silicon 3.71 12.50
Cadmium Telluride 3.31 11.16

(CdTe)
(USD 1 = RM 3.3715, Bank Negara, 1st Feb 2008)
7.2.4 Module Support Structure Cost per Meter Square and

Electrical Components, Design and Installation Cost per kWh
The scientific and technical basis of the project is sound.

Photovoltaic panels integrated into a buildings exterior (usually
the roof in equatorial countries), reduce some components of the
installation cost, and also improve the buildings integrity. Since
the overall cost is reduced, the photovoltaic modules are assumed
to be roof mounted building integrated photovoltaic system.
Several case studies were conducted by Pusat Tenaga Malaysia
under the Malaysia Building Integrated Photovoltaic Project are
evaluated to derive the cost per meter square for the mounting
structure and roof modification. The following were the case
studies considered:
i. 9.9 kWp installation at commercial shop lots at

Lot 53, Jalan SS 21/37, Damansara Utama,
Petaling Jaya, Selangor[7].
ii. 11.88 kWp installation at Putrajaya Perdana
Berhad Headquarters, Precinct 16, Putrajaya [8].
iii. 4.0 kWp BIPV System Installation at Lot 488,
Jalan Khaya, Country Heights Damansara,

Petaling Jaya[9].
iv. 4.2 kWp BIPV System Installation at No 96, Lot
5650, Lorong Indah 2, Taman Perkota, Bukit
Sebukor Melaka [10].
Averaging the cost of mounting structure which includes the

labour costs and all other miscellaneous costs would equal RM
511.30/m2 while the cost of electrical components, design and
installation which includes the labour costs and all other
miscellaneous costs would equal RM 3721.80/kWh.
7.2.5 Cost of Balance of System
For professional use the more advanced and expensive batteries are
often more cost effective in the long run. Unlike solar modules and
inverters, batteries are a mature technology and a product that is
used in volume outside the solar energy industry. Consequently,
the opportunity for major costs breakthroughs over the medium
term is limited. Across the survey of 536 different prices amongst
189 product models from 33 companies, the average price of per
watt hour output is USD 2.01 which approximates RM 6.78 (USD
1 = RM 3.3715, Bank Negara, 1st Feb 2008) per watt hour
output[11].
The rated power of the inverter is not the only factor that
determines price. The technology varies considerably from
manufacturer to manufacturer, leading to differences in efficiency,
size, weight, reliability, etc. Navigant Consultant Inc. (NCI)
conducted a survey for small inverters below the size of 10 kW.
For 2004, the price range was between USD 0.67 per watt peak to
USD 2.67 per watt peak. For the year 2005, the price range was
between USD 0.48 to USD 2.42 per watt peak. A survey involving
1189 inverters from a range 251 product models amongst 40
companies were compared to average the price of inverter per watt
peak [12]. Comparing the results between NCI and Solarbuzz, the
variation of price is USD 0.032 per watt peak which is negligible.

The value of USD 0.718 per watt peak equaling RM 2.42 per watt
peak (USD 1 = RM 3.3715, Bank Negara, 1st Feb 2008) is deemed
more accurate as the latter was a result of a survey undertaken in
2008 compared to the former undertaken in 2005.
The price index methodology is to aggregate the prices of all
charge controllers on the survey and divide by the aggregate amps
output of those same products. This produces a price per ampere
for the charge controller. As at February 2008, the average price of
the charge controller is USD 5.81 per amp which equals RM 19.59
per amp (USD 1 = RM 3.3715, Bank Negara, 1st Feb 2008) [13].
7.2.6 Array Losses Due to Mismatch, Dirt and Degradation
Physical differences between cells or cell-processing material may

arise during normal processing and different cells may have
slightly different characteristic parameters. Consequently, in an
array the output power of the combination is less than the sum of
maximum output power of constituents. This power loss may tend
to enhance with time due to degradations resulting from aging. For
example, for the ensemble of solar cells used in the present work
the fractional power loss of 12% resulting from the aging (soiling)
can be counterbalanced by series paralleling in an array oversized
by 2.4%. Taking an average between these two figures would
result in an array efficiency of 89.2% or array losses of 11.8%
[14]
.
7.2.7 Various PV Modules and Balance of System

Technologies and Efficiencies
A battery of high quality which is the valve regulated lead-acid

(VRLA) batteries with a typical lifetime of 13 years corresponding
with a maximum allowable depth of discharge of 50% is
considered [15]. The battery efficiency would average 90% and
there would be no significant increase in the near future [16].

