3rd International Conference on Ocean Energy, 6 October, Bilbao

Operational expenditure costs for wave energy projects;

O/M, insurance and site rent

G. Dalton1, R. Alcorn2 and T. Lewis3

1
HMRC, UCC,
Pouladuff Rd, Togeher, Cork, Ireland
E-mail: g.dalton@ucc.ie
2
same address
E-mail: r.alcorn@ucc.ie
2
same address
E-mail: t.lewis@ucc.ie

Abstract The object of the paper is two fold:

Define input metrics for OPEX of wave energy
The additional annual costs of operation and projects.
maintenance (O/M) as well as insurance, have not Assess impact of OPEX on net present value (NPV)
been investigated in any great detail with regard and internal rate of return (IRR) using the case study of
wave energy projects. a 50MW wave energy project.
O/M costs include annual maintenance, as well as A literature review was conducted inspecting the
overhaul and replacement costs set at defined OPEX metrics. The average results determined from
intervals throughout a project. Moreover, O/M costs that study were used as the inputs for the case study
for a wave energy project needs to be split into costs simulations. Review consisted of published reports
pertaining to the wave energy converter (WEC), from onshore wind and offshore wind, as well as wave
moorings and cables. energy reports.
For the purpose of this analysis, wave energy data The location for the case study was off the west
from the M4 buoy off the coast of Ireland for 2007 coast of Ireland. The model used for the analysis in this
was used, and was coupled with the power matrix of paper was NAVITAS, which is a Microsoft Excel [4]
the Pelamis device to produce a total energy output tool primarily developed by the author for the
in kWh for the year. The techno-economic analysis Hydraulics and Maritime Research Centre (HMRC)1
was performed by NAVITAS, a wave energy under the Charles Parsons research award. The
economics model developed by HMRC, Ireland. software calculates the wave energy output of any wave
Wave farm size of 50 MW were investigated, using a energy device, providing it has a power matrix..
feed-in tariff of 0.22/kWh. The wave energy converter (WEC) chosen for
Keywords: Wave energy, OPEX, O/M, insurance, overhaul analysis in this report was the Pelamis P1, as it is the
only WEC to date that has a published power
performance matrix. It is also the only device which
has provided some preliminary initial cost estimates,
1 Introduction which were used in the EPRI study [5] by Previsic in
2004 in California, US. A further report by EPRI was
This paper examines the input metrics of operational
conducted 2 years later (2006) by Bedard [6]. The
expenditure (OPEX) assessment for wave energy
reliability of the Pelamis power matrix has never been
projects, and reports on case study modelling
fully verified since it was first published in 2003, and
performed to examine the sensitivity of project profit
unfortunately, there has been no update of the matrix
returns to these inputs.
since. There have also been no revised initial costs
Assessment and calculation of OPEX has been a
estimates for the Pelamis device, nor has the company
very important study area for onshore wind [1]. The
volunteered to provide up to date costs. Therefore, the
determination of its impact and mitigation has been one
Pelamis device, its matrix and costings, are only used
the reasons for the huge rise in onshore wind
in the context of a case study and provide a platform
installations in Europe and globally [2, 3].
methodology to examine the papers research aims.
Research into offshore renewable OPEX has been
Revenue used for the simulations were based on the
negligible to date, with only a few reports quoting
proposed Irish wave energy feed-in tariff of
costs, with little or no analysis (a review of the reports
0.22/kWh [7].
is discussed in the next section).

1
http://www.ucc.ie/research/hmrc/
1
 3rd International Conference on Ocean Energy, 6 October, Bilbao

