Understanding the financial performance of an engineered system is a key step to its commercialization. In this study, the economic performance of the Mk1 PB-FHR using a nuclear air combined cycle to produce base load nuclear power, and highly efficient peaking power with gas co-firing, was estimated for a regulated electricity market structure.
Trang 1Nuclear Air Brayton Combined Cycle and Mark 1 Pebble Bed
Fluoride-Salt-Cooled High-Temperature Reactor economic performance in a
regulated electricity market
Charalampos Andreadesa,⇑, Per Petersonb
a
University of California, Berkeley, 4118 Etcheverry Hall, Berkeley, CA 94720, United States
b
University of California, Berkeley, 4167 Etcheverry Hall, Berkeley, CA 94720, United States
h i g h l i g h t s
Mk1 FHR performs favorably compared to both utility and IPP built NGCCs
Mk1 FHR main performance drivers: electricity price, NG price, and the discount rate
Mk1 is much more attractive in markets where NG prices are high compared to NGCCs
a r t i c l e i n f o
Article history:
Received 28 April 2016
Accepted 11 December 2016
Available online 29 December 2016
Keywords:
Nuclear economics
Nuclear Air Brayton Combined Cycle
Flexible nuclear
FHR
Regulated electricity market
NGCC
a b s t r a c t
Understanding the financial performance of an engineered system is a key step to its commercialization
In this study, the economic performance of the Mk1 PB-FHR using a nuclear air combined cycle to pro-duce base load nuclear power, and highly efficient peaking power with gas co-firing, was estimated for
a regulated electricity market structure Initially, a survey of major U.S nuclear utility holding companies’ financials was performed to estimate a credible range of input parameters In combination with the main cost parameters of the Mk1 estimated in a companion paper, a base case analysis was performed, demon-strating the economic attractiveness of the Mk1 A sensitivity study demonstrated that the main metrics
of concern were electricity price, natural gas price, and the discount rate These all pointed to possible ways to further reduce the Mk1’s investment risk, such as long term fuel contracts and improved con-struction management, in order to further increase the attractiveness of Mk1 deployment Finally, a com-parison between the Mk1 and two different natural gas combined cycle (NGCC) plants was made The Mk1 performance lies in between a utility built and an independent power producer built NGCC The Mk1 becomes a much more attractive investment than conventional NGCCs in markets where natural gas prices are high
Ó 2016 The Authors Published by Elsevier B.V This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
1 Introduction
One of the most important aspects of designing a new
commer-cial technology is understanding its revenues and long term
eco-nomic viability There are certain instances where an investor or
business is willing to accept a loss on a specific product (e.g loss
leaders, technical displays), but in general the aim is to create value
and generate profit in the long term The profit of a product
depends on two specific components, namely cost and revenue,
the difference between the two being the profit/loss This paper
assesses revenues for Mark-1 Pebble Bed, Fluoride Salt Cooled
Reactors (Mk1 PB-FHRs) coupled to nuclear air combined cycle (NACC) power conversion (Andreades et al., 2014a, 2016) Narrowing our focus to the electricity sector, the main market
of the FHR and NACC (Mk1), it is important to understand the fun-damentals of this sector’s operation and the ways in which it has evolved over its lifetime Here we focus on the U.S electricity sec-tor, although the conclusions can be generalized to other countries During the nascent years of the electricity industry at the turn of the 20th century, U.S electric utilities operated in a fiercely com-petitive environment, competing primarily in price with gas light-ing and self-generation There was discussion of appropriate rate structures, such as time-of-use and block pricing, however the need for stability and investor attractiveness pushed industry pioneers, such as Samuel Insull, to promote demand charges and government regulation of utilities as protected monopolies This http://dx.doi.org/10.1016/j.nucengdes.2016.12.013
0029-5493/Ó 2016 The Authors Published by Elsevier B.V.
⇑Corresponding author.
E-mail addresses: charalampos@berkeley.edu (C Andreades), peterson@nuc.
berkeley.edu (P Peterson).