The inverter efficiency is estimated to be around 95% while the
charge controller efficiency at 90% with a lifespan of 10 years [15].
Market efficiencies for various photovoltaic modules have been
summarized in Table 4.2. Output powers from these PV modules
greatly differ with the efficiencies of its respective modules. The
range of efficiency for PV modules of amorphous silicon and
cadmium telluride (CdTe) is between 2 4%.
Table 7.2: Summary of types of PV cells investigated
Type of PV Cell Product Model Market Power

Rating
Efficiency (W)
Amorphous silicon Uni-solar, US-116 6.20% 116
High efficiency
Low efficiency BP, MST 50 MV 6.10% 50
Cadmium Telluride BP, Apollo 980 8.60% 80
High efficiency
Low efficiency First Solar 6.90% 50
7.3 COST EVALUATION ANALYSIS
7.3.1 Initial Capital Investment
The 2 modules considered for the amorphous silicon with their

respective product model number, efficiency and power rating in
watt are listed below:
High efficiency, Uni-solar, US-116, 6.2 %, 112 W

Low efficiency, BP, MST 50 MV, 6.1 %, 50 W
Cost analysis is initially undertaken for the high efficiency module

and yields the result as observed in Figure 4.2.
Figure 7.2: High efficiency amorphous silicon
RETScreen further calculates the suggested PV array power and

PV array area. As observed from Figure 4.2, for the high efficiency
amorphous silicon, PV array power suggested is 2.86 kW and PV
array area is 46.1 m2. Next would be to determine the total net
present value of the project which is undertaken in the Cost
Analysis section of RETScreen. The result of which is observed in
Figure 4.3.
Figure 7.3: Cost analysis of high efficiency amorphous silicon
The costs of the charge controller has been neglected. Using a 36

A charge controller at an estimated average price of RM 19.59/A
would yield a cost of RM 705. RM 705 added to the initial costs
would yield a grand total net present value of RM 209,018.
The total net present value for both low and high efficiency
amorphous silicon can be summarized as in Table 4.3.
Table 7.3: Total net present value for low and high efficiency
amorphous silicon
Type of Amorphous Silicon Module Total Net Present Value
Low efficiency, Uni-solar, US-116 RM 209,409
High efficiency, NT-R5E3H RM 209,018
Averaging the costs of the total net present value would yield
RM 209,214. Hence, the average cost for a stand alone system
using amorphous silicon to supply an annual load of 3075.9 kWh
a.c is RM 209,214. Hence, the average costs for a stand alone
system using a CdTe to supply an annual load of 3075.9 kWh a.c is
RM 202,140.
7.3.2 Life Cycle Costing Analysis
The LCC method considers the initial costs and all other future
costs or cash benefits and discounts them to their present value.
Moreover, this system allows the breaking down of costs, which
enables one to carry out a sensitivity analysis study that analyzes
the impact of each of the parameters involved in the LCC on the
cost of electricity[17].
The main equation of the LCC is :
LCC = (C + M + R S) / E (1)
where,
LCC = life cycle cost (RM/kWh);
C = total initial capital (RM);
M = total maintenance cost (RM);
R = replacement cost anticipated for the whole life of the
system (RM);
S = salvage or scrap vale at the end of system life (RM) and;

