This action might not be possible to undo. Are you sure you want to continue?
THESIS
Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in
the Graduate School of The Ohio State University
By
Eyup Taymur
Graduate Program in Electrical and Computer Science
The Ohio State University
2009
Master's Examination Committee:
Professor Ali Keyhani, Advisor
Professor Charles A. Klein
Copyright by
Eyup Taymur
2009
Abstract
World energy consumption and the resulting CO
2
emissions are increasing
substantially and this increase puts in danger the ecological stability of our Earth.
Growing scarcity and rising prices of fossil fuels may lead to economical and
political instability in many countries in the near future. These problems can be
solved by contributing significantly the use of renewable energy resources. The
renewable energy resources are sufficient enough to meet the world energy
requirement. Most of the countries have recognized the new energy policy to
encourage the investment photovoltaic energy system which is one of the biggest
renewable energy resources, the government policies are providing credits,
reducing taxes or applying feedin tariff which is agreement of energy purchase
for certain period of time between the government and the investors. In this thesis,
it is desired to have a comprehensive study on energy output calculation of the
solar system. The solar radiation effect is articulated and based on monthly
average global irradiation data the daily average irradiation is found on an
inclined surface; the sun rotation relative to the Earth is analyzed in order to
calculate the optimum angle in any latitude on the Earth surface to receive the
maximum annual irradiation. Based on the given data a graphical user interface is
modeled by MATLAB GUI which calculate the energy yield by using data set for
different PV module technologies to be presented to the endusers for decision
making in PV energy investment. Based on power system and power electronic
knowledge different design proposals are modeled by using component
specification provided by the manufacturers.
iii
Dedication
This document is dedicated to my family.
iv
Acknowledgments
First of all I would like to thank to my parents and my wife for their support
during completion of my graduate study. I also would like to thank my thesis
exam committee: Professor Ali Keyhani and Professor Charles A. Klein for
giving me the opportunity to present this thesis and for their time, patience and
understanding. Special thanks to my advisor Professor Ali Keyhani for his great
support and inspiration to be able to understand Photovoltaics System, design and
application. Professor Keyhani, it has been an honor to work with you. I would
like to thank all the professors in Power area for sharing their deep knowledge and
experience with me.
v
Vita
January 1982 ..................................................Born in Batman, Turkey
2007................................................................B.S. Electrical and Electronic
Engineering, Gaziantep University
Fields of Study
Major Field: Electrical and Computer Engineering
vi
Table of Contents
Abstract .................................................................................................................. iii
Dedication .............................................................................................................. iv
Acknowledgments................................................................................................... v
Vita ......................................................................................................................... vi
Fields of Study ....................................................................................................... vi
Table of Contents .................................................................................................. vii
List of Tables ........................................................................................................ xii
List of Figures ...................................................................................................... xiv
Chapter 1: Solar Radiation ...................................................................................... 1
1.1 Angle Definitions ..................................................................................... 3
1.2 Some Astronomical Information ................................................................... 5
1.3 Calculation of the Irradiation on an Inclined Surface ................................... 7
1.3.1 PV House Example ............................................................................... 13
Chapter2: Power Electronics ................................................................................. 15
2.1 Introduction ................................................................................................. 15
2.2.1 Stepdown converter ............................................................................. 17
2.2.2 Stepup converter .................................................................................. 18
2.2.3 Buckboost converter ............................................................................ 20
2.2.4 Inverters ................................................................................................ 21
Chapter3. Design Overview of StandAlone PV Systems .................................... 33
vii
3.1. PV generators, Batteries and Loads Coupling ........................................... 33
3.2. Calculating electricity consumption ........................................................... 35
3.3. PV generator sizing .................................................................................... 36
3.3.1. Calculation of the PV generator’s yield .............................................. 36
3.4 Consideration of Cable and Conversion Losses .......................................... 39
3.4.1 Cable losses .......................................................................................... 40
3.4.2 Conversion losses ................................................................................. 40
3.5 Mismatch losses .......................................................................................... 40
3.6 Determination of Summer Excess and Winter Reserve .............................. 42
3.7 A PV House Design Example ..................................................................... 43
3.7 Cable crosssection sizing ........................................................................... 44
3.8 Sizing the battery ......................................................................................... 45
3.9 Charge Controllers ...................................................................................... 45
3.9.1 Series controllers .................................................................................. 47
3.9.2 Shunt controllers ................................................................................... 48
3.9.3 MPPT charge controllers ...................................................................... 49
3.10 Standalone PV operation principles ......................................................... 50
3.10.1 Sizing Example of StandAlone System ............................................ 51
Chapter 4: Design and Installation of Large Scale PV Systems ........................... 53
4.1 Introduction ................................................................................................. 53
4.2 GridTied PV System .................................................................................. 54
4.2.1 GridTied Inverter ................................................................................. 55
4.2.2 GridTied Inverters and PV Module Configuration ............................. 57
4.2.3 Selecting Inverter .................................................................................. 58
viii
4.3 Design Considerations and 2 MW System Installations ............................. 62
4.3.1 Module Selection .................................................................................. 62
4.3.2 Inverter Selection .................................................................................. 64
4.3.3 First Proposal of PV system ................................................................. 66
4.3.4 Description of AC Transmission Side and AC Line Model ................. 71
4.3.5 Sizing the DC Main Cable from the Array ........................................... 73
4.3.6 Second design proposal of 2.16MW system ........................................ 75
Chapter 5: Annual Energy Yield Calculation ....................................................... 80
5.1 Temperature Dependency of Efficiency ..................................................... 81
5.2 Energy Production ....................................................................................... 88
5.2.1 Cumulative solar irradiance .................................................................. 88
5.2.2 Module power rating at standard test condition ................................... 88
5.2.3 Operating temperature .......................................................................... 89
5.2.4 Maximum power point voltage dependency of irradiation .................. 89
5.2.5 Soiling ................................................................................................... 89
5.2.6 Variation of Solar Spectrum ................................................................. 90
5.2.7 Peak Solar Hours Concept and Definitions .......................................... 90
5.2.8 Temperature Dependency of Array Output .......................................... 91
5.2.9 Module orientation ............................................................................... 92
5.2.10 Mismatch Losses and Blocking/Bypass Diodes ................................. 92
5.2.11 Inverter Efficiency .............................................................................. 95
5.2.12 Cable Losses and Transmission Losses .............................................. 97
Chapter 6: Solar Cell Types and Data Sheets ..................................................... 100
6.1 Technologies ............................................................................................. 100
ix
6.1.1 Crystalline silicon cell (market share 93%) [10] ............................. 101
6.1.2 Thin film solar cell ............................................................................. 101
6.2 Different technologies and their specifications ......................................... 103
Chapter 7: Economic Evaluation and Cost Estimation Techniques ................... 104
7.1 Introduction ............................................................................................... 104
7.2 Presentworth Factor ................................................................................. 105
7.3 The Major Cost and Performance Elements of the Utility Scale PV Power
Plant ................................................................................................................. 107
7.3.1 The Performance of PV Plant ............................................................. 107
7.3.2 Initial PV Plant Investment................................................................. 109
7.3.3 PV Operation Expenses ...................................................................... 112
Chapter8: PV Sizing Simulator ........................................................................... 114
8.1 Introduction ............................................................................................... 114
8.2 Simulator Inputs and Component Prices ................................................... 114
8.3.1 DCAC Derates Factor ....................................................................... 118
8.4 Case Study for Monocrystalline............................................................... 120
8.4.1 1 km Transmission Line ..................................................................... 120
8.4.2 25km Long Transmission line Losses ................................................ 123
8.5 Case Study for PolyCrystalline ................................................................ 125
8.5.1 1km Transmission Line ...................................................................... 125
8.5.2 25km Long Transmission Line ........................................................... 127
8.6 Case Study for ThinFilm .......................................................................... 129
8.6.1 1km Transmission Line ...................................................................... 129
8.6.2 25km Long Transmission Line ........................................................... 131
x
8.7 Comparison of Case Studies ..................................................................... 133
Conclusion .......................................................................................................... 136
References ........................................................................................................... 137
Appendix ............................................................................................................. 140
xi
List of Tables
Table 1: Typical Reflectivity .................................................................................. 9
Table 2: Irradiation on Different Inclination Angle .............................................. 12
Table 3: The Optimum Angle and Maximum Irradiation for One Year ............... 13
Table 4:Average Yearly Irradiation Data ............................................................. 13
Table 5: InputOutput Relation of DCDC Converter with Duty Ratio Range .... 32
Table 6: Seasonal Electricity Usage of Equipment [19] ....................................... 36
Table 7: Z2 Factor Calculated for Different Locations [19] ................................. 37
Table 8: Z3 Factor for Deviation from Horizontal [19] ........................................ 38
Table 9: Temperature Correction Factor Z4 [19] ................................................. 39
Table 10: Design Parameters for StandAlone PV Systems ................................. 41
Table 11: Electrical Parameters and their Units ................................................... 44
Table 12: Characteristics of AS300 Module ....................................................... 63
Table 13: Characteristics of PVPowered Inverter ............................................... 64
Table 14: 13.2–132 kV Class One phase – Neutral Return Line Model .............. 72
Table 15: Satcon PV Inverters PV135 ................................................................. 75
Table 16: PV Cell Efficiencies in Different Technologies ................................. 102
Table 17: PV Module Specifications .................................................................. 103
Table 18: Payment Schedule ............................................................................... 105
xii
Table 19: Module Types and Specifications ....................................................... 115
xiii
List of Figures
Figure 1: Global irradiation values for the world (kwh/m2) [16] ........................... 2
Figure 2: World global irradiation source: www.bpsolar.com ............................... 2
Figure 3: The PV panel inclination angle on the Earth’s surface under sunshine .. 3
Figure 4: Values of Airmass calculated for specific location [26] ........................ 5
Figure 5: Sun and Earth rotation and angles description ........................................ 6
Figure 6: Irradiation components [12] .................................................................... 8
Figure 7: Semiconductor devices developments [20] ........................................... 16
Figure 8: Buck Converter Circuit ......................................................................... 17
Figure 9: Boost Converter Circuit ......................................................................... 19
Figure 10: Buckboost Converter Circuits ............................................................ 20
Figure 11: Three Phase Inverter Topology [21] ................................................... 21
Figure 12: Wave Form of PWM Sinusoidal [21].................................................. 23
Figure 13: Base Switching Vectors ....................................................................... 25
Figure 14: Determination of the Region ............................................................... 26
Figure 15: Vector Projection for Region Determination ...................................... 27
Figure 16: Vector Allocation for the Dwelling Times T
a
,T
b
and T
c
[23] .............. 29
Figure 17: Battery Charging and Discharging Conditions ................................... 34
Figure 18: Series Charge Controller ..................................................................... 47
xiv
Figure 19: Shunt Charge Controller ...................................................................... 48
Figure 20: MPPT Controller ................................................................................. 49
Figure 21: StandAlone PV Operation [1] ............................................................ 50
Figure 22: Application of PV systems [12] .......................................................... 54
Figure 23: General Structure ................................................................................. 55
Figure 24: Maximum Power Point under Different Irradiation [27] .................... 56
Figure 25: Basic Configuration for the given Example Showing the Modules,
Strings and an Array ............................................................................................. 58
Figure 26: Central Inverter with Large Scale Array Configuration ...................... 59
Figure 27: Efficiency of the Inverter under Different Irradiation and Durations [4]
............................................................................................................................... 61
Figure 28: Configuration of the Array (112 strings with 8 modules) with the
Central Inverter ..................................................................................................... 68
Figure 29: General Structure of the Arrays with the Inverters ............................. 69
Figure 30: Network Model of 2.15MW System, Single Line Diagram................ 70
Figure 31: 13.2132kV Class One PhaseNeutral line Model .............................. 72
Figure 32: Parameters for Sizing DC Main Cable [12] ........................................ 73
Figure 33: SubArray Configuration ..................................................................... 77
Figure 34: 2.16 MW PV System Line Diagram (Complete System) .................. 79
Figure 35: Shows the Error by Using the Equation (1) Calculate The PV Output in
Different Location In The World. Different Places with Their Latitude and the
Associated Error is Illustrated ............................................................................... 91
xv
Figure 36: Bypass and Blocking Diodes in PV Array ......................................... 93
Figure 37: Shading Effect on IV Characteristics of PV [16] ............................... 94
Figure 38: Inverter Efficiency [16] ....................................................................... 96
Figure 39: PV Cell Types [12] ............................................................................ 100
Figure 40: PV Panel under 20 Year Exposure of Sun [18] ................................. 108
Figure 41: Relative Solar Cell Conversion Efficiencies [18] ............................. 109
Figure 42: Area related cost [18] ........................................................................ 110
Figure 43: Land Consumption versus Capacity Factor in Different Technologies
[18] ...................................................................................................................... 111
Figure 44: O&M Cost with Different Technologies [18] ................................... 113
Figure 45: Location of Batman City with Annual Irradiation per metersquare [25]
............................................................................................................................. 117
Figure 46: Irradiation Data and Sunshine Duration for Batman City [24] ......... 117
Figure 47: Shading Effect [9] ............................................................................. 120
Figure 48: Monocrystalline for 1km Transmission Line ................................... 122
Figure 49: MonoCrystalline for 25km Transmission Line ................................ 124
Figure 50: PolyCrystalline for 1km Transmission Line ................................... 126
Figure 51: PolyCrystalline for 25km Transmission Line .................................. 128
Figure 52: ThinFilm for 1km Transmission Line .............................................. 130
Figure 53: ThinFilm 25km Transmission Line.................................................. 132
Figure 54: Area Comparison of Different Technologies .................................... 133
Figure 55: Cost Comparison ............................................................................... 134
xvi
xvii
Figure 56: Payback Time Comparison ............................................................... 135
Chapter 1: Solar Radiation
Nomenclature:
KT: the clearness index of the sky
AM: air mass factor which determines the total distance where the sunlight has to
travel to reach the Earth surface
S: the resulting coverage energy flux incident on a unit area perpendicular to the
beam outside the Earth’s atmosphere
The energy comes from the sun to the Earth is in the form of Radiation. The
amount of energy in the sunlight reaching the Earth surface is equivalent to
around 10,000 times of the world’s energy requirements. [12]
1
Figure 1: Global irradiation values for the world (kwh/m2) [16]
Figure 2: World global irradiation source: www.bpsolar.com
2
1.1 Angle Definitions
In order to make a good estimation for energy yield of photovoltaic systems, the
angle definitions and solar radiation should be understood perfectly. The total
irradiations which are provided in different official sources are the irradiation of
the certain flat surface on the world. When sun moves around the angle of the sun
reaches the Earth will change every second of the motion. In this case the
radiation will also change depending on the sun motion.
Figure 3: The PV panel inclination angle on the Earth’s surface under sunshine
αs= solar azimuth angle
α= azimuth angle of the PV generator
3
β= tilt angle of PV generator
γs= solar elevation from the horizon
Due south is given by 0 degree, east is 90 degree and west is +90 degree. Solar
altitude is very important parameter to estimate the real irradiation on a horizontal
surface, because the intensity of the solar radiation on any surface is dependent on
the solar elevation angle. While the Earth is rotating the elevation angle of the sun
changes every month in the year and every day in the months and every hour in
the day. The air mass factor determines the total distance where the sunlight has
to travel to reach the Earth surface.
AH = 1¡sin(ys)
gives the relation between air mass and elevation angle.
