fast automated demand response to enable the … automated demand response to enable the integration...
TRANSCRIPT
Fast Automated Demand Responseto Enable the Integration of
Renewable ResourcesCaFFEET
October 26, 2011
Lawrence Berkeley National Laboratory
Demand Response Research Center
David Watson
Sponsored by California Energy Commission, US Dept of Energy
1
2
1. Background & Goals
2. Automated DR: Proven Results & Future Challenges
3. Grid Operator Programs
4. AutoDR Resource Estimates
5. Conclusions
Overview
3
Background
• California Renewable Portfolio Standards increasing to 33% by 2020
• Wind & Solar resources are variable and intermittent
• Grid requires over 4 GW of ancillary service to maintain grid stability
• Automated Demand Response pilots demonstrated ancillary services
Goal of Scoping Study – Develop preliminary estimate and feasibility of
capacity of AutoDR in California as a resource for renewables
integration.
Background & Goals
7/9/2008
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
12:0
0 A
M
1:0
0 A
M
2:0
0 A
M
3:0
0 A
M
4:0
0 A
M
5:0
0 A
M
6:0
0 A
M
7:0
0 A
M
8:0
0 A
M
9:0
0 A
M
10:0
0 A
M
11:0
0 A
M
12:0
0 P
M
1:0
0 P
M
2:0
0 P
M
3:0
0 P
M
4:0
0 P
M
5:0
0 P
M
6:0
0 P
M
7:0
0 P
M
8:0
0 P
M
9:0
0 P
M
10:0
0 P
M
11:0
0 P
M
12:0
0 A
M
Wh
ole
Bu
ild
ing
Po
wer (
kW
)
CPP MA Baseline
4$/kWh
Price Signal
OpenADRData
Model InternetComms
PricingData Models
Physical Communications
Control Strategies
Client ServerClient
ServerAuto-DR
Building
Action
Building
Action
End-Use Controls
Signaling- continuous, 2-way, secure messaging system for dynamic prices, emergency and reliability signals. One-way applications are under development
Client-server architecture - uses open interfaces to allow interoperability with publish and subscribe systems
Current system - uses internet available at most large facilities or broadcasting points.
Hardware retrofit or embedded software - many clients fully implemented with existing XML software
Open Automated Demand Response (OpenADR)
Efficiency and DR Continuum
OpenADR
Daily
Energy
Efficiency
Daily
Time-Of-
Use
Energy
(TOU Rates)
Dynamic
Peak Load
Management
(Dynamic
Rates)
Scheduled
Demand
Response
Real-Time
Demand
Response
(Ancillary
Reserves)
Regulation
(Ancillary
Services)
Service Levels
Optimized
Time of Use
Optimized
Service Levels
Temporarily Reduced
Increasing Levels of Granularity of Control
Increasing Speed of Telemetry
Increasing Interactions with Grid (OpenADR & Smart Grid)
Resources Sold Back to Grid
OpenADR
6
1. Traditional A/S methods:
• Thermal generation plants use fossil fuels, & high cost.
• Grid-scale storage is environmental friendly, but cost is high
(US$1500 - $4000 / kW)
2. Potential Advantages of AutoDR systems:
• Lower first cost (US$75 - $300 / kW) for current programs
• Lower operating costs
• Lower carbon footprint
• Leverages multi-purpose systems for energy efficiency
Traditional Ancillary Services & OpenADR
7
Fast
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
0:0
0
1:0
0
2:0
0
3:0
0
4:0
0
5:0
0
6:0
0
7:0
0
8:0
0
9:0
0
10:0
0
11:0
0
12:0
0
13:0
0
14:0
0
15:0
0
16:0
0
17:0
0
18:0
0
19:0
0
20:0
0
21:0
0
22:0
0
23:0
0
Po
wer
[MW
]
Seattle Municipal Tower Mckinstry Target - T1284 3/10 BL OAT Reg BL
Test Period
Early & Cold
AutoDR Research & Commercialized Deployments have proven:
• Multi-year performance during Peak Periods (~100 MW existing capacity)
• Fast response: < 5 min. (Participating Load Pilot)
• Cold winter mornings: 7:00 AM – 10 AM (Pacific Northwest Pilots)
OpenADR: Proven Results
0
100
200
300
400
500
600
0:0
5
0:4
5
1:2
5
2:0
5
2:4
5
3:2
5
4:0
5
4:4
5
5:2
5
6:0
5
6:4
5
7:2
5
8:0
5
8:4
5
9:2
5
10:0
5
10:4
5
11:2
5
12:0
5
12:4
5
13:2
5
14:0
5
14:4
5
15:2
5
16:0
5
16:4
5
17:2
5
18:0
5
18:4
5
19:2
5
20:0
5
20:4
5
21:2
5
22:0
5
22:4
5
23:2
5
Time of Day
De
ma
nd
(k
W)
Actual 5 min. Data Hourly Forecast with reduction Forecasted Data
8
1. Economic incentive structure unclear
2. Resource varies based on time, temperature
• Few data about off-peak DR
3. A/S requirements may increase cost of AutoDR
• Monitoring, verification, telemetry
• Dedicated network connections
4. Portfolio management
• Load shaping
• Geographic issues (Sub-LAP)
Challenges for AutoDR as Ancillary Services
AutoDR
for Existing
CAISO products
Service Response Time Duration
Reg Up
Start <1 min.
