Levelized Cost of Energy

Analysis of Marine and

Hydrokinetic Reference Models
Preprint
D. S. Jenne and Y.-H. Yu
National Renewable Energy Laboratory

V. Neary
Sandia National Laboratories
To be presented at the 3rd Marine Energy Technology Symposium
(METS 2015)
Washington, D.C.
April 27‒29, 2015

NREL is a national laboratory of the U.S. Department of Energy

Office of Energy Efficiency & Renewable Energy
Operated by the Alliance for Sustainable Energy, LLC
This report is available at no cost from the National Renewable Energy
Laboratory (NREL) at www.nrel.gov/publications.

Conference Paper
NREL/CP-5000-64013
April 2015

Contract No. DE-AC36-08GO28308

 NOTICE

The submitted manuscript has been offered by an employee of the Alliance for Sustainable Energy, LLC
(Alliance), a contractor of the US Government under Contract No. DE-AC36-08GO28308. Accordingly, the US
Government and Alliance retain a nonexclusive royalty-free license to publish or reproduce the published form of
this contribution, or allow others to do so, for US Government purposes.
This report was prepared as an account of work sponsored by an agency of the United States government.
Neither the United States government nor any agency thereof, nor any of their employees, makes any warranty,
express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of
any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately
owned rights. Reference herein to any specific commercial product, process, or service by trade name,
trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation,
or favoring by the United States government or any agency thereof. The views and opinions of authors
expressed herein do not necessarily state or reflect those of the United States government or any agency thereof.

This report is available at no cost from the National Renewable Energy

Laboratory (NREL) at www.nrel.gov/publications.

Available electronically at SciTech Connect http:/www.osti.gov/scitech

Available for a processing fee to U.S. Department of Energy

and its contractors, in paper, from:

U.S. Department of Energy

Office of Scientific and Technical Information
P.O. Box 62
Oak Ridge, TN 37831-0062
OSTI http://www.osti.gov
Phone: 865.576.8401
Fax: 865.576.5728
Email: [email protected]

Available for sale to the public, in paper, from:

U.S. Department of Commerce

National Technical Information Service
5301 Shawnee Road
Alexandra, VA 22312
NTIS http://www.ntis.gov
Phone: 800.553.6847 or 703.605.6000
Fax: 703.605.6900
Email: [email protected]

Cover Photos by Dennis Schroeder: (left to right) NREL 26173, NREL 18302, NREL 19758, NREL 29642, NREL 19795.

NREL prints on paper that contains recycled content.

 LEVELIZED COST OF ENERGY ANALYSIS OF MARINE AND
HYDROKINETIC REFERENCE MODELS
D. Scott Jenne Yi-Hsiang Yu1 Vincent Neary
National Renewable Energy National Renewable Energy Sandia National Laboratories
Laboratory Laboratory Albuquerque, New Mexico, United
Golden, Colorado, United States Golden, Colorado, United States States

1Corresponding author: [email protected]

