Energy Storage
at Duke Energy
Tom Fenimore, PE
Emerging Technology Office
September 18
th
, 2014
ETO Technology Screening Funnel
Initial
Testing
Technology
Screening
Technology
Development
Field
Testing
Deployment into
Business unit
Energy Storage Current State
Initial
Testing
Technology
Screening
Technology
Development
Field
Testing
Deployment and
Implementation
Simple function
applications (frequency
regulation) have started
to be deployed in
certain markets (ie. PJM,
ERCOT, CAISO) – Duke
Energy’s Notrees’
Project
Integration of Energy Storage into
the vertically integrated utility ecosystem
to maximize value
is still under development
Energy Storage Benefits
Generation
T & D
End User
Frequency Regulation
Defer System Upgrades
Provide Back Up Power
Renewable Smoothing
Improve Reliability
Utilize lower retail rates
Energy Shifting
Renewable Smoothing
Spinning and Non-
spinning Reserves
Improve Power Quality
(Volt / VAR management)
Limit Peaker Plant Builds
Through pilots we
understand…
Capital Costs
O & M Costs
Installation Hurdles
Operational Issues
Value Streams
…to develop
Business models
Regulatory models
Understand benefits
4
How did
Duke Energy
get here?
Over 25
6
Five differe
locatio
>25
6
Six unique
demonstration projects
installed
functions
Six different
battery chemistries
tested
5
nt
ns
How did
Duke Energy
get here?
Benefit:
Connected on:
• Transmission
• Distribution
• Customer
• With renewables
Diverse chemistries in different capacities
kW
kWh
200 400 600
800
200
400
600
800
25kW/25kWh
Lithium Ion
250kW/750kWh
Lithium Polymer
36MW/24MWh
Advanced Lead Acid
200kW/500kWh
Li Iron Phosphate
200kW/500kWh
Sodium Nickel
75kW/42kWh
Lithium Titanate
How do
SIZE LOCATION
CHEMISTRY FUNCTION
enable value?
on the grid
demonstrated
Notrees Wind Farm Project
Notrees, TX
Major system components:
• 36 MW / 24 MWh
• Xtreme Power Advanced Lead Acid Technology
• Co-located at site of 156 MW Wind Farm in Notrees,
Texas
• Began commercial operation in December 2012
• 50:50 Cost share with DOE
Applications being tested
1 – Ancillary Services: participating in a series of ERCOT
(Texas ISO) market tests to learn how to structure an
efficient market that enables energy storage to provide
ancillary services to the grid.
2 – Energy Shifting: Charging and discharging to maximize
the value of energy delivered to the grid based on timing.
3-Avoiding Wind Curtailment : Using storage to store wind
energy in order to avoid orders to cease providing power to
the grid.
Example: Texas and ERCOT (Electric Reliability Council
of Texas)
• Texas has seen a large amount of wind growth over the
past 10 years.
• This has the created the need for fast responding
resources such as energy storage to help stabilize the
grid
Texas alone has more
wind than France.
15
Example of Daily Wind Output in ERCOT
Wind is highly variable
(the graph to the right
shows a daily wind
profile in Texas)
which makes balancing
supply and demand very
difficult. Fast
responding resources
such as battery
energy storage can help
solve this problem.
16
Example of Battery’s Output and Response Time
Duke Energy’s 36 MW / 24 MWh battery charging and discharging based off market signals to
help with grid stability
17
Discharging
Charging
McAlpine Energy Storage System
McAlpine Creek Retail Substation, Charlotte, NC
Major system components:
• 200 kW / 500 kWh system capacity
• BYD battery and inverter system
• All components integrated within on container
• Lithium-iron-phosphate battery (BYD)
Interconnection:
• Located on a 24 kV distribution circuit
• Interconnected immediately outside of the substation
• Adjacent to 50 kW solar facility on McAlpine test circuit
BYD battery
200 kW/500 kWh LiFePO4
Inverter/Controls
Integrated within one container
System attributes
• In service 4Q 2012
x
• Interconnected next to a 50 kW
solar facility in a planned
islandable micro-grid scheme
that will use the battery for grid
frequency/voltage regulation.
Applications being tested
1 – consolidated inverter/battery construction at a low price
2 – energy shifting applications
a) dispatched based on schedule, local load peaks, etc
3 – integration with solar in a microgrid
a) will be configured with switches, solar, and load to create
an autonomous microgrid that disconnects from the circuit
4 – solar output smoothing/firming
Solar
Islandable Microgrid Schematic
SS
McAlpine Crk.
