Energy 35 (2010) 2164–2173

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Modelling the existing Irish energy-system to identify future energy

costs and the maximum wind penetration feasible
D. Connolly a, *, H. Lund b, B.V. Mathiesen b, M. Leahy a
a
Department of Physics, University of Limerick, Castletroy, Limerick, Ireland
b
Department of Development and Planning, Aalborg University, Denmark

a r t i c l e i n f o a b s t r a c t

Article history: In this study a model of the Irish energy-system was developed using EnergyPLAN based on the year
Received 10 September 2009 2007, which was then used for three investigations. The first compares the model results with actual
Received in revised form values from 2007 to validate its accuracy. The second illustrates the exposure of the existing Irish energy-
21 January 2010
system to future energy costs by considering future fuel prices, CO2 prices, and different interest
Accepted 23 January 2010
Available online 2 March 2010
rates. The final investigation identifies the maximum wind penetration feasible on the 2007 Irish energy-
system from a technical and economic perspective, as wind is the most promising fluctuating renewable
resource available in Ireland. It is concluded that the reference model simulates the Irish energy-system
Keywords:
Modelling accurately, the annual fuel costs for Ireland’s energy could increase by approximately 58% from 2007 to
Simulating 2020 if a business-as-usual scenario is followed, and the optimum wind penetration for the existing Irish
Reference model energy-system is approximately 30% from both a technical and economic perspective based on 2020
Irish energy-system energy prices. Future studies will use the model developed in this study to show that higher wind
Cost penetrations can be achieved if the existing energy-system is modified correctly. Finally, these results are
Maximum wind penetration not only applicable to Ireland, but also represent the issues facing many other countries.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction significance for the Irish energy-system. In line with this, the aim of
this work is to develop a model of the Irish energy-system to
By 2020, Ireland1 has an obligation under EU initiatives to propose how these targets are met, and analyse the implications of
supply 20% of its total primary energy consumption from renew- these targets. The first step required in this process is to create
able sources [1]. Also, the Kyoto protocol only allows Ireland to a reference model by simulating a historical year, and ensuring that
increase its CO2 emissions by 13% compared to 1990 levels [2] and the model is functioning accurately. Therefore, this paper discusses
in 2006, Ireland was 26.7% above 1990 levels [3]. As a result, the the need for a model of the Irish energy-system and subsequently
Irish government set a number targets for energy in 2007 [4]. These describes why EnergyPLAN was chosen to create this model. Finally,
include: 30% of fuel from biomass at the three state-owned peat three initial investigations were carried out using the model
power-plants by 2015, no oil in electricity generation by 2020, 15% developed: a comparison was made between the 2007 reference
of electricity from renewable sources by 2010 and 33% by 2020, model developed and the actual performance of the Irish energy-
500 MW of ocean energy by 2020, combined heat and power (CHP) system in 2007, the sensitivity of the existing Irish energy-system
needs to be expanded to 400 MW by 2010 and 800 MW by 2020, 5% to various economic parameters was investigated, and finally, the
of heat demand must come from renewable sources by 2010 and maximum wind penetration feasible on the existing Irish energy-
12% by 2020, 5.75% of energy from biofuels by 2010 and 10% by system was identified.
2020, and finally, a 20% reduction in overall energy demands by
2020. By setting these targets, the next step is identifying how
2. The Irish energy-system
these targets can be met, and if they are met, what is their

The island of Ireland is located in the North-West of Europe and

is divided into two countries: Northern Ireland and Ireland. In
* Corresponding author. Tel.: þ353 87 6401379; fax: þ353 61 202423. November 2007 the single electricity market operator (SEMO) was
E-mail addresses: [email protected] (D. Connolly), [email protected] (H. Lund),
[email protected] (B.V. Mathiesen), [email protected] (M. Leahy).
formed [5], which created a single electricity market on the island
1
Ireland refers to the Republic of Ireland only throughout this paper, unless of Ireland. However, prior to this, the energy-systems in each of
otherwise stated. these countries were only connected via an electric interconnector.

0360-5442/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.energy.2010.01.037
 D. Connolly et al. / Energy 35 (2010) 2164–2173 2165

Table 1 (excluding industrial processes) was 31.5 TWh, with 0.05% of this
Efficiencies calculated for power-plants of different fuel-type [7–10]. supplied by solar thermal, 13% supplied by electric heating, and 87%
Plant type Capacity Electricity Fuel used Efficiency supplied by individual boilers. The transport sector in Ireland was
(MW) generated (TWh) (TWh) (%) almost completely powered by oil in 2007, including 31.34 TWh of
Natural gas PP 3525 13.38a 28.72 46.58 diesel, 22.33 TWh of petrol, and 12.13 TWh of jet fuel. The only
Coal PP 852 5.35 12.95 41.31 other fuel used was biofuel which provided 0.4% (0.249 TWh) of the
Peat PP 345 2.11 5.05 41.78
transport demand. It is evident from the structure outlined here
Oil PP 1014 1.94 4.18 46.41
Wind 724 1.88 – – that the Irish energy-system is very segregated: other than electric
Natural gas CHP 273 1.83 6.08 30.10a heating there is no significant interaction between the electricity,
Net import 220b 1.31 – – heat and transport sectors. This is the situation within numerous
Hydro 216 0.66 – – other countries also [6], and results in both an inefficient and a rigid
Pumped hydro 292 0.35 0.55 63.63
Biomass Co-combusted 0.14 0.41 34.15
energy-system.
a
A prime example of inefficiencies within the Irish energy-
The thermal efficiency of the natural gas CHP plants is 53.36%.
b system occurred within the electricity sector in 2007. Electricity
This is the maximum interconnection capacity used between Ireland and
Northern Ireland in 2007. generation in 2007 included a net import of 1.31 TWh from
Northern Ireland as displayed in Table 1 [7–10]. Therefore,
27.19 TWh of electricity was generated within Ireland, with 84% of
As the model in this paper was developed as a reference model by this coming from condensing power-plants using fossil fuels. As
simulating the year 2007, this paper only covers the energy-system condensing power-plants can only produce electricity, the effi-
in Ireland, although future models will include the electricity sector ciencies of these plants are relatively low: as displayed in Table 1
of Northern Ireland also. the average efficiency is 44%. In comparison, the CHP facilities in
Ireland has a population of approximately 4.4 million people Ireland have an efficiency of 82.8%, but they are only used to
and an area of approximately 70,000 km2. In 2007, the electricity produce 6.5% of the electricity demand. Due to the widespread use
demand was 28.5 TWh with a peak load of 5085 MW and of condensing power-plants and lack of CHP plants, 50.5% of the
a minimum load of 1800 MW. The heat demand in Ireland for 2007 energy used in the Irish electricity sector in 2007 was wasted due to

Fig. 1. The structure of the EnergyPLAN tool [18].

