1

The Role of Energy Storage in the Modern Low-Carbon Grid

Paul DenholmNational Renewable Energy Laboratory

Strategic Energy Analysis and Applications CenterJune 12, 2008

2

Today’s Discussion: The Role of Storage in the Past, Present & Future

Grid

Fraction of Energy from Wind and Solar

0% 1% 20% 100%

Studied storage valuable but not necessary

PresentPast Near Future?

Less studiedstorage increasingly valuable and

at some point necessary

Low Carbon Future

Topics to Discuss:• Grid applications of storage & valuation• Potential changes in storage valuation created by renewable energy• Storage technologies

3

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 24 48 72 96 120 144 168

Hour & Day

Lo

ad (

Fra

ctio

n o

f A

nn

ual

Pea

k)

0

500

1,000

1,500

2,000

2,5003,000

3,500

4,000

4,500

5,000

5,500

6,000

Lo

ad

(M

W)

Winter Spring Minimum

Summer Maximum

3 Weeks in Colorado (Xcel) 2003

The Challenge

Challenge is to meet this demand, increasingly with technologies that don’t behave the way we would like them

to behave

4

Grid of the Past

Fraction of Energy from Wind and Solar

0% 1% 20% 100%

Challenges of meeting variations in demand provided early motivations for storage (<1980)

- Limited flexibility steam plants (no CTs)- Projected nuclear builds (100s of GW)- Fuels Act (no gas)

PresentPast Near Future?

5

Early Storage Build Out

Conventional Pumped Hydro: ~ 20 GW

CAES: 1 Plant (110 MW)

Others (<100 MW total): A few batteries, SMES, mostly for local power quality issues

6

Recent History (1980s-2000)

• Limited Activity– PHS sites mostly gone– Cheap natural gas– Limited nuclear build out – Exotic technologies remain

costly

• Cheaper and/or easier to meet variation in load and capacity requirements with conventional generation resources

0

1

2

3

4

5

6

1970 1975 1980 1985 1990 1995 2000

Ele

ctri

c S

ecto

r N

atu

ral G

as P

rice

($/

tCF

)

Fuel Use Act

7

Grid of the Present

Fraction of Energy from Wind and Solar

0% 1% 20% 100%0

1

2

3

4

5

6

7

8

1998 2000 2002 2004 2006E

lect

ric

Sec

tor

Nat

ura

l Gas

Pri

ce (

$/tC

F)

Renewed Interest in Storage- Emergence of energy and ancillary service markets- High natural gas prices

Present

8

Potential Values of Energy Storage

• “Capturable” in existing markets– Energy Arbitrage– Capacity– Frequency regulation– Spinning and non-spinning reserve

• Less-Capturable / Regulated Services – Black start– T&D deferral– All of the above in regulated service territories

9

0

10

20

30

40

50

60

70

80

0 6 12 18 24

Hour

Wh

ole

sale

Pri

ce (

$/M

Wh

)

Hold

Sell

Buy + Double Spin

Spin

Valuation of Energy Storage

8 MWh Purchased + 6 Hours of

Double Spin = 53.3

6 MWh Sold = $251.8

Daily Net Revenue (no O&M) = $369.2

8 Hours of Spin = $64

Revenues have increased 76% by combining services

1 MW, 6 MWh Device, 75%

efficiency

10

Storage in Today’s Grid

• Can provide a variety of services• Proper valuation must consider multiple

services (energy + capacity)• Still economic, technical, regulatory &

institutional impediments to larger deployment• As with any potential solution it needs to meet

investment criteria and beat the competition• Storage is useful but not a required

component of the grid

11

Grid of the Near Future

Fraction of Energy from Wind and Solar

0% 1% 20% 100%

• Growth in renewables- dominated by wind

• Perceived need/opportunities for “firming”• Concerns about adequate T&D

Near Future?

12

Will Opportunities for Storage Increase with RE Deployment?

• Will there be increased opportunities for storage? – Increased variation in price – Increased need for capacity & ancillary

services– Greater need for T&D

13

Characteristics of Wind

• Variable and uncertain output• Limited dispatchability (can curtail when

needed) • Significant ramp rates (but generally over hours

not minutes)• Wind output is not usually coincident with peak

load– Wind is an energy resource, not a capacity resource– Somewhat anti-correlated with load

14

Possible RE Impacts that Increase Opportunities for Storage

• Increased need for ancillary services due to ramping and variability

• Increased variation in prices due to anti-correlation with load

• Increased T&D requirements due to low capacity factors

• How to evaluate???

15

Utility Wind Integration Studies

• Simulate system with and without wind– Use expensive commercial software

that includes existing generation mix, transmission system

– Get lots of wind simulations from a commercial vendor

– Spend several 100ks to several M$

• Evaluate costs of forecast errors, additional reserves etc.