ABSTRACT Although their resource potential is

In 2010, the U.S. Department of Energy initiated significant, MEC technologies are at early stages of
the development of six marine and hydrokinetic development and will require further research to
(MHK) energy converter reference models that be economically competitive with other electricity
are device point designs of well-known marine generating technologies. To help mitigate this
energy converters (MEC). Each device was challenge, the U.S. Department of Energy (DOE)
designed to operate in a specific marine resource, initiated the reference model (RM) project. The
instead of a generic device that can be deployed at objectives of this project were to: 1) analyze
any location. This method allows each device to be nonproprietary devices that would be made
used as a reference model to benchmark future available to the general public to allow for
devices. The six designs consist of three current technical and economic benchmarking; and 2)
energy converters and three wave energy assess the potential cost of energy and identify
converters. The reference model project has cost-reduction pathways and areas where
generated both technical and economic data sets additional research could be applied to best
that are available in the public domain. The accelerate technology development to market
methodology to calculate the levelized cost of readiness. Six MHK device point designs were
energy for the reference model project and an developed in the RM project, including three
overall comparison of the cost of energy from current energy converters (CECs) and three wave
these six reference-model designs are presented energy converters (WECs). Each RM model was
in this paper. designed for specific marine resources, which
were modeled after existing U.S. locations.
INTRODUCTION
Studies have shown that marine and The RM study includes structural analysis,
hydrokinetic (MHK) renewable energy has the power output estimation, a power conversion
potential to provide a significant contribution to chain system, and mooring designs. The results
the electricity supply in the United States. The were used to estimate device capital cost and
Electric Power Research Institute (EPRI) annual operation and maintenance (OpEx) costs.
estimates that the total magnitude of the Device performance and costs were used for the
recoverable wave resource is approximately 1,170 economic analysis that included costs for
Terra-Watt hours per year [1], while the designing, manufacturing, deploying, and
combined ocean current, ocean tide, and U.S. river operating single and commercial-scale MEC arrays
resource is approximately 500 Terra-Watt hours for up to 100 devices [6–8].
per year [2-4]. Combined the total MHK resource
is nearly 1/3 of the U.S. electricity demand of Following a preliminary levelized cost of
approximately 4,000 Terra-Watt hours per year energy (LCOE) comparison for RM designs 1–4
[5]. The available resource has renewed interest [9], the objective of this study was to investigate
in research and development (R&D) efforts to the comparison of LCOE for all six RM models. In
develop marine energy conversion (MEC) this paper, the RM model design concepts are
technologies. reviewed. After describing the methodology for
estimating the LCOE, the cost breakdown of the six

1
This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications.
 RM models for a single unit and for 10, 50, and More details of the device design for the RM
100 unit arrays is presented. Finally, a study on models are described in the RM project reports
the overall LCOE comparison for 10-megawatt [7–9]. Brief descriptions of the six models studied
(MW) commercial-scale MEC arrays is presented. in the RM are provided here.

RM MODEL DESIGN CONCEPT • RM1: A dual-rotor, axial-flow tidal turbine

The three CECs that were studied include a (horizontal axis) designed for the reference
horizontal-axis tidal turbine, a vertical-axis location modeled after the Tacoma Narrows
riverine turbine, and a horizontal-axis open-ocean in Puget Sound, Washington.
current turbine. The three WECs include a floating • RM2: A dual-rotor, cross-flow river turbine
body point absorber, a pitching flap device, and an designed for the reference location modeled
oscillating water column (OWC). The schematic of after a section of the lower Mississippi river
the CECs and WECs studied in the RM project are near Baton Rouge, Louisiana.
shown in Figure 1 and Figure 2, respectively. • RM3: A two-body floating-point absorber
designed for the reference location modeled
after a wave site near Eureka, in Humboldt
County, California.
• RM4: A moored glider with four axial-flow
turbines designed for an ocean current
resource modeled after the Florida Strait
within the Gulf Stream off the southeast coast
of Florida near Boca Raton.
• RM5: A floating, oscillating surge WEC
designed for the reference location modeled
after a wave site near Eureka, in Humboldt
County, California.
• RM6: A floating Backwards Bent Duct Buoy
OWC designed for the reference location
modeled after a wave site near Eureka, in
Humboldt County, California.

The data created around these devices was

intended to give future researchers and
FIGURE 1. SCHEMATIC OF CECS (RM1, RM2, RM4) developers a set of reference data for MECs. It is
(ADOPTED FROM [6]) important to note that although each device went
through a rigorous design process, they did not
incorporate any advanced materials, components,
or control strategies. Instead, the six devices
considered were used for techno-economic
benchmarks of similar device configurations.

COST OF ENERGY METHODOLOGY

The lack of MEC devices being deployed in the
United States can be attributed to the high cost of
converting the resource into electricity. Although
the marine resource is free, similar to wind and
solar, the cost of converting that resource depends
on many factors. One of the objectives associated
with the RM project was to understand the largest
cost drivers for each technology at different array
scales.