Reach bid <10
min
15 - 60
min
Reg
Down
Start <1 min.
Reach bid <10
min
15 - 60
min
Non-Spin < 10 min 30 min
Future Spin
~ Instant Start
Full Output <10
min
30 min
AutoDR
AutoDR
AutoDR
CAISO Programs and AutoDR
Examples of Needs for
AutoDR
- Shift Load to Night
- Daily Peak Management
- Ramp Smoothing
AutoDR Terminology & Link to Batteries
Shed Energy reduced during specified period (e.g., reduced lighting) - net energy reduction
Shift Energy moved to different time period- minimal change in consumption
Charge Energy use to store load (e.g., pre-cool or charge batteries)
Discharge Energy storage supplies local loads (thermal or electrical) or provides power to grid
11
1. Determine total electric load profiles for commercial and
industrial sectors & key end-uses
1. Determine % of loads that could be controlled using current
site infrastructure, and AutoDR technology
1. Determine % shed for each controllable load
Estimated Shed = (Load) x (%Controllable) x (%Shed)
Methodology for Resource Analysis
12
1. Statewide load data, field test results & engineering
judgment used to estimate potential
2. Multipliers selected based on existing or planned CAISO
products for renewables integration
1. Commercial building type and end-uses evaluated
2. Industrial load shapes evaluated based on case studies and
scoping studies
Duration Ramp time
2 hour 15 min.
20 min. 5 min.
Methodology Assumptions
End Use Type Modulate On/Off
Max.
Response
Time
HVAC
Chiller
SystemsSetpoint Adj. 15 min.
Package Unit Setpoint Adj.Disable
Compressors5 min.
Lighting
Dimmable Reduce Level 5 min.
On/Off Bi-Level Off 5 min.
Refrig/Frozen
WarehouseSetpoint Adj. 15 min.
Data Centers
Setpoint Adj.,
Reduce CPU
Processing
15 min.
Ag. Pumping
Turn Off
selected
pumps
5 min.
Wastewater
Turn Off
selected
pumps
5 min.
End Uses & Response
Shed Estimates: Highest & Lowest hours of the Year
0
0.5
1
1.5
2
2.5
HV
AC
Ligh
tin
g
Ref
rige
rati
on
Dat
a C
ente
rs
Ag
Pu
mp
ing
Was
tew
ater
Tota
l
De
ma
nd
Re
spo
nse
Ca
pa
city
(G
W)
3 a.m. - November 28th2 hour 20 minute
0
0.5
1
1.5
2
2.5H
VA
C
Ligh
tin
g
Ref
rige
rati
on
Dat
a C
ente
rs
Ag
Pu
mp
ing
Was
tew
ater
Tota
l
De
man
d R
esp
on
se C
apac
ity
(GW
)
2 p.m. - September 23rd2 hour 20 minute
0
0.5
1
1.5
2
2.5
HV
AC
Ligh
tin
g
Ref
rige
rati
on
Dat
a C
ente
rs
Ag
Pu
mp
ing
Was
tew
ater
Tota
l
De
ma
nd
Re
spo
nse
Ca
pa
city
(G
W)
2 pm - September 23rd2 hour 20 minute
0
0.5
1
1.5
2
2.5
HV
AC
Ligh
tin
g
Ref
rige
rati
on
Dat
a C
ente
rs
Ag
Pu
mp
ing
Was
tew
ater
Tota
l
De
ma
nd
Re
spo
nse
Ca
pa
city
(G
W)
3 am - November 28th2 hour 20 minute
Highest hour (hot)
(existing controls)
Lowest hour (cold)
(existing controls)
Highest hour (hot)
(Increased Controls)Lowest hour (cold)
(Increased Controls)
0.8 GW
0.25 GW
0.4 GW
1.8 GW
Total
Total
Total
Total
Shed Estimates by End UseShed Estimate Highest Hour = 0.9 GW
Average 20 Minute & 2 Hour - Current Controls
HVAC
Lighting
Refrig WH
Data Ctrs
Ag Pumping
Wastewater
Shed Estimate Lowest Hour = 0.15 GW
Average 20 Minute & 2 Hour - Current Controls
HVACLighting
Refrig WH
Data CtrsAg Pumping
Wastewater
HVAC
Lighting
Refrig WH
Data Centers
Ag Pumping
Wastewater
Shed Estimate Highest Hour = 1.85 GWAverage 20 Minute & 2 Hour - Increased Controls
HVACLighting
Refrig WH
Data Centers
Ag Pumping
Wastewater
Shed Estimate Lowest Hour = 0.41 GWAverage 20 Minute & 2 Hour - Increased Controls
Hottest Hour Coldest Hour
Now
Future
0.8 GW 0.25 GW
1.8 GW 0.4 GW
16
• Preliminary estimate: AutoDR could provide 0.25 to 0.8 GW of
AutoDR ancillary services in the existing stock throughout CA
• Investments to improve controls that are currently
“unreachable” to AutoDR could double the shed potential to 0.4
to 1.8 GW
• Use of open interoperable standards (OpenADR) lowers initial
cost and assures persistence of DR resources
• Need to link advanced controls for efficiency and DR
Summary and Conclusions
17
• Additional end-use load evaluations, especially
ancillary services
• Closer analysis of economics
• Continue to explore load and DR predictions
• Additional off-peak data from field tests & surveys
• Geographic considerations
Future Research
18
Questions? Comments?
