Bookchapter Fke01

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 Chapter 1 The Role Power Electronics in Future Energy Systems and Green Industrialization Malik Elbuluk and Nik Rumzi Nik Idris Design and Development of a Maximum Power Point Tracking Converter for Building Integrated Photovoltaics Tan Chee Wei, Tim C. Green and Carlos A. Hernandez-Aramburo Analysis of Perturbation Step Size for Perturb and Observe Photovoltaic Maximum Power Point Tracking Algorithm Tan Chee Wei, Tim C. Green and Carlos A. Hernandez-Aramburo Load Sharing Characteristic of Single Phase PV Inverter Connected to Three Phase Grid M. Imran Hamid, Makbul Anwari and Taufik Power Quality Behavior of Single Phase Fed Adjustable Speed Drive from Grid of PV Generation Makbul Anwari, M. Imran Hamid and Taufik
v 1
Chapter 2
21
Chapter 3
40
Chapter 4
58
Chapter 5
71
Contents
Chapter 6
Development of a 5 kW Inverter for AC Power Supply Applications Awang Jusoh, Naziha Ahamd Azli and Zainal Salam Cost Comparison for a Stand-Alone Residential Photovoltaic System Between Silicon and Cadmium in Malaysia A Vigneswaran, Md Shah Majid, Hasimah Abdul Rahman and Mohd Yusri Hassan A Modular Structured Multilevel Inverter for Fuel Cell Applications Naziha Ahmad Azli and Abdul Halim Mohd Yatim
84
Chapter 7
97
Chapter 8
114
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 microturbines, 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
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
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 bidirectional 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.
10
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 voltagecontrolled 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
11
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
12
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
13
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
14
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, nonconventional 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.
15
REFERENCES [1] [2] http://www.eren.doe.gov/der/. 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. IEEE Distributed Resources and Electric Power Systems Interconnection(P1547). http://grouper.ieee.org/groups/scc21/1547/index.html R. Lasseter, Dynamic models for micro-turbines and fuel cells Power Engineering Society Summer Meeting, Vol. 2, pp. 761 766, 2001. 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. M. Suter, Active filter for a micro-turbine, Twenty-Third International Telecommunications Energy Conference INTELEC 2001, pp. 162 165, 2001. 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. G.A. Smith, P.A. Onions, D.G. Infield, Predicting islanding operation of grid connected PV inverters, IEE ProceedingsElectric Power Applications, Vol. 147, No. 1, pp. 16, Jan. 2000. 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.
[3]
[4]
[5]
[6]
[7]
[8]
[9]
16
[10] Huang, S. J., Pa, F.S., Design and operation of gridconnected 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
17
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
Power Electronics Conf. and Exp., APEC 2002, Vol. 2, pp. 1081-1086, 2002. 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. L. E. Lesster, Fuel cell power electronics, www.satcon.com/news/pr/fuelcellarticle.pdf, SatCon Technology Corporation, Cambridge, Massachusetts, USA. 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. 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. R. Hebner, J. Beno and A. Walls, Flywheel batteries come around again, IEEE Spectrum, Vol. 39, No. 4, pp. 46 51, April 2002. 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. 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. 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. A. Abu-Siada, W.W.L. Keerthipala, and W.B. Lawrance,
18
[28]
[29]
[30] [31]
[32] [33]
[34]
[35] [36] [37] [38]
[39]
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. R.Solow, A Contribution to the Theory of Economic Growth." Quarterly Journal of Economics 70 (February 1956), pp. 65-94. R.V. Ayres and A.V. Kneese, Production, Consumption and Externalities, American Economic Review 59(1969): 282297. P. Roomer, "Endogenous Technological Change," Journal of Political Economy 98, October 1990, S71-S102. V.Ruttan, Technology, Growth, and Development: An Induced Innovation Approach, Oxford University Press, 2001. D. North, Institutions, Institutional Change and Economic Performance, Cambridge University Press, 1991. IEEE Standard 519-1992 Recommended practice and requirements for harmonics control in electrical power systems. 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. E. Goodstein, Economics and the Environment, John Wiley, 2002. B. W. Arthur, Increasing returns and path dependence in the economy, University of Michigan Press, 1994. D. Bromley, Environment and Economy: Property Rights and Public Policy, Blackwell, 1991. R. W. Kahn and G. L Hester, Marketable permits: lessons for theory and practice, Ecology Law Quarterly 16, 1989: 380-391. B.A. Larson and M.K. Knudson, Whose price wins:
19
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48] [49]
institutional and technical change in agriculture, Land Economics 67, No.2, 1991: 213-224. 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. H. Lee and N. Darani, Electricity Restructuring and the Environment, Harvard University Policy Group, December 1995. 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. 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. J-M. Martin, Energy technologies: systemic aspects, technological trajectories, and institutional frameworks, Technological Forecasting and Social Change 53, 1996, pp. 81-95. S. Banzhaf, D. Burtraw, and K. Palmer, Efficient Emission Fees in the U.S. Electricity Sector, Resources for the Future Discussion Paper, October 2002. 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. 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. S. Stoft, Power System Economics, Wiley-IEEE Press, 2002. S. A. McCusker, B.F. Hobbs and Y.Ji, Distributed Utility
20
[50]
[51] [52]
[53]
[54]
Planning Using Probabilistic Production Costing and Generalized Benders Decomposition, IEEE Transactions on Power Systems, 17 (May 2002): 497-505. 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. http://www.lumina.com. H. Hashimoto, A spatial Nash equilibrium model in P. Harker Ed. Spatial Price Equilibrium: Advances in Theory, Computation and Application, Springer-Verlag, 1995. 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. 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
22
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
23
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
24
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
400
10
200
10
12
14
16
18
0 20
Time (hour)
Figure 1 Solar irradiance and ambient temperature measured on 8 October 2006 at Imperial College London
Temperature ( C)
25
Solar Irradiance / Time (W/m 2/s)
8 7 6 5 4 3 2 1 0
0.01 0.005 0
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.
Temperature / Time (C/s)
26
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 1 0.8 0.6 0.4 0.2 0 0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350
Day
Energy (kWh/m2/Day)
North
South
West
East
Day
Yearly PV Energy Yield (kWh/m 2)
200 150
Optimum tilt angle Facade
100 50
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
27
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)
28
Start Sense VPV(k) and IPV(k) Calculate
PPV(k) = VPV(k) IPV(k)