2166 D. Connolly et al. / Energy 35 (2010) 2164–2173

Table 2 Table 4
Comparison of average monthly electricity demands obtained from the EnergyPLAN Comparison of total fuel consumed in Ireland in 2007 and the EnergyPLAN
model and actual values for 2007. simulation.

Month Average monthly electricity demand (MW) Difference Difference Fuel 2007 fuel consumption EnergyPLAN fuel Difference
(MW) (%) (TWh) consumption
Actual 2007 EnergyPLAN 2007 TWh %
January 3564 3559 5 0.14 Oil 105.22 104.44 0.78 0.74
February 3576 3573 3 0.09 Natural gas 49.92 50.41 0.49 0.98
March 3414 3386 28 0.82 Coal/Peat 25.70 25.76 0.06 0.23
April 3079 3084 5 0.18 Biomass 2.77 2.83 0.06 2.17
May 3029 3025 4 0.14 Renewables 2.54 2.59 0.05 1.97
June 2991 2970 21 0.71
July 2937 2947 10 0.34
August 2964 2960 4 0.15
Irish electricity grid for the year 2020 using the WILMAR energy
September 3094 3105 11 0.36
October 3279 3281 2 0.07 model [16]. The objective of this study was to identify the effects of
November 3515 3508 7 0.20 large wind penetrations on the island of Ireland in relation to
December 3531 3519 12 0.35 overall operation, costs, and emissions. Meibom et al. concluded
that a wind penetration of 42% was feasible on the island of Ireland
by 2020, which will reduce overall operation costs and the CO2
losses [7]. From an economic point-of-view, this amounts to emissions compared to 2007. In summary, a number of studies have
approximately V420 million of wasted fuel each year, which will already been carried out analysing the integration of wind energy
only become more severe if a transition is not made to local in Ireland. However, both Gardner et al. [13] and ESB National Grid
renewable energy resources. [14] used predicted data when analysing the year 2007, while
In relation to the rigidity of the Irish energy-system, the flexi- Meibom et al. [15] focused on feasible wind penetrations in the year
bility of the existing system needs to be improved to enable large- 2020. Therefore, the aim of this study is to outline the wind
scale renewable energy penetrations by integrating the electricity, penetrations that can be achieved on the 2007 Irish energy-system,
heat, and transport sectors more effectively. This is discussed in along with the economical and technical implications, using actual
detail in [11] by analysing the benefits of technologies such as CHP, historical data. This will illustrate the wind penetrations that can be
heat pumps, electric vehicles, hydrogen, etc., in combination with achieved immediately, without any major alterations to the Irish
large-scale renewable energy. The rigidity of the existing Irish energy-system.
energy-system is evident in Table 1, as renewable energy only
supplied 9.5% of the electricity generated in 2007 even though 3. Methodology
onshore wind alone could potentially supply 130% of the electricity
demand in Ireland [12]. Therefore, it is imperative that alternatives To create a model of the Irish energy-system, a suitable energy
are proposed to utilise these resources more effectively. Conse- tool needed to be identified. An investigation was carried out to
quently, by developing a model of the Irish energy-system future identify which tool would be the most suitable with the following
alternatives can be examined to propose a more efficient, envi- key objective: to identify how Ireland could integrate the most
ronmentally friendly, and economical energy-system. renewable energy into its energy-system. A detailed report
Finally, a number of other studies have been already investi- comparing the functionality of 37 different tools has been
gated the feasibility of integrating wind energy onto the Irish completed [17], where it was concluded that EnergyPLAN was the
electric grid. In 2003, Gardner et al. [13] investigated the effects of most suitable tool to meet this objective. Therefore this comparison
additional wind energy in Ireland and identified that the most will not be discussed in detail here, but instead the primary reasons
costly aspects of increasing the wind penetration are transmission that EnergyPLAN was chosen are outlined, followed by a brief
reinforcement, wind curtailment, capital costs, and operating costs. summary of the tool itself.
In 2004, Electricity Supply Board (ESB) National Grid [14] also Firstly, EnergyPLAN considers the three primary sectors of any
analysed the costs and implications for conventional power-plants national energy-system: electricity, heat, and transport. As outlined
associated with increasing the wind penetration in Ireland. The in [11,18], the integration of these three sectors is crucial in order to
report concluded that increasing the wind penetration in Ireland
from 0% to 11.7% would increase the total generation costs by V196
Table 5
million, while peaking and mid-merit power-plants would require
CO2 emissions for Ireland in 2007 and CO2 emissions from the EnergyPLAN
more frequent start-ups, need increased ramping, and have lower simulation.
capacity factors. Finally, in 2007, Meibom et al. [15] modelled the
Fuel Consumption [9] CO2 emission CO2 emitted
(TWh) factor [7] (kg/GJ) (Mt)

Table 3 Gas oil 45.188a 73.30 11.92

Comparison of electricity produced for Ireland in 2007 and the EnergyPLAN Electricity 25.867 150.83 14.05
simulation. Gasoline 22.325 70.00 5.63
Natural gas 18.424a 57.10 3.79
Production unit 2007 production [9] (TWh) EnergyPLAN Difference Jet kerosene 12.134 71.40 3.12
production Kerosene 10.620 71.40 2.73
TWh %
2007 (TWh) Coal 4.354a 94.60 1.48
Power-plants 23.56 23.54 0.02 0.08 Fuel oil (residual oil) 4.295a 76.00 1.18
Coke 3.637 100.80 1.32
a
Onshore wind 1.88 1.86 0.06 3.20 Sod Peat 2.167 104.00 0.81
Offshore wind 0.08 LPG 1.853a 63.70 0.42
Peat Briquettes 0.992 98.90 0.35
Industrial CHP 0.93 0.93 0.00 0.00
Naphtha 0.012 73.30 0.003
Hydro power 0.66 0.65 0.01 1.52 Total 46.80
a a
Onshore and offshore data could not be obtained separately. Excludes fuel required for electricity generation.
 D. Connolly et al. / Energy 35 (2010) 2164–2173 2167

