Dublin Institute of Technology
ARROW@DIT
Articles
Dublin Energy Lab
2010-04-01
Projected Costs of a Grid-Connected Domestic PV
System Under Different Scenarios in Ireland, Using
Measured Data from a Trial Installation
Lacour Ayompe
Dublin Institute of Technology, lacour.ayompe@dit.ie
Aidan Duffy
Dublin Institute of Technology, aidan.duffy@dit.ie
Sarah McCormack
University of Dublin, Trinity College, sarah.mccormack@tcd.ie
Michael Conlon
Dublin Institute of Technology, Michael.Conlon@dit.ie
Follow this and additional works at: http://arrow.dit.ie/dubenart
Part of the Power and Energy Commons, and the Urban, Community and Regional Planning
Commons
Recommended Citation
Ayompe, L., Duffy, A., McCormack, S., & Conlon, M. (2010). Projected costs of a grid-connected domestic PV system under different
scenarios in Ireland, using measured data from a trial installation. Energy Policy, vol. 38, pp. 3731-3743. doi: 10.1016/
j.enpol.2010.02.051
This Article is brought to you for free and open access by the Dublin
Energy Lab at ARROW@DIT. It has been accepted for inclusion in Articles
by an authorized administrator of ARROW@DIT. For more information,
please contact yvonne.desmond@dit.ie, arrow.admin@dit.ie.
This work is licensed under a Creative Commons AttributionNoncommercial-Share Alike 3.0 License
JEPO
4104
ARTICLE IN PRESS
XML-IS
Energy Policy ] (]]]]) ]]]–]]]
1
Contents lists available at ScienceDirect
3
Energy Policy
5
journal homepage: www.elsevier.com/locate/enpol
7
9
11
Projected costs of a grid-connected domestic PV system under different
scenarios in Ireland, using measured data from a trial installation
13
L.M. Ayompe a,n, A. Duffy a, S.J. McCormack b, M. Conlon c
15
a
Department of Civil and Structural Engineering, School of Civil and Building Services, Dublin Institute of Technology, Bolton Street, Dublin 1, Ireland
Department of Civil, Structural and Environmental Engineering, Trinity College, Dublin 8, Ireland
c
School of Electrical Engineering Systems, Dublin Institute of Technology, Kevin St, Dublin 8, Ireland
17
b
19
21
23
25
31
Article history:
Received 14 August 2009
Accepted 24 February 2010
This paper presents results of a study of projected costs for a grid-connected PV system for domestic
application in Ireland. The study is based on results from a 1.72 kWp PV system installed on a flat
rooftop in Dublin, Ireland. During its first year of operation a total of 885.1 kWh/kWp of electricity was
generated with a performance ratio of 81.5%. The scenarios employed in this study consider: a range of
capital costs; cost dynamics based on a PV module learning rate of 20 75%; projections for global
annual installed PV capacity under an advanced and moderate market growth conditions; domestic
electricity cost growth of 4.5% based on historic data; and a reduction of 25% or 50% in the CO2 intensity
of national electricity production by 2055. These scenarios are used to predict when system life cycle
production costs fall to grid prices (grid parity).
Average NPV and electricity generation costs ranged from h14,330 and 0.58 h/kWh and were close
to zero and 0.18 h/kWh for a system installed in 2009 and 2030, respectively. However, under
optimistic conditions NPVs are positive for systems installed after 2021 and grid parity occurs in 2016.
Findings are compared with similar international studies.
& 2010 Published by Elsevier Ltd.
F
Keywords:
PV system
Net present value (NPV)
Electricity cost
Grid parity
Q2 Life cycle cost
O
29
a b s t r a c t
O
27
a r t i c l e in f o
PR
33
35
39
D
37
TE
1. Introduction
41
In April 2008, the Irish Government indicated its interest in
micro-generation by announcing the implementation of a
micro- and small-scale electricity generation programme. Pilot
trials were due to be installed in 2009 for domestic scale PV
systems (Sustainable Energy Ireland, 2008b). Furthermore, in
February 2009 the Irish Government and the largest Irish
electricity supplier, the Electricity Supply Board (ESB) introduced
a feed-in tariff of 19h cents per kWh for electricity from
micro-generation (Department of Environment Heritage and Local
Government, 2009).
In order to evaluate the performance of PV systems in Ireland it
is imperative that both field trials which provide information on
the annual energy yield of typical installations as well as studies
to determine the economics and environmental benefits of PV
systems be undertaken. This information is necessary for
evidence-based policy design and implementation.
During operation, PV systems generate electricity without the
emission of greenhouse gases such as CO2, while displacing
EC
43
45
R
47
R
49
O
51
C
53
U
57
59
61
Q1
63
65
69
electricity generated from conventional power plants. The
adoption of PV systems offers significant benefits to household
in terms of reduced energy bills and to society as a whole in terms
of reduced greenhouse gas emissions (Ren et al., 2009). Numerous
issues remain still to be resolved if PV micro-generation is to be
implemented on a large-scale within the residential sector. Some
of these issues include: high electricity generation costs; high
capital cost, high CO2 abatement costs; lack of an adequate tariff
structure for electricity sale; grid integration; and load profile
mismatch with electricity generation.
The main objective of this paper is to provide an insight into
the projected energy, economic and environmental performance
of PV systems for domestic applications in Ireland based on
system cost, system performance, solar radiation data, CO2
emission and energy output dynamics. This would provide useful
information to potential investors and policy makers.
N
55
n
Corresponding author. Tel.: + 353 14023940.
E-mail addresses: lacour.ayompe@dit.ie (L.M. Ayompe), aidan.duffy@dit.ie
(A. Duffy).
67
71
73
75
77
79
81
83
85
87
2. Methodology
Performance results of a 1.72 kWp PV system installed in
Dublin, Ireland were used to determine the annual performance
ratio and total in-plane solar insolation which are important
parameters used to determine annual energy generation. The total
energy generated throughout the PV system’s life of 25 years was
0301-4215/$ - see front matter & 2010 Published by Elsevier Ltd.
doi:10.1016/j.enpol.2010.02.051
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
89
91
93
95
JEPO
4104
ARTICLE IN PRESS
2
1
3
5
7
9
11
13
15
17
19
XML-IS
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
Table 1
Economic parameters.
67
Description
Unit
Symbol
Value
Module price
BOS cost factor
BOS replacement cost factor
BOS component life time
Interest rate
Discount rate
Variable cost factor
Peak power
Global radiation
Standard radiation
Annual degradation of energy yield
Performance ratio
Base year electricity export price (feed-in tariff)
Base year grid supplied electricity price
Annual growth rate of grid supplied electricity price
Annual growth rate of feed-in tariff
PV system life time
Percentage of on-site electricity use
h2009/kWp
%
%
Year
%
%
%
kWp
kWh/m2yr
kW/m2
%
%
h2009/kWh
h2009/kWh
%
%(h2009)
Years
%
Cm
kbos
kbosrpl
Nr
i
d
kv
Ppeak
Gm
Istc
s
Q
Pel,ex
Pel,im
rim
rex
N
Eon
57007 1100
54
70
10
5
5
1
1.72
1035
1
0.82%
80.8
0.19
0.16
4.5
0
25
96
69
71
73
75
77
79
81
83
85
21
CO2 emissions due to embodied energy can be ignored from a
47
national emissions policy perspective since PV modules and
system components are not manufactured in Ireland.