Cost of Energy Estimate

LCOE is a term that DOE uses to determine the
“break even” cost for a technology assuming a
minimum rate of return. Analysis was performed
FIGURE 2. SCHEMATIC OF WECS (RM3, RM5, RM6) in the RM project to determine the cost of
(ADOPTED FROM [6-8])

2
This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications.
 electricity production for single unit, 10-unit, 50- river current turbine, and RM4 is an open ocean
unit, and 100-unit array sizes. Using these array current turbine. The different resource conditions
sizes for each point design allows for a detailed result in different capacity factor, which measures
breakdown of initial capital expenditures (CapEx) how much average electricity a MHK device
and annual operating expenditures (OpEx). This generates for a period of time relative to the
breakdown is necessary as it gives two of the four electricity it can produce at the rated power
inputs required to calculate LCOE. The other two during the same period. Table 2 lists the capacity
inputs include the annual energy production factor for the CEC models in the RM project and
(AEP) provided to the grid, and the fixed charge the rated power (installed capacity) for these
rate (FCR). The AEP in the RM project was devices.
estimated based on the reference site resource
[10–13] and the results from numerical TABLE 2. CEC CAPACITY FACTOR AND RATED
simulations and experimental tests [14–17]. The POWER FOR A SINGLE UNIT
FCR equates to the annual return that is needed to
meet investor requirements. Included in the FCR Assumed Rated Power
are the real discount rate, inflation, tax rates, Capacity Factor (kilowatts)
depreciation, and project life. The simplified LCOE RM1 0.3 1115
can be represented using these inputs [18]: RM2 0.3 90
RM4 0.7 4000
(FCR × CapEx) + OpEx (1)
LCOE = The smallest of the three is RM2 at only 90
AEP
kilowatts (kW); RM4 is designed for 4 MW, taking
CapEx and OpEx Costs advantage of the high capacity factor because of
CapEx and OpEx costs are further broken the constancy of the Gulf Stream in the Florida
down into a cost breakdown structure (CBS) that Strait. Note that the wide rated power range and
was developed in the RM project. The categories different resource conditions can create a bias
of the CBS used for all the reference models are when looking at LCOE for a particular array size.
shown in Table 1. For example, a single RM4 device delivers
approximately the same order of magnitude of
TABLE 1. COST CATEGORIES FOR CAPEX AND OPEX AEP as the 100-unit RM2 array. Nevertheless, the
trends associated with array size are still valuable
CapEx OpEx in understanding cost drivers and potential cost
Development Insurance reductions for different devices.

Post installation Moving from a single-unit deployment to a 10-

Infrastructure
environmental unit deployment reduces LCOE between 65%-
Mooring/foundation Marine operations 80%, depending on the device. The biggest
contributor to this cost reduction is that OpEx and
Device structural infrastructure costs are shared by multiple units.
Shore-side operations
components The percentages of these costs, as a function of
Power take-off Replacement parts LCOE, continue to drop as more units are
deployed, although the largest reduction in LCOE
Subsystem integration happens between single- and 10-unit arrays, as
Consumables
& profit margin shown in Table 3.
Installation -
TABLE 3. CEC LCOE FOR SINGLE-UNIT, 10-UNIT, 50-
Contingency - UNIT, AND 100-UNIT ARRAY

1-Unit 10- 50- 100-Unit

CapEx costs are broken down even further
Unit Unit
depending on the specific design. This structure
($/kilowatt-hour)
allowed the RM project team to focus on specific
RM1 1.99 0.40 0.20 0.17
costs associated with each category, allowing for a
RM2 2.67 0.78 0.42 0.35
more refined analysis for each device.
RM4 0.67 0.24 0.17 0.15
CURRENT ENERGY CONVERTER COST OF ENERGY
To demonstrate the change in the breakdown
The CECs (RM1, RM2, and RM4) represent
for a different array-size deployment, the LCOE
three different resource types for current energy.
breakdowns for a single-unit and 100-unit array
RM1 is a tidal current turbine, RM2 is a smaller