David S. Watson
Lawrence Berkeley National Lab
Demand Response Research Center
http://drrc.lbl.gov/
OpenADR Alliance
http://www.openadr.org/
DR Control Strategies Evaluated in Previous Research
19Lawrence Berkeley National
Laboratory
HVAC Lighting Other
Building use Glo
ba
l te
mp.
ad
justm
en
t
Du
ct
sta
tic p
res.
Incre
ase
SA
T I
ncre
ase
Fa
n V
FD
lim
it
CH
W te
mp
. In
cre
ase
Fa
n q
ty.
red
uctio
n
Pre
-co
olin
g
Co
olin
g v
alv
e lim
it
Bo
ile
r lo
cko
ut
Slo
w r
eco
ve
ry
Exte
nd
ed
sh
ed
pe
riod
Co
mm
on a
rea
lig
ht
dim
Off
ice
are
a lig
ht
dim
Tu
rn o
ff lig
ht
Dim
ma
ble
ba
lla
st
Bi-le
ve
l sw
itch
ing
No
n-c
ritica
l p
roce
ss s
hed
ACWD Office, lab X X X X X X X
B of A Office, data center X X X X X
Chabot Museum X X
2530 Arnold Office X X
50 Douglas Office X X
MDF Detention facility X
Echelon Hi-tech office X X X X X X X X
Centerville Junior Highschool X X
Irvington Highschool X X
Gilead 300 Office X
Gilead 342 Office, Lab X X
Gilead 357 Office, Lab X X
IKEA EPaloAlto Furniture retail X
IKEA Emeryville Furniture retail X
IKEA WSacto Furniture retail
Oracle Rocklin Office X X
Safeway Stockton Supermarket X
Solectron Office, Manufacture X X
Svenhard's Bakery X
Sybase Hi-tech office X
Target Antioch Retail X X
Target Bakersfield Retail X X
Target Hayward Retail X X X X
Walmart Fresno Retail X X
Global Temperature Adjustment Amounts for Auto-DR sites
68.0
69.0
70.0
71.0
72.0
73.0
74.0
75.0
76.0
77.0
78.0
79.0
80.0
81.0
82.0
Customer
Te
mp
era
ture
(F
)
Normal
Moderate
High
CPP
Ave. temp. increase (mod): 2.75 F
Ave. temp. increase (high): 1.88 F
DBP
Ave. temp. increase: 2.3 F
DBP CPP one-level GTACPP two-level GTA
Re
tail
Muse
um
Off
ice
1 Re
tail
1A
Off
ice
3
Go
vern
me
nt
1
Re
tail
2
Sch
oo
l
Lab
/Off
ice
2
Go
vern
me
nt
2
Go
vern
me
nt
3
Go
vern
me
nt
4
Re
tail
3
Off
ice
4
Lab
/Off
ice
Av
era
ge
Av
era
ge
Av
era
ge
Simulation tools and strategy
guides have been developed
Shed Estimates: Summer & Winter 24 hr. Profiles
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
De
ma
nd
Re
spo
nse
Ca
pa
city
(G
W)
Demand Response CapacityIncreased Controls
Hot Summer Weekday
2 hour 20 min
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
De
man
d R
esp
on
se C
apac
ity
(GW
)
Demand Response CapacityIncreased Controls
Typical Winter Weekday
2 hour 20 min
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
De
ma
nd
Re
spo
nse
Ca
pa
city
(G
W)
Demand Response CapacityCurrent Controls
Hot Summer Weekday
2 hour 20 min
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
De
man
d R
esp
on
se C
apac
ity
(GW
)
Demand Response CapacityCurrent Controls
Typical Winter Weekday
2 hour 20 min
Summer (Increased Controls) Winter (Increased Controls)
Winter (Current Controls)Summer (Current Controls)
0.8 GW
1.8 GW
0.4 GW
0.25 GW