E = electricity produced (kWh/annum).
The above values have to be converged to their present values by

using either of the following two equations:
For a fixed amount of money to be paid in n years, the present
value interest factor is given by [17]:
(2)
where,
PVIFk,n = present value interest factor of one unit of money at
a discount rate k at year number n.
For an annual recurrent expenditure, the present value interest

factor annuity is given by [17]:
(3)
In order to choose a suitable value for the discount rate, k, it

would be necessary to know the highest rate of return, known as
the nominal discount rate, m, on the money invested had it not
been spent on the project. This was taken as equal to the maximum
interest rate charged by the local commercial bank of 3.8%. Also,
the inflation rate, i, during the initial year which is known as year
number zero has to be clearly defined. This was obtained from the
Bank Negara Malaysia as being 2.5 % in 2007.
The net discount rate can then be calculated from the equation [17]:
(4)
k = 3.71 %
The length of period of the analysis is chosen to be the best

service life of the longest living component. In the case of PV
system comparison, the useful life of a PV module is in the range
of 20 30 years [18]. As for the, n, of the PV modules it is 25 years,
so this is the period chosen for this analysis.
As for the, n, of the battery control unit or charge controller and
the inverter, it is subjected to be replaced every 10 years [16]. The
lifetime for inverters and charge controllers is also 10 years[19]. As
for the, n, of the inverter and charge controller it is 10 years.
Valve regulated lead acid cells (VRLA) which are popularly
installed for stand-alone systems, have average life cycle of 13
years [20]. Since VRLA is selected as the battery type, n, for the
battery is 13 years.
System maintenance is another recurring cost. This category
includes equipment maintenance, site maintenance, system
supervision etc. For the PV module, very little maintenance is
required. The batteries in the system will require inspection and
topping up, approximately every 3 6 months [21]. In our analysis a
typical value of USD 2.80 which approximates RM 9.45 (USD 1 =
RM 3.3715, Bank Negara, 1st Feb 2008) per month is assumed [19].
Salvage value is the systems net worth in the final year of its life
cycle period. A salvage value of 20 % of the original cost is
assigned for the PV module, structure, batteries and cable, charge
controller, labour costs and wires following the current common
practice and guidelines given by Sandia National Laboratories.
For the amorphous silicon module stand alone system, the
salvage value of 20 % from the initial cost of RM 209,214 would
yield RM 41,842.80.
Applying equation (1) the total LCC cost in 25 years would be as

follows:
LCC25years = RM 209,214 + RM 1843 + RM 74,891.70 + RM

11,129.10 + RM 7731.25 + RM 489.76 + RM
340.23 RM 41,842.80
= RM 263,796.24
The LCC cost per year would be equal to LCC25years / PVIFAk,n,
LCC 1year = RM 263,796.24 / 16.11

= RM 16,374.69
Cost per kWh would be divided by the annual load of 3075.9 kWh
Cost/kWh = RM 16,374.69 / 3075.9 kWh

= RM 5.32/kWh
The Cost/kWh for an amorphous PV module for a stand alone

system supplying annual load of 3075.9 kWh is RM 5.32. The
Cost/kWh for a CdTe PV module for a stand alone system
supplying annual load of 3075.9 kWh is RM 5.20.
7.4 CONCLUSION
The cost of generating per kWh energy using a stand alone system
for a typical four member family household with an annual
consumption of 3075.9 kWh of energy is RM 5.26/kWh with CdTe
technology marginally lower by RM 0.16 for every kWh
generated. This value compared with several other European and
Mediterranean countries are reasonable. According to Lazou and
Papatsoris [18], the predicted levelised energy cost for a stand alone
system in 2005 would range between USD 1.039/kWh to USD
8.177/kWh. Upon conversion to Ringgit Malaysia, these values
equal RM 3.51/kWh and RM 27.60/kWh (USD 1 = RM 3.3715,
Bank Negara, 1st Feb 2008). This shows that the cost of generating
per kWh of energy in Malaysia would be 49.86 % more expensive
than the lowest Mediterranean or European nation but 81.16 %
cheaper. This justifies that RM 5.26/kWh is reasonable.
REFERENCES
(1) United Nation Development Program, Malaysia, 2004,