The resulting coverage energy flux incident on a unit area perpendicular to the
beam outside the Earth’s atmosphere is known as a solar constant,
S= 1S67 w¡m
2
[11]. As it is known the total power falling on a unit area from
radiant source is called irradiance. When the solar radiation travels through the
atmosphere to reach the Earth’s surface, a part of incident energy will be lost by
scattering or absorption by air molecules, clouds and particles which are called
aerosols.
Direct (beam) radiation is the radiation which is not reflected or scattered and
reaches directly to the surface is called diffuse radiation. The scattered radiation
reaching to the ground surface is called diffuse radiation. The radiation which is
the reflection of the radiation reaches to the ground surface and being received by
the objects is called albedo. The sum these radiations (diffuse, direct, albedo) is
called global radiation. The air mass specifies the clearness of the atmosphere.
The integration of the irradiance over a period of time gives us the irradiation.
4
Figure 4: Values of Airmass calculated for specific location [26]
1.2 Some Astronomical Information
The Earth revolves in an elliptical orbit around the sun; the plane of this orbit is
elliptical. The Earth completes one cycle travelling around the sun in one year.
The angle between the equatorial plane and celestial plane is 0 = 2S.4S° .
The celestial sphere around the Earth shows the relative position of the sun and
Earth. The angle between line joining the centers of the sun and the Earth and the
equatorial plane is called declination angle o. This angle will be zero at 20/21
march (vernal) and at 22/23 September (autumnal) equinoxes. On these days sun
will rise exactly in the east and will set in the west. At summer solstice, 21/22
June, o will be 23.45° and at winter solstice will be 23.45°.
5
S
W
N
E
Horizon
Equator Plane
Sun’s Daily path
Solar Noon
Zenith
φ
α
δ
Nadir
Figure 5: Sun and Earth rotation and angles description
The Earth rotates about the polar axis in the rate of one revolution per day. The
instantaneous position of the sun on the world can be described by ‘’æs'', sun
angle. Solar angle is the angle between the meridian passing through the sun and
the meridian of the area. At solar noon hour angle is zero, and increases toward
the east.
α in the figure shows the elevation angle and φ is the latitude for a given
geographical location on the Earth’s surface. During the daily motion, the solar
declination o is assumed to be constant and equal to its midday’s value. The
following geographical relationship can be used to formulize the sun position and
relate them to a i the irr d ation on an inclined surface [11].
Sino 0z = sino . sinç + coso . cosç. cosw = cos
Cosç = (sino. sinç sino)¡ (coso. cosç)
6
From these two equations it is found that the solar hour angle can be found by the
formula [11].
æs = cos
1
tanφ.tano) (
Where æs is sunrise hour angle and – æs is sunset hour angle for the day of the
year.
1.3 Calculation of the Irradiation on an Inclined Surface
The solar radiation data are generally given in the form of global radiation on a
horizontal surface. Global daily irradiation is denoted by G. If the PV panels are
positioned with an angle on a horizontal surface the total global irradiation
received by the PV will change.
7
Figure 6: Irradiation components [12]
The data G for the site is used to determine the diffuse and beam contributions to
the global irradiation by using Bo as a reference and KT is clearness index. The
attenuation of the solar radiation through the atmosphere at a given area on the
Earth can be described by KT=G/Bo in that month. When Bo is calculated the
variation of extraterrestrial irradiance on account of the eccentricity of the Earth’s
orbit is considered. The diffuse irradiation can be obtained by using the diffuse
fraction index D/G of the global irradiation. D/G is the universal function of
clearness index KT. B=GD gives the separate beam and diffuse radiation.
8
Finally, angular dependent of each component will give the diffuse and beam
irradiation on an inclined surface. And also by using the reflectivity of
surrounding area the total albedo is calculated. The total is found by adding
albedo, beam and diffuse irradiation.
The inputs for calculations are daily global irradiation G for a day, the middle of
each month can be chosen to give the average monthly irradiation and the solar
constant S, which is 1367w/m
2
, the site’s geographical latitude and the solar
declination angle o for the day of the year.
Table 1: Typical Reflectivity
GROUND COVER REFLECTIVITY
Dry bare ground 0.2
Dry grassland 0.3
Desert sand 0.4
Snow 0.50.8
Bo is the irradiation received over one day by a unit horizontal surface area
outside t s o p [11]. of he Earth’ atm s here is calculated using the expression below
Bo = 24¡pi. S. {1 +uSS. cos(2. pi. Jn¡S6S)]. {cosç. Coso. sinws
+ws. sinç. sino]
where d
n
is the number of day starting from January 1 d
n
=1 and December 31
d
n
=365.
Since we have the daily irradiation data G from KT=G/Bo the clearness index can
be found. For the disuse irradiation D, d from the formula [11] it can be foun
Ð¡0 = 1 1.1S. KI
9
The beam irradiation also can be found simply by subtracting D from G, B=GD.
After B is found for the horizontal surface, the beam irradiation B(β) on a south
facing panel inclined at an angle β to the horizontal surface is given by [28].
B([) = B.
cos(ø [) . coso. sinæs(ø [)] +æs(ø [). sin (ø [)sino
cosø . coso. sinæs +æs. sinø. sino
Ð([) =
1
. Ð. (1 +cos ([))
2
R([) =
1
2
. B. p. (1 cos[)
Where D([) is the diffuse and R(β) is the albedo radiation. Finally the global
irradiation can be found b h ponents. y summing up all the t ree com
0([) = B([) +Ð([) +R([)
Since ws (hour angle) is used for one day inclination angle is required and the
inclination angle δ can b f o 3]. e ound by the foll wing formula [1
o = 2S.4S. sin (S6u. (284 +n)¡S6S)
10
The overall formulas to put in the program for calculation of irradiation on an
inclined surface will be as following:
1. o = 2S.4S. sin (S6u. (284 +n)¡S6S)
Sol l io for the given d ar dec inat n angle is found ay n.
2. æs = cos
1
(tonø. tono) and æs' = cos
1
(tan (ø [). tono)
and where æs is the solar declination angle and φ is the geographical latitude
of the location on the Earth.
Then we find æo= min{ ws, ws’ }. Since the Earth rotates one revolution per
day, this corresponds to 360 longitude makes 24 hours. This means the day length
is 2.ws/15.
3. Bo = 24¡pi. S. {1 +uSS. cos(2. pi. Jn¡S6S)]. {cosç. Coso. sinws +
ws. sinç. sino]
Where n is the day of the year.
4. KI = 0¡Bo KT is the clearness index, and G is the solar irradiation data
of e r th a ea.
5 0. 1.1SKI). . Ð = (1 
6. B Ð. = 0
7. B([) = B.
cos(ø[).cos6.sIn(mo)+os.sIn (ø[)sìn6
cosø .cos6.sìnos+o .sìnø.sìn6 s
Ð([) =
1
. Ð. (1 +cos ([))
2
R([) =
1
. B. p. (1 cos[)
2
11
0([) = B([) +Ð([) +R([)
The average irradiation of Batman city, Turkey for each month is given below
G=[1900,2690,4070,5050,6240,7040,6840,6040,5270,3730,2410,1800]
Table 2: Irradiation on Different Inclination Angle
Inclinati
on
Angle
Jan
Feb
Mar
Apr
May
June
July
Aug
Sept
Oct
Nov
Dec
Ave.
year
wh/m
0 1900 2690 4070 5050 6240 7040 6840 6040 5270 3730 2410 1800 4233
6 2054 2863 4258 5164 6261 6973 6809 6137 5510 3992 2616 1965 4550
12 2194 3016 4414 5237 6231 6851 6721 6180 5699 4221 2803 2117 4640
18 2318 3146 4533 5266 6150 6671 6574 6166 5835 4414 2968 2252 4691
24 2425 3253 4616 5252 6016 6433 6368 6095 5915 4570 3110 2371 4702
30 2513 3334 4661 5194 5830 6137 6104 5967 5940 4686 3226 2472 4672
36 2582 3390 4667 5093 5595 5787 5784 5784 5910 4760 3316 2553 4602
42 2631 3419 4636 4951 5312 5387 5411 5546 5823 4794 3379 2613 4492
54 2666 3396 4459 4546 4618 4451 4526 4922 5487 4734 3420 2671 4158
60 2653 3344 4316 4288 4214 3928 4024 4542 5241 4641 3398 2667 3938
72 2563 3164 3929 3678 3320 2807 2935 3671 4608 4337 3270 2595 3406
84 2393 2887 3421 2965 2356 1653 1795 2689 3810 3885 3035 2440 2778
90 2281 2716 3129 2582 1870 1099 1237 2175 3361 3609 2880 2333 2439
In the given table the inclination angle of the PV panel is betha is changing from 0
degree to 90 degree. It is shown that the maximum irradiation as a total of twelve
months is at 24 degree of inclination angle which is 4702 Wh/m
2
. This amount is
almost 3343 Wh/m
2
greater than the irradiation on a horizontal surface. If the
angle of inclination can be changed every month of the year the yield will be
almost 4930.3 Wh/m
2
which is very substantial increase of irradiation on PVs.
12
Table 3: The Optimum Angle and Maximum Irradiation for One Year
Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec
Total
Ave.
2667 3424 4669 5267 6261 7014 6840 6181 5941 4795 3421 2672
4930.3
Wh/m2
54 44 33 18 6 3 0 12 30 45 51 57 β(opt.)
As it is seen in the Table 3 the total irradiation falling on the surface at the
specified angles will be more than when they are at an only one fixed angle. In
later steps, the total revenue will be discussed to analyze if different technology is
applied to change the angle every month is possible and also feasible in terms of
economics of PV systems. An example of a PV house is given below to show the
effect of inclination angle.
1.3.1 PV House Example
In this example is desired to show the energy yield increment by changing the
angle of inclination.
Data for the location:
Irradiation Data is given for the months as a wh/m
2
.
Table 4: Average Yearly Irradiation Data
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1900 2690 4070 5050 6240 7040 6840 6040 5270 3730 2410 1800
The solar house power capacity is 15kW.
13
The geographical latitude, L=38 north.
Assume the PV modules used has an efficiency of 17%.
Power rating= 300w
Module area=2.43 m
2
Assume the performance ratio of the module type is 0.77 which includes wiring
losses, power mismatch, soiling loss and inverter loss.
Result:
Total modules are for 15kw system is (15kw/300w)*2.43=121.5m
2
.
The calculation of irradiation for this example is done for different inclination
angles and the results are given in Table 2.
As it is seen in Table 3 the optimum angle is found at 24 degree of panel
inclination. This gives 4702 wh/m
2
average annual irradiation for the location.
The average annual irradiation for 0 was 4232 wh/m
2
. The difference will be
extra which is 47024232=470wh/m
2
.
The annual extra revenue from 15kW will be;
470x121.5x365=20,843.325kwh energy.
14
Chapter2: Power Electronics
2.1 Introduction
The world energy consumption is increasing while fossil fuel is limited. The main
resource of the world energy in the near future is going to be renewable energy
sources such as wind, solar and geothermal energy. Although these new energy
technologies are promising, their energy production is still more expensive than
conventional energy productions. In this case power electronic technologies are
needed to reduce the losses as much as possible, and to deliver and distribute the
power very efficiently. Power electronics had changed rapidly during the last
thirty years and the number of application is increasing due to developments of
the main components of power electronics, which all semiconductor devices and
the microprocessor technology.
15
Figure 7: Semiconductor devices developments [20]
The power electronics application in photovoltaic system is using the most
efficient technique to extract the maximum power from the PV cell. This is
achieved by using MPPT control unit in the converter or inverter side. The MPPT
techniques will be summarized in the following section.
16
2.2 Power Electronic Circuits
2.2.1 Stepdown converter
Stepdown converter which is the basic converter is known as buck converter. As
it is understood from its name the main function of this converter is to convert the
input DC voltage level to another and lower voltage level. The main components
in buck converter are semiconductor switch S, diode D, inductor filter L and
capacitor filter C as shown in figure8.
Figure 8: Buck Converter Circuit
The state of the converter where the inductor current is never zero called
continuous conduction mode (CCM). According to Faraday’s law the voltage of
inductor in any period of time is zero then;
(Is Io)  Ð  I +Io(1 Ð)  I
17
where D is duty cycle, D=ton/T and 1D=toff/T.
In conduction current mode the current IL is always greater than zero. When the
average value of the inductor current is low or the switching frequency is very
low the converter may enter the discontinuous conduction mode. In the buck
converter the boundary between continuous conduction and discontinuous
conduction mode can be found in the following formula [21].
Ib = (1 Ð) 
R
2¡
The minimum value for the capacitance to satisfy the continuous conduction
mode will be found by [21]
Cmin =
(1 Ð)  Io
8  Ir  I  ¡
2
These two equations are very important in the design consideration of converters.
2.2.2 Stepup converter
Stepup converter is called boost converter which consists of inductor L, capacitor
C, controllable semiconductor S, diode D and load resistance R.
18
Figure 9: Boost Converter Circuit
As depicted in the Figure 9, when the switch is on the inductor current is
increasing and when the switch is off the inductor current will flow through the
diode to be deployed in RC circuit.
Using again Faraday’ r e , s law for the inducto in boost conv rter
Is  Ð  I = (Io Is)  (1 Ð)  I
From which the converter transfer function Vo/Vin=1/(1D) is found.
The boundary condition for CCM in boost converter is found by
Ib = (1 Ð)
2
 Ð 
R
2¡
The minimum value for the capacitor can be found by
Cmin = Ð 
Io
Ir  ¡  R
19
2.2.3 Buckboost converter
Nonisolated buckboost converter which consists of input voltage Vs, inductor L,
capacitor C, load R and controllable switch S is shown in the Figure 10.
Figure 10: Buckboost Converter Circuits
The inductor current condition yields the equation below.
Is  Ð  I +Io(1 Ð)  I = u
From this equation it is found that the transfer function equation for the input
output relation will be Vo/Vin=D/(1D)
The boundary condition for the inductor for continuous conduction mode is
Ib =
(1 Ð)
2
R
2¡
Since the circuit structure is the same as boost converter the minimum capacitance
for buckboost converter to stay in CCM can be found from the same formula.
20
2.2.4 Inverters
The function of inverters is to convert the DC to different AC levels. Single phase
voltage source inverters are used for lowvoltage application whereas threephase
is used for medium to high voltage application. In voltage source inverter
application, the current, phase, frequency and voltage need to be always
controllable. By this control topology it is desired to give as pure as possible
sinusoidal output in desired phase, voltage and frequency.
Figure 11: Three Phase Inverter Topology [21]
The standard topology of three phase inverter is shown in Figure 11, three
switching legs are demonstrated where S1 and S4, S3 and S6 and S5 and S2 can’t
be closed at the same time, because this will result in shortcircuiting the DC
voltage supply.
21
2.2.4.1 Sinusoidal PWM technique
In this inverter technique the triangular waveform is compared with three
sinusoidal references voltages. The peak value of the triangular waveform is V
t
and the peak value of reference sinusoidal is V
r
with desired frequency of f
o
.
When V
a,r
is greater than V
t
, S1 is closed and S4 is open, when V
b
,
r
is greater than
V
t
S3 is closed and S6 is open, when V
c
,
r
is greater than V
t
,
S5 is closed and S2 is
open. The waveform of sinusoidal PWM technique is shown in Figure 12 below.
