Embed Size (px)
Citation preview
NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.
Do we really need storage to operate the renewable grid of the future?
Paul Denholm
Third Annual TrottierSymposium in Sustainable Engineering, Energy and Design
March 8, 2016
3
What is the Electric Grid?
• Features –
Supply must always meet demand (sort of)
Large hourly and season variations in electricity demand
Operating reserves to maintain system stability and reliability
A very large rotating machine spinning at 60 Hz
Source: Mullane & O’Malley
4
Four Independent North America Grids
4
5
And Many Balancing Areas
5
6
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0
10000
20000
30000
40000
50000
60000
0 24 48 72 96 120 144 168
Load
(Fra
ctio
n of
Ann
ual P
eak)
Load
(MW
)
Hour
Summer Maximum Winter Spring Minimum
Hourly electricity demand for three weeks in the ERCOT (Texas) Grid in 2005
Each Balancing Area Must Constantly Balance Variation in Demand
7
Demand Patterns Similar for Much of the U.S.
NYC (Con Ed) Demand in 2005
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
0 24 48 72 96 120 144 168
Dem
and
(MW
)
Hour
Summer Maximum Spring Minimum Winter
8
A Few Locations in North America are Winter Peaking
Ontario - 2015
0
5000
10000
15000
20000
25000
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
Load
(MW
)
Load
(Fra
ctio
n of
Ann
ual P
eak)
Hour & Day
Winter Maximum Fall Minimum
Summer Maximum
9
Traditional System Operation
9
Variations in demand traditionally met by thermal and hydroelectric plants
GE Energy 2010 (WWSIS)
10
How Can We Possibly Make the Grid Work with Lots of VG?
1. Claim it isn't possible, or it is possible but lots of storage is needed
2. Just build lots of renewables and see what happens
3. Perform actual science, math, engineering, and analysis
10
11
Framework – Net Load
Net load- what’s left over when you add wind and solar
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
12 AM 3 AM 6 AM 9 AM 12 PM 3 PM 6 PM 9 PM
Meg
awat
ts
Hour
Load without VG Net Load (Unconstrained)Total Solar Wind
Load, solar, and wind profiles for California on March 29 in a scenario with 11% annual wind and 11% annual solar
assuming no curtailment Denholm et al. 2016
12
The Most Famous Version
Source: CAISO 2013
1313
Impacts of Renewables on the Grid
13
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
50000
1-Apr 8-Apr 15-Apr
MW
Load Wind Net Load
Variation in wind output increases net load ramp ate (Increases in this period from 4,052 MW/hour to 4,560 MW/hour)
Uncertainty in wind output increases uncertainty in net load
to be met with conventional generators
Ramp Range (Increases in this two-week period from 19.3 GW/day to 26.2 GW/day)
Four major impacts of variable generation (VG) on the grid:1) Increased need for operating reserves2) Increase in hourly ramp rate3) Increase in uncertainty of net load4) Increase in ramp range
1414
Quantifying the Impacts of Renewables on the Grid
• How much fuel is actually saved?
• What is the actual economic value?
• Is “backup” needed? Doesn’t this add costs and emissions?
• How much renewables can even be used?
• If the wind blows more at night do you need to store some for use in the day?
• Is storage needed?