The inverter is replaced after 10 years.
37
39
D
35
TE
33
EC
31
R
system’s life.
R
49
O
51
3. PV system performance
55
3.1. System description
57
The grid connected photovoltaic system used in this study was
installed at the rooftop of the Focas Institute building, Dublin
Institute of Technology, Ireland. It consisted of 8 modules
covering a total area of 10 m2 with an installed capacity of
1.72 kWp. The Sanyo HIP-215NHE5 modules were each of 215 Wp
capacity and comprised 72 solar cells made of thin
mono-crystalline silicon wafer surrounded by ultra-thin amorphous silicon layers. The modules had an efficiency of 17.2% under
standard test conditions and were connected in series. The
unshaded modules were fixed, inclined at an angle of 531 equal
61
63
65
U
N
C
53
59
F
45
29
O
The current feed-in tariff is applicable throughout the PV
27
O
43
25
to the latitude of Dublin, facing south at an azimuth angle of zero
degrees. The PV modules were not cleaned throughout the
monitoring period. A single phase Sunny Boy SB 1700 inverter
was used to convert DC to AC which was fed directly into the
building. The inverter had a maximum efficiency of 93.5% and
maximum AC power of 1700 W. In-plane total solar radiation was
measured using a Sunny SensorBox. Additional sensors for
measuring ambient temperature, wind speed and temperature
at the back of one of the PV modules were connected to the
SensorBox. The solar radiation sensor had an accuracy of 78%
and a resolution of 1 W/m2. The PV module temperature sensor
was a PT 100-M type with accuracy of 70.5 1C while the ambient
temperature sensor was a JUMO PT 100 U type with accuracy of
70.5 1C. The anemometer was a Thies small wind transmitter
with accuracy of 75%. Fig. 1 shows the PV modules and inverter
installation.
PR
41
estimated taking into consideration an annual output drop as a
result of module degradation as a result of exposure to ultraviolet
radiation. A PV system life of 25 years was chosen in this study
since it is same as the performance warranty period of many
module producers (EPIA, 2008).
The average installed cost of PV systems in Ireland in 2009, a
learning rate of 20 75% and two scenarios for PV system annual
installed capacity growth were used to project the system’s
installed cost until 2030. These projected costs together with the
economic parameters in Table 1 were used to calculate projected
net present value and the life cycle cost of electricity generation
for different years of installation at which the PV system is
installed.
Net life cycle greenhouse gas (GHG) emissions were calculated
taking into consideration two scenarios reflecting different trends
in the GHG intensity of electricity production in Ireland. The net
life cycle GHG emissions were used to evaluate GHG abatement
costs associated with PV generated electricity in different years of
installation.
Key assumptions:
23
87
89
91
93
95
97
99
101
103
3.2. Solar radiation and performance ratio
The PV system was monitored between November 2008 and
October 2009. Fig. 2 shows variation of monthly average daily
total in-plane long-term solar insolation for Dublin, performance
ratio and total in-plane solar insolation over the monitored
period. A horizontal surface in Dublin receives on average 920
peak sunshine hours annually. Long-term monthly average daily
total in-plane (531) solar insolation values were obtained using
RetScreen4 software (RETScreen International, 2009). The PV
system’s annual performance ratio was 81.5% while the annual
total in-plane solar insolation over the monitored period and the
long-term annual average for Dublin were 1043 and 1034 kWh/
m2, respectively. The annual total in-plane solar insolation over
the monitored period was 0.83% higher than the long-term
average so the corrected long-term performance ratio of 80.8%
was used in this study.
105
107
109
111
112
113
114
115
116
3.3. PV system cost
117
The installed cost of a roof mounted, grid-connected PV system
depends on its capacity; type of PV modules; nature of the
building on which it is to be installed; cost of balance of system
(BOS) components and installation cost. A survey of 16 crystalline
PV systems installed under the government pilot trial programme
administered by Sustainable Energy Ireland in 2009 with
capacities between 1–6 kWp was conducted. Results showed that
118
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
119
120
JEPO
4104
ARTICLE IN PRESS
XML-IS
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
3
1
67
3
69
5
71
7
73
9
75
11
77
13
Fig. 1. PV modules and inverter installation.
79
15
81
17
5
Monitored period
Long-term average
100
Performance ratio
83
90
27
29
31
80
85
70
87
60
50
2
40
30
1
0
Nov
35
Dec
Jan
Feb
Mar
PR
33
Apr
May
Jun
Jul
Aug
Performance ratio (%)
25
3
F
23
4
O
21
89
91
93
95
20
O
In-plane solar insolation (kWh/m2/day)
19
Sep
10
97
0
99
Oct
Month
Fig. 2. Monthly average daily performance ratio, total in-plane solar insolation over the monitored period and long-term total in-plane solar insolation for Dublin.
53
55
57
59
61
63
65
EC
R
R
51
O
49
C
47
N
45
the average installed PV system cost including value added tax
was 8750 71700 h/kWp with BOS accounting for 30–40%. In this
study 35% was chosen as the average BOS cost factor of the PV
system. The average PV module cost was 5700 71100 h/kWp with
BOS accounting for 54% of this cost. An economic analysis was
conducted using three PV module costs notably 4600, 5700 and
6800 h/kWp. BOS of a PV system consists of all the systems or
engineering components apart from the PV modules or cells. It
primarily consists of an inverter to transform the direct current
(DC) output from the PV array into a form of alternating current
(AC) electricity that can be synchronized with and connected to
the electric utility grid. It also involves support structures and all
the cost of labour involved in system installation (Shum and
Watanabe, 2008).
Despite an impressive growth in annual PV installation in
Europe, Ireland still lags with virtually little or no installations. In
2008, the cumulative installed PV capacity in Ireland was
0.4 MWp made up of 0.1 and 0.3 MWp of grid-connected and
off-grid capacity, respectively. The installed photovoltaic power
per inhabitant in Ireland was 0.09 Wp/inhabitant while the EU 27
average was 19.2 Wp/inhabitant (Eurobserv’er, 2009). The small
PV market size in Ireland indicates why PV system prices are
relatively higher than those in other countries with more
established markets such as Germany, Spain and Italy.
In order to take into account cost dynamics of the PV system,
experience curve analysis was used. The concept of learning-by-
U
43
TE
39
41
101
D
37
103
105
doing expresses that accumulating the deployment or use of a
technology increases the corresponding experience, which
typically results in the optimization of the process involved. In
particular, technology improvements are often economic in
nature and thus result in cost reductions, so that changes in cost
or price are usually used as a proxy for learning-by-doing Ferioli
et al. (2009). Learning curves are usually expressed as
(van der Zwaan and Rabl, 2003; Bhandari and Stadler, 2009;
Ferioli et al., 2009)
Cðxt Þ ¼ Cðx0 Þðxt =x0 Þb
ð1Þ
where xt is the cumulative installed PV module capacity at year t;
b the learning parameter or learning elasticity parameter or rate
of innovation; C(xt) the PV module cost per kWp at year t; C(x0)
the PV module cost at an arbitrary starting year; x0 the cumulative
installed PV module capacity at an arbitrary starting point.