Total initial cost (IC) of a project is defined in this create a percentage relation to the total COE per MWh
study as the sum of all projects costs which are (see paragraph below). However, $/MWH is not the
necessary to deploy device(s) ready for operation. The simplest metric and from an economic perspective
costs include: costs of components (WEC, cable, and assumes that the total energy output is already
mooring), cost installation, boats and towing, initial calculated. Its disadvantage is that the O/M result is
permits, management fees and decommissioning). location specific, and will change for the same device
Many of the costs of the project are derived as a used in different locations, which might result in
percentage cost of the WEC, which is abbreviated as confusion.
ICwec (explained more thoroughly in Methodology and The next most popular metric is O/M based as a
inputs). O/M costs in this study are calculated for each percentage of the initial cost (IC). The advantage of this
component (WEC, cable and mooring) rather than O/M method is in its simplicity, and that it is uniform in
for the entire project. Calculation for O/M for a operation in any location, and thus easier to use in cash
component is taken as a percentage of the IC cost of flow sheets. The metric has two main disadvantages:
that component, with the inclusion of its installation % of IC quoted in the literature does not normally
cost (explained more thoroughly in Methodology and specify the capital used i.e. whether it is a percentage of
inputs). An example of the abbreviation for initial costs the device IC or the total project IC.
for cable is as follows: The metric does not reflect costs specific to a
location.
(IC+instal)cable= initial cost of cable + cost of cable The % of IC figure is at present arbitrarily chosen
installation. for economic analysis and not based on actual
evidence.
Results extracted from this study must be taken as The final method compares in % the O/M cost in
indicative and relative, keeping in mind that the main $/MWh to the total COE. It is rarely used in reports and
focus of the paper is an analysis of the impacts of assumes that COE is already analysed and discussed.
OPEX on wave energy project returns. Conclusions drawn from the data presented in Table
1 are as follows:
Average wave energy O/M:
2 Literature review on operational o $30/MWh (exception is the low quote
expenditure (OPEX) from EPRI [5] of $8/MWH).
o O/M percentage of IC range from 1.5-5%
(with a max of 10% from Wang [24]).
2.1 Operation and maintenance Average onshore wind O/M:
This section reviews the literature available on o $8-15/MWh (exception is the EWEA [18]
operation and maintenance (O/M). Operation and report quoting 40/MWh).
maintenance is defined in this case study as all annual o O/M percentage of IC range from 1-3.5%
costs required to maintain optimum mechanical (with a max of 7% from Lemming [19]).
performance of a device/ series of devices/ wave farm. Average offshore wind O/M:
In this report, it will include all planned and unplanned o $8-16/MWh (exception is the UK [23]
maintenance. The logistics of these two sub-categories report quoting 3/MWh).
will not be explored in this paper. o O/M percentage of IC range from 3-4.5%
Metrics relating to O/M expense are described by .
either of 4 following metrics: Of the three metrics reviewed, O/M based on the %
$/MWh (/MWh): This metric provides a cost based IC is the most unambiguous. Sensitivity analysis of
on the relationship between the total initial cost of the O/M can be recommended for both wind and wave
project and the annual energy output. ranging from 1-5%.
% of initial cost (IC2): O/M is calculated as a Conclusions from the $/MWh are not as clear.
straight percentage of capital cost. Onshore and offshore wind appear to have the same
% of the total OPEX. $/MWh, although their % of IC figures differ. It would
% of cost of electricity (COE): The metric provide a be perhaps expected that $/MWh for offshore wind
percentage O/M cost in relation to the total COE for the should be higher than onshore wind due to the higher
project. It requires both the O/M and COE based in IC. This anomaly can be explained by the following
$/MWh. example and explanation: a 5MW rated wind turbine
The most commonly quoted metric is $/MWh. Its will be expected to produce a higher energy output in
main advantage is that it can be used as a performance an offshore location than an onshore location, due to an
indicator, as the result is directly proportional to the expected higher capacity or load factor. Although a
device performance at the location. It can be used to 5MW wind turbine may have similar IC costs for both
onshore and offshore applications, the installation and
2
connection costs for offshore will be higher. Thus, the
IC means the total initial costs of the item in question. Therefore it
can refer to either a component such as cable or the entire project. higher energy output of the offshore turbine will offset
Care therefore needs to exercised in defining what the IC refers to the higher IC costs for that location, producing similar
when used. For example, ICwec used later defines the initial cost of /MWH results as its onshore equivalent. The higher
the WEC alone.
2
 3rd International Conference on Ocean Energy, 6 October, Bilbao

/MWh for wave energy imply that wave energy IC are

much higher than offshore wind, and are not
compensated by the slightly higher energy returns or
capacity factor.

Wind/ Location Author Referenc $/MWh % of total % of

wave e ICa COE
Wave USA Bedard, [8, 9] 36
Siddiqui
USA Bedard [10] 29-54
USA Oregon [11] 24 1.4% 14%
Europe Batten [12] 5%
Canada Dunnett [13] 2%
USA EPRI [5] 8.6 4.5%
UK Carbon Trust [14] 16.6 () 1.5%

OnshoreWi USA Bedard [10] 9

nd USA Bolinger [15] 30 (1980)
USA Bolinger [15] 8 (2000s)
Europe EWEA [16] 25%
Germany Albers [17] 8-16 () 1.8-3.6%
Europe EWEA [18] 40 ()
Europe Lemming [19] 1-7%
Denmark EWEA/RISO [1] 5-15-45 ()

Offshore Europe EWEA [20] 16 () 3.3% 26%

Wind UK Van Bussel [21] 4-4.5%
Netherlands Rademakers [22] 8-16 () 25-30%

UK Dale [23] 3b ()
Table 1: Literature review of operation and maintenance cost for onshore and offshore wind, and wave energy studies. Three metrics
a
are presented: $/MWh, % of initial cost IC and % of cost of electricity (COE). All costs are in $US unless stated otherwise in brackets.
% quoted are assumed to be a % of the total IC, although not clearly defined in reports. b Result based on $24/kw/yr quoted in the
paper.

insurance as 13% of total OPEX [26]. The Irish Wind

Energy Assoc (IWEA)3 quote insurance costs
2.2 Insurance costs typically work out around 15,000 per MW for the
The costs of insurance in a much under-researched development of the project and the first year of
area for the whole renewable energy sector. It is operation with a progressive reduction in cost after year
observed that there are 2 main ways of quoting one. Finally, the EPRI report [5] quotes $40/MWh and
insurance costs for cash flow analysis; % of IC, 2% of IC for insuring the Pelamis in the Oregon
/MWh. project.
The Carbon Trust produced two reports quoting
insurance. The first reports quotes a list of insurance 2.3 Site rentals
types and expenses as follows [14]: This Irish governments paper on offshore generating
All risk insurance at 2% of IC. station costs [27] quotes commercial rents based on the
Cost overrun insurance at 3% of the first year nominal output of each turbine 3,800p.a. on a rating of
revenue. 1 MW, or a percentage of gross revenue (2-2.5%) being
An operational insurance of 0.8% of the IC. paid as rental over the site sought. It is anticipated that
Business interruption insurance of 2% of energy
revenue. 3
The second report [25] quotes the insurance http://www.iwea.com/index.cfm/page/planning_regulationsandadmin
component of total OPEX at 14%. EWEA quote is
3
 3rd International Conference on Ocean Energy, 6 October, Bilbao

the wave energy rental costs will be similar to offshore Ocenica (CEO) which was 77% owned by a
wind rental costs, and were thus used in the modeling subsidiary of Babcock and Brown Limited and 23% by
section for wave energy in this report. Pelamis Wave Power Limited. Since the financial crisis
accelerated in the last quarter of 2008, Babcock and
Brown Limited has had its shares suspended and has
3 Methodology and inputs been in a managed process of selling its assets5. The
E.ON energy group are the second major developer
3.1 Project location and device using the Pelamis. E.ON plans to deploy the latest
Pelamis design, the P2, in the European Marine
The location for the data collection was the M4 buoy Energy Centre (EMEC) in 2010. Future plans for the
off Belmullet, 55 deg N, 10 deg W or approximately 25 EMEC and the Orcadian Wave Farm will consist of
km off the coast of Mayo, water depth approx 150m four Pelamis generators supplied by PWP to
(Figure 1). ScottishPower Renewables supported by 4m funding
from the Scottish Executive providing.