14
15
How We Do Grid Integration Science
o Software that simulates a large interconnected grid considering thousands of generators, and transmission
15
PV science machine
Grid integrationscience machine
16
Grid Simulation Requirements
• Very expensive commercial software package that includes existing generation mix, transmission systemo ~10 Vendors/Software packageso Annual licenses routinely exceed $100ko Massive databaseo ~5 hours to >400 days per simulation
• Names includeo “security constrained unit commitment and economic dispatch”o “production cost model”o “chronological dispatch model”
17
Steps to Performing a Realistic Analysis of Renewables on the Grid
1. Acquire detailed solar and wind datao Use lots of wind and solar simulations to consider spatial diversity
o Sub-hourly wind and solar data across large amounts of the U.S. didn’t exist before a few years ago
2. Calculate change in reserve requirementso Use standard industry methods for calculating changes in
regulation reserve requirements based on variability o Consider new methods of addressing
longer term variability and uncertainty
3. Modify data sets to incorporate morerealistic generator performance
0 50 100 150 200 2506
7
8
9
10
11
12
13
14
Generation (MW)
Hea
t Rat
e (m
mbt
u/M
Wh)
Front Range CC Unit 2
Lew et al. 2013
18
Steps to Performing a Realistic Analysis of Renewables on the Grid
4. Hit go and wait…•
18
19
Simulation Outputs
• Did the grid work?• Did you drop load or violate reserve
requirements?
• What was the impact of forecast error or variability on cycling costs and emissions?
• Did you actually use all the renewable generation?o How much curtailment?
• Did a bunch of bad things happen to indicate storage is needed?
19
20
Example Dispatch in Colorado
Denholm et al. 2014
21
Power Flow and Transmission
NELWAY
BRIDGER
COLSTRIP
GOSHEN
BORAH
KINPORT
MIDPOINT
SUMMERLAKE
MALIN
CAPTJACK
MERIDIAN
ALVEY
ALLSTON
KEELER
PEARL
ROUNDMT
OLINDA
JOHN DAY
MARION
LANE
GRIZZLY
BUCKLEY
THEDALLES
OSTRANDER SLATT
McNARY(1169 MW)
BOARDMAN
PAUL /CENTRALIA
RAVER
MONROE
CUSTER
ECHOLAKE
CHIEFJOE GRAND COULEE
SCHULTZ
HANFORD
ASHE
VANTAGE
LOWMON
LITGOOSE
LOWGRANITE
TAFT
GARRISON
DWORSHAK
TOWNSENDBROADVIEW
BELL
DIXONVILLE
BENLOMOND
NAUGHTON
ANACONDA
ATLANTICCITY
ROCKSPRINGS
MONUMENT
MUSTANG
SPENCE
BILLINGS
YELLOWTAIL
CUSTER
GREATFALLS
OVANDO
HOTSPRINGS
CABGORGE
NOXON
LOLO
HELLSCANYON
ROUNDUP
OXBOW
BROWNLEE
BOISE
ENTERPRISE
WALLAWALLA
LAGRANDE
HATWAI
MOSCOW
BENEWAH
RIVERTON
BUFFALO
OREGONBASIN
THERMOPOLIS
WYODAK
CASPER
SHERIDAN
PLATTE
DAVE JOHNSTON
MILES CITYDC TIE
GRANTSPASS
COPCO
LONEPINE
ROSS
ONTARIO
CALDWELLBURNS
WANETA
BOUNDARY
VACA-DIXON
TRACY
TESLA
TABLEMT
LOSBANOS
MOSSLANDING
TERMINAL
MONA
FOUR CORNERS
90SOUTH
CAMPWILLIAMS
BONANZA
HUNTINGTON
SIGURD
IPPGONDER
HARRYALLEN
MACHACEKFTCHURCHILL
AUSTIN
PAVANT
HUNTER
SYLMAR
ADELANTO
GLENCANYON
VALMY(562 MW)
HUMBOLDT
TRACY
VALLEYROAD
CROSS-OVER
AMPS
JEFFERSON
DILLONPETERSONFLATS
DRUM
WEEDJCT
CASCADE
RESTON
OLYMPIA
INGLEDOW
ROCKYREACH
MIDWAY
LIBBY
HUNGRY HORSE
CRAIG
SAN JUAN
HAYDEN
LARAMIERIVER
ARCHER
AULT
RIFLE
MONTROSE
PINTO
CURECANTI PONCHA
SIDNEY
STORY
L E G E N D:
500KV
+-500KVDC
345KV230KV115-161KV
LANGE
WESTHILL
STEGAL
COMANCHEMIDWAY
DANIELS PARKMALTA
SMOKY HILL
PAWNEE(530 MW)
VALMONT
DILLON
BEAVER
WARNERHILL TOP
BORDERTOWN
REDBUTTE
FLAMINGGORGE
TREASURETON
K-FALLSCO GEN
BOYLE
NLEWISTON
DIABLO
GATES
MIDWAY
RINALDI
VINCENT
VICTORVILLE
LUGO
MIRALOMASERRANO
VALLEY
DEVERS
MIGUEL IMPERIALVALLEY
MOJAVE
EL DORADO
MCCULLOUGHMEAD
MARKETPLACE
NAVAJO
MOENKOPI
YAVAPAI
TABLE MESA
PALO VERDE
WESTWING
FLAGSTAFF
PINNACLE PEAK
CHOLLA
NORTH GILA LIBERTY
KYRENE SILVERKING
CORONADO
SOUTH
BICKNELL VAIL
GREENLEE
SPRINGERVILLE
SAGUARO
TORTOLITA
PARKER
PRESCOTT
ROUNDVALLEY
SELIGMANDAVIS
CAMINO
EAGLEMT.