Contents lists available atScienceDirect Nuclear Engineering and Design
j o u r n a l h o m e p a g e : w w w e l s e v i e r c o m / l o c a t e / n u c e n g d e s
Trang 2structure, in which utilities are guaranteed cost recovery at a
reg-ulated rate – cost plus x – allowed them to be shielded from
com-petition, take advantage of economies of scale, and expand Thus
was the status quo for the next seven decades Around the
1990s, an interest in electricity market liberalization and
deregula-tion took shape due to success in deregulating other industries,
such as telecommunications, trucking and commercial aviation,
and a resurgence in competitive pricing in electricity markets
was in vogue This shifted electricity pricing away from average
cost (AC) based to marginal cost (MC) based The changing nature
of modern electricity markets was and remains compounded by
the large scale introduction of intermittent renewable energy
sources Flexible and quickly ramping capacity is needed to
main-tain grid stability, since traditional fossil fuel sources have physical
ramp constraints and battery reserves are not well suited to utility
scale capacities and demands
The Mark 1 (Mk1) NACC is a novel power conversion system
based on a modified General Electric (GE) 7FB natural gas (NG)
tur-bine The turbine is retrofitted to accept external heating from a
heat source in the range of 600–700°C, in this application an
FHR, while also maintaining its ability to combust NG or other
combustible fuel When coupled to the 232 MWt Mk1 PB-FHR,
NACC provides 100 MWeof baseload electricity with a 42%
effi-ciency, and a boosted power output of 240MWe under NG
co-firing with a NG-to-electricity conversion efficiency of 66%, well
above current state of the art NGCCs A full technical description
of the NACC can be found inAndreades et al (2014b, 2014c) The
NACC, with its ability to peak on-demand and provide flexible
capacity make it an attractive and well suited candidate for the
current and future low carbon electricity markets, with high
pene-tration of intermittent renewable energy sources To assess the
Mk1 economic allure vis-à-vis its operating and physical benefits,
this study aimed at initially quantifying the Mk1’s revenue under
certain hypotheses and constraints in a regulated electricity
mar-ket A description of the methodology used to perform the revenue
estimation is given, followed by a summary of the relevant
operat-ing and cost inputs from a companion paper (Andreades, 2015)
The revenue and profitability results are then presented, followed
by discussion of the Mk1 results and a comparison made to its
main competitors
2 Methodology
In order to create a regulated market revenue model, an
indus-try standard commercial software package, THERMOFLEX/PEACEÒ,
was used (Thermoflow) Once a baseline NACC configuration was
established based on the Mk1 PB-FHR commercial point design
(Andreades et al., 2014a, 2016), and as detailed in Andreades
et al (2014b, 2014c), relevant cost estimates were given A market
survey of major U.S nuclear utilities was performed to obtain a
plausible range of financing and electricity market data A base
case was run with average values to establish a baseline reference,
followed by a sensitivity study on each parameter separately Two
additional cases were run, an ‘optimistic’ and a ‘pessimistic’ one, in
order to bound the results Finally, a comparison was carried out
between the NACC and a NGCC power plant based on the GE 7FB
of similar power output, in order to establish how well the
pro-posed design performed against its assumed main competitor All
currency units are set to 2014 USD
3 Input data
The first step to performing a profitability analysis is assessing
costs of the system in question, as given by Eq.(1)
ProfitðLossÞ ¼ Revenue Cost ð1Þ
The relevant costs for the Mk1 were estimated in a companion paper and a summarized version is presented inTable 1
The next step is to appropriately identify and estimate financing numbers and structures that fit such a project and as required for input by THERMOFLEX/PEACEÒ’s, ‘Economic and regional costs’ tab Some basic operating assumptions were made The Mk1 is anticipated to have a 60 year lifespan; however, THERMOFLEX/ PEACEÒis limited to a 40 year assessment In lieu, one can simply extrapolate the 40 year results to a 60 year lifetime For the pur-poses of this study and for added conservatism a 40 year lifetime was assumed
The first year of plant operation was assumed to be 2021, fol-lowing an assumed 5-year construction period, for a 12-unit plant THERMOFLEX/PEACEÒ does not account for staggered construc-tion/operation which would provide added realism and thus results are conservative as initial revenue is generated at a later date, rather than as individual units come online Such a modeling approach can be considered as a counterbalance to potential con-struction delays
The NACC is anticipated to operate in a load-following mode due to its flexible capacity provided by its ability to produce peak-ing power by injectpeak-ing NG or other liquid and gaseous fuels when quick ramping is needed by the electricity grid For this study it was assumed that the 12-unit Mk1 NACC station ran at either
1200 MWenuclear capacity or at a full 2832 MWeco-fired capacity The capacity factor of the plant was assumed to be the 10-year nuclear industry average of 90%, with range between 80% and 95% (Nuclear Energy Institute, 2014) The Mk1’s online refueling capability might enable a higher capacity factor, but current indus-try average was used for the base case instead for conservatism Typically, nuclear installation depreciation terms are set at
15 years (Department of Commerce Bureau of Economic Analysis,
2004) The Nuclear Energy Institute is proposing lowering this term to 7 years, as it affects a plant owner’s tax expense (Fertel,
2004) A shorter depreciation term allows for a larger accounting expense each year and therefore reduced taxes in earlier years A
30 year high was used for the depreciation range
Debt terms for nuclear facilities are typically set at 15 years (OECD-NEA, 2009; IAEA, 1993) Longer terms allow for longer peri-ods to repay and service the debt and are therefore more attractive
A 30 year maturity date was used on the high side, while the
15 year term was used as the base and lower range
The following three financing components, namely debt per-centage, debt interest rate, and discount rate, are usually highly project specific and in many cases confidential to the parties
Table 1 Overview of Mk1 costs.