Learning curves are derived by fitting Eq. (1) to cost and
production data observed in the past. The starting point then
ideally corresponds to the first unit of production. In practice,
however, it often proves more appropriate to choose a later (but
still early) stage of deployment for t ¼0, and for the purpose of
estimating future cost reductions on the basis of learning curves,
it can be convenient to use the present cumulative production as
starting point Ferioli et al. (2009).
The learning rate (LR) is defined as the relative cost reduction
(in %) after each doubling of cumulative production, and is given
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
107
109
111
112
113
114
115
116
117
118
119
120
JEPO
4104
ARTICLE IN PRESS
4
LR ¼ 12
b
ð2Þ
7
With every doubling of cumulative production, costs decrease
to a value expressed as the initial cost multiplied by a factor called
the progress ratio. The progress ratio (PR) is given as (van der
Zwaan and Rabl, 2003; Bhandari and Stadler, 2009)
9
PR ¼ 1LR ¼ 2b ¼ 2
21
23
25
27
29
31
33
35
37
39
41
43
45
Electricity domestic general rate in Ireland increased from
9.5h cents/kWh in 1989 to 18.1h cents/kWh in October 2008 and
then dropped to 16.0h cents/kWh in 2009 (Sustainable Energy
Ireland, 2008a). This represented an annual average escalation of
0.43h cents/kWh (4.5%). However, a linear best fit trend on the
electricity general domestic rate data from 1989 to 2009 gives the
projected cost shown in Fig. 5.
3.5. Economic analysis
Annual installed capacity
€5,700 (LR:15%)
€5,700 (LR: 25%)
€4,600 (LR:25%)
€6,800 (LR:15%)
€6,800 (LR:25%)
€4,600 (LR:15%)
R
300
250
5000
200
4000
150
3000
100
2000
50
0
2030
2029
2028
2027
2026
2025
2024
2022
2023
2021
2020
2019
2017
2018
2016
2015
2013
2014
2011
2012
2009
0
2010
79
81
87
89
91
97
99
101
103
105
107
111
61
65
77
95
The economic analysis presented in this paper assumes that it
would be economically viable to invest in a grid connected PV
system when the net present value is positive. The net present
value (NPV) is the sum of discounted single and annual cash flows
over the service life of the PV system less the initial capital cost. It
is an indicator of the PV system’s worth, with a positive value
signifying that benefits accrued will exceed costs over its
economic life. The higher the NPV, the greater the financial
benefit with an NPV greater than zero indicating a profitable
investment (Rogers, 2001; Twidell and Weir, 2006).
This study is based on PV system performance, solar radiation
and market data for Dublin, Ireland. Values used in the analysis
1000
63
75
93
O
PV module cost (€/kWp)
59
N
57
U
55
C
53
73
85
6000
51
71
83
7000
49
69
109
R
47
67
3.4. Electricity cost trend
PV module annual installed capacity (GWp)
19
Learning at PV module level makes no distinction between
global and local learning, since most of the module manufacturing
is done by internationally operating companies and there is
extensive exchange of scientific and technological information on
module technology (Bhandari and Stadler, 2009). Shum and
Watanabe (2008) argue that BOS learning is mostly local in
nature while module learning is relatively global. Therefore, BOS
learning can mostly be attributed to cumulative experience of
system design, integration and installation attained through
greater system integration and a reduction in the number of
BOS parts.
A learning rate of 20% with a sensitivity range of 75% that
accounts for uncertainties of PV technologies and cost development as recommended by Neij (2008) was used in this study.
Future PV module costs were evaluated using learning rates of
15% and 25% (progress ratio of 0.85 and 0.75, respectively). In
order to extrapolate future costs of PV modules, it is important to
estimate future global installation of PV systems.
Global PV electricity generating technology has sustained an
impressive annual growth rate compared with other renewable
energy generating technologies. Total global installed capacity of
grid-connected PV systems was 3.5, 5.1, 7.5 and 13 GWp in 2005,
2006, 2007 and 2008, respectively (Renewable Energy Policy
Framework, 2009). EPIA (2008) developed an advanced and a
moderate scenario for future growth in global installed PV
capacity. The advanced scenario is based on the assumption that
continuing and additional market support mechanisms will lead
to a dynamic expansion of worldwide PV installed capacity. Under
this scenario, average growth rates of 40%, 28% and 18%
were proposed for the periods of 2007–2010, 2011–2020 and
2021–2030, respectively. On the other hand, the moderate
scenario envisages the development of PV against the background
of a lower level of political commitment. Average growth rates of
F
17
ð3Þ
0
PR
15
Q
D
13
0
TE
11
Cðx Þ
lnCðx t Þ=lnQ t
EC
5
30%, 21% and 12% were proposed for the same periods,
respectively. Annual installed PV capacity would therefore be
2.4, 6.9, 56 and 281 GWp for the advanced scenario and 2.4, 5.3,
35 and 105 GWp for the moderate scenario in 2007, 2010, 2020
and 2030, respectively.
Figs. 3 and 4 show projected values for global annual installed
PV capacity and costs for the moderate and advanced scenarios,
respectively. The two scenarios were used to estimate future costs
of PV modules until 2030. Under the moderate scenario, the cost
of PV modules per kWp drops from h4600, h5700 and h6800 in
2009 to h2058, h2550 and h3042, respectively, for a learning rate
of 15% and to h1108, h1373 and h1638, respectively, for a learning
rate of 25% in 2030. Under the advanced scenario, the cost of PV
modules per kWp drops from h4600, h5700 and h6800 in 2009 to
h1687, h2091 and h2494, respectively, for a learning rate of 15%
and to h779, h966 and h1152, respectively, for a learning rate of
25% in 2030.
as Ferioli et al. (2009)
O
3
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
O
1
XML-IS
Year
Fig. 3. PV module cost and global annual installed capacity against year of installation for the moderate scenario.
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
112
113
114
115
116
117
118
119
120
JEPO
4104
ARTICLE IN PRESS
XML-IS
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
7000
5
6000
PV module cost (€/kWp)
3
7
9
11
13
15
Annual installed capa city
€5,700 (LR:15%)
€5,700 (LR: 25%)
€4,600 (LR:25%)
€6,800 (LR:15%)
€6,800 (LR:25%)
€4,600 (LR:15%)
67
300
250
5000
200
4000
150
3000
100
2000
50
1000
17
71
73
75
77
79
81
0
2029
83
2030
2028
2026
2027
2024
2025
2023
2022
2021
2020
2018
2019
2017
2016
2014
2015
2013
2012
2011
2010
2009
0
19
69
PV module annual installed capacity (GWp)
1
5
85
Year
21
Fig. 4. PV module cost and global annual installed capacity against year of installation for the advanced scenario.
