Taming WECC's Carbon Diet: Are We Losing Weight Yet?

Amy Sopinka is a post-doctoral research fellow at the Pacific Institute for Climate Solutions. She completed her Ph.D. at the University of Victoria. Her dissertation work focused on the economic and environmental effects of wind integration in the weakly interconnected Alberta and British Columbia grids. She obtained an M.A. from McGill University and a B.A. from Queen’s University. Lawrence Pitt earned his B.Sc. in Applied Physics and M.Sc. in Electrical Engineering from the University of Alberta and a Ph.D. in Plasma Physics from the University of Victoria. Dr. Pitt has been involved in research and development for a variety of energy systems: laserfusion; combustion systems and alternative-fuel vehicle systems design; hydrogen systems, and grid integration of large-scale renewables. At various times, he has been a Combat Systems Engineer in the Navy, navigated a tall ship cruising the Pacific, developed and coordinated the early years of the engineering co-op program at UVic, and served as Research Coordinator at UVic’s Institute for Integrated Energy Systems (IESVic). Dr. Pitt is currently Associate Director of the Pacific Institute for Climate Solutions.

96

Taming WECC’s Carbon Diet: Are We Losing Weight Yet? What is the comparative cost of achieving carbon emissions reductions in the Western Interconnect via growth in variable energy resources versus meeting incremental demand by either coal-fired or natural gasfired generation? An analysis suggests that more emissions are mitigated with the use of VERs, but the cost of achieving those reductions is significantly higher. Amy Sopinka and Lawrence Pitt

I. Introduction The North American bulk power system is comprised of four interconnected regions: the Western Interconnect, the Eastern Interconnect, Texas, and Quebec. Arguably, the largest and most diverse of those areas is the Western Interconnect, which spans from northern Alberta and British Columbia in Canada to the Baja California peninsula in Mexico and includes all or part of 14 U.S. states in between. The varied geography and geology of the region creates a wide range of possible energy sources. The

Pacific Northwest is dominated by mountains and rivers and therefore has many large storage hydroelectric facilities. The interior portion of the Western Interconnect that runs from Alberta southward has substantial coal and natural gas deposits and these areas tend to have a higher proportion of fossil fuel capacity and generation. Several U.S. states within the region have nuclear generating capacity; however, the use of this technology is prohibited in British Columbia and the province of Alberta currently has no plans to introduce nuclear electricity

1040-6190/$–see front matter # 2014 Published by Elsevier Inc., http://dx.doi.org/10.1016/j.tej.2014.04.004

The Electricity Journal

production into its generation mix. Low-carbon energy sources, particularly the variable energy resources (VER) of wind, solar, and run-of-river technologies, are increasingly the focus of WECC policymakers, fostered by the underlying conviction that adding zero-carbon resources will reduce emissions in the electricity sector. In the work that follows, we roughly estimate the quantity of carbon abatement expected from state and provincial policies that are designed at increasing low-carbon capacity and generation. We do not consider the carbon production impacts of connecting VERs to the grid or of firming their output once they enter service, both of which would have the effect of raising the cost of CO2 abatement with the addition of VERs. In addition, we provide a lower bound on the cost of reducing emissions in a system with significant lowcarbon assets.

II. The Current Generation Mix in the Western Interconnect Geography and geology play a large role in determining the generation mix in the Western Interconnect, which, in turn, impacts the magnitude of carbon emissions attributed to the electricity sector. To adequately evaluate the diverse generation mix across the three countries, accurate and consistent data was required. This information became available in 2002 when October, 2014,

Vol. 27, Issue 8

the Western Electricity Coordinating Council (WECC), the region’s reliability entity, was created; the most recent accurate data across all jurisdictions is for the year 2011. Using data from the Energy Information Administration (EIA), the Mexican National Institute of Statistics and Geography, the Alberta Electric System Operator (AESO), and various Crown corporations in British Columbia, production volumes were estimated and aggregated into four categories: ‘‘coal, etc.’’ includes all non-natural-gas fossil fuels, while ‘‘renewables’’ incorporates low- and zerocarbon resources technologies such as storage, hydroelectric, wind, solar, run-of-river, geothermal, and biomass. The remaining two groupings comprise nuclear and natural gas technologies. The total amount of low-carbon resources within the system, as defined above, is equal to the sum of the renewable and nuclear categories. After compiling the available data, a [(Figure_1)TD$IG]complete picture of the generation