3.2 Power matrix

The total annual energy output (AEO) for the year
was calculated in NAVITAS by multiplying each cell
point of the scatter plot of hours with the corresponding
cell of a WEC power matrix. For this report, the power
matrix of the Pelamis was used [28]. The Pelamis
power matrix is presented in Table 2. Power peaks at
New 750 kW for a number of sea states.
location for Wave energy input (WEI) is calculated using the
M4 buoy following Equation 1:

Equation 1 WEI=0.55Hs2Tz

Where Hs is mean significant wave height, and Tz is

the mean zero crossing period The Pelamis scatter plot
uses Te, which is the energy in the period, and is
calculated from Tz in Equation 2.

Figure 1: Map of location of the M series buoys around Ireland4. Equation 2 Te = 1.2*Tz

Wave energy output (WEO) for the Pelamis device

The Pelamis wave energy converter is developed is a product of the energy input and the apparent
and manufactured by Pelamis Wave Power (PWP) width of a WEC is 3.5m, and is calculated using
(formerly known as Ocean Power Delivery Ltd), an Equation 3.
Edinburgh-based company originating from the Wave
Power Group at the University of Edinburgh in 1998 Equation 3: WEO = WEI*3.5
[28, 29]. Major shareholders include; Emerald
Technology Ventures, Norsk Hydro Technology 3.3 Financial inputs
Ventures, BlackRock Investment Managers, 3i, Carbon
Trust, Nettuno Power, Tudor Global, Scottish Initial costs (IC) for the WEC (ICwec) only reflects
Enterprise and Eon. the purchase of the Pelamis device from the
The Pelamis is a semi-submerged snake-like device manufacturer. The IC of the Pelamis WEC chosen for
consisting of articulated cylindrical sections linked by this report was 1,600,0006 obtained from the 2004
hinged joints. The version used in this assessment is EPRI report in California [5]. The figure included costs
the first prototype, the P1, which was 120m long, 3.5m for both the steel sections and all the internal
in diameter, rated at 750kW and weighed 700ton. PWP components.
had their first major demonstration project in Table 3 displays the initial cost/kW for the Pelamis
Aguadoura, Portugal. The multiple Pelamis units device, displaying how the cost/kW varies with both
making up the Aguadoura wave farm constitute both time and size of farm deployed. As costs for 2004 for
the worlds first, multi-unit, wave farm and also the the Pelamis are only available, costs for the proceeding
first commercial order for wave energy converters. The years were based on multiples of the price of steel.
Aguadoura project is owned and operated by a joint
venture company called Companhia da Energia 5
http://www.guardian.co.uk/environment/2009/mar/19/pelamis-
wave-power-recession
4 6
http://www.met.ie/marine/MarineWx2005.pdf, WEC $1,565,000 + steel sections $850,000. US currency
http://www.marine.ie/home/publicationsdata/data/buoys/ conversion to Euro was 1.50 (at Nov 2009).
4
 3rd International Conference on Ocean Energy, 6 October, Bilbao

Period (Te)

1 2 3 4 5 6 7 8 9 10 11 12 13
Height (Hs) 0.5 0 0 0 0 0 0 0 0 0 0 0 0 0
1 0 0 0 0 0 29 37 38 35 29 23 0 0
1.5 0 0 32 65 83 86 78 65 53 42 33
2 0 57 115 148 152 138 116 93 74 59
2.5 89 180 231 238 216 181 146 116 92
3 129 260 332 332 292 240 210 167 132
3.5 354 438 424 377 326 260 215 180
4 462 540 530 475 384 339 267 213
4.5 544 642 628 562 473 382 338 266
5 726 707 670 557 472 369 328
5.5 750 750 737 658 530 446 355
6 750 750 750 711 619 512 415
6.5 750 750 750 750 658 579 481
7 750 750 750 750 613 525
7.5 750 750 750 750 686 593
8 750 750 750 750 625
8.5 750 750 750 750
9 750 750 750
9.5 750 750
10 750
10.5
11
11.5
12

Table 2: Power matrix for the Pelamis WECS [28].

2004 2006 2008 2010

1 MW farm WEC only IC $2100/kW $4300/kW $6500/kW $4300/kW
Total project IC $4000/kW $8000/kW $11,200/kW $8000/kW

50 MW farm 50 WEC only IC $1400/kW $2800/kW $4100kW $2800/kW

Total Project IC $3000/kW $5000/kW $7600/kW $5000/kW
Table 3: The initial cost (IC) of the device only and the total
project cost, for a 1 MW project and 50 MW project. Cost
estimates for 2004, 2006, 2008 and 2010 are displayed. 2010
capital costs are used in this report.