Fig. 2. Annual operating costs of the 2007 Irish energy-system for various fuel and CO2 prices, using a real interest rate of 3%.

achieve large-scale penetrations of renewable energy. However, simulation over a period of one year. The structure of EnergyPLAN
Ireland has very little integration between these sectors and is illustrated in Fig. 1 [18]. General inputs are the demands,
therefore to utilise more renewable energy, it is essential that the renewable energy sources, energy station capacities, costs, and
energy-system is integrated more effectively. With this in mind, a number of optional regulation strategies. Outputs are energy
EnergyPLAN has a key advantage over a number of others tools balances and the resulting annual productions, fuel consumption,
considered. Secondly, EnergyPLAN has already been used to analyse import/export of electricity, and the total costs of the system
several energy scenarios which are similar to the long-term including income from the exchange of electricity. EnergyPLAN
objectives of this research. These include analysing the effects of uses analytical programming rather than iterations so the calcu-
large wind penetrations [19], the optimum combination of various lations are completed in a very short period of time. Also, the
renewable energy technologies in an energy-system [20,21], the simulation process in EnergyPLAN has been kept relatively simple
benefits of energy storage [22–24], the benefits of integrating the by aggregating all units in the various sectors mentioned and
electricity and heat sectors [25,26], as well as the transport sector hence, not considering the differences in single units and the
[27], and finally the pathway towards a 100% renewable energy- transmission among them. Otherwise, EnergyPLAN provides an
system for Denmark [28,29]. These are typical of the studies advanced representation of the entire energy-system, by using
required to identify how Ireland can work towards its 2020 energy hourly distributions of heat demands, electricity demands, wind
targets and beyond. Finally, EnergyPLAN was also chosen for this production, wave production etc., as well as detailed operational
study as it can be adapted to simulate a wide variety of national strategies. A more detailed description of EnergyPLAN can be
energy-systems. For example, EnergyPLAN has previously been found at [31].
used to analyse the energy-systems in Denmark, Estonia, Germany, The technical inputs, assumptions, and sources used to create
Poland, Scotland, and Spain [30]. As these energy-systems use the reference model of the Irish energy-system in this paper are
similar technologies to the Irish energy-system, it was evident that only relevant to Ireland. Consequently, these are not discussed
EnergyPLAN could be used for this study. here, but instead a full description of the technical data used is
The main purpose of EnergyPLAN is to assist in the design of publicly available at [32]. More applicable to a global audience is
national or regional energy-planning strategies on the basis of the financial data used which is provided in the Appendix, as well
technical and economic analysis, resulting from the imple- as the accuracy of the model created which is illustrated in the
mentation of different energy-systems and investments. Ener- results. Finally, the ‘cost sensitivity’ and ‘maximum wind pene-
gyPLAN is a deterministic input/output tool which uses an hourly tration’ results obtained in this study for Ireland are extremely

Fig. 3. Annual operating costs of the 2007 Irish energy-system for various fuel and CO2 prices, using a real interest rate of 6%.
2168 D. Connolly et al. / Energy 35 (2010) 2164–2173

4.1. Accuracy of the Irish reference model

Once the inputs were gathered, the reference model was

simulated on a one-hour time resolution over the year 2007. The
initial inputs have been continuously updated to improve the
accuracy of the simulation, until a high level of accuracy was
obtained. In this section, a comparison was made between the most
recent version of the reference model and the actual figures from
2007.
The first parameter that was compared was the electricity
demand. The total electricity generated for 2007 (28.5 TWh),
including a 1.31 TWh net import was being simulated correctly in
the model. Also, the distribution of the electricity generated over
the year was being simulated correctly, as indicated by the average
Fig. 4. Critical excess electricity production for increasing wind penetrations on the monthly electricity demands displayed in Table 2.
2007 Irish energy-system. Once it was verified that the electricity demand was being
simulated correctly, the electricity produced from various units was
compared. As seen in Table 3, the total electricity generated from
applicable to numerous other countries. This is due to the similar the various production units is very similar in both the actual 2007
structure of their segregated energy-systems i.e. the electricity, figures [9] and the results from the reference model. The only
heat, and transport sectors are very independent one another, as significant difference occurred for wind power production, which is
displayed in [6]. Therefore, the results obtained in this study most likely attributed to the 8.5% variation in installed wind
illustrate the issues ahead for numerous countries, in relation to capacity at the beginning and end of 2007.2 As power-plants
energy costs and integrating large amounts of fluctuating contributed such a large proportion of the electricity supply,
renewable energy. a further comparison was made for them.
Power-plant production could not be compared individually
4. Results because EnergyPLAN aggregates the power-plants within an
energy-system and consequently, the production from each power-
Once the data was gathered and the model created using plant is not available from the results. Therefore, electricity
EnergyPLAN, three analyses were completed. Firstly, a comparison production was not compared for each power-plant, but instead the
was made between the reference model created, and the actual annual fuel consumed by each fuel-type of power-plant was
performance of the Irish energy-system in 2007. This was compared: the fuel-types were natural gas, coal, oil, and biomass.
completed to ensure that EnergyPLAN was capable of providing an From this comparison it was clear that the model provides an
accurate simulation of the Irish energy-system and follows a less accurate representation of the actual events on the Irish energy-
detailed comparison which was completed in [33]. Secondly, the system in 2007, as the largest difference that occurred was 0.47%,
sensitivity of the existing Irish energy-system to various economic which was for oil based power-plants.
parameters was analysed: these include fuel prices, CO2 prices, and After the electricity sector was analysed, the heat and transport
interest rates. This analysis was completed to identify future energy sectors were compared with the reference model. However, all heat
costs within the Irish energy-system. Finally, as wind is the most in Ireland is produced by individual boilers and all transport is
abundant fluctuating renewable energy available in Ireland, a study powered by conventional vehicles. Therefore, due to the lack of
was completed analysing the maximum wind penetration that integration between the sectors in the Irish energy-system, no
could be achieved in Ireland, from a technical and economical hourly simulations are necessary in the heat or transport sectors.
point-of-view. The objective of this investigation was to illustrate The only input required is the annual fuel requirements which are
the potential for immediate integration of a fluctuating renewable used as inputs in EnergyPLAN. Therefore, comparing the Ener-
energy onto the Irish energy-system, without any major redevel- gyPLAN results with the actual data from 2007 would result in no
opments. To reinforce this argument, the impact of large fluctuating difference, as it would be the same data. Therefore, for the heat and
renewable energy penetrations on conventional power-plants is transport sectors, the accuracy of the model needs to be based on
also examined in detail. the assumptions made while constructing the input data, not on the
figures produced by the model. As these are very specific to the Irish
energy-system, these are not discussed here but can be freely
obtained from [32].
Next the total fuel consumption within the Irish energy-system
is compared with those calculated in EnergyPLAN. As seen in
Table 4, the total fuel consumptions from actual 2007 figures and
from the reference model are very similar for all fuels: the largest
relative-difference occurred for biomass at 2.17%.
Finally, the actual CO2 emissions for Ireland in 2007 were
compared with those from the EnergyPLAN simulation. The total
energy-related CO2 emissions for Ireland in 2007 were calculated as
46.8 Mt using fuel consumptions from [9] and emission factors
from [7], as seen in Table 5. In comparison, EnergyPLAN calculated