mix within the Western Interconnect in 2002 was developed (Figure 1). y combining the renewables and nuclear generation segments, it is evident that as early as 2002, 43.6 percent of electricity generation in the Western Interconnect came from low-carbon resources. However, despite this significant contribution, policies were created at both the state and provincial level aimed at further reducing carbon emissions via the required integration of additional low-carbon resources. In Alberta, the generation mix within the energy-only deregulated electricity market depends exclusively on investor behavior, although the Province has legislated carbon reductions from existing specified emitters of greater than 100,000 tons of CO2e per year (CanLii, 2013). In British Columbia, the Clean Energy Act requires that at least 93 percent of generation within the province must come from clean or renewable technologies (Province of British Columbia, 2010). In the

B

NUCLEAR 9.8%

COAL ETC. 34.3%

756.2 TWh

RENEWABLES 33.8% NATURAL GAS 22.0%

Figure 1: Generation Mix (percent) in 2002 by Type and Total TWh of Production

1040-6190/$–see front matter # 2014 Published by Elsevier Inc., http://dx.doi.org/10.1016/j.tej.2014.04.004

97

[(Figure_2)TD$IG] 120 100

GW

80 60 40 20 0

2002

2003

HYDROELECTRIC

2004

2005

NUCLEAR

2006

GEOTHERMAL

2007

2008

WIND

2009

2010

2011

SOLAR AND OTHER RENEWABLES

Figure 2: Installed Low Carbon Capacity by Type (GW)

U.S., renewable portfolio standards (RPS) are formed at the state level and can vary widely. For example, California’s current RPS demands 33 percent renewable energy by 2020 and the state also instigated a cap-andtrade system for carbon allowances which began in January 2013 (California Environmental Protection Agency, 2011). By comparison, Montana has an RPS that requires 15 percent of electricity generated by 2020 be from qualifying facilities; Utah created a goal (rather than a legislated requirement) of 20 percent renewable production by 2020, while Idaho currently has no renewable requirements (US Department of Energy, 2013). In order to meet state RPS and provincial clean energy policy goals, it is expected that lowcarbon capacity in the region will need to increase significantly. For example, it is estimated that California will require nearly 24 GW of installed renewable capacity to meet its 33 percent RPS target, and most of the increase is expected to come in the form of VERs (CPUC, 2009). 98

a. Existing low-carbon capacity in the Western Interconnect The impact of low-carbon policies on the electricity sector can be shown through an examination of installed capacity levels. Low-carbon electricity sources in the Western Interconnect are depicted by type in Figure 2, and the effect of state and provincial energy policies becomes apparent toward the end of the series, with the expansion of wind, solar, and other renewable capacity. ver the 10-year period, lowcarbon capacity increased by 16.7 GW, and nearly 90 percent, or 14.4 GW, of that increase was due to the addition of [(Figure_3)TD$IG] VER capacity. Clearly,

O

however, hydroelectric facilities, shown by the black bars in Figure 2, were a key resource in the WECC mix throughout the observed period, comprising approximately 73 percent of the region’s low-carbon capacity (and 30 percent of the area’s total installed capacity) even prior to the instigation of energy policies dictating the use of low-carbon resources. b. Low-carbon generation in the Western Interconnect While installed capacity figures show the potential to generate electricity, the production volumes reveal the actual output from those generating facilities. In Figure 3, historic low-carbon production volumes by

Figure 3: Historic Generation by Low Carbon Resources in TWh

1040-6190/$–see front matter # 2014 Published by Elsevier Inc., http://dx.doi.org/10.1016/j.tej.2014.04.004

The Electricity Journal

[(Figure_4)TD$IG]

Figure 4: Growth in Capacity by Generating Type

generating type are given for the Western Interconnect between 2002 and 2011. he solid black bars at the bottom of the figure represent output from hydroelectric units. In 2002, 67 percent of low-carbon generation with the Western Interconnect came from hydroelectric resources and that fraction has remained relatively constant over the 2002–2011 period. By 2011, and mostly as the result of the growth in VER resources, lowcarbon, non-hydroelectric generation increased by 22.8 percent over 2002 levels.