Description Single unit 12 Unit Capital construction costs
Preconstruction costs 80,484,991 263,622,515 $ Total direct cost 214,846,727 2,578,160,727 $ Indirect cost 142,462,635 1,709,551,614 $ Total contingency 71,461,872 857,542,468 $ Total capital investment 509,256,225 5,408,877,325 $ Specific capital investment
(nuclear)
Specific capital investment (CF) 2133 1870 $/kW Production Costs
Total annual O&M 62,086,683 311,631,799 $ Fuel cost (annual) 7,750,516 93,006,192 $ Decommissioning cost (annual) 1,165,920 13,991,046 $ Overall production cost 71,003,119 418,629,037 $ Marginal production cost 81.05 39.82 $/
MW h Bolded numbers are the key comparison metrics used to compare electricity
Trang 3gen-involved Additionally, no major nuclear construction of new
plants beyond Vogtle in Georgia and VC Summer in South Carolina,
has happened in the United States the past two decades so reliable
or relevant numbers are in short supply Although Watts-Bar 2 was
completed in 2015 with an approved cost of $4.7bn, it does not
provide relevant information since its construction began in 1972
with a 22 year hiatus between 1985 and 2007 (World Nuclear
Association, 2016) The approach taken to estimate these numbers
was to study and average the financial statements and fillings of
the major nuclear utilities, as shown inTable 2
In order to use these numbers correctly, a corporate financing
structure rather than a project financing structure was assumed
This assumption is particularly appropriate for small modular
reac-tor stations like the Mk1 PB-FHR station, because the capital placed
at risk before initial electricity is delivered to the grid is
substan-tially smaller than for conventional large reactor plants Corporate
financing usually implies gentler financing terms and higher
lever-age since lenders have a recourse on the utility’s balance sheet, not
just the project itself
Debt percentage was estimated from each company’s most recent public annual balance sheet from the basic accounting
Eq.(2)
Assets¼ Liability þ Equity ð2Þ
Next, debt interest rate and tax rate were obtained from nuclear utility 10k filings Finally, the discount rate was assumed to be the rate of return to equity investors To approximate the discount rate, the weighted average cost of capital (WACC), as defined in
Eq (3), was obtained from financial analysis companies (GuruFocus.com LLC, 2015) and modified to yield our estimate
WACC¼EquityAssets CostequityþLiabilityAssets Costliability ð1 TaxrateÞ ð3Þ
The overall results for all major U.S nuclear utility holding com-panies are presented inTable 3
The three remaining inputs that were estimated were the price
of carbon tax and the prices of electricity and natural gas The price of carbon was assumed to vary between zero, to reflect the current U.S status quo, and 120 $/tCO2e as a ceiling with a base of 40 $/tCO2e (Interagency Working Group on Social Cost of Carbon, 2013) The natural gas price range was obtained from the historical range of prices from the past 15 years (Chicago Mercentile Exchange, 2015) An electricity price range was obtained from the 2015 Annual Energy Outlook (U.S Energy Information Administration, 2014) The input parameters are sum-marized inTable 4
A final note is that escalation rates for costs were assumed to be constant and equivalent to inflation
4 Results The results for the base case assumptions are presented in
Table 5, for both baseload and co-fired operation
What is apparent from the results inTable 5is that under con-stant full capacity operation the Mk1 seems to be an attractive
Table 3
Summary of financial data for major U.S nuclear utility holding companies.
Name NYSE Total assets [bn $] Total equity [bn $] Equity [%] Debt [%] Tax rate [%] WACC [%] Cost of debt [%] ROR [%]
a Not a publicaly traded company, Data from 2014 10K filings.
Table 4
Financing and market input parameters.
Table 2
Major U.S nuclear utility holding companies.
a
Number of plants is different to number of reactors Each nuclear plant can have
one or more reactors on site.