3
This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications.
 are plotted in Figure 3 and Figure 4, respectively. three categories in addition to an increase in
The breakdowns show that the infrastructure and energy production (e.g., greater availability,
OpEx costs are the primary costs in the single-unit increased capacity per structure, and so on).
case, whereas the device structure and power
take- off (PTO) are the primary cost drivers for the WAVE ENERGY CONVERTER COST OF ENERGY
100-unit array scenario. In particular, the OpEx RM3 is a two-body floating-point absorber,
cost for the single unit is dominated by the RM5 is a floating oscillating surge device, and RM6
environmental monitoring cost because of the lack is an OWC. The WEC devices (RM3, RM5, and
of operational environmental data for the MEC RM6) are all based on the same resource near
devices being deployed. This raises the Humboldt County, California. The resource
uncertainty regarding the environmental impact conformity creates a smaller deviation of installed
from MHK devices and requires in-depth studies capacity when compared with the CEC models.
and monitoring. For the 100-unit array, all three Table 4 lists the capacity factor for the WEC
CECs have a similar LCOE breakdown because the models in the RM project and the rated power for
infrastructure cost diminishes for large array these devices. Because the same capacity factor
deployment. RM1 and RM4 are designed for ocean and wave resource were used when designing
environments, which have similar breakdowns. these three WECs and estimating their power
On the other hand, RM2 is intended for river output, the comparison of the rated power is not
currents, which have a similar 100-unit as significant between the WEC devices as it is for
breakdown but differ from the other CECs in the the CEC devices.
single-unit deployment because of higher
development cost and lower infrastructure cost. TABLE 4. WEC CAPACITY FACTOR AND RATED
POWER FOR A SINGLE UNIT
Assumed Rated Power
Capacity Factor (kW)
RM3 0.3 286
RM5 0.3 360
RM6 0.3 370

The WECs also have a large percentage of

single-unit LCOE attributed to annual OpEx costs.
In addition to the high OpEx costs, another
significant cost driver for single-unit installations
is the development costs, particularly the
environmental cost for permitting and leasing.
Unlike CECs, which have designs similar to wind
FIGURE 3. RM4 LCOE BREAKDOWN (SINGLE UNIT)
turbines and can use the experience from the
wind energy industry, WECs have a higher
development cost that is attributed to the
uncertainty caused by the variety of WEC sizes
and working principles [19].

Similar to the CEC models, the WECs show a

significant LCOE reduction as array size increases
from single- to 10-unit deployments. The
reduction in LCOE as a function of array size for
RM3, RM5, and RM6 are shown in Table 5.

TABLE 5. LCOE FOR THE SINGLE-UNIT, 10-UNIT, 50-

UNIT, AND 100-UNIT ARRAYS
FIGURE 4. RM4 LCOE BREAKDOWN (100 UNIT) 1-Unit 10- 50- 100-Unit
Unit Unit
As the arrays increase up to 100 units the cost ($/kilowatt-hour)
distribution of LCOE shifts so that approximately RM3 4.36 1.41 0.83 0.73
three-quarters of LCOE is a result of the device RM5 3.59 1.44 0.77 0.69
structure, PTO, and annual OpEx costs. Although RM6 4.79 1.98 1.20 1.06
the cost of others components may be reduced,
the greatest reduction in LCOE will be due to these

4
This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications.
 The reasoning for the relative difference in
RM3 and RM5 LCOE when looking at single-unit
and 10-unit arrays can be attributed to the
estimated pre- and post-installation
environmental costs. Some of this can be
attributed to the varying foundation design, with
the rest being caused by variations in assumptions
between the three WEC devices [19, 20]. Although
the assumptions have a significant impact at the
single-unit scale, there is little impact on LCOE as
the array goes to 100 units.