22
Figure 12: Wave Form of PWM Sinusoidal [21]
ma is the modulation index ich is Vr/Vt [22]. If V
r
is greater than V
t
and wh
mo =
Ir
It
u ¸ mo ¸ 1
The output of the inverter is then given by;
Ion = mo 
IJc
2
 sin (2  n  ¡o  t)
23
Ibn = mo 
IJc
 sin (2  n  ¡o  t 12u)
2
Icn = mo 
IJc
2
 sin (2  n  ¡o  t +12u)
If Vr is less than Vt in this case the modulation is called over modulation and
mo = Ir¡It
The output voltage in this ca l se wil be
Io, 1 = 4¡pi  (IJ¡2).
2.2.4.2 Space vector PWM
The output voltages achievable by the three level inverter can be represented in
the space as a set of vectors as shown in Figure 13. These vectors correspond to
switching combinations for the inverter switches. Figure 13 shows these base
vectors V
1
through V
6
and the two zero vectors which correspond to switching
positions resulting in zero output voltage.
24
Figure 13: Base Switching Vectors
One cycle of the output voltage can be represented by six sectors (60
o
each). The
desired output voltage is represented by a rotating reference voltage or vector V
which is approximately calculated by using three adjacent vectors .For the
purpose of specifying the base vectors used to represent the reference one, four
region in each of the six sectors are defined,
25
Figure 14: Determination of the Region
Determination of the region is based upon the length and the angle of V
ref
and is
done by calculating its projections m
1
and m
2
as shown in Figure 15.
26
Figure 15: Vector Projection for Region Determination [23]
where m1 = mn  [cos o –
sìn(u)
√3
¸ onJ m2 = 2  mn  sin(o)¡√S .
Based upon the magnitudes of m
1
and m
2
components the region is determined as
follows:
If m
1
, m
2
and (m
1
+m
2
) < 0.5, then V is in Region 1,
If m
1
> 0.5, then V is in Region 2,
If m
1
and m
2
< 0.5 and (m
1
+m
2
) > 0.5, then V is in Region 3,
If m
2
> 0.5, then V is in Region 4.
The switching or dwelling times I
u
, I
b
onJ I
c
for the three base vectors
I
u
, I
b
onJ I
c
that are used to implement the reference vector are calculated as
follows:
I
s
· F = I +I · I I
c
· I
c
27
u
· I
u b b
+
I
s
= I
u
+I
b
+I
c
where, T
s
is the total switching time.
These switching times for the four regains are found as follows,
Region 1:
Io = 2. m Is. sin((n¡S) )
Ib = Is. ( sin +n¡S))
. o
(1 2  m (o
Ic = 2. m. Is. sino
Io = 2. Is(1 m sin(o +n¡S))
Region 2:
Ib = 2  Is  m sino
Ic = Is  (2  m sin(n¡S o)) 1)
Io = Is (1 2  m sino)
Region 3:
Ib = Is (2  m sin(n¡S +o) 1)
Ic = Is (1 +2  m sin(o n¡S))
Io = Is (2  m sin(o) 1)
Region 4:
Ib = 2  m Is  sin((n¡S) o)
Ic = 2Is(1 (m sin(o +n¡S))
28
where m is the modulation index.
Therefore T
a
, T
b
and T
c
can be determined and the switching sequence
corresponding to the appropriate base vector will be applied to the inverter
switches for the specified amount of time.
The vectors which are applied to the inverter switches for T
a
,T
b
and T
c
periods are
shown in Figure 16 , and they are designated by “a”, “b” and “c” respectively.
Figure 16: Vector Allocation for the Dwelling Times T
a
,T
b
and T
c
[23]
The peak value of the fundamental component of the phase output voltage is
related to the dc link voltage through the modulation index m ,
m =
√SI
¡c]
=
√SI
¡c]
uu 2I
ÐC
7
u ¸ m ¸ 1
The maximum output voltage obtainable is when m=1;
29
I
¡c]_mux
=
7uu
√S
= 4u1.1 I pcok
The switching frequency is chosen to be 1.2 kHz.
2.2.4.2.1 A System Sizing Example
In these examples it is desired to show DCDc operation for varying voltage to a
fixed voltage level of the boost converter output of 120 V, 240V and 600V.
• Assume a 10kW, 30kW and 100kW systems.
• The switching frequency is 25kHz.
• Assume that the input voltage for the converter is varying from 0 to 48
volts.
Figure 17: The Configuration of DC to AC Conversion
For the Boost Converter:
Case 1:
P=10kW
Vin= 048Vdc
Vout=120Vdc
30
For the boost converter e t by; th inputoutpu ratio is calculated
I
s
 Ð  I +
o
 (1 Ð)  I = u I
Io
Iin
=
1
1 Ð
Ð =
Io Iin
Io
D
max
=(1200)/120=1 and D
min
=(12048)/120 6 =0.
Then the duty ratio will be in the range of u.6 ¸ Ð ¸ 1
Case 2:
P=30kW
Vin=048Vdc
Vout=240Vdc
By using the same equations above for the duty ratio, D;
D
max
=(2400)/240=1 and D
min
=(24048)/240=0.8
Then D will be in the range of u.8 ¸ Ð ¸ 1
Case 3:
P=100kW
Vin=048Vdc
Vout=600Vdc
D
max
=(6000)/600=1 and D
min
=(60048)/600=0.92
31
Then D will be in the range of u.92 ¸ Ð ¸ 1
For The ACDC Inverter:
Assume the output voltage from Case 3 is applied to the inverter for the 120 Vac
and 240 Vac output.
Case 1:
f=60Hz
Vin=120 Vdc
Vout=120Vac
For the inverter;
Ion = mo 
IJc
2
 sin (2  n  ¡o  t)
Ibn = mo 
IJc
 sin (2  n  ¡o  t 12u)
2
Icn = mo 
IJc
2
 sin (2  n  ¡o  t +12u)
For one phase Ipcok = Iout  √2, Vout=120*√2=169.7 Vac
Ma=169.7/(600/2)=0.565
Table 5: InputOutput Relation of DCDC Converter with Duty R R atio ange
u.6 ¸ Ð ¸ 1 P=10kW Vin=048Vdc Vout=120Vdc f=25kHz
u.8 ¸ Ð ¸ 1 P=30kW Vin=048Vdc Vout=240Vdc f=25kHz
u.92 ¸ Ð ¸ 1 P=100kW Vin=048Vdc Vout=600Vdc f=25kHz
32
Chapter3. Design Overview of StandAlone PV Systems
When considering planning standalone PV system the most important factor
should be energy balance of the system which means balancing the energy
consumption with the supply. Since the solar energy is changing and fluctuating
daily the solar irradiation for the given place should be calculated realistically in
order to meet the load requirement. A PV generator leads very long payback time
for the system, because it is not used for its whole life span. For this reason using
an auxiliary power generation unit will be more effective in the energy
investment. A combination with a wind turbine can be a good solution
3.1. PV generators, Batteries and Loads Coupling
A small PV generator with reverse diode and battery is connected as shown in the
figure below. The reverse diode is a protector for reverse current when there is no
PV generation otherwise the current will flow from the battery to the array which
will dissipate as a heat. Therefore, the reverse diode is to protect the battery from
these losses and avoid the PV from thermal losses.
33
Figure 18: Battery Charging and Discharging Conditions
In Figure 18, with switch change over different load can be connected to the PV
system. In case ‘’B’’ the current for the load is delivered completely by PVs. The
load ‘’A’’ needs greater current, since the PV can only give part of the required
current to the load. Therefore, the battery will discharge in this case current flows
from the battery to the load. In the load ‘’C’’ case, the current generated from the
PVs is more than what is required for the load. Thus, the generators will charge
the battery by the surplus of current from the PVs. Power electronic controllers
are used in standalone PV systems for higher performance. The charge controller
is used to protect the battery from over charging and also low discharging. MPPT
is used in the inverter to optimize the utilization of the PV output by extracting
the maximum power from the modules.
34
3.2. Calculating electricity consumption
During the year the fluctuations in the irradiation requires standalone PV system.
The consumption has to be calculated in different irradiation durations such as in
different seasons or in different months or at least between the extreme values of
summer and winter when estimating the consumption.
The calculation of the solar energy is based on the weakest month of the year
based on the location and global irradiation data. The radiation calculation for
different geographical latitude and inclination angles will be calculated in the
following section. The first step in standalone system design should start by
calculating the energy consumptions of all the appliances and other equipment
and their power range should be listed in Table 6.
For instance the list for individual power consumers will be as follow.
35
Table 6: Seasonal Electricity Usage of Equipment [19]
Consumer Power (watt) Operating hours, daily Energyused daily
Three lamps in
living room
3*12=36 w
Summer
1
Winter
3
Summer
36wh
Winter
108wh
One lamp in a
room
12 0.5 0.5 6wh 6wh
Two lamps 2*7=14 1 1 14wh 14wh
Refrigerator 50 6(effective) 0 300wh 0
TV 50 2 2 100wh 100wh
Water pump 60 1 0.33 60wh 20wh
Total 186w 11.5 6.83 720wh 248wh
3.3. PV generator sizing
After the demand for electricity is determined, the PV generator should be
determined accordingly. The sizing procedure will base on the nominal power of
the modules so that the required power rating can be meet by different
combinations of the modules in series or parallel.
3.3.1. Calculation of the PV generator’s yield
Monthly average totals for daily horizontal irradiation is listed below, in Table 7
Z2 kwh/m2/day for different locations in the world.
36
Table 7: Z2 Factor Calculated for Different Locations [19]
Locations Jan Feb Mar Apr Ma
y
Jun July Aug Sep Oct Nov Dec
Birmingham,
UK
0.65 1.18 2.00 3.47 4.35 4.53 4.42 3.87 2.67 1.48 0.83 0.45
Brisbane,
Australia
6.35 5.71 4.81 3.70 2.90 2.43 2.90 3.61 4.93 5.45 6.33 6.32
Chicago, US 1.84 2.64 3.52 4.57 5.71 6.33 6.13 5.42 4.23 3.03 1.83 1.45
Dublin, Ireland 0.65 1.18 2.26 3.60 4.65 4.77 4.77 3.68 2.77 1.58 0.77 0.45
Glasgow, UK 0.45 1.04 1.94 3.40 4.48 4.70 4.35 3.48 2.33 1.26 0.60 0.32
Houston, US 2.65 3.43 4.23 5.03 5.61 6.03 5.94 5.61 4.87 4.19 3.07 2.48
Johannesburg,
south Africa
6.94 6.61 5.90 4.80 4.35 3.97 4.26 5.10 6.13 6.45 6.57 7.03
London, UK 0.65 1.21 2.26 3.43 4.45 4.87 4.58 4.00 2.93 1.68 0.87 0.48
Los Angeles,
US
2.84 3.64 4.77 6.07 6.45 6.67 7.29 6.71 5.37 4.16 3.13 2.61
Melbourne,
Australia
7.13 6.54 4.94 3.20 2.13 1.93 2.00 2.71 3.87 5.26 6.10 6.68
New York, US 1.87 7.71 3.74 4.73 5.68 6.00 5.84 5.39 4.33 3.19 1.87 1.48
Philadelphia,
US
1.94 2.75 3.81 4.80 5.55 6.10 5.94 5.42 5.37 3.23 2.13 1.68
Phoenix, US 3.29 4.36 5.61 7.23 8.00 8.17 7.39 6.87 5.97 4.84 3.57 2.97
Sydney,
Australia
6.03 5.54 4.23 3.07 2.61 2.33 2.55 3.55 4.63 5.87 6.50 6.13
Toronto,
Canada
1.58 2.54 3.55 4.63 5.77 6.30 6.29 5.45 4.03 2.68 1.37 1.16
Vancouver,
Canada
0.84 1.75 3.00 4.27 6.03 6.50 6.52 5.42 3.80 2.06 1.03 0.65
The values above in Table 7 are given for horizontal surface, if the location is
tilted then it requires to be multiplied by a factor z3, therefore z2 is given for
horizontal surface and z3 is a factor for tilt correction. For instance, the sun
supplies 4kw per meter square of horizontal area. In august the time between
sunrise and sunset is almost 14 hours. The standard irradiation 1000w/m2 can be
applied for 4 hours instead of 14 hours of fluctuating irradiation for one day. If the
37
module is 50w laid flat on the ground and 1000w/m2 is applied for 4 hours then
the daily yield will be 4*50=200wh.
By interpreting the Table 7 as hours per day with a standard illumination of
1000w/m2, it is only multiplied by the nominal power of the modules or solar
generators. If the total nominal power of the modules is 0.5kw for 4 hours of
standard illumination the total yield is 0.5kw*4h/day=2kwh/day in august. The tilt
correction is an important factor in yield calculation to adjust the nonsouth
facing panel laid. Tilt angle correction factor (Z3) for different directions is given
below in Table 8 Z3 is the correction factor.
Table 8: Z3 Factor for Deviation from Horizontal [19]
Module
orientation
tilt angle
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
South, 30 1.44 1.40 1.17 1.08 1.00 0.96 0.97 1.03 1.17 1.30 1.47 1.42
South, 45 1.57 1.5 1.19 1.05 0.94 0.90 0.91 1.00 1.18 1.37 1.61 1.55
South, 60 1.63 1.54 1.15 0.98 0.85 0.81 0.83 0.92 1.14 1.38 1.68 1.61
South
west/south
east 30
1.37 1.33 1.15 1.07 1.00 0.97 0.98 1.03 1.15 1.25 1.40 1.36
South
east/south
west 45
1.48 1.42 1.16 1.05 0.95 0.91 0.92 1.00 1.16 1.31 1.51 1.46
West/east,
30
1.01 1.01 0.99 0.98 0.97 0.96 0.96 0.97 0.99 1.00 1.01 1.00
West/east,
45
0.99 1.00 0.96 0.95 0.93 0.92 0.92 0.94 0.96 0.98 1.00 0.98
West/east,
60
0.95 0.96 0.91 0.89 0.88 0.86 0.86 0.88 0.92 0.94 0.96 0.94
Eventually, the cell temperature deviation from standard conditions should be
taken into account. Since the standard test temperature for a PV module is 25
C, the correction factor is calculated for the average temperature over 25
38
centigrade, which reduces the power generated by the PVs. The average cell
temperature deviations for the months are listed below for London [19].
Table 9: Temperature Correction Factor Z4 [19]
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1.00 1.00 0.98 0.96 0.93 0.90 0.88 0.88 0.90 0.94 0.97 0.99
Summary,
The ideal energy yield of the generator is calculated by the product of the nominal
power of the PV generator in the following formula.
EiJcol = Pp:  Z2  ZS  Z4
Where,
Ideal energy yield of PV generator, Eideal, in kwh/day
Nominal power of the PV generator, Ppv in kWp
Factor allow for the location and month, Z2, h/day
Factor for tilt correction, Z3
Factor for deviation of cell temperature, Z4
3.4 Consideration of Cable and Conversion Losses
The electricity yield calculation has been done before, now the net produced
electricity for the costumer has to be calculated by taking into account the cable
losses and other effects on the output.
39
3.4.1 Cable losses
When sizing the cable the restriction for the losses should be 3 percent. This is the
goal for loss reduction in low voltage system applications. In standalone system
the module yield is reduced twice by 3 percent. One is in the way from the
generator to the battery via charge controller and second is in the way from the
battery to the load via charge controller. Therefore the overall losses due to
cabling can be assumed as 6 percent of the total yield.