Calculate
Ppv_diff = |PPV(k) - PPV(k-1)| Ppv_diff > ?

Yes Yes No
Ppv > PMPP/2 ?

No
Multiply with
= 0.015
Yes Yes
Multiply with
Bypass
= 0.003 Ppv_diff > 0 ?

No Yes
=1
No 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:
29
I = kr
P PV
_ diff
P PV
(2) (3)
PPV _ diff = Ppv(k ) Ppv(k 1)
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 currentreference-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 stepup MPPT converter
30
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.
31
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
32
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.
33
Figure 7 Experimental result during MPPT operation at initial startup
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.
34
(a)
I-V Characteristic
PV Current (A)
3 2 1 0
10
15
20
25
PV Voltage (V)
P-V Characteristic
PV Power (W)
50 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
35
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
PV Current (A)
3 2 1 0
10
15
20
25
PV Voltage (V) PV Power (W)

50 40 30 20 10 0 0 5 10 15 20 25 P-V Characteristic
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
36
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
PV Current (A)
3 2 1 0
10
15
20
25
PV Voltage (V)
P-V Characteristic
PV Power (W)
50 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
37
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. 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. 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,
[2]
[3]
38
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 powerpoint tracking," Energy Conversion, IEEE Transactions on,
39
[13] [14]
[15]
[16]
vol. 17, pp. 514-522, 2002. "Photovoltaics in Buildings A Design Guide," DTI, Report No. ETSU S/P2/00282/REP, 1999. 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. 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. C. Gherasim, T. Croes, J. Van den Keybus, J. Driesen, and R. J. M. Belmans, "Development of a flickermeter for gridconnected wind turbines using a DSP-based prototyping system," Instrumentation and Measurement, IEEE Transactions on, vol. 55, pp. 550-556, 2006.
40
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 stepsizes 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 stepsizes 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)
42
V + I RS V + I R S I = I L I o e nVT 1 RP
(2)
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) Lumped series resistance ( ) RS RP Lumped shunt resistance ( )
Rs
I
+
IL
+ _
ID
Ish 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.
43
(a) I-V characteristic