T

III. Hypothetical Future Generation Mix in the Western Interconnect To determine the future impact of state and provincial low-carbon policies, we examine one possible scenario: the Western Interconnect in 2020. Generation and capacity values for the year 2020 are from the Western Electricity Coordinating Council’s 2020 Study and are denoted October, 2014,

Vol. 27, Issue 8

henceforth as 2020F (WECC, 2011). The capacity additions and generation data reflect information assembled by from integrated resource plans of WECC balancing authorities as state renewable portfolio standard requirements. To understand how capacity within the Western Interconnect is expected to evolve, the actual 2011 capacity data are plotted against forecasted capacity values for 2020F. These are shown in Figure 4. Between 2011 and 2020, an estimated 4 GW of fossil fuel capacity and 35 GW of lowcarbon sources will be added to the system. Total installed coal capacity will increase by nearly 1 GW while there will be a net increase of just over 3 GW of natural gas capacity, although the magnitude of this increase will depend on how California replaces its retired nuclear capacity. The announced permanent retirement of the 2.2 GW San Onofre Nuclear Generating Station (SONGS) will reduce the 2020F nuclear capacity value to less than what is shown

in Figure 4 (Southern California Edison, 2013). It is expected that the capacity that was provided by SONGS will come from natural gas units. The largest growth of capacity is expected in the form of VER technology. Of the almost 39 GW of new capacity, nearly 34 GW will be from variable energy resources. The effect of the region’s changing capacity on the generating mix in 2020 was forecasted by WECC and is shown in Figure 5. rom its 2020 report, the WECC has estimated that low-carbon generation will comprise 50 percent of total

[(Figure_5)TD$IG]

F

NUCLEAR 7.8% COAL ETC. 29.7%

981.4 TWh

RENEWABLES 42.2%

NATURAL GAS 20.3%

Figure 5: Expected Generation Mix by 2020F (percent) by Type

1040-6190/$–see front matter # 2014 Published by Elsevier Inc., http://dx.doi.org/10.1016/j.tej.2014.04.004

99

volumes. This represents an increase of 2.5 percent over 2011 levels; a modest increase given the addition of 35 GW of low-carbon capacity. Production from coal-fired facilities is expected to remain relatively constant; its share in the generation mix increasing very slightly from 29.5 percent to 29.7 percent by 2020. This may be due to the fact that the price of mine mouth coal in Alberta and the Powder River Basin coal in Wyoming will remain competitive even against low natural gas prices (Platts, 2013). Natural gas generation is expected to fall slightly in proportion of total generation, from 23.1 percent to 20.1 percent. The share of generation currently attributed to nuclear technology will likely decrease as the San Onofre generating units have been retired permanently. If natural gas units are used to replace the output from SONGS, the share of nuclear generation will decrease and the share of natural gas generation in the portfolio mix will increase slightly over what is shown in Figure 5.

from fossil fuel-fired generation. In this section, we examine the cost of reducing a ton of CO2 in the Western Interconnect, given that the region is already replete with low-carbon energy sources. The first step is to determine the cost of the resources that will be required to meet the nearly 115 TWh of incremental demand expected by 2020. Because the 2020F scenario created by WECC envisions both increases and decreases across generation types, we simplified the analysis by creating three scenarios and allowing only one category of generation to change in each case we examined. ur base case is one in which we assume that all new demand will be met by coal-fired generation and production volumes from other technologies are held constant at their 2011 levels; we denote this 2020 COAL. In the next scenario, incremental demand is met exclusively by natural-gas-fired generation (2020 NAT GAS). Finally, we examine the case where coal, natural gas, hydro, and nuclear volumes are

O

held constant, and the supply gap is met by both VERs and from geothermal and other renewable generation; we call this 2020 RENEW. A key point to remember in the analysis that follows is that the year 2020 is treated as a single future point of reference, and as such we do not account for costs, generation, and emissions for all the years between 2011 and 2020 but look at 2020 as a one moment in the future. A comparison of the three hypothesized scenarios against the actual production volumes in 2011 is given in Table 1, where bold values denote the differences amongst the scenarios. The range of average levelized costs per MWh associated with each of the low-carbon technologies is from the Energy Information Administration (EIA) and is shown in Table 2 (EIA, 2013). These cost estimates do not include any available production or investment tax credits. With respect to the costs of the coal and natural-gas-fired generation, we assume that the existing underutilized fleet (either

Table 1: Generation in TWh by Type for 2011 and Future Scenarios.