3.4.2 Conversion losses
The conversion of electrical energy into chemical energy will take place in the
battery and from the chemical energy to back again electrical energy is a process
which has a certain loss. As this involves the temperature, age, construction
details and depth of charge and discharge this loss will reduce the system output
by a factor of almost 0.9 [19].
3.5 Mismatch losses
During the operation of PV system the voltage level is changing and causes a
mismatch. This means the radiation and temperature change all the time and the
voltage level changes and eventually the operation can’t be always in its
maximum power point. This drift in voltage is called mismatch and the factor for
mismatch loss can be taken 0.9 [19]. By using MPP tracker control in the inverter
it is possible to reduce the losses caused by this factor.
The PV generator is designed based on the requirement of daily estimated energy
per day as kwh/day. The average required energy for the summer and winter can
be used for the design.
40
Pp: =
w
Z2  ZS  Z4  I
Where I = I1  Ic  Im or it can be written as I = u.94  u.9u  u.9u = u.76
can be used to find the net ener stomer for usage. gy available to the cu
Ercol = EiJcol  I
or
Ercol = Pp:  Z2  ZS  Z4  I
Table 10: Design Parameters for StandAlone PV Systems
Summary of design overview
Characteristics Symbol Unit
Average daily energy
consumption
W Kwh/day
Real energy yield of PV
generator
Ereal Kwh/day
Ideal energy yield of PV
generator
Eideal Kwh/day
Nominal power of PV
generator
Ppv kWp
Factor of normalized
irradiation
Z2 h/day
Factor for temperature and
irradiation deviation
Z3 ….
Factor for line and
conversion mismatching
Z4 ….
41
Summer excess: SE, kwh/day
Winter excess: WE, kwh/day
3.6 Determination of Summer Excess and Winter Reserve
In standalone PV system the determination of summer excess and winter reserve
is very importa o s nt f r izing the battery and PV arrays.
S  E = Ercolsummcr wsummcr(in kwb¡Joy)
wE = Ercolwintcr wwintcr(in kwb¡Joy)
Where W is daily average energyconsumption as indicated in Table 10.
A small house example can summarize the procedure for designing of standalone
PV system. A PV generator in a small house in London from the Table 5 has
0.5kw nominal power the incident angle is southface 45 . In summer season
presented by august the net energy can be
0.5kw*4h/day*1*0.88*0.76=1.34kwh/day. In winter season which illustrated by
December, the normalization factor for effective irradiation Z2 is 0.48h/day, the
energy can be calculated as 0.5kw*0.48h/day*1.55*0.99*0.76=0.28kwh/day.
Thus we can conclude that the winter requirement for small house would be
0.28kwh/day. If it is seen in Table 6 where the energy requirement of small house
is illustrated for winter and summer, it can be understood that the energy
production by 0.5kw PV generator can meet the energy requirement in winter
which is 0.28kwh/day.
42
In Table 4 it can be seen that the summer requirement is 720wh/day and the
energy yield is calculated above is 1340wh/day. In summer there will be a big
amount of surplus in energy yield almost 160%.
The data required for energy balance of the system
1. Daily energy consumption should be listed in a table.
2. Factor Z2 normalization of the radiation at the locations should be known.
3. Factor Z3 correction of the orientation of PV generators.
4. Factor Z4 cell temperature and irradiation correction factor.
5. Overall factor V for lines, conversion and mismatch is required. V=0.76 is
cal t cula ed before based on assumptions.
Pp: = (Joily cncrgy consumption w)¡(Z2  ZS  Z4  I)
3.7 A PV House Design Example
A small house of 3kW system is designed in Phoenix, US.
Assumptions:
The maximum energy consumption is 3kW.
The tiltangle on the framework of the house is south45 .
The ASE300 module type is used having 0.77 performance ratio.
The design will be based on June.
The temperature factor Z
4
is assumed to be 1 because of the lack of daily
temperature data for the location.
43
Result:
Z
2
factor for Phoenix, US is shown in Table 7 which is 8.17.
Z
3
is also shown in Table 8 for southface 45 tilt angle 0.90.
V is the performance ratio 0.77 and Z
4
is 1.
Pp: =
EncrgyConsumption
Z2  ZS  Z4  I
=
Suuu
8.17  u.9  1  u.77
= S29.8w
3.7 Cable crosssection sizing
Table 11: Electrical Parameters and their Units
Electrical parameter Symbol Unit
Line length L M
Power transmitted in the
line
P W
mm
2
Line crosssection A
m/(Ω*mm
2
) Electrical conductivity K
Percentage line loss (3%) …. …..
System voltage V V
The cable crosssection can be calculated be
A = I 
P
(S% I
2
 k)
For the design the cable loss is assumed to be 3% previously to allow the cables to
be able to handle additional power when it is required to add, should be
considered.
44
3.8 Sizing the battery
In standalone PV systems the battery plays an important role in matching the
load requirement for the system. In the locations where there is a big fluctuation
in irradiation we will need at least 23 days reserve for the summer months and 3
5 days reserve for the winter months [12].
If the battery capacity’s unit is ‘’Ah’’ and total energy is ’’wh’’ then Ah=wh/V
where, V is the system voltage.
The battery capacity is Cn, the following equation can be used to calculate the
capacity [12].
Cn = 2  w F¡In
where;
W is the average daily consumption
F is the factor for the reserve days, how many days wanted to be reserved for
V
n
is the system voltage
3.9 Charge Controllers
The most important feature of charge controller is to measure the battery voltage
and protects the battery against the overcharging. This can be achieved by the
following ways [12]
45
Switching off the PV array when the charge cutout voltage is exceeded, short
circuiting the PV array with a shunt controller and adjusting the voltage with an
MPP charge controller.
The reserve diode which prevents the battery to be discharged via the array during
low irradiation level is integrated to the charge controller.
Operation of batteries over long time of operation requires a charge controller to
be flexible [19]. The charge cutoff and discharge cutoff voltages are dependent
on the state of charge of the battery.
The main jobs of the charge controller are;
Allow the optimum charge for the battery.
Protect the battery from the overcharge.
Prevent the battery from unwanted discharge and from deep discharge.
Get information of state of charge of batteries.
46
3.9.1 Series controllers
Figure 19: Series Charge Controller
When the voltage level reaches to the charge cutoff level the power from the PV
generator is blocked by the switch in S1 in Figure 19. After the voltage drops
again below the charging cutoff voltage level S1 switches back on.
47
3.9.2 Shunt controllers
Figure 20: Shunt Charge Controller
When the charge cutoff voltage is reached charge controller continuously reduces
the power of the module. Since it reduces the power continuously the unwanted
power is shortcircuited via the array, this creates heat in the system. This method
is usually used for battery when charging is safe and swift [19].
48
3.9.3 MPPT charge controllers
Figure 21: MPPT Controller
During the operation of the PV array the temperature and irradiation are changing
continuously, this result in changing the IV curve of the PV array consecutively
the maximum power point of the curve should be tracked to exploit the energy
from the PV more efficiently. MPP tracker is used with DC/DC converter by
regulation the voltage every a few minutes and passes through the characteristics
of PV array to determine the maximum point. DC/DC converter gets the power
from the specified point on the curve.
49
3.10 Standalone PV operation principles
A standalone PV system is shown in Figure 22 with all necessary components
such as PV panels, DCAC inverter, DCDC converters, MPPT controller and
battery.
Figure 22: StandAlone PV Operation [1]
where,
Ga is irradiation and Ta is the ambient temperature.
The PV panels are connected in series and parallel combinations to achieve the
desired power capacity. DCAC inverter is used to convert DC input from the PV
generator to AC output at certain voltage level and frequency to be used for
consumers. By the use of the battery, the excess power generated by the PV
system is stored to be used when required. The DCDC bidirectional converter is
50
boosting the voltage level from the battery side to the higher level of inverter
input side or it lowers the voltage from the inverter side to battery input side in
order to be able to charge the battery in an appropriate voltage level. The MPPT
controller is used to get the maximum power from the PV generator during the
operation.
3.10.1 Sizing Example of StandAlone System
As shown in the figure above the PV generator output is connected to a boost
converter which has a fixed output of 400Vdc. The battery will be stored from
this point via stepdown converter. The grid will be supplied via the inverter to
specified voltage and frequency level.
Assumptions:
The system capacity will be 15kW.
The PV generator has an output range of 120V380V and connected to boost
converter.
Grid system frequency is 60Hz and voltage is 120Vrms.
The selected battery voltage is 12V for each.
Results:
For the boostconverter;
Vo/Vin=1/(1D) in this case to have a fixed voltage in the output the duty ratio
operating range D
max
=(400120)/400=0.7, D
min
=(400380)/400=0.05
For the inverter;
51
Vac=(Vdc/2)xMa for this particular example ma=169.7/400=0.424
For the buckboost converter;
Vo/Vin=D/(1D), D=Vo/(Vo+Vin)=240/(240+400)=0.375 duty ratio is set.
For the battery;
The input voltage via buckboost converter is 0.6x400=240Vdc
Which means 240/12=20 batteries are needed in series.
The resulting design is shown in Figure 23.
Figure 23: StandAlone PV System Design Example
52
Chapter 4: Design and Installation of Large Scale PV Systems
4.1 Introduction
Photovoltaic technology which converts sunlight into electricity is growing very
fast among other energy sectors. This sector is being improved because of
ignitions of increasing oil prices and large amount of carbon emission from
different energy utility systems. In this research, it is planned to give the readers
essentials of PV systems in all aspects and consideration of the real design and
installation for 2MW large scale system.
PV systems can be classified into two categories; standalone systems and grid
connected systems. In standalone system the produced energy from solar system
is matched with the total energy demand. Since the energy yield from PV often
does not match with the load, extra storage systems are used to store the energy
when it is not used completely. If the system is by the other type of energy such as
wind or fuel cell, it is called a photovoltaic hybrid system.
In gridconnected systems the public electricity grid functions as an energy store.
It is estimated that by the German solar energy society after the countries set up
the feedin tariff for PV generation to encourage the investments, most of the
power generation will base on PV system in the following years. The figure
shows different application of PV systems.
53
Figure 24: Application of PV systems [12]
4.2 GridTied PV System
In a typical gridtied solar system, the DC electricity produced by the PV array is
usually fed by cables into a PV array combiner box where they are connected
together. A cable form this junction box feeds the DC electricity to the gridtied
inverter. The inverter converts DC to AC which is either consumed by the
building loads and appliances or fed onto the grid. The inverter is connected to
main AC circuit breaker panel / fuse box or directly to the incoming cables from
the grid.
54
Figure 25: General Structure
4.2.1 GridTied Inverter
Gridtied inverters convert DC into single phase or three phase AC electricity at a
voltage and frequency suitable to be fed onto the grid. They are designated in
different sites depending on the size of PV array they are connected to.
4.2.1.1 Functions of Inverters
1. Convert DC to AC at a desired voltage and frequency.
2. To maximize the output of the array under varying conditions of solar
insulation in which it will operate by tracking the maximum power point
of IV curve of the array in which it operates.
3. To ensure nonhazardous operation complying the electrical codes.
4.2.1.2 Technical Requirements
1. Generation of a pure sine wave, synchronous with the sine wave of the
grid.
2. Accurate tracking of the MMP of the array IV curve.
55
3. High frequency operation at full and lower loads.
Tracking the maximum power point is an important aspect of the inverters which
are used in PV systems.
As shown in the figure in each temperature and irradiation values the maximum
power point in IV curve. Different techniques are used. The most general is to
sweep the array current, measure the array voltage and current and deduce the
maximum power point.
Figure 26: Maximum Power Point under Different Irradiation [27]
56
4.2.2 GridTied Inverters and PV Module Configuration
General selection of inverter for large scale gridtied system is central inverter.
This is because of the compatibility and capability of carrying high power starting
from 5 kW and more up to 500 kW.
• When modules are connected in series the output current will be the same
as one module and the output voltage will be the sum of all the module
voltages.
• When, modules are connected in parallel the total current will be the sum
of all the modules, and the total voltage will be the same as one module.
In gridtied systems modules are usually connected in series strings. The
maximum voltage of a string of modules must be lower than the maximum
voltage of a string of modules must be lower than the maximum input voltage
rating of inverter.
Example:
Assume the specification for the PV module is given below.
P
nominal
=200 W
V
OC
=72 V (@10
o
C)
Assume an inverter has a maximum Dc input of 600 V dc.
How could 3.2 kW array be assembled?
Solution:
3.2 kW/200W = 16 units of modules
8×72 = 576 maximum output voltage of one string is less then and appropriate for
the given inverter type.
So we should use two 8module strings in parallel to meet the voltage requirement
with the inverter.
57
Figure 27: Basic Configuration for the given Example Showing the Modules,
Strings and an Array
4.2.3 Selecting Inverter
4.2.3.1 Central Inverters
In large scale PV system all modules in the array are connected to single inverter.
Array will consist of several strings which are then connected to PV combiner
box before being connected to inverter. For large scale system the best way is to
use high capacity inverter as the central inverter.
58
The advantage of using central inverter:
In large scale PV system
• High array outputs up to now can be handled with a few inverters only.
• Termination of the DC cables in the PV combiner box is relatively straight
forward.
The Principal features of central inverter:
• Centralized installation
• Combined series and parallel connection of modules
• Suitable for modules which have the same electrical characteristics
• Suitable for PV array where it is subject to a uniform regime of solar
insulation.
Figure 28: Central Inverter with Large Scale Array Configuration
59
Inverter PV Array Compatibility, Inverter Size and Location
Both the inverter and PV array need to be compatible with each other.
• The size of the inverter should not be less than 90% of the peak wattage of
the array. For instance, if the array peak wattage is 100 kW the inverter
should be in the range of (90 kW or 95 kW).
• The inverter’s MPP range must match the operating voltage of the array.
• The inverter must be compatible of withstanding the maximum array
voltage and current.
When the inverter is overloaded or overheated, it derates itself.
This means that the inverter is no longer able to process a portion of the array
power in the summer when the power at a peak.
Important: if the inverter or located at a place where temperature cannot be
reduces and stays at more than 70
o
C, the inverter will regulate the output of the
array down to protect itself. Generally inverters have coolers to protect them from
high temperature.
4.2.3.2 Inverter Technical Specifications and Efficiency
The following specifications are important to match the inverter with the array
1. On the PV array (DC output) , or input side of the inverter
 DC nominal power and DC peak power.
 DC nominal current and DC peak current.
 DC nominal voltage and DC peak voltage.
 The MPP voltage range (maximum @10
o
C and minimum @ 70
o
C )
60
2. The grid side (output side of the inverter)
 AC nominal power output and AC peak power output.
 AC nominal current output and AC peak current output.
 Inverter efficiency over a range of loads 5%, 10 % ,
20%,….,100%,110%
Inverters operate at different efficiencies, depending on the load. This is
expressed in the inverter’s efficiency curve.
Figure 29: Efficiency of the Inverter under Different Irradiation and Durations [4]
61
4.3 Design Considerations and 2 MW System Installations
4.3.1 Module Selection
PV module selection criteria:
• The performance warranty in case of any problems,
• Module replacement ease,
• Compliance with natural electrical and building codes.
• Manuals should be available to see the quality and characteristics of
module.
In this particular design, the most important issue has to be considered in PV
module selection is area occupancy. When Silicon and Thinfilm are compared
Thinfilm module occupies almost 1.35 times more area than a Silicon PV
module.