Dalton et al. [30] showed that the price of steel

tripled from 2004 to 2008 and then dropped to double
figures after 2009. Thus a factor of two was used for
2010 initial costs.
The majority of simulations in this case study model
a 50MW installed capacity wave farm. A 1MW project
costs more per kW at approximately 8000/kW than a

5
 3rd International Conference on Ocean Energy, 6 October, Bilbao

50MW project at 5000/kW, due to bulk discount rates low risk projects. Low risk projects have their discount
discussed later. The capital costs used in this report are rate calculated in the following Equation 5:
approximately double those quoted for other wave BR + f
energy reports and offshore wind (see paper by Dalton Equation 5 DR =
1 f
et al. [30] for more detail).
The remainder of other costs were calculated as a
Where BR is the borrowing rate given for a loan and
percentage of the IC of the WEC (ICwec) (Table 4),
f is the inflation rate. By defining the discount rate in
and were based on costs derived from a previous paper
this way, inflation is factored out of the economic
by the author, Dalton et al. [30]. This method allowed
analysis during the project lifetime. All costs therefore
for simplified cost and sensitivity analysis.
become real costs, meaning that they are in defined in
terms of constant Euros. The assumption is that the rate
WEC parameter % of ICwec of inflation is the same for all costs. A general inflation
rate of 3% [31] was used for the project.
Mooring 10%
Cabling 10% Farm size Discount rate
Replacement costs 90% 0-5MW 14%
Spare parts 2% 6-10MW 12%
Sitting and permits 2% 11-20MW 10%
GHG investigations 0.05% 21-50MW 8%
Decommissioning fees 10% 50MW 6%
Grid connection 5% >50MW Equation 5
Table 6: Discount rates.
Management fees 10% of total IC
Table 4: Costs of WEC infrastructure, calculated as a percentage
of the ICwec. The majority of Previsic [5] costings were used in
the report. 3.4 Grid Sales Revenue
Feed-in tariff (FIT) refers to the regulatory,
A summary of the project inputs can be viewed in minimum guaranteed price per kWh that an electricity
Table 5. utility has to pay to a private, independent producer of
renewable power fed into the grid [32]. It is defined in
Borrowing or interest 7.5% this report as the full price per kWh received by an
rate independent producer of renewable energy including
Inflation rate 3% the premium above or additional to the market price,
Project years 15a but excluding tax rebates or other production subsidies
paid by the government. Grid sales are a credit, and are
Table 5: Input values for the project. (a The DCENR tariff of
0.22/kWh is available for 15 year duration)
added to other negative cost values for the each year.
The sales are the product of the following two
variables:
The discount factor (DF) translates expected The total energy produced each year, referred to as
financial benefits or costs in any given future year into the annual energy output (AEO).
present value terms. The total nominal profit is The electricity tariff rate from the utility company.
adjusted for cash depreciation by multiplying the total In this report, FIT of 0.22/kWh was assessed.
nominal profit by a discount factor. The rate of 0.22/kWh is the rate presently proposed
DF is calculated using the discount rate, and is by the Irish government [7]. The tariff spans a 15 year
calculated by Equation 4. project and is available until 20157.
1
Equation 4 DF =
(1 + DR) n 3.5 Cost reduction for multiple devices
The cost of purchasing multiple devices are cheaper
The discount rate (DR) is an interest rate than buying a singular device, due to discounts
commensurate with perceived risk used to convert provided any manufacturer to encourage multiple
future payments or receipts (within a project lifetime) purchases. The discount is based on a cumulative
to present value. The discount rates for various project factorial reduction in price. The cumulative total of n
sizes are listed in Table 6. Small WEC projects are number of devices is the sum of the discounted costs,
deemed high risk, as they are at the beginning of their derived in Equation 6.
commercialisation phase, and thus have high discount
rates. For this project, wave farms greater than 50MW
are deemed fully commercial and thus categorised as
7
Personal communication from Richard Browne (DCENR).
6
 3rd International Conference on Ocean Energy, 6 October, Bilbao

n components: WEC, Cable and Mooring. As defined

Equation 6 Total ICwec = 1 Pn * ICwec n earlier, IC for a component will include installation
costs, and this be abbreviated as follows e.g. for WEC
will be (IC+instal)wec .
P is the percent reduction used in IC costing for For this assessment, 4 rates for O/M (1%, 3%, 5%,
WEC, derived in Equation 7: 10%) are examined to assess their impact on net present
value of a project. The majority of the modelling in this
report apply a uniform % for all three component
categories. However, there is one simulation which
(ln(tf ) / ln( 2 ) inspects varying percentages. This study does not
Equation 7 P=N analyse planned or unplanned O/M. The rates are
multiplied with the appropriate component IC
Where N is the number of WEC components and tf according to Equation 9:
is the technology factor. The technology factor of most
industrial products is mostly between 0.85 and 0.95. Equation 9
For this project, 0.9 was chosen. Total O / M = ((IC + instal )WEC * O / M WEC %) + (( IC + instal ) MOORING
* O / M MOORING %) + (( IC + instal ) CABLE * O / M CABLE %)

3.6 Cable costs

Overhaul (sometimes called refit)
In this paper, a simplistic cable costs estimation of
In these case study simulations, the WEC devices are
10% of the ICwec was used.
taken out of the water and receive a general overhaul
Preliminary simulations using NAVITAS of detailed
(or refit) every 4 years. The cost of the overhaul per
cable costs have indicated that cable costs converge to
device was taken as 10% of IC. Annual overhaul costs
10% for large projects. However, small projects of the
are based on the average overhaul cost averaged over
5MW may have cable costs as high as 33% of IC.
the entire project life, in Equation 10:
Detailed cable costing for wave energy farms is an
important area of research and the subject is deemed
Equation 10
the topic matter for another research paper.

sum(total overhaul cos ts )

3.7 Learning Curve Annual Overhaul cos t =
The cost of producing a device decreases in project years
proportion to the level of global and national expertise Replacement
developed over time. This case study is taken at one In this series of simulations, the WEC devices are
point in time, thus using learning curve rates is not completely replaced every 10 years. The cost of
appropriate here. The size of farm modelled in the replacement per device was taken as 90% of IC.
paper is 50MW and is assumed to be an immature Annual replacement costs are based on the sum of the
technology. A high discount rate is used which will replacement cost averaged over the entire project life,
account for this level of technology status. in Equation 11:

