Grid Integration of Renewable
Energy Generation
U.S. Experience
1
Grid Integration of Renewables
• Why is this relevant?
• How much will it cost?
• What are the challenges and proposed
solutions?
2
%: 0
1 )
(,0
%1
)
*
+!" # ! !,
%'
&' (!" # !($
(
&
$" (!" # ! $)
#
# !( 3
), !" 0 4/56 # !(7
%"
*
,
9 3
"
".
"#
'#
'
!(
8
%
&
'
() *
&
+, - 5
"= /
)
/
0
6
*
(.
' <
-
4
'
7
12 ($" # ! $%
">
%?
"!
&
'
() *
,
-
@
$
0
!1 0
%(
"%
04
%
%
(!" # !($
!" # ! $%
;
. ("
' (!" # !($
!
:
&
'
() *
4&
+ , - 56
*
?:
?!
&
+, -*
12 3
44 %
.' /
8
0
4'/
06
$
*†
3
Federal Legislation
• U. S. House of Representatives passed H.R.
2454, The American Clean Energy and
Security Act on June 26, 2009.
• Contains a national Renewable Energy
Standard requiring 6% renewables in 2012 and
increasing to 20% by 2020.
• U. S. Senate has not yet acted.
4
Recent Development of Renewables
• In 2007, wind generation was 30% of new
installed capacity in U.S.
• 145,000 MW of wind generation proposed
over the next 10 years.
• BPA balancing authority will reach 25% wind
penetration in 2009.
5
Costs of Renewable Integration
Date
Study
Capacity
Penetration
(%)
Regulation
Cost
($/MWh)
Load
Following
Cost
($/MWh)
Unit
Commitment Cost
($/MWh)
Gas
Supply
Cost
($/MWh)
Total
Operational
Cost Impact
($/MWh)
May 2003
Xcel-UWIG
3.5
0
0.41
1.44
na
1.85
Sept. 2004
Xcel-MNDOC
15
0.23
na
4.37
na
4.60
June 2006
CA RPS
4
0.45
trace
na
na
0.45
Feb. 2007
20
0-0.69
trace
na
na
0-0.69
June 2003
GE/Pier/CAIA
P
We Energies
4
1.12
0.09
0.69
na
1.90
June 2003
We Energies
29
1.02
0.15
1.75
na
2.92
2005
Pacificorp
20
0
1.6
3.0
na
4.60
April 2006
Xcel-PSCo
10
0.20
na
2.26
1.26
3.72
April 2006
Xcel-PSCo
15
0.20
na
3.32
1.45
4.97
Dec. 2006
MN 20%
31
Jul. 2007
APS
14.8
Source: NREL
4.41
0.37
2.65
1.06
na
4.08
6
Recent Cost Studies
• 2009 Pacificorp study showed costs ranging
from $9.96 to $11.85/MWh.
• 2008 Portland General Electric study showed
at a 26% penetration of system capacity, costs
were $11.75/MWh.
7
One Final Cost Study
• 2008 Avista study showed at 30% penetration
of system capacity, costs at $10.75/MWh.
• However, if feathering is available, costs
decline from $10.75/MWh to $8.75/MWh.
• And, if a shorter market time step is available,
costs decline further from $8.75/MWh to
$3.50/MWh.
8
Contributing Cost Factors
• Regulation costs averaged $33-$60/MWh in
2008 compared to reserve standby costs of
$1.50/MWh. (Regulation=minute-to-minute balancing service. Reserve
standby=10 to 30 minutes.)
• $60 billion of transmission investment needed
for wind to reach 20% market share. (DOE estimate)
9
Challenge #1: Resource is distant from
load
10
Challenge #2: Wind Forecasting
• Wind plant output uncertainty is different than
demand forecasting. Demand forecasts are
typically small and do not change much over
time.
• Example
– Demand forecast error for a 12 hour forecast is
normally 3% and unlikely to be more than 10%.
– Wind forecast error for a 12 hour forecast could be
20% or as much as 100%.
Source: NERC
11
Challenge #3: System Flexibility
• Variable nature requires increased reserves.
• Retirement of thermal resources to meet CO2
goals, needed for regulation services.
• Wind ramps down as loads ramp up.
12
13
California Wind, Solar, Load
Output, July 2003
Source: NERC
14
Solution 1a: Build Transmission
• “Tens of thousands of miles of new
transmission is needed.” (NERC)
• Designing and deploying optimal new
transmission infrastructures
– 765 kV
– High-voltage direct-current (HVDC)
15
Proposed 765 KV system
Source: AWEA
16
Solution 1a continued . . .
