Intelligent Efficiency
Conference
3A Integrating Nanogrids and Microgrids
into the Modern Grid
Raymond Kaiser, Amzur Technologies
Opening Remarks
Track A: Integrating Distributed
Resources
Intelligent Efficiency
Conference
3A Integrating Nanogrids and Microgrids
into the Modern Grid
Michael R. Starke, PhD, Oak Ridge National
Laboratory
Microgrid Research
Track A: Integrating Distributed
Resources
3 Presentation_name
What is a microgrid? – State of the Art
“A microgrid is a discrete energy system consisting of distributed energy sources (including demand management, storage, and generation) and loads capable of operating in parallel with, or independently from, the main power grid”
– General Microgrids
4 Presentation_name
Micro/Nano Grids – More Recent
Microgrid Controller
Nanogrid Controller
5 Presentation_name
Example of Nanogrid
6 Presentation_name
Next Generation: Networked Microgrids
Microgrid Controller A
Microgrid Controller B
Microgrid Controller C
Distribution Management
System
Intelligent Efficiency
Conference
3A Integrating Nanogrids and Microgrids
into the Modern Grid
Kurt Roth, Fraunhofer Center for Sustainable
Energy Systems
SUNDIAL
Track A: Integrating Distributed
Resources
energy.gov/sunshotenergy.gov/sunshot
An Integrated SHINES System Enabling High Penetration Feeder-Level PV
Kurt Roth, Ph.D.
ACEEE Intelligent Energy ConferenceDecember 6, 2016
energy.gov/sunshotenergy.gov/sunshot 9
High Penetration PV is:
Sources: Steward Health Care, Thermofab, Wikimedia Commons.
Rated PV Power = Peak Facility Loads
energy.gov/sunshotenergy.gov/sunshot10SHINES Kickoff Meeting 2016
10
Challenge: PV Intermittency
Source: Curtright and Apt (2008).
Power ramps up to 50% of peak output in one
minute
energy.gov/sunshotenergy.gov/sunshot
1
1
Challenge: Solar Surplus on Sunny Spring Days
0
250
500
750
1,000
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
Ho
url
y kW
h
Hour
April 24Facility [kWh]
PV Production [kWh] • Worcester, MA climate
• “Typical” April 24th
• Big Box Retail (simulated)
• PV = 1,000 kW
• Building Peak = 1,000 kW
• Thigh = 73oF
Sources: DOE/OpenEI, Fraunhofer
calculations.
energy.gov/sunshotenergy.gov/sunshot
1
2
The Solution: Storage + Integrated optimized system control
• 1MW of Managed Loads
• 1MW of PV Solar
• 0.5/1.0MWh of storage
Sources: National Grid, Steward Health Care, Wikimedia Commons.
energy.gov/sunshotenergy.gov/sunshot
1
3
Solution: Mitigate Solar Surplus
• Mix of C&I facilities
• PV = 1,000 kW
• Building Peak = 1,000 kW
• July 21, 2015
Sources: DOE/OpenEI, Fraunhofer
calculations.