3.8 Cash Flow Analysis Equation 11

Annual cash flow (ACF) is the sum of the revenue
sum(total replacement cos ts )
in and OPEX. This is summarised in Equation 8: Annual Re placement cos t =
project years
Equation 8 A schedule of overhaul and replacement expenses
ACF = IC OPEX + revenue (FIT ) for WEC, cable and mooring are presented in Table 7.
Insurance
Where IC is the total project initial cost, and OPEX For this assessment, 4 rates for insurance (1%, 3%,
includes O/M, insurance, utility charges and rent. 5%, 10%) are examined to assess their impact on net
Total project initial costs present value of a project. The rates are multiplied with
Figure 2 provides a break down of the initial costs in the total project initial cost according to Equation 12:
percentage of the IC. The ICwec totals 55% of the IC Equation 12 Insurance = IC * insurance %
in this study.
Site rental
3.8.1 OPEX For this paper, the Irish governments [27] rental
The OPEX inputs for the case study were based on costs for offshore generating station was used of 2.5%
average inputs derived from the literature review of gross revenue, according to equation Equation 13:
section. Equation 13 Rent = Gross revenue * Rent %
Annual O/M and Spare Parts
Annual O/M in this case study is derived from the
sum of O/M of the following 3 major project Utility O/M and capacity charges

7
 3rd International Conference on Ocean Energy, 6 October, Bilbao

Costs for Utility O/M and capacity charges were Capacity charges taken from [33, 34] for the west of
taken from Eirgrid and ESBI Network reports [33, 34]. Ireland was 8000/MW.
Line works and cable costs related to the kV and
MW of the project are multiplied by the overhead cable
lengths which in this study was 30km (Table 8).

Onshore transmission and Management fees 9%

interconnection 3%

Subtotal of off-shore
cabling 5%

Mooring installation cost

2%

Mooring 5%

Siting and permits 1% WEC initial cost 55%

GHG plans and

investigations 1%

Spare parts 1%

WEC installation cost 18%

Figure 2: Percentage breakdown of IC.

Replace Overhaul Replace Overhaul Replace Overhaul

Year Discount Initial cost Cost Cost Cost Cost Cost Cost
Factor WEC WEC Mooring Mooring Cable Cable
0 1.0000 -251,414,738
1 0.9434
2 0.8900
3 0.8396
4 0.7921 -13,799,613 -1,379,961 -1,379,961
5 0.7473
6 0.7050
7 0.6651
8 0.6274 -13,799,613 -1,379,961 -1,379,961
9 0.5919
10 0.5584 -124,196,518 -12,419,652 -12,419,652
11 0.5268
12 0.4970 -13,799,613 -1,379,961 -1,379,961
13 0.4688
14 0.4423
15 0.4173

Table 7: Schedule of overhaul every 4 years and replacement every 10 years, for WEC, cable and moorings, for 50 MW project at
0.22/kWh. (for simplicity, overhaul still takes place in year 12 despite the fact that replacement occurred in year 10).

4 Scatter plot of hours for the site

Table 9 presents the scatter plot of hours for the
M4 buoy located off the West coast of Ireland, for

8
 3rd International Conference on Ocean Energy, 6 October, Bilbao

2007. The values in this table vary form from those lies within the periods of 5-7 seconds and wave
reported in an earlier paper by the author [30], as a heights of between 1- 3.5m. The average power of
different method was used to bin the hours. In the the waves for the M4 location in 2007 was 58kW/m.
first paper, the median value for Hs and Tz was used. Figure 3 presents the monthly average energy output
of the wave, showing a large variation between
The revised method used in this paper floors the winter and summer outputs.
Hs and Tz values. The highest frequency of hours

Tz (s)
1 2 3 4 5 6 7 8 9 10 11 12 13
0.5 52 130 25
1 8 41 164 387 151 59 11
1.5 5 44 82 488 360 236 75 12
2 26 65 293 503 258 142 31 4
2.5 76 107 354 386 129 26 5 3 1
3 30 45 226 496 201 41 1 2
3.5 1 26 44 352 227 82 9
Hs (m) 4 23 19 181 252 99 19 3 1
4.5 9 9 71 210 99 19 4
5 2 6 6 158 123 30 3 1
5.5 3 81 144 29 10 3
6 4 21 100 44 7 4
6.5 4 6 47 42 11 4
7 4 27 57 16 3
7.5 3 14 33 19
8 3 29 21 5
8.5 1 21 17 2
9 4 15 2 2
9.5 1 9 1
10 2 3 1
10.5 2 1
11 2
11.5 1

Table 9: Scatter plot of hours for M4 buoy off the west coast of Ireland, 2007.

combined expenses if averaged out over the 15 year

140 project account for 46% of total OPEX. O/M and
insurance each drop to mid 20% of OPEX. O/M costs
120
Energy in the wave (kW/m)

are within the average% OPEX quoted in the literature

100 review if overhaul and maintenance are included.
80
Table 10 presents the breakdown of the expenditure
results from a /MWh perspective. The O/M costs of
60
40/MWh were approximately equal to the average
40 quoted in the literature review for wave energy in Table
1, implying that the 3% used in the simulations was as
20
high as what would be recommended for a project.
0 Insurance costs simulated at 3% of IC resulted in
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
47/MWh and 26% of OPEX and (Table 10). The
Figure 3: Average monthly energy in the wave (kW/m) or wave literature review has only one quoted result using 2%
energy input (WEI) for the M4 buoy, 2007. of IC, which if simulated in the case study results in
19% of total OPEX (in brackets) and 31/MWh. This
5 Results result is slightly higher than that quoted by EWEA and
IWEA of 13% and 14% of OPEX respectively [25, 26]
(see Table 1 literature review). When viewed from the
5.1 /MWh and % of OPEX
metric /MWh, the 31/MWh modelled result
Figure 4 displays the breakup of the total OPEX modelled for 2% insurance is close to that quoted by
costs based over the 15 project. It can be observed that the EPRI report [5] of $40/MWh.
where there is no overhaul and replacement, O/M and
insurance account for the majority of OPEX, at
approximately equal percentages (42%-48%
% of total
respectively) (Figure 4A). If overhaul and replacement OPEX /MWh
of the device is considered (Figure 4B), their total Annual O/M 22% 40
9
 3rd International Conference on Ocean Energy, 6 October, Bilbao

Overhaul 11% 20
Replacement 34% 61
a
Insurance 26% (19%) 47 (31)
Utility charges 2% 4
Rent 3% 6
Total 100% 177
Table 10: Results for the break down on expenditure costs for a
50MW wave farm, using O/M at 3% of IC and insurance at 3%
of IC. aresults for insurance modelled at 2% of IC are in
brackets.