• Cost allocation methods are hot topic of
debate.
• Cost causer/beneficiary pays
– August 6, 2009 decision by 7th Circuit Court of
Appeals overturns FERC on PJM proposal for new
500kV line.
• Socialize
– Texas spread costs of new transmission to bring
wind power to population centers.
17
Solution 1b: Planning
• Plan on a wider scale
– Regional/interconnection-wide footprint
– Federal stimulus money ($60 million)
• Combine resource and transmission planning
• Add integration goal to existing goals of
reliability and economic efficiency
– Western Renewable Energy Zone example
18
Solution 2a: Improve Long-term
Forecasting
• Long-term planning could be improved by
identifying best methods for capacity planning
with high-wind scenarios.
• Stochastic, Monte Carlo, and risk-based
techniques could be used to produce numerous
simulations to capture variability of wind.
19
Solution 2b: Improve Short-term
Forecasting
• Integrate into operations
– Use wind and solar forecasting tools in control
room operations and operational timeframes.
– EPRI (Electric Power Research Institute) worked
to develop and test wind forecast algorithms for
short-term (5-min over 3 hours, updated every 5
min) in California for 5 specific wind regions in
the state. More study required.
20
Solution 2c: Other Forecasting
Improvements
•
•
•
•
•
Quality of input data
Locational wind measurements
Real-time power output data
Wind turbines down for maintenance
Provide to operators every 15 minutes
21
Solution 3a: Aggregation
• Consolidate balancing areas aggregates wind
over a larger geographical area reducing the
overall variability on the system.
• Minnesota found by consolidating 4 balancing
areas into one, it would reduce the need for
regulation services by 50%.
22
Aggregation continued . . .
• Improving transmission interconnections,
provides access to other Balancing
Authorities’ reserves.
• Area Control Error diversity interchange
program
– Allows multiple balancing authorities in the West
to share their ACE and net out variations in the
load/resource balance.
(ACE=The instantaneous difference between net actual and scheduled interchange.)
23
24
Solution 3b: Firming Resources
• Build new resources
– Flexible, gas-fired generation
– Hydro resources
• Increase use of pumped hydro
– Western Study showed at 35% renewable
penetration an increased use of existing pumped
hydro, but that current capacity was adequate.
(NREL)
25
Solution 3c: Sub-Hourly
Schedules/Markets
• Hourly markets balance more often with
expensive regulation service versus load
following services.
• In Northwest U.S., determined that 10-minute
markets compared to hourly markets would
reduce integration costs by 40-60%.
• MISO, CAISO, PJM, NYISO operate 5 minute
energy markets.
26
Solution 3d: Operational Protocols
Change BA rules to allow non-spinning reserve
and supplemental operating reserves to be used
to compensate for large wind ramps instead of
regulation services. (Ready to respond within
10 to 30 minutes.)
27
California Wind Ramps
Source: NERC
28
Operational Protocols continued . .
.
• BPA has established protocols in Dispatch
Standing Orders to limit variable generators to
stay within balancing reserve limits.
– Over-generate to schedule, reduced to specified
levels when BPA reaches 90% of its decremental
reserve capability.
– Under-generate to schedule, curtailed down to
actuals when BPA reaches 90% of its incremental
reserve capability.
29
Solution 3e: New tariffs
• Conditional firm service allows use of
transmission capacity outside of predetermined peak-use times.
– Since wind often generates off-peak, this could be
a significant benefit.
30
New Tariffs continued . . .
• On February 16, 2007, imbalance penalties
were changed by FERC in Order 890.
• Imbalance charges escalate as the imbalance
increases and are based on incremental cost.
• Intermittent resources are exempt from the
highest deviation band.
31
Solution 3f: Storage
• Compressed Air Energy Storage
– Charge underground reservoir during off-peak
hours.
– Additional exploration of below-ground storage
locations is needed.
– To capture excess output, storage locations must
be near wind or solar resource.
• Concentrating solar storage advances
32
Solution 3g: Demand Response
• Ercot called on 1200 MW of curtailable load to
restore system frequency during incidence in
Texas in February, 2008.
• Demand Response can minimize economic
impacts of brownouts because businesses can
avoid unplanned outages by voluntarily
shifting load.
• Increase use of automated Demand Response.
33
Summary
• Grid integration of renewables will continue to
be important as increasing amounts of
renewables are built to meet renewable energy
standards.
• While a number of challenges arise with
increased penetration of renewables, a number
of potential solutions are available for
consideration.
34