Building Integrated Photovoltaic Technical Application
Project, TNB Research Sdn Bhd.
(2) Malaysia, Germany Chamber of Commerce and /industry,
Quarterly Report 11-12-07, 2007
(3) Olivia Mah, 1998, Fundamental of Photovoltaic Materials,
National Solar Power Research Institute
(4) Sandia Natinal Laboratories, 1998, Standalone PV Systems.
A Handbook of Recommended Design Practices,
Albuquerque, New Mexico, USA
(5) Solarbuzz, 2008a, Solar Module Price Highlight,
www.solarbuzz.com/solarmoduleprices.htm, 22 February
2008
(6) Bo-Varga, 2006, Nanotechnology Impact on Solar Power
Silicon Valley Solar Ventures
(7) Azah Ahmad, Vincent Tan, Chen Wei Nee, Daniel Ruoss,
2007a, 9.9 kWp BIPV System Installation at Damansara
Utama Shoplots, Damansara Utama, Malaysia Building
Integrated Photovoltaic Project
2007d, 11.88 kWp BIPV System Installation at Putrajaya
Perdana Berhad HQ, P16, Putrajaya, Malaysia Building
Integrated Photovoltaic Project
2007b, 4.0 kWp BIPV System Installation at Lot 488, Jalan
Khaya, Country Heights Damansara, Petaling Jaya,
Malaysia Building Integrated Photovoltaic Project
2007c, 4.2 kWp BIPV System Installation at No 96, Lot

5650 Loronh Indah 2, Taman Perkota, Bukit Sebukur
Melaka, Malaysia Building Integrated Photovoltaic Project
(11) Solarbuzz, 2008c, Battery Price Highlights,
www.solarbuzz.com/batteryprice.htm, 22 February 2008
(12) Solarbuzz, 2008b, Inverter Price
Highlight, www.solarbuzz.com/inverter.htm, 22 February
2008
(13) Solarbuzz, 2008b, Charge controller Price Highlight,
www.solarbuzz.com/chargecontrollerprices.htm, 22
February 2008
(14) Solarcentury, 2006, Solar energy Factsheets.
www.solarcentury.co.uk/knowledge_base/search, 23
February 2008
(15) D.A.S. Rand, P.T. Moseley, J. Garche, C D Parker, 2004,
Valve Regulated Lead-Acid Battery, pp 1-10, Elsevier BV
(16) N.D Kaushika, Anil K Rai, 2005, An Investigation of
Mismatch Losses in Solar PV Cells Networks, Bharati
Vidyapeeth School of Multidisciplinary Research, B.V.
College of Engineering, New Delhi, India
(17) R.W. Larson et. al 1992, Economics of Solar energy
Technologies, American Solar Energy Society (ASES)
(18) Anathasia A Lazou, Anastassios D Papatsoris, 2000. The
economic of Photovoltaic Stand-alone residential
Households: A Case Study of Various European and
Medittaranean Locations. Department of Electronics,
University of York, Heslington, York, UK
(19) Charles Iskander and Edward Scerri, 1996, Performance
and Cost Evaluation of a Stand-Alone PV in Malta,
Institute of Energy Technology, University of Malta
(20) F. Lasnier and T. G. Ang, 1990. Photovoltaic Engineering
Handbook, CRC Press, USA
(21) Hammon R, D. Srinivasan, A Harris, K Whitefield, J

Wohlgemuth,
1997, Effects of Soiling on PV Module and Radiometer
Performance, Pp 1121-1124. Conference Record of the 26

IEEE Photovoltaic Specialists Conference
8
A MODULAR STRUCTURED
MULTILEVEL INVERTER
FOR FUEL CELL APPLICATIONS
Naziha Ahmad Azli
Abdul Halim Mohd Yatim
8.1 INTRODUCTION
In recent years, focus on fuel cell as an alternate source of energy

has increased due to the foreseeable shortages in fossil fuel and
nuclear energy sources, limitation in potentially viable
hydroelectric power sites and the need to produce power in a more
environmental friendly manner. Many power utilities and
independent power producers are in fact planning multi-megawatt
systems based on alternate sources of energy and for them to make
an impact on future energy supply, they need to be utility
interactive at medium to large power ratings [1]. As the fuel cell
produces dc voltage and current, power conditioning is essential
through inverters with control techniques that must ensure fixed
voltage with minimum specified total harmonic distortion (THD)
at a fixed frequency is maintained on its output despite non-
linearity or unbalance of the load. Various inverter circuit
topologies and control techniques for fuel cell application have
been proposed in technical literature based on the so called
conventional two or three level structured inverters [1-6]. For
higher output voltage capacity and reduction in harmonic
distortion, these inverters are connected in series using
transformers [3,5]. While such structure is advantageous in terms
A Modular Structured Multilevel Inverter for Fuel Cell Applications 115
of the electrical isolation provided by the transformers between the