For instance consider Schott ASE300D6F/50 (Silicon Module) and First Solar
FS272 module (thin film). Schott has a power of 300W with 2.43 m
2
surface
areas and most solar Thinfilm has a power of 69.3 W with an area of 0.72 m
2
.
When these data are considered it is seen that the land required by Schott module
will be almost 35 % less.
Electric Data for ASE300D6F/50
The electrical data applies to standard test considerations (STC):
Irradiance at the module level of 1000 W/m
2
, spectrum AM 1.5 and a cell
temperature of 25
o
C:
62
Table 12: Characteristics of AS300 Module
Power (max) 300 W
Voltage @ Max. Power Point
50.6 V
Current @ MPP 5.9 Amp
V
oc
(Open Circuit Voltage) 63.2 V
I
sc
(Short Circuit Current) 6.5 Amp
Cell Temperature Coefficient
Power T
k
(P
p
) 0.47 %/
o
C
OpenCircuit
Voltage
T
k
(V
oc
) 0.38 %/
o
C
ShortCircuit
Current
T
k
(I
sc
) 0.1 %/
o
C
Limits
Max. System Voltage 600 V DC
40
o
C to 90
o
C Operating module Temp.
Equivalent wind resistance Wind speed : 120 mph
63
4.3.2 Inverter Selection
As discussed before, for high power applications large capacity inverter is
appropriate to use in this system. PVpowered inverter is one of the best inverters
having integrated MPPT central unit and synchronous with the grid. 260 MW size
in selected to use minimum number of inverters in this system.
4.3.2.1 Electrical Specifications
Table 13: Characteristics of PVPowered Inverter
Continuous output power 260 kW
Weighted CEC efficiency 96.5 % (test)
Maximum DC input voltage (V
DC
) 600 V
DC peak power tracking range (V) 295500 V
DC Nominal current (A) 918 Amp
AC Nominal voltage (V) 480
AC operating range (V) 422528
AC frequency range (Hz) 59.360.5
Harmonic Distortion < 3 %
64
As seen in the data sheet the input DC voltage range of the inverter is 295Vdc to
500Vdc. The output voltage is 480 volts. When the formulas in equations below.
Ion = mo 
IJc
2
 sin (2  n  ¡o  t)
Ibn = mo 
IJc
2
 sin (2  n  ¡o  t 12u ̊)
Icn = mo 
IJc
2
 sin (2  n  ¡o  t +12u ̊)
For V
in
,
1
=295 volts and output voltage is 480 volts ma=2*Vout/Vin,1
If we assume that the voltage is boosted to the level by duty ratio of 0.3 then the
actual inverter input voltage will be;
Io
Iin
=
1
1 Ð
The Vdc for the inverter will be 29S¡(1 u.7) = 98S.S minimum and Suu¡(1 
u.7) = 1666.6 volts. I s t e m i e calculated by, n this ca e h odulation ndex can b
Homox = 2  48u¡98S.S = u.976S
Homin = 2  48u¡1666.6 = u.S76
The modulation index value w n h ill be i t e range of
u.976S ¸ mo ¸ u.S76
65
4.3.3 First Proposal of PV system
Problem description:
In this proposal we will design a large scale PV system which has a size of
≈2MW. The data sheet for the module type ASE300 Scottsolar is given on the
appendix. The arrays will be sized and the inverters simultaneously. The
maximum modules will be used in the array for large scale inverter size and the
restrictions of the inverter input voltage and current will be considered.
The grid voltage is 20kV at 60 Hz frequency. The power will be carried from the
PV side through the inverters having output voltage of 460Vac.
The cabling, transformer and inverter sizing also will be discussed in detail.
Solution:
1. Number of modules :
Peak power of Schott AS300 module; P
max
=300 W
Estimated modules to use in 2 MW system =
2000000
300
= 6,666 modules.
2. Voltage values (high, low, nominal outputs) of modules must be
determined.
Max. Voltage is determined for (10
o
C), V
oc
, and the lowest is in summer
(high temperature) time.
V
oc
, at low temperature (10
o
C) has to be calculated.
V
MPP
, I
MPP
, V
oc
for (25
o
C) is given in data sheet, which are
[V
MPP
=50.6 V, I
MPP
=5.9 A, V
oc
=63.2 V @25
o
C]
Voltage temperature coefficient T
c,Voc
= 0.38 %/
o
C
Current temperature coefficient T
c,Isc
= 0.1 %/
o
C
Power temperature coefficient T
c,PP
= 0.47 %/
o
C
66
And Maximum System Voltage for this system is 600 V.
∆I
1
= 2S (1u) = SS , ∆I
1
= 7u 2S = 4S
c
(Ç ) = I
o
1u
o
C = 6S.6 +I
c.voc
× ∆I
1
6S.6 +SS × u.S8 = 76.9 I
PP
C ∆ S I
M
(Ç1u
o
) = Su.6 +u.S8 × I
1
= Su.6 + S × u.S8 = 6S.9 I
I
MPP
(Ç7u
o
C) = Su.6 u.S8 × ∆I
1
= Su.6 4S × u.S8 = SS.S I
3. Since the system is very large (2 MW) the inverter should be selected to
be as large as possible.
260kW inverter is available in the market; however PVpowered is
selected.
From the data sheet of PVPowered inverter:
Maximum DC input voltage V
DC
= 600 V,
V
pv,MPP
lower = 265 V
V
pv,MPP
upper = 500 V
DC nominal current, I
DC nominal
= 918 Amps.
4. How many modules in one string? How on ay? many strings in e arr
Hox. numbcr o¡ moJulcs =
I
p¡,uppc¡
(o¡ in:crtcr)
I
MPP
(1u
o
C)
=
Suu
6S.9
= 7.82
 8 moJulcs in o string
67
Hox. numbcr o¡ moJulcs =
I
ÐC,mux
(o¡ in:crtcr)
I
0C
(1u
o
C)
=
6uu
= 7.8 moJulcs
76.9
Hox. numbcr o¡ strings in porollcl =
918 ( in:crtcr mox. Jc input)
S.9 (mox. currcnt o¡ cocb moJulc)
= 1SS strings
Maximum number of modules in one array is; 155×8=1250 modules
Total power of one array = 1240 × 300 W = 372 kW which is very large for a
single inverter not appropriate!
We have to consider estimated power for one array such as 270 kW:
270×0.95=260 kW, therefore inverter is appropriate.
If we calculate the estimated power for the array to the 8 modules, then string
power will be
270 kW/ (8×300) =112.5@ 112 strings in parallel can be used for this design.
Figure 30: Configuration of the Array (112 strings with 8 modules) with the
Central Inverter
68
Total number of modules in this case is; 112×8=896 modules, then
896×300=268.8 kW which is suitable for 260 kW inverter.
If we use 110 strings and 8 modules in one string, then total power will be
8×112×300=268.8 kW for one array , 268.8 ×0.965=260.7 kW which is
good for 260 kW inverter.
If we use 8 inverters, that is, 8 arrays;
Total power will be 8×268.8=2.15 MW which is the system power for
maximum efficiency usage.
Figure 31: General Structure of the Arrays with the Inverters
69
The reason for 2.15 MW proposals is that since our system with these inverters
can handle up to 2.15 MW better to use maximum number of modules to reduce
the cost as overall and to utilize the system more efficiently.
Figure 32: Network Model of 2.15MW System, Single Line Diagram
70
Multiple Interconnected Inverters: The highest efficiency in gridtied inverters is
at their full load condition. If the load reduces the efficiency will decrease. If each
array shown in the single line diagram is connected to one inverter and there is no
interconnection between the inverters, when the solar insulation reduces the array
output will decrease. However, in multiple interconnected inverter method, this
situation is overcome by coordinating communication between the inverters and
when the arrays output is less they will feed only one inverter as much as the
output power increases they will switch on one more inverter to meet the power
requirement from the arrays. This continues until it comes to the point of the
maximum power during the time and all the inverters are switched on running at
full load condition. The same sequence is also for when the output of the arrays
are decreasing they start to switch off the inverters one by one to match the output
power of the array with the inverters. This communication will give us two
different advantages; first inverter lifetime will increase, second the inverters will
run in their highest efficiency.
4.3.4 Description of AC Transmission Side and AC Line Model
From 2.1 MW PV capacity, (295500) VDC input from the arrays to the
inverters, connect all the inverters to the same bus at 460 VAC , then step up the
voltage to 20 kV with 5 local bus network . Then transfer the power through an
overhead line for 30 km to the nearest transmission line at 63 kV. At 63 kVAC
bus we have 10 bus local networks on 20 kV bus. For 13 km transmission line
different models were tried and Magpie is selected.
71
Table 14: 13.2–132 kV Class One phase – Neutral Return Line Model
Conductor DC
Resistance
Inductance
(Ω/km)
L
Susceptance
(S/km)
C
Current
ratings
(Ω/km)
Magpie 1.646 j 0.755 j 1.45e7 100 Amp
Squirrel 1.3677 j 0.78 j 6.9e7 130 Amp
Gopher 1.0933 j 0.711 j 7.7e7 150 Amp
Figure 33: 13.2132kV Class One PhaseNeutral line Model
The high voltage in transmission line side is 63kV in this system. According to
this transmission line model voltage drop has to be calculated and checked if it is
within the acceptable range (3%5%).
Ztb = Z +Z2)
72
1  Z2 ¡ (Z1
I1 +I2 = I
I2  Z2 I1  Z1
Ztrons. linc = Ztb¡km Sukm
=
For further increase of the system assume the transformer is rated 2MW, in this
case 2Hw¡S = 6SkI¡√S  I  p¡ assume pf=0.95 then I=19.29A which is less
than 100A
S Iout = Iin I1  (R +XI)¡km ukm Iout = 62.u48kI
9 IoltogcÐrop (%) = 1uu  (6SkI 62.u48kI)¡6SkI = 1.S1 %. As
seen, the voltage drop is less than 3% which is acceptable in transmission
line selection.
4.3.5 Sizing the DC Main Cable from the Array
The DC main cable has to be able to carry the maximum current produced by the
array.
Figure 34: Parameters for Sizing DC Main Cable [12]
73
Generally, the DC main cable is selected to be greater than 1.35 times of the short
circuit current of the array at STC (standard test condition). The cross section of
the cable is selected according to the permitted current carrying capacity. Then
the cross section of DC main cable is derived from;
A
dc,cubIc
= 2  IJc
cubIc

I
n
2
(:  Pp: Pm)  k
the loss factor V= 1% or 2% the cross section is calculated and the result will
round to the next closest value from the standard such as 2mm
2
, 2.5mm
2
if the
result is 2.3mm
2
. Then 2.5mm
2
is selected.
The actual cable loss can be calculated from the following formulas:
P
dc,cubIc
= 2  IJc
cubIc

I
n
2
(AJc
cubIc
)  k
P
dc,cubIc
= 2  IJc
cubIc

P
p¡
2
(AJc
cubIc
 Impp
2
)  k
74
4.3.6 Second design proposal of 2.16MW system
Design Description:
In this proposal, we will design 2.16MW PV system by using ASE300 module
type which is shown on appendix. PV array, a smaller array constructed with
compatible inverter size. The desired inverter characteristic is given in the table
below. The nearest grid high voltage is 34.5kV with 60Hz frequency. All the
technical requirements will be discussed in details.
Solution:
The inverter selected for this design is given below Satcon PV135.
Table 15: Satcon PV Inverters PV135
Continuous output power 135 kW
Weighted CEC efficiency 96.2 % (test)
Maximum DC input voltage (V
DC
) 600 V
DC peak power tracking range (V) 310600 V
DC Nominal current (A) 454 Amp
AC Nominal voltage (V) 208
AC operating range (V) 183229
AC frequency range (Hz) 59.360.5
Harmonic Distortion < 3 %
75
The open circuit voltage and short circuit current of ASmodule is given
Voc=63.2 and Isc=6.5.
In this case if we connect 9 modules in parallel;
The output voltage of the array which is equal to the voltage of one string will be
9*63.2=568.8 which is slightly less than the maximum input voltage of the
inverter. Now, parallel configuration can be considered. If 50 strings with 9
modules each are connected in parallel 50*6.5=325 Amps which is also less than
the maximum current capacity of the inverter. Based on this combination, one
array will have 50*9=450 modules and 450*300=135000 watts which is
compatible with the inverter we have selected.
The cable sizes are selected based on NEC, 2008, 310.15 Appendix.
76
Figure 35: SubArray Configuration
77
The connection cable of one subarray, which is 135 kW array, to the inverter has
a size of #350kcmil from NEC table for the given voltage and current level of 568
volts and 325 amps.
The inverter has an output voltage of 208 volts and the voltage in this level with
150kVA local transformer will be boosted to 480 volt and 200 A current. For this
reason the circuit breaker connected to the end of 135kVA transformer as shown
in figure above. The output of the circuit breaker will necessitate a cable size of
#3/O THHN type, from NEC table, which is compatible for the given voltage and
current level.
These 4 subarrays construct a 540kW big array. The output of one big array,
540kW, will be connected again to local transformer with a size of 500kVA
nominal power. These transformers will convert the generated power from 480
volts level to the 34.5kV grid voltage level. The final design will be as shown in
Figure 36 below.
78
Figure 36: 2.16 MW PV System Line Diagram (Complete System)
79
Chapter 5: Annual Energy Yield Calculation
Nomenclature:
N: number of days in the year
Ul: the unit area thermal loss coefficient
Hd: daily irradiation on PV
α
¡: PV array protective covering transmittance for insulation calculation
: PV cell surface absorbance
β= temperature coefficient of efficiency
Tr= rated temperature when the cell being manufactured
Tm= mean air temperature
Gsc=1367w/m
2
the solar constant
KT: clearness index
H: monthly average daily irradiation on a horizontal surface
H
ç: The geographical latitude of the location
o= monthly average daily irradiation
K= ratio of heat dissipation
80
5.1 Temperature Dependency of Efficiency
Basically, the output of the PV system can be estimated by just multiplying the
PV rated capacity by the standard regional factor [15]. Since a small change in the
output will affect the inverter efficiency, in large scale systems, it is desired to
have more detail calculation for the output. If the rated capacity of PV is C, the
regional factor RF is in the unit of kwh/month/kw of capacity on the basis of
average annual. When ηi is the inverter efficiency, its value will be different
under different power ratings. Its value will affect the output in DCAC transfer.
E is the average monthly output h/month. , measured in kw
E = C  RF  p
ìn¡

For more details, the temperature, insulation, suboptimum panel slope and
characteristics of local climate should be all considered simultaneously.
E = A 
∑p
ìn¡
  Ii
N
where;
E= the average daily output for certain period of time
A= panel area
η
inv
= average hourly panel efficiency
I
i
= the integrated insulation for the hours
N= number of days in the integration and the sum is hourly (daylight hours) for
the month
81
The weather condition is different from one day to another within a month. The
biggest difference in weather conditions which is able to affect the PV output will
be the difference between the months.
The hourly cell efficiency ηi is a function of cell and array design, temperature
and insulation. It can be characterized by the equations below [15], where ηi is the
rated cell efficiency, β is the temperature coefficient of the efficiency.