BLYTHE
KNOB
GILATIJUANA
METROPOLIJUAREZ
LOMAS
CIPRES
LAROSITA
SAN LUIS
MEXICALI
INTERGENSEMPRA
MERIDIAN
CHEEKYE
MALASPINA
DUNSMUIR
SAHTLAM
GOLD RIVER
ARNOTT
CLAYBURNROSEDALE
WHALEACH
BRIDGERIVER
NICOLA
KELLYLAKE
100 MILEHOUSE
SODACREEK
BARLOW
WILLISTON
GLENANNAN
TELKWA
SKEENA
PRINCE RUPERT
KITMAT
KEMANO
SAVONA
MICA
REVELSTOKE
ASHTONCREEK
SELKIRK CRANBROOK
INVERMERE
NATAL
PEIGAN N. LETHBRIDGE
LANGDON
JANETSARCEE
REDDEER
BENALTO
BRAZEAU
BICKERDIKE KEEPHILLSELLERSLIE
W. BROOKS
WAREJTN. JENNER
EMPRESS
SHEERNESS
EAST EDMONTONCLOVERBAR
LAMOUREUX
DEERLAND
WHITEFISHLAKE
MARGUERITELAKE
RUTH LAKE
MITSUE
N. CALDER
N. BARRHEAD
LITTLESMOKY
LOUISECREEK
SAGITAWAH
WABAMUNSUNDANCE
MCKINLEY
P.E.G.S.AMBROSIA WEST
MESA
B-A
NORTON
OJO
TAOS
BLACKWATER
ARTESIAAMRAD
CALIENTENEWMAN
ARROYO
DIABLO
LUNA
HIDALGO
LEUPP
EL CENTRO
KENNEDY
PEACE CANYON
PEACE RIVER
BATTLE RIVER
METISKOW
LEUPP
ELCENTRO
KDY 5CX3
PCN500
GMS500
BAT RV79
METIS644
LEUPP
ELCENTRO
KDY 5CX3
PCN500
GMS500
BAT RV79
METIS644
Red indicates high costs and transmission constraints
Source: RMATS Study
22
Example Simulation - Solar PV in the Summer
22
0
10,000
20,000
30,000
40,000
50,000
60,000
Bas 2% 6% 10%PV Penetration and Hour
Gen
erat
ion
(MW
)
PV
GasTurbinePumpedStorageHydro
CombinedCycleImports
Coal
Nuclear
Wind
Geo
Base (no PV) 2% 6% 10%
Simulated Dispatch in California for a Summer Day with PV Penetration from 0-10%
Denholm et al. 2008
23
Reserve Violations can Indicate Loss of Reliability
2424
First Generation of Wind Integration Studies (<2010, up to about - 20% Penetration)
• Focused on basic operability and “integration costs”.