70 60
25 C 50 C
PV Power (W)
50 40 30 20 10 0 0 5 10
Solar Irradiance Increases
Temperature Increases
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 Maximum power (Pmax) Voltage at Pmax (Vmpp) Current at Pmax (Impp) Short circuit current, (ISC) Open-circuit voltage (VOC) Temperature coefficient of ISC Temperature coefficient of VOC NOCT (Nominal Operating Cell Temperature) Value 65 W 17.6 V 3.69 A 3.99 A 22.1 V (0.065 0.015) %/C -(80 10) mV/C 47 2 C
44
Recent Trends in Renewable Energy and Power Electronics Research 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]. Online 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
45
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 dV dV dV dP at the MPP, =0 dV
(3)
46
dP >0 dV dP <0 dV
to the left of the MPP, to the right of the MPP (4)
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.
47
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.
48
Figure 4 Configuration of the Current-Mode-Controlled-Boost DCDC 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)
49
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:
I = kr
PPV
_ diff
PPV
_ diff
PPV
= P pv ( k ) P pv ( k 1 )
(7) (8)
50
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.
51
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
52
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 stepsize (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.
53
(a) simulated response applying 0.008 fixed step-size
(b) simulated response applying 0.001 fixed step-size
54
(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)]
55
3.6
CONCLUSION
An improved Perturb and Observe MPPT algorithm with a currentmode 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.
56
REFERENCES "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 powerpoint 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
[1]
57
[11]
[12]
[13]
[14]
[15]
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. C. Hua, J. Lin, and C. Shen, "Implementation of a DSPcontrolled photovoltaic system with peak power tracking," Industrial Electronics, IEEE Transactions on, vol. 45, pp. 99-107, 1998. 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. 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. 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. 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.
58
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
60
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
61
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.
62
Central Inverter Configuration
String Inverter Configuration
Multi-string Inverter Configuration
Team concept Configuration
dc ac
dc ac
dc ac
dc ac
dc dc
dc dc dc ac
dc dc
dc ac
dc 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
63
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.
Cr e t ( ) ur n A
30
Upper Band
20
10
Lower Band
0
Inverter Current
-10
-20
-30 0.04
0.042
0.044
0.046
0.048
0.05 0.052 Time (s)
0.054
0.056
0.058
0.06
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.
64
Cre t ( ) ur n A
40 30 20 10 0 -10 -20 -30 -40 0.02
Sinusoidal triangle reference Inverter current
0.022
0.024
0.026
0.028
0.03 0.032 Time (s)
0.034
0.036
0.038
0.04
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.
65
A a
b B
c C
Three Phase Grid
Iref
Plant A
Current Control T(oC) 50 Ga 800
Gate In3 Sinus
Iref
Product2
Out1
PLL
+ In1 v -
Ga + T
i + A A
Breaker3
A DC Out ( + ) DC Out ( - ) B B B
PV Array A
Inverter
Filter
Iref1 Iref
Transformer
T(oC)1 50 Ga1 800
Plant B
Current Control1
Sinus Gate In3
Product1
PLL1
+ Out1 In1 v -
Ga + T DC Out ( + ) DC Out ( - ) -
g +
i A A
2 1
Breaker1
PV Array B
LCL Filter1
Iref2 Iref
Plant C
T(oC)2 50 Ga2 800
Current Control2
Sinus Gate In3
Product3
PLL2
+ Out1 In1 v -
Ga + T DC Out ( + ) DC Out ( - ) -
i + A A
Breaker2
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.
66
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.
67
25 20 15 10 Current ( Amp ) 5 0 -5 -10 -15 -20
Grid Current
PV Current
Load Current
0.01
0.02
0.03
0.04
0.05 Time (s)
0.06
0.07
0.08
0.09
0.1
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
68
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 0.05 0.1 0.15 0.2 Time (Sec) 0.25 0.3 0.35
400 200 V o lt a g e (V o lt ) 0 -200 -400
0.05
0.1
0.15 0.2 Time (sec)
0.25
0.3
0.35
80 Current (Am p) 60 40 20 0 -20 -40 -60 -80 -100 -120
IL=Ipv +Igr Ipv
Igr
t1
t2
t3
t4
0.2 0.25 0.3 0.35
0.05
0.1
0.15
Figure 8 Load current sharing between grid and plant on phase a when plant switched
69
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-T116:2007. [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
70
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 VSIPWM 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
72
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
73
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
74
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
75
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
76
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:
77
Ta
Ga
Gate
Ga
A
+
a aA
bB
DC Out ( + )
B -
B c cC
DC Out ( - )
C
C n
node 10
Solar Array
PV Inverter
Filter
Transformer
Grid Load
Speed Control
+
C
A B
Load
B Tm A C
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.
78
Motor Speed Rpm
1500
1000
500
0 1 1.2 1.4 1.6 1.8 2
ASD input Current Current (A)
30 20 10 0 -10 -20 -30 0.9
1.1
1.2
1.3
1.4
1.5
1.6
Time (sec)
Grid Voltage During ASD Starting
2000 1000 0 -1000 -2000 0.95 1 1.05 1.1 1.15 1.2
Voltage (Volt)
Grid Current During ASD Startin
20 10
Current (A)
0 -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
79
loads are from non linear type such as rectifier.