Rent
6%
Utility charges
4%

Annual O/M
42%

Insurance
48% Overhaul
0%
Replacement
A 0%

Rent
Utility charges 3%
Annual O/M
2%
22%
Insurance
27%

Overhaul
11%

Replacement
35%
B

were simulated at 3% of IC of components nad rent at 2.5% of

Figure 4: Break up of OPEX expenditures, A no overhaul and IC,.
replacement, B - overhaul (10% of IC) every 4 years,
replacement (90% of IC) every 10years. O/M and insurance

10
 3rd International Conference on Ocean Energy, 6 October, Bilbao

5.2 NPV and IRR 250 15.0%

200

Results displayed in Figure 5 indicate that for a 50 150

10.0%

MW project at a FIT of 0.22/kWh, O/M above 3% 100

does not result in a positive cash flow or positive NPV. 5.0%

NPC (M)

IRR (%)
50
O/M at 1% and 3% result in lengthy payback times of
12-14 years. (No overhaul or replacement costs were 0

1%
2%
3%
5%
10%
1%
2%
3%
5%
10%
1%
2%
3%
5%
10%
1%
2%
3%
5%
10%
0.0%

considered in this simulation). -50

5MW 20MW 50MW 100MW
Figure 6 displays NPV and IRR results for various -100
-5.0%
combinations of project sizes and O/M costs, based on -150

a FIT of 0.22/kWh and no overhaul and replacement -200 -10.0%

costs. For projects 20 MW and under, the NPV never NPC IRR

achieves a positive value. A positive NPV results for

50 MW on O/M ranging from 1-3% only. Despite the Figure 6: NPC and IRR for varying O/M expenses and different
size wave farms. 0.22/kWh and no overhaul and replacement
high apparent NPV for 100 MW scenarios, threshold
IRR of 10% is only reached for two of the simulations; 0
100MW project at 1% and 2% O/M. It can be 0 2 4 6 8 10 12 14

concluded that at 0.22/kWh, return on investments

-50
(IRR) are extremely vulnerable to O/M costs over 1%

Cumulative cash flows (M)

of IC.
Figure 7 displays the resultant cash flows for -100

modelling replacement of WEC, cable and mooring at O/M 1%

O/M 3%
10 years, and overhaul planned every 4 years. Results -150
O/M 5%
indicate that the 3 schedules of overhaul have little O/M 10%

impact on the cumulative cash flows over the 15 year -200

period. However, the replacement of the components at
year 10 has a significant impact, resulting in all O/M
-250
scenarios never reaching a year of positive cash flow.
The result is interesting, in that it might be assumed
that the cost impact of replacement in year 10 would be -300

minimal due to the fact that the discount rate impact is

Figure 7: Cumulative cash flows and payback times for a range
relatively high by year 10. of O/M simulating a schedule of replacement and overhaul
expenses (FIT at 0.22/kWh).
100

50 Figure 8 presents the analyses of the cash flow

implications of splitting O/M expenses into their
Cumulative cash flows (M)

0
0 2 4 6 8 10 12 14
subcomponents or WEC and cable (for the sake of
-50
simplicity in the graph, mooring was left out). It is
O/M 1% evident from the results that O/M for the WEC
-100
O/M 3%
O/M 5%
component has the most impact, where an increase of
O/M 10% O/M from 1% to 5% for WEC alone pushes the cash
-150
flow projection completely in the negative. A similar
increase in cable costs has a much lesser effect.
-200
The impact of annual insurance costs is assessed in
-250 Figure 9, at FIT of 0.22/kWh. It is evident that even a
modest 1% insurance cost pushes the cash flow for the
-300 15 years into the negative.
The ramifications of a 3% insurance costs at an FIT
Figure 5: Cumulative cash flows for 50MW project at 0.22/kWh
of 0.22/kWh on NPV and IRR are examined in Figure
no overhaul and replacement expenses.
10 and is compared to the results described for 0%
insurance displayed in Figure 6. The insurance
premium eliminates almost all possibility of positive
NPV except for low O/M costs for the largest 100MW
project. IRR fails to reach the 10% threshold in any
scenario, unlike IRR results in Figure 6.
As discussed in the introduction, insurance costs of
5% to 10% are common in the maritime industry and
results from these simulations indicate that severe
consequences on cash flow projections will eventuate,
with current FIT at 0.22/kWh.

11
 3rd International Conference on Ocean Energy, 6 October, Bilbao

50

100
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Cumulative cash flows (M)

50
Cumulative cash flows(M)

-50

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 -100
-50 No Utility O/M Charges
1% WEC, 1% Cable Utility O/M charges
5% cable, 1% WEC -150
-100
5% WEC, 1% cable
5% WEC, 5%cable -200
-150

-200 -250

-250 -300

-300
Figure 11: Cumulative cash flows inspecting impact of ESB O/M
charges (O/M kept ay 3% of IC, 0.22/kWh FIT).
Figure 8: Cumulative cash flows at varying levels of O/M for
either WEC and or cable.