Ponnequin PeetzPonnequin Peetz

16

Load Impacts

Greater Ramp Rates

Limited Capacity

Reduced Fuel Use

GE Study of ERCOT

17

Forecast Error

Forecast error results in over- or under-commitment & increased costs

18

Example Results (Arizona Public Service)

19

Comparison of Cost-BasedU.S. Operational Impact Studies

* 3-year average; total is non-market cost** highest integration cost of 3 years; 30.7% capacity penetration corresponding to 25% energy penetration; 24.7% capacity penetration at 20% energy penetration*** found $4.37/MWh reduction in UC cost when wind forecasting is used in UC decision

Date Study Wind Capacity Penetra-tion (%)

Regula-tion Cost ($/MWh)

Load Following Cost ($/MWh)

Unit Commit-ment Cost ($/MWh)

GasSupplyCost($/MWh)

Tot Oper. Cost Impact($/MWh)

May ‘03 Xcel-UWIG 3.5 0 0.41 1.44 na 1.85

Sep ‘04 Xcel-MNDOC 15 0.23 na 4.37 na 4.60

June ‘06 CA RPS 4 0.45* trace na na 0.45

Feb ‘07 GE/Pier/CAIAP 20 0-0.69 trace na*** na 0-0.69***

June ‘03 We Energies 4 1.12 0.09 0.69 na 1.90

June ‘03 We Energies 29 1.02 0.15 1.75 na 2.92

2005 PacifiCorp 20 0 1.6 3.0 na 4.60

April ‘06 Xcel-PSCo 10 0.20 na 2.26 1.26 3.72

April ‘06 Xcel-PSCo 15 0.20 na 3.32 1.45 4.97

Dec ‘06 MN 20% 31** 4.41**

Jul ‘07 APS 14.8 0.37 2.65 1.06 na 4.08

20

Load Impacts & Price Volatility

21

Price/Load Relationship

0

50

100

150

200

250

0 10000 20000 30000 40000 50000 60000

Load (MW)

Wh

ole

sale

Pri

ce (

$/M

Wh

)

0

5

10

15

20

25

30

35

18000 20000 22000 24000 26000Load (MW)

Wh

ole

sale

Pri

ce (

$/M

Wh

)

More wind results in lower off-peak prices, and a

greater overall difference in on/off-peak prices…

..And sometimes $0 cost energy (curtailed

wind)

22

Important Takeaways: How Does Wind Increase Opportunities for Grid Storage?

• Dedicated RE storage generally not justified– Integration cost penalties are not so high that they

themselves would pay for storage…Wind “adder” less than $5/MWh in cases up to 20% capacity (even 20% energy in some cases)

– existing flexible generation limits the impact– Spatial diversity smooths aggregated wind output

reducing short-term fluctuations to hour time scales

• What wind does do is add to the existing opportunities for energy storage

23

Increased Opportunities for Storage

• Small increase in A/S market due to wind

• Increased arbitrage opportunities– Larger swings in prices– More hours of low cost electricity, including

zero cost energy from curtailment

• Increased opportunities for T&D deferment/alternative

24

Storage in the Near Future Grid

• Applications still dominated by whole grid applications

• RE begins to have an impact– Increase in A/S requirements– Increase in short-term price volatility– More hours of low-cost energy

• Storage increasingly attractive but not yet a required component of the grid

25

Grid of the Future

Fraction of Energy from Wind and Solar

0% 1% 20% 100%

• Carbon constraints & high priced fossil fuels• Massive growth of all RE sources including wind, solar

PV, CSP, geothermal- RE as the dominant electricity source

• Competition from non RE baseload technologies including IGCC/CCS & Nuclear

26

What Happens When Extremely Large Amounts of RE is Deployed?

• More interaction between RE sources and baseload plants– More curtailment, perhaps significant– Need to evaluate whole new grid architectures– Enabling technologies become necessary

• Analysis can be challenging– Good wind data over large regions is limited

• An analysis of PV can provide and example of the potential need for storage or other enabling technologies…

27

PV Coincidence With Load - Summer

0

10

20

30

40

50

0 12 24 36 48Hour

Lo

ad

(G

W)

Normal Load Net Load with PV PV Output

16 GW simulated PV system providing 11% of system’s energy

28

PV Coincidence With Load - Spring

0

5

10

15

20

25

30

0 12 24 36 48Hour

Lo

ad

(G

W)

Normal Load Net Load with PV PV Output

2000 Normal Min Load

Potentially curtailed PV

29

Why is RE Curtailed?