0.1
0.05
C urrent (Am p)
-0.05
-0.1 0.95
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
80
25 T HD V 20 T HD I
THD (% of Fund)
15 10 5 0 0.5 0.8 1 Speed (pu) 1.2 1.5
30 THD V 25 THD I
THD (% of Fund)
20 15 10 5 0 0.25 0.5 0.75 Loa d (pu) 1 1.25
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.
81
800 750
DC Voltage (Volt)
700 650 600
G= 800 W.m-2
550 500
G= 500 W.m-2
G= 900 W.m-2
0.5
1.5
2.5 Time (Sec)
3.5
4.5
80 70 60
DC Current (Amp)
50 40 30 20 10 0.5 1 1.5 2 2.5 Time (Sec) 3 3.5 4 4.5 5
Figure 7 DC output of pv array when irradiance is adjusted
2000 1000 0 -1000 -2000
V oltage (V olt)
0.5
1.5
2.5
3.5
1500
Speed (Rpm )
1000
500
0.5
1.5
2 Time (Sec)
2.5
3.5
Figure 8 ASD Input voltage and motor speed when irradiance is adjusted as Fig. 7.
82
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.
83
[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.
84
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
86
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
87
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.
L S1
V dc
S3
vPWM
vo
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 50Hz Square PWM LEG 1
Gate Drive
S1 S2
PWM LEG 2
Gate Drive
S3 S4
Figure 2 Control signals and gates
Figure 3 Control and PWM signals
88
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
89
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 Schmitttrigger 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.
90
Figure 6 Dead time circuit
Input Pin 3 7408 Vt Pin 9 7414
Pin 8 7414
+15V Output VGS -15V Deat time,td
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.
91
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
where k = 1 ..
(1)
mf 2
By using (1), the switching angle, i.e. the rising and falling
92
edge of the kth pulse can be calculated as: Rising edge, 1k = k
(2) (3)
Vm(t) Vc(t)
Falling edge, 2 k = k +
k
2
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-C167CSLM (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
93
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).
94
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
95
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
96
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 STANDALONE 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 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
2
98
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.
99
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.
100
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
101
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 (USD) 3.71 3.31 Price per Watt Peak (RM) 12.50 11.16
Amorphous Silicon Cadmium Telluride (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. ii. iii. 9.9 kWp installation at commercial shop lots at Lot 53, Jalan SS 21/37, Damansara Utama, Petaling Jaya, Selangor[7]. 11.88 kWp installation at Putrajaya Perdana Berhad Headquarters, Precinct 16, Putrajaya [8]. 4.0 kWp BIPV System Installation at Lot 488,
102
iv.
Jalan Khaya, Country Heights Damansara, Petaling Jaya[9]. 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
103
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
104
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 Amorphous silicon High efficiency Low efficiency Cadmium Telluride High efficiency Low efficiency Product Model Uni-solar, US-116 BP, MST 50 MV BP, Apollo 980 First Solar Market Efficiency 6.20% 6.10% 8.60% 6.90% Power Rating (W) 116 50 80 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.
105
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.
106
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.
107
Table 7.3: Total net present value for low and high efficiency amorphous silicon
Type of Amorphous Silicon Module Low efficiency, Uni-solar, US-116 High efficiency, NT-R5E3H Total Net Present Value RM 209,409 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);
108
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.
109
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.
110
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,
111