Tc=cell average temperature for the hour
Tr= is the temperature at which the ce l efficien y is rated (STC, 25 C) l c
ηi = ηi1 β(Tci Ti) +γlog
10
(II)
]
If these two equati a e combined; ons r
E = _
ηi  A
N
]  Ii
[ (Ici Ioi)  Ii
[ (Ioi Im)  Ii
[ (ImIr)  Ii +y Ii  log
10
(II)
]
From different prospective it is preferable to calculate E from the following
formula [15];
E = p 
A
N
 ∑Ii And p = pr1 β(Tc Ta) β(Ta Tm) β(TmTi) +
γlog
10
(II)
]
The monthly average cell efficiency can be calculated by [15]
a. = (Ic Io)  ∑Ii ∑(Ici Ioi)  Ii
b. i Im)  Ii
82
(Io Im)  ∑I = ∑(Ioi
c. log
(I∑Iì)
= ∑Ii  log
Iì
TcTa can be used to determine the monthly average difference between the cell
temperature and ambient temperature during the daylight hours.
T
ci
T
ai
is dependent upon insulation on solar cell and thermal loss from the array
to the environment.
If solar gain in PV array is equated to the sum of electrical output and thermal
loss,
Where Ul is the unit area thermal loss coefficient, α is the PV surface absorbance
and ¡=array pro . tective covering transmittance for the insulation
= o  ¡  Ii  pi +ul  (Ici Ioi) (1) o  ¡  Ii
α  ¡  Ii
∑Ii
= Tci Tai (2)
0l  (Tc Ta)¡( α  ¡) =
∑Iì
2
∑Iì
(3)
The monthly value of KT is calculated
∑Ii
2
∑Ii
= u.219 +u.8S2KT
A tilt correction factor Cf is determined in the computer program in the following
form;
C¡ = 1 u.uuu117(SmS)2
83
where;
S=actual panel tilt
S
m
=the optimal panel tilt
Determination of Ul, based on manufacturer’s data sheet from standard test
protocols.
NOCT symbolizes the nominal operating cell temperature normally its value is
4045 C, and I
NOCT
is the insulation during the test conditions with the cell its
value is generally in the 800 or 1000w/m
2
.
ul¡o = I
N0C1
¡(N0CI I
uì¡
)
T
air
is the ambient temperature, in general Ul is assumed to be 20w/m
2
C
1
[6]. By
rearranging equations, (1), (2), (3) [6].
Io Im =
∑Ii  log
Iì
∑Ii
log(I) = u.64u u.u7S2KI
γ, the coefficient of the efficiency is very small and can be removed from the
equation…
p = pr  1 [((Ic Io) (Io Im) (ImIr))]
Consequently the output calculation can be summa
Ic Io = {o  ¡  (u.219
rized by above equations
+u.8S2KI  C¡)]¡ul [6]
 C¡ = 1 u.uuu117 (SmS)2
p = pr  1 [((Ic Io I
84
) (Io m) (ImIr))]
Emontb = N]  p]  EJ]
N: number of days,
Hd: daily irradiation on PV
α
¡: PV array protective covering transmittance for insulation calculation
: PV cell surface absorbance
Ul: assumption (20m
2
K
1
)
β= temperature coefficient of efficiency
Tr= rated temperature when the cell being manufactured
Tm= mean air temperature
Gsc=1367w/m
2
the solar constant
KT: clearness index
H: monthly average daily irradiation on a horizontal surface
Ho= monthly average daily irradiation
KT=H’/Ho
Eo = 864uu 
0s
pi
 _1 +u.uSS [2  pi 
n
S6S
¸_(cosç  coso  sinæs +æs
 sinç  sino)
ç: The geographical latitude of the location
δ: tilt angle
Hd: monthly n average daily diffuse radiatio
EJ = E'  1.S91 S.S6uKI +4.189KI
2
2.1S7KI
2
]
If sunset hour angle is less than 81.4 degree then;
85
EJ = E’  1.S11 S.u22KI +S.427KI2 1.821KIS]
After that the daily irradiation is broken into hourly values by formula from
CollaresPereira and Rabl [15] for global irradiance
rt =
pi
24
 (o +b  cosæ)  (cosæ cosæs)¡(sinæs æs  cosæs)
o = u.4u9 +u.Su16  in(æs pi¡S)
b = u.66u9 u.4767  sin(æs pi¡S)
s 
rt: i
s: (in radians) is the hourly angle
s the ratio of hourly data to daily total global irradiation
æ
æ: is the midpoint of the hour angle at which the calculation is made
rJ =
pi
24

cosæ cosæs
sinæs æs  cosæs
rd: the ratio of hourly total radiation to daily diffuse radiation
for each hour of the day the average global horizontal radiation H, its diffuse
component Hd, beam component Hb is given by [15];
E = rt  E’
EJ = rt  E’
Eb = E EJ
Finally the equation of hourly irradiance on a plane of PV is found by [15]
Et = Eb  Rb +EJ 
1 +cos[
2
+E  p 
1 cos[
2
Rb is the ratio of the the beam radiation on the PV array to that on horizontal
surface
86
Rb =
cos0
cos0z
θ: incidence angle of beam irradiance on the array
θ
z
: the zenith angle of the sun
Once tilted irradiance for all hours of the day is computed, the daily total Ht is
obtained by summing all the irradiances for all hours.
ηp: average e ficiency, a fu
ηp = ηi(1 βp(Tc Ti)) ,
f nction of Tc
Ta: mean monthly ambient temperature
Tc:
is temperature coefficient of module efficiency
average module temperature
βp
βp, ηi and NOCT depend on the PV cell characteristics.
Example:
Assume the outside temperature is 43 C and the cell temperature is 50 C.
Th
βp=0.4 %/ C
e module type is monocrystalline having an efficiency of 13%.
NOCT=45
Find the efficiency for the specified cell and ambient temperature.
87
Solution:
ηp = ηi(1 βp(Tc Ti)) is given for efficiency as a function of temperature.
pp = 1S  (1 u.4  (44.1 4S)) = 7.28 %
5.2 Energy Production
The power produced by the PV systems is dependent on various factors. The
comprehensive study completed on identifying the factors affecting the annual
electricity generation is done recently.
5.2.1 Cumulative solar irradiance
The surface orientation and tracking of PVs are major factors which determine the
annual production. In the program which has been done in MATLAB to calculate
the tilt angle shows that the optimum tilt angle will increase the total irradiation
for Batman city of Turkey from almost 4233 wh/m
2
to 4702 wh/m
2
. Refer to
Table 2 in Chapter 1.
5.2.2 Module power rating at standard test condition
There is almost 5 percent error in power rating of PV modules. This shows that
the power rating of the module can never operates at its specified maximum
power rating. Different projects have been completely done in different sizes of
PV systems and they verified this truth.
88
5.2.3 Operating temperature
Analysis of different PV technologies showed that the annual production is
reduced because of the operating temperature by almost 2 to 10 percent [8]
resulted from wind speed, module design, mounting technique and outside
temperature.
5.2.4 Maximum power point voltage dependency of irradiation
Different technologies have certain reaction to certain irradiation levels. For
instance, asi, CdTe modules are analyzed and it is seen that they have almost 10
percent larger power output in low irradiation levels compared to other
technologies.
5.2.5 Soiling
Since the soiling reduces the irradiation which is absorbed by the PV, the annual
energy production is reduced by almost up to 7 percent.
89
5.2.6 Variation of Solar Spectrum
The hourly solar variation affects the energy production of PV. Amorphous has
the highest sensitivity and almost 3 percent changes is observed
.
5.2.7 Peak Solar Hours Concept and Definitions
The first approximation analysis and design consideration of PV systems is
dependent on peak solar hours [16]. It is useful to use this concept in first order
sizing of the flatplate arrays which operate under global irradiation. The
magnitude of peak solar hours is equal to the amount of an equivalent day with
1000kw/m
2
of irradiance, resulting in the same value of the daily radiation. These
parameters have unit of time when it is given as hours it will have the same
numerical value of as given in kwh/m
2
day. The whole year can be estimated for a
PV array which exposed to a sol r radiation by [6];
Eo = ∑ (PSEi)  Po
365
ì=1
(1)
Where (PSH)i is the value of parameter PSH for day i and Po is the nominal array
power under standard test conditions. The normalized instantaneous power output
PA/Pmax depends on temperature and irradiance, therefore this equation (1) is
only a close approximation. By using the NOCT concept, the cell temperature can
be estimated by using ambient temperature and cell temperature coefficient.
When the only factor irradiation, is considered for energy yield there will be a
certain percentage of reduction in PV output, because of not considering the
temperature effect. Figure 37 [16] is an analysis of different locations which gives
the error for different values.
90
Figure 37: Shows the Error by Using the Equation (1) Calculate The PV Output in
Different Location In The World. Different Places with Their Latitude and the
Associated Error is Illustrated [16]
5.2.8 Temperature Dependency of Array Output
The temperature effect in PV output comes from temperature dependence of open
circuit voltage w b d [ hich can e escribed by 6]
Ioc(Ic) = I +JIoc¡JI  (Ic 2S) oc(SIC)
Where Tc is cell temperature, JIoc¡JI is temperature coefficient
If an accurate measured value is not known the following theoretical expression
can be used [6].
91
JIoc
JI
=
Ioc Ego y  kb  Ic
Ic
 numbcr o¡ cclls in tbc moJulc
Where E
go
is energy gap, the thermal energy k
b
*T
c
are given in electron volts.
γ is 2 for silicon [6], for typical module with 36 crystalline cells, the equation
gives a value of approximately 80mV/⁰C. The cell temperature can be estimated
from the ambient temperature Ta and the irradiance G with the use of NOCT
nominal operating temperature.
Ic = Io +
N0CI 2u
8uu(or1uuu)
 0
NOCT in centigrade and G w/m
2
, the generally value of NOCT is around 48 [8]
common for PV modules
5.2.9 Module orientation
Fixed tilt arrays: as discussed before the output of dc generator depends on the
inclination angle of the array which increases or reduces the total irradiation
received by the array. A discussion is made before from the result of MATLAB
program showing the tilt angle effect on irradiation. The simulation showed that
the maximum annual irradiation is received at an angle of 2426 tiltangles for
that particular geographical location.
5.2.10 Mismatch Losses and Blocking/Bypass Diodes
When the PV modules are connected in series and parallel different combinations
some losses arise because of nonuniform illuminations of the modules.
92
Consequently, the sum of the output power of the array will be less than the sum
of output power of individual modules. The mismatch losses that resulted from
nonuniform illumination are shown below [7].
Figure 38: Bypass and Blocking Diodes in PV Array
The Figure 38 shows the small combination of PV modules constructing an array
with one blocking diode for each string and one bypass diode for each module.
93
Figure 39: Shading Effect on IV Characteristics of PV [16]
In Figure 39, it shows the IV characteristic one array composed of two string and
2 modules in one string. The graph (a) shows that all the modules are illuminated;
the second graph (b) shows that one model is shaded and wit a bypass diode, the
third graph (c) shows all the modules are illuminated but no bypass diode is used,
the fourth graph (d) shows the IV characteristic of four illuminated modules and
one shaded module with a bypass diode across the shaded. These different
scenarios show the function of bypass diodes for more efficient operations.
94
5.2.11 Inverter Efficiency
DC characteristic of the inverter, the minimum and maximum input current and
voltage with maximum power point range should be considered when sizing the
array. The AC energy produce e V a e found [16]. d by th P arr y can b
EAC = p1  Po  PSE
Po is the nominal power of the array at standard test conditions. PSH is the
average value of peak solar hours at condition. v ff ciency. η1 is the in erter e i
p1 = Pout¡(Pout ut +k2  Pout
2
) +ko +k1  Po
Pout = Poc¡PI
The instantaneous AC output power is normalized to AC output of inverter P
1
.
The k
o
represents self consumption factor, independent of output power. K
1
and k
2
represent also losses which are linear with load power such as voltage drops,
whereas k
2
represents ohmic losses.
95
Figure 40: Inverter Efficiency [16]
Typical inverter efficiency as a function of power generation in Figure 40 from
university of Southampton, S1aR facility [16] with k
o
=0.013, k
i
=0.02, k
z
=0.05. As
it is seen in the Figure 40 the efficiency of the inverter is low in low power range.
This is a choice of inverter operating is generally chosen to be less than the array
power Po of the array. This is based on the losses in the inverter when it converts
the power from DC to AC and self consumption of PV systems is the reason for
lower rating of the inverter. That is why the sizing of the inverter is mostly
dependent on the latitude of the location. Which specify the output of the array,
for instance the recommended value of ratio PI/Po is 0.650.8 for northern Europe
countries and 0.750.9 is more suitable for midEurope latitudes [16].
96
5.2.12 Cable Losses and Transmission Losses
For GW and MW scale the transmission line losses should be considered and
calculated. Because this kind of system is usually established in a remote area,
therefore the transmission loss will be a big part in loss calculation. In PV systems
the irradiation is changing during the day and output power is changing
consecutively. Thus, the output current of PV is assumed to be proportional to
irradiation values. Since, the transmission losses are calculated by I
2
*R, the daily
average irradiation may not be appropriate to apply, so the root mean square can
be applied.
The Berlage model [17] can be used to calculate daily irradiation which is used to
make a theoretical clear day’s curve.
Egtb = ¡
AM
 Ion  Cos0z  +u.S  (1 ¡
AM
)  Ion 
cos0z
1 1.4  ln(¡)
E
¡= atmospheric transmittance value
gth
= global horizontal irradiance (kW/m
2
)
AM
0z= zenith angle
= air mass
I
on
= direct component of horizontal irradiance
The sunny and cloudy days are divided by using the rate of sunshine ‘’r’’ [17].
_(0s
2
 Jn  r +0c  Jn  (1 r))
Jn
= ¸0sr
2
+0c  (1 r) = 0o:c, n
G
s
= sunny day irradiation (kWh/day)
97
G
c
=cloudy days irradiation (kWh/day)
G
ave,n
= RMS of monthly irradiation (kWh/day)
r= rate of sunshine
n= the number of month
Then the RMS value of the day’s irradiation is calculated by [9].
_
0o:c1
2
 S1 +0o:c2
2
 28.2S………+0o:c12
2
 S1
S6S.2S
2
The daily average irradiation is calculated in equation above ↑. The current will
be calculated by using the resulting irradiation data.
I
2
 R = n  J  l  K  t
I= permitted current
l= length of transmission line
t= rise in temperature
d= outside diameter of transmission line
w= irradiance (w/cm
2
)
K= ratio of heat dissipation
R=resistance of final temperature and length
η= radiation coefficient of full black
hr= ratio of heat dissipation by heat emission
98
K = _br 
w
nt
]  p +bw
br = u.uuuS67 
__((2S7 +I +t)¡1uu)
4
[
27S +I
1uu
¸
4
]_
t
bw =
u.uuS72
[27S +I +
t
2
¸
0.123
 ¸I¡J
T is the ambient temperature in Celsius and V is the wind speed.
5.2.12.1 Cable and Transmission Losses Calculations
For short distances;
PI = S  I
2
 R  I
is simply enough to calculate the loss.
Where L is the length of the transmission line and I is the power current.
For longer distances it is more suitable to use [17].
PI = S  R  I  ( I
2
nc  cosç +1¡S  Ic
2
)
Ic= ch
Cosç is the power factor which is generally assumed 0.9 in loss estimation.
arging current at sending end
C= capacitance of one line e two ines D = distance between l
C = u.u241S¡(log
Ðc¡¡
) 1u
E is calculated by line voltage multiplied by 1/√S
99
Chapter 6: Solar Cell Types and Data Sheets
Figure 41: PV Cell Types [12]
6.1 Technologies
The manufacture of solar cells based on three different types of material. The first
one is a semiconductor which absorbs light and converts it into electronhole
pairs. The second is semiconductor junction which separates photogenerated
carriers (electron and hole), and the third is the contacts on front and back of the
100
cell that allow the current to the external circuit. Two main categories of
technology are specified by the semiconductor types.