Figures 5 and 6 compare the LCOE

FIGURE 6. RM5 LCOE BREAKDOWN (100 UNIT)
breakdowns for RM5 at single-unit and 100-unit
array deployments. The identical resource and 10-MW SMALL COMMERCIAL-SCALE COMPARISON
similar installed capacity makes for a better side- The differences in the resource conditions and
by-side comparison of the WEC devices. Although the device rated power make it unfair to compare
there are subtle differences between RM3, RM5, the LCOE for the six MHK point designs studied in
and RM6, the percentage breakdown of LCOE is the RM project. To investigate the LCOE cost
similar. The largest deviation is the contribution of reduction due to increasing array size, a
structural costs; the WEC's LCOE is dominated by polynomial curve fit of the four array sizes was
structural costs. The RM3 structural costs account used. The LCOE is plotted against the installed
for 37% of the LCOE at 100-unit deployments, capacity in Figure 7. The LCOE for each RM model
whereas RM5 and RM6 are 41% and 52%, was estimated at 10 MW for small commercial-
respectively. Much of this deviation can be scale, and the results are listed in Table 6. The
associated with overdesign, particularly for RM6 LCOE for the 10-MW CECs are in the range
[8]. Unlike CECs, the mooring cost has a bigger between $0.3 per kilowatt-hour (kWh) and
impact than that from PTO. High mooring costs $0.5/kWh. In comparison, the LCOE values for the
are expected in these devices because all three same 10-MW installed capacity WECs are in the
WECs use floating platforms. Floating designs range between$ 1.0/kWh and $1.5/kWh.
inherently require high mooring loads to resist
power dissipation via device movement,
particularly for RM5, where the device requires a
taut mooring [6]. The relative costs of the PTO and
device structure are shifted because of the large
structural costs of the WEC devices. It is likely that
a more structurally optimized design will reduce
these costs.

FIGURE 7. RM MODELS' LCOE FOR DIFFERENT

INSTALLED CAPACITIES

TABLE 6. LCOE ESTIMATE FOR THE 10-MW SMALL

COMMERCIAL-SCALE
RM Model LCOE ($/kWh)
FIGURE 5. RM5 LCOE BREAKDOWN (SINGLE UNIT) RM1 0.42
RM2 0.31
RM3 0.98
RM4 0.48
RM5 0.98
RM6 1.47