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) (2) (3) (4) (5) (6) (7) United Nation Development Program, Malaysia, 2004, Building Integrated Photovoltaic Technical Application Project, TNB Research Sdn Bhd. Malaysia, Germany Chamber of Commerce and /industry, Quarterly Report 11-12-07, 2007 Olivia Mah, 1998, Fundamental of Photovoltaic Materials, National Solar Power Research Institute Sandia Natinal Laboratories, 1998, Standalone PV Systems. A Handbook of Recommended Design Practices, Albuquerque, New Mexico, USA Solarbuzz, 2008a, Solar Module Price Highlight, www.solarbuzz.com/solarmoduleprices.htm, 22 February 2008 Bo-Varga, 2006, Nanotechnology Impact on Solar Power Silicon Valley Solar Ventures 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 Azah Ahmad, Vincent Tan, Chen Wei Nee, Daniel Ruoss, 2007d, 11.88 kWp BIPV System Installation at Putrajaya Perdana Berhad HQ, P16, Putrajaya, Malaysia Building Integrated Photovoltaic Project Azah Ahmad, Vincent Tan, Chen Wei Nee, Daniel Ruoss, 2007b, 4.0 kWp BIPV System Installation at Lot 488, Jalan Khaya, Country Heights Damansara, Petaling Jaya, Malaysia Building Integrated Photovoltaic Project Azah Ahmad, Vincent Tan, Chen Wei Nee, Daniel Ruoss,
(8)
(9)
(10)
112
(11) (12) (13) (14) (15) (16)
(17) (18)
(19) (20) (21)
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 Solarbuzz, 2008c, Battery Price Highlights, www.solarbuzz.com/batteryprice.htm, 22 February 2008 Solarbuzz, 2008b, Inverter Price Highlight, www.solarbuzz.com/inverter.htm, 22 February 2008 Solarbuzz, 2008b, Charge controller Price Highlight, www.solarbuzz.com/chargecontrollerprices.htm, 22 February 2008 Solarcentury, 2006, Solar energy Factsheets. www.solarcentury.co.uk/knowledge_base/search, 23 February 2008 D.A.S. Rand, P.T. Moseley, J. Garche, C D Parker, 2004, Valve Regulated Lead-Acid Battery, pp 1-10, Elsevier BV 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 R.W. Larson et. al 1992, Economics of Solar energy Technologies, American Solar Energy Society (ASES) 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 Charles Iskander and Edward Scerri, 1996, Performance and Cost Evaluation of a Stand-Alone PV in Malta, Institute of Energy Technology, University of Malta F. Lasnier and T. G. Ang, 1990. Photovoltaic Engineering Handbook, CRC Press, USA Hammon R, D. Srinivasan, A Harris, K Whitefield, J Wohlgemuth, 1997, Effects of Soiling on PV Module and Radiometer
113
Performance, Pp 1121-1124. Conference Record of the 26 IEEE Photovoltaic Specialists Conference
114
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 nonlinearity 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
116
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.
117
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
118
N and M. M = (N-1)/2 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. (1)
Figure 2 Functional block diagram of the fuel cell power conditioner system
119
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.
120
1.5
m1(t) c(t)
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)
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
121
For example, for a 5 level modular structured inverter, 2 modules is required, thus the modulation signals are m1(t) = | Am sin(2fm)t| m2(t) = | Am sin(2fm)t| - 1 (2)
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 ma = Am /(M.Ac ) (3) (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
122
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.
c (t)/m 1(t)
0.5
0 0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
p1(t)
-1
0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
g11
-1
0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
g21
-1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02 Time(s)
123
c (t)/m 2(t)
-1 0 1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
pw2(t)
-1 0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
g12
-1
0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
g22
-1
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
c (t)/m 1(t)
0 0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
p1(t)
-1
0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
g11
-1
0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
g21
-1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02 Time(s)
124
c (t)/m 2(t)
-1 0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
p2(t)
-1
0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
g12
-1
0
1
0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02
g22
-1
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.
125
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
Norm aliz ed fourier coeff.
1.4
1.2
0.8
0.6
0.4
0.2
0 0
10
15
20
25 30 35 Harmonic order
40
45
50
55
Figure 6 5 level modular structured inverter (a) output voltage (b) frequency spectrum for ma = 0.4.
126
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. 19201926, 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
127
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 nonconventional 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. 497505, 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, 114116, 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