IV. Cost of Low-Carbon Sources The oft-stated reasoning behind low-carbon energy policies is that by forcing the addition of lowcarbon capacity and therefore creating low-carbon generation in the system, these volumes effectively displace carbon emissions that would have come 100

2011

2020 COAL

2020F WECC

2020 NAT GAS

2020 RENEW

Coal Other Fossil Fuels

246.9 8.4

361.8 8.4

290.9 0.4

246.9 8.4

246.9 8.4

Natural Gas Hydro

200.1 265.1

200.1 265.1

199.5 246.8

315.0 265.1

200.1 265.1

Nuclear

72.7

72.7

76.4

72.7

72.7

Geothermal Wind

16.9 36.3

16.9 36.3

35.7 76.7

16.9 36.3

43.4 93.3

20.1 866.4

20.1 981.4

55 981.4

20.1 981.4

51.6 981.4

Generation Type

Solar + Other Renewables Total

1040-6190/$–see front matter # 2014 Published by Elsevier Inc., http://dx.doi.org/10.1016/j.tej.2014.04.004

The Electricity Journal

Table 2: Average Levelized Costs for Low-Carbon Technology in 2011$/MWh. Generation Type

Minimum

Onshore Wind

Average

Maximum

$73.50

$86.60

$99.80

Solar Geothermal

$112.50 $81.40

$144.30 $89.60

$224.40 $100.30

Nuclear Hydro

$104.40 $58.40

$108.40 $90.30

$115.30 $149.20

$62.50

$67.10

$78.20

Natural Gas

coal or natural gas, depending on the scenario) can provide the required energy at a cost equal to the heat rate in MMBtu/MWh multiplied by the fuel cost in $/ MMBtu. These are outlined in Table 3. e find that to meet the supply gap with existing coal resources would cost approximately $2,877.29 million, while using natural gas generation in place of the incremental coal would cost about $3,194.97 million. Variable energy resource and geothermal technology would be the most expensive way to meet the 2020 demand, with the cost ranging from $9,473.1 to $12,690.5 million.

W

V. Carbon Emissions and Estimated Cost of Emissions Reductions Historical carbon emissions in the Western Interconnect were Table 3: Heat Rates and Fuel Cost by Generating Technology.

calculated by determining the amount of generation by fuel type and multiplying those values by the EIA emissions factors shown in Table 4 (EIA, 2000). We then used that same methodology to determine the estimated emissions under the three scenarios outlined in the previous section. These results, along with the emissions estimated from the 2020F scenario are plotted as bars in Figure 6. System emissions, given in tons of CO2 per MWh, are represented by the dotted line. The historical variability in the emissions levels is in large part due to fluctuations in the amount of hydroelectric production, which in turn is the result of precipitation levels. The firm energy used to replace hydroelectric production historically has come fossil-fuelfired units. The difference Table 4: Emissions Factors for Various Energy Market Generating Technologies. Generating Technology

tCO2/MWh

Generating

Heat Rate

Fuel Cost

Coal Gas

0.894 0.526

Technology

(MMBtu/MWh)

($/MMBtu)

Other Fossil

1.542

Hydro Wind

0 0

Coal Natural Gas

October, 2014,

10.8 7.94 Vol. 27, Issue 8

$2.32 $3.50

between the 2020 COAL and 2020 RENEW emissions levels is the estimated impact of adding 115 TWh of low-carbon generation, and is calculated to elicit a 102.7 megaton (MT) reduction. Meeting demand with natural gas rather than coal-fired generation would reduce emissions by 42.2 MT. At present this carbon mitigation figure includes the low-carbon production that would have come from SONGS in California. If the nuclear output is replaced by natural-gas-fired facilities, then the CO2 abatement figure will differ from the estimated 42.2 and 102.7 MT values. sing both the cost information and the aggregated CO2 abatement figures from above, we estimate that the abatement cost associated with mitigating 42.2 MT using natural gas generation is approximately $8/ tCO2. The addition of low-carbon resources to displace coal-fired generation reduces emissions by just over 102.7 MT, and the cost of achieving those emissions has an associated energy cost of between $9.473.1 and $12,690.5 million, resulting in a per unit emissions reduction cost in the range of $64 to $96/tCO2. Note that these lower-bound cost estimates do not include any specified tax incentives such as the production and investment tax credits which are available for some technologies in some Western Interconnect jurisdictions and the addition of these subsidies will raise the cost

