Intelligent Efficiency Conference - aceee.org · Intelligent Efficiency Conference 3A Integrating Nanogrids and Microgrids into the Modern Grid Michael R. Starke, PhD, Oak Ridge National

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

austinshines@austinenergy.com

www.austinenergy.com/go/shines

Thank you!

Raymond Kaiser

Director Energy Management Systems, Amzur Technologies

941.320.9866

Raymond.Kaiser@amzur.com

Marc Collins

Senior Principal Consultant, DNV GL

416-522-3064

Marc.Collins@dnvgl.com

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