inverters and the load, the transformers too are the main
contributor to problems such as bulkiness, high loss and high cost
to the overall power conditioning system [7].
A non conventional topology know as multilevel inverter is

identified as an alternative for the fuel cell power conditioner. The
general concept of the multilevel inverter is to synthesise a
sinusoidal voltage from several levels of voltages. By having more
than two voltage levels to build the sinusoidal voltage, harmonic
distortion of the waveform is significantly minimised. There are
several types of multilevel inverters but the one considered in this
paper is the modular structured type which consists of single phase
full-bridge inverters with separate dc sources, connected in series.
By considering this inverter topology, the dc input voltage to the
inverter thats required from the fuel cell can be significantly
reduced without reducing its ac output voltage. Some of the
advantages of the modular structured multilevel (MSM) inverter
can be seen from the aspect of the inverter component count and
the modularity and simplicity of its circuit layout. Of course,
several fuel cell output of lower voltages are required in this
topology, depending on the number of levels and the type of
system i.e. single phase or three-phase, that the inverter is designed
for. Some of the applications of the MSM inverter encompassing
various control strategies are as reported in [7] as a static var
generator, [8] as a high power converter for plasma stabilisation in
nuclear fusion experiments and [9] as a solar arrays and batteries
power conditioner for space environment.
This research is concerned with the modulation strategy aspect of

controlling the MSM inverter. The following section describes its
circuit topology in general and its application in a fuel cell power
conditioner. A single carrier instead of multicarrier [10,11]
modulation strategy proposed for the MSM inverter based on
sinusoidal natural sampling pulsewidth modulation (PWM)
technique. is elaborated in section 3. The use of the hybrid PWM
switching scheme [12] in determining the gating signals for the

power devices is also highlighted in this section. Results of a
digital simulation work performed using MATLAB/SIMULINK to
assist in the verification and analysis of the proposed modulation
strategy are presented in section 4. This is then followed by some
concluding remarks on the work carried out in section 5.
8.2 MODULAR STRUCTURED MULTILEVEL

INVERTER
8.2.1 BASIC CONFIGURATION
Figure 1 shows the single phase configuration of the MSM

inverter. This multilevel inverter topology is chosen for the fuel
cell power conditioner not only for the reasons mentioned in the
previous section but also due to its unique structure that allows
each module consisting of the single phase full-bridge inverter to
be controlled separately, using the proposed single carrier
modulation strategy.
Figure 1 Single phase structure of the MSM inverter
The output phase voltage of the MSM inverter is actually the

summation of the output voltage of the respective modules that are
series connected. The number of modules (M) which is equal to
the number of dc sources required depends on the number of
levels (N) of the MSM inverter. It is usually assumed that N is odd
as this would give an integer valued M which would simplify
further analysis. As an example, for an output voltage consisting of
five levels which include +2E, +E, 0 , -E and +2E, the number of
modules needed is 2. Equation (1) gives the relationship between
N and M.
M = (N-1)/2 (1)
8.2.2 FUEL CELL APPLICATION
Similar to other alternate energy sources such as solar and wind,

fuel cell too produces dc voltage that is unregulated. On the other
hand, to ensure high performance of the control strategies imposed
on the MSM inverter, the separate dc sources for each module has
to be constant and ripple free. This paper concentrates on the dc to
ac conversion of the fuel cell output voltage, thus assuming that
the dc voltage is constant. Figure 2 however gives an overall
description of the fuel cell power conditioning system in terms of
functional block diagrams, based on a 5 level modular structured
inverter.
solar irradiance.
Figure 2 Functional block diagram of the fuel cell power conditioner

system
8.3 SINGLE CARRIER MULTI LEVEL PWM

TECHNIQUE
8.3.1 THE MODULATION PROCESS
The single carrier multilevel PWM technique is actually