87
23
89
25
40
27
35
Actual cost
Projected cost
33
35
37
F
25
20
15
10
95
97
99
101
103
D
5
39
93
O
31
30
O
29
PR
Electricity cost (€cents/kWh)
91
105
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
2015
2017
2019
2021
2023
2025
2027
2029
2031
2033
2035
2037
2039
2041
2043
2045
2047
2049
TE
0
41
EC
43
107
Year
Fig. 5. Electricity general domestic rate and projections in Ireland from 1989 to 2050.
45
55
57
59
R
R
O
53
C
51
N
49
are shown in Table 1. The PV used in this analysis had a capacity
of 1.72 kWp and generated 885.1 kWh/kWp during its first year of
operation. The average annual electricity consumption of a
representative domestic dwelling in Ireland is 5591 kWh. An
analysis run on the daily energy generated by the PV system on
15 min interval and the energy demand profile for the representative dwelling revealed that over a year, 96% of the energy
generated will be consumed on-site while 4% will be exported to
the grid. Over the lifespan of the PV modules, exposure to
ultraviolet radiation would lead to degradation of power output. A
study conducted by Osterwald et al. (2002) showed a linear
dependence of maximum power degradation of 0.82% per year for
mono-crystalline Si modules due to exposure to ultraviolet
radiation.
U
47
61
109
111
and system component inefficiencies or failures. The PR of a PV
system indicates how close it approaches ideal performance
during real operation and allows comparison of PV systems
independent of location, tilt angle, orientation and their nominal
rated power capacity (Blaesser, 1997). Performance ratio is
defined by the following equations as (Eicker, 2003; Nakagami
et al., 2003):
Q¼
Zsys EAC GSTC
¼
ZSTC
Gt EDC,STC
112
113
114
115
ð4Þ
where, Zsys is the system efficiency (%); ZSTC the efficiency under
standard test conditions (%); EAC the AC energy output (kWh);
EDC,STC the DC energy output under standard test conditions
(kWh); Gt the total in-plane solar radiation; GSTC the total solar
radiation under standard test conditions.
116
117
118
119
63
65
3.5.1. Performance ratio
The performance ratio (Q) indicates the overall effect of losses
on a PV array’s normal power output depending on array
temperature and incomplete utilization of incident solar radiation
3.5.2. Net present value
The present value of total cost of the PV system is the sum of
the present value of costs associated with the PV module, initial
BOS, replacement cost of BOS and variable cost. It is assumed that
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
120
JEPO
4104
ARTICLE IN PRESS
6
13
0.8
39
0.7
41
43
45
47
61
0.4
0.3
83
103
105
107
109
111
112
0.2
113
114
0
2005
O
N
U
59
0.5
C
53
57
0.6
81
Actual emissions
0.1
51
55
Projected emissions (50%)
R
49
Projected emissions (25%)
R
CO 2 emissions (kg/kWh)
37
79
87
2055
35
77
CO2 emissions associated with electricity generation in Ireland
89
dropped from 0.728 kg CO2/kWh in 2005 to 0.644 kg CO2/kWh in
2006 and then 0.538 kg CO2/kWh in 2007 with the introduction of
91
the All-Island Single Electricity Market. This drop in CO2
emissions was as a result of fuel switching from coal, oil and
93
peat to gas and renewables. In 2007, electricity generation in
Ireland was based on coal (18%), gas (55%), oil (6%), peat (6%),
95
renewables (11%) and combined heat and power (4%) (Electricity
Supply Board. Annual Report, 2007). This fuel mix shows that Q3 97
there is still potential for CO2 emission reduction in electricity
generation. In order to investigate the effect of decarbonizing
99
electricity generation in Ireland, two scenarios for CO2 emissions
101
2050
33
75
85
3.6. Greenhouse gas emission analysis
2045
31
ð10Þ
73
F
29
N
Gm X
ð1sÞn1
ISTC n ¼ 1
71
O
27
En ¼ Ppeak Q
D
25
where Ct is the total PV system cost (h); Cmt the present value of
cost associated with PV module (h2009); CBOS the present value of
cost associated with the initial investment on BOS (h2009); CBOSrep
the present value of BOS replacement cost (h2009); Cv the present
value of total variable cost (h2009); GHGEF the greenhouse gas
emission factor.
The present value of total revenue from the system during its
useful life is the sum of the present value of revenue of PV
generated electricity consumed on-site and the present value of
revenue of electricity exported to the grid. The size of the PV
system and the electricity demand profile determine the quantity
of electricity it generates that is used on-site to displace grid
supplied electricity at a cost of Pel,im for a given year. The present
value of total revenue (Rt) of the PV system is given as
"
nX
¼N
Gm
ð1 þnrim Þð1sÞn1
P
Eon
Rt ¼ Ppeak Q
ISTC el,im
ð1 þ dÞn
n¼1
2020
23
Cmðn þ Nr Þ
CmðnÞ
TE
21
k¼
ð9Þ
The total power output (En in kWh) by the PV system over its
service life is given as
2015
19
The PV module price reduction factor (k) is given as
EC
17
ð8Þ
Ct
CE ¼
En
67
69
3.5.3. PV electricity cost
The cost of electricity generated (CE in h/kWh) is the ratio of
the life cycle cost of the PV system to the total power output from
the PV system over its service life and is given as
ð6Þ
2010
15
NPV ¼ Rt Ct
and written as
) (
)3
2 (n ¼ N
nX
¼ Nr
X 1þ iðNn þ 1Þ
1 þiðNr n þ 1Þ
þ
6
7
Nr ð1þ dÞn
6 n ¼ 1 Nð1 þ dÞn
7
n¼1
6
7
(
)
Ct ¼ Cm Ppeak 6 7
nX
¼N
6
7
k
k
1
4 k BOS BOSrep þ kv ð1þ kBOS Þ
5
n
ðNr Þ
ð1
þ
dÞ
ð1 þ dÞ
n¼1
ð7Þ
The net present value (NPV) of the PV system is given as
2040
11
ð5Þ
#
O
9
ð1þ nrex Þð1sÞn1
þ Pel,ex ð1Eon Þ
ð1 þdÞn
n¼1
2035
7
Ct ¼ Cmt þ CBOS þCBOSrep þ Cv
nX
¼N
PR
5
money used to buy the PV modules and BOS components are
obtained from a bank loan. The total life cycle cost of the PV
system (Ct) is given as (Bhandari and Stadler, 2009)
2030
3
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
2025
1
XML-IS
115
Year
116
Fig. 6. Actual and projected CO2 emissions associated with grid-supplied electricity in Ireland.
117
Table 2
Parameters for GHG emission analysis.
118
Description
Unit
Symbol
Value
119
Grid supplied electricity emission factor for 2007 excluding losses
Grid transmission and distribution losses
PV electricity distribution losses
kg CO2/kWh
%
%
GHGEF
Lgrid,TD
LPV,D
0.538
9
2
120
63
65
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
JEPO
4104
ARTICLE IN PRESS
XML-IS
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
1
3
5
7
9
electricity exported to the grid are reduced by the distribution
losses on the low voltage distribution lines since this electricity
ends up within the PV system’s neighbourhood.
reduction were considered in this study namely: 25% and 50%
reduction of 2007 emissions between 2007 and 2055 when a PV
system with a service life of 25 years installed in 2030 will be
decommissioned as shown in Fig. 6.
