PG&E’s PG&E’s 2009 Participating 2009 Participating Load Pilot Load Pilot
Dec 31, 2015
PG&E’s PG&E’s 2009 Participating Load 2009 Participating Load
PilotPilot
2
Overview
Regulatory Context
Pilot Characteristics
Lessons
Next Steps
3
Regulatory Context
December 2007
Federal Energy Act of 2007 (EISA 2007)
February 2008
CPUC directs CA IOUs to further integrate Demand Resource in wholesale markets
October 2008
FERC issues No. 719
December 2008
CPUC approves pilots that integrate in CAISO wholesale market
4
Participating Load Pilot
Can we create demand response for the wholesale market with the attributes of
supply side alternatives?
5
“What is Participating Load?”
Load acting as a resource to:
Participate in the wholesale market and offer ancillary service and energy products
Load resource must be scheduled daily on an hourly basis, even without the DR load reduction:
Separated from overall load schedules
Must meet all CAISO technical specifications; e.g., 5-minute settlement metering, 10-minute responses, 4-second real time demand meter data, etc…
6
Pilot Characteristics
Customer recruitment:
Retail (HVAC), Industrial (Process),Local Gov. Office (HVAC)
No co-generation or back up generation
Already on Auto-DR
Customer benefits:
Receive Capacity + Energy (if called) incentives
Able to receive real-time data
Convenience and ease by having pre-determined load shed strategies; peace of mind
CAISO ADS
7
Pilot Characteristics
For real time meter data, PG&E used real time equipment to harvest meter data from end use sites transmit to the CAISO.
Transmit meter data in 4-seconds.
Communication must be running at all times even without DR bids.
Setup node points in CAISO’s EMS and Full Network Model (FNM) database.
For notification of a dispatch events, once CAISO initiates the event, Auto-DR triggers the customers’ pre-determined load shed strategies within the customers’ EMS and sends an email alerting the customer of an event.
8
Pilot Operations
Pilot ran from July 29th to October 31st
Day-Ahead Non-Spinning Bid during Monday – Friday Bid in as both contingency and economic
9
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
:05
10
:45
11
:25
12
:05
12
:45
13
:25
14
:05
14
:45
15
:25
16
:05
16
:45
17
:25
18
:05
18
:45
19
:25
20
:05
20
:45
21
:25
22
:05
22
:45
23
:25
Time of Day
Dem
an
d (
kW
)
Actual 5 min. Data Hourly Forecast with reduction Forecasted Data
PLP EVENT
Example of PLP Event
10
Lessons: Technology
Successful system integration:
Real-time (4-second instantaneous) data delivered from customers site to CAISO EMS systems
Seamless dispatch from CAISO to customers
Customers did not recognize that an event was dispatched
Feedback mechanism control between Auto-DR server to customer
11
Lessons: Operations
Customers responded when called by the CAISO
On average, resources were able to respond within 4 minutes of receiving the dispatches
A total 16+ hours were called for non-spinning reserves
More experience should lead to less forecasting errors
12
Lessons: End Use Resources
Retail and office building customers are ideal candidates
Dominant use of HVAC load
Industrial participant less predictable
High customer satisfaction:
Since everything was automated, customers felt participation was easier
Events were relatively short and not disruptive
13
Next Steps
2Q 2010: Continue to work on rules surrounding direct participation with the CPUC
3Q 2010: CAISO is scheduled to release and operate “Proxy Demand Resource” which will allow direct participation
1Q 2011: Field study for pilot demonstration on intermittent renewables
1Q 2011: Submit 2012 – 2014 DR portfolio proposal to CPUC. Portfolio will include programs that will have mechanism for bidding into wholesale market