based on the classical unipolar PWM switching technique [13].The
main idea behind it is to use several sinusoidal modulation signals
m(t), while having only one triangular carrier signal c(t). The
number of modulation signals needed is equal to the number of
modules in the MSM inverter. These sinusoidal modulation signals
have the same frequency fm and the same amplitude Am.
However, there exist an offset between each of the modulation
signal at its starting point as shown in Figure 3(a) and (b),
depending on the number of modules.
1.5
m1(t)
c(t)
1
0.5
m 2(t)
-0.5
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
Time(s)
(a)
2
1.5
m1(t)
c(t)
1
0.5
m 2(t)
-0.5
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
Time(s)
(b)
Figure 3 The single carrier multilevel PWM technique for (a) ma =

0.8 (b) ma = 0.4
For example, for a 5 level modular structured inverter, 2 modules is required,

thus the modulation signals are
m1(t) = | Am sin(2fm)t| (2)

m2(t) = | Am sin(2fm)t| - 1
The carrier signal is a triangular one set at frequency fc and amplitude Ac.
During the first half of the MSM inverter output fundamental period,
comparison between the modulation signals and the carrier signal gives 1 if
m(t) > c(t), 0 otherwise whereas in the second half of the fundamental period,
comparison between the modulation signals and the carrier signal gives -1 if
m(t) > c(t), 0 otherwise.Equations (3) and (4) defines the amplitude modulation
ratio ma and the frequency modulation ratio mf for the single carrier multilevel
PWM technique.
mf = fc / fm (3)
ma = Am /(M.Ac ) (4)
As ma ranges between 0 and 1, Am ranges between 0 and M while Ac remains

fixed at 1. For the 5 level modular structured inverter, two pulsewidth
modulated waveforms are obtained. These waveforms can then be used to
derive the actual gating signals for the power devices in each module which
finally gives the actual MSM inverter output voltage. Depending on the value of
ma there may be instances whereby only one pulsewidth modulated waveform is
obtained while the other emits a zero signal. This occurs when ma is less than or
equal to 0.5. In this case, the single carrier multilevel PWM technique works
exactly as the classical unipolar PWM switching technique which gives only a
three level output voltage namely +E, 0 and -E for the MSM inverter.
8.3.2 SIMULATION RESULTS
A 5 level modular structured inverter is simulated in

MATLAB/SIMULINK using the proposed single carrier
multilevel PWM technique. For simulation purposes, it is assumed
that the dc voltage input to each module is V = 100 V, the output
voltage fundamental frequency f1 = fm = 50 Hz while the carrier
frequency fc = 2kHz (mf = 40). As discussed earlier, comparison

between m1(t) and m2(t) with c(t) gives two pulsewidth
modulated waveforms p1(t) and p2(t). The gating signals (g11 and
g21) for power devices S11 and S21 and (g12 and g22) for power
devices S12 and S22 is then determined according to Table 1 to
allow implementation of the hybrid PWM switching scheme.
These parameters are illustrated in detail in Figure 4 and Figure 5
for ma = 0.4 and ma=0.8 respectively. The gating signals for S31,
S41, S32 and S42 are not shown as they obviously complement
that of S11, S21, S12 and S22 accordingly.
1
c (t)/m 1(t)
0.5
0
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
p1(t)
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
g11
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
g21
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
Time(s)
c (t)/m 2(t)
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
pw2(t)
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
g12
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
g22
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
Time(s)
Figure 4 Switching techniques and gating signals of a 5 level

modular structured inverter with ma = 0.4
2
c (t)/m 1(t)
0
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
p1(t)
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
g11
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
g21
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
Time(s)
c (t)/m 2(t)
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
p2(t)
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
g12
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
1
g22
0
-1
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
Time(s)
Figure 5 Switching techniques and gating signals of a 5 level

modular structured inverter with ma = 0.8.
Figure 6 and 7 shows the output voltage and frequency spectrum