U

1040-6190/$–see front matter # 2014 Published by Elsevier Inc., http://dx.doi.org/10.1016/j.tej.2014.04.004

101

500 450 400 350 300 250 200 150 100 50 -

COAL Emissions

0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 -

NG Emissions

OTHER FOSSIL Emissions

tCO2/MWh

MT

[(Figure_6)TD$IG]

tCO2/MWh

Figure 6: Estimated Carbon Emissions Both Historically and Under Varying Scenarios

of meeting demand using renewable resources. A recent study of the U.S. energy system by the National Academy of Sciences found that the production tax credit (PTC) and investment tax credit (ITC) combined yielded a $4–5 billion dollar revenue loss per year for the U.S, while the resulting CO2 reductions were small (National Academies Press, 2013). Overall, when the revenue lost as a result of the PTC/ ITC is divided by the reduction in CO2 emissions, just under $250 in revenues are lost per ton of CO2 reduced . . . the fiscal cost per ton of CO2 reduced is high relative to other, more efficient approaches (p. 70)

VI. Conclusion Electricity generation within the Western Interconnect is significantly low-carbon, due to the massive amounts of storage hydroelectric facilities located in the Pacific Northwest and along the west coast of the United States. At the same time, public policies are driving additional amount of low-carbon capacity into the system, at a significant cost. This new capacity is predominantly variable in nature and must be backstopped by firm energy sources (Sopinka and Pitt, 2013). Policies that attempt to replicate the WECC choice of renewables without utility-scale storage are bound to fail, as the generation mix

within the WECC is unique and cannot be replicated widely. Emissions are expected to increase in the Western Interconnect even with the vast amounts of lowcarbon capacity that are forecasted to come on line by 2020. We estimate that the cost of carbon abatement using low-carbon resources is approximately $64 and $94 per ton, a figure that is significantly higher than the existing market price of carbon. In British Columbia the combustion of fossil fuel is taxed at $30/tCO2 for the near future while California’s cap-and-trade system has CO2 permits valued at about $14/tCO2 for the 2016 vintages (Reuters, 2013). Managing the incremental demand with naturalgas-fired generation instead would also cause emissions to rise over the 2011 values; however the cost of emissions abatement would be about $8/tCO2. re we taming the WECC’s carbon diet? Slowly perhaps, but the main issue seems to be the cost of the diet plans.

A

Appendix. Cost assumptions Tables A1–A5&

W

ithin the Western Interconnect, the relatively high cost of emissions reductions results from the abundant low-carbon energy resources that current reside within the region, the cost of reducing a ton of carbon from this already low-carbon system requires a significant effort at a substantial cost.

102

Table A1: Levelized Costs in 2011 $/MWh for Installation by 2018 (EIA). Generation Type

Minimum

Average

Maximum

$73.50

$86.60

$99.80

Solar Geothermal

$112.50 $81.40

$144.30 $89.60

$224.40 $100.30

Nuclear Hydro

$104.40 $58.40

$108.40 $90.30

$115.30 $149.20

$62.50

$67.10

$78.20

Onshore Wind

Natural Gas

1040-6190/$–see front matter # 2014 Published by Elsevier Inc., http://dx.doi.org/10.1016/j.tej.2014.04.004

The Electricity Journal

Table A2: Generation by Type in TWh. TWh

2011

2020 COAL

2020 NAT GAS

2020 RENEW

COAL

246.9

361.8

246.9

246.9

NG HYDRO

200.1 265.1

200.1 265.1

315.0 265.1

200.1 265.1

NUCLEAR GEOTHERMAL

72.7 16.9

72.7 16.9

72.7 16.9

72.7 43.4

WIND

36.3

36.3

36.3

93.3

OTHER RENW SOLAR

18.6 1.5

18.6 1.5

18.6 1.5

47.8 3.8

OTHER FOSSIL TOTAL

8.4 866.4

8.4 981.4

8.4 981.4

8.4 981.5

Table A3: Emissions in Megatonnes (MT) by Scenario. EMISSIONS (MT)