Table 2 shows the parameters used to calculate the quantity of
avoided CO2 emissions by the PV system. Avoided CO2 emissions
associated with grid supplied electricity are essentially those
displaced by PV generated electricity used on-site. On the other
hand, avoided CO2 emissions associated with PV generated
15
3.6.1. Avoided greenhouse gas emissions
PV systems produce electricity in a CO2 neutral way during
their service life although some CO2 emissions arise during the
production of PV modules and other equipment. In this study the
embodied emissions due to the production of the PV system
components are neglected since the components are not
manufactured in Ireland and do not therefore affect national
CO2 emission figures. The total quantity of avoided greenhouse
gas emissions (GHGavoided in kg CO2/kWh) during the PV system’s
service life is the sum of avoided emissions due to the quantity of
displaced grid-supplied electricity by the PV system used on-site
(including transmission and distribution losses) plus the quantity
of avoided emissions due to electricity exported to the low
Global PV market growth
Advanced
Moderate
Learning 25%
‘Moderate 25’
‘Advanced 25’
15%
‘Moderate 15’
‘Advanced 15’
Rate
67
69
11
13
7
17
Fig. 7. PV learning rate and global growth scenarios.
71
73
75
77
79
81
83
19
87
-2000
89
-6000
35
-10000
(0.46)
(0.49)
-12000
-14000
(0.57)
(0.60)
97
99
(0.57)
101
(0.67)
(0.68)
Moderate 25
Advanced 15
103
(0.71)
(0.72)
Advanced 25
105
TE
Moderate 15
Fig. 8. Net present value and electricity generation cost against PV module cost under different scenarios. Life cycle electricity production costs (h/kWh) are shown in
parenthesis.
EC
45
5,000
Moderate 25
Advanced 15
Advanced 25
111
R
49
107
109
Moderate 15
47
112
113
O
Net present value (€)
R
0
51
114
-5,000
U
N
C
53
59
95
(0.59)
D
-20000
43
57
(0.48)
-18000
39
55
(0.46)
-16000
37
93
O
-8000
O
33
91
F
27
Net present value (€)
-4000
31
€ 6,800
0
25
29
85
€ 5,700
PR
23
41
PV module cost (€ /kWp)
€ 4,600
21
115
-10,000
116
117
-15,000
118
61
65
2030
2028
2029
2027
2026
2025
2024
2023
2022
2021
2019
2020
2018
2017
2016
2015
2014
2013
2012
2011
2010
2009
-20,000
63
Year of installation
Fig. 9. Net present value for different years of installation under different scenarios for PV module cost of h4600 per kWp.
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
119
120
JEPO
4104
ARTICLE IN PRESS
8
1
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
4. Results and discussion
voltage grid (including distribution losses). It is calculated using
Eq. (11) given as
3
5
N
Gm X
ð1sÞn1 GHGEF,n Eon ð1 þ Lgrid,TD Þ
ISTC n ¼ 1
þ ð1Eon Þð1 þLgrid,TD LPV,D Þ
67
Four scenarios were developed to model the effect of PV
module cost reduction based on a moderate and advanced growth
in annual installed global PV capacity and a learning rate of either
15% or 25% (i.e. 20%75%). The scenarios are illustrated in Fig. 7.
System NPVs and life cycle electricity generation costs were
calculated for the four scenarios based on the PV system costs and
a system lifespan of 25 years.
GHGavoided ¼ Ppeak Q
7
ð11Þ
9
11
13
15
17
XML-IS
NPV
GHGavoided
71
73
75
3.6.2. Greenhouse gas emissions abatement cost
The total cost of greenhouse gas emissions abatement (CGHG) is
calculated using the negative values of NPV which indicate the
amount of money required to make the investment on the PV
system economically viable. It is calculated using Eq. (12) given as
CGHG ¼
69
ð12Þ
4.1. PV system installed in 2009
77
Fig. 8 shows system NPVs and electricity generation costs
against PV module costs under the different scenarios for PV
systems installed in 2009. It can be seen that NPVs decrease with
PV module costs for a given scenario. The values in brackets
represent the life cycle costs of electricity generation in h/kWh.
79
81
83
19
Moderate 15
21
Moderate 25
Advanced 15
85
Advanced 25
5,000
87
23
89
0
F
O
O
2029
2030
2027
2028
2026
2024
2025
2023
D
2015
2016
2013
2014
2011
2012
2009
103
TE
105
107
EC
45
109
Moderate 15
5,000
Moderate 25
Advanced 15
Advanced 25
111
R
47
49
112
O
Net present value (€)
R
0
51
U
N
C
53
59
101
Fig. 10. Net present value for different years of installation under different scenarios for PV module cost of h5700 per kWp.
43
57
97
Year of installation
41
55
2010
-20,000
39
95
99
2022
35
37
93
PR
-15,000
2020
33
-10,000
2021
31
2019
29
91
-5,000
2018
27
2017
Net present value (€)
25
113
114
-5,000
115
-10,000
116
117
-15,000
118
61
65
2030
2028
2029
2027
2026
2024
2025
2023
2022
2020
2021
2019
2018
2017
2016
2015
2013
2014
2011
2012
2009
2010
-20,000
63
Year of installation
Fig. 11. Net present value against year of installation under different scenarios for PV module cost of h6800 per kWp.
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
119
120
JEPO
4104
ARTICLE IN PRESS
XML-IS
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
1
3
5
7
9
11
viable to invest in the PV system under the conditions assumed in
this study. However, the cost reduction potential of PV modules due
to market growth, learning by doing and technological developments (resulting in a learning rate of 2075%) together with the
likely long-term increase in the value of electricity assumed to be
(4.5% per annum) has the effect of improving the NPV for systems
installed post 2009. Therefore, the future values of NPV and life cycle
electricity generation costs for systems installed after 2009 were
calculated for each of the scenarios (see Fig. 7) and capital cost
projections (see Figs. 3 and 4). Results obtained show that the
NPV increases with year of investment and systems become
economically viable under some scenarios and PV module costs
within the time horizon analysed (up to 2030). The negative
values of NPV indicate the level of additional capital support
required to make investment in the PV system economically viable.
The support required decreases as the year of investment increases
beyond 2009.
Electricity generation costs also decrease with increases in
learning rates for a given PV module price. However, electricity
generation costs which increase with PV module cost are lower
for the advanced scenarios than the moderate scenarios while
holding learning rates constant.
The system’s NPV reduces from h10,173 for PV module cost
of h4600 per kWP to h18,718 per kWp for the Advanced 25 and
Moderate 15 scenarios, respectively. The negative NPVs showed
that investment in the PV system in 2009 was not economically
viable. The average cost of PV electricity generation was 0.47, 0.58
and 0.70 h/kWh for PV module costs of h4600, h5700 and h6800
per kWp, respectively.