of the 5 level modular structured inverter for ma = 0.4 and
ma=0.8. The even valued mf chosen for the 5 level modular
structured inverter gives an output voltage waveform that is odd as
well as quarter-wave symmetry whereby from the frequency
spectrum of Figure 6 and Figure 7, it can be depicted that only the
odd harmonics exist. Such features however are not obtained if odd
valued mf is utilized.
100
80
60
40
20
vo(t)
-20
-40
-60
-80
-100
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
Time(s)
1.8
1.6
1.4
Norm aliz ed fourier coeff.
1.2
0.8
0.6
0.4
0.2
0
0 5 10 15 20 25 30 35 40 45 50 55
Harmonic order
Figure 6 5 level modular structured inverter (a) output voltage (b)

frequency spectrum for ma = 0.4.
8.4 CONCLUSION
A MSM inverter topology for fuel cell applications has been

proposed in this paper as a different approach in targeting for a
higher powered generation system. Its circuit configuration has
allowed it to be controlled using the single carrier multilevel PWM
technique successfully which is verified by the results of the
digital simulation. For a 5 level modular structured inverter, only 3
combination of carrier and modulation signal is needed in this
control technique compared to 5 signals required when the
multicarrier multilevel PWM technique is utilised. It is believed
that the ability to control each module of the MSM inverter
separately using this technique, is actually an added advantage
especially in its hardware implementation development.
REFERENCES
[1] R. Naik , N. Mohan, M. Rogers and A.Bulawka. A novel grid

interface, optimized for utility-scale applications of
photovoltaic, wind-electric and fuel-cell systems. IEEE
Transactions on Power Delivery. Vol. 10, No. 4, pp. 1920-
1926, 1995.
[2] J. M. Carrasco et. al. Microcomputer control of a fuel cell
power system. IEEE 20th International Conference on
Industrial Electronics, Control and Instrumentation. 1994.
[3] N. Eguchi and T. Emura. Self-commutated inverter for fuel
cell power plant. IEEE IAS Conf. Rec. 1986.
[4] S. Igarashi, K. Kuroki, T. Hatta and H. Mogi. Interconnection
inverter consisting of large capacity dc/dc converter and HF
PWM inverter for fuel cell power plant. 1992*
[5] M. Edani, M. Yamamoto and N. Eguchi. DDC controller for
inverters in fuel cell dispersed generation plant. IEEE Conf.
Rec. of the Power Conversion Conference. Japan 1993.
[6] W. Shireen and M. S. Areffen. An utility interactive power
electronics interface for alternate/renewable energy systems.

IEEE Transactions on Energy Conversion. Vol. 11, No. 3, pp.
643-649, 1996.
[7] F. Z. Pheng, J. Lai, J. W. McKeever and J. VanCoevering. A
multilevel voltage-source inverter with separate DC sources
for static var generation. IEEE Transactions on Industry
Applications. Vol. 23, No. 5, pp. 1130-1138, 1996.
[8] M. Marchesoni, M. Mazzucchelli and S. Tenconi. A non-
conventional power converter for plasma stabilization.
PESC88 Conference Record. Japan 1988.
[9] D. Casini and M. Marchesoni. Sliding mode multilevel control
for improved performances in power conditioning systems.
IEEE Transactions on Power Electronics. Vol. 10, No. 4, pp.
453-463, 1995.
[10] G. Carrara, S. Gardella and M. Marchesoni. A new
multilevel PWM method: A theoretical analysis. IEEE
Transactions on Power Electronics. Vol. 7, No. 3, pp. 497-
505, 1992.
[11] V. G. Agelidis, D. M. Baker, W. B. Lawrance and C. D.
Nayar. A multilevel PWM inverter topology for photovoltaic
applications. ISIE97. Portugal 1997.
[12] R. Lai and K. D. T. Ngo. A PWM method for reduction of
switching loss in a full-bridge inverter. IEEE Transactions
on Power Electronics. Vol. 10, No. 3, pp. 326-332, 1995.
[13] N. Mohan, T. M. Undeland and W. P. Robbins. Power
Electronics Converters, Applications and Design. John
Wiley 1995.
[14] M. Marchesoni. High-performance current control
techniques for applications to multilevel high-power voltage
source inverters. IEEE Transactions on Power Electronics.
Vol. 7, No. 1, pp. 189-204, 1992.
Indeks 127
INDEX
Adjustable speed drive, 163, 164

network of adjustable speed
drives, 163, 164
Energy storage devices, 2,8
Fuel Cell, 1, 5-6, 8, 10, 15, 114-