2011

2020F

2020 COAL

2020 NAT GAS

COAL

220.6

259.9

323.3

220.6

220.6

NG HYDRO

105.3 0.0

105 0.0

105.3 0.0

165.7 0.0

105.3 0.0

NUCLEAR

0.0

0.0

0.0

0.0

0.0

GEOTHERMAL WIND

0.0 0.0

0.0 0.0

0.0 0.0

0.0 0.0

0.0 0.0

OTHER RENW SOLAR

0.0 0.0

0.0 0.0

0.0 0.0

0.0 0.0

0.0 0.0

OTHER FOSSIL

13.0

0.6

13.0

13.0

13.0

338.8

365.5

441.5

399.3

338.8

42.2

102.7

TOTAL Change over coal case

2020 RENEW

References

Table A4: Total Energy Costs by Scenario Given EIA Costs in Millions (shown in Table 2). COSTS

Min

Average

Max

2020 COAL

$2,877.29

$2,877.29

$2,877.29

2020 NG 2020 VAR

$3,194.97 $9,473.06

$3,194.97 $10,890.61

$3,194.97 $12,690.50

Table A5: Costs per tCO2/MWh for Carbon Abatement. Scenario

Min

Emissions reductions costs using NG Emissions reductions costs using RENEW

October, 2014,

Vol. 27, Issue 8

Average

Max

$7.52

$7.52

$7.52

$64.22

$78.02

$95.54

California Environmental Protection Agency, 2011. Overview of ARB Emissions Trading Program. Available from http://www.arb.ca.gov/ newsrel/2011/cap_trade_over view.pdf. CanLii, 2013. Specified Gas Emitters Regulation, Alta Reg 13/2007. Available from http://canlii.ca/t/83nf. CPUC, 2009. 33% Renewables Portfolio Standard Implementation Analysis: Preliminary Results. Available from http://www.cpuc.ca.gov/NR/ rdonlyres/1865C207-FEB5-43CF99EB-A212B78467F6/0/33Per centRPSImplementationAnalysisIn terimReport.pdf.

1040-6190/$–see front matter # 2014 Published by Elsevier Inc., http://dx.doi.org/10.1016/j.tej.2014.04.004

103

EIA, 2000. Guidelines Volume II – U.S. Department of Energy. Available from ftp://www.eia.doe.gov/pub/ oiaf/1605/cdrom/pdf/gg-apptables.pdf. EIA, 2013. Levelized Cost of New Generation Resources in the Annual Energy Outlook 2013. Available from http://www.eia.gov/fore casts/aeo/electricity_genera tion.cfm. National Academies Press, 2013. [64_TD$IF]Effects of U.S. Tax Policy on Greenhouse Gas Emissions. Available from http://www.nap. edu/catalog.php?record_id =18299. Platts, 2013. Western US Fuel Mix Changes. Available from http://

www.platts.com/news-feature/ 2013/naturalgas/coalgas/index. Province of British Columbia, 2010. [159_TD$IF]Clean Energy Act. Available from http://www.bclaws.ca/EPLi braries/bclaws_new/document/ ID/freeside/00_10022_01#section2. Reuters, 2013. UPDATE 2-California Carbon Permits Sell for Record High Price. Available from http:// www.reuters.com/article/2013/ 05/21/california-carbon-idUSL2 N0E21VR20130521. Sopinka, A., Pitt, L., 2013. Variable [16_TD$IF]energy resources: VERy interesting impacts for the Western Interconnect. [162_TD$IF]Electr. J. 26 (5) 1–6, http:// dx.doi.org/10.1016/j.tej.2013. 04.015.

Southern California Edison, 2013. Southern California Edison Announces Plans to Retire San Onofre Nuclear Generating Station. Available from http://www.songs community.com/news2013/ news060713.asp. US Department of Energy, 2013. [16_TD$IF]Database of State Incentives for Renewables and Efficiency. Available from http://www.dsireusa. org/documents/summarymaps/ RPS_map.pdf. WECC, 2011. 10-Year Regional Transmission Plan: 2020 Study Report. Available from http://www.wecc. biz/library/StudyReport/Docu ments/2020%20Study%20Re port.pdf.

The first step is to determine the cost of the resources that will be required to meet the nearly 115 TWh of incremental demand expected by 2020. 104

1040-6190/$–see front matter # 2014 Published by Elsevier Inc., http://dx.doi.org/10.1016/j.tej.2014.04.004

The Electricity Journal