Retail
Sola
r
Battery
120/208 V
480/277 V
24 kV
24 kV
Ckt 2414
Electronic switch w/
DG protective settings:
“DG Switch”
Normally Closed
75 kVA
(3 x 25 kVA)
Fire
Station
200 kVA
500 kWh
50 kW
Electronic switch to
disconnect from grid:
“Islanding Switch”
Normally Closed
500 kVA
Equipment
Legend
Microgrid
Controller
Manual
Disconnect
Point of common
coupling
Existing
New
Sense
& Control
AC
DC
AC
DC
Inverter
Inverter
19
Microgrid Controller HMI
Rankin Energy Storage System
Rankin Ave. Retail Substation, Mount Holly, NC
Major system components:
• 402 kW / 282 kWh system capacity
• FIAMM sodium nickel chloride battery
• 12 Zebra bus batteries connected in parallel
• 1.25 MVA S&C Electric Company Inverter (SMS)
Interconnection:
• Located on a 12.47 kV distribution circuit
• Interconnected immediately outside of the substation
• Circuit contains a 1.2 MW solar facility ~3 miles away
System attributes
• Installed Dec 2011, in service Mar 2012
• Remotely operable
• ZEBRA bus batteries by FIAMM
for stationary application
development
• Contains fiber connection to
substation relaying; no
connection to the solar facility
on the circuit
1000 kVA transformer
Auxiliary power load center
Steps up 480 V inverter
120V/240V service
output to 12.47 kV
Battery container
Inverter/Controls
402 kW/282 kWh NaNiCl
Storage Management System (SMS)
batteries (12 cells)
1.25 MVA capacity/1.0 MVAR capacity
Applications being tested
1 – centralized solar-induced power swing mitigation
a) senses substation real power loading and uses battery to
“smooth” rapid ramp rates caused by cloud-induced solar
intermittency
b) no direct connection to the solar – designed to smooth
power swings from multiple dispersed solar sites on a circuit
2 – active VAR/power factor management
3 – combined watt/VAR voltage control
a) compensation for rapid solar-induced voltage changes
kW (circuit loading)
(Battery Output)
2550
2500
800
2450
2400
600
2350
discharge
charge
2300
400
2250
2200
200
kW
2150
2100
0
2050
2000
-200
08:00 08:07 08:14 08:21 08:28 08:36 08:43 08:50 08:57 09:04 09:12
Original Circuit Load - Before Smoothing
Target Circuit Load
Measured Circuit Load - After Smoothing
Target Battery Output
Actual Battery Output
Marshall Energy Storage System
Marshall Steam Station, Sherrills Ford, NC
Major system components:
• 750 kWh / 250 kW system capacity
• Kokam Superior Lithium Polymer Batteries
• 1.25 MVA S&C Electric Company Inverter
(SMS)
Interconnection:
• Located on a 12.47 kV distribution circuit
• Separate but adjacent medium-voltage interconnection
from 1.0 MW solar facility
• Located at the end of a distribution feeder
System attributes
• Installed May 2012, in service July 2012
• Remotely operable
• Battery and inverter independently
sourced (both vendors to Duke)
• Located at the Marshall solar test site
where multiple solar technologies are
being field tested on a sealed coal-ash
landfill
1000 kVA transformer
Inverter/Controls
Steps up 480 V inverter
Storage Management System (SMS)
output to 12.47 kV
1.25 MVA capacity/1.0 MVAR capacity
Battery container
750 kWh/250 kW Lithium Polymer
Includes Batt. Mgt. System
1.2 MW solar facility
Applications being tested
1 – energy shifting
a) for system-level arbitrage
b) for local operational constraint management
c) based on forward-looking economic algorithm
2 – solar output smoothing and firming
a) for local feeder voltage management
b) solar-induced power swing mitigation
3 – active VAR/power factor management
4 – combined algorithms / optimization
a) combined energy shifting and smoothing algorithm
b) use of distributed logic with economic, substation, and
local input parameters
Battery
Clay Terrace Energy Storage System
Clay Terrace Mall, IN
Major system components:
• 75 kW / 42 kWh system capacity
• Toshiba lithium titanate battery
• 10 kW roof-mounted solar
• Eaton 50 kW, Siemens 3.3 kW PEV charging stations
Interconnection:
• Behind a commercial meter (customer sited)
• Interconnected at 480V, 3-phase transformer
• Located in the parking lot of a shopping mall
System attributes
• Installed 3Q 2012, in service 4Q 2012
• Designed to
manage and
optimize the
combined energy
profile of solar,
PEV charging, and
storage.
75 kW / 42 kWh Toshiba Li-Titinate
10 kW solar roof-top
PEV DC Fast charging station
50 kW Eaton unit
Level 2 PEV charging
station
J1772 up to 3.3 kW charging
Applications being tested
1 – active management of combined solar, storage and PEV
charging
a) testing energy management system and sizing of a
behind-the-meter system
2 – energy shifting
3 – customer-sited installation aspects
Battery Responding to DC Charge Event
Community Energy Storage Systems
Two units installed
McAlpine 24 kV circuits, Charlotte, NC
Major system components:
• 25 kVA inverter system – S&C Electric Company
• Kokam 25 kWh / 25 kW lithium ion battery
• Battery located in underground vault
Interconnection:
• Interconnected at 120V/240V split single phase
• Configured to serve up to five customers on 50 – 75 kVA
padmount transformers
• Initially connected to one customer each for testing
System attributes
• Installation: unit 1 - Oct 2011; unit 2 - Dec 2011
• Remotely operable and
monitored via DMS
• Demonstrating underground
battery vault configuration
Inverter/control unit
50 kVA secondary
25 kVA connected at 120V/240V
transformer
Battery vault
Battery (underground in vault)
4-feet deep, open bottom
25 kW / 25 kWh Kokam Li-ion battery
pack
Applications being tested
1 – automatic voltage managment
a) automatically injects/consumes VARs to maintain voltage
within a specified setpoint
2 – islanding/back-up power
a) automatic islanding during a grid outage
3 – distributed energy shifting
a) various energy shifting applications using a network of
distributed batteries
4 – control system for distributed storage
a) using distributed communications network to monitor
and dispatch the battery
Need to Consider
• Physical vs. Virtual Energy Storage
• How will energy storage compete / work
together with:
– Smart Invertors
– Demand Response
– Devices which perform autonomous frequency
regulation
Conclusions
• The electric grid is changing: Electric generation is
becoming more “de-centralized” – moving closer to
the end user
• Balancing supply and demand requires a highly
interconnected ecosystem with constant
communication between assets
• Understanding how energy storage can seamlessly be
integrated in this ecosystem is still under development