6.1.1 Crystalline silicon cell (market share 93%) [10]
Crystalline silicon (csi) has been used as a light absorbing semiconductor in most
solar cells, even though it is relatively poor absorber of light. It is now being
improved by huge knowledge of microelectronics industry and producing high
efficiency (1118%) solar cells.
There are two types of crystalline silicon mainly used in the industry; mono
crystalline, its production process is by silicon wafers up to 150mm diameters to
350 microns thick from a high purity single crystalline boule. The second one is
multicrystalline silicon technology.
In two types of silicon the process is, a semiconductor homojunction is formed
by diffusing phosphorus (ntype dopants) into the top surface of the boron doped
(ptype) si wafers. Screenprint contacts are applied to the front and back of the
cell; front contact is designed to allow maximum light exposure of the si materials
with the minimum electron loss. The most efficient production cells use mono
crystalline csi with laser grooved, buried grid contacts for maximum light
absorption and current collection. Each of the csi cells generates about 0.5V and
then 36 cells are connected together in series to form a module producing 12 volt
to charge any 12V battery.
6.1.2 Thin film solar cell
Since the crystalline silicon wafer has very high cost, the market industry started
to look for cheaper materials to use for solar cell. By selecting strong and light
101
absorber materials the cost is significantly reduced. The most common materials
are amorphous silicon (asi), still silicon but in different form, or polycrystalline
materials; such as condium telluride (CdTe) and copper indium (gallium) (CIS
and CIGS). The thin film semiconductor layers are deposited on to either coated
glass or stainless steel sheet.
6.1.2.1 Amorphous Silicon
The most developed thin film solar cell technology and PIN singlesequence
layer is its simple structure. Such cells are suffering from the degradation when
exposed to the sun. Thin film cells are laminated to produce a weather resistant
and environmentally robust module. Although they are less efficient (5.58.5%)
they are cheaper CSi. Besides their lower efficiency they occupy larger area than
the conventional silicon cell of almost 30% more area. The reason for having
larger area is their lower material cost and larger substrate size. Many thin film
technologies have come with the best efficiency in laboratory, 13% and prototype
module 10% efficiency.
Table 16: PV Cell Efficiencies in Different Technologies
102
6.2 Different technologies and their specifications
Table 17: PV Module Specifications
Company name Technology Efficiency Power
rating
Area
First solar Thinfilm
(CdS/Cdte)
10.76 % 77.5 w 0.72 m
2
GE energy Polycrystalline 13.7 % 200 w 1.4567
m
2
Mitsubishi Electric Polycrystalline 13.7 % 190 w 1.382 m
2
ET solar module Monocrystalline 14.5% 185 w 1.276 m
2
Shine solar PV
modules
Monocrystalline 12.4% 210 w 1.688 m
2
Shine solar PV
modules
Polycrystalline 13.94% 270 w 1.937 m
2
Sunwize PV
modules
Monocrystalline 14.09% 180 w 1.276m
2
1SOLTECH Monocrystalline 14.06% 230 w 1.635m
2
Alfa solar Polycrystalline 15.2% 210 w 1.444m
2
Alfa solar pyramid Polycrystalline 15.3% 237 w 1.6m
2
Atersa PV modules Polycrystalline 13.36% 130 w 0.9726m
2
Atersa PV 222p Polycrystalline 13.63% 222 w 1.628m
2
Canadian solar
Allblack CS5A
Monocrystalline 13.3% 170 w 1.277m
2
ET solar modules Polycrystalline 16.86% 215 w 1.47m
2
Tenesol 1.673m
2
Polycrystalline 14.6% 240 w
1.575m
2
Nex power Thinfilm 6.5% 100 w
Titan energy
system
Polycrystalline 14.3% 145 w 1.0127m
2
Titan energy
system
Polycrystalline 14.6% 240 w 1.6434m
2
Titan energy
system
Thin film CIGS
120
8.1% 120 w 1.4732m
2
Titan energy
system
Thin film
Amorphous
5.9% 120 w 2.097m
2
103
Chapter 7: Economic Evaluation and Cost Estimation Techniques
7.1 Introduction
The cost of PV includes acquisition cost, operating cost, maintenance cost and
replacement cost. There are two important effects on the value of money when it
is invested.
1. Inflation rate, i, the measure degradation value of money.
2. Discount rate, d, related to amount of interest that could be saved by that
amount of money.
If an initial amount of money is invested at a rate of 100d % per year, where d is
the percentage rate expressed as a function. After n years the value of investment
will be
N(n) = No  (1 +J)
n
N(n) amount of money will not be able to purchase the same amount of products
as it would purchase when the investment was made. Then the inflation rate
should be taken into account;
C(n) = Co  (1 +i)
n
If Co is the cost of the item at time of investment, Cn is the cost of the item after n
years.
104
7.2 Presentworth Factor
If Co=No present factor Pr= ((1+i)/(1+d))
n
for the item Co to be purchased n
years later the present worth can be calculated by
Pw = (Pr)  Co
) Ict x = (1 +i ¡(1 +J)
Pw = Co  (1 +x +x
2
+x
3
……. . x
n1
)
Po =
Pw
Co
=
1
1 x
x
ì
«
ì
=
1
1 x
=n
x
n
 x
ì
«
ì=0
Po = (1 x
n
)¡(1 x)
Pa is the accumulative present worth factor.
If money is borrowed;
If the total borrowed money is Co for the total cost of investment by the interest
rate of ‘’i’’, the payments at the end of k
th
year van be represented as shown in the
table below.
Table 18: Payment Schedule
Year The paid
amount
Interest
payment
Total payment Balance
1 A1 i*Co A1+Co*i CoA1
2 A2 i*(CoA1) A2+(CoA1) CoA1A2
. . . . .
. . . . .
n An i*(Co
……A
n1
)
An+i*(Co…..A
n
1
)
0
105
Setting the total payme e h e solution for A1. nt of ac year to be equal, it giv s a
A1 +i   i  A1 Co = A2 +i Co
A2 = A1  (1 +i)
In general the total annual payme e rm of; nt can be writt n in the fo
An = A
n1
 (1 +i) = A
1
 (1 +i)
n1
Let x = i +1
Co = A1  x +A1  x
2
+·………A1  x
n1
= A1  (1 +x +·………+x
n1
)
Then A
1
= Co 
1x
1x
n
Substituting x=i+1 yield the equation;
ANNPHI = Co  i 
(1 +i)
n
(1 +i)
n
1
Major inputs for the investments;
• Area related costs which scale the physical size of the system, panel sizes,
inverters size and finally the total size.
• Grid connection costs, it scales the power of the system, how much power
will be invested in which technology, all the equipment that will be used
in the system such as inverters, PV panels, switchgear, transformer and
cablings.
• Project related cost, general overhead, sales, marketing and site design
which are generally fixed for similar sized projects.
106
7.3 The Major Cost and Performance Elements of the Utility Scale PV Power
Plant
7.3.1 The Performance of PV Plant
The life time energy produced by PV power plant is composed of plan location,
annual performance of given capacity, component degradation and system
lifetime.
7.3.1.1 Capacity Factor of the System
The capacity factor is a key point of solar projects economics. The annual
capacity factor for a PV power plant is calculated as [18];
onnuol kilowott bours gcncrotcJ ¡or cocb kilowott AC o¡ pcok copocity [
kwb
kwp
¸
876u bours in o ycor
It is a function of insulation at the location of the PV power plant, performance of
PV panels (efficiencies), the orientation (tilt angle) and the availability of the
power plant to produce.
7.3.1.2 PV Performance and Lifetime
A good estimation of PV panel performance over time is critical to investors.
Silicon cell has the largest operating history of any other solar cell technology. It
is shown in the Figure 42 [18] a monocrystalline silicon PV panel after 20 years
exposure, it doesn’t show any visual degradation. Most of the investors finance
107
the solar system by the assumption of 0.5 to 1 percent degradation rate of output
per year.
Figure 42: PV Panel under 20 Year Exposure of Sun [18]
7.3.1.3 System Performance Prediction
Besides the calculation of PV output, the system component’s performance plays
an important role in making decision of PV projects. The PV plants performance
is uptime, installation, ambient temperature, soiling, inverter and especially panel
efficiency.
108
7.3.2 Initial PV Plant Investment
7.3.2.1 PV Panels
The rapid growth of PV industry led the PV cost to decline in the recent a few
years. The cost of panels and cells are based on depreciation, yield, labor,
chemical consumption, electricity cost and materials. Conversion costs are
improved by shorter and more efficient processes such as higher throughput
production lines, larger plant sizes driving scale economics and greater
automation.
Figure 43: Relative Solar Cell Conversion Efficiencies [18]
109
7.3.2.2 Area Related Cost
This cost related is related to the scale of PV power plant to be established. These
are nonpanel cost expenditure such as steel, foundations, mounting hardware,
plant installation, shipping, warehousing and electrical components used to
connect the panels. It is actually highly correlated with prices of steel, concrete as
well as transportation expenses.
Area related cost can vary considerably from site to site and from one country to
another. This is because of the geographical surface structure of the place whether
it is very flat or hills.
Figure 44: Area related cost [18]
An example of 1 TW PV system is given to show area related cost. The figure
shows how the area related costs are leveraged in different technologies.
110
7.3.2.3 Land Use Cost
The land has been used to be inexpensive in the past. It is becoming a big cost
nowadays as a land area, however, because of government policies encouraging
investment on renewable energy, in most countries the land is provided by the
government. The case is different in some different countries where the solar
investment is largely done by different companies. As PV power plant investors
started to buy lands in south Europe and SouthKorea, due to high demand on in
the land PVsuitable land price increased almost 300 percent and southwest
desert land has sold by $23,000 per acre for flat land [7].
There are two important properties which affect the consumed land by PVpower
systems. Solar power efficiency and system ground covered ratio (GCR). GCR is
the ratio of solar panel area to the land. Flat mounted PV modules use the land
more efficiently and have the maximum GCR but have lower capacity factor
which means lower utilization of the solar irradiation for energy yield.
Figure 45: Land Consumption versus Capacity Factor in Different Technologies
[18]
111
The figure above shows the land consumption versus capacity factor for a power
plant producing 1TWh electricity per year. It is seen that with very high efficiency
PV panels, up to 70 % less land is required for a given capacity factor
configuration. With tracker system almost 30 percent more annual energy is
produced, on the other hand more or the same amount of land is used with lower
efficiency and flat plate system.
7.3.2.4 Connection Costs to the Grid
This cost is related to the total cost of inverters, transformers, wiring system
which is directly related to copper price and transmission lines, switchgear and all
the equipment used in constructing PV power system.
7.3.3 PV Operation Expenses
The operation and maintenance (O&M) of the PV power plant is straighter
forward than the other parts which are explained before.
There are three factors affecting the O&M costs;
• System peak power dominated by the inverter maintenance
• System annual energy production density
• General site related items
If the system capacity factor is improved, the O&M cost automatically reduces
via higher utilization of fixed equipment.
112
The annual energy production density plays an important role in economics of the
PV power systems. Annual energy production density=kwh/m
2
/year.
Figure 46: O&M Cost with Different Technologies [18]
113
Chapter8: PV Sizing Simulator
8.1 Introduction
The PV simulator has been programmed using MATLAB graphical user interface
based on the calculation of irradiation on an inclined surface for a given latitude
and global irradiation data. All the calculation and equations are used which are
mentioned in previous chapters and the assumptions for efficiency calculation and
losses are stated based on references indicated in previous chapters.
It is desired to make large scale PV sizing which will be useful for the end users
to have very close results of the system in the stage of decision making. Different
semiconductor technologies are used to see different cases in order to select the
best design for the desired capacity design in terms of economical issue, area
related costs, smallest payback time consideration and the total system cost.
8.2 Simulator Inputs and Component Prices
The inverters price and module price are used from the www.solarbuzz.com
which has always the updated market prices $/kW. In our simulator three different
semiconductor technologies are selected, monocrystalline; which has the highest
efficiency, polycrystalline and thinfilm which has the lowest efficiency, but
cheaper price and largest surface area. The data sheets for the PV modules which
114
are used in this simulation are NH100AX1 thinfilm, Mitsubishi Electric190W
polycrystalline and Sunwize180W monocrystalline PV modules.
Table 19: Module Types and Specifications
Efficiency Power Area Voc Isc Price
$/W
1.382 m
2
Mitsubishi 13.4 % 185 W 30.8 V 8.23 A 2.48
1.276 m
2
Sunwize 14.09 % 180 W 44 V 5.3 A 2.70
1.575 m
2
Nex NH
100AX1
6.5 % 100 W 101 V 1.65 A 1.76
The feedin tariff as an incentive is given in most of the countries to encourage
the investment in PV systems. In this simulator program it is assumed to be 35
cent per kW which is the offer from energy ministry of Turkey inside the country
for any investor in PV systems. In this agreement the government signs a contract
with the investor, which is a guarantee of purchasing the electricity with certain
price and with a certain period of time. By the consideration of this feedin tariff
the cashflow is calculated for 30 years of operation of the system. Since the
electricity production is reducing by almost 0.9 % every year the cashflow is also
reducing by the same percentage.
The grid voltage is given as another option for the user to be able to calculate the
transmission line losses from the system. The distance and the grid connection
voltage are as inputs and the calculation is made based on the Gopher
transmission line model which is given in the Table 14. The reason for selecting
this model was, since the electricity generation of PV system is very much
expensive, almost 3 or 4 times higher than the conventional production, the
115
investment in higher ampacity will reduce recover its cost in very short operation
period of the system. Additionally, higher ampacity will allow the investor to be
able to increase the capacity without constructing another transmission line
system.
The program receives global irradiation data for each month also based on the
global irradiation for the location in specified latitude. The irradiation on an
inclined surface is calculated for different angle and the optimum angle for the
design is selected. In the transmission line losses calculation, since the monthly
energy yield from the system has been calculated, the monthly energy as a kWh is
divided by the number of days in the month and the daily average energy yield is
used as an input power for transmission lines for more precise calculation.
The irradiation data for Batman city, Turkey, is given for different case studies in
Turkey. In these case studies, different technologies will be selected and distance
will be taken into consideration in order to compare the losses also.
The transmission line nearby the PV system is selected to be 34.5kV.
116
Figure 47: Location of Batman City with Annual Irradiation per metersquare [25]
Figure 48: Irradiation Data and Sunshine Duration for Batman City [24]
117
8.3 General Assumption for PV System Sizing Simulator
8.3.1 DCAC Derates Factor
It is the efficiency of the overall power output of the PV system. In our sizing
simulator the assumption for DC rating will be 0.95, for the inverter and
transformer overall will be 0.94, for mismatch the assumption will be 0.98, for
diode correction 0.995, DC wiring will be 0.98 and AC wiring will be 0.99,
system availability is 0.98. Finally the overall efficiency of the system will be
η=0.786.
The first loss due to DC rating nameplates is the variation of the maximum power
of the manufacturer’s data sheet and the power that is obtained from the real
operating condition of the module. It is proved that the loss because of that is
almost 5 percent [9] and also what lessens the rating power is degradation of the
module itself by being under sun exposure and not operating under standard test
conditions.