0
0 2 4 6 8 10 12 14
6 Conclusions
-50
Navitas simulations results for /MWh were
Cumulative cash flows (M)

approximately equal to the average reported in the

-100
literature review (40/MWh O/M and 47/MWh for
Insurance 1%

-150
Insurance 3% insurance). The total cost for all OPEX was 87/MWh
Insurance 5%
Insurance 10%
(177/MWh including overhaul and replacement).
-200
These results reflect the high initial costs used in the
case study, and remained high despite the high energy
-250
output from the devices at the chosen case study
location. The results were twice as high as offshore
-300
wind. The author has reservations about the widespread
use of the /MWh metric, as the metric is location
Figure 9: Cumulative cash flows for various levels of insurance specific. It is suggested the metric is best used when
(O/M kept ay 3% of IC, 0.22/kWh FIT). assessing devices of similar MW rating in the same
location. The metric has similar limitations as the cost
of electricity (COE) metric, which was discussed in a
The analysis of utility O/M expenses is often
published paper by Dalton et al [30]. O/M of 3% IC
forgotten about in cash flow analysis. The, the impact
produces /MWh results around the average quoted I
on cash flows in this case study is minimal, as is
the literature.
evident in Figure 11.
The results from the simulations indicate that NPV
100 10.0%
of a project is very sensitive to changes in O/M cost,
8.0%
50
with a limit of 3% before NPV goes negative. Larger
6.0%
0 wave farms of 100MW or more will make projects
1%
2%
3%
5%
10%
1%
2%
3%
5%
10%
1%
2%
3%
5%
10%
1%
2%
3%
5%
10%

4.0%

-50
more profitable from an IRR perspective due to cost
2.0%
NPC (M)

5MW 20MW 50MW 100MW

savings, as well as being more tolerant to higher O/M
IRR (%)

0.0%
-100
costs.
-150
-2.0%
The O/M of the WEC is the most significant
-200
-4.0%
contributor to overall O/M (in comparison to cable and
-6.0%
mooring) as expected due to the higher IC of that
-250
-8.0% component.
-300 -10.0% Variation in insurance costs is the other major factor
NPC IRR
which has significant impact on NPV. Case study
Figure 10: NPC and IRR for 3% insurance, at varying O/M simulations indicate that the current FIT of 0.22/kWh
expenses and different size wave farms, 0.22/kWh and no may not be sufficient if insurance rates of 1% and over
overhaul and replacement. are levied.
Replacement cost is the other OPEX cost which is
not often considered in assessments. In the present case
study modelling, over 1/3 of the total OPEX cost for
the entire project was due to device replacement in the
10th year.
In conclusion, project profits are very sensitive to
annual OPEX, especially if overhaul and replacement
costs are accounted for. Results indicate that device
designers will need to choose whether to opt for longer

12
 3rd International Conference on Ocean Energy, 6 October, Bilbao

lasting more expensive devices which require lower content/uploads/2010/01/EPRI-Wave-Energy-

annual maintenance costs, or cheaper devices with Economics.ppt#839,5,Learning
shorter device lifetimes requiring overhaul mechanisms [11] Oregan Wave. Value of Wave Power. Oregan
that enable easy and cheap retrieval from ocean site to Wave Energy Utility Trust, 2010;
maintenance dock. Care needs to be exercised in the http://www.oregonwave.org/our-work-
careful choice of OPEX categories and their definition. overview/market-development/utility-market-initiative/
More detailed research is required to determine exact [12] Batten WMJ, Bahaj AB. An assessment of
insurance and O/M costs as well specifics of device growth scenarios and implications for ocean energy
service times and device lifespan as well as weather industries in Europe. Sustainable Energy Research
windows, before more reliable costs estimates for wave Group, School of Civil Engineering and the
energy projects can be confidently assessed. Environment, University of Southampton, Report for
CA-OE, Project no. 502701, WP5, 2006;
http://eprints.soton.ac.uk/53003/
7 References [13] Dunnett D, Wallace JS. Electricity generation
from wave power in Canada. Renewable Energy 2009;
[1] Morthorst PE. Costs and prices. EWEA and 34:1: p.179-195.
RISO, 2003; [14] Carbon Trust. Oscillating water column wave
http://www.ewea.org/fileadmin/ewea_documents/docu energy converter evaluation report. Marine Energy
ments/publications/WETF/Facts_Volume_2.pdf Challenge, ARUP, EON, 2005;
[2] Dalton G, Gallachir BP. Building a wave http://www.carbontrust.co.uk/SiteCollectionDocuments
energy policy focusing on innovation, manufacturing /Various/Emerging%20technologies/Technology%20D
and deployment. Renewable and Sustainable Energy irectory/Marine/Other%20topics/OWC%20report.pdf
Reviews 2010; In Press, Uncorrected Proof. [15] Bolinger M, Wiser R. Annual report on U.S.
[3] EWEA, Blanco G, Kjaer I. Wind at work: Wind wind power installation, cost, and performance trends.
energy and job creation in the EU. European Wind NREL, 2007; http://eetd.lbl.gov/ea/ems/reports/ann-rpt-
Energy Association, 2008; wind-06-ppt.pdf
http://www.ewea.org/fileadmin/ewea_documents/docu [16] EWEA. Wind power economics. EWEA, 2008;
ments/publications/Wind_at_work_FINAL.pdf http://www.ewea.org/fileadmin/ewea_documents/docu
[4] Microsoft. Excel. Washington, US: Software ments/publications/factsheets/factsheet_economy2.pdf
produced by Microsoft; 2006. [17] Albers A. O&M cost modelling, technical
[5] Previsic M. System level design, performance, losses and associated uncertainties. German Windguard
and costs of California Pelamis wave power plant. Consulting, 2008;
EPRI, 2004; http://www.ewec2009proceedings.info/allfiles2/554_E
http://oceanenergy.epri.com/attachments/wave/reports/ WEC2009presentation.pdf
006_San_Francisco_Pelamis_Conceptual_Design_12- [18] EWEA. Operation and Maintenance Costs of
11-04.pdf Wind Generated Power. 2010; http://www.wind-
[6] Bedard R. EPRI ocean energy program, energy-the-facts.org/en/part-3-economics-of-wind-
possibilities in California. EPRI, 2006; power/chapter-1-cost-of-on-land-wind-
http://oceanenergy.epri.com/attachments/ocean/briefin power/operation-and-maintenance-costs-of-wind-
g/June_22_OceanEnergy.pdf generated-power.html
[7] DCENR. Minister Ryan launches major new [19] Lemming J, Morthorst PE. O&M costs and
Ocean Energy initiative. DCENR, 2008; economical costs of wind turbines. European Wind
http://www.dcenr.gov.ie/Press+Releases/2008/Minister Energy Conference. 1999.
+Ryan+launches+major+new+Ocean+Energy+initiativ http://books.google.com/books?hl=en&lr=&id=AhuLV
e.htm rUMP7UC&oi=fnd&pg=PA294&dq=/mwh+operation
[8] Bedard R, Previsic M, Hagerman G. North +and+maintenance+wave+wind&ots=QExHLII_4U&si
American Ocean Energy Status. EPRI, 2007; g=d0OwGU1S2MlgmxV1CCMTAerqvTQ#v=snippet
http://www.oceanrenewable.com/wp- &q=maintenance%20&f=false
content/uploads/2008/03/7th-ewtec-paper-final- [20] EWEA. The cost of energy generated by
032907.pdf offshore wind power - Wind Energy the facts. RISO,
[9] Siddiqui O, Bedard R. Feasibility assessment of 2010; http://www.wind-energy-the-facts.org/en/part-3-
offshore wave and tidal current power production: a economics-of-wind-power/chapter-2-offshore-
collaborative public/private partnership. Piscataway, developments/the-cost-of-energy-generated-by-
NJ, USA. IEEE. 2005 IEEE Power Engineering offshore-wind-power.html
Society General Meeting (IEEE Cat. No. 05CH37686). [21] van Bussel GJW, Zaaijer MB. Reliability,
2005. Vol. 2: 2004-10. Availability and Maintenance aspects of large-scale
http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=0 offshore wind farms, a concepts study. Marine
1489368 Renewable Energies Conference, Newcastle U.K. 2001.
[10] Bedard R. Ocean wave power/ energy http://ocw.tudelft.nl/fileadmin/ocw/courses/OffshoreWi
economics. EPRI, 2009; ndFarmEnergy/res00055/MAREC_2001_OM_Paper.p
http://hinmrec.hnei.hawaii.edu/wp- df