• Integration costs are modest (typically less than $5/MWh). Ongoing questions as to what this even means….
• Spatial diversity smooths aggregated wind output reducing short-term fluctuations to hour time scales
• Almost all the wind can be used (very little curtailment)
• Additional reserves have a modest impact on operational costs
2525
Second Generation• Higher penetration (up to 35% penetration of wind and solar)
• Examines impact of increased system flexibility
• General Conclusions
• Extensive co-operation will be needed
• We may be nearing the flexibility limits of the grid as it exist today
• High solar penetrations are more difficult than high wind
• Curtailment may be the primary limitation for economic deployment of wind and solar
26
Limits to VG Penetration - Curtailment
• There are no technical limits to how much VG can be put on the grid – only economic limits
• You can always find a piece of hardware to solve the problem (including storage….)
• At high penetration, economic limits will likely be due to curtailment
27
WWSIS II High Wind Case (8% solar, 25% wind)
http://www.nrel.gov/electricity/transmission/western_wind.html
Curtailment
Lew et al. 2013
2828
WWSIS II High Wind Solar (25% solar, 8% wind)
Solar is 60% PV and 40% Concentrating Solar Power with 6 hours thermal storage
Curtailment
Lew et al. 2013
29
Sources of Curtailment
Too much supply, not enough demand, when considering:• Ramp constraints• Transmission constraints• Minimum output levels from hydro and thermal
generatorso This also includes the need to operate partially loaded
capacity to maintain system reliability
• Many of these challenges are institutional in addition to technical
3030
Current System FlexibilityLimited by Baseload Capacity
Price/Load Relationship in PJM
Below Cost Bids
0
50
100
150
200
250
0 10000 20000 30000 40000 50000 60000
Load (MW)
Who
lesa
le P
rice
($/M
Wh)
0
5
10
15
20
25
30
35
18000 20000 22000 24000 26000Load (MW)
Who
lesa
le P
rice
($/M
Wh)
Plant operators would rather sell energy at a loss than incur a costly shutdown. Wind and solar may be curtailed under these conditions.
30
31
Impacts of “Must-Run” Generation
0
10000
20000
30000
40000
50000
60000
70000
143
987
713
1517
5321
9126
2930
6735
0539
4343
8148
1952
5756
9561
3365
7170
0974
4778
8583
23Potentially met by solar
Off limits to solar
CAISO estimate of must run generation in current system
This translates into 41% of CAISO load “off limits” to wind and solar
32
Curtailment with Limited Flexibility
Used and curtailed VG in California on March 29 in a scenario with 11% annual wind and 11% annual solar
0%
5%
10%
15%
20%
25%
1-Jan 1-Feb 1-Mar 1-Apr 1-May 1-Jun 1-Jul 1-Aug 1-Sep 1-Oct 1-Nov 1-Dec
Frac
tion
of D
aily
Sol
ar E
nerg
y Cu
rtai
led 11% Annual Solar
15% Annual Solar
Denholm et al. 2016
33
Curtailment Increases RapidlyMarginal curtailment = curtailment of all incremental VG moving from one penetration level to the next Total curtailment = curtailment rate of all PV installed on the system at a certain penetration level
Marginal and average curtailment due to overgeneration under increasing penetration of PV in California with limited grid flexibility
0%
10%
20%
30%
40%
50%
60%
70%
6% 8% 10% 12% 14% 16% 18% 20% 22% 24%
Annu
al S
olar
Cur
tailm
ent
Annual Solar Energy Penetration
Marginal Curtailment
Total Curtailment
Denholm et al. 2016
34
Consequences of RE curtailment
• Technically easy to do (at least on utility-scale renewable energy generation)
• But reduces economic benefits measured by either increased cost of decreased benefit
35
Impact of VG curtailment on LCOECurtailed energy means less can be sold and incremental costs of additional PV rise dramatically
Marginal and average PV LCOE (based on SunShot goals) due to overgeneration under increasing penetration of PV in California with
limited grid flexibility
6
7
8
9
10
11
12
13
14
15
6% 8% 10% 12% 14% 16% 18% 20% 22% 24%
PV C
ost D
ue to
Cur
tailm
ent
(Cen
ts/k
Wh)
Annual Solar Energy Penetration
Marginal Cost
Total Cost
Denholm et al. 2016
36
Avoided Generation and Fuel
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
10% 15% 20%
Avoi
ded
Gen
erat
ion
(MW
h pe
r MW
h of
Po
tent
ial P
V G
ener
atio
n)
Annual PV Penetration
total gas
total coal
total other
total gen
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
10% 12% 14% 16% 18% 20% 22%
Avoi
ded
Fuel
(BTU
per
kW
h of
Pot
entia
l PV
Gen
erat
ion)
Annual PV Penetration
Gas
Coal
Other
total fuel
37
Avoided Generation Costs
-10
0
10
20
30
40
50
60
70
10% 12% 14% 16% 18% 20% 22% 24%Avoi
ded
Prod
uctio
n Co
st o
f PV
per u
nit
of P
oten
tial G
ener
atio
n ($
/MW
h)
Annual PV Penetration
vom
start
fuel
emissions
total
This is the avoided production (variable) costs of PV
3838
Increasing PV Value and Avoiding Curtailment
• While storage provides an “obvious” answer to the problem of supply-demand coincidence, there are a number of options
Denholm et al. 2010
39
Flexibility Supply Curve Concept
39
Denholm et al. 2010
40
Flexibility Supply Curve Concept
40
You probably do this firstCochran et al. 2015
41
Mitigation Options
Type Description
Generator flexibility Ability of conventional generation to vary output over various time scales
Storage flexibility Ability to store energy during periods of low demand and release that energy during periods of high demand
Geographic flexibility Ability to use transmission to share energy and capacity across multiple regions
Load flexibility Ability to vary electricity demand in response to grid conditions
42
Impact of Increased Flexibility Dropping the minimum generation level increases the amount of load served by PV
Net load on March 29 in a scenario with 15% annual solar increasing the grid flexibility
0
5,000
10,000
15,000
20,000
25,000
30,000
35,000
12 AM 3 AM 6 AM 9 AM 12 PM 3 PM 6 PM 9 PM
Net
Loa
d (M
W)
Hour
Limited Flexiblity
Enhanced Flexibility
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
12 AM 3 AM 6 AM 9 AM 12 PM 3 PM 6 PM 9 PM
Sola
r Cur
tailm
ent (
MW
)
Hour
Limited Flexiblity
Enhanced Flexibility
Denholm et al. 2016
43
Increased Flexibility = Increased PenetrationPV penetration of 25% with less than 20% marginal and 5% total curtailment
Marginal and average curtailment due to overgeneration under increasing penetration of PV in California with enhanced grid flexibility
0%
10%
20%
30%
40%
50%
60%
70%
6% 8% 10% 12% 14% 16% 18% 20% 22% 24% 26% 28% 30% 32%
Annu
al S
olar
Cur
tailm
ent
Annual Solar Energy Penetration
Limited Flexibility(Marg.)Limited Flexibility(Total)Enhanced Flexibility(Marg.)Enhanced Flexibility(Total)
Denholm et al. 2016
44
And More Competitive Costs
Marginal and average LCOE due to overgeneration under increasing penetration of PV in California with enhanced grid flexibility
6
7
8
9
10
11
12
13
14
15
6% 8% 10% 12% 14% 16% 18% 20% 22% 24% 26% 28% 30% 32%
PV C
ost D
ue to
Cur
tailm
ent
(Cen
ts/k
Wh)
Annual Solar Energy Penetration
Limited Flexibility(Marg.)Limited Flexibility(Total)Enhanced Flexibility(Marg.)Enhanced Flexibility(Total)
Denholm et al. 2016
45
Increased Flexibility Increases VG Value
0
10
20
30
40
50
60
70
10% 12% 14% 16% 18% 20% 22% 24% 26% 28% 30%
Avoi
ded
Prod
uctio
n Co
st o
f PV
per u
nit
of P
oten
tial G
ener
atio
n ($
/MW
h)
Annual PV Penetration
Base
Increased OperationalFlexibility
0
10
20
30
40
50
60
70
80
90
10% 15% 20% 25% 30% 35%
Tota
l Val
e P
V pe
r uni
t of P
oten
tial
Gen
erat
ion
($/M
Wh)
Annual PV Penetration
Increased OperationalFlexibility
Base
Operational value
Total value
4646
Curtailment as a Function of Flexibility
Average curtailment rate as a function of VG penetration for different flexibilities in ERCOT
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
0% 10% 20% 30% 40% 50% 60% 70% 80%
Fraction of System Electricity from Wind
Frac
tion
of W
ind
Gen
erat
ion
Curta
iled
80% flexiblity (12 GW min load)
90% flexibility (6 GW min load)
100% flexibility (0 min load)
Denholm and Hand 2011
4747
Different RE Mixes Improves Supply/Demand Coincidence
47
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
20% 30% 40% 50% 60% 70% 80%Fraction of System Electricity from Solar and Wind
Frac
tion
of V
G C
urta
iled
0/10020/8030/7040/6060/4080/20
Solar / Wind Mix
Denholm and Hand 2011
4848
How High Can We Go?
• Can renewables themselves largely de-carbonize the electric sector?
• Results from various studies indicate that beyond 35% VG, new sources of flexibility will be needed for economic deployment of renewables
• Need to perform scenario analysis to consider all options and mixes of renewable resources
49
Renewable Electricity Futures
Technology cost & performanceResource availabilityDemand projectionDemand-side technologiesGrid operationsTransmission costs
Black & Veatch
Technology Teams
Flexible Resources
End-Use Electricity
System Operations
Transmission
ABB inc.GridView
(hourly production cost)
rooftop PVpenetration
2050 mixof generators
does it balancehourly?
ImplicationsGHG Emissions
Water UseLand Use
Direct Costs
Capacity & Generation 2010-2050
Mai et al. 2012
50
RE Resource Supply from 30% - 90% Electricity
Additional variability challenges system operations, but can be addressed through increased use of supply- and demand-side flexibility options and new transmission.
Mai et al. 2012
51
A Transformation of the U.S. Electricity System
RE generation from technologies that are commercially available today, in combination with a more flexible electric system, is more than adequate to supply 80% of total U.S. electricity generation in 2050—while meeting electricity demand on an hourly basis in every region of the country.
Mai et al. 2012
52
How? Build New Transmission…
52
-1
-0.5
0
0.5
1
0 2190 4380 6570 8760Tr
ansm
issi
on u
sage
/ ca
paci
ty
Hours
CO to KS MT to ND
Mai et al. 2012
53
Deploy Dispatchable Renewables
Source: Denholm et al (2012)
54
0
50
100
150
200
250
Apr-27 00:00
Apr-27 12:00
Apr-28 00:00
Apr-28 12:00
Apr-29 00:00
Apr-29 12:00
Apr-30 00:00
Apr-30 12:00
Pow
er (G
W)
Coal
0102030405060708090
Apr-27 00:00
Apr-27 12:00
Apr-28 00:00
Apr-28 12:00
Apr-29 00:00
Apr-29 12:00
Apr-30 00:00
Apr-30 12:00
Pow
er (G
W)
Gas CC
0
1
2
3
4
5
6
Apr-27 00:00
Apr-27 12:00
Apr-28 00:00
Apr-28 12:00
Apr-29 00:00
Apr-29 12:00
Apr-30 00:00
Apr-30 12:00
Pow
er (G
W)
Gas CT
05
1015202530354045
Apr-27 00:00
Apr-27 12:00
Apr-28 00:00
Apr-28 12:00
Apr-29 00:00
Apr-29 12:00
Apr-30 00:00
Apr-30 12:00
Pow
er (G
W)
Coal
02468
1012141618
Apr-27 00:00
Apr-27 12:00
Apr-28 00:00
Apr-28 12:00
Apr-29 00:00
Apr-29 12:00
Apr-30 00:00
Apr-30 12:00
Pow
er (G
W)
Gas CC
0
1
2
3
4
5
6
Apr-27 00:00
Apr-27 12:00
Apr-28 00:00
Apr-28 12:00
Apr-29 00:00
Apr-29 12:00
Apr-30 00:00
Apr-30 12:00
Pow
er (G
W)
Gas CTBa
selin
e sc
enar
io
80%
RE-
ITI s
cena
rio
Design for cycling
Mai et al. 2012
55
Harness Responsive Demand (smart grid?)
55
0
20
40
60
80
100
120
Baseline
30% R
E
40% R
E
50% R
E
60% R
E
70% R
E
80% R
E
90% R
E
Con
tribu
tion
to O
pera
ting
Res
erve
requ
irem
ent (
GW
)
Generators Storage Interruptible Load
Mai et al. 2012
56
0
2
4
6
8
10
12
14
16
18
Curt
ailm
ent (
TWh)
EastWECCERCOT
Accept inevitable curtailment in the spring.
• 8-10% of wind, solar, hydropower curtailed in 2050 under 80% RE scenarios
Mai et al. 2012
57
• RE Futures develops about 80 GW of new storage, in addition to the 20 GW of pumped storage existing in the U.S.
• Lower cost storage would be more competitive
• We don’t understand the opportunities of a world with low cost energy storage
..And yes, develop new storage
Mai et al. 2012
5858
Energy Storage Can Reduce VG Curtailment
580%
5%
10%
15%
20%
25%
30%
35%
40%
20% 30% 40% 50% 60% 70% 80%Fraction of System Electricity from Wind&Solar
Frac
tion
of V
G C
urta
iled
No Storage4 hours8 hours12 hours24 hours
Denholm and Hand 2011
59
Cost Optimal Storage Deployment?
Now - Conventional Pumped Hydro: ~ 20 GW
Future 80-140 GW depending on REF scenario
Mai et al. 2012
60
But we are just not there yet….
Value of bulk storage (sited on the transmission network) in today’s system
800
1000
1200
1400
1600
1800
2000
2200
2400
Base Device
(300 MW)
Increased Efficiency
No PHS Double Natural
Gas Price
2X NG and No PHS
Brea
keve
n ca
pita
l cos
t fo
r an
8-ho
ur
stor
age
devi
ce ($
/kW
)
High
Mid
Low
Divide by 8 to get $/kWh installed cost
$200/kWh installed cost
Denholm et al. 2013
61
0
10,000
20,000
30,000
40,000
50,000
60,000
70,000
12 AM 4 AM 8 AM 12 PM 4 PM 8 PM 12 AM 4 AM 8 AM 12 PM 4 PM 8 PM
Meg
awat
ts
Hour
Load without PV
14% PV
Storage/PV Synergy
Less time between peak and off-peak period
0
1
2
3
4
5
6
0% 5% 10% 15% 20% 25% 30%
Hour
s of S
tora
ge N
eede
d to
Rel
iabl
y Re
duce
Net
Dem
and
Annual PV Penetration (Potential)
2650 MW
1325 MWNarrower peak means less storage needed
Denholm et al. 2016
62
A Tipping Point for Storage?
The adoption of storage may be primarily driven by its ability to offset conventional capacity aided by the increase in value associated with VG deployment
9001,0001,1001,2001,3001,4001,5001,6001,7001,8001,9002,0002,1002,2002,3002,400
225250275300325350375400425450475500525550575600
8 10 12 14 16 18 20
4-Ho
ur B
atte
ry C
ost (
$/kW
)
4-Ho
ur B
atte
ry C
ost (
$/kW
h)
Battery Lifetime (Years)
20% PV $1500/kW CT
15% PV $1500/kW CT
20% PV $900/kW CT
15% PV $900/kW CT
Denholm et al. 2015
63
Conclusions: What I Think I Know About the Grid and Storage (<35%)
• Numerous studies have demonstrated the feasibility of 35% RE
• New methods of grid operation are required• Significantly increased cooperation across large
areas• More ramping of thermal units• Storage is probably not the least-cost option
for increased integration
64
What I Think I Know About the Grid and Storage (>35%)
• Less explored territory• Curtailment rates increase• Any and all sources of grid flexibility will be needed
• Demand Response• Long distance transmission?
• Value of storage increases• At some point low cost sources of flexibility will be
exhausted and storage will be an increasingly attractive means of utilizing wind and solar • Non obvious sources of storage may be cost-competitive
(Thermal storage in buildings, CSP with TES)• Storage adoption may be driven by its ability to
replace conventional capacity
66
Selected References• Brinkman, G., J. Jorgenson, A. Ehlen, J. Caldwell (2015). “The California 2030 Low Carbon Grid Study,” NREL TP-
6A20-64884• CASIO (2013). What the Duck Curve Tells us about Managing a Green Grid. • Denholm, P., K. Clark, and M. O’Connell. (2016). On the Path to SunShot: Emerging Issues and Challenges in
Integrating High Levels of Solar into the Electrical Generation and Transmission System. Golden, CO: National Renewable Energy Laboratory. NREL/TP-6A20-65800.
• Denholm, P., M. O'Connell, G. Brinkman, J. Jorgenson (2015). Overgeneration from Solar Energy in California: A Field Guide to the Duck Chart. NREL/TP-6A20-65023
• Denholm, P., V. Diakov, R. Margolis (2015) The Relative Economic Merits of Storage and Combustion Turbines for Meeting Peak Capacity Requirements under Increased Penetration of Solar Photovoltaics. NREL/TP-6A20-64841
• Denholm, P., J. Jorgenson, M. Hummon, T. Jenkin, B. Kirby, O. Ma, M. O'Malley, and D. Palchak. (2013) Value of Energy Storage for Grid Applications. NREL Report No. TP-6A20-58465.
• Denholm, P., and M. Hand. (2011) “Grid Flexibility and Storage Required to Achieve Very High Penetration of Variable Renewable Electricity” Energy Policy 39, 1817-1830.
• Denholm, P. and M. Mehos. (2011) “Enabling Greater Penetration of Solar Power via the Use of Thermal Energy Storage” NREL Report No. TP-6A20-52978.
• Denholm, P., E. Ela, B. Kirby, and M. Milligan. (2010) “The Role of Energy Storage with Renewable Electricity Generation” NREL/TP-6A2-47187.
• Denholm, P., R. M. Margolis and J. Milford. (2008) “Production Cost Modeling for High Levels of PhotovoltaicsPenetration” NREL/TP-581-42305.
• Denholm, P., and R. M. Margolis. (2007) “Evaluating the Limits of Solar Photovoltaics (PV) in Electric Power Systems Utilizing Energy Storage and Other Enabling Technologies”. Energy Policy. 35, 4424-4433.
• Mai; R. Wiser; D. Sandor; G. Brinkman; G. Heath; P. Denholm; D.J. Hostick; N. Darghouth; A. Schlosser; K. Strzepek 2012 Renewable Electricity Futures Study. NREL/TP-6A20-52409
• Lew, D.; Brinkman, G.; Ibanez, E.; Florita, A.; Heaney, M.; Hodge, B. M.; Hummon, M.; Stark, G.; King, J.; Lefton, S. A.; Kumar, N.; Agan, D.; Jordan, G.; Venkataraman, S. (2013). “Western Wind and Solar Integration Study Phase 2.” NREL Report No. TP-5500-55588.
Contact:Paul [email protected]