energy.gov/sunshotenergy.gov/sunshot
1
4
Architecture – Major Components
BATTER
Y
STORA
GE
PLANT
MASTER
CONTROLLE
R
C&I
FACILITY
LOAD #1
C&I
FACILITY
LOAD #2
C&I
FACILITY
LOAD #3PV ARRAY +
Inverter (1MW)
FEEDER
energy.gov/sunshotenergy.gov/sunshot
1
5
Architecture – Major Components
BATTER
Y
STORA
GE
PLANT
MASTER
CONTROLLE
R
C&I
FACILITY
LOAD #1
C&I
FACILITY
LOAD #2
C&I
FACILITY
LOAD #3PV ARRAY +
Inverter (1MW)
FEEDER
energy.gov/sunshotenergy.gov/sunshot
1
6
Architecture – Major Components
BATTER
Y
STORA
GE
PLANT
MASTER
CONTROLLE
R
C&I
FACILITY
LOAD #1
C&I
FACILITY
LOAD #2
C&I
FACILITY
LOAD #3PV ARRAY +
Inverter (1MW)
FEEDER
SUNDIAL
GLOBAL
SCHEDULER
UTILITY & ISO
COMMUNICATIONSWEATHER
FORECAST
energy.gov/sunshotenergy.gov/sunshot
1
7
Architecture – Major Components
BATTER
Y
STORA
GE
PLANT
MASTER
CONTROLLE
R
C&I
FACILITY
LOAD #1
C&I
FACILITY
LOAD #2
C&I
FACILITY
LOAD #3PV ARRAY +
Inverter (1MW)
FEEDER
SUNDIAL
GLOBAL
SCHEDULER
UTILITY & ISO
COMMUNICATIONSWEATHER
FORECAST
FACILITY LOAD AGGREGATION & MANAGEMENT ENGINE (FLAME)
energy.gov/sunshotenergy.gov/sunshot
1
8
SunDial: A Vision for Integrating Hundreds of GW of Solar
SunDial Objectives
• Create extensible framework to readily integrate loads, storage, and PV
• Test and pilot business models and market mechanisms to enable high PV penetration
Market Transformation: A transparent, low-friction market for storage / solar integration on the feeder level
• Flexible with respect to markets: multiple use cases, vendors, and business models
• Potential T&D deferral
• Avoided system upgrades
• Virtual Power Plant , etc.
• Flexible with respect to asset location, ownership, and type
Year-long Demonstration Project
Intelligent Efficiency
Conference
3A Integrating Nanogrids and Microgrids
into the Modern Grid
Lisa Martin, Austin Energy
Austin SHINES
Track A: Integrating Distributed
Resources
2
A bit about Austin Energy
DOE SunShot & SHINES Vision
21
The projects will work to dramatically increase solar-generated electricity that can be
dispatched at any time – day or night – to meet consumer electricity needs while
ensuring the reliability of the nation’s electricity grid
SHINES Conceptual Architecture
Utility Scale Energy
Storage + PV
Commercial Energy
Storage + PV
Residential Energy
Storage + PV
Illustrative
DER Management
Platform
22
Potential DER Control System Applications
23
Peak Loss Avoidance
Voltage Support
forTrans.
Peak LoadReduction
(4-CP) Energy Arbitrage
LMPOpportunities
Regulation Up/Down
Fast Frequency Response
EmergencyResponse
Service
SolarVariance
WindVariance
CongestionMgmt
VoltageSupport
Power FactorCorrection
Loss Avoidance
Harmonics
Back-upPower
DemandCharge
Reduction
Time of Use
Trans.
Constraint
Avoidance
Austin Energy
considered 19
applications
during conceptual
design of its DER
Control System
Thank you
24
Lisa MartinAustin SHINES Project Manager
www.austinenergy.com/go/shines
Thank you!
Raymond Kaiser
Director Energy Management Systems, Amzur Technologies
941.320.9866
Marc Collins
Senior Principal Consultant, DNV GL
416-522-3064
Visit ACEEE on the Web:
www.aceee.org
Residential Components
26
*Market transactions in the SHINES project will be
simulated only and included in LCOE analysis.
DG-DERO Aggregator (Third-Party)
Pecan Street Aggregator
Sites x6
PV and ESS
Direct Utility Control
Sites x12
PV only
ERCOTSimulated*
Autonomous Sites x6PV only
Auto Auto
Electricity
Information
Value
Legend
Commercial Components
27
3rd Party Aggregator Sites – 400kW 5x – 30kW
2x – 125kW
Dispatch Priority: Customer value propositions
Direct Utility Control Sites – 155kW1x – 30kW
1x – 125kW
Dispatch Priority: Utility reliability needs
API or
DNP3
*Market transactions in Austin SHINES will be
simulated only and included in LCOE analysis.
DEROAggregator(Third-Party)
ERCOTSimulated*
Electricity
Information
Value
Legend
**estimated
Utility-scale Components
28
*Market transactions in the
SHINES project will be
simulated only and
included in LCOE analysis.