13
 3rd International Conference on Ocean Energy, 6 October, Bilbao

[22] Rademakers L, Braam H, Zaaijer M, Van

Bussel G. Assessment and optimisation of operation
and maintenance of offshore wind turbines. EWEC.
2003.
http://www.ecn.nl/docs/library/report/2003/rx03044.pd
f
[23] Dale L, Milborrow D, Slark R, Strbac G. Total
cost estimates for large-scale wind scenarios in UK.
Energy Policy 2004; 32:17: p.1949-1956.
[24] Wang U. Making a splash: PG&E dives
headlong into wave power project Renewable Energy
World, 2010;
http://www.renewableenergyworld.com/rea/news/articl
e/2010/05/making-a-splash-pg-e-dives-headlong-into-
wave-power-project
[25] Carbon Trust. Cost estimation methodology.
ENTEC, 2006;
http://www.carbontrust.co.uk/SiteCollectionDocument
s/Various/Emerging%20technologies/Technology%20
Directory/Marine/MEC%20cost%20estimation%20met
hodology%20-%20report.pdf
[26] Intelligent Energy Europe. Wind Energy - the
facts DEWI, 2010; http://www.wind-energy-the-
facts.org/en/part-3-economics-of-wind-power/chapter-
1-cost-of-on-land-wind-power/operation-and-
maintenance-costs-of-wind-generated-power.html
[27] Irish Government. Offshore electricity
generating stations - note for intending developers.
Department of Marine and Natural Resources, 2000;
www.sei.ie/uploadedfiles/RenewableEnergy/policydoc
ument.pdf
[28] Pelamis. Ocean Energy, 2008;
http://www.pelamiswave.com/
[29] Pelamis, Carcas M. The Pelamis Wave Energy
Converter. Ocean Power Delivery Ltd, 2007;
http://hydropower.inl.gov/hydrokinetic_wave/pdfs/day
1/09_heavesurge_wave_devices.pdf
[30] Dalton GJ, Alcorn R, Lewis T. Case study
feasibility analysis of the Pelamis wave energy
convertor in Ireland, Portugal and North America.
Renewable Energy 2010; 35:2: p.443-455.
[31] Lambert T, Lilienthal P. HOMER: The micro-
power optimisation model. Software produced by
NREL; 2004. www.nrel/gov/HOMER and
http://www.mistaya.ca/homer/MicropowerSystemMod
elingWithHOMER.pdf.
[32] Sijm JPM. The performance of feed-in tariffs to
promote renewable electricity in European countries.
ECN, 2002; http://www.wind-
works.org/FeedLaws/Netherlands/ECNFeedLawsc020
83.pdf
[33] ESB Networks. Schedule of operation and
maintenance charges. 2009;
http://www.esb.ie/esbnetworks/downloads/esbnetworks
_operation_maintenance_charges_2009.pdf
[34] Eirgrid. Statement of charges. 2009;
http://www.eirgrid.com/media/2009%20Statement%20
of%20Charges%20v1.0%20%E2%80%93%20CER%2
0approved.pdf

14