116, 118-119, 126
Harmonics, 163 164

total harmonic distortion, 166
39
harmonic current, 144 45
harmonic spectrum, 167,
170 172
Inverter
multilevel inverter, 114-116
single phase PV inverter, 58,
59, 67
Maximum Power Point

Tracking, 21, 25, 40, 41
Point of common coupling, 164,

168-172
Power quality, 163, 165, 169,
173
Power electronics, 9-10, 14-15

Bookchapter Fke01 PDF

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Bookchapter Fke01 PDF

Enviado por

Direitos autorais:

Formatos disponíveis

First Edition 2008

MAKBUL ANWARI 2008

Perpustakaan Negara Malaysia Cataloguing-in-Publication Data

Recent trends in renewable energy and power electronics research / editor

Editor: Makbul Anwari

Diatur huruf oleh / Typeset by

Diterbitkan di Malaysia oleh / Published in Malaysia by

Dicetak di Malaysia oleh / Printed in Malaysia by

Chapter 1 The Role Power Electronics in Future Energy 1

Chapter 2 Design and Development of a Maximum 21

Chapter 3 Analysis of Perturbation Step Size for 40

Chapter 4 Load Sharing Characteristic of Single Phase 58

Chapter 5 Power Quality Behavior of Single Phase Fed 71

Chapter 6 Development of a 5 kW Inverter for AC 84

Chapter 7 Cost Comparison for a Stand-Alone 97

Chapter 8 A Modular Structured Multilevel Inverter 114

Power electronics technology is used for efficient control of

electronics engineers. To facilitate future referencing, ENCON has

Part I : Modeling and Control of Power Converters and Drives

ENCON will strive to maintain, update and expand the series.

1.1 FUTURE ENERGY SYSTEMS

Presently most of the worlds energy supply comes from fossil

generation unit, it has to be converted to a suitable form by

1.1.1. Micro Turbines

Micro-turbines are small combustion turbines with outputs power

Figure 1. Power electronic converters for micro-turbine

1.1.2 Photovoltaic (PV) Systems

Figure 2. Power electronic converters for micro-turbine

1.1.3 Fuel Cells

Figure 3. Power electronic converters for fuel cells system

Figure 4. Singly fed induction generator wind turbine

1.1.4 Wind Turbines

1.2 ENERGY STORAGE DEVICES

DERs respond relatively slowly to a sudden increase in load

1.3 POWER FLOW AND CONTROL

The DER system consists of power electronic converters and a

energy sources available as discussed in the previous section.

When connected to the grid, the power electronics must maintain

Figure 5. Power flow and control of the energy sources, ESDs,

1.3.1. Stand-Alone Mode

system is too expensive or impossible, for example, rural areas in

Figure 6. Power flow and control: stand-alone mode operation

1.3.2. Grid-Connected Mode

is the dc load that may be connected to the dc link via a dc-dc

by the controller for optimum energy utilization and cost. During

Figure 7. Power flow & control: grid-connected mode

1.4. GREEN INDUSTRIALISATION

Green industrialization is a broader term concerned with

sustainable and environmentally friendly. The advantage is the

Power electronics is a key and enabling technology for many

[10] Huang, S. J., Pa, F.S., Design and operation of grid-

Power Electronics Conf. and Exp., APEC 2002, Vol. 2, pp.

Application of a super-conducting magnetic energy storage

institutional and technical change in agriculture, Land

Planning Using Probabilistic Production Costing and

In urban areas, the implementation of PV technology,

PV modules under varying weather conditions. MPPT converters

2.2 BUILDING INTEGRATED PHOTOVOLTAIC

The term building integrated photovoltaics refers to

1. Financial appeal costs are combined for a building material

surfaces using semi-transparent PV products

The integration of PV into a building can be done by the following

The realization of BIPV projects and applications has been