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Bookchapter Fke01

Hochgeladen von

Copyright:

Verfügbare Formate

First Edition 2008

MAKBUL ANWARI 2008

Diterbitkan di Malaysia oleh / Published in Malaysia by

Makbul Anwari Faculty of Electrical Engineering Universiti Teknologi Malaysia 2008

FUTURE ENERGY SYSTEMS

Recent Trends in Renewable Energy and Power Electronics Research

The role power electronics in future energy systems

Recent Trends in Renewable Energy and Power Electronics Research

Figure 1. Power electronic converters for micro-turbine technology

The role power electronics in future energy systems

Figure 2. Power electronic converters for micro-turbine technology

Recent Trends in Renewable Energy and Power Electronics Research

Figure 3. Power electronic converters for fuel cells system

Figure 4. Singly fed induction generator wind turbine

The role power electronics in future energy systems

Recent Trends in Renewable Energy and Power Electronics Research

ENERGY STORAGE DEVICES

POWER FLOW AND CONTROL

The role power electronics in future energy systems

Recent Trends in Renewable Energy and Power Electronics Research

The role power electronics in future energy systems

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

Recent Trends in Renewable Energy and Power Electronics Research

The role power electronics in future energy systems

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

Recent Trends in Renewable Energy and Power Electronics Research

The role power electronics in future energy systems

Recent Trends in Renewable Energy and Power Electronics Research

The role power electronics in future energy systems

Recent Trends in Renewable Energy and Power Electronics Research

[35] [36] [37] [38]

The role power electronics in future energy systems

Recent Trends in Renewable Energy and Power Electronics Research

Design and Development of a Maximum Power Point Tracking Converter

Recent Trends in Renewable Energy and Power Electronics Research

Design and Development of a Maximum Power Point Tracking Converter

Recent Trends in Renewable Energy and Power Electronics Research

Solar Irradiance (W/m2)

Design and Development of a Maximum Power Point Tracking Converter

Solar Irradiance / Time (W/m 2/s)

Figure 2 Derivative of solar irradiance and temperature from Fig. 1

PV ORIENTATION AND TILT ANGLE

Temperature / Time (C/s)

Recent Trends in Renewable Energy and Power Electronics Research

Yearly PV Energy Yield (kWh/m 2)

Optimum tilt angle Facade

Design and Development of a Maximum Power Point Tracking Converter

CURRENT-MODE CONTROLLED MPPT CONVERTER

Recent Trends in Renewable Energy and Power Electronics Research

Start Sense VPV(k) and IPV(k) Calculate

PPV(k) = VPV(k) IPV(k)

Ppv_diff = |PPV(k) - PPV(k-1)| Ppv_diff > ?

Ppv > PMPP/2 ?

= 0.003 Ppv_diff > 0 ?

VPV(k) - VPV(k-1) > 0 ?

VPV(k) - VPV(k-1) > 0 ?

I = kr (Ppv_diff / Ppv) Iref(k) = Iref(k-1) + (I)

Design and Development of a Maximum Power Point Tracking Converter

PPV _ diff = Ppv(k ) Ppv(k 1)

Figure 5 Configuration of the Current-Mode Controlled DC-DC stepup MPPT converter

Recent Trends in Renewable Energy and Power Electronics Research

MPPT SYSTEM OVERVIEW

Design and Development of a Maximum Power Point Tracking Converter