The inverter and transformer efficiencies are given in the data sheets. Generally
the values for their efficiencies are ηt= 0.95 to 0.99 for the transformer and
η
inv
= 0.92 to 0.96 for the inverter and in our assumption it will be
η=0.99*0.94=0.93 for the overall of inverter and transformer.
As discussed in previous sections when the PV modules are connected together
due to various effects such as nonuniform illumination which causes non
uniform distribution of irradiation on each module in the array and not operating
at their maximum operating conditions. This will affect the output power by the
reduction of almost 2%. That is why the power loss assumption for the mismatch
will be almost 2 percent that is η=0.98 due to mismatch.
118
Diodes used in the systems as a blocking and also bypass influence the power
output of the system. This loss is because of the dissipation power when trying to
block the reverse current from resistance losses in electrical connections.
DC wiring loss is an assumption of how much power loss is accounted for the
resistance in the DC side of the system. The DC side contains the wiring among
the PV modules and between the arrays and the inverters.
Soiling is also a factor which causes a loss in PV systems. The reason for that is
the accumulation of the dust and different matters on the PV array’s surface
results in reducing the total received irradiation by the PV cells. In our assumption
the factor caused by the soiling will be assumed 0.95.
System availability is important when the revenue is considered for long term.
Because the system sometimes will need maintenance and will not be available
during that time this means in certain period of time there will not be output from
the system. The factor in this regard should be considered and in our assumption
it will be 2 percent of reduction for the whole output.
In the output calculation of the system shading effect should also be considered.
This loss is due to the shading on the PV modules caused by objects or the nearby
PV modules. The figure below shows the derates factors due to shading effect, the
GCR (ground cover ratio) is determined as the array area over the total area used
for the system. The appropriate values in the figure can be picked up to be used in
our case studies.
119
Figure 49: Shading Effect [9]
8.4 Case Study for Monocrystalline
8.4.1 1 km Transmission Line
System Description:
Global irradiation data for Batman city as shown in Figure 48 are entered. The
monocrystalline module type is selected, the specifications for monocrystalline
is given in Table 17. The desired PV capacity is 2MW.
The system is connected to 34.5kV grid system which is given in the input.
120
The distance from the PV system to the grid is 1km which is for the short
distance.
The feedin tariff is selected to be 35 dollarcents per kWh and assume that the
government is buying the electricity with that price for 10 years.
121
Simulation Result:
For specified latitude the optimum angle for the panels is 24 degree which gives
the maximum annual electricity output for the system.
Figure 50: Monocrystalline for 1km Transmission Line
122
The simulation result shows that for monocrystalline silicon technology the total
cost of the system is estimated to be 7.45853 million dollars. The total area
needed for the system is 17013 meter square or 4.204 acres. 1 acre is equal to
4046.8 m
2
. Based on the outputs from the system which is than calculated for the
cash flow on the price of 35 cent per kWh which is set in the program the cash
flow is indicated for 30 years period. In the first simulation the distance to the grid
is selected short, according to this selection the transmission losses appears to be
35,242.4 kWh per year which results in almost 12,334 dollars per year.
8.4.2 25km Long Transmission line Losses
System Description:
In this simulation we will design 2MW PV system with monocrystalline PV
module type. The transmission line voltage is 34.5kV and the distance from PV
system to the grid connection is 25km. it is desired to understand the effect of
transmission line losses on annual revenue. The feedin tariff is selected to be 35
cents per kWh. The irradiation data is used for Batman city as given in Figure 46.
123
Simulation Result:
Figure 51: MonoCrystalline for 25km Transmission Line
124
The simulation result in Figure 51 shows that the total system cost is 7.45853
million dollars. The area is also the same as 17,013.3 m
2
or 4.204 acres. The
transmission line losses can be seen that affected the system annual output which
eventually affects the paybacktime also. The transmission line loss is seen in
figure case2 which is 8.81061x10
8
Wh, that is 881,061 kWh. If this energy wasn’t
lost in transmission and was sold by the price specified in feedin tariff, 35 cent,
the revenue would be 308,371 dollars per year. The payback time of the system
increased to the value of 10.511 years which was 7.434 for 1km transmission line
loss.
8.5 Case Study for PolyCrystalline
8.5.1 1km Transmission Line
System Description:
This case study is based on polycrystalline module technology for 2 MW
systems. As discussed before, the efficiency of the polycrystalline is less than
monocrystalline so the output from monocrystalline is higher than the output
from polycrystalline. But in terms of cost the price for polycrystalline is less
than the price of monocrystalline as a $/kW. This simulation is done for poly
crystalline for 1km transmission line with a grid voltage of 34.5 kV. The same
feedin tariff, 35 cents, for electricity purchasing will be applied to compare the
result with different technologies.
125
Simulation Result:
Figure 52: PolyCrystalline for 1km Transmission Line
It can be obviously seen that from the Figure 52, the total system cost will be
6.97453 million dollars. The total area needed for the system is 17,456.8 m
2
,
126
which is 4.3137 acres. The optimum angle for the inclination of the panels is still
the same, 24 degree, because there is no change in latitude and monthly average
irradiation data. The transmission line loss is 35,078.2 kWh per year for 1km
transmission line length. The amount of cash lost due to transmission losses is
12,277.37 dollars. The payback time is also found that to be 6.968 years. The 30
years electricity yield and 30 years cash flow is shown as a result of the
simulation in figure50.
8.5.2 25km Long Transmission Line
System Description:
In this simulation the polycrystalline module technology is applied, for 2 MW
systems. The grid voltage is 34.5kV, feedin tariff is 35 cents, irradiation data is
for Batman city in Figure 46 and latitude is 38. The transmission line length is
25km.
127
Simulation Result:
Figure 53: PolyCrystalline for 25km Transmission Line
128
The simulation result shows that the transmission line loss changes the annual
revenue substantially. The total transmission line loss is 876,955kWh dissipates in
transmission. The amount of cash lost in transmission is shown to be around
306,934.25 dollars. The payback time of the system is increased to around 9.842
years. The energy yield for 30 years and cash flow based on the amount of
electricity generated for 30 years is also illustrated. It has to be kept in mind that
the decrease in energy yield from the first year toward 30
th
year is because of the
degradation of the module itself.
8.6 Case Study for ThinFilm
8.6.1 1km Transmission Line
System Description:
In this case study thinfilm technology is used for 2MW systems. Thinfilm
technology is relatively cheaper technology compared to polycrystalline and
monocrystalline but the space occupied by thinfilm is almost twice of silicon
technologies. The inputs for this simulation again will be 34.5 kV grid system
voltages, the feedin tariff will be 35 cents per kWh, the transmission line length
will be 1km and irradiation data is given in Figure 46 for Batman city and the
latitude is 38.
129
Simulation Result:
Figure 54: ThinFilm for 1km Transmission Line
130
The simulation result shows that there is considerable increase in area needed for
the system. The space needed in this case is 34,020 m
2
, which is almost 8.4066
acres. Since the output of the system is related to the module area and efficiency
and the increase in area is more than the decrease in efficiency in thinfilm, it can
be seen that the electricity yield of thinfilm is 2,960,690 kWh per year. The total
cost of the system is 5.39053 million dollars for 2MW system because of the
cheapest price of thinfilm as a $/kW. The transmission line loss for 1km is shown
that it is 37,023.3 kWh per year and the amount of cash lost in 35 cents of feedin
tariff policy is 12,058.2 dollars per year. The payback time of the system is
5.2438 years. The energy yield for 30 years of operation is listed in Figure 54 and
the cash flow from the system for 30 years is also listed in Figure 54.
8.6.2 25km Long Transmission Line
System Description:
In this case study thinfilm technology is used for 2MW systems. Thinfilm
technology is relatively cheaper technology compared to polycrystalline and
monocrystalline but the space occupied by thinfilm is almost twice of silicon
technologies. The inputs for this simulation again will be 34.5 kV grid system
voltages, the feedin tariff will be 35 cents per kWh, the transmission line length
will be 25km and irradiation data is given in Figure 46 for Batman city and the
latitude is 38.
131
Simulation Result:
Figure 55: ThinFilm 25km Transmission Line
132
Results of the simulation shows that the energy loss in transmission is 925,583
kWh per year and the amount of cash lost in transmission in 35 cents feedin tariff
policy is almost 323,954 dollars per year. The payback time of the system in this
case will be 7.492 years. The annual yield of 2MW system for 30 years is listed
in Figure 55 and the cashflow is also listed in the figure from the first year to the
end of 30
th
year.
8.7 Comparison of Case Studies
The area occupied by PVs is shown in Figure 56. As can be seen thinfilm module
technology has the biggest area need for 2MW system. Since the efficiency is low
and the energy yield is directly related to efficiency and surface area, to satisfy the
same power output more area will be required. 34020 meter square is almost
8.406 acres which is twice as much as the area needed by the other technology.
34020
17456
17013
0
5000
10000
15000
20000
25000
30000
35000
40000
Thin‐Film Poly‐Crystalline Mono‐Crystalline
S
Q
U
A
R
E
M
E
T
E
R
S
AREA
AREA
Figure 56: Area Comparison of Different Technologies
133
The figurecost shows the cost of 2MW PV system with different technologies. It
is shown that the minimum cost is in thinfilm technology and the maximum is
seen in monocrystalline module type. It has to be known that the area cost wasn’t
included in the system cost. The reason for that in developed countries the area
related cost is a considerable percentage of the total cost while in developing
countries the area is almost free given by the government for incentive in PV
investment. If the area cost of the required system is big enough the biggest area
requirement was for thinfilm then the cost of the thinfilm technology would be
higher and in some areas it could be higher cost than other technologies because
of the area requirement.
5.3905
6.9745
7.4585
0
1
2
3
4
5
6
7
8
Thin‐Film Poly‐Crystalline Mono‐Crystalline
M
i
l
l
i
o
n
‐
D
o
l
l
a
r
s
2 MW SYSTEM COST
cost
Figure 57: Cost Comparison
The payback time is shown in Figure 58, payback comparing different
technologies for 2MW system. It can be clearly seen that the thinfilm technology
134
has lower time duration of payback because of cheaper price of material as $/kW
it has. As mentioned earlier, if the area related cost was included in total cost this
picture would change and depending on the area the payback time of thinfilm
would increase more than the other technologies.
5.243
6.368
7.434
10.511
9.842
7.492
0
2
4
6
8
10
12
Thin‐Film Poly‐Crystalline Mono‐Crystalline
y
e
a
r
s
Tr‐line:1km
Tr‐line:25km
PAYBACK TIME
Figure 58: Payback Time Comparison
135
Conclusion
The Photovoltaics system is gaining importance while the silicon price is going
down. As a biggest energy potential, the Photovoltaic system can be improved to
be more efficient. The solar radiation has to be understood for PV modules to get
the maximum irradiation in optimum inclination angle. The technical details of
the array configuration and component sizing in PV systems are also very
important for efficient power generation, because the factors affecting the PV
output are related to inverter maximum power point tracking, the PV array
configuration, and losses. In technical design, all the losses are considered and
alternative ways are recommended to reduce the losses.
The simulator has been programmed based on energy yield calculation as
discussed in details. The most used PV technologies, monocrystalline, poly
crystalline and thinfilm, are used with their efficiencies and surface areas. The
simulator provides the estimation of energy for 30 years period for the
investment. The area related cost wasn’t included because of variable land prices
all over the world. This tool will give very close estimation of PV output and cost
relation for the endusers to invest in Photovoltaics.
136
References
[1] Ali Keyhani, Integration of Green, Renewable Energy in Electric Power
Systems, Dec 2009, WileyVCH.
[2] Tomas Markvart, Solar Electricity, 2000
[3] Mukund R. Patel Wind and Solar Power Systems, 1999
[4] Falk Antony, Cristian Durschner, KarlHeinz Remmers, Photovoltaics for
Professionals, Solar Electricity Systems, Marketing, Design and Installation, 2007
[5] Frede Blaabjerg, Florin Iov, Remus Teodorescue, Zhe Chen, ‘’Power
Electronics in Renewable Energy Systems’’, Aalborg University, Institute of
Energy, IEEE transaction, 2006.
[6] Analysis of Factors Influencing The Annual Energy Production of
Photovoltaic Systems, David L. King, William E. Boyson and Jay A. Kratochvil,
Sandia National Laboratories, Albuquerque, NM. 871850752
[7] Todd Woody, the SouthWest Desert’s Real Estate Boom, Fortune Magazine,
July 11, 2008.
[8] Roger A. Messenger, Jerry Ventre, Photovoltaic Systems Engineering, 2004
[9] www.nrel.gov/pv/
[10] http://www.solarbuzz.com/solarprices.htm
[11] Tomas Markvart, Solar Electricity, 1994
[12] Deutsche Gesellshaft Für Sonnenenergie, Planning and Installing
Photovoltaic Systems: A Guide for Installers, Architects and Engineers, January
2008
[13] D. Yogi Goswami, Principle of Solar Engineering, January 2000, CRC Press.
[14] http://re.jrc.ec.europa.eu/pvgis/apps3/pvest.php
137
[15] Francis M. Vanek, Louis D. Albright, Energy Systems Engineering:
Evaluation and Implementation, 2008
[16] T. Markvart and L. Castaner, Practical Handbook of Photovoltaics,
Fundamentals and Applications. Elsevier, 2003.
[17] Kosuke Kurokawa, Keiichi Komoto, Peter V. Vleuten, Energy from the
Desert: Practical Proposals for Very Large Scale Photovoltaic Systems, 2007
[18] SunPower Systems, The Drivers of The Levelized Cost of electricity for
UtilityScale Photovoltaics, August 2008
[19] Deutsche Gesellshaft Für Sonnenenergie, Planning and Installing
Photovoltaic Systems: A Guide for Installers, Architects and Engineers, January
2008
[20] Frede Blaabjerg, Florin lov, Remus Teodorescu, Zhe Chen, Power
Electronics in Renewable Energy Systems, Aalborg University, Institute of
Energy Technology Pontoppidanstraede 101, DK9220 Aalborg East, Denmark,
IEEE transactions, 2006
[21] M. H. Rashid, Power Electronics Handbook, Academic Press, 2001.
[22] N. Mohan, Power Electronics: Converters, Applications and Design, Wiley,
NJ, 2003.
[23] A.Kocalmis and S.Sunter, Simulation of a Space Vector PWM Controller For
a ThreeLevel VoltageFed Inverter Motor Drive, IEEE transactions , 2006.
www.eie.gov.tr [24]
[25] www.epdk.gov.tr
[26] T. Danowicz, T. Rodziewicz, and M. Z¥BkowskaWac£Awek, Faculty of
Microsystem Electronics and Photonics, SolarLab, Wroc³aw University of
Technology 11/17 Janiszewskiego Str., 50372 Wroc³aw, Poland, University of
Opole, 5 Miko³ajczyka Str., 45271 Opole, Poland
[27] Maximum Power Point Tracking For Photovoltaic Optimization Using
Extremum Seeking, Steven L. Brunton1,Clarence W. Rowley1, Sanjeev R.
138
Kulkarni1, and Charles Clarkson2 1Princeton University, Princeton, NJ 08544
2ITT, Space Systems Division, Rochester, NY 14606
139
Appendix
140
141
142
143
144
145
146
147
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.