0

0.2

0.4

0.6

0.8

1

1 8760Hours at Load

Fra

cti

on

of

Pe

ak

Lo

ad

Peak

Intermediate

Base

Hypothetical MinimumGenerator Loading

50% or more of a system’s energy is generated here

Price approaches

$0

30

RE Curtailment

0

0.2

0.4

0.6

0.8

1

1 8760Hours at Load

Fra

cti

on

of

Pe

ak

Lo

ad

SolarWind

IGCC, Nuclear

31

PV Curtailment

-0.25

0

0.25

0.5

0.75

1

0 8760Hours at Load

Ne

t L

oa

d R

ela

tiv

e t

o P

ea

k L

oa

d

0%

10%

20%

Fraction of Energy from PV

32

Wind Curtailment

33

PV Curtailment and Cost Impacts

0%

5%

10%

15%

20%

25%

0% 5% 10% 15% 20% 25%

% System Energy from PV

PV

Sy

ste

m C

ap

ac

ity

Fa

cto

r (%

)

1.0

1.5

2.0

2.5

3.0

PV

Sy

ste

m R

ela

tiv

e C

os

t

Marginal CF Marginal Cost

34

Supply Demand Coincidence

• Provides the ultimate limit to RE penetration• Generation does not occur when desired

– Too little supply during periods of high demand– Too much supply during periods of low demand

• Storage can effectively “shift” supply to demand– But so can other enabling technologies…

35

Options for Increased Use of RE

0.0

0.1

0.2

0.3

0.4

0.5

0 4 8 12 16 20Hour

Lo

ad (

Rel

ativ

e to

An

nu

al

Pea

k D

eman

d)

Usable PV

Surplus PV

Load Met withConventionalGenerationNormal Load

Minimum Load

0.4

0.1

0.2

0.3

Load Shifting

Energy Storage

Increased Flexibility

AdditionalLoad

36

Flexibility Supply Curve

Improved Pricing

Demand Response

Gas Generation

Coal Cycling

Existing Hydro

Pumped Hydro

Gas Storage

Increasing RE Penetration

LowCost

HighCost

The relative order of these is conceptual only.

Ice

Heat

Heating

Transportation

Demand Side

Flexibility

Supply Side

FlexibilityCSP

Markets

Thermal

Storage

New

Loads

Electricity

Storage

Electricity

Storage

Existing

Storage

RE

Curtailment

Thermal

Storage

Flexible

Generation

37

Supply Side Storage Technologies

?

CSP Thermal

38

Three Classes of Electricity Storage Devices

• Short-term storage for ancillary services• Distributed storage for local applications• Bulk energy storage

39

Short-term Devices (30 min or less)

• Devices that can provide frequency regulation and spinning reserve

• Flywheels, batteries & capacitors

Beacon Flywheel for Frequency

Regulation

40

Distributed Storage (<10 MW or so)

• Provide both capacity and energy services• Local T&D appears to be a primary

application• Primarily batteries

– Flow batteries– NaS– Other battery chemistries?

41

Bulk Energy Storage

(Courtesy of TVA)

Compressed Air Energy Storage (CAES)

Limited growth opportunities for PHS

42

What is CAES?

• Hybrid storage/generation system using gas turbine technology

• Requires natural gas– 4000-4400 BTU fuel per kWh out

• 0.7-0.75 kWh electricity in per 1 kWh out• Use of natural gas makes round trip efficiency difficult

to define, but 75% is a reasonable value• Requires underground storage

– Salt domes, aquifers, hard rockAlternative fuels possible including biofuels and “adiabatic” CAES using thermal energy storage

• Alternative configurations pursed by EPRI and others

43

Proposed CAES Projects

Norton, OH 2700 MW (merchant plant – arbitrage and A/S)

Iowa Stored Energy Park~300 MW, $800-$900/kW

44

Possible CAES Geology

Robert Haug (ISEP): “Suitable aquifer storage sites are

difficult to find…………..but not impossible”

Succar & Williams

2008

45

Concentrating Solar & Thermal Storage

High-efficiency storage can be

added at relatively low cost

Huge resource- several times the nation’s electric

demand

46

Parabolic Trough Power Plant w/ 6hr Thermal Storage

HX

HotTank

ColdTank

47

End Use Storage

• Thermal Storage– Hot or Cold– Dispatchable– Low Cost

• Electric Transportation and PHEVs

48

PHEVs as Dispatchable Load

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 24 48 72 96 120 144 168

Hour

Lo

ad (

% o

f A

nn

ual

Pea

k L

oad

)

Summer Maximum

Winter

Spring Minimum

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 2000 4000 6000 8000

Hours at Load

Lo

ad (

% o

f P

eak)

50%

30%

10%

No PHEVs

49

Storage in the Future Grid

• Inherent conflict between competing sources of low-carbon energy– Wind and solar need flexible generation– IGCC and nuclear reduces system flexibility

• Flexibility will have to come from a variety of sources on both the supply and the demand side

• Storage is one of a large number of enabling technologies that can increase the utilization of RE generation

• Still unclear at what point storage becomes the best option, but its probably less than 100%!

50

Questions?