Trang 4investment, with a net present value (NPV) of $18.8bn, a levelized
cost of electricity (LCOE) of $0.045/kW h and a breakeven price of
NG of $21.6/GJ (22.8 $/MMBtu), i.e the price of NG up until which
the plant remains profitable On the other hand, under base load
operation the Mk1 is an unprofitable investment and is thus
dis-counted for the rest of this analysis A note to make is that the
Mk1 will run in its peaking mode only when prices of electricity
are above the price where electricity revenues exceed natural gas
costs, and it is quite unlikely that it will run at full capacity at all
times
The results for the co-fired deviating cases are presented
graph-ically for compactness inFigs 1–5 The effect of each input variable
on selected major dependent variables is illustrated
Fig 1demonstrates that NPV is most sensitive to discount rate, electricity price, and NG price Electricity price affects the revenue cash flow of the Mk1, with reduced prices resulting in reduced rev-enue NG prices affect the cost stream of the Mk1, with increased prices reducing profit margin The results vary dramatically as parameters vary Other input parameters have much smaller, yet not negligible effects A counterintuitive result is the relatively small effect of an imposed carbon tax (12% range in NPV from base case) compared to the dominating parameters The high conver-sion efficiency of NG to electricity allows the NACC to burn less
NG and thus reduce its carbon tax cost and thus mitigate a more pronounced impact; however, if considered separately, an approx-imately 10% swing in NPV is still significant
Table 5
Mk1 base case (40 $/tCO2e) economic performance under peaking and baseload operation.
Break-even NG LHV price @ input electricity price 22.8 (21.6) 2.68 (2.53) $/MMBtu ($/GJ)
Trang 5Fig 2demonstrates that return on investment (ROI) is most
sen-sitive to NG price, electricity price, and capacity factor for co-fired
operation This makes sense since electricity price and capacity
fac-tor determine the revenue stream, while NG price affects the cost stream and thus the time and amount needed to pay back investors
in a timely manner, in turn affecting return on investment
Fig 2 Return on investment of project.
Trang 6Fig 3illustrates that the LCOE is most sensitive to NG price,
dis-count rate, and to a lesser extent carbon price A carbon tax is a
price adder by definition Higher discount rate increases LCOE by
increasing returns demanded by investors NG price is directly linked to the cost of producing electricity It is imperative to note that only high NG prices push LCOE to exceed 0.05$/kW h, but on
Fig 4 Years to payback equity.
Trang 7the other hand can lower it down below 0.04$/kW h All other
parameters keep LCOE within that range
As NG price increases it keeps imposing a larger production
cost, up to a certain point where it then makes the plant
unprof-itable (i.e the break-even price of NG) If electricity market
feed-back is considered, the BL capacity provided by nuclear heat will
have a larger profit margin as NG price increases, since the price
will be set by competing NGCCs or GTs, as the NACC’s co-fired
capacity has a much higher thermal efficiency
Fig 4shows that NG price and electricity price affect years of
payback of equity by shifting the profit margin on cost and revenue
for production costs All other variables have negligible effects and
a range of payback on co-fired operation between a year and two years
Fig 5depicts the effects of discount rate, electricity price, and capacity factor on the break-even NG price at input electricity price As discount rate increases, the NG price threshold is lowered, thus becoming more restrictive As electricity price and capacity factor increase, so does the NG threshold, meaning that the co-fired Mk1 can remain profitable even at higher NG prices, up to
$31/GJ
The bounding cases, where all input parameters were set to either optimal or worst case, are presented in Table 6 What becomes immediately clear is that the range of results is too broad
Table 6
Bounding case financial results for Mk1.