2
Fig. 5. Primary energy supply on the 2007 Irish energy-system for increasing wind There was an 8.5% increase in wind capacity in Ireland in 2007 from 723.8 MW
penetrations. to 785.2 MW.
 D. Connolly et al. / Energy 35 (2010) 2164–2173 2169

Table 6
CEEP, PES, COMP, and costs for various wind penetrations on the 2007 Irish energy-system.

Wind energy, Wind penetration (%) CEEP (TWh/year) PES (TWh/year) COMP DPES/DCEEP (–) System costsa 2007 fuel System costsa 2020 fuel
TWh (MW) prices (MV/year) prices (MV/year)
2 (724)b 6.53 0.00 188.92 – 8849 12,184
5 (1846) 17.56 0.00 185.27 – 8839 12,101
6 (2222) 21.07 0.00 184.14 – 8837 12,076
7 (2598) 24.58 0.04 183.19 24.25 8837 12,055
8 (2973) 28.09 0.15 182.47 6.55 8842 12,041
9 (3349) 31.60 0.38 182.04 1.87 8853 12,037
10 (3725) 35.11 0.74 181.89 0.42 8871 12,041
11 (4100) 38.62 1.23 181.97 0.16 8894 12,054
12 (4476) 42.13 1.81 182.26 0.50 8921 12,074
a
Based on a CO2 price of $30/t and a real interest rate of 6%.
b
This is the wind energy on the reference energy-system for 2007.

the CO2 emissions for Ireland in 2007 as 47.21 Mt. This is 0.88% analysing the costs of the future costs in the Irish energy-system
(0.41 Mt) higher than those calculated from the statistics, and thus along with 6% to illustrate the sensitivity of the results. Finally,
indicates that the reference model is providing an accurate repre- the annual operating costs calculated are not absolute energy-
sentation of the Irish energy-system. system costs as they do not include costs outside of the indi-
After completing the comparison between the reference model vidual technologies, i.e. transmission lines, but they do enable the
and the actual 2007 figures, it was concluded that the model was various alternatives to be directly compared as this is a business-
accurate: the largest difference recorded was 2.17%. Therefore, the as-usual scenario. Also, investment costs were not included for
model was used in this paper to analyse the economic sensitivity the transport sector.
and maximum feasible wind penetrations for the existing Irish By analysing the results displayed in Figs. 2 and 3, the sensitivity
energy-system. of the existing Irish energy-system to various economic parameters
can be identified. It is evident from these results, that fuel prices
4.2. Economic sensitivity of Irish energy-system form the most substantial part of the Irish energy-system’s annual
costs. In addition, the CO2 price also has a significant effect on the
In this section the reference model was used to identify the overall system costs, with CO2 penalties costing approximately the
sensitivity of the Irish energy-system to fuel prices, CO2 prices, same, or more, than the investment costs for each scenario ana-
and interest rates using the cost data discussed in the Appendix. lysed. Also, the annual O&M costs make up a very small amount of
For the simulation, no technical alterations were made to the Irish the overall system costs. This is partly due to the exclusion of
energy-system between now and 2020 so the effects of a busi- investment costs in the transport sector, but also due to the
ness-as-usual scenario could be illustrated. To illustrate the segregated system currently in use in Ireland. As the power-plants
economic consequences of a business-as-usual scenario, the first used in the Irish energy-system are designed for a single purpose
step was to calculate the annual operating costs of the Irish they are relatively easy to operate i.e. condensing power-plants
energy-system using the historical oil and CO2 costs for 2007, produce electricity only, boilers produce heat only, and conven-
which were $69.33/bbl and $30/t respectively [34]. Afterwards, tional vehicles provide transport. This causes a relatively simple (i.e.
the annual operating costs were also calculated for future oil and low investment and O&M costs) but inefficient (i.e. high fuel costs)
CO2 prices: oil prices of $100/bbl and $110/bbl were used as these system.
are predicted prices for 2010/2015 and 2020 respectively [34], This costs analysis also indicates that future oil and CO2 prices
while a second CO2 cost of $60/t was used as it is double the will have huge implications on the future cost of energy in Ireland.
existing CO2 cost of $30/t, which is still lower than many of the By 2020, if the cost of oil is $110/bbl and the cost of CO2 remains at
estimated costs required to stabilise CO2 emissions at an accept- $30/t, then the annual cost of fuel for the Irish energy-system will
able level [35]. In addition, the calculations were completed using be 44% higher than they are today. If the CO2 price does double in
two real interest rates: 3% (see Fig. 2) and 6% (see Fig. 3). The order to mitigate climate change, then the annual cost of fuel for the
energy regulator in Ireland, Commission for Energy Regulation Irish energy-system in 2020 will be 58% more expensive than today.
(CER), typically uses a real interest rate of approximately 3–4% This economic risk illustrates how important it is for Ireland to
when assessing the economic potential for constructing new reduce its dependence on all fossil fuels, and convert to a renewable
power-plants [36,37]. Therefore, 3% was also used in this study for energy-system.

Table 7
Hourly fluctuations in power-plant output for various wind penetrations.

Wind energy Wind Max ramp Max ramp Number of hours with ramp ups Number of hours with ramp downs
(TWh) penetration (%) up (MW) down (MW)
>1000 MW >500 & >250 & >1000 MW >500 & >250 &
<1000 MW <500 MW <1000 MW <500 MW
2a 6.77 785 692 0 241 756 0 7 759
5 17.56 868 1042 0 258 806 1 23 895
6 21.07 897 1157 0 271 812 1 32 923
7 24.58 934 1272 0 282 810 1 50 927
8 28.09 1051 1386 1 282 794 1 59 922
9 31.60 1179 1501 1 280 782 1 83 891
10 35.11 1307 1616 2 274 778 1 89 886
11 38.62 1387 1730 2 275 762 1 102 878
12 42.13 1336 1845 6 263 756 1 111 865
a
This is the wind energy on the reference energy-system for 2007.
2170 D. Connolly et al. / Energy 35 (2010) 2164–2173

Table 8
Fuel prices used for 2007, 2010/2015 and 2020 [34,38].

(V/GJ) Crude oil ($/bbl) Crude oil Fuel oil Gas oil/diesel Petrol/JP Coal Natural gas Biomass
2007 69.33 9.43 6.66 11.79 12.48 1.94 5.07 6.30
2010/2015 100 13.60 9.60 17.00 18.00 3.19 8.16 7.01
2020 110 14.96 10.56 18.70 19.80 3.11 9.16 7.45

4.3. Maximum feasible wind penetration on Irish energy-system 0.15 TWh. Therefore, the additional 1 TWh of wind has caused the
system to generate twice as much energy that it can no longer
The final analysis completed using this reference model was an consume than it has saved. Consequently, it is evident that the
investigation into the maximum wind penetration that could be largest feasible wind penetration defined by COMP is 31.6%: after
achieved on the 2007 Irish energy-system. The critical excess this point the reduction in energy consumption is less than the
electricity production (CEEP)3 and the primary energy supply increase in CEEP (i.e. COMP < 1), and before this point the reduction
(PES)4 for the entire Irish energy-system were recorded, as the in energy consumption is larger than the increase in CEEP (i.e.
annual wind energy production was varied from 0% to 105% COMP > 1). For some systems this could be classified as a conser-
(0–30 TWh) of the annual electricity production in Ireland: see vative approach: if fossil fuel prices are extremely high, then the
Fig. 4 and Fig. 5, respectively. wasted energy created by the addition of wind could be classified as
The CEEP results in Fig. 4 and Table 6 illustrate that no excess acceptable. Therefore, the costs of the Irish energy-system were
electricity occurs on the Irish energy-system up to a wind energy also analysed to compare with COMP.
penetration of approximately 21%. After this point, the CEEP Using a CO2 price of $30/t and a real interest rate of 6%, which is
increases relatively slowly until a wind penetration of approxi- the worst case scenario for investing in wind power in the future,
mately 50%, when the gradient increases substantially. Mean- the cost of the existing Irish energy-system was calculated for
while, the lowest fuel consumption shown in Fig. 5 and Table 6 various wind penetrations using 2007 and 2020 fuel prices, which
occurs at a wind penetration of approximately 36%, with similar is displayed in Table 6. Firstly, these results indicate that the
increases in fuel demand above and below this point. An optimum existing Irish energy-system is more expensive than alternative
wind penetration could not be identified using these initial energy-systems with higher wind penetrations, under both 2007
results, but boundaries could be created. The minimum wind and 2020 fuel prices. Secondly, although a wind penetration of
penetration that is advantageous to the Irish energy-system is at 31.6% was identified previously as the optimum from a technical
least 21% as up to this point there is no CEEP and fuel consumption point-of-view, a wind penetration of 21% is the optimum from an
is reducing. Also, the maximum wind penetration that should be economic point-of-view when considering 2007 fuel prices.
used is approximately 36%, as at this point both the CEEP and PES However, when using 2020 fuel prices, the technical ‘optimum’
are increasing. Therefore, the ‘optimum technical’ wind penetra- becomes the same as the economic ‘optimum’, at a wind penetra-
tion for the existing Irish energy-system in 2007 is between 21% tion of 31.6%. This is due to the sensitivity of the existing Irish
and 36%. energy-system to fuel prices, as previously displayed in Figs. 2 and
To identify the technically optimum point, a compromise is 3. The economic results from Table 6 indicate that as fuel prices
needed between the increase in CEEP and the resulting impact on increase to 2020 levels, the most economical energy-system will be
the PES, as the wind penetration increases. As a result, a compro- achieved in Ireland by integrating the maximum wind penetration
mise coefficient, COMP, was created which is the ratio between the that is technically feasible. It is worth noting that the ‘optimum’
reduction in PES, DPES, and the increase in CEEP, DCEEP, as the wind wind penetrations discussed here are only for the 2007 Irish
penetration increases from one simulation to the next: energy-system. Future alterations to the Irish energy-system have
not been included (but will be in future studies), and hence larger
DPES feasible wind penetrations are likely to be identified in future
COMP ¼ (1)
DCEEP studies.
The COMP coefficient illustrates the benefits of adding wind Finally, the next issue discussed is the effect of increasing wind
capacity (i.e. a reduce energy consumption and hence a lower PES), penetrations on conventional power-plants. The hourly demand on
against the disadvantages of adding wind capacity (i.e. increasing power-plants was analysed for different wind penetrations, and the
the fluctuating power on the energy-system and hence increasing scale and frequency of power-plant ramping was analysed. The
CEEP). The maximum feasible wind penetration occurs when the results are displayed in Table 7 and show that higher wind pene-
COMP coefficient is less than 1. Therefore, the COMP coefficient trations increase the demand for ramping power-plants both up
states that wind energy should no longer be added to the system and down. For a wind penetration of 31.60% (which was previously
when, the increase in electricity that you are forced to export is deemed the optimum wind penetration in the existing Irish
greater than the reduction in energy that you need to power your energy-system), the power-plants in the Irish energy-system would
system. From the results in Table 6 for example; when the wind need to be able to ramp up by a maximum of 1179 MW, and ramp
energy on the reference system is increased from 9 TWh to 10 TWh, down by a maximum of 1501 MW. In addition, the power-plants
then the increase in forced electricity exports is increased by would need to be able to ramp up/down between 500 MW and
0.36 TWh and the total energy demand is only decreased by 1000 MW 280/83 times respectively, and between 250 MW and
500 MW 782/891 times respectively. It is evident that ramping
demands on power-plants between 250 and 500 MW, for a wind
penetration of 31.60% and the reference model are very similar.
3
CEEP is the amount of excess electricity produced that could not be used in the However, there are significant increases in ramping requirements
energy-system. The consequences of CEEP are forced export (if adequate intercon-
of þ/500 MW as wind penetrations increase. Therefore, ramping
nection capacity does not exist) or stopping the wind turbines to reduce
production.
capabilities within this range should be a key issue for future
4
PES is the total fuel consumption within the energy-system, which includes power-plants that are built on the Irish energy-system, to
electricity, heat, and transport. encourage the integration of more renewable energy.
 D. Connolly et al. / Energy 35 (2010) 2164–2173 2171

Table 9
Fuel handling costs [38].

V/GJ Fuel oil Gas oil/diesel Petrol/JP Coal Natural gas Biomass
Power stations (central) 0.228 0.228 – 0.067 0.428 1.160
Distributed CHP, district heating & industry 1.914 1.807 – – 1.165 1.120
Individual households – 2.905 – – 2.945 6.118
Road transport – 3.159 4.257 – – 11.500 [39]
Airplanes – – 0.696 – – –

Table 10
Investment, fixed O&M, and variable O&M costs for Irish condensing power-plants [41].

Plant type Investment costs Fixed O&M costs Variable O&M costs 2007 Irish capacity/
(MV/MW) (V/MW/year) (V/MWh) fuel-type
Steam turbine, coal fired, advanced steam process, 2004 1.100 16,000 1.800 852.5 MW/Coal; 806 MW/Oil
Steam turbine, coal fired advanced steam process, 20% 1.200 22,000 3.000 345.6 MW/Peat
co-firing of biomass, 2004
Gas turbine single cycle (40–125 MW), 2004 0.485 7350 2.500 719 MW/Gas
Gas turbine combined cycle (100–400 MW), 2004 0.525 14,000 1.500 2806 MW/Gas
Gas turbine combined cycle (10–100 MW), 2004 0.700 10,000 2.750 208 MW/Gas

Table 11
Costs (excluding taxes) of individual heating systems for the reference model of the Irish energy-system.

Fuel-type Size Cost including installation (V) Lifetime (years) O&M costs (V/year)
Oil 26 kW 14,750 15 110
Biomass 19 kW 19,500 15 110
Natural gas 26 kW 14,750 15 110
Solid fuel 21 kW 15,300 15 110
Electric boiler 12 kW 15,500 15 0
Electric heaters 20 kW 6000a 20 0
Solar thermal 2400 kWh/year 5900 35 55
a
Does not account for electric transmission upgrades that may be necessary for widespread installations.

5. Conclusions Although these results have provided some excellent indications

for the future, the next step will be to develop alternatives based on
The development of an energy model for the Irish energy- future energy-systems and not the existing energy-system. Alter-
system has been discussed in this study. Firstly, the accuracy of native energy-systems will then be designed with a particular focus
the model has been verified by comparing the results of the model on the integration of renewable energy. As the CEEP, PES, and hence
with actual statistics from the year 2007. From these results it is the COMP coefficient are system dependent, i.e. dependent on the
clear that the model is sufficiently accurate for use in future studies system analysed, changes on both the production side and the
that will focus on the integration of renewable energy into the Irish consumption side will affect the values. Therefore, it is anticipated
energy-system. After validating the accuracy of the model, this that much higher renewable energy penetrations will be identified
study then illustrated the huge vulnerability of the Irish energy- for the Irish energy-system, when future alternative scenarios are
system to future fuel prices and CO2 prices. Results illustrate that incorporated into the model. Future alternatives could involve
fuel costs could increase by at least 44% in 2020 due to fuel price analysing any of the following in Ireland: large-scale CHP and
increases alone, and by an additional 14% if CO2 prices are doubled district heating, electric vehicles, energy storage, increased inter-
to mitigate climate change. Afterwards, the maximum wind connection, introduction of demand side management, etc.
penetration feasible on the Irish energy-system from a technical
and economic perspective was identified. It was concluded that Acknowledgements
based on 2020 fuel prices, both the technically and economically
optimum wind penetration occurs at approximately 30% for the The findings in this article result from a research project at the
existing Irish energy-system. Finally, the effect of high wind University of Limerick, which has been supported financially by the
penetrations on conventional generation was investigated. Here, it Irish Research Council of Science, Engineering & Technology (IRC-
was concluded that the ramping capabilities of the power-plants SET) and Science Foundation Ireland (SFI) under Grant Number 06/
needs to increase by approximately double to incorporate a wind CP/E007. Preliminary results of the paper were presented and dis-
penetration of approximately 30%. If this can be achieved with cussed at the Joint Action on Climate Change (JAOCC) conference in
existing power-plants, then this wind penetration can be achieved Aalborg, Denmark, on the 10th June 2009 [33].
now on the existing Irish energy-system. Finally, using this model
will be used in future studies to identify even larger renewable
Appendix. Cost data used
energy penetrations for Ireland.
To complete the economic analysis in this study, a number of
6. Future work costs had to be gathered. The purchasing costs for each fuel were
obtained for the years 2007, 2010/2015, and 2020 and are displayed
All the results displayed in this paper were obtained by ana- in Table 8. These are predicted prices from the International Energy
lysing the existing Irish energy-system based on the year 2007. Agency [34] and the Danish Energy Authority [38].
2172 D. Connolly et al. / Energy 35 (2010) 2164–2173

The crude oil price was used to identify the cost of fuel oil, diesel, [7] Howley M, Ó’Gallachóir B, Dennehy E. Energy in Ireland 1990–2007. Sustain-
able Energy Ireland. Available at: http://www.sei.ie/Publications/Statistics_
and petrol. As these fuels are refined from crude oil, their prices are
Publications/EPSSU_Publications; 2008.
proportional to the crude oil price and hence, the price ratio [8] EirGrid. System operations. EirGrid. Available at: http://www.eirgrid.com/
between each of these and crude oil typically remains constant. operations/systemperformancedata/downloadcentre; 2009.
Therefore, the following ratios recommended by the Danish Energy [9] Sustainable Energy Ireland. Energy balance 2007. Sustainable Energy Ireland.
Available at: http://www.sei.ie/Publications/Statistics_Publications/2007_Energy_
Authority were used to calculate these prices [38]: ratio of crude oil Balance; 2008.
to fuel oil was 1–0.7, crude oil to diesel was 1–1.25, and crude oil to [10] Barriscale A. Personal communication at the energy policy statistical support
petrol was 1–1.33. Also, the fuel handling costs were obtained from unit in sustainable energy Ireland (SEI). Dublin, Ireland; 2008.
[11] Mathiesen BV, Lund H. Comparative analyses of seven technologies to facili-
the Danish Energy Agency [38] and Sustainable Energy Ireland [39] tate the integration of fluctuating renewable energy sources. IET Renewable
as displayed in Table 9. Power Generation 2009;3(2):190–204.
After consulting with the Irish revenue office, it was found that [12] Electricity Supply Board International (ESBI), Future Energy Solutions, Energy
Research Group (University College Dublin). Renewable energy resource:
there are currently no taxes (other than value added tax) placed on Ireland to 2010 and 2020. Sustainable Energy Ireland. Available at: http://
fuels in Ireland. Also, at the time of this study there was no carbon www.sei.ie/uploadedfiles/FundedProgrammes/REResources20102020MainReport.
tax in Ireland. However, Ireland does participate in the European pdf; 2004.
[13] Gardner P, Sodin H, Higgins A, McGoldrick S. The impacts of increased levels of
carbon trading scheme so these costs had to be accounted for. To wind penetration on the electricity systems of the Republic of Ireland and
identify this cost, historical trading prices were obtained from [40] Northern Ireland. Commission for Energy Regulation. Available at: http://
and an average value of V20.63/tCO2 ($30/tCO2) was found for the www.cer.ie/cerdocs/cer03024.pdf; 2003.
[14] Electricity Supply Board National Grid (ESBNG). Impact of wind power
year 2007.
generation in Ireland on the operation of conventional plant and the economic
The condensing power-plant costs were obtained from a report implications. EirGrid. Available at: http://www.eirgrid.com/media/2004%20
completed by the Danish Energy Authority, Ekraft System, and wind%20impact%20report%20(for%20updated%202007%20report,%20see%20
Eltra5 [41], and are displayed in Table 10. Using these details the above).pdf; 2004.
[15] Meibom P, Barth R, Brand H, Hasche B, Ravn H, Weber C. All-island grid study:
aggregate cost of condensing plant on the Irish energy-system was wind variability management studies. Department of Communications Energy
calculated as 0.733 MV/MW, with a fixed O&M cost of 14,081 V/ and Natural Resources (Ireland). Available at: http://www.dcenr.gov.ie/
MW/year and a variable O&M cost of 1.84 V/MWh. Energy/LatestþNews/All-IslandþGridþStudyþPublished.htm; 2008.
[16] Risø National Laboratory. WILMAR: wind power integrated in liberalised
The onshore wind and offshore wind costs were also obtained electricity markets. Risø National Laboratory. Available at: http://www.wilmar.
from the Danish Energy Authority [42]: investment costs for risoe.dk.
onshore wind are 1.2 MV/MW and offshore wind is V1.6 MV/MW, [17] Connolly D, Lund H, Mathiesen BV, Leahy M. A review of computer tools for
analysing the integration of renewable energy into various energy systems.
while the fixed O&M costs are 6 V/MWh for onshore wind and Applied Energy 2010;87(4):1059–82.
8.70 V/MWh for offshore wind.6 The investment cost for the hydro [18] Lund H. Choice awareness and renewable energy systems. Aalborg: Depart-
power in Ireland was obtained from the British Hydropower ment of Development and Planning, Aalborg University; 2009.
[19] Lund H. Large-scale integration of wind power into different energy systems.
Association [43]: the investment cost for hydro stations below Energy 2005;30(13):2402–12.
100 MW is 1.765 MV/MW, the fixed O&M costs are approximately [20] Lund H. Large-scale integration of optimal combinations of PV, wind and
2.7% of the investment, while the variable O&M costs are approx- wave power into the electricity supply. Renewable Energy 2006;31(4):
503–15.
imately 1.3% of the investment. The costs for pumped hydroelectric
[21] Lund H, Munster E. Management of surplus electricity-production from
energy storage in Ireland were found from [44] as V0.473.6 MV/ a fluctuating renewable-energy source. Applied Energy 2003;76(1–3):65–74.
MW and V7.89/GWh for the initial investment, 3000 V/MW for the [22] Blarke MB, Lund H. The effectiveness of storage and relocation options in
fixed O&M cost, and 3 V/MWh for the variable O&M cost. renewable energy systems. Renewable Energy 2008;33(7):1499–507.
[23] Lund H, Salgi G. The role of compressed air energy storage (CAES) in future
Finally, the investment and O&M costs were obtained for the sustainable energy systems. Energy Conversion and Management 2009;50(5):
individual heating systems used in the 2007 Irish energy-system as 1172–9.
displayed in Table 11. This ensured that all costs within the heating [24] Lund H, Salgi G, Elmegaard B, Andersen AN. Optimal operation strategies of
compressed air energy storage (CAES) on electricity spot markets with fluc-
and electricity sectors were accounted for in the reference model. tuating prices. Applied Thermal Engineering 2009;29(5–6):799–806.
[25] Lund H, Munster E. Modelling of energy systems with a high percentage of
CHP and wind power. Renewable Energy 2003;28(14):2179–93.
[26] Lund H, Munster E. Integrated energy systems and local energy markets.
References Energy Policy 2006;34(10):1152–60.
[27] Lund H, Kempton W. Integration of renewable energy into the transport and
[1] Europa. Boosting growth and jobs by meeting our climate change commit- electricity sectors through V2G. Energy Policy 2008;36(9):3578–87.
ments. Europa. Available at: http://europa.eu/rapid/pressReleasesAction. [28] Lund H, Mathiesen BV. Energy system analysis of 100% renewable energy
do?reference¼IP/08/80&format¼HTML&aged¼1&language¼EN&guiLanguage¼ systems – the case of Denmark in years 2030 and 2050. Energy 2009;34(5):
en; 2009. 524–31.
[2] Sustainable Energy Ireland. Policy drivers. Sustainable Energy Ireland. [29] The Danish Society of Engineers. The Danish society of engineers’ energy plan
Available at: http://www.sei.ie/Renewables/Renewable_Energy_Policy/Policy_ 2030. The Danish Society of Engineers. Available at: http://ida.dk/sites/
Drivers; 2009. climate/introduction/Documents/Energyplan2030.pdf; 2006.
[3] O’Leary F, Howley M, Ó’Gallachóir B. Energy in Ireland: key statistics. [30] DESIRE. Dissemination strategy on electricity balancing for large scale inte-
Sustainable Energy Ireland. Available at: http://www.sei.ie/Publications/ gration of renewable energy. Available at: http://www.project-desire.org.
Statistics_Publications/EPSSU_Publications; 2008. [31] Aalborg University. EnergyPLAN: advanced energy system analysis computer
[4] Department of Communications, Marine and Natural Resources (Ireland). model. Aalborg University. Available at: http://energy.plan.aau.dk; 2009.
Government white paper: delivering a sustainable energy future for Ireland. [32] Connolly D, Lund H, Mathiesen BV. Modelling the Irish energy-system: data
Department of Communications, Marine and Natural Resources (Ireland). required for the EnergyPLAN tool. University of Limerick. Available at: http://
Available at: http://www.dcmnr.gov.ie/Energy/EnergyþPlanningþDivision/ www.dconnolly.net/publications.html; 2009.
EnergyþWhiteþPaper.htm; 2007. [33] Connolly D, Lund H, Mathiesen BV, Leahy M. Developing a model of the Irish
[5] SEMO. The single electricity market operator. SEMO. Available at: http://www. energy-system. In: Proceedings of the joint action on climate change, Aalborg,
sem-o.com; 2009. Denmark; 2009.
[6] International Energy Agency. Electricity information 2009: with 2008 data. [34] International Energy Agency. World energy outlook 2008. International
International Energy Agency. Available at: http://www.iea.org/Textbase/ Energy Agency. Available at: http://www.worldenergyoutlook.org/2008.asp;
Publications/free_new_Desc.asp?PUBS_ID¼2036; 2009. 2008.
[35] IPCC. Climate change 2007: mitigation of climate change. In: Metz B,
Davidson OR, Bosch PR, Dave R, Meyer LA, editors. Contribution of working
group III to the fourth assessment report of the intergovernmental panel
5
Ekraft System and Eltra have been combined to form a single Danish TSO called on climate change. Cambridge, United Kingdom and New York, NY, USA:
Energinet.dk. Cambridge University Press. Available at: http://www.ipcc-wg3.de/activity/
6 publications/ar4/working-group-iii-fourth-assessment-report; 2007.
This does not include the balancing costs associated with wind power.
 D. Connolly et al. / Energy 35 (2010) 2164–2173 2173

[36] Commission for Energy Regulation. Best new entrant price 2006: consultation [41] Danish Energy Agency, Ekraft System, Eltra. Technology data for electricity and
paper. Commission for Energy Regulation. Available at: http://www.cer.ie/en/ heat generating plants. Danish Energy Agency, Ekraft System, Eltra. Available
electricity-wholesale-market-current-consultations.aspx?article¼422eba10- at: http://www.energinet.dk/NR/rdonlyres/4F6480DC-207B-41CF-8E54-BF0B
ba7d-479d-8bb5-72ad3e2af8f2; 2005. A82926D7/0/Teknologikatalog050311.pdf; 2005.
[37] Commission for Energy Regulation. Best new entrant price 2007: consultation [42] Danish Energy Agency. Basisfremskrivning af Danmarks energiforbrug frem til
paper. Commission for Energy Regulation. Available at: http://www.cer.ie/en/ 2025 [Forecast of the Danish energy supply until 2025]. Danish Energy
electricity-wholesale-market-decision-documents.aspx?article¼e9c113b7-e8 Agency. Available at, http://www.ens.dk/graphics/Publikationer/Energipolitik/
77-453e-ad4a-5eec93263140; 2006. Basisfremskrivning_2007_170108/index.htm; 2008 [in Danish].
[38] Danish Energy Agency. Forudsætninger for samfundsøkonomiske analyser på [43] Salmond N. Personal communication at the British Hydropower Association;
energiområdet [Prerequisites for socio-economic analysis of energy]. Danish 2008.
Energy Agency. Available at: http://www.ens.dk/sw80140.asp; 2009 [in Danish]. [44] Gonzalez A, Ó’Gallachóir B, McKeogh E, Lynch K. Study of electricity storage
[39] Hamelinck C, van den Broek R, Rice B, Gilbert A, Ragwitz M, Toro F. Liquid technologies and their potential to address wind energy intermittency in
biofuels strategy study for Ireland. Sustainable Energy Ireland. Available at: Ireland. Sustainable Energy Ireland. Available at: http://www.sei.ie/Grants/
http://www.sei.ie/uploadedfiles/InfoCentre/LiquidbiofuelFull.pdf; 2004. Renewable_Energy_RD_D/Projects_funded_to_date/Wind/Study_of_Elec_Stor
[40] PointCarbon. Historic prices. PointCarbon. Available at: http://www.pointcar age_Technologies_their_Potential_to_Address_Wind_Energy_Intermittency_
bon.com/news/historicprices; 2009. in_Irl; 2004.