13
4.2. Net present value and electricity cost projections
15
17
9
The negative NPV values for the PV system installed in 2009
presented in Section 4.1 show that at present it is not economically
67
69
71
73
75
77
79
81
83
19
85
21
Grid electricity
Moderate 15
Moderate 25
Advanced 15
Advanced 25
0.8
87
0.7
89
0.6
91
23
F
O
0.3
97
41
2030
2029
2028
2027
2026
2025
D
2016
2017
2015
2014
TE
2013
2011
2012
2009
2010
0.0
101
2024
0.1
37
99
2023
35
2022
0.2
39
95
O
0.4
2021
33
93
PR
31
0.5
2020
29
2018
27
2019
Electricity cost (€/kWh)
25
105
Year of installation
Fig. 12. PV-generated and grid-supplied electricity cost against year of installation under different scenarios for PV module cost of h4600 per kWp.
EC
43
Grid electricity
45
0.7
Advanced 15
Advanced 25
111
112
0.6
113
O
N
U
59
C
53
57
Electricity cost (€/kWh)
51
55
Moderate 25
R
49
107
109
Moderate 15
R
0.8
47
103
0.5
114
0.4
115
0.3
116
0.2
117
0.1
118
0.0
119
65
2030
2029
2028
2027
2026
2025
2024
2023
2022
2021
2020
2019
2018
2016
2017
2015
2014
2013
2012
2011
2010
63
2009
61
Year of installation
Fig. 13. PV-generated and grid-supplied electricity cost against year of installation under different scenarios for PV module cost of h5700 per kWp.
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
120
JEPO
4104
ARTICLE IN PRESS
10
1
3
5
7
9
11
13
15
17
XML-IS
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
kWp. Fig. 12 shows that for an initial PV module cost of h4600 per
kWp, the life cycle cost of PV generated electricity drops
from between 0.46 and 0.49 h/kWh in 2009 to between 0.08
and 0.27 h/kWh in 2030. Grid parity is projected to occur in 2026,
2023, 2018 and 2016 for the Moderate 15, Advanced 15, Moderate
25 and Advanced 25 scenarios, respectively.
Fig. 13 shows that for an initial PV module cost of h5700 per
kWp, the life cycle cost of PV generated electricity drops from
between 0.57 and 0.60 h/kWh in 2009 to between 0.10 and 0.27 h/
kWh in 2030. Grid parity is projected to occur in 2028, 2024, 2019
and 2017 for the Moderate 15, Advanced 15, Moderate 25 and
Advanced 25 scenarios, respectively.
Fig. 14 shows that for an initial PV module cost of h6800 per
kWp, the life cycle cost of PV generated electricity drops from
between 0.67 and 0.72 h/kWh in 2009 to between 0.11 and 0.32 h/
kWh in 2030. Grid parity is projected to occur in 2030, 2023 and
2021 for the Advanced 15, Moderate 25 and Advanced 25
scenarios, respectively.
4.2.1. Net present value
Figs. 9–11 show variation in NPV against year of installation
under different scenarios for initial (2009) PV module costs of h4600,
h5700 and h6800 per kWp, respectively. Fig. 9 shows that for an
initial PV module cost of h4600 per kWp, the investment would be
economically viable under the Advanced 15, Moderate 25 and
Advanced 25 scenarios in 2030, 2023 and 2021, respectively. Fig. 10
shows that for an initial PV module cost of h5700 per kWp, the
investment would be economically viable under the Moderate 25
and Advanced 25 scenarios in 2026 and 2023, respectively. Fig. 11
shows that for an initial PV module cost of h6800 per kWp, the
investment would be economically viable under the Moderate 25
and Advanced 25 scenarios in 2029 and 2024, respectively.
4.2.2. Electricity cost
Figs. 12–14 show variations in PV-generated and grid-supplied
electricity costs against year of installation under different
scenarios for PV module costs of h4600, h5700 and h6800 per
67
69
71
73
75
77
79
81
83
19
85
21
Grid electricity
Moderate 15
Moderate 25
Advanced 15
Advanced 25
0.8
87
0.7
89
0.6
91
23
F
O
0.3
97
41
2029
2030
2028
2027
2026
2025
2024
101
2023
2016
D
2014
2015
TE
2013
2012
2011
2009
2010
0.0
99
2022
0.1
37
2021
0.2
35
39
95
O
0.4
2020
33
93
PR
31
0.5
2019
29
2018
27
2017
Electricity cost (€/kWh)
25
Year of installation
Fig. 14. PV-generated and grid-supplied electricity cost against year of installation under different scenarios for PV module cost of h6800 per kWp.
EC
43
45
O
51
65
111
112
N
113
114
6
115
4
116
117
2
118
2030
2029
2028
2027
2026
2025
2024
2023
2022
2021
2020
2019
2018
2017
2016
2015
2013
2014
2012
0
2011
63
107
8
2010
61
50%
10
2009
59
U
57
C
53
55
25%
R
49
105
109
R
47
Avoided CO2 emissions (tCO2 /kWp )
12
103
Year of installation
Fig. 15. Avoided CO2 emissions against year of installation for 25% and 50% projected reduction in grid-supplied electricity emissions between 2007 and 2055.
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
119
120
JEPO
4104
ARTICLE IN PRESS
XML-IS
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
1
4.3. Avoided greenhouse gas emissions
3
Fig. 15 shows avoided total life cycle CO2 emissions against year
of installation for a 25% and 50% projected reduction in grid-supplied
electricity emissions between 2007 and 2055. The results show a
reduction in the quantity of avoided CO2 emissions due to the
projected reduction in GHG emissions associated with electricity
generation. It is seen that the total quantity of avoided CO2
emissions for the PV system installed in 2009 reduces from 10.4
and 9.7 tCO2/kWp for the 25% and 50% cases, respectively, to 7.8 and
6.3 tCO2/kWp if the PV system was installed in 2030. This shows that
reducing the intensity of GHG emissions associated with the
generation of grid supplied electricity reduces the attractiveness of
PV systems in mitigating CO2 emissions.
15
Figs. 16–18 show variations in CO2 emissions abatement costs
against year of installation for a 25% and 50% projected CO2
21
Moderate 15
(50%)
U
N
C
53
83
101
2030
2029
103
2027
2015
2014
TE
99
105
Year of installation
EC
107
109
R
111
112
113
Advanced 25
(25%)
800
O
51
1000
R
49
CO 2 emissions abatement cost ( €/tCO 2 )
1200
47
81
97
Fig. 16. CO2 emissions abatement cost against year of installation for 25% and 50% projected CO2 emission reduction and an initial PV module cost of h4600 per kWp.
45
59
2013
2011
2012
2010
2009
41
57
F
200
0
55
O
400
39
43
79
95
2028
37
77
93
Moderate 15
(25%)
Advanced 25
(25%)
2026
35
600
2025
33
Advanced 25
(50%)
2024
31
75
91
O
29
73
800
PR
27
71
89
1000
D
25
69
87
2018
CO 2 e m i s s i o ns a b a t e m e nt c o s t (€/tCO 2 )
23
67
85
1200
2017
19
4.4. Greenhouse gas abatement cost
2016
17
2023
13
2022
11
2021
9
2020
7
emission reduction and initial PV module costs of h4600, h5700
and h6800 per kWp. It can be seen that CO2 abatement costs
decrease for all scenarios with increasing year of installation;
abatement costs also decrease with decreasing PV module
installation costs. In the best case, no subsidies are required
from 2021 onwards.
Fig. 16 shows CO2 emissions abatement costs against year of
installation for an initial PV module cost of h4600 per kWp for
both a 25% and 50% reduction in the CO2 intensity of
electricity production by 2055. In the case of a 25% reduction
in CO2 emissions, abatement costs decrease from between
h569 and 523h/tCO2 in 2009 to between 230 and 0h/tCO2
in 2020 and to between 55 and 0h/tCO2 in 2030; costs
are higher for the Moderate 15 scenario in all cases. For 50%
reduction in CO2 emissions, abatement costs decrease
from between 624 and 569h/tCO2 in 2009 to between 269
and 0h/tCO2 in 2020 and finally to between 70 and
0h/tCO2 in 2030; again costs are highest for the Moderate
15 scenario.
2019
5
11
Moderate 15
(25%)
Advanced 25
(50%)
114
600
Moderate 15
(50%)
115
116
400
117
200
118
61
65
2030
2029
2028
2027
2026
2025
2024
2023
2022
2021
2020
2019
2018
2017
2016
2015
2014
2013
2011
2012
2010
2009
0
63
Year of installation
Fig. 17. CO2 emissions abatement cost against year of installation for 25% and 50% projected CO2 emission reduction and an initial PV module cost of h5700 per kWp.
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
119
120
JEPO
4104
ARTICLE IN PRESS
12
XML-IS
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
1
CO 2 e m i s s i o ns a b a t e m e nt c o s t (€/tCO 2 )
1200
3
5
7
9
11
13
15
17
67
1000
69
Advanced 25
(25%)
Advanced 25
(50%)
800
71
Moderate 15
(25%)
73
Moderate 15
(50%)
600
75
77
400
79
200
81
83
2030
2029
2028
2026
2027
2025
2024
2023
2022
2021
2020
2019
2017
2018
2016
2014
2015
2013
2012
2011
2010
2009
0
19
85
Year of installation
21
Fig. 18. CO2 emissions abatement cost against year of installation for 25% and 50% projected CO2 emission reduction and an initial PV module cost of h6800 per kWp.
87
23
35
37
39
41
43
45
47
R
49
4.5. Comparative analysis
59
C
N
57
Table 3 shows projected costs for PV electricity generation for
rooftop system in different locations (European Photovoltaic
Industry Association and Green Peace 2008). Current life cycle
electricity generation costs for Dublin are at least 50% higher than
equivalent locations in Europe (Berlin and Paris). The disparity in
electricity generation costs however decreases somewhat after
2010.
U
55
O
51
53
61
5. Conclusion
63
Results of a study on projected economic and environmental
performance of a 1.72 kWp pilot trial rooftop mounted PV system
in Dublin, Ireland has been presented. Measured performance
data was used to evaluate the performance ratio and correct same
65
F
Location
O
O
Berlin
n
Dublin
Paris
Washington
Hong kong
Sydney/Buenos
Aires/Bombay/Madrid
Bangkok
Los Angeles/Dubai
PR
33
89
Table 3
Projected PV electricity generation costs for rooftop system (h/kWh).
n
91
Sunshine hours
2010
2020
2030
900
920
1000
1200
1300
1400
0.35
0.53
0.31
0.28
0.24
0.22
0.20
0.28
0.18
0.15
0.14
0.13
0.13
0.18
0.12
0.10
0.09
0.08
1600
1800
0.20
0.17
0.11
0.10
0.07
0.07
93
95
97
99
101
Current study.
103
D
31
TE
29
EC
27
Fig. 17 shows CO2 emissions abatement costs against year of
installation for an initial PV module cost of h5700 per kWp for
both a 25% and 50% reduction in the CO2 intensity of electricity
production by 2055. In the case of a 25% reduction in CO2
emissions, abatement costs decrease from between 759 and
697h/tCO2 in 2009 to between 358 and 56h/tCO2 in 2020 and to
between 168 and 0h/tCO2 in 2030; costs are higher for the
Moderate 15 scenario in all cases. For 50% reduction in CO2
emissions, abatement costs decrease from between 827 and
759h/tCO2 in 2009 to between 418 and 65h/tCO2 in 2020 and
finally to between 213 and 0h/tCO2 in 2030; again costs are
highest for the Moderate 15 scenario.
Fig. 18 shows CO2 emissions abatement costs against year of
installation for an initial PV module cost of h6800 per kWp for
both a 25% and 50% reduction in the CO2 intensity of electricity
production by 2055. In the case of a 25% reduction in CO2
emissions, abatement costs decrease from between 948 and
871h/tCO2 in 2009 to between 485 and 125h/tCO2 in 2020 and to
between 281 and 0h/tCO2 in 2030; costs are higher for the
Moderate 15 scenario in all cases. For 50% reduction in CO2
emissions, abatement costs decrease from between 1030 and
948h/tCO2 in 2009 to between 568 and 146h/tCO2 in 2020 and
finally to between 356 and 0h/tCO2 in 2030; again costs are
highest for the Moderate 15 scenario.
R
25
105
using long-term solar radiation data. Measured PV output was
used to dynamically evaluate the quantity of electricity generated
by the PV system that would be used on-site relative to a 15 min
average electricity demand profile for domestic dwellings in
Ireland with annual average consumption of 5591 kWh. It was
determined that 96% of the energy generated would be used onsite with only 4% available for export. Over an estimated useful
service life of 25 years, an annual average of 759 kWh/kWp of
electricity would be generated by the PV system with output loss
due to UV degradation accounting for 9.2% of total generation
assuming no losses.
An average PV module installed cost of 5700 71100h/kWp
obtained from a survey of 16 systems installed in 2009 was used.
A current feed-in tariff of 0.19h/kWh was used which assumed to
remain constant throughout the analysis period. Grid supplied
electricity was projected to increase linearly at an annual rate of
4.5% based on 20 years of historic data. Four scenarios for PV
module cost growth were developed based on PV module learning
rates of 15% and 25% as well as a moderate and an advanced
global PV market growth. The quantity of GHG emissions avoided
and GHG abatement costs were evaluated assuming a 25% and
50% reduction in GHG emissions associated with grid-supplied
electricity between 2007 and 2065. The economic and environmental performance of the PV system was evaluated for years of
installation between 2009 and 2030 inclusive assuming a 25 year
lifespan (with production up to 2055).
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
107
109
111
112
113
114
115
116
117
118
119
120
JEPO
4104
ARTICLE IN PRESS
XML-IS
L.M. Ayompe et al. / Energy Policy ] (]]]]) ]]]–]]]
9
11
13
15
17
19
21
23
25
27
29
31
33
35
45
47
49
TE
EC
43
Bhandari, R., Stadler, I., 2009. Grid parity analysis of solar photovoltaic systems in
Germany using experience curves. Solar Energy 83 (9), 1634–1644.
Blaesser, G., 1997. PV system measurements and monitoring the European
experience. Solar Energy Materials and Solar Cells 47 (1–4), 167–176.
Department of Environment Heritage and Local Government, 2009. People power.
/http://www.dcenr.gov.ie/Press + Releases/People +power + - +Minister + Rya
n +announces + incentives+ for+ micro-generation.htmS (accessed on 20/03/
2009).
Eicker, U., 2003. Solar Technologies for Buildings.. John Wiley and Sons, England.
Electricity Supply Board. Annual Report 2007: ESB power generation review.
/http://www.esb.ieS (accessed on 30/10/2008).
Eurobserv’er, March 2009. Photovoltaic barometer.
EPIA, 2008. Solar generation V. /http://www.epia.orgS (accessed on 13/12/2008).
R
41
References
R
39
D
37
55
57
59
61
63
65
67
69
71
73
75
77
Annual degradation of PV yield (s): Percentage at which the maximum power
degrades due to exposure to ultraviolet radiation (%);
Balance of system (BOS): All PV system components other than the PV modules or
cells;
BOS cost factor (kbos): BOS components cost relative to the cost of PV modules;
Discount rate (D): Interest rate used to discount future cash flows (%);
Feed-in tariff (FIT): Policy mechanism designed to pay the producer of electricity at
a specific rate usually higher than the purchase price;
Global radiation (Gm): Sum of beam (direct) and diffuse (indirect) solar radiation
(kWh/m2/yr);
Greenhouse gases (GHG): Gases in the atmosphere that absorb and emit radiation
within the thermal infrared range;
Grid parity: When the cost of PV generated electricity equals that of grid supplied
electricity;
Interest rate (i): Rate charged or paid for the use of money expressed as an annual
percentage of the principal (%);
Net present value (NPV): Sum of present value of costs associated with the PV
module, initial BOS, replacement cost of BOS and variable cost (h);
Peak power (Ppeak): Maximum rated power under standard test conditions;
Performance ratio (Q): Indicator of how close a PV system approaches ideal
performance during real operation and allows for comparison of PV systems
independent of location, tilt angle, orientation and their nominal rated power
capacity;
PV system life (N): Period during which a PV system produces energy (years);
Standard radiation (Istc): Total solar radiation under standard test conditions for PV
modules (1 kW/m2) ;
Variable cost factor (kv): Proportion of the initial investment used annually for
system maintenance and insurance.
U
N
C
O
51
53
Glossary
F
7
O
5
Ferioli, F., Schoots, K., et al., 2009. Use and limitations of learning curves for energy
technology policy: a component-learning hypothesis. Energy Policy 37 (7),
2525–2535.
Nakagami, H., Ishihar, O., et al., 2003. Performance of residential PV system under
actual field conditions in wester part of Japan. In: International Solar Energy
Conference, Hawaii, USA.
Neij, L., 2008. Cost development of future technologies for power generation—a
study based on experience curves and complementary bottom-up assessments. Energy Policy 36 (6), 2200–2211.
Osterwald, C.R., Anderberg, A., et al., May 2002. Degradation analysis of weathered
crystalline-silicon PV modules. National Renewable Energy Laboratory, NREL/
CP-520-31455. /http://www.nrel.gov/docs/fy02osti/31455.pdfS.
Ren, H., Gao, W., Ruan, Y., 2009. Economic optimization and sensitivity analysis of
photovoltaic system in residential buildings. Renewable Energy 34 (3),
883–889.
Renewable Energy Policy Framework (REN21), 2009. Renewables global status
report: 2009 update. Paris.
RETScreen International, 2009. Renewable energy project analysis software.
/http://www.retscreen.netS.
Rogers, M., 2001. Engineering Project Appraisal. Blackwell Science, Oxford, UK.
Shum, K.L., Watanabe, C., 2008. Towards a local learning (innovation) model of
solar photovoltaic deployment. Energy Policy 36 (2), 508–521.
Sustainable Energy Ireland, 2008a. Domestic fuel datasheet. /http://www.sei.ieS
(accessed on 17/10/2008).
Sustainable Energy Ireland, 2008b. Micro and small scale-generation pilot trial
consultation. /http://www.sei.ie/index.asp?locID=1708&docID=-1S (accessed
on 20/11/2008).
Twidell, J., Weir, T., 2006. Renewable Energy Resources. Taylor and Francis, New
York.
van der Zwaan, B., Rabl, A., 2003. Prospects for PV: a learning curve analysis. Solar
Energy 74 (1), 19–31.
O
3
Results obtained showed that domestic grid-supplied electricity
cost is projected to grow from 0.16h/kWh in 2009 to 0.27h/kWh in
2030 and 0.45h/kWh in 2065. PV generated electricity costs decrease
from 0.60 and 0.57h/kWh in 2009 to 0.27 and 0.10h/kWh in 2030 for
the Moderate 15 and Advanced 25 scenarios, respectively. Grid
parity would then occur in 2030 and 2019 for the Moderate 15 and
Advanced 25 scenarios, respectively. The system NPVs increase from
h14,926 and h13,734 in 2009 to h2971 and h2731 for the
Moderate 15 and Advanced 25 scenarios, respectively, during the
same period. The corresponding normalized NPVs would therefore
increase from 8678 and 7985h/kWp in 2009 to 1727 and
1588h/kWp, respectively. For all four scenarios considered, the
average PV generated electricity costs drop from 0.58 to 0.18h/kWh
while the average NPV increases from h14,330 to h35 if the PV
system is installed in 2009 and 2030, respectively. The normalized
average NPV increases from –8331 to 20h/kWp for PV systems
installed in 2009 and 2030, respectively. Grid parity then occurs in
2024. The total quantity of avoided CO2 emissions for the PV system
installed in 2009 reduces from 10.4 and 9.7 tCO2/kWp for the 25%
and 50% cases, respectively, to 7.8 and 6.3 tCO2/kWp if the PV system
was installed in 2030. The life cycle cost of CO2 abatement drops
from between 523 and 1030h/tCO2 in 2009 to between 0 and 568h/
tCO2 in 2020 and finally to between 0 and 356h/tCO2 in 2030.
A comparative analysis shows that current PV system costs are
at least 50% higher in Ireland than other jurisdictions with similar
climatic conditions. This suggests that the market is not operating
optimally, possibly due to the low numbers of buyers and sellers
or due to a lack of market information such as system price and
quality. Significant decreases in capital costs might occur,
therefore, if many buyers and sellers entered the market and
information were more freely available. This would require
market intervention such as subsidies (e.g. grants or low cost
loans), higher feed-in tariffs, or regulation. However with
a current marginal cost of abatement of between 523 and
1030h/tCO2, widespread deployment would be costly and would
not represent best value for money for the taxpayer.
PR
1
13
Please cite this article as: Ayompe, L.M., et al., Projected costs of a grid-connected domestic PV system under different scenarios in
Ireland, using measured data from a trial installation. Energy Policy (2010), doi:10.1016/j.enpol.2010.02.051
79
81
83
85
87
89
91
93
95
97
99
101