Cumulative net cash flow 180,443,800,000 109,520,200,000 20,482,060,000 USD
Break-even fuel LHV price @ input electricity price 32.84 21.6 4.878 USD/GJ
Ò
Trang 8to make any definitive conclusions, other than that both cases are
unlikely
5 Discussion
What is distilled from the results in the previous section is that
three parameters recurrently affect the economic performance of
the Mk1, namely NG and electricity prices, and discount rate
Addi-tionally, these parameters are mainly market and not operation
(capacity factor) driven Although other variables should also be
kept in mind, they do not have as pronounced an effect as the aforementioned This leads us to concentrate on understanding how one might try to positively affect each parameter, if at all possible
The price of NG or combustible fuel is to a large extent external and can only be set advantageously through long term delivery contracts, rather than being purchased on the volatile spot market
In many international markets, natural gas prices are sufficiently high that the Mk1 economics will be attractive
Electricity price to a certain extent can be affected by bringing
on or taking off the Mk1’s flexible capacity, when the opportunity cost might warrant it However, there needs to be a strong consid-eration to avoid applying market power to manipulate spot prices
of electricity, since a full Mk1 plant has a significant peaking reserve capacity
The amount of remuneration demanded by equity investors, namely through the discount rate, signals the risk perception of the individual project in tandem with that of the industry as a whole What the Mk1 needs to accomplish to become attractive
is to convince investors of a superior risk profile vis-à-vis its com-petitors Steady and added revenues from its flexible capacity, reduced capital investment due to stacked construction and oper-ation, economies of series due to factory production should all help
to reduce the risk profile of the Mk1 Furthermore, steady, manage-able, and predictable costs should also be a priority
On an industry front, nuclear needs to be perceived as investor friendly, through streamlined regulation, improved construction and supply chain management A deeper elaboration on these issues is beyond the scope of this study
Fig 7 Three pressure steam turbine for combined cycle THERMOFLEX/PEACEÒschematic.
Table 7
NGCC operation and financial parameters.
IPP Utility
Total overnight cost 1055 1055
Trang 95.1 Comparison to conventional NGCC
In order to obtain a better perspective of the previous economic
results of the Mk1, a comparison between it and its competitors
was deemed necessary The most obvious candidate to compare
the Mk1 to was an unmodified NGCC based on the GE 7FB This
comparison essentially shows whether the Mk1’s more efficient
burning of NG during peak operation compared to a conventional
NGCC could make up for the added capital cost of the nuclear
com-ponent of the system A THERMOFLEX/PEACEÒmodel using a five
2 1GE 7FB and steam turbine configuration was used to match
the power output of the Mk1 at 2800 MWe, as shown inFigs 6
and 7
The main input parameters for the NGCC were selected from
industry averages of advanced NGCC from the Energy Information
Administration’s 2015 Annual Energy Outlook (U.S Energy
Information Administration, 2015) and Alstom (Bozzuto, 2006)
and are presented inTable 7 A utility and independent power
pro-ducer (IPP) financing structure were used to compare the two
dif-ferent scenarios of constructing NGCCs
The base case results of the co-fired Mk1 were compared to those of the NGCC and are presented inTable 8 A sensitivity study was performed on the price of carbon, electricity price, and NG price and compared to the Mk1 graphically inFigs 8–10, with all metrics being normalized to the values of the Mk1 base case Under base case assumptions with a 40 $/tCO2e carbon price, the co-fired Mk1 compared favorably to both an IPP NGCC installa-tion and a utility built NGCC The Mk1 managed a NPV of 1$8.8bn
as compared to $4.4bn of an IPP NGCC and $12.7bn of a utility NGCC Return on investment was nearly double for the NGCCs due to the smaller initial capital required compared to the Mk1 The LCOE of the Mk1 was lower than an IPP NGCC and slightly higher than a utility owned NGCC The break-even price of NG, i.e the price of NG up until which the plant remains profitable, was significantly higher than both NGCCs, demonstrating that the Mk1 would perform very favorably in markets with high NG prices, such as Europe and Japan The main difference between the IPP and utility NGCCs is the longer running period of the utility plant, which led to more favorable financial results This should also be considered when looking at the results for the Mk1, since
Table 8
Financial performance of competitors under base case assumptions.
Cumulative net cash flow 109,520,200,000 30,452,090,000 61,093,990,000 $
Break-even fuel LHV price @ input electricity price 21.60 11.81 13.63 $/GJ ($/MMBtu)
Trang 10only two thirds of its operational lifetime were accounted for in
this study An additional fact to consider is that the Mk1 managed
to produce more power with a smaller carbon footprint compared
to a NGCC of a similar size, due to its improved power conversion
efficiency and producing part of the power by non-emitting
nuclear fuel
Changing electricity price can have a dramatic effect on the
per-formance of plants as pointed out in the previous section As
shown inFig 8, as electricity price increased all plants performed better All results were normalized to the Mk1’s base case results The IPP NGCC could not compete with the co-fired Mk1 on NPV even under high electricity prices The utility owned NGCC also struggles to compete with Mk1 as electricity prices vary The same basic comparative outcomes as the base case still applied under the electricity sensitivity study
Fig 9 Normalized financial parameters under NG price variation.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
NGCC
Ulity NGCC
NGCC
Ulity NGCC
NGCC
Ulity NGCC
NGCC
Ulity NGCC
Carbon tax variaon