DERO
ERCOTSimulated*
DG-
IC
DG-
IC
Kingsbery ESS
1.5 MW / 3MWh
Single 46’ ISO container
Kingsbery
Community
Solar
2 MW
Rooftop
Solar
@Mueller
Mueller ESS
1.5 MW total**
Modular container design
Electricity
Information
Value
Legend
energy.gov/sunshotenergy.gov/sunshot
2
9
A Market for Aggregated, Feeder-Scale Demand-Side PV Support
Multiple potential business models accessible to multiple participants
• Potential T&D deferral
• Avoided system upgrades for storage- and load-
aggregated PV
• Virtual Power Plant
• Robust alternative to net metering
• Multiple markets: day ahead, real time, demand response, capacity
• Bid into markets as a single controllable aggregated resource
• Future localized market for grid support
SunDial enables assets…
…from different owners…
…at different locations…
…to engage in cooperative
business models
energy.gov/sunshotenergy.gov/sunshot
3
0
Different Use Cases
Use Case Goal Battery Storage FLAME
PV IntermittencyLimit max. rate of change to <10%/min
Seconds to minutes~5-15 minutes (fans, pumps, lighting)
Feeder-scale Load Shaping
Limit net power flow and morning/evening ramps
15 min to 4+ hours15 min to 4 hours (pre-cooling, HVAC)
Peak Load Shaving / Demand charge reduction
Match generation and loads
15 min to 4+ hours15 min to 4 hours(pre-cooling, HVAC)
Volt-Var Optimize voltage Real/Reactive power n/a
Illustrative Examples
energy.gov/sunshotenergy.gov/sunshot
3
1
The Concept
Physically decouple storage, PV, and load management
• Global Scheduler: Feeder-scale global optimization engine
• Optimization over varying timescales and use cases
• Leveraging PV, storage, AND aggregated load management resources
• FLAME: Facility load aggregation and management engine
• Based on an existing, proven demand response aggregation business model
• Plant Master Controller: Local, fast, site-level control of PV and storage
• Utilizing standard utility-scale PV/Storage control and integration capability
• Newly developed interoperability interfaces
Enables a transparent, broadly scalable mechanism to achieve and simplify feeder-
scale integration of PV, loads, and battery storage
energy.gov/sunshotenergy.gov/sunshot
3
2
SunDial Global Scheduler
Works for Different Use Cases
• PV intermittency mitigation
• Load Shaping
• Peak Load Reduction
• And more…
Determines System State
(Current & Predicted Future)
• Solar resource
• Battery
• Loads and Load Sink/Shed Potentials
• Grid Constraints, Pricing
Performs Optimization
• Minimize cost based on objective function defined by the current use case
• Shrinking horizon scheduling approach
• Updated according to new information at subsequent scheduling steps.
Generates Control Signals
• PMC, FLAME, Battery
Implemented as an extension of, e.g., PNNL’s VOLTTRON distributed control and sensing platform
energy.gov/sunshotenergy.gov/sunshot
3
3
Meeting SHINES FOA Technical Targets
• LCOE: $0.14/kWh with $1.55/W solar; $0.10/kWh with
$1.00/W solar in MA
• Efficiency: 90% RT efficiency achievable
• Displace ~25% of electrochemical storage throughput with load
management
• approaches or exceeds 100% RT efficiency
• Co-located storage on the primary side of the MV transformer
• Component lifetimes:
• Limit cycling on battery through load management
• Account for replacement in lifetime LCOE calculations
energy.gov/sunshotenergy.gov/sunshot
3
4
Project Outcomes
• Standardized interoperability interface for integration of aggregated loads
with DG
• Develop new, low-friction market mechanism for localized PV support
services
• Leverage aggregated resources to reduce interconnection complexity
• Commercial implementation of distribution-scale DSM aggregation engine
for integration with solar
• Demonstrate technical and commercial feasibility of scalable approach for
decoupled solar, storage, and load management
energy.gov/sunshotenergy.gov/sunshot
3
5
Facility Load Aggregation & Management Engine (FLAME)
Statistical representation of expected portfolio loads and shed/sink potentials and their costs, over time
Curtailment script within customer acceptance parameters
Building model calibrated with prior load control events
Predicted loads based on historic building data and exogenous weather factors
Portfolio Resource
Automated Facility Dispatch
Load Sink & Shed Potential
Load Forecast