5
This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications.
 [3] Haas K., 2011, Assessment of Energy
CONCLUSIONS Production Potential from Tidal Streams in
A study on the LCOE for six MEC reference the United States.
model point designs is presented in this paper. [4] Jacobson P. T., Ravens T. M., Cunningham K.
The six designs consist of three CECs and three W., Scott G., 2012, Assssment and Mapping of
WECs. The LCOE was estimated based on four the Riverine Hydrokinetic Resource in the
primary inputs, including CapEx, OpEx, AEP, and Continental United States.
FCR. The cost breakdown of the six RM models [5] U.S. Energy Information Administration, 2014,
was analyzed for up to a 100-unit array, and a Electric Power Monthly.
study on the overall LCOE comparison of the [6] Neary V. S., Previsic M., Jepsen R. a., Lawson M.
models for 10-MW commercial-scale arrays were J., Yu Y.-H., Copping A. E., Fontaine A. a.,
presented. Hallett K. C., 2014, “Methodology for Design
and Economic Analysis of Marine Energy
The study shows that CECs (e.g., 10 MW Conversion (MEC) Technologies,” 2nd Marine
installed capacity) are within the range of early Energy Technology Symposium, Seattle, WA.
market adoption primarily because the CEC [7] Yu Y.-H., Jenne D. S., Thresher R., Copping A.,
technologies are more mature then WECs. Much of Geerlofs S., Hanna L., 2015, Reference Model
the maturity associated with CEC is a result of the 5: Oscillating Surge Wave Energy Converter,
knowledge gained from offshore and land-based National Renewable Energy Laboratory
wind technology; however, technology (NREL), Golden, CO.
advancements that will lead to significant LCOE [8] Bull D., Smith C., Jenne D. S., Jacob P., Copping
reductions are still needed to be widely A., Willits S., Fontaine A., Brefort D., Copeland
competitive. Cost reductions for CEC devices will G., Gordon M., Jepsen R., 2014, Reference
most likely result from improving OpEx strategies Model 6: Oscillating Wave Energy Converter,
and reducing PTO cost. WEC devices, on the other Sandia National Laboratories, Albuquerque,
hand, are further behind on the market readiness NM.
scale, and there is little convergence on a standard [9] Neary V. S., Previsic M., Jepsen R. A., Lawson
WEC technology, particularly with respect to and M. J., Yu Y.-H., Copping A. E., Fontaine A. A.,
device size. In addition, the WECs studied in this Hallett K. C., Murray D. K., 2014, Methodology
RM project are most likely overdesigned for Design and Economic Analysis of Marine
structurally, as mentioned in the last section. The Energy Conversion (MEC) Technologies,
systems can be further improved by implementing Sandia National Laboratories, Albuquerque,
advanced control strategies to optimize their NM.
power performance. In addition to the cost [10] Neary V. S., 2011, Reference Inflow
reduction from OpEx and PTO, structure design Characterization for River Resource
innovation, and power performance improvement Reference Model: Reference Model 2 (RM2),
are two important areas that need additional Oak Ridge National Laboratory, Oak Ridge,
research to accelerate WEC technology Tennessee.
development to market readiness. [11] Polagye B., and Thomson J., 2013, “Tidal
energy resource characterization:
ACKNOWLEDGEMENTS methodology and field study in Admiralty
This work was supported by the U.S. Inlet, Puget Sound, WA (USA),” Proc. Inst.
Department of Energy under Contract No. DE-AC36- Mech. Eng. Part A J. Power Energy, p.
08-GO28308 with the National Renewable Energy 0957650912470081.
Laboratory and Contract No. DE-AC04-94-AL85000 [12] Raye R. E., 2002, “Characterization study of
with Sandia National Laboratories. Funding for this the Florida Current at 26.11 north latitude,
work was provided by the DOE Office of Energy 79.50 west longitude for ocean current
Efficiency and Renewable Energy, Wind and Water power generation,” Florida Atlantic
Power Technologies Office. University.
[13] Dallman A., and Neary V., 2014, “Initial
REFERENCES Characterization of the Wave Resource at
[1] Jacobson P., Hagerman G., Scott G., 2011, Several High Energy U.S. Sites,” 2nd Marine
Mapping and Assessment of the United States Energy Technology Symposium, Seattle, WA,
Ocean Wave Energy Resource, Electric Power pp. 1–7.
Research Institute. [14] Hill C., Neary V. S., 2014, U . S . Department of
[2] Haas K., 2013, Assessment of Energy Energy Reference Model Program RM1 :
Production Potential from Ocean Currents Experimental Results, Sandia National
along the United States Coastline. Laboratories, Albuquerque, NM.

6
This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications.
 [15] Hill C., Neary V., 2014, U . S . Department of Performance Reporting for Marine and
Energy Reference Model Program RM2 : Hydrokinetic Technologies,” 1st Marine
Experimental Results, Sandia National Energy Technology Symposiumt,
Laboratories, Albuquerque, NM. GMREC/MTES.
[16] Yu Y., Lawson M., Li Y., Previsic M., Epler J., [19] Copping A., Geerlofs S., Hanna L., 2014, The
Lou J., 2015, Experimental Wave Tank Test Contribution of Environmental Siting and
for Reference Model 3 Floating- Point Permitting Requirements to the Cost of
Absorber Wave Energy Converter Project, Energy for Wave Energy Devices: Reference
National Renewable Energy Laboratory Model 5, Pacific Northwest National
(NREL), Golden, CO. Laboratory (PNNL).
[17] Bull D., Gunawan B., Holmes B., 2014, [20] Copping A., Geerlofs S., 2011, The
Experimental Confirmation of Water Column Contribution of Environmental Siting and
Natural Resonance Migration in a BBDB Permitting Requirements to the Cost of
Device, Sandia National Laboratories, Energy for Marine and Hyrokinetic Devices:
Albuquerque, NM. Reference Models #1, #2, and #3, Pacific
[18] LaBonte A., Connor P. O., Fitzpatrick C., Northwest National Laboratory (PNNL).
Hallett K., Li Y., 2013, “Standardized Cost and

7
This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications.