Page 1
Renewables Grid Integration:Achieving High Penetration
New Mexico Legislative CouncilScience, Technology and Communications
Committee
Abraham Ellis
Sandia National Laboratories
[email protected]
Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company,
for the United States Department of Energy’s National Nuclear Security Administration
under contract DE-AC04-94AL85000.
Page 2
Presentation Outline
• Wind and Solar Generation Potential
• Penetration Levels
• Distribution System Impacts
– Example: Voltage Control
• System-Level Impacts
– Integration Cost
– Variability and Uncertainty
• Technical Benefits to the Grid
• Conclusions
Page 3
Wind Generation PotentialTotal US: 35,000 MW (2009)
AWEA - Year End
2009 Market Report
Page 4
Solar Generation Potential
SEIA - US Solar Industry
Year in Review 2009
Page 5
Renewables In The Big Picture
• By energy produced, 69% of electricity comes from fossil fuels (2009)
• Only 4% comes from “Other Renewables”
– Wind (mostly)
– Biomass
– Geothermal
– Solar
• High penetration levels in
some parts of the system
• Can penetration levels be increased significantly?
Source: EIA
Page 6
Penetration Level “Limits”• There are no absolute technical “limits”
– Cost and risk determine practical limits
• What is the cost and who is willing to pay?
• What is the risk and who is willing to bear?
– There are technical solutions (often costly) to mitigate reliability and safety impacts
• However, there are technical challenges!
• How much renewable generation can be integrated at a reasonable cost and risk?
– It depends on many factors!
Page 7
Penetration Level• Penetration level is a way to quantify amount of
renewable generation deployed on the system
– For distribution system deployment
• Renewable capacity / peak load (most common)
• Renewable capacity / minimum load , station capacity, feeder capacity
– For system-wide deployment
• Renewable energy / total load energy over one year (most common)
• Renewable capacity / peak load or total generation capacity
Page 8
Distribution System Impacts• Renewable deployment in distribution systems is
principally solar PV
• Depending on many factors, high penetration PV in distribution systems could lead to…
– Customer voltages that are too high or hard to control• Voltage raise most likely to encountered first, especially in long
feeders
• Variability can cause excessive operation of voltage control equipment
• Impact depends on where the solar generation is connected
– Degradation on protection performance
– Increased likelihood of electrical “islands”
– Flow in excess of feeder/regulator/transformer capacity
Page 9
Utility
Substation
Vo
lta
ge
Voltage Raise Issue• Voltage along the feeder must be maintained within
service limits (standards)
Power Flow
Near the
Substation
Near the end
of the feeder
Service
Range
Page 10
Vo
lta
ge
Near the
Substation
Voltage Raise Issue• High amounts of PV generation (or other DG) can
cause voltage to raise above service limits
Power Flow
Near the end
of the feeder
Utility
Substation
Service
Range
Page 11
Voltage Control Issue • Long feeder, large DG (wind) at end of feeder
– Simulation shows excessive voltage regulator tap operations
0
10 00
20 00
30 00
40 00
-591
-491
-391
-291
-191
-91
0 20 000 40 000 60 000 80 000
A ctive a n d R e a ctive P o w e r
E lec t ro t ek C on c ep t s ® T O P , T h e O u t p u t P roc e s s o r®
P3
-(
kW
)
(
W)
Q3
-(
kv
ar
)
(
VA
r)
T im e (s )
0.9 7
0.9 8
0.9 9
1.0 0
1.0 1
1.0 2
1.0 3
0.9 6
0.9 8
1.0 0
1.0 2
0 20000 40 000 60000 800 00
Feeder Voltage and Regulator Tap Changes
E lectrotek Con cepts ® TO P , The O utp ut P roces sor®
Ta
p-
(p
u)
(
V)
T ime (s)
Source: EPRI
Page 12
Voltage Control Issue• Short feeder, large DG (PV) close to station
– In this case, voltage regulation is less problematic, even though PV penetration level is very high!
01 02 03 04 05 06 07 08 09
64
66
68
70
72
74
Voltage (
kV
)
Days
01 02 03 04 05 06 07 08 09-5
-4
-3
-2
-1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Real P
ow
er
(kW
)
Interconnection AC Vabc Line-to-Line (kV) vs. AC Real Power (kW) '3/1/2010'
P V (rms)
Load level
without solar
Voltage
Load – Solar output
Page 13
Voltage Control Issue• Possible technical solutions
– Locate large solar generation closer to substation or connect directly to station with dedicated feeder
– Operate voltage regulators in “DG mode”
– Allow PV generators to control voltage or power factor
• More expensive alternatives
– Upgrade feeder circuit with large conductor
– Disconnect or reduce solar output when feeder or customer voltage is too high
– Apply energy storage
Page 14
System-Wide Impacts• System-level concerns
– PV variability and uncertainty could make it more difficult and costly to operate generators
– A large amount of solar generation may trip during a system disturbance (voltage ride-trough issue)
– System may become less stable as a large amount of conventional generation is displaced
Page 15
Integration Cost• System is good at managing variability and uncertainty
associated with the load
• Additional variability & uncertainty from renewable generation can increase operating cost
Time (hour of day)
0 4 8 12 16 20 24
Syste
m L
oad
(M
W)
seconds to minutes
Regulation
tens of minutes to hours
Load
Following
day
Scheduling
Days
Unit
Commitment
Page 16
Integration Cost• Impact of adding variability and uncertainty is called
“integration cost”– Factors: Penetration level, system conditions and additional
variability/uncertainty
Summary of Integration cost for large-scale wind
$0
$1
$2
$3
$4
$5
$6
$7
$8
$9
$10
0% 5% 10% 15% 20% 25% 30% 35%
Wind Penetration (Capacity Basis)
Inte
gra
tion
Co
st (
$/M
Wh
)
Arizona Public Service
Avista Utilities
California RPS
Idaho Power
MN-MISO
Pacificorp-2004
Pacificorp-2007
Puget Sound Energy
We Energies
Xcel-MNDOC
Xcel-PSCo-2006
Xcel-PSCo-2008
Xcel-UWIG
Xcel-PSCo-2008 at
2006 Gas Prices
Source: LBNL
Page 17
Solar Variability• Solar intensity (irradiance) at a given location can vary
a little or a lot depending on sky conditions (clouds)
Note: irradiance is not a
good indication of short-
term variability from
large plants or
geographically-
dispersed plants.
Page 18
Solar Variability• Geographic diversity greatly reduces variability (both
solar and wind)
– Example for residential PV systems in Germany
(Wiemken et al. 2001)
Page 19
Integration Cost• Technical solutions
– Geographical dispersion
– Manage generation resources over wider footprint
– Use state-of-the-art forecasting
– Increase flexibility of conventional generation
– Increase participation in well-functioning electricity market
– Tap load response potential
• Less cost-effective– Curtail variable generation when needed
– Large-scale energy storage
Page 20
Examples of Very High Penetration
Ota City, Japan: 2 MW PV on single feeder (553
homes, 3.85 kW average PV system)
Lanai , Hawaii: 1.2 MW PV system on 4.5 MW island
grid supplied by old diesel generators
Page 21
Examples of Very High Penetration
Ireland: >1 GW wind capacity in 7 GW peak
load island system
Penetration level by energy
approaching 15%
Instantaneous penetration level
approaching 50%
Source: Mark O’Malley
Page 22
Technical Benefits to The Grid• Renewable generation could support grid reliability,
performance, and efficiency… to some extent
– Off-load distribution and transmission lines and transformers; reduce losses (depending on location!)
– Help manage voltage and maintain system stability
– Serve load
• By combining it with modern technologies (Smart Grid concepts, Energy Storage, etc), the technical value to the grid can be increased
Page 23
Reduce Station/Feeder Loading
0
200
400
600
800
1000
1200
1400
1600
1800
0
2
4
6
8
10
12
0:00 6:00 12:00 18:00 0:00 6:00 12:00 18:00 0:00
Irra
dia
nce
(kW
/m2
)
Net
Lo
ad (
MW
)
Standard Time
TuesdayMonday
• In the SW, PV output matches commercial load peak…
Load Load with 20% PV
in the feeder
Solar Irradiance
Page 24
0
200
400
600
800
1000
1200
1400
1600
1800
0
1
2
3
4
5
6
7
8
0:00 6:00 12:00 18:00 0:00 6:00 12:00 18:00 0:00
Irra
dia
nce
(kW
/m2
)
Net
Lo
ad (
MW
)
Standard Time
Sunday Monday
Reduce Station/Feeder Loading
• … timing is less optimal for residential load peak
Load
Load with 20% PV
in the feeder
Solar Irradiance
Page 25
Conclusions• Actual impacts of high penetration renewable
generation depends on many factors– Penetration level
– Characteristics of the system
– Degree of variability and uncertainty of renewable generation
• Most common technical issues:– Distribution system deployment: voltage management
– System-wide deployment: managing variability and uncertainty
• There are no absolute technical limits, but there are technical challenges